How We Can Tackle Housing Insecurity through the Power of Data Science

Manhattan, New York buildings

The New America team has been driving an effort to visualize the scale and breadth of housing instability and displacement across the United States, while also telling the stories of communities impacted by this loss. Initially, The Rockefeller Foundation connected them to DataKind so that they could compile a Housing Loss Index, pulling in normalized eviction and foreclosure data, which ranks U.S. counties by the severity of housing loss in the 5 year period between 2014 and 2018. This is the first attempt to consolidate these two leading indicators of housing loss into a single measure, and so will provide a much more comprehensive picture of housing instability than either eviction or foreclosure data can in isolation. They will visualize this index as a national heat map.

In addition to the national picture, with DataKind onboard, they began to work with local partners in three cities across the U.S. – Indianapolis, Phoenix, and Winston-Salem – to develop deep-dive case studies of housing loss and insecurity in those cities down to the census tract level. When Covid-19 hit and the instability crisis only became amplified, this work only became more important. The Deputy Mayor of Indianapolis had allocated $15 million of the city’s CARES funding to housing assistance and quickly had to decide which neighborhoods in his city most urgently needed to relief funds. 

With DataKind’s support, the New America team was able to create a census tract level heat map for Indianapolis that showed where evictions, mortgage foreclosures, and tax foreclosures had been happening over the last five years, pull out census tracts that were especially hard hit and where both evictions and foreclosures were high, and also demographic and socioeconomic predictors of housing loss – all within 24 hours.