Introduction

One of the important areas of recent development in the field of input–output analysis has been the modelling, in a regional context, of linkages between industrial output and household activity. Household consumption has been shown to be a very important component of the demographic-economic system (see, e.g., Hewings, 1986; Batey and Madden, 1981) and one which is rarely accorded the attention it deserves in terms of resources devoted to analysis and data collection. The linkages between industrial activity and household activity are traditionally modelled in input–output analysis by treating households as an ordinary industry which consumes industrial products and produces labour services. Additional rows and columns are added to the interindustry flows matrix in what has come to be called the extended input–output model. A core problem in applications of this model to the regional context has been how to model the impact of newly employed workers on a regional economy. Morrison (1973), for example, treated new workers as in-migrants attracted to the new town of Peterborough in the UK during the early 1970s. Since then, however, the world economy has suffered from two major recessions, and attention has become focussed on the pool of unemployed workers which now exists in many intra-national regions. The so called type IV input–output model has been developed in an attempt to capture explicitly the consumption profiles of employed and unemployed workers (see, e.g., Batey and Madden, 1981; van Dijk and Oosterhaven, 1986).