EQ Vol.14: Eyes on the Street: Testing Jane Jacobs

Contributing Writer: Zachary Golden | May, 2024

The “eyes on the street” theory hypothesizes that the presence of people in a public space deters crime.[1] American writer Jane Jacobs introduced the theory in her 1961 book The Death and Life of Great American Cities. The book is part polemic against the Modernist style of urban planning (developed by architect Le Corbusier and his colleagues, then crystallized into 20th-century planning orthodoxy) and part treatise on the nature of cities. It was one of the first and most influential works to criticize Modernist urbanism, with its emphasis on separating land uses and modes of transportation — along with planning practices such as slum clearance, urban renewal, public housing, and downtown highways. Jacobs and her contemporaries initiated a critical discourse which has continued over the past six decades. Jacobs introduced concepts which have influenced social science fields such as urban economics and environmental psychology.[2] To approach the eyes on the street theory, we begin with her informal explication and then discuss the academic literature that builds upon it.

Jacobs saw cities as complex and dynamic environments influenced by the interplay of many actors. In this way, her view is similar to that of economics (titles of her other books include Cities and The Wealth of Nations and The Economy of Cities). She begins her treatise with an exploration of the uses of sidewalks and parks in the city.

Why are some streets safer than others? She poses the question like an economist– in terms of incentives and their logical influence on behavior: “How much easy opportunity does [a given street] offer to crime?” Law enforcement is not always present, but informal social behavior also deters crime (along with other antisocial activity). In the context of an urban street, a public presence is one such deterrent. Environmental psychologists call this natural surveillance — people organically observe their surroundings. A criminal who thinks he will be seen should be less inclined to offend. A street full of life and activity has many natural observers, multiplying this effect.[3]

The basic theory is this: “A well-used city street is apt to be a safe street. A deserted city street is apt to be unsafe.” Certain qualities, then, can make a street well-used and thus safe. Jacobs describes three:

  1. “There must be a clear demarcation between what is public space and what is private space.” (“so that the area needing surveillance has clear and practicable limits”)
  2. “There must be eyes upon the street, eyes belonging to those we might call the natural proprietors of the street.”
  3. “The sidewalk must have users on it fairly continuously, both to add to the number of effective eyes on the street and to induce the people in buildings along the street to watch the sidewalks in sufficient numbers.”

 

These last two depend upon land use. Retail, for instance, allows “natural proprietors” to emerge. Store owners and employees are capable of observing from sidewalk-facing buildings and have a particular sense of concern for their street.[4] Public-serving land uses such as retail also draw a broad public to the sidewalks. If these are diverse, they draw people in at various times, ensuring a public presence “fairly continuously” throughout the day and week. Bars, restaurants, housing, parks, churches, offices, and so on draw people in at different times. Such a mix of land use creates not only a lively public life, but a greater level of safety.

Consider as a counterexample city neighborhoods and districts dominated by a single use. Large residential neighborhoods that lack bars, restaurants, and corner stores have empty streets at night. These feel and often are unsafe as a consequence. Office complexes, university campuses, public housing projects, and other isolated districts are sometimes plagued by safety concerns, or require formal measures such as private security (university police departments, gated communities, doormen for residential or office towers, etc.) to compensate for the lack of organic security provided by a lively street.

Jacobs’s theory has a strong intuitive appeal. It invites some empirical questions. Can we find evidence, for instance, that retail deters crime? What combinations of land use are more or less effective at creating a safe and lively public realm? While the eyes on the street theory is concerned chiefly with safety and crime, a better understanding of the links between land use and patterns of behavior can have broad implications.

A recent paper provides some evidence and suggests a path for further research. Chang and Jacobson (2017) investigate whether marijuana dispensaries draw crime to their area (as more states legalize marijuana, this has become a prominent concern). The authors use data from Los Angeles, where a city-wide policy forced some dispensaries to close and allowed others to remain open. The city placed a cap on the number of dispensaries, and told those which had registered with the city after a certain date to close.

The selection criteria was not correlated with local characteristics which might influence crime levels. The two groups of dispensaries – those which closed and those which remained open – were selected quasi-randomly. This, therefore, is a natural experiment. The authors obtained data on which dispensaries closed and data on crime types, dates, and locations. They tested whether crime rates in a radius around each dispensary were different between the two groups after the closure date.

