Long-term Effect of Redlining Policy on Air Pollution Exposure

3048 

Contributed Abstract 

Paper 

Xiaodan Zhou (1), Shu Yang (1), Brian Reich (2)

(1) North Carolina State University, Department of Statistics, N/A, (2) North Carolina State University, N/A

Shu Yang  
North Carolina State University, Department of Statistics

Brian Reich  
North Carolina State University

Xiaodan Zhou  
North Carolina State University, Department of Statistics

Xiaodan Zhou  
N/A

This study assesses the potential long-term environmental effects of redlining policies (1935-1974) on present-day PM2.5 air pollution levels. Enacted in the 1930s, there are only a few low-quality pre-treatment covariates recorded in survey. Consequently, traditional methods fails to sufficiently account for unmeasured confounders, potentially skewing causal interpretations. Moreover, the time lapse of 75 years between the policy action and the pollution measurement further obscures causal links. By integrating historical redlining data with 2010 PM2.5 levels, our study aims to discern whether a causal link exists. Our study addresses challenges with a novel spatial latent framework, using the income level and percentage of Black population in survey as proxies to reconstruct pre-treatment socio-economic factors. We establish identification of a causal effect under broad assumptions, and use Bayesian MCMC to quantify uncertainty. Our method promises to enhance the validity of causal claims by rigorously adjusting for confounders. Anticipated findings will illuminate the effects of redlining policy on contemporary air quality.

Bayesian causal model|Spatial latent factor|proxy variable|Redlining policy|air pollution exposure|

Section on Statistics in Epidemiology

Causal Inference