There are many discussions and has being argued for long time that crime moves in response to targeted law enforcement (Repetto, 1974; Hakim & Rengert, 1981; Barr & Pease, 1990), however Eck (1993) and Hesseling (1994) argues that although the empirical evidence is often less convincing than the rhetoric. There are arguments that due to the negative effects of the enforcement strategies that has actually resulted in a diffusion of crime rather than a displacement of crime; it is believed that the crime reduction may due to the spill over due to the police operation into areas not directly targeted by the law enforcement action (Clarke & Weisburd, 1994; Green, 1995; Ratcliffe, 2002). Until recently there has been no systematic approach in acquiring empirical support for either side of the argument, although there has been a need for a methodology that is solid enough to clarify any displacement or diffusion of benefits and being simple enough to be applicable across a range of situations and user accessibilities. The introduction of the weighted displacement quotient (WDQ) from Bowers and Johnson (2003) attempted to address many of these issues.
Normally crime incidents are neither uniformly distributed over space nor randomly distributed. Rather, they occur in some form of clusters or trends. Numerous researchers have identified various sociological, temporal, and behavioural characteristics of offenders and crime locations that inform the non-random nature of crime clustering as Chainey and Ratcliffe (2005) have discussed in their research. Comparing both rational choice perspective (Cornish & Clarke 1986, 1987; Clarke & Felson, 1993) and discussion of routine activities theory (Cohen & Felson, 1979; Felson, 1998) Brantingham and Brantingham (1993a) elaborated the idea that crimes occur as criminally motivated when individuals come into contact with suitable targets as they travel between places and through areas that are known to them and wherein they are comfortable (awareness spaces). Further, they stated that crimes occur within cities where criminal behaviour is highly patterned and frequently localized (Brantingham & Brantingham, 1993b). Brantingham & Brantingham (1993a) argued that certain areas will experience disproportionate amount of different types of crime due to the newly setup design on urban areas and also the people within the cluster therein. If this idea is combined with place-specific attempts by crime prevention practitioners to reduce crime then it is really concerned with the displacement of crime.
The central tenet of displacement concerns is that criminally motivated offenders, when prevented from committing crime at a particular location or in a particular way, will not be dissuaded from criminality but will simply be displaced to commit crime in some other manners. Repetto (1974), Hakim and Rengert (1981) identified five forms of displacement as the result of blocked criminal opportunities. The first three are temporal displacement where the time of the offense changes, spatial displacement where the location changes, and target displacement where the target changes. The other types of displacement include method displacement, where the method the offender uses to commit the offense changes, and finally type displacement, where the offender changes from one type of offense to another (Repetto, 1974; Hakim & Rengert, 1981). From the perspective of action-oriented researchers aiming for a crime prevention outcome, Barr and Pease (1990) took a more theoretical approach in considering displacement as both a frustrating side effect and a predictable effect of specific policies and a manipulative tool of crime control. Barr and Pease (1990) concluded that displacement or deflection of crime is simply a reorganization of the space and time in which a potential criminal comes in contact with a potential victim, and that displacement is valuable not only for the planning of short-term strategies, but because it illuminates the choices that have been made which lead to the present pattern of crime (Barr & Pease, 1990). Many crime displacement researches tend to test for spatial displacement (Hesseling, 1994) but there is rarely research non-spatial example as discussed by Yang (2008). Yang (2008) focused on economic risks and loss resulting from changes in Philippines customs procedures, identifying predominantly method displacement, where changes and increases in enforcement of customs procedures caused adaptations in methods of evasion of duty payments. Spatial displacement is often articulated by practitioners and criminology researchers as a negative outcome from a crime prevention operation.
Ratcliffe (2002) cited various examples of researchers who have encountered a negative response of practitioners to suggestions of displacement. The general concern expressed is that crime displacement eliminates the effectiveness of crime reduction operations by pushing crime to nearby areas or other crime types. This assumption is often predicted on the overly optimistic notion that without displacement no crime would have occurred at all, a notion that places greater faith in crime prevention strategies. As Clarke (1995) noted that the uncritical acceptance of crime displacement may also mean that increases in crime, which might have occurred anyway, have sometimes been wrongly attributed to displacement. The incorrect view that crime displacement is inevitable which do not result in any net reduction in crime is a negative outcome of a crime prevention operation. It has not been effectively countered by a research tradition that although blessed with evidence, lacks appropriate tools and a public voice. Where crime displacement does occur, it is usually the result of situations that forces criminals to access nearby opportunities not within the confines of the enforcement operation. When criminals are forced into secondary and, by implication, less effective target choices, this will theoretically result in less crime. This idea was echoed by Eck (1993), who in his summary of thirty-three studies of crime displacement found that crime displacement is least likely to occur in the direction of unfamiliar places, times, targets and behaviours; what he termed as familiarity decay (Eck, 1993). Eck (1993) concluded that prevention and crackdown efforts focused on unique situations will have less displacement than prevention or crackdown efforts focused on general situations.
