Outdoor lighting can enhance function, safety and in some settings, if properly designed for the purpose, it can enhance security. However, street lighting, by its very nature, is focused on vehicular and pedestrian travel safety. Nonetheless, much of the public assumes that street lighting can deter crime despite the fact that there is no reliable objective evidence to support that conclusion.  A great deal of unnecessary street lighting has been deployed on the basis of this assumption with adverse consequences for climate change, damage to wildlife and habitat, human health and the esthetics of our urban communities, not to mention wasting limited public funds. Where crime is a serious consideration, we should employ effective countermeasures, and avoid unnecessary street lighting that is more likely to attract and/or facilitate crime than prevent it. Responsible lighting practices, as defined by the Illumination Engineering Society, will make our cities and towns safer, more efficient, ecologically sound and esthetically pleasing.
We need to distinguish between safety and security, terms which are often ambiguously interchanged in discussing outdoor lighting. In this document “security” relates to criminal conduct, while “safety” alludes to accidental personal injury.

Security lighting is a complex subject.  The Illumination Engineering Society (IES) has a 61-page manual dedicated to the subject. (IESNA G-1-03) While your local power company typically offers “security lighting” this is usually nothing more than a barnyard light hung on an existing pole, or in a commercial setting, a flood light aimed nearly horizontal across the target property, creating glare, shadows and trespass on adjoining property.  It is highly doubtful that such lighting is effective for security and may, in fact, be counterproductive.  The reason that the IES recommends using lighting on motion sensors for residential security purposes instead of static always on lighting. (IES RP-33-14, 4.6.1)

But, our focus here is on street lighting.  Street lighting’s purpose is to promote travel safety on roadways and adjacent sidewalks. (ANSI/IES RP-8-14, 1.2&1.3) An early example of this concern is the bequest of a civic-minded seventeenth-century citizen of Billingsgate London who earmarked in his will four pounds a year in perpetuity to maintain a lamp in an area where he had once fallen and broken a leg.(i)

Before the advent of street lighting, a nighttime urban traveler might hire a lantern bearer or “link-boy” to navigate city streets. Unfortunately, there was the risk that an unscrupulous link-boy might lead a client into an alley for a mugging – an early example of lighting facilitating rather than deterring crime.(ii)

Nevertheless, many in the public have come to assume that increased security is a collateral if not a primary benefit to lighting residential streets. This is understandable, humans are not nocturnal animals and natively fear the dark. For humans, light is good and darkness synonymous with evil. How could light not make us more secure?

This instinctive assumption has made it easy for those with an economic interest in street lighting as well as politicians (who have long found street lighting to be an easy visible way of appearing to do something about the ever-present concern for crime) to promote the concept that street lights deter crime. In consequence, streetlights are installed where there is no safety need and shine all through the night on vacant streets.  This, along with other careless uses of artificial light, has contributed to polluting the developed world with light at night.

Today there is almost no naturally dark night sky in the continental United States East of the Mississippi River and south of upper New England.(iii) It has been estimated that 80% of the US population has lost the ability to personally experience the Milk Way Galaxy – arguably nature’s grandest spectacle.  This has lead to concerns about the potential psychological, emotional, cultural and esthetic cost of removing the reality of the universe over our heads from our daily lives. (iv)

But there are other more widely shared and more tangible concerns that have emerged of late.  Outdoor lighting at night is a significant contributor to greenhouse gas emissions.  One 100 watt light bulb burning all day can consume nearly half a ton of coal in a year and generate a ton of CO2.  The AMA has twice warned of adverse impacts to wildlife and human health associated with turning night into day with artificial light.(v) These concerns have prompted the Federal Highway Administration to review its roadway lighting recommendations.(vi)

While the costs are becoming more apparent the benefits of street lighting, at least as relates to security, are elusive at best as crime remains with us and nowhere is it more concentrated than in the heart of our light polluted cities, where it is higher than in the less well-lit suburbs, where it is higher than in rural communities.  If artificial light deters crime why does this pattern exist and why does crime repeatedly fall during large-scale power outages?(vii) Indeed, the term lunatic predates artificial lighting to describe those with odd behavior, doubtless derived from the mischief that would emerge at night around the full moon.

These facts should cause us to ask, have our assumptions about lighting at night and security been wrong? Is it possible that less light at night on our streets would provide a less friendly environment for criminals? 

