Modeling Homelessness
Homelessness is a ubiquitous issue in the modern world. Although focused in more urban areas, there are homeless individuals in every state. Anecdotally the public’s idea of how people end up being homeless leans more to the feeling that those individuals are: lazy, drunken, mentally ill, former convicts, illicit drug users, and or degenerates. In published academia, research on the subject says the causes of homelessness are economic in nature.
Coming from our literature review we gathered economic data to come up with a model. Out of the data we gathered we found five variables that are influential drivers of homelessness. Although there may be individual reasons for becoming homeless, this macro-approach can help policymakers understand how to approach solutions. Since there are multiple causes of homelessness, addressing it takes a multifactor approach. There are many ways governments and nonprofits address homelessness. The efficacy of their programs is beyond the scope of this paper.
The five influential variables are: the Gini index value, average rent prices, violent crime rate, housing supply, and the percentage of marginally attached workers.
● Gini Index: The spread of wealth among a given population. A higher Gini index value the higher the spread of statistical dispersion of wealth.
● Average rent prices: The average rent price that is listed in Zillow.
● Violent crime rate: The rate of violent crime per 1million people.
● Housing Supply: The number of houses that exist in a given area.
● Percentage of marginally attached workers: The number of workers that are marginally attached divided by the total of workers in the given population.
More important than the β values of our model, is the number in relation to zero. Positive values indicate that as they rise, homelessness will rise. As there is an increase in the Gini index, average rent prices, violent crime, and percentage of marginally attached workers there is a higher rate of homelessness. A negative value indicates that as there is a fall, then there is an increase in homelessness. The only negative β value was attached to the housing supply variable.
This is the executive summary from a paper I co-authored with Nate Allen, Ben Phan, Gabriel Nogueras, Hadly Bringhurst, Jan Otrusinik, and Jessica Wolford. The full paper and citations is available upon request.
