How to Think about Homelessness

Stephen Eide

Fall 2018

Homelessness is one of the most striking exceptions to America's ongoing urban renaissance. However successful officials in major cities have been on fronts such as economic development, crime, and education reform, they've proven to be much less so when it comes to handling homelessness. Indeed, the situation cities now face regarding homelessness may very well be worse than the one they confronted during the "bad old days" of urban decline.

It's certainly different. "Modern" homelessness really took shape in the early 1980s. Initially understood as an aftereffect of that period's economic downturn, the persistence of homelessness forced governments to recognize that they had an entirely new, intractable problem on their hands. Layered on to the problem of poor, alcoholic, single men — the "pre-modern" homeless who inhabited "skid row" neighborhoods — were an array of new sources and causes of homelessness, such as the vast increase in the number of single-parent families and the deinstitutionalization of the mentally ill. Add in a contraction of the low-rent housing market, and the storm becomes perfect.

Since the challenge of modern homelessness emerged, the preferred solution of most advocates has been permanent housing, meaning subsidized rental units that are not subject to any formal time limits. There is no question that affordable rents must be part of the solution, perhaps especially for homeless people with serious, untreated mental illnesses. But it is just as illogical to offer everyone who falls into homelessness a permanent housing benefit as it would be to offer everyone who loses his job a guaranteed income. Homelessness is not just a housing problem, and our policy solutions should reflect this reality.


America's industrialization throughout the late-19th and early-20th centuries relied heavily on a large pool of transient laborers. They concentrated in cities, and their housing needs were met mainly by the private sector. Lodging houses and single-room-occupancy hotels (SROs) rented out rooms on a nightly basis. Their accommodations were sparse, but they afforded a modicum of privacy at a price within reach of even very low-income individuals. Over the course of decades, neighborhoods with abundant lodging-house and SRO stock declined in size and reputation and became known as "skid row" districts. Examples included West Madison Street in Chicago, Scollay Square in Boston, and, most famously, the Bowery in New York City. By the late 1950s, skid row was largely populated by aged men who were worn down by years of manual labor, suffered from high rates of alcoholism, and for whatever reason remained detached from their families.

Rising levels of prosperity and a reduced need for transient labor reduced skid row's size, but government policy provided the coup de grâce. Urban-renewal programs launched a direct assault on these districts by demolishing and redeveloping buildings that used to house "bums," "vagrants," and "derelicts." Regulations aimed at improving housing conditions made the development of new units unprofitable unless they were backed by government subsidies. Beds in Bowery lodging houses declined from 11,219 to 2,400 between 1949 and 1993; SRO units in the West Madison area of Chicago fell from 22,836 to 2,629 between 1960 and 1980.

Skid row was an ad hoc private-sector solution to the problem of how to house the losers of industrialization. It had many failings, as should be obvious to any modern-day visitor to the original "Skid Row" near downtown Los Angeles. (L.A. made the deliberate decision in the 1970s to hang on to its skid row to preserve its SRO stock and enhance it with social services. Skid Row is reputedly the largest single concentration of homelessness in America, though it's home to only a minority of L.A.'s homeless population.) But the beginning of wisdom on modern homelessness policy consists in adopting a measure of humility toward government's efforts to fill the gap created by skid row's collapse.

For all its grave troubles, skid row did avert homelessness. Although hard historical data on unsheltered homelessness are extremely spotty, many scholars believe that only a very small number of people slept on the streets. The old SROs seem to have been more effective at providing temporary housing than the free emergency-shelter programs of more recent vintage, many of which are funded by the government. In New York City, thousands of individuals continue to sleep on streets despite the city's "right to shelter" law, and they often cite unsafe or unsanitary conditions in its $2 billion shelter system as their reason. In addition to keeping them housed, skid row provided destitute adult men with access to pawn shops and bars, as well as opportunities for day labor. Left without a neighborhood of their own, skid row's former residents, or those who in past times would have lived on skid row, now inhabit public spaces such as parks, train stations, and libraries. Tent encampments have become common sights in Western cities, despite widespread concerns over the threat they pose to the social order and standards of public health.

