Measuring Population Estimates of Housing Insecurity in the United States: A Comprehensive Approach

Abstract

This paper develops a simple comprehensive housing security scale based on a seven dimension definition of housing security set forth by Cox et al. (2017). We compare our scale to other common measures of housing insecurity and find that failing to use a comprehensive, multidimensional measure could result in substantial bias in prevalence rates of housing insecurity. We also find that while the categories overlap, they do not do so perfectly, such that one dimension, like housing affordability, can capture, or represent, all other dimensions. Location also seems to matter in expected ways. In particular, rural, exurban, and central city locations experience the most housing concerns across domains. Moreover, we find that failure to capture housing insecurity along a multidimensional scale might undercount housing-insecure households in certain locations. Finally, using the housing insecurity scale we develop, we find that single households, poor households (i.e., income less than two times the poverty line), black households, Hispanic households, undocumented immigrants, and less educated individuals experience more severe forms of housing insecurity. In addition, we find that older adults are also more likely to experience low housing security. This provides some validation that our measure is trending with well-established poverty measures.