The effects of location and economic cluster development on Native American entrepreneurship Vincent J. Pascal and Daniel Stewart

Abstract: This study investigates the relationship between indigenous entrepreneurship and the location of indigenous firms near economic clusters. In recent years there has been a fundamental shift in the way regional economic development is viewed. Promoting economic cluster formation has become the new strategy for regional competitive advantage and entrepreneurial growth. Cluster development has been shown to facilitate entrepreneurship and business performance, yet no research has been undertaken to investigate whether all entities benefit from local cluster development, especially less well represented parties such as Native American entrepreneurs, who are often located in rural areas away from economic clusters. The findings support the contention that Native American entrepreneurs located near economic clusters enjoy competitive advantages not experienced by their counterparts outside of cluster economies. The implications for Native American entrepreneurship and culture are discussed and areas for future research proposed. Keywords: indigenous entrepreneurship; economic clusters; cluster development; Native American Vincent J. Pascal is Associate Professor of Marketing, Department of Management, Eastern Washington University, N 668 Riverside Blvd, Suite A, Spokane, WA 99202, USA. E-mail: [email protected]. Daniel Stewart is Assistant Professor of Management, Gonzaga University, 502 E Boone Ave, Spokane, WA 99258-0009, USA. E-mail: [email protected].

The study of indigenous entrepreneurship has been gaining momentum in recent years, partly due to its practical and political implications. Indigenous entrepreneurship has been defined as subsistence or other employment or economic activity by members of an indigenous minority. However, in the USA, relatively few studies have been undertaken to understand the Native American entrepreneur or to identify factors that either facilitate or hinder the rate of entrepreneurial activity within Native American communities (for exceptions, see Garsombke and Garsombke, 2000; Stewart and Schwartz, 2007). By comparison, interest in regional economic development has long motivated research, particularly from

economic and public planning perspectives. As a result, in recent decades we have seen considerable interest by researchers in exploring the phenomenon and consequences of regional clustering of industries, or ‘industrial districts’ (Hendry et al, 2000). Some of this interest was given impetus by Michael Porter’s (1990) seminal book, The Competitive Advantage of Nations. In his book, Porter contends that nations derive significant competitive and economic advantages when there are concentrations of firms (economic clusters) in home markets of similar or related industries. These competitive advantages derive from the locational relationships of firms within these clusters that result in benefits from knowledge spillover, ease of access to skilled labour,

ENTREPRENEURSHIP AND INNOVATION Vol 9, No 2, 2008, pp 121–131

121

Native American entrepreneurship

better acquisition and assembly of the inputs of production, and competitive pressures to innovate and increase productivity. It has been argued that clusters facilitate complementarities between the activities of cluster participants, producing competitive advantages for cluster members, which can serve to motivate entrepreneurial behaviour (Porter, 2000). What is ambiguous is whether these benefits extend to all business ventures located proximate to clusters. While it is commonly understood that cluster development may serve as a means of improving the economic and competitive position of regions, no research has attempted to see whether the benefits of cluster development extend to all subpopulations within the economy, especially those subpopulations that have traditionally been marginalized in social and economic benefit. In the USA, indigenous Native American entrepreneurs are particularly underresearched, partly due to their comparatively low rates of self-employment. Consequently, identifying factors that serve to motivate or dampen Native American entrepreneurship should be of particular interest to researchers for both their practical implications and their potential influence on the public policy discourse. It is for these reasons that investigations of indigenous entrepreneurship and factors that encourage or hinder its development are of particular import and serve as the motivation for this study.

