Jump to ContentJump to Main Navigation
Making Public Places SaferSurveillance and Crime Prevention$

Brandon C. Welsh and David P. Farrington

Print publication date: 2009

Print ISBN-13: 9780195326215

Published to Oxford Scholarship Online: May 2012

DOI: 10.1093/acprof:oso/9780195326215.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: null; date: 17 September 2019

Safer Streets, Safer Cities: Policy Choices for America

Safer Streets, Safer Cities: Policy Choices for America

(p.111) 8 Safer Streets, Safer Cities: Policy Choices for America
Making Public Places Safer

Brandon C. Welsh

David P. Farrington

Oxford University Press

Abstract and Keywords

This chapter explores key policy choices and challenges that U.S. cities face in the use of major forms of public-surveillance approaches such as closed-circuit television (CCTV) surveillance cameras and improved street lighting to prevent crime in public places. It suggests that not all forms of surveillance are as potentially intrusive, and raises questions about the infringement of civil liberties and other social costs. However, these social costs need to be weighed against any crime-prevention benefits that accrue from the different forms of surveillance.

Keywords:   public surveillance, U.S. cities, CCTV surveillance cameras, street lighting, crime prevention, public places, civil liberties, social costs

On the basis of the highest quality research evidence available of the effects on crime of the five major forms of public area surveillance, a few general conclusions can be drawn. First, closed-circuit television (CCTV), improved street lighting, and the defensible space practice of street closures or barricades seem to be effective in preventing crime. Second, security guards are promising in preventing crime. Third, place managers appear to be of unknown effectiveness. For city managers, police chiefs, urban planners, business owners, or others, this may be useful information if a decision needs to be made about implementing one or the other measure. (Hopefully, as Mark Moore [2002] argued in another context, a cost-effectiveness analysis or comparative cost-benefit analysis would also be carried out to inform this decision.1) For some, this information may be far too limited.

What may prove more helpful to policy makers and practitioners is information about the specific conditions under which these surveillance measures are most effective in preventing crime. The present research shows that CCTV is effective in preventing crime in car parks; improved street lighting is effective in city and town centers and residential/public housing communities; and the defensible space practice of street closures or barricades is effective in inner-city neighborhoods. (Just as important (p.112) are those conditions under which surveillance measures are not effective. We discuss this shortly.)

Also of importance is evidence that shows that CCTV and improved street lighting are more effective in reducing property (especially vehicle) crimes than in reducing violent crimes. Street closure or barricade schemes are effective in reducing both property and violent crimes. The weight of the evidence suggests that security guards are promising when implemented in car parks and targeted at vehicle crimes.

Striking a Balance between Crime Reduction and Social Costs

Andrew von Hirsch (2000) argues that two major issues confront the “proper uses and limits” of surveillance for crime prevention in public places. His discussion takes place in the context of CCTV, but it is relevant to the other forms of public area surveillance that we cover in this book. The first issue pertains to privacy concerns. In chapter 2, we expanded on privacy concerns to include other social costs that may infringe on public interests or violate legal or constitutional protections; we return to these points here. The second issue concerns the matter of the “legitimising role of crime prevention,” or as von Hirsch (2000, p. 61) posits, “To what extent does crime prevention legitimise impinging on any interests of privacy or anonymity in public space?” We investigate this matter here.

Our analysis begins with CCTV. In car parks, the one setting in which CCTV is effective in preventing crime (specifically vehicle crime), there may be little resistance to the installation of CCTV cameras. In part, this is because this public space is used for one rather inconsequential purpose— parking vehicles. It is also the case that a car park is a well-defined and clearly marked physical space, meaning that individuals know that it is a car park and can choose whether to park their vehicle there (providing there are other alternatives). These points stand in sharp contrast with how individuals come into contact with CCTV in other public settings (see following discussion).

One could take issue with CCTV cameras in public car parks. For the period of time that the vehicle is parked, including leaving and returning to the vehicle and exiting the car park, the individual is being monitored in some capacity. To some or possibly many this is an invasion of privacy; (p.113) to others, this is a slight inconvenience for the added safety that is afforded their vehicle by the presence of the cameras. It is also noteworthy that most of the studies (four out of six) did not measure either displacement of crime or diffusion of benefits. On balance, the crime reduction benefits of the use of CCTV in car parks seem to outweigh any social costs.

CCTV in other public settings, such as city and town centers, public housing communities, and transportation facilities, evokes more resistance on the basis of threats to privacy and other civil liberties and are associated with a larger number of social harms, including the reinforcement of the notion of a fortress society and the social exclusion of marginalized populations (Clarke, 2000). Indeed, it is often these settings that are at the center of the debate over how best to strike a balance between the potential crime reduction benefits and social costs associated with CCTV (Savage, 2007).

