An empirical model of residential location and the journey-to-work in a metropolitan area

In this paper we present and estimate a single equation model designed to explain the residential location behaviour of individuals living in a multi-centered metropolitan area. We test the model for heads of households and non-heads of households separately, as well as for the total working population, in order to isolate differences in commuting behaviour between primary and secondary wage earners. The relationships are estimated from 1971 Census, cross-section data using least squares multiple regression. The data pertain to the Toronto Census Metropolitan Area (CMA) and sixty-three designated zones therein.The estimation results reveal that although location-rents prove to be significant in the individual's residential location decision-making process, their effect was limited to discouraging those employed in or near the central business district (CBD) from living close to work. The existence of secondary employment centres did not have the same significant effect in bidding up location-rents. Furthermore, contrary to the standard theory, the results suggest that residential location decisions are made in response to the availability of collective residential opportunities and workers' preferences for specific residential attributes rather than by reference to the “transportation cost—housing cost” trade-off. Among the socio-economic variables, age of the worker is found to be most significant in affecting journey-to-work distance. The results provide some evidence that non-heads are, to a certain extent, more sensitive to urban structural constraints in their commuting behaviour.