Background and hypotheses Native American entrepreneurship Much of the recent interest in Native American economic development resides at the level of the individual entrepreneur. Entrepreneurship among Native American populations has been viewed as one means of improving economic conditions for these indigenous groups. Native American gaming has been an important source of economic activity in the USA. However, not all Native Americans benefit from gaming, since it is restricted to certain parts of the country and, where found, gaming pacts with state governments generally prohibit individual casino ownership. Consequently, any study of Native American entrepreneurship by necessity must focus outside of the gaming industry. Native American entrepreneurs For the purposes of this study, the Native American entrepreneur is defined as an owner of a small to medium-sized enterprise (SME), rather than using Schumpeter’s (1934) definition of the entrepreneur as the bearer of new or creative products. This is in part due to the scarcity of SME ownership among Native

122

Americans. In fact, only one in 100 Native Americans can be considered a small business owner (O’Hare, 1992). Across the USA, there are estimated to be only 170,083 reservation-based micro-entrepreneurs (Adamson and King, 2002). Thus, being a business owner is, by itself, a creative endeavour that breaks from the normal economic activity of the population. This condition may be partly due to social and cultural factors that create unique barriers to would-be Native American entrepreneurs. For example, there is a dearth of entrepreneurial role models within the Native American community. This is partly a consequence of the legacy of Native American subjugation brought about by European expansion into North America beginning in the sixteenth century. As a result of westward expansion and the Indian wars of the nineteenth century, Native Americans were forced on to tracts of reservation land where dependence upon the US government for resources became the tradition. Entrepreneurial spirit was supplanted by dependence upon social welfare programmes subsidized and controlled by the US government. Still other social and cultural differences have been found to create distinctive barriers to Native American entrepreneurship. For example, Native American entrepreneurs have viewed themselves as generally having less business education, lower aspirations, poorer communication skills and more racial discrimination than non-Native American entrepreneurs (Garsombke and Garsombke, 2000). Thus, Native American entrepreneurs see themselves as having greater barriers to entrepreneurial success than do non-Native Americans. The obstacles mentioned above do not appear to be uniformly experienced within the Native American community. That is, there appear to be differences in beliefs between Native American entrepreneurs who are independent of tribal funding and those who are tribally funded. In a study of Native American entrepreneurship in the north-west USA, Stewart and Schwartz (2007) found that Native American entrepreneurs who were supported with tribal funding perceived higher levels of technical and cultural support than did their counterparts who were independent. However, the quality of technical support available was perceived to be relatively low. Their findings bolster other research (eg Adamson and King, 2002), which found that access to financing coupled with limited business expertise and experience constituted significant barriers for Native American entrepreneurs who were reservation-based. Several other research reports, including the Harvard Project and a Corporation for Enterprise Development (CFED) study on Native American entrepreneurship (Malkin et al, 2004) have concluded that business training and technical support are some of the most important yet most

ENTREPRENEURSHIP AND INNOVATION Vol 9, No 2

Native American entrepreneurship

underdeveloped aspects of entrepreneurship development programmes in Native American communities. Access to financial resources continues to be a major obstacle to Native American venture success. There are several factors that contribute to this circumstance. One problem relates to the consequence of reservation geography. Most reservations tend to be rural. As such, access to lending institutions tends to be more restricted than in more urban areas. In addition, relatively few reservations have on-site financial institutions, and those that do tend to lack competition. As a consequence, Native American entrepreneurs located on reservations tend to be given loans from these institutions (to the degree that they are given) with interest rates that are prohibitively high. Furthermore, many other commercial institutions will not serve Native American entrepreneurs who live on reservations due to their higher perceived risk and due to their general lack of collateral. As tribal land is held in trust by the US government, Native Americans living on reservations cannot access their land for use as collateral for common forms of business financing. Other research appears to reinforce the notion that Native American entrepreneurs face unique cultural challenges with respect to entrepreneurship that are not faced by their non-Native American counterparts. Native American culture tends to be less individualistic and more collectivistic in attitude (Stewart and Schwartz, 2007). That is, Native American entrepreneurship is often motivated by economic (as well as non-economic) benefits that may accrue to the extended family as well as to other community members (Malkin et al, 2004). For example, Native American entrepreneurs are more likely to excuse poor employee performance in order to maintain harmony with family members or the tribal community. This approach to conventional business practice may put the Native American entrepreneur at odds with the business goals of the traditional entrepreneur. In fact, even the accumulation of wealth, a common entrepreneurial objective, may conflict with traditional Native American aspirations. It is not uncommon for Native American businesses to extend credit or even sales revenue to extended family members. For instance, Malkin et al (2004) found in their study that tribal members reported uneasiness with accumulating greater wealth than other tribal members. In sum, Native American entrepreneurs face a unique set of technical, financial and cultural challenges with regard to SME success. Economic clusters and their importance Although previously overlooked as a factor in Native American entrepreneurship, there has been a relatively

