most such programs probably prevent some homelessness. But what they do, indis-
putably if in undocumented ways, is help some poor people manage their depriva-
tion a little better.
their own ways, difficult to evaluate. On the surface, at least, they seem expensive
and no more demonstrably efficient than the indicated programs we have criti-
cized. Even so, the evidence to date suggests, above all, that the most effective
levers for homelessness prevention are instruments of housing and income. Writ
large in the form of housing, employment, income maintenance, and tax policy,
such broad programs would affect the many rather than the few and lift vagrant
boats on the flood tide. A selected strategy like subsidies for households with
worst-case housing needs (akin to what Skocpol [1991] calls "targeting within uni-
versalism") would not solve the problem of eligibility thresholds that arises in all
programs that are not absolutely universal, but it would reach a high proportion of
those at risk and would, we believe, markedly decrease homelessness.
at the end of the last century.
attained does it make sense to consider other objectives. Passage of legislation
number of units renting for less than $300 (adjusted for inflation) decreased by 13% from 1996 to 1998,
leading to a loss of 950,000 such units; (3) cuts in federal support for affordable housing led to a drop
of 65,000 in the number of HUD-assisted households from 1994 to 1998; and (4) private owners are
dropping out of the HUD-assisted project-based subsidy program.
(with amounts adjusted for inflation) than in any year since 1979, the first year for which comparable
data are available (Greenstein, Prim, & Kayatin, 2000). See the website of the Center for Budget and
Policy Priorities <http://www.cbpp.org> for a fuller discussion. Whereas child poverty decreased from
1995 to 1997, the proportion of children in families in extreme poverty (below half the poverty line) in-
creased (Sherman, Amey, Duffield, Ebb, & Weinstein, 1998). Work is no protection against housing
difficulties. According to the 1997 American Housing Survey, 3.9 million households living in
unsubsidized units had incomes below 30% of the area median, despite earning at least the equivalent of
the minimum wage for 40 hours per week, year round. Over two-thirds of these households paid at least
30% of their incomes for housing, and one-quarter paid over 50% (Joint Center for Housing Studies,
2000, p. 23). Twombly, Pitcoff, Dolbeare, and Crowley (2000) reported the number of hours per week
of work at the minimum wage, or alternatively, the full-time "housing wage," needed to afford a
two-bedroom rental at the fair-market rent in every community in the country. The fair-market rent is a
low-average rent calculated by HUD, and the authors used the HUD standard that no more than 30% of
income should go to rent. They also showed the proportion of renters unable to afford the fair-market
rent for units of various sizes. For example, in San Francisco, a person earning the California minimum
wage of $5.75 per hour ($.60 above the federal minimum) would need to work 195 hours per week to
afford a two-bedroom unit; 56% of actual San Francisco renters cannot afford a unit of this size. Even in