SME competitive strategy and location behavior: an exploratory study of high-technology...

Introduction

Regional development of high technology-based industries has attracted the attention of researchers in the fields of industrial management, entrepreneur-ship, economics, and urban planning for the past two decades. More recently, the rapid national and global proliferation

of high-technology centers, incubators, and "technopolis" communities has highlighted the need to more fully understand the driving forces behind this growth (e.g., Tamasy 2007; Drejer 2005; Lofsten and Lindelof 2003; Stuart and Sorenson 2003; Biggiero 2002; Preer 1992). One important factor impacting the competitive strategies and location decisions of both high-technology and traditional manufacturing firms is the development of flexible, modular, and cluster-based manufacturing technologies (Small 2007; Dasci 2005; Galbraith and DeNoble 2002; Wadhwa and Rao 2000; Boyer 1999; Boyer et al. 1997; Dean and Snell 1996). These advances have allowed some manufacturing firms, particularly small to medium-sized technology-based enterprises, to become much more fluid in their location decisions, thus freeing them of encumbering long-term commitments to a particular site or region (Galbraith and DeNoble 1995; McDonald 1986). Like any other business investment, the flexibility of production switching with low change-over costs, as well as the ability to quickly shift operations between diversified locations, each with varying input cost structures and output distribution benefits, now provides a valuable location-related real option (Nembhard, Shi, and Aktan 2005; Wu and Lin 2005; McGrath and MacMillan 2000; McGrath 1997; Lei, Hitt, and Godhar 1996). Not surprisingly, many technologically advanced manufacturing companies report that they regularly reevaluate the merits of their existing location networks as part of an ongoing, dynamic planning process rather than assuming location to be primarily a depreciated fixed asset (MacCormack, Newman, and Rosenfeld 1994).

The ability of an organization to refocus or relocate all or part of its operations within a relatively short timeframe has elevated the importance of the location issue not only as a component of competitive strategy, but also as a way to understand the rapidly changing industrial constituencies of various communities (Alcacer and Chung 2007; Cumbers, Mackinnon, and Chapman 2003; Vereecke and van Dierdonck 2002; Ginsberg, Larsen, and Lomi 2001; Ferdows 1997a; Lei, Hitt, and Godhar 1996; Bartmess et al. 1994; Kogut and Kulatilaka 1994; Galbraith and DeNoble 1992; DeMeza and Van der Ploeg 1987). Following this line of research, the purpose of this study is to investigate whether significant linkages exist between selected aspects of manufacturing and competitive strategy and related infrastructure requirements at both the regional- and site-specific levels of analysis.

Literature Review

Early approaches to understanding industrial development and location decision-making are rooted in Weber's (1929) neoclassical assumption that the maximization of owner wealth is critical to the location decision of manufacturing facilities. Since Weber's work, early location and regional development economic theory expanded on the notion of "agglomeration," where firms were believed to cluster in regions primarily because of demand and supply considerations (e.g., Goodman and Bamford 1989). More recently, as in the case of the first-generation high-technology centers of California's Silicon Valley and Boston's Rte. 128, interest has switched to the influence of intellectual resources, social, and venture capital networks to explain agglomeration tendencies (e.g., Alcacer and Chung 2007; Drejer 2005; Cumbers, Mackinnon, and Chapman 2003; Sorenson 2003; Stuart and Sorenson 2003; Suarez-Villa 2002; Baptista 1996; Saxenian 1994 1991; Jarboe 1986; Malecki 1984; Dorfman 1983). With the rise of second- and third-generation technology centers, research studies have started investigating the influence of factors related to more ambience and personal lifestyle issues (e.g., Baptista 1996; Gripaios et al. 1989; Hall et al. 1987; Schmitt et al. 1987; Jarboe 1986; Galbraith 1985).

While the agglomeration line of regional development research has contributed greatly to our understanding of location decisions, this approach still contains many of the neoclassical, "black-box" views of the firm by assuming relatively homogeneous firm-specific strategic content (Jaffe, Trajtenberg, and Henderson 1993; Krugman 1991). It was Schmenner (1982), however, who explicitly argued that plant location decisions should be understood within the context of a firm's dynamic corporate strategy rather than as a simple cost-function. This foundation has led some researchers to argue for a more strategic-oriented, transaction cost analysis of location decisions and entrepreneurial clustering (McCann, Arita, and Gordon 2002).

Though recognizing the potential for differences in location behavior, the vast majority of empirical research still implicitly assumes that high-technology firms are essentially homogeneous in their location strategies, thus commonly pooling data either by sector or size for analysis while ignoring other firm-specific characteristics. However, there is increasing evidence of significant differences in spatial tendencies among firms even within the same industrial sector. Accordingly, several more sophisticated frameworks have been offered that purport to shed light on high-technology location behavior. One important early approach is the spatial application of product life cycle theory, where the stage of product maturity is argued to affect location requirements (Begg and Cameron 1988; Galbraith and DeNoble 1988; Malecki 1981). Country culture and preferences also appear as possible determinants of high-technology location behavior (Nohria and Ghoshal 1994; Haigh 1990), as does plant mission and operational strategy (Ketels 2005; Vereecke and van Dierdonck 2002; Brush, Maritan, and Karnani 1999; Khurana and Talbot 1998; Ferdows 1997b; Galbraith and DeNoble 1995). Recent explorations have also examined the role of the social capital embedded in entrepreneurs' networks of relationships as a possible determinant of co-location patterns in high-technology industries (Stuart and Sorenson 2003).

In addition, location requirements vary by the level of geographical analysis (Badri, Davis, and Davis 1995; Schmenner 1982). One decision typically deals with the regional area (e.g., county or state) or a limited set of areas typically beyond a normal commuting distance, whereas another focuses on the actual site, industrial center, or research park. It is often the complex packaging of these two considerations, regional and site specific, that leads to the final location decision.

A line of research has also suggested that technology content is important in understanding location decisions (Barrios, Gorg, and Strobl 2006; Ketels 2005; Meijboom and Vos 1997; Markusen, Hall, and Glasmeier 1986) and has been argued by some to be the most appropriate framework (Taylor 1987). Within the context of location decisions, previous studies have typically defined technology content as R & D investment; however, the argument can also be equally applied to process technology. Specifically, manufacturing systems can be measured on a variety of levels, including their flexibility (i.e., providing the ability to quickly switch production between multiple products) and modularity (i.e., represented by a series of stand-alone technologies (e.g., in production cells) versus a single interrelated, nonseparable process).

