I wrote an investigative piece about the actual effectiveness of CCTV for the May issue of Wired Magazine (published in April). I have reprinted it below or you can see it in its full glory on the Wired website.
Investigation: A sharp focus on CCTV
By Heather Brooke|01 April 2010
As the major political parties jostle for position in the run-up to the general election, it’s clear that the way the next government monitors and controls information about us will fundamentally shape British society in the next decades.
Both the Blair and Brown administrations have pursued policies of setting up giant, centralised databases, such as the national Automatic Number plate Recognition (ANPR) system, which tracks vehicles through an expanding network of cameras across Britain’s roads, and the massive communications register behind the Interception Modernisation Programme, intended to log all UK telephone and internet traffic.
The Conservatives, meanwhile, have pledged to scrap the National Identity Register — a database that underlies the proposed ID card, which will store 50 items of personal information about every citizen — along with the children’s database, ContactPoint. (This extensive datastore, which was introduced after the Victoria Climbié inquiry in 2005, records every child’s name, gender, date of birth, address and parental contacts, along with educational and health details.) The Tories have also announced a scaling-back of the DNA database, in order to remove individuals who have never been convicted of a crime. Yet for all this maneuvering, the major parties have been uncharacteristically quiet on the most controversial of all the invasions of UK citizens’ privacy — CCTV.
Although it’s widely supposed that over the past decade there has been a significant increase in the number of surveillance cameras in the UK, it wasn’t until last year that hard numbers emerged via a Freedom of Information request. Big Brother Watch (bigbrotherwatch.org.uk), an anti-surveillance campaign group, found that the number of council-owned cameras had risen from 21,000 to 60,000 in less than ten years — equal to one CCTV camera for every 1,000 people in the country. Its report demonstrated a trebling of investment in local CCTV — even though Home Office research published in 2002 suggested that CCTV has a negligible impact on reducing crime.
Nevertheless, as public perception equates CCTV to tackling crime and antisocial behaviour, most MPs are happy to show up to the unveiling of a new surveillance system in their constituencies. The UK has more CCTV cameras per capita than any European country, yet figures released in July 2009 by the European Commission and United Nations showed Britain’s recorded rate of violent crime surpassed any other country in Europe. Does CCTV do anything to make us safer? If so, at what cost?
CCTV is seen either as a symbol of Orwellian dystopia or a technology that will lead to crime-free streets and civil behaviour. While arguments continue, there is very little solid data in the public domain about the costs, quantity and effectiveness of surveillance. However, using the Freedom of Information Act (FOI), it is possible to use local examples to understand the wider trend. For example, Falkirk Council in Scotland spends the following on its set-up: £11,000 for the erection of a pole and electrical connections; unit costs of approximately £3,000 per camera; £1,500 for a fibre-optic connection; and £750 to connect each camera to a central monitoring facility. That’s an initial outlay of more than £16,000 per camera. To support each of the devices, a council must add the cost of the control room where the footage is monitored, staff, tapes or other storage facilities, monitoring and recording software, retro fitting and replacing hardware or software.
Peter Fry, the director of industry association the CCTV User Group, estimates the average cost of a camera to be £20,000, although FOI requests reveal great variation between councils. Moray Council in Scotland spends a frugal £10,000 per camera, for example, whereas the cost in Midlothian is £100,000. Edinburgh City Council budgets £25,000 per camera.
Edinburgh currently has 185 public cameras, at an estimated initial outlay of £4.6 million — not counting monitoring, retrofitting and replacement. But London is CCTV central: Wandsworth has the highest number, 1,113. By Fry’s estimates, Wandsworth has spent more than £22 million installing them — the equivalent of 1,100 police officers at the starting salary of £20,000 per year. And Fry says that maintenance of control rooms will set Wandsworth back between £350,000 and £400,000 per year. According to the CCTV User Group figures, the City of London (with 619 cameras) will have invested more than £12 million setting them up, and spends £2.25 million annually to maintain them.
