SiKi zEi'Y THE COMPETITIVE BEHAVIORS OF SMALL RETAILERS: EXAMINING THE STRATEGIES OF LOCAL MERCHANTS IN RURAL AMERICA Michael J. Rubach University of Central Arkansas mrubach@mai Iuca. edu Jeffrey E. McGee The University of Texas at Arlington jmcgeeuta. edu ABSTRACT Mass merchandisers, discount chain stores, category killers, and e-commerce are changing the retail industry through innovations in marketing and service. Driven by increased competition and discriminating customers, small, local retailers are searching for ways to gain and sustain competitive advantage. This study examines the competitive behaviors and performances of 236 retailers located in rural Nebraska. The findings indicate that retailers which adopted competitive behaviors with a goal of low cost/low price leadership were most successful. However, retailers using a combination or value-oriented strategy were also successful. Retailers which adopted competitive behaviors that constitute solely a differentiation strategy and those which had no clearly defined strategy were the worst performers. These findings suggest that in order to achieve competitive advantage, rural retailers should follow a value-oriented strategy, being effective at both cost reduction and competitive pricing while offering unique products and services. INTRODUCTION Shifting demographics and changes in consumer purchasing patterns, coupled with new and stronger competition, are putting increased pressure on local merchants. E-commerce and mail order threaten the very concept of "brick and mortar" retailing. Large mass merchandisers and discount chain stores are altering the landscape of retailing across America (McGee, 1996). In these increasingly competitive environments, small local retailers are searching for new formulas to achieve competitive advantage. Despite the challenges of the competitive environment, many retailers are thriving, and retailing continues to be an attractive avenue for entrepreneurs. However, little empirical research has attempted to systematically identify which specific competitive behaviors actually help small merchants cope with the changes occurring within the retailing industry. The literature provides little guidance on how small local merchants can compete effectively in this dynamic environment. This study attempts to help fill this void by examining the relationships between performance and the competitive behaviors of 236 retailers located in rural Nebraska. 65 Journal of Small Business Strategy Vo/. /2, No. 2 Fall/Winter 2002 Models of competitive advantage should attempt to capture the finer-grained patterns of activities (Mauri & Michaels, 1998; Porter, 1996). Prior research has addressed the wide variety of issues related to the appropriateness of competitive behaviors among larger organizations and manufacturing and other non-retail businesses. This study's contribution is its examination of the competitive behaviors of local retailers. While retailers enact identifiable business level strategies, they most likely do not set down or consciously develop these strategies. This study's premise is that the existence and effectiveness of certain competitive behaviors among small merchants may not be readily apparent. Specifically, this study first identified distinguishable patterns of competitive retail behavior using taxonomy- building procedures. These procedures produced four generic retailing types which reflect how the retailers compete. These four retailing types were then compared across a series of performance measures to determine which competitive behaviors, if any, lead to competitive advantage (i.e., superiority in performance outcomes). STRATEGIES FOR RETAILERS lt should not be presumed that small retailers cannot be innovative in response to the competitive behaviors of larger competitors such as large mass merchandisers, category killers and discount chain stores (Kirchhoff, 1994). Small businesses, despite the liabilities of newness and smallness, adopt distinguishable competitive methods or behaviors, whether articulated or not (McGee & Rubach, 1997; McDougall & Robinson, 1990). These competitive behaviors are a major factor in their performances (Droge, Vickery, & Markland, 1994). While there is some evidence of a general lack of variety among the competitive behaviors associated with small business success, small business owners do have options in their sources of advantage (Chaganti, Chaganti, & Mahajan, 1989), and the range of choices may be more robust than previously thought (Carter, Steams, Reynolds, & Miller, 1994). Competitive behaviors, methods, or items are viewed as potential strategic abilities which a firm acquires, sustains, or improves with a goal of either differentiation or lower costs (Droge et al., 1994; Miller, 1988; Porter, 1980). Early research debated whether small businesses should avoid direct competition with large firms and adopt "niche" strategies. Small firms, with limited resources, skills, and capabilities, were advised to design specialized products or services for targeted markets which were generally overlooked by larger, more established firms (Carter et al., 1994). Competition is dynamic and firms advance by continuously taking actions and responding to the actions of their competitors. Strategic management research focuses on the competitive