ENTREPRENEURSHIP AND INNOVATION Vol 9, No 2

large amount of attention paid to the effects of economic clustering on entrepreneurship within the general US population. In the USA, the idea that geographic concentrations of industries within regions could be a source of competitive advantage long pre-dates the current interest in clusters. For instance, Marshall (1895) argued that economies of scale could be derived from the spatial advantages of location by similar firms and suppliers, as well as from the increased economic intensity resulting from agglomeration of these firms. The current conceptualization of economic clusters is the product of several streams of theory, of which the nature of regional economic interaction is but one (Colgan and Baker, 2003). In 1990, Porter suggested that a new approach to economic development was needed and that in advanced economies regional clusters of related industries would be the source of new jobs, income and export growth (Waits, 2000). Three forces were cited for a new approach to economic development: globalization, rapidly changing technologies and declining living standards (Waits, 2000). While globalization is reflected in the growing interdependencies of economies around the world, technological innovation is causing fundamental structural shifts in the national economy; and while previous generations have always expected and achieved a better standard of living than the generations before them, most recently this expectation has diminished as income and wage growth have slowed, and in some cases reversed. Clusters defined. Although a general understanding as to what constitutes a cluster exists, research on the subject has not produced unanimity with respect to a specific definition of the concept. The difficulty in achieving a consensus arises from the difficulties in identifying cluster boundaries. For example, clusters can be described as ‘geographic concentrations of interconnected companies and institutions in a particular field’ (Porter, 1998, p 78) or as groups of related industries located in the same region (Held, 1996). A rather inclusive view of clusters is provided by the National Governors Association (2002, p 9), which defines a cluster as ‘a geographically bounded concentration of similar, related or complementary businesses, with active channels for business transactions, communications and dialogue, that share specialized infrastructure, labor markets and services, and are faced with common opportunities and threats’. Regardless of the lack of a specific definition, there is some consensus among researchers that clusters represent geographic concentrations of businesses that share related production inputs, specialized labour pools, distribution and communication channels and network associations.

123

Native American entrepreneurship

Identifying clusters. The process of identifying a cluster usually begins by measuring the number of firms and employees by industrial sector using association directories and existing databases (eg County Business Patterns or Employment Service 202 Reports). These data are used to assess the relative concentration of related firms in a particular location. The ratio of employment or companies in a cluster to the same ratio for the nation generates a location quotient. A location quotient greater than 1 denotes a higher-than-average concentration of related firms in an area, indicating a cluster (National Governors Association, 2002). Cluster types. The literature is consistent in describing three cluster categories that account for the majority of cluster development. Traded clusters are those developed around traded industries. Traded industries are not resource-dependent and sell products and services across regions and sometimes across countries. They locate based upon competitive considerations, and employment in these industries varies markedly by region (Porter, 2003). Examples of traded industries include motion picture and videotape production and automobile assembly, as seen in Table 1. On the other hand, local clusters develop around local industries where employment is generally evenly distributed across all regions and is roughly proportional to the regional population. These industries provide goods and services primarily to the local market or region in which the industry is located (ie real estate, restaurants, retail). Lastly, natural endowment clusters develop from industries associated with the utilization of natural resources, and consequently employment is located primarily where the natural resources are found. Examples of natural resource industries include logging, mining and snow skiing. Clusters and competitive advantage. It has been argued that clusters provide constituent firms with a competitive advantage not afforded to more dispersed firms (those not in a cluster). Porter (2000) contends that these competitive advantages arise from three factors influenced by clusters: (a) increases in current productivity of constituent firms or industries; (b) increasing new business formation that supports innovation and productivity growth; and (c) stimulating new business formation that supports innovation and expands the cluster (p 21). The literature is consistent in reporting that clusters provide for increased returns to scale (Hill and Brennan, 2000; Porter, 1998, 2000, 2003; Stuart and Sorenson, 2003; Waits, 2000). These returns are derived from increased productivity in cluster firms resulting from better access to specialized inputs and employees. In this respect, locational advantages accrue from being

