Individual determinants of homelessness: A descriptive approach

In this paper, I utilize a novel data set collected in 2012 to investigate the determinants of homelessness intensity among those who are street homeless in Costa Mesa, California. By restricting the sample to only those who are already homeless and introducing the concept of a homelessness intensity measure, I exploit variation in individual characteristics and usage of a Check-In Center to predict the intensity of homelessness (defined as the ratio of actual homeless nights to potential homeless nights). By making use of a novel application of the weighted least squares estimator, I am able to estimate coefficients that are design-unbiased for the population values. By using an instrument, I find that the Check-In Center has a large negative effect on the intensity of homelessness. The self-reported cause of homelessness is a determining factor as well. I also find, as would be expected, that education and the presence of family nearby lessens the intensity of homelessness.