These developments actually have two important theoretical impacts on location behavior. First, as discussed earlier, firms with flexible and/or modular systems may no longer consider their facility a fixed asset but rather a discretionary asset that can be quickly altered as competitive and technological pressures demand. This allows firms to more "thinly slice" their co-location decisions by having multiple sites, each one appropriate for a specific value-adding activity (Ricart et al. 2004). Second, just as the flexible nature of today's high-technology manufacturing processes encourages the integration of location decisions into a corporation's competitive strategy, it also provides a powerful incentive for certain types of strategies that can be particularly important for Small and Medium-Sized Enterprises (SMEs) (Marri, Irani, and Gunasekaran 2007). For example, a strong theoretical argument for the geographical diversification of flexible manufacturing operations can be made based upon risk-hedging (Wu and Lin 2005), market opportunity shifts and input switching (Feng and Yamashiro 2006), and the transferability of firm-specific capital (such as high-value transportable equipment) and technical know-how within a flexible manufacturing environment (Nembhard, Shi, and Aktan 2005; Galbraith and DeNoble 2002; Kogut and Kulatilaka 1994; DeMeza and Van der Ploeg 1987). All of these arguments underline the need to examine location decisions as a firm-specific asset within the context of competitive strategy (Ketels 2005; Ricart et al. 2004).

In this paper, we therefore investigate the notion that spatial behavior and location preferences, both regional and site-specific, are directly correlated to a firm's competitive strategy, or as Ketels (2005) notes, "the best location for a company is uniquely dependent on its own strategy" (p. 879). We therefore include several different dimensions of competitive strategy considered to be most relevant to SME technology-intensive operations. These are classified under the categories of technology, manufacturing, and market competition strategies.

One key decision is related to the nature of a firm's technology strategy. This is often related to the level of research and development investment and targeted level of technological advancement when compared with competitors (Lin, Chen, and Wu 2006; Kurokawa, Pelc, and Fujisue 2005; Cooke 2003; Suarez-Villa and Karlsson 1996).

In terms of manufacturing strategies, two aspects are often discussed. The first deals with the flexibility and modularity of the manufacturing processes, which allow more discretion in decisions about optimal lot size, tooling switchovers, and length of production runs. Some firms have tended toward manufacturing strategies of high flexibility and modularity, whereas others within the same industry have opted for more traditional assembly line manufacturing strategies (Marri, Gunasekaran, and Sohag 2007; Dasci 2005; Galbraith and DeNoble 2002; Dean and Snell 1996). The other component relates to the labor content of the manufacturing processes, with the classic trade-off between labor and capital inputs as factors of production. High-technology firms have been shown to use different strategies involving combinations of the two inputs; some remain labor-intensive on the basis of highly skilled labor whereas others emphasize manufacturing processes that are more machine- (capital-) intensive.

Finally, regarding market competition strategies, this study employs the idea of generic strategies (Porter 1985), where firms can obtain competitive advantage by emphasizing strategies of either cost leadership or differentiation.

Research Hypotheses

Our underlying thesis is that location criteria are associated with the various dimensions of a firm's strategy. Within this context, four specific hypotheses are developed.

Regarding technology strategies, it is recognized that some firms elect technology leadership strategies by investing in substantial research and development (R & D) and product development. These firms are likely to focus their product development initiatives on creating innovations based on state-of-the-art technologies combined with possible first-mover advantages (Lieberman and Montgomery 1988). Since the seminal work of Marshall (1920), technology-leadership strategies are known to depend on the efforts of highly skilled knowledge-based employees, which, in turn, are associated with networks that create key externalities such as a pool of specialized workers, availability of venture capital, and expert legal advice (Lindelof and Loftsen 2003). Such motivations are not likely to be as important for technology-follower firms, who may place a higher premium on strategies that involve low-cost manufacturing of proven products and technologies.

On the same vein, firms that pursue a high-value/high-cost market strategy (differentiation, Porter 1985) need to emphasize investments in product development that create a perception of uniqueness. This strategy usually requires a significant focus on providing customers with specialized, premium products. In technology-intensive industries, differentiators tend to also be technology leaders, concentrating their efforts on the R & D function of their value chain while possibly obtaining similar advantages from the participation in technology networks or clusters (e.g., Barrios, Gorg, and Strobl 2006; Drejer 2005; McCann, Arita, and Gordon 2002; McCann 1997; Turok 1993). In contrast, firms following a cost leadership strategy aim at establishing a cost structure that allows them to manufacture products at the lowest possible cost. For these firms, the product R & D functions are typically less emphasized (Porter 1985). Thus our first hypothesis:

H1: Firms that pursue (a) a technology R & D leadership strategy or (b) a high-value/high-cost strategy (Porter's differentiation) will place greater emphasis on criteria related to proximity of technology networks and availability of technical labor, whereas firms pursuing (c) a technology R & D follower strategy or (d) a low-cost/-price strategy (Porter's cost leadership) will place greater emphasis on traditional location criteria such as input and property costs, financial incentives, and proximity of manufacturing labor. These relationships will hold true for both regional and site-specific location decisions.

Firms following technology-leadership and/or differentiation strategies will also likely employ more highly skilled, highly educated individuals with consumption patterns tending toward higher-end products and services, as well as an increased interest in quality of life and macro-environmental issues (e.g., Holt 1998; Suarez-Villa and Fischer 1995; Coleman 1983). Attracting and retaining an educated and scientific workforce is critical for firms pursing technology-leadership and/or differentiation strategies.

H2: Firms that pursue (a) a technology R & D leadership strategy or (b) a high-value/high-cost strategy (Porter's differentiation) will place a greater emphasis on ambience issues such as lifestyle, desirability of area for executive living, high-technology appearance in a facility, and personal amenities in their location criteria. This relationship will hold true for both regional and site-specific location decisions.

As previously discussed, the flexibility and modularity that are defining characteristics of recent, advanced manufacturing technologies have allowed firms more discretion in their location decisions. McDonald (1986), for example, suggests the "floating factory" as a good metaphorical representation of the dynamic nature of the modern plant location decisions. We therefore argue that for companies adopting these flexible and modular technologies, market competitive considerations become more important location factors than those related to the more classical location criteria such as land costs, tax incentives, and proximity of manufacturing labor.

H3: Firms that adopt more flexible and modular manufacturing strategies will place less emphasis on traditional location criteria such as input and property costs, financial incentives, and proximity of manufacturing labor than firms adopting less flexible manufacturing strategies. This relationship will hold true for both regional and site-specific location decisions.