What now causes homelessness? Being poor is obviously a crucial risk factor, but homelessness can't be explained by poverty alone. Detroit's poverty rate is close to twice that of New York City's. And yet 9 out of every 1,000 New Yorkers were counted as homeless in the most recent Department of Housing and Urban Development figures, compared to only 3 out of every 1,000 Detroit residents. In terms of the economic factors that drive homelessness, what matters is the gap between rent and income, and how much rental housing is available to those with extremely low incomes. Housing quality has been on a steady march upward over the decades. Low-income Americans are likely to be much better housed than in the past, but, not coincidentally, they are also more likely to be unhoused.

Homelessness is a problem in all urban areas, but those in which it's a crisis tend to be vibrant, with one measure of that vibrancy being a highly competitive market for rental apartments. The three cities with the largest homeless populations — New York, Los Angeles, and Seattle — are all generally regarded as healthy and prosperous in urbanist circles. Together, they account for a full 26% (over 140,000) of the nationwide count. In terms of the unsheltered homeless, climate is another decisive factor. About 55% of America's unsheltered homeless population lives in Florida and California, but only 18% of the population as a whole.

But housing-market dynamics and climate cannot fully account for the scale of the current crisis. The skid-row crowd never went away but was rather redistributed throughout the city, as noted above. Added to their ranks were the people hardest hit by certain social changes in the second half of the 20th century, most notably the deinstitutionalization of the mentally ill and the rise of the single-parent family.

Since the mid-1950s, our institutionalized population has dropped by over 90%. We refer to the transition of mental-health care's locus from the inpatient to outpatient setting as "deinstitutionalization," a term that gives a misleadingly deliberate-sounding cast to a process that was anything but. As psychiatrist Jeffrey Geller explained in 2014 testimony to Congress, deinstitutionalization was "not initiated as a considered policy; it was an accident of history." Excessive optimism about the promise of anti-psychotic drugs, a government financing system that incentivized outpatient care over inpatient even for the most severe cases, and legal changes that made it difficult to commit someone against his will were the underlying drivers of deinstitutionalization.

Many regard it as one of the greatest catastrophes in social policymaking in American history, citing as evidence the high rate of mental illness among America's homeless population. HUD's 2017 estimate suggested there are 111,902 homeless individuals with serious mental illness, or 20% of the homeless population. (Others place the rate at about one-third, noting the practical challenge of counting the number of people living under overpasses and in transit systems, as well as that of diagnosing their mental conditions.) In grim confirmation of deinstitutionalization's role in causing modern homelessness, some governments wound up converting former psychiatric wards into homeless shelters.

And mental illness not only precipitates homelessness, it also perpetuates it. As Ann Braden Johnson put it in her 1990 book Out of Bedlam, "[M]aybe it's easier to be homeless if you're crazy." The homeless endure smells, tastes, sensations, and experiences that for many of us would constitute "rock bottom," but then decline to take steps to improve their situation. In former times, many of the mentally ill homeless, quite likely the vast majority of them, would have been committed to long-term care in a state asylum. The seriously mentally ill may not compose the majority of the homeless, but they make up most of the hardest cases — the "service resistant" who persist in refusing offers of help.

Another social development behind the rise of homelessness is the tripling, over the last 50 years, of the number of families headed by single parents (the population as a whole grew by two-thirds). Homelessness policy commonly addresses single adults and families differently, with the latter often seen as being more of a purely economic case. But while it may be true that heads of homeless families do not experience the same rate of behavioral-health disorders as homeless single adults, many would not have wound up homeless had they been members of a family with two parents.