Implicit in the assumption that light provides security, even if it is only lighting a narrow band of a street, is that criminals are like cockroaches; they will scurry away from light to hide in the dark for fear of being detected and not come out again until it is dark. But, of course, criminals are not cockroaches.  They don’t wear stripped shirts and masks as Dastardly Dan is depicted in power company adds either.  On reflection, it is silly to think lighting the street would in any way concern a criminal casing your neighborhood; to the contrary.
Sure, criminals seek not to be detected doing their thing, but the strategy for that is to operate when and where no one is looking. The reason no doubt, contrary to popular belief, most burglaries occur in the daytime when homes are vacant and they can work inconspicuously by the light of day.(viii)

No, criminals need light like everyone else, a fact implicit in Dastardly Dan’s flashlight.  If criminals need light, it should be obvious that lighting my attract rather than deterrent crime.  The precise point made in a 1997 Report to Congress, by the National Institute of Justice in which, after an exhaustive review of the literature addressing the impact of lighting on open spaces in general, concluded: “We may speculate that lighting is effective in some places, ineffective in others, and counterproductive in still other circumstances. The problematic relationship between lighting and crime increases when one considers that offenders need lighting to detect potential targets and low-risk situations (Fleming and Burrows 1986).”…continuing…”We can have very little confidence that improved lighting prevents crime, particularly since we do not know if offenders use lighting to their advantage.”(ix)

This conclusion is born out by the literature. The study most often cited in support of street lighting as a deterrent to crime was commissioned by the British Home Office. The study did a meta-analysis of street lighting improvement projects in American and British cities occurring over several preceding decades. It documented a reduction in crime in about half the projects but no effect in the other half. Puzzling until you learn that in each instance of an apparent positive impact, crime dropped as much in the day as at night.  This led to the authors to a conclusion frequently ignored by those sighting it in support of street lighting as a security device. That being that the observed benefit, when present, was likely due to an impact on community pride arising from the investment made in those particular communities, “rather than increased surveillance or deterrent effects.”(x) This raises the question, if the light itself is not of benefit could the money spent on lighting be better spent in ways that more directly impact the social and economic causes of poverty that would be less damaging to the environment?

All the instances reviewed by the meta analysis above involved long after the fact analysis with resulting gaps in information.  A more rigors case study was undertaken by the Illinois Criminal Justice Authority in evaluating the impact of an alley lighting program planned for the City of Chicago. The study had two different control baselines: the incidence of crime in the alleys to be illuminated before and after the illumination, and the before and after incidence of crime in alleys in similar neighborhoods that were left un-illuminated. The study found that crime of all types rose significantly in the illuminated alleys compared to pre-illumination and compared to the non-illuminated alleys.(xi)

An even more recent study takes us back to Britain where an increasing number of local councils in England and Wales have been turning off or dimming streetlights late at night during “light curfews” to save money and reduce carbon emissions. The National Institutes of Health Research funded a study to evaluate the impact on both crime and vehicle accidents. Covering fourteen years and 62 councils, the study found no discernible increase in crime or accidents.(xii)
While not a street study, another recent American study with an aggressive statistical analysis evaluated the effect of the “Summer Night Lights” project intended to reduce crime in Los Angeles city parks by lighting them at night. Depending on which of several statistical methods were used, and the year considered, the best the study could come up with was a .33% daytime and .45% nighttime reduction in crime. The conclusion: the Summer Night Lights was not cost-effective.(xiii)

While anecdotal, the experience of a number of individual local governments on both side of the Atlantic suggests that light curfews reduce crime at night. Typical is the Bristol Post report that the police in Bristol England “have discovered that criminals are afraid of the dark.” After the city started turning street lights off between midnight and 5 am, crime not only did not rise, as some feared, it dropped as much as 20%.(xiv) These results have crossed the Atlantic into this country.(xv)

Similarly, numerous school districts in  Texas, California and Washington state turned campus lighting off at night to save money and found that vandalism and burglary drop dramatically often to near zero.(xvi)
In the face of mounting evidence that streetlights are not effective in crime prevention, advocates have shifted the argument to making people feel more secure. This is a worthy consideration if people are in fact secure. However, a false sense of security can be a disservice. Throwing up some lights as a substitute for effective countermeasures to crime can result in people feeling secure when they are not.
Once one thinks about it, all this leads to a rather obvious conclusion. The limitation of lighting, as a security device, lies in the fact that lights can’t see. We don’t need empirical studies no know that a person out at night can better protect themselves, both as to safety and security, if their environment is illuminated. Lighting sidewalks, parking lots and nighttime outdoor spaces when and where there are people present, makes sense and can enhance function, security, and safety. Likewise, if there is a reasonable expectation of criminal activity being observed, such as the presence of a night watchman at a warehouse, or an active night scene, lighting may well deter crime. But lighting deserted streets in the dead of night can in no way deter crime and may well invite or even facilitate it as seen in Chicago’s alleys by attracting or facilitating criminal activity.
Accordingly, IES recommends dimming street lighting during the night as activity levels change, and turning off area lighting after normal activity ceases. (xvii)
A final note, while streetlighting is likely not effective as a security device, poor street lighting is worse than useless. Uneven lighting, typical of street lighting in most residential areas, creates contrasting areas of dark and light that can inhibit witness observation by making the dark areas less visible than if there were no street lighting at all. While over-lighting streets with improperly shielded fixtures, typical of much residential street lighting, can produce glare reducing the visibility of adjoining areas.  If you are going to light, light properly.(xviii)


Crime is not the progeny of darkness, but rather has socioeconomic roots. What works and does not work to reduce crime reflects its source. (xix)
It is clear that there are very real consequences to lighting the nighttime environment and accordingly we must find ways to rationalize our use of light at night to extract its real benefits while minimizing the adverse impacts. 

