Nick Tilley

The Crime Drop and the Security Hypothesis

Major crime drops were experienced in the United States and most other industrialized countries for a decade from the early to mid-1990s. Yet there is little agreement over explanation or lessons for policy. Here it is proposed that change in the quantity and quality of security was a key driver of the crime drop. From evidence relating to vehicle theft in two countries, it is concluded that electronic immobilizers and central locking were particularly effective. It is suggested that reduced car theft may have induced drops in other crime including violence. From this platform, a broader security hypothesis, linked to routine activity and opportunity theory, is outlined.

Handbook of crime prevention and community safety

With all due apologies to the aggregate age/crime curve, there is little denying that there is great variability between offenders and within offenders over time with respect to offending over the life course. Many persons offend, but only do so once. Others offend two or three times, and then stop. And a much smaller set of persons offend with high frequency and over longer periods of time. But much like a roller-coaster’s initial large incline to the top, participation and frequency of offending also exhibit declines over the life course until some finite (but difficult to identify without fail) endpoint. This decline in offending, is what is commonly referred to (but not always conceptually agreed upon) as desistance from crime. The topic of desistance is not necessarily a new one as it can be traced back to seminal works in the 19th century by both Quetelet (1831/1984) and Beccaria (1764/1992). However, it did not begin to take hold, at least empirically, until the Philadelphia Birth Cohort Study by Wolfgang et al. (1972/1987), who were the first to study transition probabilities and recidivism in a large sample of boys followed to late adolescence. Research on desistance, however, began to flourish after the U.S. National Research Council’s report on criminal careers (Blumstein et al., 1986), in which the Panel outlined a new paradigm that parceled out an individual’s offending career into distinct dimensions, with desistance emerging as a critical feature. Not surprisingly, much theoretical work was devoted to studying desistance, and policy-relevant discussion ensued in earnest, again not surprisingly because the justice system in general, and the correctional apparatus in particular, has a keen interest in curtailing offenders from persistent offending. Unfortunately, the study of desistance requires longitudinal data spanning at least the first two decades of life, but ideally would follow individuals much longer as antisocial behavior continues (to zig and zag) well into middle to late-middle adulthood. It was only until the last quarter century that such data sets “aged,” much like a fine wine or bourbon, and permitted such analyses both in the US and abroad. In this chapter, we provide an overview of the theoretical and conceptual issues surrounding the topic of desistance, as well as a review of the theoretical and empirical work regarding it. As well, we pay particular attention to the role of desistance in prevention, intervention, rehabilitation, and punishment programs and policies, and we highlight a few directions for empirical research. In doing so, we take a look at whether it is possible to “force the plant,” as the Gluecks (1937/1966, p. 205) wondered nearly 80 years ago, “so that benign maturation [desistance] will occur earlier than it seems to at present.”

Exploring the international decline in crime rates

This paper examines aggregate crime trends and variation around them from 1988 to 2004 for 26 countries and five main crime types using data from the International Crime Victims Survey. Multilevel statistical analysis is used to identify the main trends. Major drops in crime were experienced in many countries from the early to mid-1990s onwards. The current analysis estimates that between 1995 and 2004, the mean international crime incidence reductions were: 77.1 percent in theft from cars, 60.3 percent in theft from person, 26.0 percent in burglary, 20.6 percent in assault and 16.8 percent in car theft. The study results suggest that, with the exception of burglary, all examined crime types fell by roughly the same rate across countries. The sample’s small number of countries in Latin America, Africa and Asia experienced even steeper reductions in burglary than occurred in Europe, North America and Australia.

Why the Crime Drop

The “crime drop” is the most important criminological phenomenon of modern times. In North America, Europe, and Australasia, many common crimes have fallen by half or more since the early 1990s, albeit with variation in the specifics. Seventeen explanations are examined here including demographics, policing, imprisonment, drug markets, and lead poisoning. Pioneering research relevant only to the United States now appears, with the benefit of hindsight, somewhat parochial. Sixteen of the 17 hypotheses fail one or more of four evidence-based standardized tests on which they are assessed. The one that passes is the security hypothesis, underpinned by crime opportunity theories. Here there is strong evidence that vehicle theft fell because of more and better security, and mounting evidence that improved security was critical in reducing burglary and other acquisitive crime. Many crime types are interrelated, while most criminal careers are dominated by property crime, so removing these volume crimes might be expected to reduce violence.

