Field Notes

question

Does online displacement behave like crime displacement?

StatusWorkingUpdatedStands on2 sourcesSupportsnotes

Displacement carries an empirical record from situational crime prevention: relocation appears in roughly a quarter of the observations that test for it, the opposite spillover appears about as often, and where relocation occurs it is usually smaller than the intervention’s own effect. Those numbers were measured on physical places, offenders, and police work. The vault applies the concept to online restrictions on lawful conduct, where the actor is an ordinary user and the alternative is one tap away.

The question is whether the base rates transfer, and it matters because the vault currently borrows the concept without borrowing a warrant for the numbers.

Why the answer is not obvious either way

The mechanism should transfer. Both settings involve a controlled point, an uncontrolled alternative, and an actor weighing effort against reward.

The magnitudes may not. Search cost is the parameter the criminological finding depends on: displacement stays partial because a comparable opportunity nearby is often absent, riskier, or more effort. Online, a substitute is discoverable in seconds and costs nothing, which is an argument for higher displacement than the crime literature found. Pulling the other way, the online population is not a population of motivated offenders, so much of the restricted behaviour may simply not be pursued elsewhere.

What is already known

Children’s circumvention behaviours online is the closest measurement. Among UK 11- to 17-year-olds, 53 percent had chosen platforms that imposed no check or an easily tricked one, against 10 percent using location-changing tools. That establishes the form of displacement online — destination change rather than defeating the gate in place — but not its completeness, because the study did not measure whether the substituted platform delivered comparable exposure.

What would settle it

  • A pre- and post-enforcement measurement that observes a defined catchment of substitute services, chosen in advance, rather than only the gated platforms.
  • An estimate of the diffusion side: whether a gate on major platforms reduces exposure on services it never covered.
  • Harm-weighted rather than volume-weighted comparison, so that benign and malign displacement are distinguished instead of netted.

Until then, treat the crime-prevention base rates as a caution against asserting displacement without measuring it, not as an estimate of how large online displacement is. Do under-16 social-media bans reduce measured harm and Does age verification reduce harms other measures cannot would each be answerable in part by the same measurements.

Built on 2 sources (1 archived here, 1 external).

Working out connections…