behaviors of firms and on firm performance as an outcome of a series of competitive actions (D'Aveni, 1994). Traditional strategic thought suggests that organizations should pursue a singular strategic approach of either low cost leadership or differentiation (Porter, 1980, 1985; Cappel, Wright, Wyld, & Miller, 1994). Low cost leadership suggests that firms seek to be the low cost producer/provider in a market or industry. For retailers, this often translates into a no-frills approach because the primary means of attracting customers is through lower prices. This competitive behavior is offen associated with cost controls and economies of scale. Achieving cost leadership typically requires a high relative market share and aggressive pricing (Porter, 1980; Miller & Friesen, 1986). Above average returns are anainable because cost leaders can match the prices of their most eAicient competitors. Retailers can typically achieve cost advantages by adopting the following competitive behaviors: inventory control methods, efficient transportation systems, purchasing practices, such as quantity discounts, elTicient staffing, use of new technologies, which include point-of- sales technologies and computers, and efficient use of Aoor space (Cappel et al., 1994). This 66 Journal ofSmall Business Strategy Vol. 12, No. 2 Fall/IVinter 2002 approach is oAen identified with mass merchandisers such as Home Depot, Kmart, and Lowe's. Small discount chains such as Dollar General and Family Dollar have also succeeded by implementing a low cost strategy. Differentiation, on the other hand, characterizes firms striving to develop a product or service that is unique. Differentiation can be based on image or exclusivity, grounded in high quality merchandise, unique products, or superior service (Brennan & Lundsten, 2000). Its aim is to create brand loyalty and price inelasticity. Differentiation is tailored to the retailer's target market segment and customer characteristics. It is oAen identified with specialty retailers such as The Body Shop, Sharper Image, and Brookstone. It is also most closely identified with small retailers that adopt competitive behaviors aimed at specific consumers or product niches. The methods used to differentiate are almost unlimited. Porter (1980) groups all forms of differentiation into one category, while Mintzberg (1988) argues that low-cost leadership is but another form of differentiation, and that cost leadership, based upon cost minimization, does not create competitive advantage by itself. For Mintzberg, low-cost leadership must result in below average prices to provide competitive advantage. Kotha & Vadlamani (1995) found empirical support for Porter's typology of distinct generic business level strategies, but little support for Mintzberg's typology. For purposes of this study, low-cost leadership is deemed to produce low price leadership, and the constructs of low cost leadership and low- price leadership have been consolidated in this study as one competitive behavior. Questions remain as to which competitive behaviors are most appropriate for small retailers. While small business strategists have begun to examine which specific competitive behaviors a firm should adopt, the classification schema of generic business strategies continues to be useful in characterizing the goals of competitive behaviors on the simplest and broadest level. The generic strategies framework accentuates the need for companies to choose specific competitive methods and to make trade offs in adopting specific competitive behaviors (Porter, 1996). Clearly Defined Competitive Behaviors Historically, strategic management theory has suggested that using a single, clearly defined strategy was most appropriate in achieving competitive advantage, so experts advised organizations to adopt clearly defined strategies that emphasized either superior differentiation or lower overall costs and prices. The early literature advised small business owners to pursue a narrow or focused strategy, avoiding direct competition with large firms (Broom & Longnecker, 1979). Firms were advised to concentrate on specialized products or localized operations or on market segments where service and customization could create unique advantages (Hosmer, 1957). Firms lacking emphasis and clarity or following inconsistent strategies or strategies with no apparent focus are characterized as "stuck in the middle" or "straddlers" (Porter, 1985; Hodgetts, 1999). These firms suffer lower profitability because they lack market share, do not possess the resolve to pursue a low-cost approach, have not differentiated their products or services, or have not focused upon a specific segment. By trying to be all things to all people, they set themselves up for mediocrity (Hodgetts, 1999;Porter, 1985). Prior research has generally supported the notion that a lack of strategic clarity (i.e., straddling or being stuck in the middle) can be detrimental to an organization's performance. For example, Robinson and Pearce's (1988) cross-industry study strongly indicated that firms which pursued inconsistent strategies were underperformers. Hooley, Lynch, and Jobber's 67 Journal ofSmall Business Strategy Vo/. /2, No. 2 Fa///IV/uter 2002 (1992) study of UK retailers provided similar empirical support for the pursuit of a single, clearly