Spatial Dependence and Neighborhood Effects in Mortgage Lending: A Geographically Weighted Regression Approach

Current research on mortgage lending disparity is mostly based upon the process-based approach on discrimination, and the outcome-based approach on lending disparity and redlining. In recent years, the outcome-based model has received much attention, particularly on the relationship between intra-metropolitan geography and mortgage lending outcomes. From a policy perspective, the theoretical and empirical evidence on lending disparity is of great importance. However, there exists a mismatch between theoretical models, which focus on racial preferences, and empirical studies, which are essentially reduced form without adequate information. Besides, it may be difficult to unravel the effects of neighborhood race and other attributes. So far most studies conducted with the Home Mortgage Disclosure Act (HMDA) data ignore determinants of geographic variations in lending outcome, or simply attribute them to local variations in risk. This study intends to investigate spatial dependence and neighborhood effects of mortgage lending disparities in the Southern California Five-county Region. In so doing, it assesses indicators of primary mortgage market activity and their determinants for the region as a whole and for the sub-regions inside it. The study compiles data from the 2002 HMDA and the 2000 U.S. Census to undertake a variety of analyses, including computation, assessment, and mapping of social-economic characteristics, as well as home mortgage origination, denial rates, and secondary market purchase rates by census tracts among sampled areas and population cohorts. Cluster analysis on those social-economic and mortgage parameters show distinctive patterns of spatial clustering among tracts across the region. In observing these blueprints of spatial dependence, the study further undertakes a geographically weighted regression (GWR) to analyze the spatial non-stationarity of the determinants of variability in primary market loan denial rates across locations for the year 2002. The modeling result reveals that significant spatial non-stationarity exists between mortgage denial rates and the social-economic determinants. Specifically, the study finds that those census tract-level attributes, including income, population, age, racial composition, housing stock, etc., show significant and varying impacts on mortgage denial rate pattern by spatial clusters. In particular, higher values of spatially varying coefficients on racial composition on traditionally underserved areas, such as south-central Los Angeles, and central cities of outer counties, shed lights on the concerns of redlining. The study concludes that mortgage lending pattern is better understood by the geographically-weighted model than traditional Ordinary Least Square (OLS) regression approaches on lending outcome, which ignore the spatial correlation among local determinants.