Displacement

Displacement

Displacement, or more specifically crime displacement, is the phenomenon of criminal activities shifting in some form as a response to crime prevention initiatives such as increased policing, neighbourhood watch, and target hardening. Displacement is an important phenomenon to consider when designing crime prevention initiatives because if crime is simply displaced to other locations, times, methods, crime types, or victims, crime prevention is a zero-sum game for society. Moreover, if displacement of some form and degree does occur the societal benefits to crime prevention are not as great. This forces us to ask whether crime prevention initiatives should be undertaken or not, as well as what crime prevention initiatives should be undertaken given that we expect some form of displacement to occur. This entry discusses the context of displacement in terms of theoretical expectations, the forms of displacement, methodological issues with measuring displacement, and evidence for the existence of displacement.

The Displacement Model of Crime

The fact that researchers and practitioners discuss the possibility of displacement means that there is some underlying belief in the structure of crime that would lead to such a phenomenon. Consequently, there is a model of displacement in the criminological literature that is sometimes referred to explicitly and, sometimes, implicitly. Steven Lab outlines the four primary assumptions of such a model, a useful starting point for considering the impact of crime prevention initiatives.

The first assumption is that offenders have a need to commit a certain number of offenses. The place, time, crime classification, method, and target are not particularly important, crime must simply be performed. Such a view almost necessarily implies some form of pathology or anti-social characteristic within offenders that force them to satisfy this need. Steven Lab uses the term inelastic to describe this need for committing crime. Inelastic is a term used in choice theory (most commonly in microeconomics) to describe a product that must be consumed at a certain level. The best example of an inelastic good is insulin. Regardless of the price, a diabetic must consume a certain quantity of insulin to avoid health complications. In the context of the offender, regardless of the “price” of offending (risks, payoffs, etc.) crime must be committed.

The second assumption is that offenders are willing to be mobile. This simply means that offenders are willing to operate in different locations, at different times, use different tactics, select different targets, and perform different types of crime. Consequently, if a crime prevention initiative is undertaken, then offenders must be flexible and be generalists.

The third assumption is critical and become very important when questioning the displacement model of crime: offenders act rationally. Though a contentious statement at times, the rational offender is the only type of offender that is displaced. Otherwise, the offender will continue to undertake crime in the manner before any crime prevention initiative began. Derek Cornish and Ronald Clarke point this out through the use of the term choice-structuring properties. Choice-structuring properties of crimes involve the assessment, by offenders, of the potential payoffs of a crime and its corresponding risks. It is because of any changes in payoffs and risks that offenders choose to displace their criminal activities.

The fourth, and last, assumption is that alternative places, times, methods, crimes, and targets are available to the offender. If the second assumption of willing to be mobile is satisfied, then this fourth assumption must also be satisfied for crime to be displaced.

Questioning the Displacement Model of Crime

The purpose of questioning the displacement model of crime is not to deny that displacement exists, or that some offenders are always willing to commit crime no matter how it comes about. Rather, the purpose of questioning this model is for the purposes of showing that displacement is not absolute.

The assumption of the inelastic commission of crime assumes that the root of all crime is within the offenders. As mentioned above, this could be some pathological condition or some other anti-social characteristics. Both Marcus Felson and Ronald Clarke challenge this view. They state that it is opportunity that is at the root of crime. In his book Crime and Everyday Life, Marcus Felson goes into great detail to show how this is the case. They do not deny that motivation for committing crime exists, but that motivation means nothing without the presence of suitable targets. If criminal activities are at least partially opportunistic and/or unplanned, the assumption of the inelastic commission of crime is violated.

The assumption of mobility is also easily challenged. Though the methods used to commit a crime, the crime type chosen, and the victims of crime have some degree of substitutability, location and time are not always flexible. Patricia and Paul Brantingham outlined the importance of geography to criminal offenses, describing the geometry of crime. Offenders do not commit crime randomly in their environment. Rather, crimes are committed in the places they frequent most often as well as the pathways between them. In the context of time, there are certain commitments that people have. Consider a young offender in school; if the criminal opportunities between 3 and 6pm are removed, the young offender has few options to commit other crimes.