defined strategy over a muddled or inconsistent strategy. Finally, Conant, Smart, and Solano-Mendez's (1993) study of small apparel retailers also suggests that firms which have chosen to compete in clearly defined ways outperform those firms whose competitive behaviors are characterized by a lack of emphasis or clarity. The patterns of competitive behavior should constitute a specific strategy (Droge et al., 1994). Theory suggests that firms which follow a discrete/distinct strategy will outperform those firms which follow a strategy that positions them as "stuck in the middle" or "straddlers." This leads to the first hypothesis: Hypothesis it Rural retailers whicli follow a discrete generic strategy will outperform rural retailers which have an inconsistent or muddled strategy. Value-Oriented or Combination Competitive Behaviors Although there is general agreement that an organization's competitive behaviors should constitute a clearly defined strategy, many experts argue that companies should not limit their competitive tactics to either differentiation or cost/price leadership. Instead, they suggest a value-oriented strategy which encompasses both differentiation and cost/price leadership. Porter (1980, 1985) has argued that low cost and differentiation strategies were diametrically opposed, and that firms in most circumstances could not effectively pursue both strategies (Hodgetts, 1999). Murray (1988), Wright (1987), and Miller (1992), all counter that mixed or hybrid strategies (value-oriented strategies) have distinct advantages and that pursuing a discrete strategy may be dangerous, leading to lower performance. For example, the goal of differentiation is consistent with a low-cost strategy or position especially when the product or service can be differentiated, switching costs are low or reasonable, and there is a potential for cost reductions (Hill, 1988). Thompson and Strickland (2001), among others, have delineated a strategy which contains aspects of both differentiation and low-cost. We have called this generic business strategy: value-oriented (see Figure I ). Customers are becoming increasingly discriminating in their purchasing habits and are demanding more product or service for less money. Competitiveness among providers of such products or services is quickly becoming dependent upon a merchant's ability to offer higher quality products or unique products at lower costs. Providing value to the customer has emerged as the primary goal of many organizations, and a multifaceted or combination strategy provides an attractive means of achieving this objective. Achieving both cost leadership and differentiation may seem inconsistent because differentiation is usually costly. For Porter (1996), the generic business strategies framework continues to identify the need to choose between these contradictory strategies and make tradeo/Ts between these incompatible positions. While Porter admits that a company can pursue both lowest cost and highest quality at the same time, he continues to argue that it is rare that a company can improve both simultaneously, unless the competition is inept (Hodgetts, 1999). However, there is empirical support for a combination or value-oriented strategy. Miller and Dess (1993), for example, found that not only are combinations of generic strategies possible, but that the combination strategies are also quite profitable, especial)ye combination of low cost and high differentiation. Experts have suggested that 68 Journal ofSmall Business Strategy Vol. 12, No. 2 FalllIVinter 2002 Figure I The Five Generic Competitive Strategies TYPE OF COMPETITIVE ADVANTAGE BEING PURSUED Lower cost Differentiation A Broad Cross-Section I A Narrow Buyer Segment (or Market Niche) retailers can use a combination of low-cost behaviors together with differentiation (Pitelis & Taylor, 1996; Cappel et al., 1994). Moreover, the evidence suggests that certain retailers have adopted competitive behaviors characterized by efforts at both differentiation and cost leadership (Helms, Haynes, & Cappel, 1992). A retailer can pursue a strategy based upon image or exclusivity and high quality merchandise, while maintaining low costs through the competitive methods of efficient staffing, space use, and inventory controls. The cost leader can preempt rivals by creating entry barriers, while differentiation creates a sustainable competitive advantage through its responsiveness to customer tastes. The value-oriented or combination strategy has been identified with image conscious retailers such as Dillard's Department Stores (Cappel et al., 1994) and Marks & Spencer, the UK apparel retailer (Pitelis & Taylor, 1996). The potential for cost savings exists in a rural market. Selling costs tend to be lower due to family ownership and operation, and property rental costs and employee wages are also favorable (Lowry, 2000). These factors lend themselves to the pursuit of a low-cost focus and the adoption of a combination or hybrid strategy. This leads to the second hypothesis: Hypothesis 2: Rural retailers whose competitive behaviors emphasize a value orientation comprising both a high degree of differentiation and efforts at pricelcost leadership will out perform all other rural retailers. METHODOLOGY Context and Sample The unit of analysis for this study is the local store and its