Economic and Noneconomic Factors Influencing Geographic Differentials in Homelessness: An Exploratory State‐Level Analysis

• Published date: 01 March 2020
• DOI: http://doi.org/10.1111/ajes.12320
• Author: Gigi M. Alexander,Richard J. Cebula
• Date: 01 March 2020

Economic and Noneconomic Factors

Influencing Geographic Differentials in

Homelessness: An Exploratory

State-Level Analysis

By RichaRd J. cebula* and GiGi M. alexandeR†

abstRact. In this exploratory state-level empirical study for the United

States, the authors estimate a pooled time-series/cross-section

framework, with control variables for population size and population

growth, for the years 2015–2016. Within this context, the least squares

estimates lead to the following tentative findings: (1) homelessness is

positively associated with the overall cost of living, on the one hand,

and the average rent level, on the other hand; (2) homelessness appears

to be an increasing function of the percent of the population without a

high school diploma but a decreasing function of the percent of the

population with a bachelor’s degree or a higher level of formal

education; (3) homelessness is a decreasing function of labor market

freedoms reflecting the degree of union density and union power, on

the one hand, and excessive government employment beyond that

needed solely for productive and protective services, on the other hand;

(4) homelessness is positively associated with personal freedom from

incarceration and arrest; and (5) homelessness is negatively associated

with income, as higher income reduces homelessness. Based on these

findings, preliminary policy implications are also provided.

Introduction

With over half a million persons considered homeless in the United

States, homelessness continues to be a challenge across the nation

American Journal of Economics and Sociology, Vol. 79, No. 2 (March, 2020).

DOI: 10.1111/ajes.12320

© 2020 American Journal of Economics and Sociology, Inc.

*Associate Editor, Journal of Economics and Finance, co-author with Jim Koch

ofRunaway College Costs, Johns Hopkins University Press, in press, 2020; Walker/Wells

Fargo Endowed Chair in Finance, Department of Accounting, Economics, and Finance,

Jacksonville University, Jacksonville, Florida. Email:rcebula@ju.edu

†Department of Accounting, Economics, and Finance, Jacksonville University,

Jacksonville, Florida. Email:gigialexander44@gmail.com

512 The American Journal of Economics and Sociology

(U.S. HUD 2015, 2016). Public and private agencies have sought to

find remedies for homelessness, but the problem persists despite these

efforts. One element impeding progress is insufficient knowledge of

the factors that contribute to the problem. This article hopes to make

a small contribution to better understanding of the factors that are

involved in homelessness by making state-level comparisons. Before

turning to our specific findings, we first examine some of the work

that has been previously undertaken to understand homelessness.

Previous Analyses of Homelessness

Homelessness has received considerable attention over the years not

only in the media but in other contexts as well. For example, a variety

of government agencies and policy groups have directed attention

toward this issue: the U.S. Department of Health and Human Services

(U.S. HHS), the U.S. Department of Housing and Urban Development

(U.S. HUD), the National Network for Youth (NNY), and the National

Center for Homeless Education (NCHE). There is also a rather exten-

sive academic/scholarly literature revolving around the issue of home-

lessness, with much of it seeking to investigate, identify, and better

understand factors that influence the incidence of homelessness. This

literature investigates such diverse dimensions as the price of rental

housing, rent control, poverty/welfare, employment issues, climate,

gender, and educational attainment.

An early example of empirical work on factors that impinge on

the level of homelessness is a study by Grimes and Chressanhtus

(1997) that endeavors to assess the impact of rent control on home-

lessness, as measured by the S-Night Count, a technique discussed

below. While acknowledging its limitations, Grimes and Chressanhtus

(1997: 26) adopt the perspective that the S-Night Count “reflect[s] mea-

sures of two displaced populations normally perceived as the chronic

homeless—the shelter count and the street count.” Accordingly, the

authors provide estimates for three measures of homelessness (each

as a percent of total city population): 1) the shelter count; 2) the

street count; and 3) the total count (the shelter count plus the street

count) for the 200 cities included in the dataset. Their findings indicate

that rent control does in fact exercise a statistically significant positive