The last hypothesis sums up a general relationship between the importance of labor as a factor of production in firms' manufacturing and product development strategies and the location decision.

H4: Firms that have high labor content will place greater emphasis on proximity to labor pool, labor productivity, and labor costs. This relationship will hold true for both regional and site-specific location decisions.

As labor content can be either technical or nontechnical in nature, no specific hypotheses are developed related to the other aspects of location.

High Technology in Scotland

The Scottish area was considered interesting for the study because it represents a geographically broad and relatively successful high-technology manufacturing region. A relatively small country of approximately five million people, Scotland has historically been associated with the agricultural, heavy manufacturing, mining, and shipbuilding industries. In the late 1980s, the government implemented a number of programs to both attract international technology-based firms and expand its local high-technology entrepreneurial base, particularly in light of the aggressive European Union's harmonization efforts during that period and the fact that Scotland enjoyed a high concentration of engineering-, health-, and science-oriented universities. High-technology employment has steadily increased since this time. While approximately 13.6 percent of Scotland's labor force is employed in the manufacturing sector (compared to 15.1 percent of Great Britain), Scotland's employment of science, health, and technology processionals has increased dramatically. For example, currently the electronics sector accounts for about 12 percent of Scotland's manufacturing employment and 50 percent of its exports (Scottish Enterprise 2002). The life sciences sector is close behind and expected to overtake electronics in the near future (Scottish Enterprise 2006). Other important technology sectors in Scotland include software, optics, and instruments.

Though this area has not been as extensively examined as Silicon Valley and other well-known high-technology centers, there is a small but growing scholarly literature addressing Scottish high-technology development (e.g., Raines, Turok, and Brown 2001; McCann 1997; Turok 1993). And although some external validity questions may arise regarding the regional focus, the findings of our study should have a clear applicability to areas with similar geographical characteristics such as Ireland and other northern European regions. The site-specific criteria, however, are expected to be more geographically independent and valid across all geographic regions.

Sample

High-technology companies located in Scotland were surveyed regarding their technology and competitive strategy, regional location behavior and expectations, and infrastructure requirements. A list of firms was obtained from the Scottish Enterprise in Glasgow (the Scottish Development Agency) and screened for our industry criteria. For purposes of this project, we have selected a fairly restrictive definition of high technology and included only those industries of computers, communication equipment, electronic components, optics, and instruments. Because we are primarily concerned with firms that actually produce a definable product, firms such as pure R & D laboratories, computer and electronics consulting and services, and biotechnology were excluded from the present analysis. The remaining firms were randomly contacted by phone, and approximately 50 or about 35 percent of the firms contacted agreed to participate in the survey. The senior executive in each firm was then interviewed in person.

All firms in the sample performed manufacturing operations, with an average of 62 percent of all employees engaged in manufacturing activities. Though approximately two-thirds of the sample was composed of subsidiaries, the vast majority of these subsidiaries marketed and sold their output to unrelated customers. Specifically, 8 percent of the employees in the sample were engaged in marketing and sales functions (82 percent of the subsidiaries in the sample had an on-site sales force), and 75 percent of the subsidiaries in the sample sold more than 80 percent of their production in external transactions. This clearly indicates a high degree of business autonomy among the subsidiary firms in the sample. In addition, we excluded any firm selling more than 90 percent of their output to the parent firm. The resulting sample contained 44 usable responses, 15 independent firms, and 29 independent acting subsidiaries. The size of our sample is comparable to most single-country empirical research (e.g., Suarez-Villa and Karlsson 1996) and falls within the range of many broader location studies that rely upon primary survey data (e.g., Vereecke and van Dierdonck 2002).

Many of the sample firms, including the subsidiaries, also performed product development on site, with an average of 10 percent of the employees in the sample engaged in R & D. The remaining employees were classified as administrative or management. The average size of the firms surveyed was relatively small, about 260 employees and can thus be classified as small-to-medium sized enterprises. The median age of the plant was less than 10 years.

Variables

Dependent Variables: Regional and Site-Specific Location Criteria

Unlike the vast majority of previous spatial tendency studies that examined only regional location (or confound regional dimensions with more site-specific concerns), we explicitly focused on both regional and site-specific decisions. For this study, the survey instrument consisted of two sets of questions for identifying the salient factors employed by high-technology firms in their location decisions.

The first set, consisting of 26 questions, dealt with infrastructure issues that potentially could influence the regional location decision. Here, the selection of relevant factors was based on a combination of Schmenner's classic 1982 study and the survey instruments used by Brush, Maritan, and Karnani (1999), Galbraith and DeNoble (1995), and Schmitt et al. (1987). The second set of 26 questions addressed factors relevant to specific site location decisions. All location criteria factors were measured by a five-point Likert scale. The respondents were asked to rate each factor on "how important it is for their location decision." The survey questionnaire was administered in person to the senior executive from each of the sample firms.

A factor analysis was performed on the original 26 regional decision variables. Nine factors were estimated (eigenvalues > 1.00), accounting for over 83 percent of the cumulative variance. These regional decision factors were labeled Employee Labor Productivity, Availability of Technical Labor, Energy Input Costs, Labor Input Costs, Availability of Funding Incentives, Regional Ambience, Proximity to Technology Firms, Proximity to Suppliers/Buyers, and Management (Chief Executive Officer) Desires to Live in Area. Similarly, a factor analysis was also performed on the original 26 site-specific decision variables. Eight factors were estimated (eigenvalues > 1.00), accounting for over 82 percent of the cumulative variance. These site-specific decision factors were categorized as Industrial Site Amenities, Personal Site Amenities, High-Technology Appearance, Proximity to Labor Pool, Large Facility Size, Proximity to Other High Technology, Cost of Property, and Proximity to Housing. The variables constituting each factor and the relevant factor loadings (loadings > 0.400) are shown in Table 1.

Explanatory Variables: Strategy

Two different measures of technology strategy were examined. R & D Focus was a five-point scale asking how advanced was the firm's internal research program relative to university research; Technology Strategy was measured by a five-point scale asking the firm to characterize its technology strategy from "state-of-art, first to market strategies" to "emphasizing the manufacturing of proven designs." To categorize the combinations and interactions of the two technology subcomponents into a typology of technology strategy, a cluster analysis was performed. Cluster analysis has been used extensively in the strategy literature to empirically delineate between various types of strategic behavior. Employing a hierarchical agglomerative clustering technique resulted in a two-cluster solution. Examining the difference in mean values of the two technology variables (p < .01), the first cluster clearly reflected a grouping of firms with more advanced technology-leadership strategies, that is, "advanced R & D effort" combined with a "state-of-art, first to market strategy" (50 percent of sample), whereas the second cluster appeared more of a technology-follower strategy (50 percent of the sample). This was used in the regression equation as a dummy variable (1 = Technology Leader, 0 = Technology Follower).