competitive behaviors. This study's context is the rural market where a mass merchandiser (Wal-Mart) had recently opened a 69 Journal of Small Business Strareg3 Vol. 12, No. 2 Fall/IVinrer 2002 store. The sample was drawn from all retailers in the counties surrounding four Nebraska cities with populations of less than 25,000 inhabitants and where the nearest metropolitan area was at least twenty miles away. The sample was equivalent to a census (i.e., a total population) of the 658 retail merchants in and around those four communities. A cross- industry study was used to avoid the problem that the competitive methods utilized by the merchants evolved only from unique industry characteristics (Carter et al., 1994). Two hundred thirty-six useable responses were received, a response rate of approximately 36 percent. To detect any potential non-response bias, early and late responding firms were compared (Armstrong & Overton, 1977). This extrapolation method assumes the late or last respondents in a sample are similar to theoretical non-respondents. The similarities which were found between the early and late responders in this study can be interpreted as suggesting the absence of non-response bias (Miles & Arnold, 1991).Although this evidence indicates that non-response bias is not a major problem in this study, it cannot be assured that the non-respondents did not differ in important ways from the respondents. While it is unlikely that sample selection bias drove the results, more confident generalization would only be demonstrated by additional studies. Sam pie Profile Most of the retail stores in the sample were small in terms of annual revenues and number of employees. Nearly 64 percent of the sampled stores reported annual sales of less than $500,000. Roughly 78 percent of the businesses employed fewer than five employees with a median of three full-time employees. These retailers would be classified as small according to the Small Business Administration which defines a small retail business as a retail store having fewer than 100 employees (Gaskill, Van Auken, & Manning, 1993). These retailers are also located in rural areas. A rural area is an area located outside a United States Census Bureau's "Metropolitan Statistical Area" (MSA) (Trinh & O'onnor, 2000). All of the counties covered by the sample are not in a MSA; the only MSAs in Nebraska are those in Lincoln —Lancaster Country, and the Omaha metropolitan area - Douglas County. In terms of business longevity/age, 29 percent of the respondents reported that their stores had been in business for less than ten years, while 45 percent indicated they had been in existence for more than twenty years compared to the national average of 34'/s (see table I). Compared to retail industry standards, the sample's demographic characteristics are quite similar to national averages in the areas of employment and annual sales (Statistical Abstract of the United States, 1998). In terms of business longevity, sample respondents appear to be somewhat more mature than the national average. In any event, the sample appears to adequately reflect the population of small retailers in the United States. Survey Instrument A questionnaire was used to collect the data. A modified version of Dillman's (1978) total design method was used. The questionnaire was pretested on six retailers (not included in the sample) to determine if there were any interpretation difficulties. No problems were noted and the pretest respondents were not troubled by any of the questions or their abilities to rate their companies. A booklet comprising a cover letter of explanation and the questionnaire, together with a postage-paid return envelope, was sent to each retailer. A reminder postcard was mailed approximately one week later to all potential responders. Finally, duplicate booklets with a follow-up cover letter and postage-paid return envelope were re-sent to nonrespondents approximately three weeks later. The questionnaire items requested merchants to identify and rate their competitive behaviors along 24 dimensions using a five-point Likert scale. The scale was anchored by the labels "No emphasis" and "Major, constant emphasis." The competitive behaviors comprised 70 Journal ofSmall Business Strategy Vol. 12, No. 2 Fall/IVinrer 2002 activities which local managers/owners can implement. The dimensions of competitive behavior were adopted from previous empirical studies (Robinson & Pearce, 1988; Conant et al., 1993) because they have been shown to be valid measures for ascertaining the archetypes of generic strategies (Kotha & Vadlamani, 1995; Dess &. Davis, 1984). Slight modifications were needed to improve the appropriateness of certain items because the previous studies addressed the competitive behaviors of manufacturing firms rather than small independent retailers. Such modifications do not, however, alter this proven method for identifying which competitive behaviors constitute generic strategies. Table 1 Summary Statistics (n=236) Annual Revenues (in dollars) Percent Less than 100,000 18 100,000 to 250,000 21 250,000 to 500,000 24 500,000 to 1 million 20 1 to 5 million 13 More than 5 million 2 TOTAL 100 Employment (number of full-time employees) Percent Fewer than 5 78 Between 5 and 10 14 Between 10 and 15 5 More than 15 3 TOTAL 100 Store Age (number of