124

Table 1. Traded clusters in the US economy. Aerospace engines Aerospace vehicles and defence Agricultural products Analytical instruments Apparel Automotive Biopharmaceuticals Building fixtures, equipment and services Business services Chemical products Communications equipment Construction materials Distribution services Education and knowledge creation Entertainment Financial services Fishing and fishing products Footwear Forest products Furniture Heavy construction services Heavy machinery Hospitality and tourism Information technology Jewellery and precious metals Leather products Lighting and electrical equipment Medical devices Metal manufacturing Motor-driven products Oil, gas products and services Plastics Power generation and transmission Prefabricated enclosures Processed food Production technology Publishing and printing Sporting, recreational and children’s goods Textiles Tobacco Transportation and logistics

Source: Porter (2003).

in a cluster and provide less costly access to specialized inputs such as components, machinery, business services and skilled personnel (Porter, 2000). As such, clusters are seen as being more competent in assembling the inputs needed for more efficient delivery. The specialized inputs needed for new firm formation (eg capital, skilled labour, suppliers, etc) are more easily accessed within clusters, and cluster economies make it more attractive for new firms to locate within these concentrations (Bresnahan et al, 2001; Enright, 2000). In addition, increased productivity results from specialized access to information (both market and technical) available to cluster members. This specialized access is the consequence of co-locating with supply and technological sources located within the cluster, which in turn facilitates communication flows. Clustering has also been associated with higher levels of innovation. Porter (1998, 2000) explains that innovation results because firms within a cluster are better able

ENTREPRENEURSHIP AND INNOVATION Vol 9, No 2

Native American entrepreneurship

to perceive new buyer needs as well as new technological, operating, and/or delivery possibilities more clearly and rapidly. Other studies also reinforce the notion that clusters increase innovation (Audretsch, 2001; Audretsch and Stephan, 1996; Audretsch and Feldman, 1996; Austrian, 2000; Hendry et al, 2000; Kenney and von Burg, 1999; Steinle and Schiele, 2002; Walcott, 1999). Hendry et al (2000) demonstrated that clusters, because of their network relationships, provided a locus for innovation activity, especially for high-technology sectors. Austrian (2000) argues that in clusters, the relationships that exist between customer and supplier industries and the organizations that support them create synergies that result in more innovation. These innovations then lead to enhanced cluster competitive positions and the strengthening of local economies. Baptista and Swann (1998) showed that a firm was more likely to innovate if located in a region where the presence of firms in its own industry was strong. Clusters and Native Americans. Clusters have been shown to provide advantages to cluster firms partly due to their proximity and access to the inputs needed for opportunity exploitation (eg capital, skilled labour, suppliers). Consequently, cluster development creates a competitive environment in which the barriers to entry and success are lower for the would-be entrepreneur. However, many Native Americans live on federal reservations that are often located in rural areas away from economic clusters. Thus, it may be difficult for many Native American entrepreneurs to take advantage of the benefits associated with economic clusters. Still, many Native Americans live away from their indigenous reservations and some federally recognized reservations are located in or near urban areas that have economic clusters. Native American entrepreneurs who are located proximate to cluster economies should share the same competitive advantages presented to other firms located proximate to clusters, and thus should be competitively advantaged compared with their counterparts who are reservation-based. As such, the unique challenges faced by Native American entrepreneurs may be somewhat mitigated by the comparative advantages afforded by being located near economic clusters. Therefore, we should expect: H1: For Native American entrepreneurs there will be a positive relationship between business performance and proximity to economic clusters. One of the unique challenges faced by Native American entrepreneurs emanates from cultural expectations placed upon the Native American entrepreneur, which may be at odds with mainstream conceptions of profit-

