| Abstract: | ABSTRACT. Recent research suggests that in nonmonocentric cities compensation for commutes takes the form of both lower housing prices and higher wages. This paper develops a random utility model that predicts the probability of an actor choosing to commute between each residence and job in a metropolitan area conditional on the observed location of housing units and job sites. The model allows commuting time, origin-specific amenities, land prices, destination-specific amenities and wages to influence actors' choices. We estimate the model using maximum likelihood and generalized least squares techniques and data on commuting between each of 38 origin and 15 destination jurisdictions in the Tokyo metropolitan area. The empirical results show that, all else equal, a one percent increase in commuting time reduces the probability that a route (origin-destination combination) will be chosen by almost five percent. Origin-specific amenities are not completely capitalized into land prices and destination-specific amenities are not completely capitalized into wages. Desirable residential amenities include school quality and a low ratio of day to night population. Desirable workplace amenities include a large share of white collar jobs and a high density of employment. |