Their findings contradict the popular claim that dispensaries draw crime. When dispensaries closed, crime increased in the immediate area during the following weeks. The dispensaries that remained open saw no such increase. The authors view eyes on the street as a possible explanation; with dispensary employees and customers absent, fewer people are around to deter crime. The same logic should apply to retail in general. To test this broader hypothesis, the authors performed similar tests with restaurants that closed due to code violations. They found the same result: crime increased after the closures. Crime types likely to be deterred by bystanders – theft from vehicles and property crime – rose in particular.

The authors also incorporated Walk Scores – a measure of walkability – in their analysis, to better estimate the impact of foot traffic.[5] More walkable places saw a smaller increase in crime. Since customers of one business in these areas represent a proportionally small share of the foot traffic, the amount of foot traffic could remain high after a closure, providing some deterrent still.

This evidence supports Jacobs’s theory that retail deters crime. The businesses closed, activity decreased, and opportunistic crime increased. This analysis provides a path for further research. While this natural experiment was particular to Los Angeles, one could explore restaurant closures in other cities as the authors did here.

The results may be different in different cities. Los Angeles has a relatively low density, which might impact the results. In a denser city like New York, the density of retail stores is likely higher. Should one close, it may make only a small impact, given the large amount of foot traffic which remains. It is also possible that land uses are more diverse in New York, drawing foot traffic around the clock to a greater extent than in Los Angeles. Closures in New York might then predict a smaller increase in crime, if a robust and vibrant block remains. Denser cities, therefore, might provide less clear results. If these predictions are true, though, analysis of denser cities might also provide more evidence for the theory.

These land use concerns, however, are already accounted for by Walk Scores to some extent. Other differences between cities may impact results, such as the relative amounts of foot traffic and automobile traffic or the sizes of blocks. An analysis of other cities can provide more evidence for or against the eyes on the street hypothesis. It can also shed more light on which land use and urban design factors produce safe streets, though disentangling the many differences between cities would be a challenge in making city-to-city comparisons.

These questions have broader implications beyond crime. Certain land use types (such as retail) can potentially be shown to draw more people to an area or to deter crime. A decrease in crime — if caused by the presence of particular land uses — is a positive externality, since it benefits people other than the relevant business and its customers. If land use mixtures can reliably draw in a lively and diverse public, other benefits may follow – indeed, Jacobs suggests that urban diversity and proximity, broadly speaking, have a myriad of positive effects. While safety is significant in its own right, it is worth also investigating the other implications of local land use mixtures, in terms of Jacobs’s broader thesis that dense and diverse land uses create more vital and prosperous cities.

Notes:

[1] Jacobs’s theory is concerned primarily with streetscapes, though it can be applied to other types of public spaces as well. Her theory is not restricted to those who are physically in this public space, as natural surveillance can occur from adjacent private buildings as well, including storefronts and homes.

[2] Jacobs influenced the work of Paul Romer and Edward Glaeser, among others. See Duranton 2017 for discussion of Jacobs’s work in the context of economics.

[3] She contrasts this with a different system of informal controls in non-urban settings: these generally have smaller populations and thus less anonymity. People are not always present, but people know each other better; “a web of reputation, gossip, approval, disapproval and sanctions” operates instead.

[4] It is important that this retail (or any type of land use) is directed toward the street for this effect to occur. The suburban type of retail that is distant from the street – typically separated from it by parking lots – would allow little natural surveillance.

[5] Walk Scores rate an address based on its distance to 9 categories of amenities. A higher Walk Score indicates that the address is more walkable. See https://www.walkscore.com/methodology.shtml for more information.


REFERENCE

Chang, Tom Y., and Mireille Jacobson. “Going to pot? the impact of dispensary closures on crime.” Journal of Urban Economics, vol. 100, 21 Apr. 2017, pp. 120–136, https://doi.org/10.1016/j.jue.2017.04.001. 

Duranton, Gilles. “The death and life of Great American cities / the economy of Cities.” Regional Studies, vol. 51, no. 12, 29 Sept. 2017, pp. 1871–1873, https://doi.org/10.1080/00343404.2017.1364041. 

Glaeser, Edward L. Triumph of the City: How Our Greatest Invention Makes Us Richer, Smarter, Greener, Healthier, and Happier. Penguin Press, 2011. 

Jacobs, Jane. The Death and Life of Great American Cities. Random House, 1961.