SPATIAL AND TEMPORAL PATTERNS IN SHORT TERM EMPLOYMENT CHANGE WITHIN THE SOUTHERN ONTARIO URBAN SYSTEM

Abstract

"Regional growth theory requires the explicit introduction of two fundamental dimensions - time and space. The time dimension was successfully brought into economic theory with the release of Keynes’ assumption of a constant production capacity. Model building in the field of growth theory of national economics continues to achieve high levels of theoretical sophistication, including the empirical application of rather abstract models. Independent of these striking developments in dynamic analysis was the introduction of the spatial dimension, mainly through the work of Walter Isard in the late 1950's. These two. fundamental innovations, however, failed to be integrated. Growth theory formulated its models for a wonderland of no spatial dimension, and regional science did not bother to introduce the time dimension." (Siebert, 1969, pp. 5 - 6). The importance of incorporating temporal dynamics into building urban and regional planning models is becoming increasingly recognized. Forrester’s (1969) urban dynamics model, which gives a purely temporal, non-spatial simulation model of the city is one example of this line of development. The Lowry (1964) model of urban land use has been given a number of temporal reinterpretations such as the Tomm and Empiric models (Lowry, 1967). A third approach is the work focussing on the spatial transmission and description of business cycle impulses in urban and regional economic systems. The present study falls in line with the last approach. It involves an empirical identification of the variations in the timing and intensity of employment fluctuations existing among cities in Southern Ontario. Economic change or growth in one urban place is viewed as, at least, a partial function of changes taking place elsewhere in the urban system. The structure of urban interdependencies is conditioned by the frictions of distance, by the existence of urban size thresholds and hierarchies and by inter-market, industrial and financial linkages. These spatial relations are examined in reference to growth pole theory and to the literature on economic fluctuations in urban/regional systems. Some spatial considerations in growth pole theory are discussed first, followed by a review of a number of limitations and neglected issues found in the empirical research. Based on this theoretical and empirical review, the study attempts to show how an analysis of urban short run phenomena such as cyclical fluctuations is related to growth pole theoretical constructs and how it is useful in the empirical testing of growth pole processes and in planning applications. A conceptual framework is then outlined, structuring the way economic impulses are generated through national, regional and local mechanisms and transmitted through the urban system and how the impacts of these impulses on urban centres vary in intensity and timing. From this conceptual framework, three analytical procedures for examining certain questions about change in an urban system and for investigating spatial interdependencies in urban short run economic behaviour, are outlined. First, factor analysis as a technique for studying spatial— temporal patterns in the intensity and timing of growth among cities is presented. Second, a model that deals with the decomposition of urban time series data into three components, a long term growth trend, a national cyclical component and a regional component, is presented. Third, a model for testing for spatial-temporal growth trends (polarization trends) at the regional level is developed. Time series data, consisting of monthly industrial composite employment indicies for a five year time period from January, 1968, to December, 1972, for 29 cities in Southern Ontario, are applied to the analytical methodologies. The monthly observations are seasonally, adjusted, using dummy variables and least squares multiple regression. Some concluding statements are made in the final chapter.