• Email
  • Single Page
  • Print

The Homeless

Over the Edge: The Growth of Homelessness in the 1980s

by Martha R. Burt
Russell Sage Foundation, 267 pp., $16.95 (paper)

The Way Home: A New Direction in Social Policy

by the New York City Commission on the Homeless
(unpublished), 118 pages plus appendices pp.

Down and Out in America: The Origins of Homelessness

by Peter H. Rossi
University of Chicago Press, 247 pp., $9.95 (paper)

Down on Their Luck: A Study of Homeless Street People

by David A. Snow, by Leon Anderson
University of California Press, 391 pp., $14.00 (paper)

The Mole People

by Jennifer Toth
Chicago Review Press, 267 pp., $19.95

Rude Awakenings: What the Homeless Crisis Tells Us

by Richard W. White Jr.
Institute for Contemporary Studies, 333 pp., $24.95

Late in the 1970s Americans began noticing more people sleeping in public places, wandering the streets with their possessions in shopping bags, rooting through garbage bins in search of food or cans, and asking for handouts. By January 1981, when Ronald Reagan took office, a small group of activists led by Robert Hayes and Mitch Snyder had given these people a new name—“the homeless”—and had begun to convince the public that their plight was a serious one. Later that year America entered its worst recession in half a century, and the homeless became far more numerous. At the time, many people saw this as a temporary problem that would vanish once the economy recovered, but they were wrong. Unemployment fell from almost 10 percent in 1983 to just over 5 percent in 1989, but homelessness kept rising.

The spread of homelessness disturbed well-to-do Americans for both personal and political reasons. The faces of the homeless often suggest depths of despair that we would rather not imagine, much less confront in the flesh. Daily contact with the homeless also raises troubling questions about our moral obligations to strangers. Politically, the spread of homelessness suggests that something has gone fundamentally wrong with America’s economic and social institutions.

Because homelessness is both deeply disturbing emotionally and controversial politically, it has inspired a steady flow of books and reports by journalists, political activists, and scholars.1 These publications cover a multitude of different issues, but I will concentrate on what they can tell us about three questions: how much homelessness increased during the 1980s, why it increased, and what we can do to reduce it.


As soon as homelessness became a political issue, legislators and journalists began asking for numbers. The Census Bureau was in no position to answer their questions, because it had always counted Americans by making lists of “dwelling units” and then trying to determine how many people lived in each unit. The bureau had never made much effort to count people living in bus stations, subways, abandoned buildings, parks, doorways, or dumpsters. Nor did it try to fill this gap when public interest in the homeless exploded in the early 1980s. Even the 1990 Census, which attempted a systematic count of people in various kinds of shelters, made only a half-hearted effort to count the roughly equal number of homeless adults who were not in shelters.2

In the absence of official statistics, both journalists and legislators fell back on estimates provided by activists. In the late 1970s Mitch Snyder argued that a million Americans were homeless. In 1982 he and Mary Ellen Hombs raised this estimate to between two and three million.3 Lacking better figures, others repeated this guess, usually without attribution. In due course it became so familiar that many people treated it as a well-established fact.

Widespread acceptance of Snyder’s estimate apparently convinced the Reagan administration that leaving statistics on homelessness to private enterprise was a political mistake, and in 1984 the Department of Housing and Urban Development (HUD) undertook a survey by telephone of social relief agencies and other informants in each large American city, asking them to estimate the number of homeless people in their area. Extrapolating from these responses, HUD’s “best estimate” was that between 250,000 and 350,000 people were homeless in the nation as a whole.4

This estimate was so much lower than Snyder’s that many skeptics assumed HUD must have distorted the data. But when Ted Koppel asked Snyder where his figures had come from, this is what Richard White, the author of Rude Awakenings, quotes Snyder as having said:

Everybody demanded it. Everybody said we want a number… We got on the phone, we made lots of calls, we talked to lots of people, and we said, “Okay, here are some numbers.” They have no meaning, no value.

Nonetheless, Snyder denounced HUD’s numbers as “tripe.” If HUD’s numbers were accepted, he told Koppel, they would “take some of the power away…some of our potential impact…and some of the resources we might have access to, because we’re not talking about something that’s measured in the millions.”

Snyder was right. Big numbers are politically useful. That is why many advocates for the homeless still claim that several million Americans are homeless on a given night, even though no careful study has ever yielded a nightly count that high. White describes the repetition of these inflated estimates as “lying for justice.”