years) Percent Less than 5 8 Between 5 and 10 21 Between 10 and 20 26 More than 20 45 TOTAL 100 Three measures of performance were used in this study. The performance of the sampled firms was operationalized using subjective self-report measures (Dess, Lumpkin, & Covin, 1997). Each local retailer was asked to compare its performance to other local retailers in the community along three dimensions: net income after taxes, total sales growth over the past three years, and overall store performance/success. For this performance measure, respondents were requested to use a seven-point scale anchored by the labels "Much Better" and "Much Worse." This subjective method was chosen over objective data because small firms are often reluctant to disclose financial information, objective data are not readily available, and reporting of the data may be inconsistent across industries (Hooley et al., 1992). Also, due to the cross-industry design of this study, making objective comparisons may be misleading (Covin & Slevin, 1989). Previous research has found that subjective assessments of organizational performance are quite consistent with objective performance data both internal and external to the organization (Hooley et al., 1992; Dess & Robinson, 1984). Also, organizational performance measures when assessed comparatively or relatively tend to be more meaningful (Conant et al., 1993). The use of subjective self-reports for small business 71 Journal ofSmall Business Strategy Vol. /2, No. 2 Fall/Winter 2002 performance has been used extensively in prior research and is recognized as an appropriate methodology to determine performance (Dess et al., 1997). The questionnaire also solicited descriptive information regarding merchandise categories, age/years in operation, approximate annual sales, and number of employees. RESULTS One purpose of this study was to determine whether patterns of competitive behavior among rural retailers exist. This study did not presume that certain strategies exist or that small business owners can identify or articulate their strategies or the goals of their competitive behaviors. In all likelihood, rural merchants did not consciously develop or set down in writing their strategies. Accordingly, factor analysis of the responses was used to identify how many distinct patterns of competitive behavior actually existed. The details of the factor analysis are provided in the appendix. Another purpose of this study was to determine whether retailers are pursuing clearly defined patterns of competitive behavior. To identify the pattern, cluster analysis was employed. Nonhierarchical cluster analysis techniques are appropriate to identify similar entities from the characteristics they possess (Hair, Anderson, & Tatham, 1987), which is appropriate for identifying those retailers using a distinct, clearly defined strategy or a combination strategy (Hooley et al., 1993). The use of cluster analysis has been quite prevalent in identifying generic business strategies and their performance differences (Wagner & Digman, 1997; Wright, Kroll, Tu, & Helms, 1991; Dess & Davis, 1984; Hambrick, 1983). Prior research of generic business strategies of retailers produced factor and cluster analyses that differ from this study's results. This study used entirely different measures for discerning the presence of generic business strategies, and did not assume that a specific strategy or archetype existed. This study attempted to allow the data to identify the strategies pursued by retailers. None of the prior studies attempted to ascertain whether retailers pursued value-oriented or combination strategies. Helms et al. (1992) in their study of forty publicly-traded retailers with annual sales of more than $ 1 million used archival data and only two strategy variables (low cost or differentiation). They did not consider combination or value-oriented strategies. Their cluster analysis found that few (only eight of 40) companies followed a low cost/low price strategy. Hooley et al. (1992) in their study of UK retailers used five dimensions of marketing strategies: objectives, focus, targeting, quality, and price, which are quite different from the competitive behaviors studied herein. The details of the cluster analysis are also provided in the appendix. The cluster analysis resulted in five clusters. Each cluster grouping was labeled to capture the nature of each strategic behavior: variety discounters, broad differentiators, focused differentiators, muddled strategists/straddlers, and value-oriented merchants. The cluster profiles are presented in table 2. To determine performance ditTerences among utilized strategies, each strategic cluster was compared, using ANOVA and Tukey-Kramer paired comparisons, over the three performance measures (net income after taxes, total sales growth over the past three years, and overall store success or performance). As the ANOVA results suggest (see table 3), significant differences exist on all three performance measures. The Tukey-Kramer paired comparisons also produced identifiable patterns of performance. Retailers which followed a low-cost/low-price variety strategy (Variety Discounters), which is most akin to a low-cost leadership/low-price strategy, outperformed the other retailers on the performance measures of net income a(ter taxes and overall