In the end, street lighting’s value relates to travel safety, particularly pedestrian. Those who promote it as a security device are manipulating human fear of the dark and are steering us away from real cost-effective measures to address crime with the false promise that in more light lies security. Where crime is a serious consideration, we should seek actual solutions and countermeasures, not pander to uninformed opinion or substitute a false sense of security.

The recommended practices for outdoor and street lighting, as published by the Illumination Engineering Society, should be any city’s guide in formulating a modern science-based lighting plan. IES recommendations represent the best science-based thought on when, where and how to light. Responsible lighting will make for a safer, more efficient, ecologically and esthetically sound city.


i. John Timbs, Curiosities of London, 1867



iv. A Review of the Elements of Human Well‐Being with an Emphasis on the Contribution of Ecosystem Services, Ambio. 2012 June; 41(4): 327–340.

v. American Medical Association, REPORT 4 OF THE COUNCIL ON SCIENCE AND PUBLIC HEALTH (A-12) Light Pollution: Adverse Health Effects of Nighttime Lighting and REPORT OF THE COUNCIL ON SCIENCE AND PUBLIC HEALTH, CSAPH Report 2-A-16, Human and Environmental Effects of Light Emitting Diode (LED) Community.

vi. “Ongoing research demonstrates the impact of lighting at night as it relates to human health and the condition of wildlife and plant life. As a result, revisions are being made in our approach to light control and recommended lighting levels. This research also affects the decision‐ making process on whether, and where, lighting is beneficial.” Section 5, FHWA Lighting Handbook, 2012.

vii. Lighting and crime, part 1, B.A.J. Clark. (section 3),

(Note: D. Dan is provided with a flashlight to imply that he operates at night to encourage the installation of “security” lighting.  However, this raises two questions. If he needs light why would we provide it for him and wouldn’t he be more, not less, conspicuous if he needed to use that flashlight?)
ix. Preventing Crime: What Works, What Doesn’t, What’s Promising, A Report to the United States Congress, Prepared for the National Institute of Justice – 1997.

(Note: The 1997 study starts by referencing the predecessor agencies 1979 review of sixty studies on the topic that concluded:

“Is street lighting an effective approach to the reduction and deterrence of crime? The answer is inconclusive. The paucity of reliable and uniform data and the inadequacy of available evaluation studies preclude a definitive statement regarding the relationship between street lighting and crime.” (Tien, et al. 1979, page 93.)

x. Effects of improved street lighting on crime: a systematic review David P. Farrington and Brandon C. Welsh, 2002.
(Note: A study that briefly swung opinion in favor of a positive effect was Improved Street Lighting: Crime Reducing Effects and Cost-Benefit Analyses, Painter and Farrington, Security Journal Volume:12  Issue:4:1999.  However, upon reflection, its credibility suffered on numerous counts.  First, it was funded by electrical distribution interests. Second, it relied on neighborhood surveys rather than police statistics. The control community was substantially dissimilar in pre-lighting crime rates. The results were improbably high and potentially explainable by reversion to mean and/or the substantial change in police surveillance occurring between the study community and the control community during the before and after period. This was one of the studies relied upon by Farrington andWelch in their meta-analysis.  The other affirmative studies suffer from similar shortcomings.  The reader is referred to Outdoor lighting and Crime: Part 1, B.A.J. Clark for a complete critique.)

xi. The Chicago Alley Lighting Project: Final Evaluation Report
April 2000, Illinois Criminal Justice Information Center.

xii. Journal of Epidemiology and Community Health. DOI: 10.1136/jech-2015-206012.

xiii. Studying the Impact of the “Summer Night Lights” Program on Crime in Los Angeles, Siddheshwar “Sid” Salvi, Amherst College, 2011.


xvii. ANSI/IES RP-8-14, 5.4 and IES RP-33-14, 2.4.1 and 2.5.2-7

xviii. IES RP-33-14, 4.6.1 Safety Lighting, “Too often, people associate brighter light and glare with “safer” surroundings. In reality, more light and glare do not equate to better lighting. It can be easily demonstrated that too much light or misdirected light, actually causes a loss of visibility.”

xix. Preventing Crime What Works, What Doesn’t, Whats Promising, U.S. Dept. of Justice, Research in Brief, 1998.

xx. Benefits and costs of artificial nighttime lighting of the environment, Gaston, Environmental Reviews, Environmental Reviews, 2015, 23(1): 14-23,