Spatial Distribution of Retail Sales
Department of Finance, College of Management, National Yunlin University of Science & Technology, 123,
Section 3, University Road, Touliu, Yunlin Taiwan 640
R. KELLEY PACE
Department of Finance, E.J. Ourso College of Business Administration, Louisiana State University,
Baton Rouge, LA 70803-6308, USA
E-mail: firstname.lastname@example.org; www.spatial-statistics.com
We examine the distribution of sales for a retail chain in the Houston market using a spatial gravity model.
Unlike previous empirical studies, our approach models spatial dependencies among both consumers and
retailers. The results show that both forms of spatial dependence exert statistically and economically signiﬁcant
impacts on the estimates of parameters in retail gravity models. Contrary to the suggestions of Gautschi (1981)
as well as Eppli and Shilling (1996), our results show the importance of the distance parameter in retail gravity
models may be greatly understated. Thus, ignoring spatial dependence may lead to overestimation of the
deterministic extent of trade areas, and underestimate the importance of good locations.
Key Words: spatial autoregression, spatial statistics, spatial econometrics, gravity model, trade area
When considering opening a new store, retail executives attempt to maximize performance
across their entire store network. In particular, they wish to avoid opening a proﬁtable store
at the expense of existing stores. To avoid such a loss, executives need the spatial
distribution information of customers and competitors to accurately deﬁne trade areas for
site selection. In addition, managers of individual stores can use the spatial distribution of
customers and competitors to promote sales. From an overall market perspective, the
technology of forecasting sales can affect the location premia of retail properties.
We apply retail gravitation notions to examine empirically the spatial distribution of
Over seventy years ago Reilly (1931) published his seminal proposition,
known as Bthe law of retail gravitation.^ Retail gravity models draw an analogy with
Newton’s gravitational law to account for human behaviors related to shopping ac-
tivities. In retail gravity models, various store features such as size attract customers, just
as larger astronomical bodies have greater gravitational force. Distance between the cus-
tomers and the store diminishes this attraction, just as gravity diminishes with distance.
Many studies have examined empirically this concept of retail gravitation.
less, the literature often differs on the importance of the distance among retailers and
The Journal of Real Estate Finance and Economics, 31:1, 53–69, 2005
2005 Springer Science + Business Media, Inc. Manufactured in The Netherlands.