FOI requests filed by the Liberal Democrats in 2007 revealed that London’s CCTV cameras had cost taxpayers around £200 million over the previous decade. That cost is borne, as least initially, by the Home Office, which over the past 20 years has spent the largest chunk of its crime prevention budget on CCTV. A House of Lords report published in January this year estimated that during the 1990s the Home Office spent 78 per cent of its crime prevention funds — estimated to be in excess of £500 million — on CCTV. After cameras are set up, costs of day-to-day management and retrofitting are met locally, either through the council or using partnerships with local police.
In 1991, only five local authorities possessed public-space TV. But in 1994, prime minister John Major announced the first government-led CCTV initiative, promising investment of £20 million in four phases. By 1996, 167 local authorities had installed surveillance equipment. In 1997 the new Labour government honoured Major’s pledge for the fourth round of funding, and a year later announced a £170 million grant for CCTV, followed by another £153 million distributed between 1999 and March 2002.
It was this central funding model that drove the proliferation of public cameras. “If you give out money for CCTV, it should come as no surprise that people begin to define their problems in terms of needing a CCTV solution,” says Martin Gill, director of Perpetuity Research & Consultancy International and professor of criminology at the University of Leicester, who wrote a Home Office research report into the effectiveness of CCTV in 2005. “If you know that’s where the money is coming from, that’s how you’ll define your problem.” Yet, despite government willingness to fund surveillance equipment, there is little data from either local or central government regarding its effectiveness.
What has become clear, though, is that the government’s allocation of resources can determine whether a council deploys surveillance cameras. CCTV implementation is likely to fail when Home Office money falls through.
According to Sarb Sembhi, a security consultant and London head of the Information Systems Audit and Control Association, the funding structure — whereby investment is given at the whim of central politicians — means that many cameras are now obsolete. Finance for the purchase of cameras arrives in huge chunks, but there is no commitment to follow-up money for maintenance or technological upgrades. “The life cycle of those early cameras was about ten years, so many of them are beyond their age of use,” Sembhi says. “They should have been decommissioned or replaced, but they haven’t.”
Since 1995 to date, the CCTV initiative for England and Wales has funded 684 schemes, varying in size from £30,000 to £7 million. And it’s not just urban centres that have embraced the technology: even a small rural council such as Scottish Borders has 58 cameras. According to the BBC, some of the UK’s smallest local authorities, such as the Shetland Islands Council (101 cameras) and Corby Borough Council (90 cameras), have more surveillance equipment than the San Francisco Police Department (71 cameras). The City of London’s 619 cameras work out as more than the combined total of the police departments of Boston, Johannesburg and Dublin.
Yet not even the Information Commissioner’s Office, an independent authority that is supposed to regulate CCTV use, has any exact numbers, partly because there are myriad organisations pointing cameras at us (councils, police forces, hospitals, numerous public bodies, and thousands of private companies). There are oft-cited estimates, guesstimates and urban myths; like all myths, these figures thrive on the lack of empirical data.
Two commonly quoted statistics are that the average Londoner is caught on camera 300 times daily, and that there are 4.2 million cameras in the UK. However, the methods used to estimate these numbers are completely unscientific. The source of the first of these statistics is Simon Davies, director of Privacy International, the surveillance watchdog.
Davies came up with the figure based on a walking tour of London back in the early 90s. He had visitors from America and they were curious about the burgeoning growth of CCTV in Britain. As there were no figures available they decided to conduct a survey by counting the number of cameras they saw as they walked from Blackfriars to Bond Street.
Generalising this guesstimate to the whole city of London is, of course, erroneous. And last year, in his column in The Times, writer David Aaronovitch demonstrated that this flawed figure has taken on a life of its own. The figure was picked up by academic Clive Norris, who conducted his own survey by walking along Putney High Street in London, and by 2002 came up with the estimate of 4.2 million cameras in the UK. A separate report in 2009 by consultancy IMS Research, using global sales of CCTV cameras, estimated there to be 3.2 million cameras in the UK.