Take city and town centers, for example. Here, the coverage of the cameras is much less than in car parks, but there is often a greater number of cameras extending over a larger area. For instance, a person walking down a street in an urban center could be caught on camera 20, 30, or 40 times, depending on the concentration of cameras. In London and other major cities in the United Kingdom, this concentration as well as its widespread use in other public (and private) places figured into the estimate that the average Briton is caught on camera roughly 300 times a day (Associated Press, 2007). CCTV in this setting may also result in the social exclusion of vulnerable or marginalized populations, such as unemployed youths who are hanging out and the homeless. The fear is that instead of providing assistance for these groups to get off the street (so to speak), CCTV, among other interventions like police and security guard patrols, may push them further away from available services and cause increased harm in the form of crime, victimization, or both. On the matter of crime displacement, it was most common to find evidence of no displacement.

Our finding that CCTV is associated with a nonsignificant and rather small 7% reduction in crimes in city and town centers does not detract from the need to assess its potential social harms. This is because city and town centers remain the most popular public setting for the implementation of this form of surveillance (Savage, 2007). So the main questions that confront policy makers over the use of CCTV in this public setting seem to be: How can its effectiveness be improved? How can CCTV be less intrusive? Both questions are important but difficult to answer.

(p.114) From our meta-analysis of evaluations of CCTV in city and town centers, there is no clear indication about what may work best in this setting. However, lessons can be drawn from the effectiveness of CCTV in car parks. All six car park schemes included other interventions (e.g., improved lighting, security guards), were mostly limited to a reduction in vehicle crimes (the only crime type measured in five of the six schemes), and camera coverage was high for those evaluations that reported on it. In contrast, the evaluations of city and town center schemes measured a much larger range of crime types and only a small number of studies involved other interventions. (Too few studies reported on camera coverage.) This was also the case in public housing communities, in which we also found that CCTV was associated with a nonsignificant 7% reduction in crimes.

These findings point to the need for CCTV in city and town centers to be targeted on property crimes, targeted at specific places such as high-crime areas (as part of an effort to increase camera coverage), and combined with other surveillance measures. Regular crime analysis by the police, such as that used in CompStat, could be used to identify those places that are at greatest risk for property crimes, which, in turn, could be used to guide the implementation of video surveillance. The advent of mobile and redeployable CCTV cameras may make this a more feasible and perhaps less costly option (Waples and Gill, 2006). (In chapter 9, we discuss these and other new and emerging surveillance technologies.) This more targeted approach could also go some way toward reducing the pervasiveness of the threat to the general public’s privacy and other civil liberties. Efforts to reduce any social exclusion effect associated with CCTV in this setting would need to involve other services.

Incidentally, we did not find any evidence that supports an association between the effectiveness of CCTV and those public places in which there is less resistance to its use, for example, in car parks. In chapter 5, we discussed the possibility that a contributing factor to the difference in effectiveness between the British CCTV schemes and those in other countries may have something to do with the public in other countries being less accepting and more apprehensive of CCTV. However, this view may be changing in the United States.

In public housing communities, the stakes are even higher in an effort to balance crime reduction and social costs. This is because people live there. What we do not know is whether residents of public housing (p.115) communities are more or less supportive of CCTV than those who come in contact with it in other public settings. One view is that residents, who often do not have a say in its implementation, are distrustful of how CCTV will be used by the police or security company, not to mention the threat it presents to privacy. Another view is that residents are grateful for the added measure of security.

Similar to city and town centers, CCTV in public housing may be more effective and less intrusive if it is implemented in specific places where crime is more likely to take place. These hot spots could include the car park, playground, or park. As well, CCTV may be more effective if implemented in high-use areas, such as the main lobby, other points of entry or exit, elevators, and stairwells. Importantly, of the five schemes that measured displacement or diffusion, all of them reported that displacement did not occur.

The effects of CCTV on crime in public transportation facilities, specifically, underground railway systems or subways, raise other considerations. In our meta-analysis of the four evaluations of CCTV in this public setting, we found that it led to a sizable (23%) but nonsignificant reduction in crime. A substantial reduction in robberies and thefts in the first London Underground evaluation (an overall 61% decrease), conducted by John Burrows (1980), was the main reason for this large average effect size over all four studies. Only two studies measured displacement or diffusion, with one showing evidence of diffusion and the other showing evidence of displacement.