Empirical evidence supports a lack of total crime displacement. For example, both Hesseling’s (1994) meta-analysis and Ratcliffe’s (2005) empirical work in Canberra have concluded that crime displacement is not an inevitable outcome of a police crackdown on crime suppression operation. Green (1995), while examining the impact of a specialized multiagency response team on Cornish and Clarke (1987), conceptualized crime displacement from the perspective of rational choice theory. From their examination of the choices made by criminals in the decision to commit crimes and their choice of targets, they believed that crime displacement could best be explained and understood by concentrating on the criminal’s personal decisions and choices made in committing crime. Cornish and Clarke (1987) viewed the decisions, opportunities, costs, and benefits associated with particular offenses intention to establish the confines of crime displacement within different categories of offense. If this is the true case, then it is less possible to justify a generalized approach to crime displacement. This may requires security practitioners and researchers to be more open to the possibility that crime displacement might or might not occur with different spatio-temporal and other crime displacement typology due to the characteristics for different crime types and situations. This raises the specter that crime displacement areas might not be uniform, with the corollary that the spatial context of the wider environment surrounding a crime prevention operation must be considered.
Clarke and Weisburd (1994) discussed that the diffusion of benefits is the spread of the beneficial influence of an intervention beyond the places which are directly targeted, the individuals who are the target of control; whereas Ratcliffe and Makkai (2004) discussed that the crimes which are the focus of intervention and the time periods in which an intervention is brought (identified two types of diffusion of crime control benefits: deterrence and discouragement. Ratcliffe and Makkai (2004) also argued that deterrence is closely related to an increase in the potential criminal’s perceived risk of apprehension, and discouragement relates to an increase in the perceived effort required to complete the criminal act. Both mechanisms are capable of reducing crime in areas not specifically targeted by a crime reduction policy because the levers they pull relate to criminal perception rather than increases in actual risk or actual effort to commit crime.
Evidence for diffusion of benefits can be found in Weisburd et al. (2006) on social observations and ethnographic research that examined immediate crime displacement and diffusion effects of Jersey City interventions focused on prostitution, disorder, and drug market activity. They found a tendency among those they interviewed to resist moving and re-establishing their prostitution or drug activity in another location. Further, they noted statistically significant trends in target and catchment areas regarding the decline of disorder offenses, leading to the conclusion that the crime control efforts diffused into the areas surrounding the target sites (similar trends were seen with prostitution and drugs, although the trends in drug crime lacked statistical significance). Their results supported rational choice and routine activity as theoretical explanations, as well as giving credence to Eck’s (1993) concept of familiarity decay. However, with the exception of a few research papers, it can be concurred with the view of Hesseling from over a decade ago that most of the extant research lacks interest in, or intent to seek, evidence of a diffusion of benefits (Hesseling, 1994). This situation remains today, even with the explosion in recent years of the use of geographic information systems (GIS) and GIS science within policing and criminal justice research. This is very unfortunate situation as those actively seeking out evidence of this benefit will help the Police or law enforcement to justify their operational expenses and planning for future crime reduction strategies (Ratcliffe & Makkai, 2004). There is also the possibility of increasing the theoretical understanding of criminal response to targeted policing strategies beyond the practical policing and crime reduction benefit
Ratcliffe and Breen (2011) conducted research on the use of weighted displacement quotient (WDQ). The use of the WDQ has been shown to be of value beyond the initial application to burglary, as predicted by Bowers and Johnson (2003). Furthermore, with the inclusion of some relatively simple statistics; the research shows it is possible to enhance the WDQ and increase the statistical robustness, which helps determine the suitability of the chosen buffer area. The addition of the phi statistic helps to ascertain that the buffer and target areas are functioning relatively independent of each other. In this way security practitioners using the tool to examine police operations and crime prevention tactics are able to address better the possibility that any changes to the buffer area are not simply a corollary of changes in the target area but rather might be an independent reaction to the crime suppression operation. If there are noticeable changes to the buffer area, security practitioners are frequently interested to know whether the buffer area changes were a response to the treatment. This interest is usually confined to an assumption of complete crime displacement and rarely the more likely possibility of some diffusion of benefits. The phi statistic can inform this understanding by clarifying the relationship between target and crime displacement area; ruling out the possibility of spatial autocorrelation early in an analysis. If the phi value is low, it can be argued that the buffer area appears to be behaving largely independent of the target area. We can then use the WDQ to determine how much the treatment influenced the buffer area through crime displacement or diffusion and take this into account when assessing the value of the treatment.