Affect Prices of Various Commercial Properties Differently ? A Nonparametric Approach to a Natural Experiment

Abstract: Airports have become crucial for connecting business travelers with the global economy. The 2009 opening of rail rapid transit in Richmond, BC Canada is a natural experiment that enables us to identify how improved access to Vancouver International Airport may be capitalized into commercial real estate prices. Few known airport access benefits studies focus on commercial property values. In assessing comparative statics, our model implies the rail rapid transit opening leads to higher real estate values when the travel time change is sufficiently large relative to the fixed costs of using the rail line; otherwise the effect is non-positive. Our identification strategy consists of two parts. First we construct a nonparametric Fourier repeat sales price index covering each time period using repeat sales observations from 2005-2012 that are orthogonal to the set of observations that straddle the rail opening date. Second, our natural experiment focuses on how travel time changes affect sales price changes for repeat sales dates straddling the date of the rail line opening, while controlling for neighborhood price changes. Using the nonparametric estimation approach of Locally Weighted Regressions (as in McMillen and Redfearn (2010)) to allow for nonlinear effects, we find significantly positive marginal effects between travel time savings and sale price changes of some commercial properties, while for others this relationship is negative. We also find that on average these marginal effects increase with travel time savings increases, which validates the predictions of our model. Falsification tests indicate our identification strategy is sound.