Platform Solutions Pricing
Sign in Get Started
Evidence Guide

Do Security Cameras Deter Crime?

Yes, modestly, and only under conditions most businesses do not meet. The research is clear about where cameras work, where they do nothing, and what separates the two. We are a video analytics company and we are going to tell you the unflattering half as well.

Last updated July 2026
Start Free Trial View Pricing
The Short Answer

Cameras Cut Crime About 13 Percent, and Only When Someone Is Watching

Security cameras do deter crime, but the effect is smaller and far more conditional than the security industry implies. The largest study on the question, a 40-year systematic review and meta-analysis of 80 evaluations published by Piza, Welsh, Farrington and Thomas in 2019, found that crime decreased by approximately 13 percent in areas with CCTV compared with control areas.

That average hides everything important. Cameras reduced crime by about 37 percent in car parks and about 12 percent in residential settings, but had no statistically significant effect in city and town centers, on public transport, or in public housing. They significantly reduced property and vehicle crime, and had no significant effect on violent crime at all.

The single biggest predictor of whether cameras worked was whether anyone was actually watching them. Actively monitored systems produced significant reductions. Passively monitored systems, meaning cameras that simply record, showed no significant effect in any analysis the authors ran. Their conclusion was blunt: this is evidence against using CCTV as a stand-alone tactic.

Where Cameras Work and Where They Do Not

All figures below come from Piza, Welsh, Farrington and Thomas (2019), the 40-year systematic review published in Criminology and Public Policy, which is also the basis for the US National Institute of Justice's CrimeSolutions rating of CCTV.

Setting or crime type Effect on crime Statistically significant?
Car parks and parking lots About 37% reduction Yes
Residential areas About 12% reduction Yes
All settings combined About 13% reduction Yes
City and town centers No measurable effect No
Public transport No measurable effect No
Public housing No measurable effect No
Property and vehicle crime About 14% reduction each Yes
Violent crime and disorder No measurable effect No
Actively monitored cameras Significant reduction Yes
Passively monitored cameras (recording only) No measurable effect No

Source: Piza, Welsh, Farrington & Thomas (2019), CCTV surveillance for crime prevention: A 40-year systematic review with meta-analysis, Criminology & Public Policy 18(1), 135 to 159.

Why Recording Alone Does Nothing

This is the finding that should reshape how a business thinks about its cameras. The meta-analysis split the studies by whether the CCTV was actively monitored, meaning a person or a system was watching the feeds and could respond, or passively monitored, meaning the cameras recorded and the footage was reviewed only afterward if at all. Active monitoring produced significant crime reductions. Passive monitoring showed nonsignificant effects across all three analyses the authors ran.

That result is not mysterious. Deterrence depends on an offender's perception that they will be caught. A camera that nobody watches does not raise the odds of getting caught in the moment; it only raises the odds of being identified later, and the research on clearance rates suggests that is a thin threat. The FBI reported that 15.9 percent of property crimes were cleared by arrest or exceptional means in 2024. Someone stealing from your stockroom is, statistically, unlikely to face consequences from footage alone.

Here is the uncomfortable part for a US buyer. The same meta-analysis found no significant CCTV effect in the United States across its 24 US evaluations, and the authors point at a plausible reason: US schemes used active monitoring in only 58.3 percent of cases against 88.2 percent in the UK, and almost never paired cameras with other interventions. American CCTV underperformed because of how Americans deploy it. Cameras went up, nobody watched them, and nothing changed.

We sell software that watches camera feeds, so read the next sentence with appropriate suspicion. The honest version is this: the variable the research identifies as decisive is attention, and attention is the thing almost no business can afford to buy in human form. Nobody is going to staff a monitor wall for a four-store chain. The reason software matters here is not that AI is magic; it is that continuous attention is the mechanism the evidence points to, and software is the only affordable way to get it. If you were to hire a person to watch the cameras instead, the research would predict a similar benefit, and you would spend far more.

What Burglars Say About Cameras

The most-cited offender study is a 2012 survey by the University of North Carolina at Charlotte's Department of Criminal Justice and Criminology, which interviewed 422 incarcerated burglars in North Carolina, Kentucky, and Ohio about how they chose targets.

Cameras and surveillance equipment ranked as the single most commonly considered factor, with roughly 57 percent of respondents saying they weighed it when picking a target, ahead of alarms, people being home, and dogs. Roughly 43 percent said the presence of cameras would cause them to avoid a target altogether. On alarms specifically, about 83 percent said they would try to determine whether an alarm was present before attempting a burglary, and about 60 percent said an alarm would push them to look for a different target.

Two caveats you deserve to know, because most articles quoting this study omit both. First, it was funded by the Alarm Industry Research and Educational Foundation, an industry body with an obvious interest in the answer. Second, and more importantly, it measures what offenders say they would do, not crime that actually fell. Stated intent and measured outcomes are different things, which is precisely why the meta-analysis above matters more. Take the survey as a plausible account of the mechanism, not as proof of the result.

Cameras Are Better at Solving Crime Than Preventing It

The best evidence on cameras as an investigative tool comes from Ashby (2017), which examined 251,195 crimes recorded by the British Transport Police between 2011 and 2015. CCTV was available to investigators in 45 percent of cases and judged useful in 29 percent, which is 65 percent of the cases where it existed at all.