It is not altogether clear how CCTV could be more effective in this setting. Camera coverage was not reported in any of the evaluations; three of the four included other situational prevention measures (e.g., CCTV notices, passenger alarms) or in one case improved lighting; as already noted, crime reduction benefits accrued from a range of crimes, meaning that targeting property crimes alone may not be an effective strategy in this setting. On the matter of CCTV being less intrusive in subways, its use could be limited to those stations or lines that have the highest crime rates.

Like CCTV, the use of security guards in public places has also been criticized on the grounds that it may result in the social exclusion of vulnerable populations, such as unemployed youths and the homeless (Wakefield, 2003). The potential threat to privacy or other civil liberties associated with the presence of security guards may be somewhat less of a (p.116) concern to the general public. Although no research has investigated this question, one of the possible reasons for this could be that unlike CCTV, this technique of formal surveillance is not as invasive; it does not monitor (and record) every move one makes.

Our finding that security guards represent a promising public area surveillance technique when implemented in car parks and targeted at vehicle crimes has some limitations in an analysis of crime reduction and social costs. This is because, as discussed in chapter 7, the promising nature of security guards means that we are not recommending their wider use but rather calling for additional evaluation research to test their effects under different conditions. Policy makers and practitioners should move cautiously. It is interesting that both techniques of formal surveillance— CCTV and security guards—share some of the same features with respect to the conditions under which they are deemed effective or promising: implemented in car parks, targeted at vehicle crimes, and combined with other measures. On this basis, too, further experimentation of the effectiveness of security guards seems warranted.

Improved street lighting—by far the most effective surveillance technique in reducing crime in public space—has some advantages over other surveillance measures that have been associated with the creeping privatization of public space, the exclusion of sections of the population, and the move toward a “fortress” society (Bottoms, 1990). Street lighting benefits the whole neighborhood rather than particular individuals or households. It has no adverse civil liberties implications, and it can increase public safety and effective use of neighborhood streets at night. It is also the case that improved street lighting does not displace crime. Only 3 of the 13 studies reported some or possible evidence that lighting improvements caused crime to be displaced to the surrounding areas. Nine other studies reported no evidence of territorial displacement, with two of these also reporting at least some evidence of diffusion of crime prevention benefits to nearby areas that did not receive the intervention.

The one potential social harm associated with street lighting is that it may contribute to light pollution (Pease, 1999). We do not deny that improved street lighting may cause an increase in light pollution, and this potential social cost should be considered in decisions to implement lighting schemes. However, for many, especially those who live in urban areas, this may be a small price to pay for the crime reduction benefits that have been shown to accrue from lighting improvements.

(p.117) Like improved street lighting, the natural surveillance technique of defensible space may have few (if any) social costs. Defensible space involves design changes to the built environment to maximize the natural surveillance of open spaces afforded by people going about their day-to-day activities. It does not appear to violate personal privacy, infringe on civil liberties, or contribute to the social exclusion of groups. Conceivably some of the more structural design changes (e.g., construction of street barricades or closures, redesign of walkways, installation of windows) could evoke more resistance than some of the more mundane changes (e.g., removal or pruning of bushes in front of homes, removal of objects from store shelves or windows).

Our systematic review of defensible space found that the practice of street closures or barricades is effective in reducing crime in inner-city neighborhoods. Unfortunately, only two of the four studies reported on displacement or diffusion. In one case, there was some evidence of both displacement and diffusion, and in the other displacement did not occur. On balance, the crime reduction benefits of street barricades or closures in inner-city neighborhoods appear to outweigh any potential social costs.

Finally, we consider place managers. Our conclusion is that this surveillance technique is of unknown effectiveness in preventing crime in public places. Even so, it seems worthwhile to sketch out a rough crime reduction–social cost analysis.

Some may express concern with parking lot attendants, bus drivers, concierges, and other place managers as yet another set of eyes watching potential offenders and law-abiding citizens alike. Unlike security guards or CCTV, surveillance is a secondary function of place managers. This secondary function comes about from their presence and ability to intervene, thus presenting a deterrent threat to potential offenders. In this regard, the use of this surveillance technique may strike a fairly good balance between the public’s interest in community safety and concerns over the erosion of privacy and civil liberties. More research is needed to investigate the effects of place managers on crime in public places.

Implications for Other Western Countries

In most of our systematic reviews of the different forms of public area surveillance, studies were included from a range of Western countries. (p.118) This often allowed for an analysis of differential effects between countries, notably the United States and Great Britain. For example, in the improved street lighting review, the included studies were carried out in one of these two countries. Results of the meta-analysis showed that street lighting improvements were much more effective in reducing crime in the United Kingdom than in the United States. In the CCTV review, a larger number of countries were represented in the included studies (Canada, Norway, Sweden, United Kingdom, and the United States). The best country comparison was between the United Kingdom and other countries. Results of the meta-analysis showed that CCTV was much more effective in reducing crime in the United Kingdom than in other countries.

