Measuring the Purple Line's Effect on Residential Rents: Spatial Econometrics and Causal Inference in Urban Policy Analysis

Abstract

In this paper, I leverage a unique dataset containing unit-level observations on rental prices for units in Montgomery County, Maryland to build a series of hedonic models using a spatial econometric panel framework. Over the dataset’s time horizon (2015-2018), The Purple Line, a new light rail was announced and its plans revealed, introducing exogenous variation in land prices through a new potential amenity. I explore whether this amenity is capitalized ahead of its construction, as well as the role of spatial spillovers and, importantly, the implications of ignoring spatial effects in the context of applied policy analysis. Specifically, I show that classic methods for causal inference result in starkly different estimates for the “causal effect,” yielding dramatically different policy conclusions. Using a series of spatial econometric panel models, this analysis emphasizes the importance albeit difficulty of interpreting spatial effect parameters in a causal research design. Specifically, I show that the announcement of the Purple Line probably caused a decrease in rents for units nearest the station areas, and this effect is mistakenly estimated as an increase if spatial effects are ignored.

Keywords: Spatial Econometrics, hedonic model, causal inference, spillover effects, land-value capture