In New York City, two-thirds of the homeless population is comprised of families with children, and around 90% of those families are headed by single mothers. The single-parent family has been blamed for many social pathologies, such as underperforming schools and crime, but its role in the modern homelessness crisis may be one of the most straightforward examples of its social and economic unsustainability. Women were almost nonexistent on the old skid row. And unlike single adults, who constantly cycle in between the streets and shelters, homeless families are almost always sheltered. They usually wind up there after a "doubled-up" housing arrangement with friends or extended family has become insupportable. It should be said that the ability of poor families to tolerate crowded living conditions is not constant throughout times and cultures. New York City's Asian residents are fairly poor and endure a high rate of overcrowding, but they are nearly absent from the local shelter system. Asian families in New York are also more likely to be headed by two parents than families in other minority groups.

Most varieties of homelessness, whether in its pre-modern or more-complex modern form, can be described equally well as cases of "family-lessness." The homeless lack access to the social supports to which most of us would turn if faced with a disruption in our housing situation. In addition to being generally low-income, mentally ill individuals, drug addicts, and alcoholics have trouble sharing housing with others. They are often quarrelsome and erratic, sometimes threateningly so; they tend to burn their bridges with friends and family. They also tend to hoard and have sub-optimal personal hygiene. Thus living with roommates, which is how many single adults cope with an expensive housing market, is impractical for them.

HUD estimates that 17% of the homeless population are "chronic" cases, meaning people with disabilities who have experienced long-term or repeated spells of homelessness. Almost all chronically homeless people are single adults. For families, homelessness is usually only a temporary crisis. At the same time, those concerned with moral hazards and the threat of long-term dependence might view families as a more complicated policy challenge than, say, the mentally ill homeless, since the latter's claims on open-ended grants of government assistance are less controversial. We expect single mothers to become functioning and productive members of society in a way that we do not expect of people with schizophrenia. How should government go about ensuring maximum assistance to poor children without encouraging long-term dependence among their parents? There is of course no simple answer to that question, which confronts us with regard to all social programs involving children.

By the early 1980s, when the face of modern homelessness first presented itself in full, some specific policies targeting homelessness (as opposed to poverty more generally) began to be debated at the federal, state, and local levels. The recession of that era was the most severe since the Great Depression, but the "structural" nature of homelessness was confirmed by its persistence throughout all subsequent economic cycles. Usable nationwide data on homelessness have only been available since 2007. Though the data are generally thought to have improved since then, measuring and tracking trends continues to be a challenge with respect to unsheltered homelessness. HUD, which compiles data provided by localities, reported in its most recent "point in time" count that there were 553,742 homeless people in America. About two-thirds were sleeping in emergency shelters or transitional-housing accommodations; the rest were found sleeping outside, in subways, cars, or other "places not suitable for human habitation."

Four decades into this era of modern homelessness, effective solutions remain few and far between.


Modern homelessness is an urban phenomenon, and cities are often communities of strangers. That goes especially for downtown areas where few people live, which tend to attract the homeless because of the opportunities for panhandling and the availability of public space in plazas, parks, and the like. It is easier to tolerate erratic behavior from someone we know than from a total stranger. When standards of common decency aren't observed, public spaces cease to feel public, and become more like the private accommodations of those who feel like they're at liberty to do whatever they want there, including smoking or shooting up drugs, drinking alcohol, gesticulating wildly, shouting obscenities, sleeping, defecating, and urinating.

The rights revolution of the 1960s made it more difficult for cities to respond to demands to maintain public order, for reasons explained by James Q. Wilson (emphases in original):

Courts are institutions whose special competence lies in the discernment and application of rights. This means that to the extent courts decide matters, the drift of policy will tend to be toward liberty and away from community. The court will, typically, hear a case brought by (or on behalf of) an individual beggar, sleeper, or solicitor. Such an individual rarely constitutes much of a threat to anyone, and so the claims of communal order often seem, in the particular case, to be suspect or overdrawn.

But the effects on a community of many individuals taking advantage of the rights granted to an individual (or often, as the court sees it, an abstract, depersonalized individual) are often qualitatively different from the effects of a single person. A public space — a bus stop, a market square, a subway entrance — is more than the sum of its human parts; it is a complex pattern of interactions that can become dramatically more threatening as the scale and frequency of those interactions increase. As the number of unconventional individuals increases arithmetically, the number of worrisome behaviors increases geometrically.