Similarly, two measures of manufacturing process strategy were used. Manufacturing Flexibility was a five-point scale examining the ease of shifting production between product categories, whereas Manufacturing Modularity was a five-point scale assessing the modularity of the manufacturing process, from "a series of stand-alone technologies" to "interrelated, non-separable technologies." A cluster analysis on these variables again resulted in a two-cluster solution. The first cluster (68 percent of sample) reflected a "flexible and modular" manufacturing strategy, whereas the second cluster (32 percent of sample) indicated a less flexible, interrelated manufacturing process (p < .01, difference in means). In the regression equation, this analysis was represented by the inclusion of a variable called Flexible Manufacturing Strategy (1 = Flexible/Modular, 0 = Rigid/Interrelated).

Several other strategy-related variables were examined. A Market/Pricing strategy variable was measured as a five-point scale from "low cost leader in industry" to "offering additional value at a higher cost." Labor/Machine Intensity was measured by a five-point scale assessing the degree of "labor" versus "machine" intensity in operations.

Table 2 presents the bivariate correlations for the raw strategy variables. Not surprisingly, the R & D Focus was significantly correlated to Technology Strategy in that advanced internal R & D is positively related to state-of-the-art, first-to-market strategies (note, survey questions were opposite coded) whereas high manufacturing flexibility (Manufacturing Flexibility) was positively correlated to more modular manufacturing (Manufacturing Modularity). In addition, there were significant bivariate correlations between Technology Strategy and Manufacturing Modularity, with more advanced technologies being associated with more modular manufacturing, and between Manufacturing Flexibility and the Market/Pricing variable, where more flexible manufacturing was negatively related to low-cost pricing strategies. However, after the creation of the combined strategy clusters of Technology Strategy and Manufacturing Strategy, none of the correlations between the strategy variables were statistically significant, suggesting no serious multicollinearity between the major strategy variables used in the regression analysis.

In addition, we recognized that many previous studies on location behavior did not control for subsidiary or "component" effects (e.g., Suarez-Villa and Karlsson 1996). Though our basic research question is not dependent upon whether or not a firm is a subsidiary, that is, our hypotheses revolve around the fundamental relationship between strategy and location criteria regardless of the locus of decision-making responsibility, we wanted to control for possible effects. Thus, Subsidiary was a dummy (0-1) variable reflecting whether or not the sampled firm was technically an independent firm and is used in the study as a control variable. Several other variables, such as vertical integration and diversification, were also examined; however, these did not provide any additional significant explanatory power to the models, and for brevity purposes were excluded from the succeeding discussion.

Results and Discussion

Pooled Results

A scale for each factor was created using the variables with factor loadings greater than .400 described in Table 1. This allows for a comparison of the relative importance between the different location factors whereas the direct factor scores are normalized, and thus are difficult to compare across factors. For the regional criteria (full sample), the Availability of Technical Labor was most important (3.72), followed by Employee Labor Productivity (3.65), Availability of Funding Initiatives (3.31), Labor Input Costs (3.11), Proximity to Suppliers/Buyers (2.86), Proximity to Technology Firms (2.85), Energy Input Costs (2.70), Regional Ambience (2.59), and Management (CEO) Desires to Live in Area (2.04). For the site-specific criteria (full sample), the Proximity to Other High Technology (4.03) was the most important criterion followed by Cost of Property (3.79), Proximity to Labor Pool (3.58), Proximity to Housing (3.53), Facility Size (3.51), High-Technology Appearance (2.88), Industrial Site Amenities (2.72), and Personal Site Amenities (2.63).

In all, the pooled results appear to reflect the fundamental nature of technology-oriented manufacturing. Proximity to technological employees and the ability to access intellectual and social capital networks, combined with traditional manufacturing issues such as financial incentives and labor productivity, are seen as the most dominant location criteria. This is true at both the regional and site-specific levels of analysis. In fact, these broad results from the pooled sample are consistent with previous research regarding technology-based manufacturing spatial behaviors (Baptista 1996; MacCormack, Newman, and Rosenfeld 1994; Jarboe 1986; Rabino 1989; Galbraith 1985) and provide support regarding the representativeness of this particular sample.

Location versus Strategy

Though the relative importance of the different location factors is, in itself, interesting, we are primarily interested in examining the degree to which location criteria are coupled to the various elements of corporate strategy. To determine this, a regression analysis was performed on each of the nine regional and eight site-specific factors. As discussed earlier, explanatory or independent variables included technology strategy, manufacturing strategy, labor content, market competition/pricing strategy, and subsidiary. Table 3 shows the results of the regional location criteria regressions, whereas Table 4 shows the estimated regression models for the site-specific location criteria. Only the significant regressions are shown in these tables.

Overall, the regression results support the broad hypothesis that location criteria and competitive strategy are closely related. With respect to the regional location criteria, seven of the nine regressions produced significant results, with almost every coefficient estimated in the hypothesized direction. Estimated [R.sup.2]s of the statistically significant models ranged between 0.115 and 0.258. Consistent with our hypotheses, firms that followed technology-leadership strategies combined with more sophisticated flexible manufacturing strategies generally placed significantly less emphasis on labor productivity, energy input costs, and financial incentives. Likewise, these firms placed significantly greater emphasis on proximity to technology networks and the region's ambience.

Though not as strong in explanatory power (only four of eight equations produced statistically significant models), similar results were found for the site-specific criteria. Firms that followed more technologically advanced strategies appeared to place a significantly higher priority on the appearance of the industrial site and its proximity to housing, and less emphasis on the facility size. Combined, these findings provide support for H1a, H1c, H2a, and H3.

Our labor content hypothesis (H4) was only partially supported, however. Though the labor content variable consistently produced estimated coefficients in the hypothesized direction, none were statistically significant in the broad regional location analysis, and in the site-specific model labor content was significant in only one regression; high labor content was positively related to an industrial site's proximity to local labor sources.

The pricing strategy variables also produced results in the hypothesized direction, providing some support for H1b, H1d, H2b, and H3. With respect to regional decisions, cost leadership strategies were positively associated with traditional cost input factors. On the other hand, higher cost or differentiation strategies were positively associated with proximity to technology labor pools, other technology firms, and buyers/suppliers--an indication that firms following these strategies will tend to cluster within geographical regions or areas that have strong surrounding intellectual and social capital networks. For the site-specific models, again, though most of the coefficients were in the hypothesized direction, they were not statistically significant probably due to the effects of a relatively small sample size and the number of variables in the estimated model.