So far our discussion has focused largely on the policy implications of these results for the United States. That is, from these international comparisons, our interest has been on the lessons that can be drawn from other countries to help improve the crime prevention effectiveness of public area surveillance in the United States. Here we take a look at what some of the key results mean for other Western countries.

The combination of CCTV with other surveillance (e.g., improved lighting, security guards) or situational or social (e.g., fencing, youth inclusion projects) measures seems to be an important contributing factor to the effectiveness of the British studies. Half of the 36 British schemes used one or more other types of interventions alongside CCTV, whereas not one of the eight schemes from the other countries (five of them from the United States) used other interventions. The implication here is that CCTV on its own may not represent a sufficient deterrent threat to influence an offender’s decision-making process to commit a crime, and this seems to have been registered in the United Kingdom. No difference was found between the British and American improved street lighting schemes on their use of other interventions.

Cultural context is another important issue that may be a contributing factor to the difference in effectiveness between the British CCTV schemes and those in other countries. In the United Kingdom, there is a high level of public support for the use of CCTV cameras to prevent crime in public settings (Gill and Spriggs, 2005; Norris and Armstrong, 1999; Phillips, 1999). As noted, in the United States, at least during the period of time that the American CCTV schemes were implemented (before 2000), the public was less accepting and more apprehensive of Big Brother implications of the use of video surveillance in public places (Murphy, 2002; Rosen, 2004).

(p.119) The overall poor showing of CCTV schemes in other countries, especially in the United States, could be due in part to this lack of public support and quite possibly a lack of political support. This may have resulted in cuts in program funding or the police assigning lower priority to the schemes, for example. Each of these factors could potentially undermine the effectiveness of CCTV schemes. That there have been only five high-quality evaluations of CCTV schemes in the United States so far may say something about this lack of public and political support for the use of video surveillance to reduce crime in public places. In the United Kingdom, then, it would seem that maintaining public and political support for CCTV may be important to its effectiveness.

Although cultural context could play a role in the differential effectiveness of street lighting in the two countries (to our knowledge there have been no recent surveys of the public in either country), we believe this is not likely. This is mainly because this form of surveillance is viewed as having few (if any) perceived harmful social consequences.

There may also be something of an age effect that contributed to the difference in effectiveness between the British and American street lighting studies. With the exception of one of the eight U.S. street lighting evaluations, all of them were carried out at least 10–15 years earlier than the first U.K. street lighting evaluation. Could it be that the U.K. evaluations drew on the knowledge gleaned from the previous U.S. evaluations and the detailed review by James Tien and his colleagues (1979), and that this played some role in the effectiveness of the U.K. lighting schemes? This is possible, because there was a great awareness of this American research, which is evidenced in British-based reviews of the literature (Painter, 1996; Ramsay and Newton, 1991) and some of the British lighting studies included in our review.

Another factor that may have contributed to the difference in effectiveness between the American and British street lighting schemes is the possibility that the offenders during the 1990s (in the case of the British studies) may have been influenced by different factors compared with those over a decade ago. There may also be differences between British and American offenders.

Disappointingly, less can be said about implications for other Western countries arising from the results of the other forms of public area surveillance. This is largely owing to there being no possibility for between-country comparisons. Our finding that place managers (p.120) are of unknown effectiveness in preventing crime is more relevant to the U.K. than the U.S. context. This is because the two included studies in our review took place in the United Kingdom. As already noted, the key implication here is that this technique of public area surveillance could benefit from further evaluation research. Our finding that security guards are promising when implemented in car parks and targeted at vehicle crimes is based on two studies from different countries. For security guards, even more so than for place managers,2 the implication is that this technique of formal surveillance could benefit from further experimentation.

On the matter of the defensible space practice of street closures or barricades, which we found to be effective in reducing crime when implemented in inner-city neighborhoods, all four studies included in our review took place in the United States. Although there was no possibility for an international comparison of the effects on crime, other countries could draw lessons from the U.S. experience with this natural surveillance technique. For example, as noted, the potential social costs associated with street closures or barricades are minimal if not nonexistent. These are particularly effective in high-crime areas and with a range of crimes. In three of the four studies, they were implemented in high-crime inner-city neighborhoods (the other case involved implementation across the city). Each study produced desirable effects on overall crime or on a specific crime type (violence in one case and property in another). One other important point that may have salience for other cities across the United States or in other countries is that the effectiveness of street closures or barricades is very much viewed as a product of increased natural surveillance on the part of residents who now feel safer being outside their homes—for example, walking the streets or visiting the park.