As a result of this dynamic, homelessness policy, in contrast to anti-poverty policy more generally, often begins with a response to the public demands to address street homelessness. Progressive advocates tend to characterize these efforts as the "criminalization of homelessness." They claim that such efforts are on the rise, although, from a much longer-term perspective, municipal authorities had far more latitude to deal with "vagrants" prior to the 1960s.

The 1972 U.S. Supreme Court ruling Papachristou v. City of Jacksonville struck down a local law that had entrusted police with what the justices termed "unfettered discretion" to address vagrancy, thus "encourag[ing] arbitrary and erratic arrests and convictions." Since laws and policies that target a "status" are constitutionally suspect, cities' legal departments must take pains to craft laws that ban or regulate specific behaviors such as panhandling, sitting or lying on sidewalks, and camping, and in certain times and places. So-called "sit-lie" ordinances have been found constitutional, so long as they refrain from blanket bans of all sleeping in public places at all hours of the day within an entire jurisdiction's limits.

Another emergent source of obstacles to local efforts to manage street homelessness has been the conservative wing of the Supreme Court, as it has worked to protect free speech and religious liberty. Since Reed v. Town of Gilbert, a 2015 ruling about religion-themed billboards in Arizona, many panhandling ordinances have been struck down on grounds that they failed the standard of "content neutrality" articulated by Justice Clarence Thomas. In the view of some legal experts, however, cities may still rely on the "captive audience" doctrine to prohibit solicitations in cases where the person solicited cannot walk away and thus exercise his own right to not listen to someone else's speech.

Spending on social services and housing is necessary to address homelessness, but it will not be enough, due to the aforementioned "service resistance" dilemma. And discussions of service resistance inevitably raise the issue of civil commitment. Thanks to a series of legal developments during the late 1960s and early '70s, practically no one in America can be held in a mental institution against his will for any significant duration unless he poses a threat to himself or others. Throughout the modern homelessness era, numerous press stories have chronicled how unfortunate homeless individuals have been allowed to "die with their rights on," that is, literally deteriorate in public despite numerous outreach attempts to connect them with services. Someone in the grips of florid schizophrenia — someone who, as my colleague, mental-health advocate D. J. Jaffe puts it, "[does not] believe they are the messiah. They know it" — cannot be understood to be exercising moral freedom or rational choice in any meaningful sense. The public would likely support more leeway in civil-commitment standards, as would many advocates on the right and center-left. Nonetheless, civil-liberties groups have so far been successful in thwarting all but the most modest revisions to the modern paradigm on civil commitment.

So far the most influential policy response to modern homelessness is "Housing First," which has enjoyed bipartisan federal support since the George W. Bush administration and has been adopted in some form by dozens of cities nationwide. Housing First was developed in New York in the 1990s by Sam Tsemberis, a psychologist and former outreach worker. Years of working with the homeless convinced him that the best way to help the most challenging, service-resistant cases was to offer them immediate, no-strings-attached access to permanent housing benefits. He reasoned that housing should not be used as a carrot to incentivize compliance with sobriety, employment, or treatment regimens. All that is secondary to housing, which should come first even for — especially for — the least-cooperative cases for whom nothing else has worked.

Housing First is a "harm-reduction" approach to public policy, akin to supervised-injection facilities and sex education. It also, in the eyes of its more zealous proponents, bears a certain resemblance to proposals for replacing the dozens of safety-net programs with a cash-based universal grant or negative-income-tax program. Tsemberis sees himself as an anti-bureaucratic disrupter, but from the left. After years of funding various programs — some of them run by service providers with dubious reputations — and seeing no reduction in homelessness in their communities, many Republican and Democratic officials came to the conclusion that they might as well try open-ended grants for housing assistance that would at least move some people off the streets. The construction of over 160,000 permanent supportive-housing units over the last decade has been fueled by enthusiasm for Housing First.