HUD’s estimates were more systematic than Snyder’s, but they were still based largely on adding up guesses. The first careful count I know of came only in September 1985, when Peter Rossi made a survey of the homeless in Chicago—a city where advocates claimed that between 15,000 and 25,000 were homeless on an average night.

Rossi tried to count everyone sleeping either in a public place, such as a doorway or a park, or in a free shelter. Counting the shelter population was easy. To count the street population, Rossi had interviewers search a random sample of city blocks during the small hours of the morning, looking in every space they could reach without encountering a locked door or a security guard. This was potentially dangerous work, so his interviewers worked in pairs and were accompanied by off-duty policemen. The interviewers asked everyone they found, awake or asleep, whether they had a home elsewhere. Nine out of ten said they did. Since Rossi paid his respondents, almost everyone answered his questions.

Down and Out in America describes what Rossi found. Extrapolating from the blocks he searched, he estimated that about 1,400 homeless adults were sleeping on the Chicago streets or in publicly accessible buildings such as bus stations, airports, allnight movie theaters, and restaurants. Another 1,400 adults and children were in shelters. That brought the overall count to about 2,800. When Rossi did a second survey during the winter of 1986, cold weather had lowered the street count and raised the shelter count, but the city’s estimated homeless population was still about 2,500. Since Chicago had almost 3 million residents at the time, these estimates imply that just under 0.1 percent of the city’s population was homeless. That was about the same as HUD’s 1984 estimate for the nation as a whole.

Rossi’s procedures probably missed more than a few homeless people. No early morning street count is likely to find everyone who is homeless. In order to avoid being robbed, assaulted, rained on, or frozen, the homeless often try to conceal themselves, preferably indoors. Some sleep in basements and hallways whose owners make the space available at night in return for daytime work. Some sleep in abandoned buildings, often with makeshift locks on the door. Some spend the night in “shooting galleries” for drug users.

In New York, an early morning street count would probably miss even more people than in Chicago. Manhattan is honeycombed with underground tunnels, most of which were built in the nineteenth century either for the city’s new subway system or for the railroads that converged on the city. Many of these tunnels have been abandoned, but most can still be reached from the streets. Jennifer Toth, the author of The Mole People, explored this labyrinth during 1990 and 1991 accompanied by diverse and sometimes terrifying guides. She concluded that at least 5,000 people lived in the tunnels. An outreach program funded by the Metropolitan Transit Authority in 1990 put the figure even higher, claiming that over 6,000 people were living under Grand Central and Penn Stations alone. About twice that number of single adults sleep in the city’s shelters on an average night.

Because the homeless often try to disappear at night, many investigators interview them during the day, when they make less effort to conceal themselves. The most comprehensive survey of this kind was conducted by the Urban Institute during March 1987 in a representative sample of cities with more than 100,000 residents. Interviewers talked with randomly selected adults in shelters, soup kitchens, and “congregating sites” frequented by the homeless, such as bus stations, parks, and street corners.

Martha Burt, who directed this survey, summarizes many of its findinqs in Over the Edge. Her work suggests that only a third of all single homeless adults slept in a shelter the night before they were interviewed. Among families with children, in contrast, 96 percent had slept in either a shelter or a welfare hotel.5 If these ratios also hold for smaller communities, as they seem to, about 325,000 single adults and 65,000 members of families with children were homeless on a typical night in March 1987.6

Far more people become homeless at some time during a year than are homeless on any given night. In 1990, Bruce Link and a group of colleagues at Columbia University asked a national sample of adults with telephones about their experiences with homelessness. Three percent said they had spent time in a shelter or on the streets during the past five years.7 If we allow for the responses of people who had no telephones, that would mean six or seven million people had been homeless at one time or another between 1985 and 1990. Roughly speaking, at least 100,000 people became homeless every month, while another 100,000 returned to conventional housing.8

Link’s survey suggests that half of those who became homeless got off the streets within a couple of months. Only one in eight remained homeless more than a year. But while most spells of homelessness were quite short, the long-term homeless still account for about half of those who are homeless on any given night.9

We have no good national data on the size of the homeless population before Burt’s 1987 survey. White suspects that the number may not have risen at all. He thinks activists like Snyder and Hayes persuaded liberal journalists eager to discredit the Reagan administration that an age old problem was out of control without having any hard evidence.

When I first read White’s book, I dismissed this theory as ridiculous, on the grounds that city streets looked completely different in the late 1980s from those in the late 1970s. But what we see on the streets often depends more on police practices than on the frequency of destitution. The number of panhandlers, for example, depends mainly on the risk of arrest and how much one can earn from panhandling compared to other activities. Most panhandlers appear to live in conventional housing, and only a minority of the homeless admit to panhandling. Nor is appearance a reliable indicator of homelessness. Rossi’s interviewers rated more than half their respondents “neat and clean.”