The subsidiary variable was used essentially as a control variable in our models and was significant in several equations. Being a subsidiary was positively related to technology labor availability and energy input cost in the regional criteria models and to high-technology appearance and proximity to labor pool for the site-specific models. Because subsidiary was used as a control variable, no specific hypotheses were developed.

Conclusion

Historical approaches to understanding the location requirements and spatial development of technology-intensive environments have primarily centered on sector-specific factors such as industry life cycle and industry R & D within an underlying neoclassical economic framework. No doubt this is partly due to the fact that the vast majority of empirical studies have employed secondary data sources such as corporate formations, tax data, utility costs, and employment information collected and maintained by government or quasi-government agencies. And although recent analyses of these increasingly detailed secondary data, such as Canada's Longitudinal Employment Analysis Program, have provided a more sophisticated insight into the nature of entrepreneurial clustering and regional development (see, e.g., Pe'er and Vertinsky 2004), these studies usually de-couple the firm's strategic content from its location criteria by assuming that the underlying starting point for location analysis is the economics of the regional cluster rather than the dynamics of the locating firm's competitive strategy. Over the past two decades, however, a relatively small but continuing stream of research (Ketels 2005; Ricart et al. 2004; Vereecke and van Dierdonck 2002; Ginsberg, Larsen, and Lomi 2001; Brush, Maritan, and Karnani 1999; Ferdows 1997a; Galbraith and DeNoble 1995, 1992; McDonald 1986; Schmenner 1982) has demonstrated that location decisions, spatial development, and economic geography must also be properly understood within the broader context of a firm's competitive strategy.

In this study, we expand on this theme with a focus on small to medium-sized enterprises in high-technology industries. In particular, we empirically examine the relationship between elements of a firm's competitive strategy and its geographical requirements using a sample of high-technology manufacturing firms located in Scotland. In addition, by using primary data we attempt to understand the strategy-location issue in much more detail than previous efforts that rely upon secondary data sources. With respect to both the regional and site-specific levels of analysis, several different elements of high-technology competitive strategy--technology strategy, manufacturing strategy, labor input, and pricing--all appear to influence a firm's location criteria.

Though due to the sample-size limitations this study is clearly exploratory in nature, the results have several implications. First, as technology-intensive firms utilize increasingly flexible and modular manufacturing systems, the relationships between facility location and competitive strategy will likely become more intertwined and complex. As this trend continues, it will become even more important to understanding regional economic development and the nature of entrepreneurial clustering within the context of firm-specific assets and the overall strategic positioning of the firm. Second, as McCann, Arita, and Gordon (2002) note, firms must include in their location decisions considerations about the nature of their strategic relationships with other companies in the same region or site, depending on whether the firms in that location form clusters with characteristics based on strong or loose industrial and social exchanges. However, given that our empirical models account for less that 30 percent of the total variation in our data, it certainly suggests that other variables not investigated in our study are also important in explaining location criteria and behavior.

Finally, there are important public policy lessons. As location and competitive strategy decisions become more intertwined, local governments increasingly need to view their community's attributes as a potential strategic resource within the overall strategic orientation of the firm. In this manner, policymakers can better analyze their community or region within a framework of strategic value, contestability, and imitability to develop a clearer understanding of how to best position or "improve" their communities to attract and retain high-technology industries (Lowe 2007; Ginsberg, Larsen, and Lomi 2001). Policymakers and firms have to consider the possibility that the same factors that historically favor agglomeration or clustering, such as local information spillovers and a specialized, dynamic employee pool that provides flexibility for rapid restructuring, can now be better framed within the transaction economics from both an embedded network perspective (Raines, Turok, and Brown 2001) and within the context of a firm's dynamic competitive strategy (Ketels 2005; Ginsberg, Larsen, and Lomi 2001; Galbraith and DeNoble 1995).

With an improved understanding of these relationships, policymakers can assist in developing economically strong entrepreneurial clusters. Caution must be given, however, because tightly concentrated entrepreneurial or innovative clusters, though benefiting a local region, might have broader negative social impacts such as increasing income inequalities between neighboring regions in the same country (Suarez-Villa and Fischer 1995).

It is the thesis of this paper that location decisions and entrepreneurial clustering must be understood within the context of both competitive strategy and the developments of modern manufacturing technology. However, other factors that were not investigated in detail in the present study such as the nature of clustering economics (McCann, Arita, and Gordon 2002), how well information can be codified and communicated within a cluster (Audretsch 1998), negative and positive externalities of clustering (Pe'er and Vertinsky 2004), the evolutionary role of multinational plants (Vereecke and van Dierdonck 2002), and the embeddedness of plants within a cluster's network (Stuart and Sorenson 2003) are also relevant to the study of location decisions of SMEs. Together, these concepts paint a much more complete picture of regional and local clustering dynamics that needs further investigation.

References

Alcacer, J., and W. Chung (2007). "Location Strategies and Knowledge Spillovers," Management Science 53(5), 760-776.

Audretsch, D. (1998). "Agglomeration and the Location of Innovative Activity," Oxford Review of Economic Policy 14(2), 18-29.

Badri, M., D. Davis, and D. Davis (1995). "Decision Support Models for the Location of Firms in Industrial Sites," International Journal of Operations and Production Management 15(1), 50-62.

Baptista, R. (1996). "Industrial Clusters and Technological Innovation," Business Strategy Review 7(2), 59-64.

Barrios, S., H. Gorg, and E. Strobl (2006). "Multinationals' Location Choice, Agglomeration Economies, and Public Incentives," International Regional Science Review 29(1), 81-107.

Bartmess, A., A. Gray, K. Ferdows, and K. Cerny (1994). "The Plant Location Puzzle," Harvard Business Review 72(2), 20-30.

Begg, I., and G. Cameron (1988). "High Technology Location and the Urban Areas of Great Britain," Urban Studies 25(5), 361-379.

Biggiero, L. (2002). "The Location of Multinationals in Industrial Districts: Knowledge Transfer in Biomedicals," Journal of Technology Transfer 27(1), 111-125.

Boyer, K. (1999). "Evolutionary Patterns of Flexible Automation and Performance: A Longitudinal Study," Management Science 45(6), 824-842.