So has it worked? Several studies have documented Housing First's ability to keep around 80% of its beneficiaries stably housed for one to five years after being placed in a unit. Many also contend that the approach saves money, an argument advanced by Malcolm Gladwell in a 2006 New Yorker article. But while it's true that some "super utilizers" of public services will place less of a burden on the health-care and criminal-justice systems if they are brought in off the streets, that's not necessarily the case for every beneficiary of a Housing First program, and not even every chronically homeless individual.

Writing in the New England Journal of Medicine in December 2016, a group of physicians and researchers, all of whom embrace Housing First on moral grounds, noted that carefully designed studies of the program have not found across-the-board cost savings. The University of Pennsylvania's Dennis Culhane, whom Gladwell featured in his article, has also recently tamped down expectations about its savings. There are even questions as to how much Housing First can reduce homelessness. Studies by Kevin Corinth (now a senior economist at the White House Council of Economic Advisers) and Culhane with other researchers affiliated with the Department of Veterans Affairs have found that communities needed to bring multiple new units of permanent supportive housing online in order to trim the local homeless population by just one person.

The larger problem with Housing First concerns how it has defined down the standard of success in homelessness policy. The literature on Housing First, while affirming its success in keeping recipients "stably housed," is ambiguous as to its success with truly getting people off drugs and alcohol, onto medication, into employment, and toward a state of authentic independence. (Tsemberis himself concedes this.) And though it was originally intended to help the chronic homeless, Housing First's momentum has emboldened homeless advocates to push harder for a permanent housing-based solution for all subsets of the homeless population, including families.

Homelessness-services systems, like all anti-poverty efforts, should be structured like a ladder with the ultimate goal of upward mobility. Harm-reduction advocates always claim that the first few rungs of the ladder are missing, thus having left the hardest cases behind. Though Housing First is officially committed to "independent" living in the community, independence is an empty concept if people are not moving up the human-services ladder. Some people may not be capable of that, but for those who are, being "stably housed" is not an adequate measure of success. An occupant of a public-housing project in a neighborhood with high crime rates and low-performing schools may be "stably housed," but he is far from a state of authentic independence.


During the 2000s, a nationwide campaign formed to end homelessness, with backing from the Bush administration. In the words of the federal Interagency Council on Homelessness, "ending" homelessness means having "a comprehensive response in place that ensures homelessness is prevented whenever possible, or if it can't be prevented, it is a rare, brief, and one-time experience." A truly "comprehensive" approach to homelessness, however, is not realistic, since it would entail root-and-branch reforms to mental-health care, housing regulation, and substance-abuse policy — not to mention the restoration of the two-parent household.

Short of such far-reaching but unrealistic ambitions, four discrete goals for homelessness policy seem appropriate. First, policymakers should seek the integration of the various facets of the homelessness-services system. Otherwise, different groups focused on the problem run the risk of working at cross-purposes. Perhaps the largest challenge of homelessness-services integration lies in the distance between the public and nonprofit sectors. Charitable organizations, many affiliated with churches, have always played a large role in serving the destitute. They provide many forms of assistance directly to the homeless (such as emergency shelters and meal programs), and in doing so may sometimes help facilitate homelessness more than they reduce it. The easier you make it for someone to live under an overpass or on a sidewalk, the more homelessness you're going to get.

This is not an easy discussion to have. Advocates insist that food and other services be provided to the homeless "where they are" and with no strings attached. For their part, charities can argue that the government has not exactly covered itself in glory in dealing with homelessness and, more basically, that when one is in a position to relieve someone else's pain, one should do so. They are likely to get good press for these views. Nonetheless, we will never get anywhere on homelessness unless the challenge is considered a responsibility of the public sector. As much as possible, government should try to coordinate with these groups to integrate their efforts, such as by providing free food at shelters and other locations where homeless individuals might be connected with treatment, employment, and housing. In cities where government provision of homeless services is minimal, integration will require spending money. If a city can't rid itself of encampments, it should assert public control over them. Pinellas Hope, near St. Petersburg, Florida, is an example of a worthy encampment-style program that's run by Catholic Charities but with support from local and state governments.