The assumption of the rational offender is particularly instructive here. If offenders change their criminal activities because of the perception of increased risks or decreased rewards, then it must follow that risks can be made high enough and reward low enough to have some (or most) offenders desist from criminal activities. As such, the displacement model of crime contains an assumption that does not have to be shown to be violated. The mere presence of the assumption states that displacement is not inevitable.

Lastly, the assumption of (available) alternative targets for crime is probably the easiest to accept. As discussed below, there is evidence of displacement so there must be alternative targets available. However, not all targets are equally attractive to all offenders. This may be the simple reason why crime concentrates in particular areas of urban (and other) environments. The easy and most attractive targets are chosen first. Once removed, less attractive and more difficult targets are selected, and so on. However, this process can only go so long before the payoffs are simply not worth the risks.

Forms of Displacement

Regardless of whether one subscribes to a displacement model of crime or not, displacement comes in a number of forms. Over thirty years ago, Thomas Reppetto listed five different forms of displacement: territorial, temporal, tactical, target, and functional. Each is discussed, in turn.

Territorial, or geographical, displacement is the most common form of displacement that is expected to occur. To paraphrase David Weisburd and colleagues, crime simply moves around the corner. For example, if there is some form of crime prevention activity in one neighbourhood (neighbourhood watch, increased police presence, target hardening, etc.), criminals simply move to another neighbourhood without such crime prevention activities. Most often, territorial displacement is assumed to occur in neighbourhoods, or areas, close to the original crime area, but this is not necessarily the case. Rather, crime moves to the next best location for committing the relevant offense that may be in close proximity.

Temporal displacement occurs when offenders commit their offenses at different times of the day, days of the week, etc. This form of displacement will only occur with certain forms of crime prevention activities. These are the type of activities defined by the presence of a citizen or police patrol that only occur at certain times of the day or week. Depending on the nature of the offenders, who may also be constrained by time, temporal displacement may be expected. Crime prevention activities such as target hardening is not expected to result in temporal displacement because such an activity is always present. Additionally, if temporally-based crime prevention activities are targeted at youth that are enrolled in school (extra police patrols between 3 and 6 pm, for example), temporal displacement is not expected to occur.

Tactical displacement is a change in methods used by offenders to commit crime. This form of displacement is most likely a response to target hardening crime prevention initiatives. Changes in the technology to prevent automotive theft has shown this to be the case quite clearly with the advent of steering column locks, steering wheel locks, alarms, and electronic disabling devices. Automobiles are still stolen, just through different means.

Target displacement occurs when there is a change in victims, but no change in the area where offenders search for opportunities. A prime example would be a target hardening initiative that only some proportion of homes undertakes within a neighbourhood. If that crime prevention initiative is identifiable by the offenders (burglar alarm signage, for example), the remaining homes in that same neighbourhood would become the victims of crime. Though the distinction between target and territorial displacement is an important one, target displacement may be thought of as micro-territorial displacement: crime still occurs, it just occurs next door not in another neighbourhood.

Functional displacement occurs when an offender actually stops committing one type of crime and switches to another type of crime. In this situation, the crime prevention initiative simply removes all criminal opportunity, literally, or makes the commission of crime so difficult that it is easier for offenders to switch their activities, albeit to other criminal activities. Functional displacement may also be sub-divided into two different forms: benign or malign. Robert Barr and Ken Pease are apt to point out that even if displacement occurs, perfectly, it may actually benefit society. Though this may initially be thought of an odd statement, most people, if asked about their “preferences” for victimization would prefer one crime over another; the cost of a simple assault to the individual and society is far less than the cost of a sexual assault. Of course, functional displacement may go the other way: rather than committing automotive thefts, offenders may switch to committing armed robberies.