The effectiveness of vehicle security devices and their role in the crime drop

Car theft in the UK fell two-thirds from the mid-1990s as part of more widespread crime drops, and has been attributed to improved vehicle security. This study develops a Security Impact Assessment Tool (SIAT) to gauge the contribution of individual security devices and their combination. The metric of impact derived is termed the Security Protection Factor (SPF). Cars with central locking plus an electronic immobilizer, and often an alarm, are found to be ‘SPF 25’, that is, they were up to 25 times less likely to be stolen than those without security. That impact is greater than expected from the individual contributions of those devices, and is attributed to interaction effects. Tracking devices are found to be particularly effective but rarer. Protective effects were greater against theft of cars than against theft from cars or attempts, almost certainly reflecting the difficulty imposed on thieves by electronic immobilizers. It is suggested that this type of analysis could be usefully extended to other crime types and security combinations. The analysis also lends support to a ‘security hypothesis’ component of an explanation for the major national and international crime drops that is based in the criminologies of everyday life.

Caring communities, paradigm polemics, design debates

In a previous issue of this journal (Vol. 3, no. 2) David Farrington introduced—‘in the interests of stimulating discussions’—an outline research design for Evaluating a Community Crime Prevention Program. This article takes up the challenge, offers a critique of his ‘quasi-experimental’ approach and insists that such programs are better evaluated using designs based on ‘scientific realist’ principles. The paper also draws attention to benefits of a ‘theories of change’ approach to evaluation, pioneered in the US in conjunction with such comprehensive community initiatives.

Implementing Crime Prevention

Situational crime prevention is increasingly proving its worth in crime control, but crime rates in many countries continue to rise. Existing small-scale, but effective, measures should be more widely duplicated. There is a need at all levels to identify mechanisms that will lead to more widespread implementation of situational approaches. Legislative mandates and publicity exhortation can be used, but the demonstration to those with the authority to act that situational measures result in cost savings may be more effective. A prerequisite of such a demonstration is the availability of good data and their analysis. This points to a considerable training task and the need to introduce criminological theory and good research and evaluation practice to those with the authority and responsibility to take action against crime.

Sherman vs Sherman: Realism vs Rhetoric’

There are two Lawrence Shermans. One is a realist and the other a rhetorician. I shall refer to the realist Sherman as Sherman1 and the rhetorician Sherman as Sherman2. Sherman1’s work, I’ll suggest, is of huge value, while the directions suggested by Sherman2 pose risks to research, practice and important liberal principles. I also think there are underlying tensions between the two Shermans. Before getting to the details, I should make clear that there is a good deal of common ground between both Shermans and, for that matter, the author of this short note! All take the view that criminal justice policies and practices in general, and in policing in particular, could be substantially improved by more systematic attention to evidence about the effects of what is delivered. All take the view that better use might be made of past research, whose systematic synthesis promises more than a simple catalogue of individual findings. All are, broadly speaking, liberals, who are in favour of maximizing freedom and minimizing harm.

Applying theory-driven evaluation to the British Crime Reduction Programme The theories of the programme and of its evaluations

The Crime Reduction Programme began in April 1999 and was largely finished by March 2002. It went through a number of major changes. It also incorporated an ambitious initial evaluation component and a range of forms of evaluation activity. This article traces the changing theories in and of the programme as a whole and the changing forms of evaluation that were conducted under its auspices. It attempts to explain how and why the programme metamorphosed and how and why provisions for evaluation activity evolved as they did. It tries to distil realistic lessons for the conduct of informative evaluation in the context of large-scale government programmes. It also attempts to locate the potential for realist evaluation within such programmes.

Offenses around Stadiums: A Natural Experiment on Crime Attraction and Generation

Objectives: Inspired by ecological theories of crime, the aim of this study was to make use of a natural experiment to see if a U.K. soccer stadium generates or attracts crime in the area that surrounds it. Method: Data for theft and violent crime around Wembley stadium are analyzed to see if the rate (per-unit time and ambient population) of crime differ for days on which the stadium is used and those it is not. In addition, differences in the spatial and temporal distribution of crime are examined for these two types of days. Results: Analyses indicate that on days when the stadium is used, the rate of crime per-unit time is elevated, but that the rate per ambient population at risk is not. The spatial and temporal pattern of crime also clearly differs for the two types of days. For example, the level of crime is elevated in the surrounding area when the stadium is used relative to when it is not. Conclusions. The case study suggests that the facility studied contributes to levels of crime in the area that surrounds it. The research provides further support for ecological theories of crime and their utility in informing criminological understanding and policy-related questions.