ENTREPRENEURSHIP AND INNOVATION Vol 9, No 2

able business practices (for instance, the collectivist sharing of resources and rewards with non-employees). These differences may be more prevalent in Native American businesses that are reservation-based than in those that are not. Thus, an alternative geographic explanation for firm performance is that firms experience differential effects by being away from a reservation. One reason for this is that Native Americans who live and work away from tribal reservations have, by their choice to leave the reservation, already made a break from the cultural norms of the Native American community. Consequently, they may be less likely to feel bound by prevalent Native American business conventions that would be constantly reinforced within reservation communities. As a consequence, we might expect that, less constrained by cultural convention, Native American entrepreneurs within the greater economy would feel more free to pursue individual economic gain than those located within reservation economies. Furthermore, reservation-based firms might become insulated from the benefits of cluster environments, which facilitate benchmarking between competing firms, motivate competition and provide constituent firms with competitive advantages. Thus, we might expect that Native American entrepreneurs located away from reservation economies would exhibit better business performance than Native American entrepreneurs who are not: H2: Native American entrepreneurs located off reservations will perform better than those located on reservations. It has been shown that cluster economies provide firms with advantages that are not accorded to firms outside these economies. As a result, research shows that more new firms are created within clusters than outside of clusters. One factor that motivates firm creation is the accessibility to skilled labour needed to exploit a business opportunity. Skilled labour is more readily attracted to cluster economies due to the perceived heightened prospects for employment and, consequently, labour is more readily available to cluster firms. Because many tribal reservations are located in rural areas, access to specialized labour might be hindered due to geographic constraints. This would not be the case for Native American entrepreneurs located within cluster economies that are located in urban areas. This advantage should manifest itself in the employee base of Native American firms located within clusters: H3: For Native American entrepreneurs there will be a positive relationship between firm size and proximity to cluster development.

125

Native American entrepreneurship

Methods Data The hypotheses were tested using organizational-level data on Native-American-owned firms located in the US north-west states of Washington, Idaho, Oregon and Montana. Data on each firm were taken from the US Small Business Administration (SBA)’s Central Contractor Registry (CCR). The CCR is a self-certifying database of all firms wishing to do business with any branch of the US federal government. Data were extracted from the CCR database using the CCR’s Webbased Dynamic Small Business Search tool, which allows users to search the CCR database for firms that meet the SBA’s criteria for small business. Small business definitions vary according to each firm’s North American Industry Classification System (NAICS) code and are summarized in the SBA’s ‘Table of Small Business Standards’ (SBA, 2006). All Native-Americanowned firms in the north-western USA that were registered in the CCR were included in the sample. Model and dependent variable Hypotheses were tested using standard ordinary least squares (OLS) regression using a firm’s reported annual revenue and number of employees as dependent variables. As one might expect from data on small firms, these variables were skewed to the left, which required use of the natural logarithm in order to normalize the distribution of the dependent variable in the regression models. Independent variables On reservation. A binary variable was created to measure if a firm was located on a federally recognized American Indian reservation. These data were not available directly from the database, so we used firmlevel postal code data to infer location on a reservation. Because there is no official list of which US postal codes are on reservations, these data were created by cross-referencing geospatial maps of federal reservation areas and 5-digit Zip Code Tabulation Areas (ZCTA) using the US Census Bureau’s mapping tool on its ‘American FactFinder’ Website (Census Bureau, 2006). Rural. A firm was categorized as residing in a rural area if its postal code was not in a federally recognized Metropolitan Statistical Area (MSA) or Micropolitan Statistical Area. Thus, according to definitions used by the US Census Bureau, all firms that were located in a census area in which the population was under 10,000 were coded as rural.

