Figure 1
Fractions of Employment/Labor as a
function of traveling distance d to CBD
In the following sections we will
demonstrate that a function of this
kind can be argued to be a result of
two competitive forces. First, a
concentration of local sector activities
in a city, or a central business district,
results from a combination of
technological scale economies and specific kinds of external economies of scale that generate co-
location of local sector firms (see for example Quigley (1998)). Contrary to the Weberian tradition in
location theory, however, we take into account that economic activities are space-consuming.
Hence, the center is not concentrated to one single point in the geography, but the various kinds of
scale and agglomeration economies can be expected to fall off rapidly at a certain distance from the
city center. This explains the first part of the curve in Figure 1. The exact form of this part of the
curve of course depends on the size and dispersion of a specific city. At some distance from the city
center, however, a second principle can be expected to start dominating the picture. Like for
example in de Palma et al. (1994), it can be argued that centrally located stores set lower prices and
offer a greater variety of consumer goods than more peripherally located stores. In this paper we
focus on the price aspect. Price reductions might be due to scale economies, externalities and
competition effects. The price reductions, however, are counteracted by an increased traveling cost
as the distance to the CBD increases. The price reductions are not constant, but can be expected to
exhibit a widely distributed probability distribution over a large class of different firms and services.
The paper will be organized as follows. In Section 2 we consider the situation where the CBD is
reduced to a single point, and emphasis is put exclusively on the balance between (consumer) savings
versus generalized traveling costs. We model these effects starting from a simplified case and then
gradually adding on more structure to the model. In Section 3 we suggest how one can incorporate
the spatial component of the agglomeration part. In Section 4, we generalize the construction to the
case where there is more than one CBD. In Section 5 we use empirical findings from a region on the
western coast of Norway and try to calibrate our model with respect to these observations. Finally in
Section 6, we offer some concluding remarks.
2. Consumer savings versus traveling costs
Traditionally, urban economic models are based on the assumption of a monocentric city center, see
for example Fujita (1986) for a survey. To a certain degree this will also be the case in this paper.
We do no attempts to explain the spatial configuration of central places, and we consider the location
of a city center to be predetermined. As mentioned in the introduction, agglomerations of local
sector activities in a city center is typically argued to be a result of specific increasing returns and
external scale economies. External scale economies, or agglomeration economies, are traditionally
divided into urbanization and localization economies (Isard1956). Urbanization economies relates to
the overall economic activity in an area, while localization economies reflect interdependencies
between firms that supply/produce similar goods or services. Through the emergence of the New
Economic Geography there has been an increased focus on scale economies to explain the spatial