Boyer, K., G. Leong, W. Keong, T. Peter, and L. Krajewski (1997). "Unlocking the Potential of Advanced Manufacturing Technologies," Journal of Operations Management 15(4), 331-347.

Brush, T., C. Maritan, and A. Karnani (1999). "The Plant Location Decision in Multinational Manufacturing Firms: An Empirical Analysis of International Business and Manufacturing Strategy Perspectives," Production and Operations Management 8(2), 109-132.

Coleman, R. (1983). "The Continuing Significance of Social Class to Marketing," Journal of Consumer Research 10, 265-280.

Cooke, P. (2003). "Biotechnology Clusters, 'Big Pharma' and the Knowledge-Driven Economy," International Journal of Technology Management 25(1,2), 65-80.

Cumbers, A., D. Mackinnon, and K. Chapman (2003). "Innovation, Collaboration, and Learning in Regional Clusters: A Study of SMEs in the Aberdeen Oil Complex," Environment & Planning 35(9), 1689-1706.

Dasci, A. (2005). "Evaluation of Plant Focus Strategies: A Continuous Approximation Framework," Annals of Operations Research 126(1/4), 303-327.

Dean, J. Jr., and S. Snell (1996). "The Strategic Use of Integrated Manufacturing: An Empirical Examination," Strategic Management Journal 17(6), 459-480.

DeMeza, D., and F. Van der Ploeg (1987). "Production Flexibility as a Motive for Multinationality," Journal of Industrial Economics 35(3), 343-351.

Dorfman, N. (1983). "Route 128: The Development of a Regional High Technology Economy," Research Policy 12(6), 299-316.

Drejer, I. (2005). "Location and Collaboration: Manufacturing Firms' Use of Knowledge Intensive Services in Product Innovation," European Planning Studies 13(6), 879-898.

Feng, D., and M. Yamashiro (2006). "A Pragmatic Approach for Optimal Selection of Plant-Specific Process Plans in a Virtual Enterprise," Journal of Materials Processing Technology 173(2), 194-200.

Ferdows, K. (1997a). "Made in the World: The Global Spread of Production," Production and Operations Management 6(2), 102-109.

______ (1997b). "Making the Most of Foreign Factories," Harvard Business Review 75(2), 73-88.

Galbraith, C. (1985). "High-Technology Location and Development: The Case of Orange County," California Management Review 28(1), 98-109.

Galbraith, C., and A. DeNoble (1988). "Location Decisions by High Technology Firms: A Comparison of Firm Size, Industry Type, and Institutional Form," Entrepreneurship Theory and Practice 13, 1-47.

______ (1992). "Competitive Strategy and Flexible Manufacturing: New Dimensions in High-Technology Venture-Based Economic Development," Journal of Business Venturing 7(5), 387-404.

______ (1995). "Strategic Scope, Flexible Manufacturing, and Plant Location in Small High Technology Firms," Journal of High Technology Management Research 6(2), 281-296.

______ (2002). "Advanced Manufacturing Technology, Flexibility and the Strategic Scope of High Technology Firms: Theoretical Thoughts and Empirical Evidence from Scottish Technology Products Firms," in International Research in the Business Disciplines, Vol. 3, Strategies and Organizations in Transition. Ed. M. Ryan. Amsterdam: Elsevier Science, 27-56.

Ginsberg A., R. Larsen, and A. Lomi (2001). "Entrepreneurship in Context: Strategic Interaction and the Emergence of Regional Economies," in The Entrepreneurship Dynamic: Origins of Entrepreneurship and the Evolution of Industries. Ed. C. Schoonhoven and E. Romanelli. Stanford, CA: Stanford Univ. Press, 314-348.

Goodman, E., and J. Bamford, Eds. (1989). Small Firms and Industrial Districts in Italy. London: Routledge.

Gripaios, P., P. Bishop, R. Gripaios, and C. Herbert (1989). "High Technology Industry in a Peripheral Area: The Case of Plymouth," Regional Studies 23, 151-158.

Haigh, R. (1990). "Selecting a U.S. Plant Location: The Management Decision Process in Foreign Companies," Journal of World Business 25(3), 22-31.

Hall, P., M. Breheny, R. McQuaid, and D. Hart (1987). Western Sunrise: The Genesis and Growth of Britain's Major High Tech Corridor. Boston, MA: Allen & Unwin.

Holt, D. (1998). "Does Cultural Capital Structure American Consumption?" Journal of Consumer Research 25(1), 1-24.

Jaffe, A., M. Trajtenberg, and R. Henderson (1993). "Geographic Localization of Knowledge Spillovers, as Evidenced by Patent Citations," Quarterly Journal of Economics 108, 577-598.

Jarboe, K. (1986). "Location Decisions of High-Technology Firms: A Case Study," Technovation 4(2), 117-129.

Ketels, C. (2005). "Location, Location, Location," Global Agenda 3, 160-161.

Khurana, A., and B. Talbot (1998). "The Internationalization Process Model through the Lens of the Global Color Picture Tube Industry," Journal of Operations Management 16 (2,3), 215-239.

Kogut, B., and N. Kulatilaka (1994). "Operating Flexibility, Global Manufacturing, and the Option Value of a Multinational Network," Management Science 40(1), 124-139.

Krugman, P. (1991). Geography and Trade. Cambridge, MA: MIT Press.

Kurokawa, S., K. Pelc, and K. Fujisue (2005). "Strategic Management of Technology in Japanese Firms: A Literature Review," International Journal of Technology Management 30(3/4), 223-247.

Lei, D., M. Hitt, and J. Godhar (1996). "Advanced Manufacturing Technology: Organizational Design and Strategic Flexibility," Organization Studies 17(3), 501-524.

Lieberman, M., and D. Montgomery (1988). "First-Mover Advantages," Strategic Management Journal 9(5), 41-58.

Lin, B., C. Chen, and H. Wu (2006). "Patent Portfolio Diversity, Technology Strategy, and Firm Value," IEEE Transactions on Engineering Management 53(1), 17-26.

Lindelof, P., and H. Lofsten (2003). "Science Park Location and New Technology-Based Firms in Sweden: Implications for Strategy and Performance," Small Business Economics 20(3), 245-258.

Lofsten, H., and P. Lindelof (2003). "Determinants for an Entrepreneurial Milieu: Science Parks and Business Policy in Growing Firms," Technovation 23(1), 51-72.

Lowe, N. (2007). "Job Creation and the Knowledge Economy: Lessons from North Carolina's Life Science Manufacturing Initiative," Economic Development Quarterly 21(4), 339-353.