Lastly, Robert Barr and Ken Pease also contribute to the measurement of different types of displacement through the addition o a sixth form of displacement: perpetrator displacement. In this form of displacement, one offender stops criminal activity (incarceration, ages out of crime, begins to view criminal activities as too risky, etc.) but another takes his/her place. Alternatively, though a crime prevention initiative may make crime unattractive (too risky or too few rewards), allows another offender to see opportunity.

It should be clear from this brief discussion that it is possible to predict, to some degree, the form of displacement that will occur. Consequently, if one holds the view that displacement is not always perfect, or one hundred percent, crime prevention activities may be taken in stages to minimize overall displacement.

Methodological Issues for Measuring Displacement

Being able to classify and, potentially, predict the form of displacement is a powerful tool for those involved in crime prevention. However, actually measuring displacement can itself be problematic. Though there are other difficulties outlined in some of the further readings provided below, one methodological issue is particularly problematic: measuring significant change. The most obvious example of this problem is outlined using territorial displacement, but similar difficulties exist with other forms of displacement.

Consider an urban environment that is divided into a set of equally sized neighbourhoods, representing a grid. Suppose there are nine of these neighbourhoods organized in a square and a crime prevention initiative is undertaken in the centre neighbourhoods. The crime prevention initiative is undertaken in the middle neighbourhood because it has the greatest volume of crime (an average of ten per some unit of time) whereas the other neighbourhoods have much lower volumes of crime (averaging five per the same unit of time). Now suppose the crime prevention initiative is a success with a fifty percent reduction in crimes: now five per unit of time. However, there is perfect displacement. The five crimes that are reduced now occur in five of the adjacent eight neighbourhoods. Here’s the problem: because crime will vary over time, an increase of one crime in five of the other neighbourhoods will likely not lead to any statistically significant changes being measured. Based on averages, those assessing the crime prevention initiative may think that territorial displacement has occurred, but cannot make any meaningful statements. This problem only becomes more difficult when the number of neighbourhoods that crime may move to increases.

This measurement problem, however, is not confined to territorial displacement, as indicated above. If an assessment of increased police presence after school hours is being undertaken, then temporal displacement must be considered a possibility. Again, suppose crime falls during 3 – 6pm, but there is a corresponding increase in crime at other times of the day. If the comparison temporal period is the rest of the day, then displaced crimes that were committed three hours of the day will be counted with crimes that occur in the other twenty-one hours of the day. Measuring any statistically significant change will likely be very difficult even if perfect displacement occurs. Tactical, target, functional, and perpetrator displacement may also be difficult to measure for similar reasons; it take time for some of these forms of displacement to occur, so is it due to actual displacement or the variability of crime over time?

(A Lack of) Evidence for Displacement

This methodological difficulty, and others not discussed here, has not stopped researchers from attempting to assess the presence of displacement from crime prevention initiatives. Though there has not been a comprehensive review for over a decade, the 1990s saw three such reviews of the empirical assessments of displacement undertaken by Robert Barr and Ken Pease, John Eck, and René Hesseling, all listed in the further readings.

Because it is easiest to measure, it should come as no surprise that territorial displacement is the most common form of displacement investigated. This is followed by target displacement and offense displacement—temporal and tactical displacement are studied the least of all forms. All three of the reviews published in the 1990s come to the same conclusion: displacement is not inevitable and if present is not significant enough to have cause for concern. These are rather strong statements to be made by these authors because more often than not there is evidence of some, but not perfect, displacement. In fact, both John Eck and René Hesseling suggest that different types of crime prevention initiatives and different crimes will have different degrees of displacement. Given the methodological problem outlined above and the fact that these authors can identify differential degrees of displacement for different crime prevention initiatives and crimes, displacement should be taken seriously. Because of the methodological problem outlined above, the measurement of displacement is likely to be biased downwards.

