Geographical displacement occurs when the intervention blocks crime or disorder opportunities at a facility or in an area, and offenders move to other facilities or areas to offend. Temporal displacement also stems from successful prevention, but in this case offenders shift offending in time to other hours or days.
Offenders can find it difficult to move to some other location because easy crime or disorder opportunities are limited (Step 16). Targets may be concentrated at some places and not others (Step 16). Vulnerable potential victims can be found at some locations, but not others (Step 29). Some facilities have low behavioral controls, but others do not (Step 28). Opportunities that exist are either already hot spots or are hidden from offenders - either far away or not recognizable as fruitful places to offend.
Offenders will not usually spend time searching far from their hot spot when it is suppressed. So, if offenders move, they are most likely to move to a place close to the original hot spot. The likelihood that offenders will move to an opportunity declines the further the opportunity is from the original hot spot, as illustrated in the figure. Also, not all spaces are suitable to offenders. Opportunities are not spread evenly across the map.
In this map, the diamonds are places with characteristics like the original hot spot. Those closest to the original location are most likely to be affected most by displacement. In addition to distance, natural barriers to movement can limit displacement. In the figure, the river flowing northeast/southwest reduces the chances of displacement to the east. Knowing this, displacement countermeasures can be applied with the response at the most vulnerable locations.
If geographical or temporal displacement occurs, it is most likely to shift crime to locations and times very similar to the locations and times affected by the prevention. Such shifts require less effort, learning, and risk for offenders than shifting to very different places and times. It is more likely that offenders will try to outwait the response, which explains Lawrence Sherman's finding that the effects of crackdowns decay. If offenders cannot outwait a response, it will be the most familiar locations and times that will have the greatest chance of receiving displaced crime. As Paul and Patricia Brantingham note, it is possible to predict the most likely areas for displacement. But this requires detailed knowledge of the crime opportunities in the current situation.
If geographical displacement occurs, it can distort conclusions about effectiveness. Table 1 illustrates how this can happen. In this example there are three similar areas with equal numbers of crimes before treatment: (1) a treatment area; (2) an area adjacent to the treatment area; and (3) an area distant from the treatment area. The treated area has a decline of 25 crimes. However, the adjacent area has a 10-crime increase. This seems to suggest that if nothing had been done in the treatment area it too would have experienced a 10-crime increase. So the net reduction is 35 (the 25 crimes reduced in the treatment area and the 10-crime increase that was averted).
Decline of Geographical Displacement with Distance from Hot Spot Epicenter
Table 1: Use of Adjacent and Distant Control Areas in Controlling for Geographical Displacement
|Before||After||Difference||Estimated Net effect|
But these extra 10 crimes could have been due to geographical displacement. One would be better off using the distant control area for comparison. As a control, the distant area suggests that if no treatment were implemented, crime would not have changed in either the treatment or the adjacent areas. The implication is that the treatment caused a 25-crime decline in the treatment area, but a 10-crime increase in the adjacent area (displacement), for a combined reduction of 15 crimes. Though effective, the program is not as effective as originally estimated. Step 51 describes formulas to take account of displacement when assessing effectiveness.
Try to select two comparison areas as part of evaluations: one near the treatment area that has similar crime opportunities to detect geographical displacement (and diffusion, Step 51), and the other to serve as a control area. The control area should be protected from displacement contamination by distance or some other barrier (e.g., a highway or river). Valid selection of control and displacement areas requires you to have some idea of offenders' normal movement patterns, as the control area needs to be outside their roaming territory while the displacement area should be within it.
Temporal displacement may be easier for offenders than geographical displacement because it requires less effort. Temporal displacement can occur within a 24-hour day, if, for example, the prevention is restricted to certain times but leaves other times unprotected. It can also occur over a week. Or it can occur over longer periods.
If the evaluation compares times with a prevention response to times without a prevention response, contamination of temporal controls can take place. In Table 2, a treatment takes place on Saturday and Sunday. The average number of crimes on these days dropped by 25 crimes after treatment, while crimes on Mondays and Fridays increased by 10. Was this due to temporal displacement? Midweek days may be more valid controls because they have less in common with weekends than do Mondays and Fridays.
Waiting out the prevention is a common form of temporal displacement. Enforcement crackdowns are particularly vulnerable to this form of time shifting because they are temporary by definition. If an intervention can be maintained (unlike a crackdown), then offenders cannot wait it out. They then face the difficult option of moving to less attractive places or targets or undertaking new tactics or other crimes. If these options are too difficult, unrewarding, risky, or otherwise unattractive they may commit fewer offenses.
Table 2: Using Days of the Week to Control for Temporal Displacement
|Days of week||Before||After||Difference||Estimated Net effect|
|Treatment||Sat & Sun||100||75||-25|
|Adjacent Area||Mon & Fri||100||110||+10||-35|
Footnote text here
- Brantingham, Paul and Patricia Brantingham. (2003). "Anticipating the Displacement of Crime Using the Principles of Environmental Criminology." Crime Prevention Studies, volume 16. Monsey, NY: Criminal Justice Press.
- Eck, John (2002). Assessing Responses to Problems: An Introductory Guide for Police Problem-Solvers. Washington, DC: U.S. Department of Justice, Office of Community Oriented Policing Services. (Accessible at: www.popcenter.org and www.cops.usdoj.gov ).