Counting the uncounted
Homelessness is hard to measure because the people it describes are, by definition, hard to find. In the United States, local volunteers and staff count people in shelters and on the streets on a single night each January. The 2024 count found about 771,000 people, the highest since national counts began in 2007. Part of the recent rise came from migrants arriving in a few large cities and from disasters, but much reflected rising rents. Counts in other countries use different definitions. Finland, for instance, includes people staying temporarily with friends or relatives, which makes its numbers look larger than an American-style count would.
Different paths
People who become homeless are not one group. In a 1998 study of shelter records in New York City and Philadelphia, Dennis Culhane and Randall Kuhn found three patterns. Most people who used shelters stayed briefly and did not return, typically after a job loss, a family breakdown or an eviction. A second group cycled in and out. A small, chronically homeless group, often struggling with mental illness, addiction or disability, stayed for long periods and used a large share of shelter beds. This finding matters because the help each group needs is different: a deposit and a few months' rent can end one person's homelessness, while another needs years of support.
Why some places have more
A common view holds that homelessness is mainly caused by drugs, mental illness or poverty. Those are real and serious, but they do not explain why rates differ so much between places. West Virginia has severe opioid problems and Mississippi has high poverty, yet both have low homelessness rates, while wealthy coastal cities have very high ones. In their 2022 book, Gregg Colburn and Clayton Page Aldern argued that regional homelessness rates track housing costs and rental vacancy rates more than drug use or poverty.
A useful image is musical chairs. Individual vulnerabilities help decide who loses a seat when the music stops, but the number of chairs decides how many people end up without one. Critics of this framing say it can understate how much addiction and untreated mental illness shape life on the street, particularly for people who refuse shelter, and that cheaper housing alone will not reach them. Both points can be true at once.
Housing First
For decades many programs required people to become sober or enter treatment before they could get housing, on the theory that stability had to be earned. In 1992 the psychologist Sam Tsemberis founded Pathways to Housing in New York on the opposite idea. Housing First gives people a home without first requiring sobriety or treatment, then offers support that tenants may accept or decline. The reasoning is that recovery is hard on the street and easier behind a locked door.
The idea has been tested carefully. Canada's At Home/Chez Soi trial, run from 2009 to 2013 in five cities with more than two thousand participants, randomly assigned homeless people with mental illness to Housing First or to usual services. Those in Housing First spent far more time stably housed. But improvements in health and addiction were smaller and less consistent than gains in housing. Supporters reply that housing was the goal and that dignity matters on its own; critics say that a program judged by housing alone may leave deeper problems untouched.
Finland adopted Housing First as national policy in 2008, buying and converting buildings into permanent flats and replacing many shelters. Its count of homeless people fell from around 18,000 in the late 1980s to under 4,000 by the early 2020s, and long-term homelessness fell sharply. Skeptics note that the decline began well before 2008 and that Finland's small, well-funded welfare state differs from most.
Public space and the law
Encampments raise hard questions about public space. Residents want parks and sidewalks to be safe and usable; people with nowhere else to sleep need somewhere to be. In City of Grants Pass v. Johnson in 2024, the US Supreme Court ruled that enforcing camping bans is not cruel and unusual punishment, even where shelter beds are lacking. Some cities welcomed the freedom to clear camps; advocates warned that fines and arrests make it harder for people to find work and housing.
Worked example: the encampment by the river
A city council faces an encampment of about eighty people along a river path. Some residents demand it be cleared. A service provider reports that most of the campers have been homeless less than a year, while a dozen have been on the street for many years.
Using the level, the council could separate the groups. For the newer arrivals, quick rent help and job referrals may end homelessness within months. For the long-term group, a Housing First program with optional support is likely to keep them housed, though it may not cure addiction. Clearing the camp without either would satisfy the law after Grants Pass, but the musical chairs image predicts that most people would reappear elsewhere if the city has too few cheap homes. The council's real choice is how to add chairs while keeping public space shared.
