Homeless in America, Homeless in California

Summary

"Homeless in America, Homeless in California" is a 2001 paper by John M. Quigley, Steven Raphael, and Eugene Smolensky (all then at UC Berkeley), published in the Review of Economics and Statistics 83(1): 37–51. It is the foundational empirical statement of the housing-market theory of homelessness — the direct econometric ancestor of Colburn & Aldern (2022) — and it long predates any Georgist framing.

The authors test "the alternate hypothesis that variations in homelessness arise from changed circumstances in the housing market and in the income distribution," against the then-dominant belief that homelessness "arisen from broad societal factors, such as changes in the institutionalization of the mentally ill, increases in drug addiction and alcohol usage, and so forth."[1] Their strategy is to assemble "essentially all the systematic information available on homelessness in U.S. urban areas: census counts, shelter bed counts, records of transfer payments, and administrative agency estimates," estimate "similar statistical models using four different samples of data on the incidence of homelessness," and check whether the housing-market relationship holds across all of them.[1] The fourth dataset — an eight-year panel of California county-level AFDC-HAP caseloads — lets them use "standard panel techniques to address the unobserved heterogeneity which is not easily captured in simpler statistical models."[1]

Core Findings

• Homelessness rises with rents and falls with vacancies — across four independent datasets. "For three of these four imperfect measures of homelessness, we find that the incidence of homelessness varies inversely with housing vacancy rates and positively with the market rent for just-standard housing."[1] The result is not an artifact of any one flawed count; it survives being re-estimated on national cross-sections, county cross-sections, and a within-state panel.
• Small changes in the housing market produce large changes in homelessness. In their simulations, "a reduction in the rate of homelessness by one-fourth... could be achieved in the national sample of housing markets by a one percentage point increase in the vacancy rate (from an average of 8.4%) combined with a decrease in average monthly rent-to-income ratios from 17.5% to 16.8%."[1] The abstract's summary: "rather modest improvements in the affordability of rental housing or its availability can substantially reduce the incidence of homelessness in the United States."[1]
• The individual-pathology explanations fit the time trend poorly. The authors show that deinstitutionalization of the mentally ill largely predates the 1980s homelessness surge (inpatient rates fell faster in the 1970s than the 1980s), and that much of the mentally-ill population was "trans-institutionalized" into prisons and jails rather than onto the street.[1] Personal vulnerabilities are real, but they do not track the timing or cross-market variation of homelessness the way housing costs do.
• A rational-choice reading of the extreme lower tail. The consistent negative vacancy effect, positive rent effect, and negative cold-weather effect together "support models of homelessness that emphasize rational choice among the extremely poor."[1]

Relation to the Georgist Case

Like Colburn & Aldern, this is not a Georgist paper and says nothing about land value taxation. Its role on this wiki is as the earliest rigorous evidence — and the first to combine cross-sectional and panel identification — that how many people a community pushes into homelessness is set by the price and availability of rental housing. It anchors the market-level half of Outcome: Homelessness is a housing-cost problem, and since long-run housing costs are predominantly land costs (Knoll, Schularick & Steger), it connects the most visible urban destitution to the price of access to location.

Nuances and Limits

• The count data are imperfect. The authors are candid that all four homelessness measures are flawed ("four imperfect measures"), and that "given the nature of the underlying data, the accuracy of these precise estimates is open to question."[1] Their defense is triangulation: the same qualitative result appears in every dataset.
• Cross-sectional plus panel, but not an experiment. The California panel addresses unobserved heterogeneity, but the identification is still observational, not a randomized or natural experiment on rents.
• Policy reading is joint. The authors' own conclusion pairs supply with assistance: "homelessness may be combated by modest supply policies combined with housing assistance directed to those for whom housing costs consume a large share of their low incomes."[1] The finding is about housing-market conditions, not a claim that any single lever suffices.

Bears On

• Outcome: Homelessness is a housing-cost problem — the original cross-market and panel evidence for the market-level claim.
• Colburn & Aldern, Homelessness Is a Housing Problem — the 2022 book-length successor that reaches the same conclusion with a longer, community-level dataset.
• Hanratty (2017), Do Local Economic Conditions Affect Homelessness? — a later HUD-panel test whose fixed-effects results echo the rent finding.
• Corinth (2017), permanent supportive housing — the intervention-side complication on how far adding housing units moves the count.

See Also

• Outcome: Homelessness is a housing-cost problem
• Outcome: Rising land costs drive poverty
• Narrative: The Housing Crisis Is a Land Crisis
• Economic Rent

Sources

1. John M. Quigley, Steven Raphael & Eugene Smolensky (2001), "Homeless in America, Homeless in California," Review of Economics and Statistics 83(1): 37–51. DOI: 10.1162/003465301750160027. Free full-text working-paper version: eScholarship (UC Berkeley), item 4v61c0ws — used for the four-dataset design, the vacancy/rent findings, the one-fourth-reduction simulation, the deinstitutionalization timing argument, and the abstract's affordability conclusion; full 16-page PDF fetched and read 2026-07-11.