MEASURING URBAN AMENITIES AND DISAMENITIES:A SPATIAL HEDONIC ANALYSIS IN BOGOTA,COLOMBIA

Mobility and air quality have emerged in recent years as major concerns in Bogota. Even though government authorities recognize that pollution control and congestion reduction are priorities in their policy agenda, there is little local understanding of the monetary benefits from improving urban amenities such as air quality or accessibility to Central Business Districts (CBD).First, the hedonic pricing model (HPM) is used in conjunction with simple GIS and spatial interpolation analysis of PM10 emissions to answer policy relevant questions in a local context such as a) what are the benefits of improving accessibility to a CBD for properties located nearby rapid transit system lines?, b) what are the benefits for those properties located far away? c) How can we approximate monetary benefits from improving air quality conditions? Second, spatial econometric approaches to estimation of the hedonic price function are explored. The use of the Spatial-Lag, Spatial Autoregressive Error and Spatial Components Error specifications are motivated by different error processes that may describe local housing markets. OLS and spatial models all showed the same sign and significant coefficients for the variables of interest of this study; i.e, air quality and accessibility to employment centers. Lagrange Multiplier tests suggest that a spatial process characterized by auto correlated errors is the most likely one in this data, indicating that spatially varying omitted variables may have a significant effect on apartment rent values. Finally, motivated by two processes that were labeled as speculative overpricing and omitted variables, alternative techniques to OLS were used and an asymmetric random error in the hedonic regression was hypothesized. Thus, the regression contains a systematic composed error characterized by the usual zero mean and constant variance idiosyncratic noise and an asymmetric error component. Estimation results from a heteroskedastic frontier model from an apartment rental market suggest the presence of an asymmetric random error whose variance decreases with increased pollution levels. This result would be consistent with an omitted variable process in the rental price formation of this housing market.