An analysis of plant location decisions in wood-based companies

Abstract

A conceptual model of industrial location selection was synthesized from spatial microeconomics and business strategy. The model was empirically tested by (1) identifying and ranking important location factors, (2) quantifying the influences of strategy on the evaluation of location factors, and (3) describing how location decisions develop into an industrial location pattern. The southern plywood industry was selected for study. The research was conducted in 5 stages: (1) statistical identification of strategic groups, (2) a survey of decision makers, (3) case studies, (4) statistical analyses of existing location characteristics, and (5) a comparative analysis of the industry's intraregional development. A strategic typology was developed by cluster analysis of 24 plywood firms on 14 strategic variables. Five groups were formed for further testing of conduct in location selection. Chi-square analysis of a questionnaire survey of the 24 firms showed that the importance of different categories of location factors changed at different levels of geographic resolution. Subregion selection was market-oriented, and state selection was oriented to timber and labor supply. Site selection was aimed at minimizing timber transport cost which was often partially sacrificed to utilize an existing site, improve process integration, take advantage of community attitudes, get closer to market, or avoid timber purchase competition. Multidimensional scaling of survey responses identified underlying dimensions in the evaluation of location factors. Eight case studies verified that location factors can function as conditions, requirements, or determinants in the selection process. Eleven physical characteristics were measured for each of 73 plywood plant sites. Analyses of variance showed that strategic group affiliation accounted for significant variation in (1) competitor capacity, (2) company landownership, (3) percent of non-industrial land, and (4) market potential. Canonical correlation analysis produced several specific location to strategy links. The geographic pattern of the industry was analyzed using an integrated model of spatial oligopoly and product market evolution. Strategic group composition was compared to levels of timber purchase competition. Strategic heterogeneity was shown to vary in time and space, increasing at centers of industrial concentration.Recommendations were developed for forestry policymakers, company planners, and economics researchers.