Where it helped, it helped enormously. Robbery detection rates rose from 8.9 percent without useful CCTV to 55.7 percent with it. But the value was wildly uneven by crime type: CCTV was useful in 62.2 percent of robbery investigations and only 10.7 percent of drug investigations, with no significant benefit for drugs, fraud, or public order offenses. The main thing limiting its usefulness was simple, and it is the same in your building: large areas were not covered.

That study covers a UK rail network, not a US business, so treat the exact percentages as indicative. The pattern is what transfers. Cameras earn most of their keep after an incident, on the crimes where a visual record identifies a person, and they earn it only where the coverage actually existed. Which is a good argument for thinking carefully about placement before adding more cameras: coverage of the four loss points beats a higher camera count, something we lay out in detail for retail on our page for store security cameras.

Do Cameras Just Move Crime Somewhere Else?

Sometimes, but displacement is not the norm, and the opposite effect is about as common. The 2019 meta-analysis looked at studies that specifically tested whether crime moved to nearby areas. Among the 23 city and town center studies that examined it, 13 found neither displacement nor its opposite, 6 found a diffusion of benefits, meaning crime also fell in adjacent areas without cameras, 3 found displacement, and 1 found both. In the car park studies that tested it, results were similarly mixed.

For a business owner the takeaway is modest and practical: you are not likely to be simply exporting your crime to the store next door, and you may be helping them slightly. But you should not count on either.

One claim we deliberately do not repeat is the widely circulated figure that London's cameras solve only one crime per thousand cameras. It traces back to press coverage of an internal police briefing and we could not verify it in any primary source. It gets quoted constantly by camera skeptics and it should not be.

How to Make Cameras Actually Deter Crime at Your Business

The research does not say cameras are useless. It says cameras are a condition, not a cause. Four things separate the deployments that reduced crime from the ones that did nothing.

  1. Make sure something is actually watching. This is the decisive variable in the meta-analysis. A recorder in a back office is passive monitoring, and passive monitoring showed no significant effect. Either someone watches the feeds, or software does, and something escalates when it sees a problem.
  2. Respond visibly. Deterrence works through perceived risk of being caught. When staff and customers see that footage gets acted on, the perception changes. When incidents are logged and forgotten, it does not.
  3. Put the cameras where the evidence says they work. Parking lots showed the strongest effect of any setting, by a wide margin, and property and vehicle crime responded where violent crime did not. If your lot is uncovered, that is likely your highest-yield camera.
  4. Pair cameras with something else. The US evaluations underperformed partly because CCTV was deployed alone. Lighting, access control, staffing at the right hours, and signage all appear alongside cameras in the schemes that worked.

Do not expect cameras to solve a violence problem. The evidence is consistent that they do not, and a vendor telling you otherwise is selling past the research. Expect them to reduce property and vehicle crime meaningfully, to help enormously with investigations where coverage exists, and to do essentially nothing if they sit unwatched.

If you want the counterpart to this page on the numbers side, our sourced breakdown of retail theft statistics covers what US shrink actually consists of, and why a third of it is not theft at all. For sites that genuinely need a human to respond overnight rather than a report in the morning, we compare the options honestly in video monitoring service versus software.

Common Questions

Do security cameras deter crime?

Yes, but modestly and conditionally. The 40-year meta-analysis by Piza and colleagues (2019) found crime fell about 13 percent in areas with CCTV versus control areas. The effect was strongest in car parks at about 37 percent, absent in city centers and on public transport, and absent for violent crime. Cameras that were not actively monitored showed no significant effect at all.

Do cameras deter burglars?

Burglars say they do. In a 2012 University of North Carolina at Charlotte survey of 422 incarcerated burglars, cameras and surveillance ranked as the most commonly considered factor when choosing a target, weighed by roughly 57 percent, and about 43 percent said cameras would make them pick a different target. Note the study was industry-funded and measures stated intent, not measured crime reduction.

Do fake security cameras work?

There is no credible research showing dummy cameras reduce crime, and the mechanism the evidence supports argues against them. Deterrence in the studies tracks with active monitoring and real response, not with the visible presence of a camera housing. A fake camera also produces no evidence, no alert, and creates liability if anyone relied on it. We would not recommend one.

Do security cameras reduce violent crime?

The evidence says no. The 2019 meta-analysis found no significant effects for violent crime or disorder, and the National Institute of Justice's CrimeSolutions rates CCTV as Ineffective for violent crime while rating it Promising for property and vehicle crime. Cameras help investigate violent incidents after the fact, but they do not appear to prevent them.

Does crime just move somewhere else?

Usually not. Among the city center studies in the 2019 review that tested for displacement, most found neither displacement nor diffusion, and more found a diffusion of benefits, meaning crime also fell nearby, than found displacement. Displacement happens sometimes, but it is not the typical result and it does not cancel out the benefit.

Are cameras useful for catching criminals after the fact?

Often, where coverage exists. Ashby's 2017 study of 251,195 British Transport Police crimes found CCTV was available in 45 percent of cases and useful in 29 percent. Robbery detection rose from 8.9 percent without useful footage to 55.7 percent with it. But it added little for fraud, drugs, or public order, and its main limitation was areas with no camera coverage.

Sources

Figures verified against these sources in July 2026. Where a widely-quoted statistic could not be verified in a primary source, we left it out and said so.

Turn passive cameras into watched cameras

Active monitoring is the variable the research points to. Surveillant provides it in software, on the cameras you already own, without a monitor wall or a guard.

Start Free Trial View Pricing