MacCormack, A., L. Newman III, and D. Rosenfeld (1994). "The New Dynamics of Global Manufacturing Site Location," Sloan Management Review 35(4), 69-80.

Malecki, E. (1981). "Product Cycles, Innovation, and Regional Economic Change," Technological Forecasting and Social Change 19, 291-306.

______ (1984). "High Technology and Local Economic Development," Journal of the American Planning Association 50(3), 262-269.

Markusen, A., P. Hall, and A. Glasmeier (1986). High-Tech America: The What, How, Where and Why of the Sunrise Industries. Boston, MA: George, Allen, and Unwin.

Marri, H., A. Gunasekaran, and R. Sohag (2007). "Implementation of Advanced Manufacturing Technology in Pakistani Small and Medium Enterprises: An Empirical Analysis," Journal of Enterprise Information Management 20(6), 726-739.

Marri, H., Z. Irani, and A. Gunasekaran (2007). "Advanced Manufacturing Technology Implementation in SMEs: A Framework of Justification Criteria," International Journal of Electronic Business 5(2), 124-140.

Marshall, A. (1920). Principles of Economics. London: Macmillan.

McCann, P. (1997). "How Deeply Embedded Is 'Silicon Glen'? A Cautionary Note," Regional Studies 31(7), 695-703.

McCann, P., T. Arita, and I. Gordon (2002). "Industrial Clusters, Transactions Costs and the Institutional Determinants of MNE Location Behaviour," International Business Review 11, 647-663.

McDonald, A. (1986). "Of Floating Factories and Mating Dinosaurs," Harvard Business Review 64(6), 82-86.

McGrath, R. (1997). "A Real Options Logic for Initiating Technology Positioning Investments," Academy of Management Review 22(4), 974-996.

McGrath, R., and I. MacMillan (2000). "Assessing Technology Projects Using Real Options Reasoning," Research Technology Management 43(4), 35-49.

Meijboom, B., and B. Vos (1997). "International Manufacturing and Location Decisions: Balancing Configuration and Co-ordination Aspects," International Journal of Operations & Production 17(8), 790.

Nembhard, H., L. Shi, and M. Aktan (2005). "A Real-Options Based Analysis for Supply Chain Decisions," IIE Transactions 37(10), 945-956.

Nohria, N., and S. Ghoshal (1994). "Differentiated Fit and Shared Values: Alternatives for Managing Headquarters-Subsidiary Relations," Strategic Management Journal 15(6), 491-502.

Pe'er, A., and Vertinsky I. (2004). "Do Clusters and Non-Cluster Entrepreneurial Entries Differ?" Paper presented at the Babson-Kaufman Entrepreneurship Research Conference, Glasgow, Scotland, June, published in proceedings, 169.

Porter, M. (1985). Competitive Advantage. New York: Free Press.

Preer, R. (1992). The Emergence of Technopolis: Knowledge-Intensive Technologies and Regional Development. New York: Praeger.

Rabino, S. (1989). "High-Technology Firms and Factors Influencing Transfer of R & D Facilities," Journal of Business Research 18(3), 195-205.

Raines, P., I. Turok, and R. Brown (2001). "Growing Global: Foreign Direct Investment and the Internationalization of Local Suppliers in Scotland," European Planning Studies 9(8), 965-978.

Ricart, J., M. Enright, P. Ghemawat, S. Hart, and T. Khanna (2004). "New Frontiers in International Strategy," Journal of International Business Studies 35, 175-200.

Saxenian, A. (1991). "The Origins and Dynamics of Production Networks in Silicon Valley," Research Policy 20(5), 423-427.

______ (1994). "Silicon Valley versus Route 128," Inc. 16(2), 25-29.

Schmenner, R. (1982). Making Business Location Decisions. Englewood Cliffs, NJ: Prentice Hall.

Schmitt, M., S. Gleason, B. Pigozzi, and P. Marcus (1987). "Business Climate Attitudes and Company Relocation Decisions," Journal of Applied Psychology 72, 622-628.

Scottish Enterprise (2002). Scottish Labor Market. Glasgow: Scottish Enterprise.

______ (2006). Life Science, March 2006. Glasgow: Scottish Enterprise.

Small, M. (2007). "Planning, Justifying and Installing Advanced Manufacturing Technology: A Managerial Framework," Journal of Manufacturing Technology Management 18(5), 513-527.

Sorenson, O. (2003). "Social Networks and Industrial Geography," Journal of Evolutionary Economics 13(5), 513-527.

Stuart, T., and O. Sorenson (2003). "The Geography of Opportunity: Spatial Heterogeneity in Founding Rates and the Performance of Biotechnology Firms," Research Policy 32, 229-253.

Suarez-Villa, L. (2002). "High Technology Clustering in the Polycentric Metropolis: A View from the Los Angeles Metropolitan Area," International Journal of Technology Management 24(7), 818-842.

Suarez-Villa, L., and M. Fischer (1995). "Technology, Organization, and Export-Driven Research and Development in Austria's Electronics Industry," Regional Studies 29(1), 19-42.

Suarez-Villa, L., and C. Karlsson (1996). "The Development of Sweden's R & D-Intensive Electronics Industries: Exports, Outsourcing, and Territorial Distribution," Environment and Planning 28, 783-817.

Tamasy, C. (2007). "Rethinking Technology-Oriented Business Incubators: Developing a Robust Policy Instrument for Entrepreneurship, Innovation, and Regional Development," Growth & Change 38(3), 460-473.

Taylor, M. (1987). "Enterprise and the Product Life Cycle Model: Conceptual Ambiguities," in New Technology and Regional Development. Eds. B. Van der Knapp and E. Wever. London: Croom Helm, 75-93.

Turok, I. (1993). "Inward Investment and Local Linkages: How Deeply Embedded is 'Silicon Glen?'" Regional Studies 27(5), 401-417.

Vereecke, A., and R. van Dierdonck (2002). "The Strategic Role of the Plant: Testing Ferdow's Model," International Journal of Operations & Production Management 22(5,6), 492-514.

Wadhwa, S., and K. Rao (2000). "Flexibility: An Emerging Meta-Competence for Managing High Technology," International Journal of Technology Management 19(7,8), 820.

Weber, A. (1929). Theory of the Location of Industries. Trans. C. Friedrich. Chicago, IL: University of Chicago Press.

Wu, C., and C. Lin (2005). "Hedging Entry and Exit Decisions: Optimizing Location under Exchange Rate Uncertainty," IMA Journal of Management Mathematics 16(4), 355-367.

