multiplier process is initiated that converges towards a higher level of regional production and
employment. For a textbook presentation of economic base theory, see for instance Treyz (1993).
Any operational comprehensive model of urban and regional development has to take into account
the interdependency between location decisions of firms and households. Most of the existing large
scale model formulations are according to the basic idea in the Lowry model, where the residential
location pattern responds to changes in the spatial configuration of basic sector activity, initiating the
well-known multiplier process. Contrary to this the IMREL model (Integrated Model of Residential
and Employment Location) defines residential location decisions to be the driving force of the
development within an urban area, see Anderstig and Mattsson (1991). As a result of improved
mobility, households are argued to be less dependent on work location, while location decisions of
firms are influenced by the accessibility relative to labor supply, or, in other words, the recruiting
potential to labor.
In this paper, however, we will not enter into a discussion of what is a reasonable sequence in the
process towards a new location pattern in the study area. One important component of such a
process is nevertheless how the spatial configuration of local sector firms relates to the residential
location pattern and household shopping behavior. This is the subject of our paper. We neither focus
on location decisions of basic sector firms, nor on how the residential location decisions of
households relate to job market accessibility or retailing facilities.
Unlike the set of large-scale models in the literature, our approach is not restricted to urban, or
metropolitan, areas. Rather, we take on a more macroscopical view of the geography, as we focus
on a regional perspective, potentially including several urban areas. In this respect our approach is
corresponding to the modified version of the Lowry model that is presented in Thorsen (1998). In
Thorsen (1998) this part of the model is, however, introduced on an ad hoc basis, ignoring a set of
relevant aspects. The main ambition of this paper is to offer a refined and theoretically more
satisfying specification of the spatial configuration of local sector production, based on the shopping
behavior of households.
To be more precise, a crucial part of our construction is to model the spatial dispersion of the
fraction between employment and labor in a region with a central business district (CBD). Letting E
denote employment and L labor, we claim that the fraction E/L, viewed as a function of the traveling
distance d from the CBD can be expected to trace a graph similar to the one shown in
Figure 1.