The Unintended Consequences of Crime Prevention Measures

Crime displacement: Criminal behaviour that is observed elsewhere or at different times because of that intervention.

Diffusion of benefit: A reduction in crime among nearby locations and times that were not targeted by the intervention.

Our research identified and collated the existing evidence base for both displacement and diffusion of benefits following a crime deterrence intervention. We used a systematic approach to identify the relevant literature. The review considered peer reviewed studies that were published in English up to March 2021. The studies included an intervention designed to deter crime and at least one measure of deterrence.  

Displacement 

We found 69 studies that attempted to measure whether displacement occurred after the introduction of one or more interventions designed to deter offenders from committing crime. Of those, 38 studies found indications of some form of displacement.

There are 6 types of displacement; (a) spatial: a change in location, (b) temporal: change of activity by the time of occurrence, (c) target: crime against a different target, (d) tactical: the perpetrator adopts a separate modus operandi, (e) functional: a change in crime type, and (f) perpetrator: when opportunities for a new type of offender occur.

Spatial displacement was observed in 28 studies, half of which reported displacement effects through surveillance schemes, mainly formal police-based patrol interventions where perpetrators conducted criminal behaviour near the original target location. Closed-circuit television (CCTV) placement, the introduction of place-based lighting improvements, and target hardening efforts like gates and locks, all also reported displacement effects. Spatial displacement was also reported when interventions used a mixed measures approach to crime reduction. 

The remaining displacement types were observed much less frequently. Temporal displacement was typically observed when perpetrators perceived surveillance and formal police-based foot patrol schemes as temporary and not a continual increase in security. 

A diffusion of benefits is more likely to occur than displacement. 

Only one study produced results indicative of target displacement after a residential property-marking scheme was introduced. Whilst the rate of burglary against residential properties decreased, there was an increase in commercial burglaries. 

Tactical displacement was associated with target hardening in two studies that introduced burglary preventative measures like door-specific security upgrades. A study observing an increase in maritime guardianship/place management efforts also observed a change in piracy tactics. 

Two studies found an associated functional displacement effect; one of which occurred after the introduction of burglary security devices. Crimes such as vehicle theft, theft from a person, robbery, and bicycle theft all increased. 

Lastly, perpetrator displacement was identified in one study. After the introduction of various SCP (Situational Crime Prevention) and CPTED (Crime Prevention Through Environmental Design) measures at Rotterdam port, a shift in the perpetrator was seen. Port employees now became involved in cocaine trafficking. Employees typically either worked for port companies or government agencies and facilitated organised crime by bypassing security checks or providing information and/or access cards. 

Diffusion of Benefits 

We found 33 studies that attempted to measure whether diffusion of benefits occurred after the introduction of one or more interventions designed to deter offenders from committing crime. 30 studies had findings indicative of a diffusion of benefits. 

Formal surveillance-based interventions produced positive effects in three diffusion subtypes. Spatial diffusion of benefits was the most reported outcome across violent crime, theft, and burglary when introduced with police-based patrol schemes and increased guardianship/place manager interventions. Temporal diffusion was the next most reported benefit, mostly when introduced with marked vehicle and bicycle patrols.

Lastly, outreach efforts saw a perpetrator diffusion effect when measuring gang-involved shootings in gangs not directly targeted by the intervention.

CCTV specific interventions contributed mostly to spatial diffusion of benefits across assault, robbery, burglary, and vehicle theft crime types. One study suggests that CCTV is dominant in providing a diffusion of benefits in more serious crimes compared to disorder types. 

Target hardening schemes, such as implementing physical barriers, increasing access security measures, increasing the presence of uniformed officers at a location, and CPTED all had a spatial diffusion of benefits. 

Practitioner implications

Despite the limited evidence base, we were able to extract useful insight for practitioners:

1. A diffusion of benefits is more likely to occur than displacement. 90.9% of studies that measured its presence found indicators of a diffusion of benefit. The corresponding figure for displacement studies was 55%. 
2. A displacement effect is not indicative of a failed intervention. The focus should be on the level of harm produced by the intervention versus the level of harm prevented, which was underreported in the literature. For example, the volume of crime displaced may be lower than what was prevented in the treatment location, or the severity of the crimes could be less. 
3. We know far less about temporal, target, tactical, functional, and perpetrator displacements compared to spatial displacement. This is an important gap to fill when studying crime types that are significantly motivated by target selection, tactics, and perpetrator specific recruitment strategies, like terrorism.

An important limitation of the existing evidence base is that not all studies in this topic are robust. Many employ quasi-experimental designs making it difficult to pinpoint whether the interventions were also responsible for the displacement/diffusion of benefit.