In 2009, Peter Fry conducted another survey on behalf of the Home Office and the National Police Improvement Agency, by asking the owners of public-space systems (80 per cent of which are local authorities) for the number of cameras each of their control rooms monitored. He got an average figure of 100. He then asked for the number of control rooms and found a total of 14,050. By extrapolation, he came up with a figure of 1.4 million public-space CCTV cameras. “The 4.2 million figure is totally fallacious,” Fry says, although he admits that if private cameras were included “you might possibly get up to 4.2 million”. The regulation of surveillance is equally murky, data-wise. The law requires those who film the public to notify the Information Commissioner’s Office (ICO), the government body responsible for regulating the Data Protection Act and CCTV.
Although the ICO offers a searchable public register of companies that operate CCTV, it keeps no record of the actual number of cameras. Additionally, not all cameras have to be registered; for example, there is an exemption for any used for “domestic purposes”. In any case, the enforcement of rules governing the use or misuse of CCTV is largely moot: in the 12 years since the Data Protection Act has been in force, the ICO has not conducted a single investigation, nor will it reveal how many complaints it has received about the use of CCTV. “No complaints have resulted in formal investigations or enforcement action,” a spokeswoman for the ICO says. “We will need additional time to ascertain the number of CCTV-specific complaints.
Our current system is being reviewed to make this easier in future.” Further requests from Wired received no response. And despite the profusion of cameras, there is little data to suggest that CCTV reduces crime. “The government threw money at CCTV schemes before they studied what they actually did,” says criminologist Martin Innes, director of the Universities’ Police Science Institute. “They help in reducing auto crime, but do very little else. They don’t make people feel any safer. There’s no evidence to say that they do much at all.” Last year, detective chief inspector Mick Neville, head of the Met’s Visual Images, Identifications and Detections Office, revealed that only three per cent of the capital’s street robberies are solved using security camera footage. A Wired Freedom of Information request to the Crown Prosecution Service for the number of times CCTV footage is used in trials revealed that such information is not recorded — which means there’s no evidence that CCTV is helping to bring criminals to justice.
Academic criminologist Nic Groombridge was the author of one of the first Home Office reports into the effectiveness of CCTV in 1994. “Naturally, you couldn’t see whether CCTV worked or not,” Groombridge says. But that didn’t stop the government from rolling it out. “It works brilliantly as a memorial for people,” is Groombridge’s overall assessment of closed-circuit television. “Rubbish at crime prevention.” According to the two meta-analyses of CCTV conducted for the Home Office and published in 2002 and 2005, video surveillance has had only limited impact on crime prevention and detection.
The most frequently cited and comprehensive review of CCTV is the detailed Home Office study by Professor Martin Gill and others, published in 2005. Gill and his team evaluated 14 CCTV systems around Britain and concluded: “…Only two showed a statistically significant reduction relative to the control, and in one of these cases the change could be explained by the presence of confounding variables. Crime increased in seven areas but this could not be attributed to CCTV. The findings in these seven areas were inconclusive as a range of variables could account for the changes… including fluctuations in crime rates caused by seasonal, divisional and national trends and additional initiatives.”
The team also found that certain types of offence were affected by CCTV more than others: impulsive (e.g. alcohol-related) crimes were less likely to be reduced than premeditated crime (for example, theft of motor vehicles). Violence against the person rose and theft of motor vehicles fell in the target areas, in accordance with national trends in recorded crime. Despite this data, the government continues to invest money in CCTV. “The fact is, people wanted to be seduced,” says Groombridge. “This happens with crime prevention a lot. The councils, police, politicians — they all want to be able to say, ‘Look, we’re doing something.'” CCTV is no longer seen as the crime cure-all it was once perceived to be. Suppliers talk of “managing expectations” and urge public bodies to think strategically about the use of surveillance.
Finally it seems the Home Office is listening too; it is currently reviewing its practices, and changes to its funding criteria mean that grants will be given to surveillance only as part of projects that employ a range of other crime-prevention methods including, notably, old-fashioned human intelligence. “I find it amazing that the government and local authorities spend that kind of money promoting a system and yet wait 15 years to develop a strategy for its use,” Peter Fry says. “Normally you do the strategy first and the implementation afterwards.” Which is precisely where we were 50 years ago, when CCTV was invented — it’s only taken a few hundred million pounds of public spending to come full circle.