Shelter counts provide stronger evidence that homelessness increased. Census data suggest that five times as many people were using shelters in 1990 as in 1980. We do not have national data on the proportion of the homeless sleeping in shelters in either year, but no local study suggests that it rose by anything like a factor of five. 10 It follows that the overall homeless population must have grown.

  1. 1

    In addition to the books discussed in this essay, my shelf of recent contributions includes Gregg Barak, Gimme Shelter: A Social History of Homelessness in Contemporary America (Praeger, 1991); Alice Baum and Donald Burnes, A Nation in Denial: The Truth about Homelessness (Westview Press, 1993); Joel Blau, The Visible Poor: Homelessness in America (Oxford University Press, 1992); Philip W. Brickner, Linda Keen Scharer, Barbara A. Conanan, Marianne Savarese, and Brian C. Scanlan, editors, Under the Safety Net: The Health and Social Welfare of the Homeless in the United States (Norton, 1990); Michael Dear and Jennifer Wolch, Malign Neglect: Homelessness in an American City (Jossey-Bass, 1993); Benedict Giamo, On the Bowery: Confronting Homelessness in American Society (University of Iowa Press, 1989); Benedict Giamo and Jeffrey Grunberg, editors, Beyond Homelessness: Frames of Reference (University of Iowa Press, 1992); Stephanie Golden, The Women Outside: Meanings and Myths of Homelessness (University of California Press, 1991); Julee H. Kryder-Coe, Lester Salamon, and Janice Molnar, editors, Homeless Children and Youth: A New American Dilemma (Transaction Publishers, 1991); Henry Miller, On the Fringe: The Dispossessed in America (Lexington Books, 1991); Karen Ringheim, At Risk of Homelessness: The Roles of Income and Rent (Praeger, 1990); Steven Vanderstaay, Street Lives: An Oral History of Homeless Americans (New Society Publishers, 1992); and James Wright, Address Unknown: The Homeless in America (Aldine de Gruyter, 1989).

  2. 2

    The Census Bureau did try to count people who were visible between 2:00 and 4:00 AM in public places frequented by the homeless, but it did not ask them whether they were in fact homeless. When interviewers do ask this question, at least half those queried say that they live in conventional housing.

  3. 3

    Mary Ellen Hombs and Mitch Snyder, Homelessness in America: A Forced March to Nowhere (Washington: Community on Creative Non-Violence, 1982).

  4. 4

    See “A Report to the Secretary on the Homeless and Emergency Shelters” (Washington: Office of Policy Development and Research, US Department of Housing and Urban Development, 1984).

  5. 5

    These estimates are based on tabulations that Burt made at my request. I discuss them at length in the appendix of my forthcoming book, The Homeless (Harvard University Press). Unless I indicate otherwise, all the numbers in this review come from that book and are documented there.

  6. 6

    Burt herself put the total somewhat higher in an influential paper for the Census Bureau titled “Developing the Estimate of 500,000 to 600,000 Homeless People in the United States in 1987” (see Cynthia Taeuber, editor, Enumerating Homeless Persons: Methods and Data Needs, US Bureau of the Census, 1991). For reasons I discuss in The Homeless, I believe her data imply a one-night count closer to 400,000, but the difference is of little practical importance.

  7. 7

    Bruce Link, Ezra Susser, Robert Moore, Sharon Schwartz, Elmer Streuning, and Ann Steuve, “Reconsidering the Debate about the Size of the Homeless Population,” paper presented at the American Public Health Association meetings in October 1993. Link’s work was the basis for the Clinton administration’s widely publicized claim, in a recently leaked draft report, that seven million people had been homeless between 1985 and 1990.

  8. 8

    My monthly estimate of 100,000 is an “unduplicated” count, in which no individual appears more than once over the course of five years. If we were to include people who became homeless more than once, the monthly incidence of homelessness would be considerably higher.

  9. 9

    To see why this is so, imagine a shelter in which half the beds are occupied by people who stay for a week and half by people who stay for a year. In such a shelter, 96 percent of the new entrants will be short-term residents, but only 50 percent of the beds will be occupied by short-term residents.

  10. 10

    Irwin Garfinkel and Irving Piliavin survey these local studies in “Homelessness: Numbers and Trends,” New York: Columbia School of Social Work, 1994.

  • Email
  • Single Page
  • Print