"Mixed findings: Birmingham 14 CCTV cameras in the city centre in the early 1990s failed to reduce overall crime levels. However, recorded crime statistics indicate that CCTV reduced robbery and theft from persons, whilst incidences of theft from vehicles rose. The CCTV system was not designed to target vehicle related crime and so was not installed in many car parks. There was some displacement of crime to surrounding areas. Surveys of the public before and after CCTV installation found little change in general feelings of safety for those using the city centre during the day. Nevertheless, for those using the city centre after dark there was an increase in feelings of safety amongst those aware of the CCTV cameras. It is not clear whether these effects were a direct result of CCTV as the area was also redeveloped at the same time."
pages 40 to 57 of that second reference contain a detailed discussion of some statistics from 20 years ago in Birmingham city centre, a location I know well. Once sentence on page 50 seems to suggest displacement of street crime to car theft, only as a suggestion. However the whole section gives you an idea of how hard it is to be definite about displacement or reduction in a city centre subject to shifting building patterns and traffic routing.
Thanks. I'd imagine that there would be a noticeable shift in reported crime from areas of higher CCTV concentration to those [local] areas with lower CCTV concentration. It seems that there were no particularly clear displacement effects in the case-studies in that first doc¹.
The second doc² appears to make no clearer claims on displacement either.
Mind you we've 17 years more data now and there must be several areas where CCTV has been installed, removed (or deactivated) and possibly reinstalled.
There really must be better citations than this around. I've no time to dig for them now but these don't convince me that displacement is, if anything, more than a minor effect. Moreover those docs show a shift from direct personal robbery/theft to theft from vehicles; I'd expect that would result in less bodily harm overall at the expense of broken windows and bent door frames, probably a good result.
I've heard that argued for many years but never seen the stats. Do you have a reference for us?