126

Economic clusters. In order to measure the effects of being located near economic clusters, we include a variable that measures the number of traded economic clusters that are located in the same MSA as the focal firm. A list of the MSAs included in the study is attached in the Appendix. Traded clusters consist of traded industries (those industries that sell products and services across economic areas). A traded economic cluster is defined according to its Location Quotient (LQ), or the concentration of a cluster in each region relative to the national average. LQ was calculated according to the following formula: LQ = (Ei,j /Ei) / (Eus,j /Eus) where Ei,j refers to MSA i’s employment in industry j, Ei is the total employment in MSA i, Eus,j is the total US employment in industry j, and Eus is total US employment. Raw employment data for the LQ calculations were extracted from the Harvard Business School Cluster Mapping Project database (ISC, 2006). An economic cluster is said to exist if the LQ score for a given industry is equal to or greater than 1. In other words, if an MSA has a higher than average amount of employment in a given industry (relative to all MSAs in the USA), then it has an economic cluster for that industry. In order to control for qualitative differences between regions that had multiple clusters, a sum of the number of traded economic clusters existing in an MSA was included in the models. For firms that were not located in an MSA, this variable was coded as 0. Control variables In order to control for firm-specific attributes, we included measures of a firm’s age (in years). In the models for revenue, we controlled for size (number of employees) and in the models for size, we controlled for firm annual revenue. Firms are not required to list their primary NAICS (industry) code in the CCR database, so it was not possible to control for industry-specific effects due to a large amount of non-reported data.

Results Descriptive statistics Table 2 shows the descriptive statistics for the variables in the study. There were 858 firms in the analysis. The average firm was nearly 10 years old, with annual revenues of $1.48 million and 11 employees; 38% of the firms were located on reservations and 34% were located in rural areas.1 On average, each firm was located in a census area that had 5.55 traded economic clusters. This value ranged from zero for firms located in rural areas to a high of 18 for firms located in the Bellingham, Washington metro area.

ENTREPRENEURSHIP AND INNOVATION Vol 9, No 2

Native American entrepreneurship Table 2. Descriptive statistics. Variable Annual revenue ($) Age (in years) Number of employees On reservation Rural Number of traded clusters in firm’s census area

Mean

Std deviation

1,482,274.98 9.68 11.12 0.38 0.34 5.55

5,020,665.69 10.76 33.71 0.49 0.47 6.09

Minimum 0 0 0 0 0 0

Maximum 69,000,000 91 500 1 1 18

Note: N = 858.

Table 3. OLS estimates of LN firm revenue. Variable Constant Age (in years) Number of employees On reservation Rural Number of traded clusters in firm’s census area R-squared F

Model 1 Coefficient Std error 10.196 0.079 0.029

0.140 69.935***

0.159*** 0.011*** 0.004***

Model 2 Coefficient Std error 10.320 0.078 0.029 –0.295

0.142 47.142***

0.189*** 0.011*** 0.004*** 0.242

Model 3 Coefficient Std error 10.510 0.075 0.028 0.148 –0.978 0.154 38.874***

0.195*** 0.011*** 0.003*** 0.272 0.280***

Model 4 Coefficient Std error 9.872 0.077 0.028 0.202 –0.387 0.073

0.293*** 0.011*** 0.003*** 0.271 0.344 0.025**

0.162 33.072***

Note: N = 858; ***p < 0.001; **p < 0.01.

Multivariate analyses Table 3 presents the results of ordinary least squares regressions for the natural log of firm revenue. Model 1 includes only the control variables of age and firm size. Both were highly significant positive predictors (p < 0.001) across all models. Model 2 adds the variable for being located on a reservation. Surprisingly, the coefficient for this variable is negative but not significant. Being located on a reservation has no statistically significant relationship to firm sales. Model 3 adds the variable for being located in a rural area. This relationship is negative and highly significant; suggesting that being in a rural area hinders firm performance. It is also interesting to note that, once rural area is controlled for, the coefficient for being on a reservation turns positive. Thus, the negative effect of being on a reservation is really an artefact of being rural. Model 4 adds the variable for the number of traded clusters in a firm’s census area. Being located proximate to economic clusters is positive and highly significant (p