Craig S. Galbraith is professor of entrepreneurship and technology in the Cameron School of Business at the University of North Carolina Wilmington.

Carlos L. Rodriguez is on the faculty in the Cameron School of Business at the University of North Carolina Wilmington.

Alex F. DeNoble is on the faculty in the School of Business at the San Diego State University.

Address correspondence to: Craig S. Galbraith, e-mail: galbraithc@uncw.edu.

Table 1 Factor Loadings and Location Criteria -- Regional and Site
Specific (a)

Factor Name

Regional Location     Factor Loadings (>0.400)
  Criteria
  Employee Labor      Labor Productivity (0.791); Low Union Activity
    Productivity        (0.729); Government Attitudes Toward Business
                        (0.629)
  Availability of     Availability of Technical Employees (0.808);
    Technical Labor     Availability of Management Employees (0.633);
                        Proximity to Universities (0.713)
  Energy Input Costs  Cost of Energy (0.611); Nonhazardous Waste
                        Disposal (0.909); Hazardous Waste Disposal
                        (0.808); Water Costs (0.723)
  Labor Input Costs   Availability of Skilled Workers (0.400); Health
                        Care Costs (0.816); Labor Costs (0.584); Welfare
                        Costs (0.856)
  Availability of     Financial Incentives of Local Government (0.803);
    Funding             Local Tax Structure (0.537); Price of Industrial
    Incentives          Land (0.775); Availability of Funding Sources
                        (0.628)
  Regional Ambience   Proximity to Primary/Secondary Schools (0.876);
                        Proximity to Recreational/Cultural Activities
                        (0.845)
  Proximity to        Other High-Tech Firms in Same Industry in Region
    Technology Firms    (0.846); Other High-Tech Firms in Other
                        Industries in Region (0.789)
  Proximity of        Proximity to Customers (0.905); Proximity to
    Suppliers/Buyers    Suppliers (0.452); Transportation for Materials
                        and Products (0.805)
  Management Desires  Owner/Senior Managementt Wants to Live in Area
    to Live in Area     (0.762)
Site Specific
  Location Criteria
  Industrial Site     Parking (0.800); View (0.763); Loading Docks
    Amenities           (0.624); Visibility (0.707); Privacy (0.692);
                        Near to Services (0.901)
  Personal Site       On-Site Recreational Facilities (0.782); Food
    Amenities           Service in Facility (0.820)
  High Technology     High-Tech Appearance (0.805); Landscaping (0.558);
    Appearance          Security (0.597); Technology Industrial Theme
                        (0.707); Proximity to University (0.551)
  Proximity to Labor  Proximity to Labor Pool (0.862); Proximity to
    Pool                Public Transportation (0.618)
  Large Facility      Large Size (0.912)
    Size
  Proximity to Other  Proximity to Other High Technology (0.836); Room
    High Technology     for Expansion (0.643)
  Cost of Property    Quality Anchor Tenant in Place (0.738); Cost of
                        Property (0.632)
  Proximity to        Proximity to Housing (0.855); Proximity to Hotels
    Housing             (0.643)

(a) Cumulative Variance of Nine Regional Location Criteria Factors =
83.726 percent; cumulative Variance of Eight Regional Location Criteria
Factors = 82.682 percent.

Table 2 Correlation Matrix -- Strategy Variables

                           R & D    Technology  Manufacturing
Strategy Variable          Focus    Strategy    Flexibility

R & D Focus                 1.000
Technology Strategy        -0.225*   1.000
Manufacturing Flexibility   0.185    0.163       1.000
Manufacturing Modularity    0.134    0.278*      0.456***
Market/Pricing             -0.141   -0.051      -0.310**
Labor/Machine Intensity     0.121    0.109       0.011

                           Manufacturing  Market/  Labor/Machine
Strategy Variable          Modularity     Pricing  Intensity

R & D Focus
Technology Strategy
Manufacturing Flexibility
Manufacturing Modularity    1.000
Market/Pricing             -0.196         1.000
Labor/Machine Intensity     0.136         0.234    1.000

*p < .10 (two-tailed).
**p < .05 (two-tailed).
***p < .01 (two-tailed).

Table 3 Selected Regression Analysis -- Regional Location Criteria

                      Employee      Availability  Energy    Availability
                      Labor         Technical     Input     of Funding
Location Criteria     Productivity  Labor         Costs     Incentives

Strategy Factor
  Constant             4.081         3.927         3.598     4.408
  Subsidiary          -0.417         0.414*        0.721**  -0.161
  Technology Leader   -0.543**       0.258        -0.408*   -0.463*
    Strategy
  Flexible            -0.614**       0.209        -0.528**  -0.896***
    Manufacturing
    Strategy
  High Labor Content  -0.046         0.097        -0.008    -0.028
  Low Cost Pricing    -0.197        -0.273**       0.228*   -0.010
    Strategy
[R.sup.2]              0.191         0.140         0.258     0.209
N                     44            44            44        44

                      Proximity to  Proximity to   Management
                      Technology    Suppliers and  Desires to
Location Criteria     Firms         Buyers         Live in Area

Strategy Factor
  Constant             1.197         1.660          1.131
  Subsidiary           0.130        -0.062          0.423
  Technology Leader    0.876***      0.062          0.542*
    Strategy
  Flexible             0.186         0.207         -0.204
    Manufacturing
    Strategy
  High Labor Content  -0.029        -0.054          0.274*
  Low Cost Pricing    -0.303**      -0.315**        0.143
    Strategy
[R.sup.2]              0.276         0.115          0.147
N                     44            44             44

*p < .10 (one-tailed).
**p < .05 (one-tailed).
***p < .01 (one-tailed).

Table 4 Selected Regression Analysis -- Specific Site Criteria

                      High        Proximity  Large
                      Technology  to Labor   Facility  Proximity
Location Criteria     Appearance  Pool       Size      to Housing

Strategy Factor
  Constant             2.345       1.968      2.465     2.724
  Subsidiary           0.619**     0.492**    0.266     0.124
  Technology Leader    0.449**    -0.216     -0.494**   0.250*
    Strategy
  Flex Manufacturing   0.219       0.292     -0.234     0.424*
    Strategy
  High Labor Content   0.081       0.189*     0.067     0.058
  Low Cost Pricing    -0.141       0.173      0.155    -0.047
    Strategy
[R.sup.2]              0.184       0.256      0.162     0.130
N                     44          44         44        44

*p < .10 (one-tailed).
**p < .05 (one-tailed).