Surveillance in Private Space

This book is purposely focused on surveillance in public places. As discussed in chapter 1, this is not to deny the importance as well as the widespread use of surveillance measures to prevent crime in private domains. Instead, our specific focus on public places allows for a more comprehensive examination of one aspect of the current debate on surveillance and crime (p.121) prevention. This focus also recognizes the growing use of surveillance measures to prevent crime in public places in the United States and in other Western nations.

Each public area surveillance measure discussed in this book is also used to prevent crime in private places. For example, the first use of CCTV cameras to prevent crime took place in the private sector. In the United Kingdom, cameras were first introduced in the retail sector; in the United States, it was in banking. It is common nowadays to encounter video surveillance in a wide range of private domains, including convenience stores, shopping malls, gas stations, banks, and even homes. Security guards are more often associated with the protection of private places,3 but they are increasingly being used to protect public places (Sklansky, 2006).

Our specific focus on public area surveillance is the reason we did not assess the effects on crime of neighborhood watch programs. This highly popular form of citizen surveillance is most often used to prevent crimes at private residences. In a systematic review and meta-analysis of neighborhood watch, British criminologists Trevor Bennett, Katy Holloway, and David Farrington (2006) found this program to be effective in preventing crime.

There is still another dimension to our specific focus on public area surveillance. This pertains to the poor state of evaluation research on private area surveillance practices. We found few high-quality studies that evaluated the effects of surveillance measures on crime in private places. One of the reasons for this paucity of research is the private sector’s resistance to independent evaluation of their practices and, equally important, making any evaluations (independent or otherwise) publicly available. Although there are some excellent evaluations of the application of situational crime prevention practices in the private sector (see e.g., Eck, 2006; Hunter and Jeffrey, 1997), until such time that the private sector embraces evaluation research more fully it will be difficult to assess in any comprehensive way the effectiveness of surveillance practices in preventing crime in private places. Another reason for the poor state of evaluation research in the private sector may stem from biases of criminologists about what is interesting and useful research and the fact that governments have not fully understood that assisting private security benefits the public sector as much as the private.

(p.122) Conclusions

For some Americans, the idea of surveillance technology being used in public areas conjures up images of Orwell’s Big Brother society—a society that is constantly watching (and recording) every move, every action that one takes (Stanley and Steinhardt, 2003). To return to a point made throughout this book, not all forms of surveillance are as potentially intrusive and raise questions about the infringement of civil liberties and other social costs. Importantly, these social costs need to be weighed against any crime prevention benefits that accrue from the different forms of surveillance.

The public area surveillance measures that we found to be effective in preventing crime seem to present few (if any) social costs to the general public. Improved street lighting is perhaps the most obvious. It is the most effective of the main surveillance techniques in preventing crime, and its only drawback is that it may cause an increase in light pollution. The defensible space practice of street closures or barricades in inner-city neighborhoods is also highly effective in preventing crime (both property and violent). By all accounts, it also does not appear to cause any undue hardship to residents in these areas. The only effective use of CCTV—when implemented in car parks and targeted at vehicle crimes—may cause little resistance on the part of the general citizenry, at least compared to its use in other public settings. Indeed, this is where the debate heats up. On one hand, CCTV is shown to be not very effective in preventing crime in city and town centers, public housing communities, and transportation facilities. On the other hand, in these areas the potential social costs are most troubling.

We do not anticipate that the use of CCTV will cease anytime soon in these other public areas, nor should it. What we do hope is that consideration is given to the research evidence on other public area surveillance measures that are effective (often in combination with other surveillance interventions), and that lessons from the effective use of CCTV may be applied to other conditions and contexts. Importantly, drawing on the lessons from the effective use of CCTV may not only go some way toward improving its crime prevention effectiveness in other public settings, it may also contribute to reducing its social costs. In this respect, results of between-country comparisons may also have some relevance for other Western countries.

In the next chapter, we offer some concluding thoughts and identify future directions to help make public places safer in this age of surveillance.


(1.) Disappointingly, too few of the included studies carried out a cost-benefit analysis to allow for this important dimension to be considered here. We return to this subject in chapter 9.

(2.) This has to do with the formal meaning of promising (see chapter 7).

(3.) David Farrington and his colleagues (1993) conducted one of the few high-quality evaluations of the effects of security guards on crime in private space.