In this article we extend the Carlino and Mills and Boarnet models of local development to test for the presence and direction of rural area linkages to urban areas in Functional Economic Areas (FEAs). In a sample of southern FEAs, we detect a mix of spillover and backwash effects from urban core and fringe areas to their rural hinterlands. Rural-area population and employment both grew faster than average between 1980 and 1990 if in an FEA with a pattern of urban decentralization.

Current regional industrialization strategies encourage recruitment, small business development and business retention and expansion efforts to promote industry cluster development. The paper describes the advantages and disadvantages of promoting industry clusters as an industrial development alternative for rural areas. Advantages of successful cluster promotion include stronger external economies, a more conducive environment for industrial reorganization, greater networking among firms and more efficient use of public resources. The disadvantages of an industry cluster approach are selecting industries to target, overcoming latecomer disadvantages, and providing supporting institutions. Findings indicate that an industry cluster promotion is not appropriate for many rural communities. Areas considering cluster promotion should compare costs of initiating or expanding a cluster with the potential benefits of successful cluster development.

No abstract

Historically, employment in high-technology manufacturing industries has been concentrated in metropolitan areas. The product life cycle theory of industrial location suggests, however, that employment in this sector will decentralize as these industries mature and production processes are standardized. This study utilized shift-share analysis and Enhanced County Business Pattern data to determine the extent to which a metropolitan-to-nonmetroplitan employment shift occurred from 1975 to 1982. Employment shifts for nonmetropolitan counties of different sizes, adjacency status, and cenus regions were estimated. Also, the specific high-technology manufacturers (four-digit SIC) participating in the decentralization process were identified. The findings of this study indicate that nonmetropolitan employment in the high-tech sector increased by more than 13,000 as a result of decentralization. Small-and mediumsized nonmetropolitan counties and nonmetropolitan areas in New England, the South, and the West experienced the greatest gains in high-technology employment. Urban-to-rural employment shifts in innovative high-technology industries was restricted to primarily counties adjacent to metropolitan areas and nonmetropolitan counties in the New England and Pacific regions. INCE 1980, THE GROWTH AND LOCATIONAL DYNAMICS of S high-technology industries have received considerable attention in the economics and regional science literature. For example, Riche et al. (1983), Tomaskovic-Devey and Miller (1982) , Browne (1983), and Armington et al. (1983) have estimated the employment growth potential of high-technology industries and the ability of this sector to replace the jobs lost in the more traditional "smokestack" industries. Appelbaum (1981), Etzioni and Jargowsky (1984), Clair (1986) and Flynn (1986) examined the occupational structure of high-technology industries and forecast the sector's future occupational demands. The locational preferences of hightechnology firms and the characteristics of communities or regions which have been successful in attracting or generating these firms were analyzed David L. Barkley i s an associate professor of agricultural economics at the University of Arizona.

In this paper we investigate industry characteristics associated with the clustering of establishments in three-digit SIC manufacturing industries in nonmetropolitan areas. The dispersion parameter "k" of the negative binomial distribution is selected as the measure of industry spatial concentration. Associations between industry characteristics and spatial concentration are investigated using OLS regression analysis. Our findings indicate that the spatial clustering of establishments is positively related to industry average establishment size, reliance on natural resource inputs, labor intensity, cost shares of professional and technical employees, and cost shares of low-skilled workers. Agglomeration is negatively related to multiplant structure, employment in precision production, and reliance on local product and input markets. Copyright 2000 Blackwell Publishers