Homeless Shelters for Single Adults: Why Does Their Population Change?

Using data on the time series of entries, exits, and censuses for women and men in single-adult homeless shelters in New York City, this article examines correlates of shelter population change. Findings sometimes weakly support conventional wisdom about correlates of population change but usually run counter to it. For example, findings suggest that prosperity and good weather reduce the number of adults in shelters but that the prosperity effect is questionable and the estimated weather effect is too small to generate by itself the observed seasonal fluctuations in the number of homeless people on the streets. Placing adults from shelters into permanent subsidized housing reduces the shelter census but also increases the number of adults entering shelters. Changes in homeless shelter capacity are not found to change the homeless shelter census for women but do appear to induce some changes in the shelter population for men.