Early-life residence in mortgage-disinvested neighborhoods and mortality: a novel analysis of holc grades and ipums MLP-numident data

The National Institute on Aging awarded The Research Foundation For The State University Of New York $452,409 on September 1, 2026, under the Aging Research program (CFDA 93.866) to examine the association between early-life residential exposure to Home Owners' Loan Corporation mortgage disinvestment in the 1930s and mortality outcomes among the 1930–1940 birth cohort.

The research addresses a gap in individual-level analysis of how legal HOLC grading and duration of exposure to HOLC-disinvested neighborhoods correlate with long-term mortality. Prior aggregate-level studies found associations between historical HOLC grade D status and current resident mortality, but most current residents were not exposed to legal HOLC grading when it occurred. The project overcomes a methodological barrier: enumeration districts from the 1940 and 1950 U.S. decennial censuses are not easily mappable. The researchers will create accurate 1940 enumeration district map files for all New York City enumeration districts and new 1950 enumeration district map files for New York City and 1,174 non-New York City counties, linking individuals between the 1940 and 1950 censuses. New York City serves as the pilot study because it was the nation's largest city in 1940.

Performance occurs in Albany, New York. The period of performance runs from September 1, 2026, through August 31, 2028. This is an exploratory R21 Project Grant, which is not renewable.

Abstract

This project examines how early-life exposure to mortgage disinvestment in the 1930S or holc (home owners' loan corporation) grading (when IT began) and mortgage-disinvestment duration are associated with mortality among the 1930-1940 birth cohort. Aggregate-level studies find an association between a neighborhood's historical holc grade D status (when IT was illegal) and the mortality of its current residents. However, most people living in these historically disinvested areas were not alive when holc grading was legal and thus not exposed to IT. Formerly holc grade D neighborhoods also have been transformed by other factors since the 1930S (e.g., urban renewal, gentrification). Almost no research on the mortality consequences of legal holc grading: (a) examines individual-level data; (B) considers early-life exposure to legal holc grading in focal and adjacent areas; or (C) considers duration of exposure to holc D graded areas across the life course. Researchers are hampered because individual geographic locations (i.e., enumeration districts (EDS)) from the 1940 and 1950 U.S. Decennial censuses are not easily mappable. While digital ED maps exist in 1940 for some cities, they do not contain EDS with large apartments or group quarters and are based on grayscale maps even though original ED maps are in color, which impedes researchers from precisely measuring individual exposure to holc grades. Digital ED map files for 1950 do not exist. We propose to overcome these limitations in this exploratory R21 project. We focus on new york city (NYC) because IT was the nation's largest city in 1940, making IT an ideal pilot study. First, we create accurate 1940 ED map files for all NYC EDS; new 1950 ED map files for NYC and specific locations of individuals linked between 1940 and 1950 censuses (in 1174 non-NYC counties); construct holc scores based upon the grades for those 1940 and 1950 ED locations; and construct a novel adjacent ED holc measure in 1940 (aim 1). Second, we link holc scores to: (a) ipums MLP-numident (MLP) individual date-of-death data (between 1940 and 2007) from the national archives' social security numident file and (B) individuals' 1940 and 1950 U.S. Census records (aim 2). We also calculate end-of-life holc scores for residence at the time of death from MLP (aim 2). Third, we estimate models predicting age at death as a function of early-life focal holc scores, adjacent holc scores, and holc exposure duration, accounting for the spatial autocorrelation of holc scores and outmigration selectivity (aim 3). The proposed project documents whether 1930S legal holc grading associated with historical mortgage disinvestment in early life and holc-grade exposure over the life course are associated with individual-level mortality. IT will generate novel, public-use data, and the methodology can be used in a future R01 application for all holc-appraised cities lacking 1940 and 1950 ED maps. Knowledge generated from this work will help health practitioners and public health professionals identify populations living in mortgage-disinvested neighborhoods who may be at elevated risk of premature mortality.