The geography of unsheltered homelessness in the city: Evidence from “311” calls in New York

Abstract

We provide the first detailed evidence on the distribution of unsheltered homelessness within the city. We rely on “311” call data reporting unsheltered homeless individuals to New York City authorities, based on evidence that the distribution of calls reflects their actual locations. Centrality dominates an otherwise important role of median income in explaining variation in unsheltered homelessness across neighborhoods. Subway stations and restaurants are important as well. We also find that police respond more quickly to calls in more affluent and central neighborhoods. This suggests that the distribution of homelessness across neighborhoods could affect the city's overall response.

Neighborhood Distribution of Unsheltered Homelessness and its Temporal Changes: Evidence from Los Angeles

“…The results indicate that unsheltered homeless people in the study area are heavily concentrated in and around the city center and in neighborhoods with greater access to shelters and with lower socioeconomic status. Unlike in New York (Corinth and Finley 2020), poverty rates are positively associated with unsheltered homeless population density in the Los Angeles area. Notably, neighborhoods closer to the city center experienced a relatively large increase in unsheltered homelessness during the study period, implying a persistent spatial concentration of unsheltered homelessness.…”

“…Some urban parks serve as refuges, partly due to their shower facilities and bathrooms. Corinth and Finley (2020) revealed that subway stations also significantly explain neighborhood variations in unsheltered homelessness.…”

“…The authors maintained that social services tend to be clustered in a relatively small portion of the city, mostly in inner-city neighborhoods, leading to the concentration of homeless people near such services. Suzuki (2008) and Corinth and Finley (2020) also provided empirical evidence that spatial proximity to supportive services, such as shelters and public health centers, is significantly associated with the neighborhood distribution of unsheltered homelessness. However, it remains unclear whether homeless people choose to settle near these services or the services are provided in areas with an existing heavy concentration of the homeless population (both explanations may apply).…”

“…Previous studies have also suggested that neighborhoods with higher socioeconomic status tend to have the political resources needed to engage in anti-homeless activities (Marr, DeVerteuil and Snow 2009), which likely force homeless people to stay outside wealthier and Whiter neighborhoods. However, a recent study found a positive association between income and neighborhood-level unsheltered homelessness in New York City, although such association became statistically insignificant after adjusting for centrality (Corinth and Finley 2020). Given that the centrality of a neighborhood mediates the relationship between income and homelessness, the authors concluded that this relationship may vary by region, depending on the spatial patterns of income (i.e., whether or not poor people are concentrated near the city center).…”

“…First, whereas Corinth and Finley (2020) relied on homeless-related “311” call data to identify the locations of unsheltered homeless people, the present study utilizes the actual homeless count data systematically collected by the LAHSA. Because “311” call data are collected based on voluntary reports, they might be affected by differences in reporting behavior, particularly using 311 systems, across different population groups (O’Brien, Sampson and Winship 2015; Kontokosta and Hong 2021).…”