Second, beyond integration, some degree of "nudging" can play a role in homelessness policy, and can help reconcile it with civil liberties and respect for the social order. As is the case with saving for retirement, nudge theory affirms that there's a right thing to do (move off the streets) even though people can't be forced to do it. This approach has been mocked by many on the right because of its implicit paternalism, but paternalism is the point. There may even be a way in which nudging is a legal necessity. Efforts to regulate vagrancy-related behavior are more likely to be upheld if local authorities are providing alternative accommodations to streets and parks.

The enforcement of laws against low-level crimes should not be the extent of a city's homelessness policy, but the public will demand that it be a component of it. Moreover, street homelessness is not a condition that can be treated on an outpatient basis. Whether it's a question of connecting people with jobs, mental-health care, therapy, or housing, the street homeless need to relinquish their attachment to personal freedom (as they define it) if they are going to improve their lives. As one Hawaiian service provider and supporter of a sit-lie ordinance explained to the New York Times in 2016, "[O]ur homeless outreach teams need to motivate clients to take action." At the same time that governments create disincentives to living in public, they need to incentivize contact with social-service programs, and that will entail funding as well as creativity. Albuquerque's "There's a Better Way" program connects any homeless individual who wants a job with work for the day, but also with social services and shelter. It's a workfare program as well as an enhanced outreach effort.

Of course, incentivizing behavior among the homeless is not easy. Service resistance among the seriously mentally ill homeless is often driven by lack of insight or "anosognosia": Affected individuals are so delusional that they don't believe they're delusional. Going back to the days of the SRO, it has always been clear that the homeless, like the rest of us, place a high premium on privacy. Curiously, some of them feel like they have more privacy living in public spaces than in a shelter. On the one hand, the homeless have low levels of functionality, conventionally understood. They have wound up where they are because they were too weak to overcome forms of adversity that others are generally able to overcome. On the other hand, the homeless are extremely tough and resourceful. They can put up with conditions — filth, eating out of dumpsters, living outside for years on end — that the rest of us could not tolerate.

Third, homelessness policy must involve some housing deregulation. No one seriously disputes that communities in which low-income people have unusually limited housing options will have more homelessness, and that more regulated housing markets have higher costs. A lack of quality data on local housing regulations, and to some extent on homelessness, makes it difficult to tease out the precise relation between the two. Nonetheless, in a 2009 working paper that scrupulously considers the available data and literature, Steven Raphael of the University of California, Berkeley, argues that "regulation may be a substantial contributor to U.S homelessness levels." Outside the homelessness context, housing deregulation has become a popular policy goal among many on the right and center-left. The "YIMBY" (Yes in My Backyard) coalition's political successes have been minimal so far, but their views will continue to attract attention among policy types so long as homelessness crises persist in areas with extremely tight rental markets.

Of course, homelessness is often at crisis levels in these areas for some of the same reasons that rents are unaffordably high: regulations and modern-day labor and land costs. The benefits of increasing the housing stock, by allowing taller buildings or more densely packed construction, would eventually trickle down to the homeless, but it would take time. In a healthy housing market, there are ample used units available to the poor — units that were originally built for the middle class but have deteriorated, and which landlords believe are more economical to rent out in their condition than to pay to renovate. Tenement life may seem unimaginable to many of us now, but, as my colleague Howard Husock has often pointed out, rents were cheap and it was temporary, due to immigrants' upward mobility.

The tacit deals that many immigrant families have made throughout our history — to consume less housing now to secure a better future — cannot be made available to low-income Americans today unless we focus our deregulation efforts on housing quality. If the goal is to restore the lowest rungs of the housing ladder, and to do so as soon as possible, then we as a society may need to accustom ourselves to allowing the poor to live in conditions inferior to today's norm as well as — to evoke the case of the SRO — without amenities that middle-class households consider necessities. In the cities with the most severe homelessness crises, which are all dominated by the political left, it is not clear there is enough political support to allow that to happen.

