left homeless when the music stops.
grams that offer the scarce goods risk reallocating homelessness. Program partici-
pants are less likely to become homeless, but those moved back in line or displaced
from the queue may have been placed at increased risk. In a sample of families in
shelters in New York city, two factors predicted receipt of subsidized housing:
length of stay in shelter and assignment to a relatively small, nonprofit shelter
rather than a congregate shelter or a welfare hotel (Shinn et al., 1998). Both factors
signaled coming to the head of the housing line. "Months in shelter" reflected fami-
lies' waiting time in that line; the success of the nonprofit shelters in obtaining sub-
sidized housing for tenants reflected targeted advocacy on behalf of their families
(queue jumping). Either way, the overall prevalence of homelessness was not
changed by this reallocation of homelessness between those who were lucky
enough to have advocates or durable enough to wait their turn in the shelter system
and those who were not.
of the queue, others on the verge of homelessness may be moved back and their
risk elevated. Further, if entering shelter is seen as the quickest, most certain route
to subsidized housing, shelter entry may be promoted by the promise of queue
jumping.
preventive services experiences less homelessness than a control group, may not
demonstrate net prevention (overall reduction in incidence or prevalence) if home-
lessness has merely been reallocated. At the individual level, homelessness has
been held at bay for program participants, but at the population level, no net reduc-
tion in homelessness has occurred. Because overall prevalence rates are very hard to
measure accurately and are influenced by many factors unrelated to the operation of
a particular program in a particular area, accurate measures of reductions in the
prevalence of homelessness and unassailable attribution of observed changes
to intervention programs are both unlikely. Evaluators must instead consider what is
in effect the ecological null hypothesis--that homelessness has merely been reallo-
cated--on whatever logical or empirical grounds are available. This is most plausi-
ble when the evaluated program involves advocacy for or assignment of existing
resources to particular groups. Still, even where the reallocation hypothesis seems
persuasive, the program may show that net homelessness would truly be prevented
if critical resources were more widely available.