TITLE : ECONOMETRIC ANALYSIS OF REGIONAL SYSTEMS: EXPLORATIONS IN MODEL BUILDING AND POLICY ANALYSIS *

In today's complex and volatile economic climate, decision makers in both the public and private sectors are growing ever more dependent on economists for a sense of what the future holds in store. Businessmen and public officials alike need answers, on a regular basis, to such questions as

"What is the likely level of demand for my firm's produet?" "How much will my town be able to raise in taxes next year?" "Should we build a new elementary school in our community?" "How high will prices and unemployment be in the future?"

Economists have approached such questions in several ways, ranging from ad hoc "seat of the pants" judgments to the construction of systematic economic "models."

ln recent years, sophisticated models, employing time series and cross-sectional data, have proved invaluable for forecasting on a national scale. Input-output models (pioneered by Leontief) and econometric models (such as those constructed earlier by Tinbergen, Klein, and others) have been effectively employed to forecast national growth trends and to analyze structural problems in the U.S. and other economies.

But it is only in the last ten years that economists have begun to grapple with the problem of making economic predictions for urban and regional economies, and it is to this more narrowly focused field that this book addresses itself.

Like their national counterparts, regional economic models have progressed from ad hoc methods to more formal models. The principal types, economic base, input-output, and econometlie, are analyzed in Chapter II.

Of these techniques, the econometric have proved the most effective for national forecasting. The bulk of this monograph explores the application of these techniques to specific regional problems.

In Chapter III, I summarize a large-scale eeonometrie model for the Philadelphia region. The model, based on time series data, makes forecasts for output, employment, wages, prices, income, government activity, and other economic aggregates. In Chapter IV, the usefulness of the model is examined. Tests of its accuracy in forecasting the business cycle are presented along with a set of policy-related simulation experiments. These latter include impact analyses of the 1973 oil "crisis," revenue sharing proposals, loeal tax policy, "no growth policy," and an examination of metropolitan decentralization. Thus, the usefulness of regional econometric models for policy analysis is demonstrated.

In all, this volume brings together my efforts since 1967 in the area of macroeconometric forecasting for metropolitan regions. The work began with my doctoral dissertation at the University of Pennsylvania which was completed in 1969. Further work was initiated in 1971 and in 1973, the Philadelphia Region Econometric Project was launched under the auspices of the Economies Research Unit of the University of 'ennsylvania. The Projeet brought together economists from the business and the government communities in an effort to make regular and accurate forecasts for the Philadelphia Standard Metropolitan statistical Area. Funds generated from the Project allowed for the instruction of the model outlined in Chapter III and for the policy simulations in Chapter IV.

Hopefully, this work will provide a modest starting point for scholars, private economists, and government officials who want to build and use regional econometric models for forecasting and policy analysis.