A discrete choice model of housing selection by low-income urban renters

The purpose of this study was to (1) develop a statistical model that classifies the housing problems of affordability, quality and crowding as elements in a choice set facing low-income urban renters, (2) identify the demographic, socioeconomic and regional factors which influence housing outcomes, (3) use the resulting model to estimate the probability that an individual household faces a particular combination of housing problems, and (4) consider how the receipt of housing assistance alters those probabilities by addressing the question of targeting assistance.

The study used data from the American Housing Survey of 1989 to estimate the model. These were low-income renters who both lived in an urban area and moved from one residence to another during the prior twelve month period. The mean income level of the final sample was $14,336. Sixty-one percent of the sample had affordability problems, twenty-eight percent had quality problems and six percent had crowding problems.

The theoretical framework for the study is discrete choice analysis based on a random utility function. The conceptual framework included the development of seven, binary, logit models. These models represent a sequence of choices which the low-income renter makes when finding housing. The assumed choice hierarchy was affordability decisions followed by quality decisions and finally crowding decisions. The affordability and crowding models performed well; however, the quality model was somewhat disappointing. It appears that either quality is not easily modeled using a binary variable or households perceive quality differently than do the policy makers who establish quality guidelines. The models clearly show that affordability problems constitute the biggest hurdle for the low-income renter. Regional location is a significant factor in estimating the probability of having housing problems. Households in the western region of the United States are most likely to have multiple problems. A major contribution of this study is the focus on housing assistance and how receiving assistance alters the probability of low-income households finding basic shelter.