Homelessness in the United States: What Research Shows About Causes and Effective Responses

Homelessness in the United States has become an increasingly visible and politically charged issue. Encampments in major cities have prompted court battles, ballot initiatives, and public debates about enforcement, services, and housing. Understanding what research shows about the causes of homelessness and the effectiveness of different policy responses is essential for moving past both the punitive approaches that research consistently finds ineffective and the advocacy framings that sometimes overclaim what is known. The definition of homelessness matters for measurement and for policy. The federal government counts people who are unsheltered or staying in emergency shelter on a single night in January, the Point in Time count, which is the most widely cited measure but has known limitations. Shelter counts miss people sleeping outdoors who avoid contact with outreach workers. The count also misses people in doubled-up housing situations, staying temporarily with family or friends due to economic necessity, which researchers call the hidden homeless population. When broader definitions are used, estimates of the number of people experiencing some form of homelessness annually are substantially larger than the single-night count suggests. Housing costs are the most consistent predictor of homelessness rates at the metropolitan level. Studies comparing homelessness rates across cities and states find that areas with high housing costs and low vacancy rates have substantially higher rates of homelessness than areas with more affordable housing markets. This relationship has strengthened as housing costs have risen dramatically in many high-demand metropolitan areas. Research by economists including Gregg Colburn and Clayton Aldern finds that housing market variables explain far more of the variation in homelessness rates across cities than variables related to shelter availability, mental illness prevalence, drug use, or weather. Individual risk factors for homelessness have been studied extensively. Substance use disorders, serious mental illness, histories of incarceration, foster care involvement, domestic violence, and episodes of extremely low income all substantially increase the probability of experiencing homelessness. These factors concentrate among people who experience homelessness and are often cited as its primary causes. However, the relationship between these individual factors and homelessness is moderated by housing market conditions: the same individual vulnerabilities produce homelessness at much higher rates in expensive markets than in affordable ones. The composition of the homeless population has shifted in ways that are relevant to policy design. Chronic homelessness, defined as extended or repeated homelessness combined with a disabling condition, represents a relatively small share of people experiencing homelessness on any given night but accounts for a disproportionate share of emergency services utilization. Family homelessness, which is often shorter in duration and driven primarily by housing cost and income volatility, represents a large and often invisible share of the total. Youth homelessness, including LGBTQ+ youth who are disproportionately represented, reflects a combination of family rejection, aging out of foster care, and lack of income support. Housing First is the best-supported policy approach for addressing chronic homelessness. Housing First programs place people directly into permanent supportive housing without requiring sobriety, treatment compliance, or evidence of housing readiness as preconditions. The evidence from randomized controlled trials is unusually strong for social policy: Housing First consistently outperforms treatment-first approaches in maintaining housing stability. Housing stability does not automatically produce recovery from substance use or mental illness, and critics sometimes overstate Housing First's claims in this regard, but it is the most effective approach for keeping chronically homeless individuals housed. Emergency shelter is an immediate harm reduction tool, but research finds that shelter capacity does not reduce the overall number of people experiencing homelessness in a community over time. Shelter is a temporary intervention that addresses the immediate safety need without addressing the underlying conditions that produce homelessness. Communities that invest heavily in shelter without investing equally in permanent housing see populations cycle through shelter repeatedly without achieving stable outcomes. Enforcement-based approaches, including anti-camping ordinances, sweeps of encampments, and criminalization of homelessness behaviors, are the most common policy responses in many jurisdictions and among the least supported by research. Studies find that enforcement actions disperse encampments and temporarily move homeless individuals out of view but do not reduce the overall number of people experiencing homelessness and often disrupt the continuity of services, damage trust between homeless individuals and outreach workers, and destroy the limited possessions and documents that people need to access housing and benefits. Prevention is receiving growing attention. Research finds that relatively modest financial interventions, including one-time rental assistance, mediation with landlords, and utility assistance, can prevent housing loss for individuals and families on the margin of homelessness at much lower cost than treating homelessness after it occurs. Targeting prevention resources toward households at highest risk of becoming homeless maximizes impact per dollar invested. The research picture supports a conclusion that homelessness is primarily a housing problem rooted in the failure to produce and maintain affordable housing at the scale that housing markets require, and that effective responses center on housing with appropriate support services. Recognizing this does not minimize the complexity of individual circumstances but does redirect policy attention toward the structural conditions that produce homelessness at scale.