An additional problem for assessing the evidence for displacement, as pointed out by David Weisburd and colleagues, is that most evaluators of crime prevention initiatives do not employ methodologically sound research designs for detecting displacement. This lack of methodologically sound research designs is because of a number of constraints (time, money, and other resources), but is very real and has not been dealt with until very recently in the context of territorial displacement.

David Weisburd and colleagues set out to assess territorial displacement in the context of drug crimes and prostitution. They assumed that if territorial displacement was to occur, it would occur within these crime classifications because of the explicitly economic nature of these crimes. In their study, many of the methodological problems are dealt with at the outset of the research or dismissed based on their results. They find no evidence of territorial displacement, but do find evidence for the diffusion of benefits. Though discussed in detail in another entry, the diffusion of benefits is when crime is reduced in areas close to, but not included in, the targeted crime prevention area. As such, the methodological concern outlined above is not relevant in this particular case.

However, there are some limitations in their analysis. David Weisburd and colleagues only ask whether or not crime moves around the corner. This assumes that the second best locations for crime are geographically close to the targeted area. This is not necessarily the case. It may be that the second best area for crime is across town and easily accessible by public transit—a parking lot at another shopping mall five kilometres away, for example. Given the limitations of the displacement model of crime, outlined above, it is unlikely that perfect displacement will occur, but only measuring crime in immediately adjacent areas is limited, at best. Also, David Weisburd and colleagues find evidence for tactical displacement. Though the quantification of this form of displacement is difficult, the fact that is can be observed or measured indicates that its presence is significant.

Consequently, the current state of research evaluating displacement indicates that displacement is not a significant concern for those implementing crime prevention initiatives. Or, alternatively, displacement is not significant enough to discourage the implementation of crime prevention initiatives. However, as with most social science, very few definitive statements can be made.

Martin A. Andresen

See also Burglary, Prevention of; Crime Prevention; Crime Prevention Models; Crime Prevention Through Environmental Design; Crime Prevention, Domains of; Diffusion of Benefits; Methodological Issues in Evaluating Effectiveness of Crime Prevention; Situational Crime Prevention

Further Readings

Barr, R., & Pease, K. (1990). Crime placement, displacement, and deflection. Crime and Justice: A Review of Research, 12, 277 – 318.

Brantingham, P.L., & Brantingham, P.J. (1981). Notes of the geometry of crime. In P.J. Brantingham & P.L. Brantingham (Eds.), Environmental Criminology (pp. 27 – 53). Beverly Hills, CA: Sage Publications.

Cornish, D.B., & Clarke, R.V. (1987). Understanding crime displacement: an application of rational choice theory. Criminology, 25, 933 – 947.

Eck, J.E. (1993). The threat of crime displacement. Criminal Justice Abstracts, 25, 527 – 546.

Felson, M. (2002). Crime and Everyday Life, Third Edition. Thousand Oaks, CA: Sage Publications.

Felson, M., & Clarke, R.V. (1998). Opportunity Makes the Thief: Practical Theory for Crime Prevention. London, UK: Home Office.

Hesseling, R.B.P. (1994). Displacement: a review of the empirical literature. Crime Prevention Studies, 3, 197 – 230.

Lab, S.P. (2007). Crime Prevention: Approaches, Practices and Evaluations, Sixth Edition. Cincinnati, OH: Anderson Publishing.

Reppetto, T. (1976). Crime prevention and the displacement phenomenon. Crime and Delinquency, 22, 166 – 177.

Weisburd, D., & Green, L. (1995). Measuring immediate spatial displacement: methodological issues and problems. Crime Prevention Studies, 4, 349 – 361.

Weisburd, D., Wyckoff, L.A., Ready, J., Eck, J.E., Hinkle, J.C., & Gajewski, F. (2006). Does crime just move around the corner? A controlled study of spatial displacement and diffusion of crime control benefits. Criminology, 44, 549 – 591.