Finally, homelessness policy needs to emphasize temporary over permanent housing benefits. In addition to being more decorous than old-fashioned terms like "bum," "vagrant," or "derelict," "homeless" points in a more direct manner to a policy solution. But it also abstracts from the underlying reasons for the condition in question. Take ex-offenders, a cohort that's extremely prone to homelessness. Their problem is not that they lack permanent housing; it's that they have a limited work history due to their years behind bars. If the cost of "ending" homelessness is a broad and deep expansion in levels of government dependence, then it's far from clear that that's a deal cities should be making.

As noted earlier, the vast majority of the homeless are not chronic cases. One could even go so far as to say that the large non-chronic or "situational" (or "transitional" or "episodic") component is one way in which modern homelessness differs from the skid-row era. Our policies should reflect this by placing time limits on most housing for the homeless.

Temporary housing benefits can come in several forms: emergency shelters (typically communal-sleeping arrangements in the case of single adults, and private apartment- or hotel-style arrangements for families), transitional housing (programs that last anywhere from a few months to two years and often provide a private room, enriched by supportive services), or time-limited rental vouchers. Such programs have recently fallen somewhat out of favor in homelessness-policy circles. Throughout the Housing First era, the federal government has shifted hundreds of millions of dollars in annual funding from transitional-housing programs to permanent supportive housing. But, in addition to avoiding moral hazard and long-term dependency, temporary-housing benefits allow governments to distribute access to something that is, in the case of tight rental markets, inevitably a fixed resource. (Recipients of a housing benefit without time limits, especially when affordable housing is scarce, are liable to stay a very long time.) Even when dollars are available to fund permanent housing for the homeless, controversies over siting can stall the placement rate even as more homeless people continue to flow into the system. This is often glibly dismissed as "NIMBYism" — from the acronym for "Not in My Backyard" — but the notion that public authorities should be able to build shelters or permanent residences anywhere, without any neighborhood input, is not compatible with a sincere respect for local self-government.

High-quality temporary-housing programs, such as those operated by the Doe Fund and the Bowery Residents' Committee in New York, or Central City Concern in Portland, Oregon, may be in limited supply in some jurisdictions and may not be "scalable." When governments fund these programs, performance-based contracting that carefully monitors client outcomes is essential, so as to distinguish between strong, weak, and mediocre service providers. But in general, identifying skilled policy entrepreneurs and supporting them with public resources is essentially the charter-school model; there's no reason why it can't work just as well for the homeless.


In an odd way, the greatest obstacle to effectively addressing the challenge of homelessness is the view that homelessness should be tolerated rather than addressed at all.

Progressive activists in these debates often allege that the public's attitude toward homelessness is rooted in an inability to handle the truth about extreme poverty in America. But this is a distortion. At the root of popular intolerance of homelessness is the view that government and the public need not countenance a miserable state of affairs that is neither good for the public nor for the homeless themselves. There is something profoundly wrong with a city where vast swathes of downtown are not places where most people would feel comfortable walking with a child.

As a constitutional matter, cities are required to tolerate homelessness. As a moral matter, both policymakers and the public must act to address it. The role of political leadership is to affirm, inform, and strengthen the public's intolerance of homelessness, which is a vital resource in the struggle for a solution. Only by viewing homelessness as a public problem to be solved can we truly begin to help those in need on the streets of our cities.

Stephen Eide is a senior fellow at the Manhattan Institute and contributing editor of City Journal.


from the


A weekly newsletter with free essays from past issues of National Affairs and The Public Interest that shed light on the week's pressing issues.


to your National Affairs subscriber account.

Already a subscriber? Activate your account.


Unlimited access to intelligent essays on the nation’s affairs.

Subscribe to National Affairs.