Working Paper: “The Long-Run and Intergenerational Effects of the Institutionalization of Redlining”
Abstract: This article identifies the long-run and intergenerational effects of the U.S. federal government’s institutionalization of redlining. Using longitudinally linked federal administrative microdata to track over a million individuals across 1940, 1970, and 2000, our estimation strategy focuses on a city-level Regression Discontinuity Design that exploits an unannounced cutoff in which only cities over 40,000 population in 1930 were redline mapped. We find that redline mapping reduced the future income of Black children born into redline mapped cities by 7% and had a small negative effect on their future housing values. On the other hand, we find that redline mapping increased White individuals’ income by 3% and housing values by 36%. Overall, redline mapping increased the Black-White income gap by 36% and the Black-White wealth gap by 82%. We show that the income effects are diminished by about 70% while the wealth effects persist strongly for the children of those who were redline mapped. We then directly test potential mechanisms for these effects, including education, mortality, household formation, and migration. We find that education explains about half of the income effects, mostly to the benefit of White children, and that redline mapping caused about one-third of the Black-White mortality gap for those born in the middle of the twentieth century. As with income, the channels of education and migration are diminished – but nonetheless persist – into the second generation.
“The Long Run Effects of de jure Discrimination in the Credit Market: How Redlining Increased Crime”, Journal of Public Economics (2023)
Abstract: Today in the United States the welfare costs of crime are disproportionately borne by individuals living in predominately African-American or Hispanic neighborhoods. This paper shows that redlining practices established in the wake of the Great Depression make present-day contributions to this inequity. First I use an unannounced population cutoff that determined which cities were redline-mapped to show that redline-mapping increased present-day city-level crime. Secondly, I use a spatial regression discontinuity to show that redlining influenced the present-day neighborhood-level distribution of crime in Los Angeles, California. Lastly, I identify channels though which redline-mapping influenced crime including increasing racial segregation, decreasing educational attainment and harming housing markets.
John Anders 