Geographic differences and the location of new manufacturing facilities

Abstract

This paper introduces two innovations to the empirical study of plant location: (1) division of the decision into stages, and (2) use of plant-specific characteristics to either magnify or temper factors defined at the state level. The plant-specific characteristics derive from a study of Fortune 500 plants that were opened in the 1970s. A number of relationships are derived from considerations of expected profitability that relate site selection to state and firm characteristics. These are tested through a series of multinomial logit models. The results confirm that the plant location decision can be usefully approached as a staged process and that geographically defined differences are not sufficient, by themselves, to explain why some states do better than others in attracting new plants. When plant-specific characteristics are used to modify the state-based factors, the explanation becomes much richer and more powerful.