store performance. The "Variety Discounters" also outperformed the "Broad Differentiators" and the "Muddled Strategists/Straddlers" on three years sales growth. Retailers which followed a combination or multifaceted strategy (Value-Oriented Merchants) 72 Journal af Small Business Sira/egp Val. /2, /Vo. 2 Fall/Winier 2002 also reported significantly higher net income aAer taxes and overall store performance than groups following either a differentiation strategy (Broad Differentiators and Focused Differentiators) or the "Muddled Strategists/Straddlersy Also, except for the "Focused Differentiators," the "Value-Oriented Merchants" perceived better three years sales growth than the remaining groups. Only the "Focused Differentiators" exhibited any significantly higher performance among the three lower groups; their three year sales growth was higher. ln terms of failed strategies, the "Broad Differentiators" and the "Muddled Strategists/Straddlers" were clearly the weakest performers. Table 2 Cluster Profiles Cluster I: Variety Discauniers (n = 16) The competitive orientation of this group is one that seeks to emphasize both low prices and a variety of products. Cluster 2: Broad Differenrialars (n = 45) The competitive orientation of this group is one that seeks to offer a wide variety of differentiated products. This group of retailers does not emphasize low prices. Cluster 3: Focused Differenfiarars (n = 30) The competitive orientation of this group emphasizes differentiation and the careful monitoring of competitors'ricing and promotional activities. This group does not emphasize factor 3. Cluster 4: Muddled Siraregisis/Sfraddlers (n = 55) The competitive orientation of this group is unclear. Cluster 5: Value-Oriented Merchants (n = 88) The competitive orientation of this group is one that seeks to provide value to their customers by emphasizing both differentiation and low price. The retailers in this group are not concerned with the pricing and promotional activities of their competitors. Table 3 Results of Analysis of Variance Performance 'p. ten pifferentiators Pitferentiators 'Strategists/ Oriented Unirorior p ' (CI) (C2).,(C3) r Straddlers Merchants P-Voice Ccrnporjsprr (C4) .; (C5); C1 &AII; 5.30 3.9g 4.03 3.90 4.66 3 99"'s&C2,C3, after taxes Three year 5 14 3.85 4.09 3 63 4 53 3 girs ~ CI,C3,C5 sales growth C2,C4 C 1&AII; 5.10 3.65 3.04 3 4g 4 11 3 1orrr C5&C2 C3 performance '"p & 00/ 73 Journal ofSmall Business Strategy Vo/. /2, No. 2 Fall/Winter 2002 Table 3 (continued) Descriptive Statistics for Performance Variables Performance Measure Mean Standard Deviation Net income after taxes 3.35 1.25 Three years sales growth 4.13 1.38 Overall store performance 3.88 1.57 To summarize, a strategy emphasizing low costs/low prices m~a be viable. Only sixteen retailers belonged to this group, however, so this particular finding may be somewhat misleading. A value-oriented or combination strategy, on the other hand, is clearly an effective means of achieving a competitive advantage. The results also clearly suggest that a muddled or ill-defined strategy should be avoided. DISCUSSION The purpose of this study was twofold. First, we attempted to determine if rural merchants exhibited distinct patterns of competitive behavior and then determine whether following a discrete generic strategy produced superior performance. The results of the factor and cluster analysis suggest that they do. To reiterate, we were able to group the sampled retailers into the following five clusters based on their distinct patterns of competitive behaviors: "Variety Discounters," "Broad Differentiators," "Focused Differentiators," "Muddled Strategists/Straddlers," and "Value-Oriented Merchants." More importantly, we discovered that rural merchants pursuing ill defined or muddled strategies were consistently out performed by firms pursuing clearly defined strategies. This finding supports hypothesis one, reflecting that firms suffer if their competitive behaviors lack emphasis and clarity. Our second purpose was to see if certain strategies or patterns of competitive behavior were more successful than others. Specifically, we predicted that retailers exhibiting a combination of differentiation and low-cost /low-price leadership competitive behaviors, termed "Value- Oriented Merchants," would outperform all other merchants. This prediction was only partially supported because the "Variety Discounters" actually outperformed all other merchants including the "Value-Oriented Merchants." Notwithstanding this finding, "Value- Oriented Merchants" reported better performance than every other type of merchant. In other words, merchants whose patterns of competitive behavior were characterized by a desire to offer low prices and a differentiated shopping experience were more successful than most other retailers. This finding is of particular interest because it suggests that the days when high prices meant high quality may be gone (McGee tk Snook, 2000). Today's retail consumers are seeking value, and it is up to individual merchants to offer a shopping experience that matches the consumer's perception of value. Retailers generally understand that their pricing policies have a direct impact on their stores'mage and profitability. However, they are oflen less confident about whether their merchandising and pricing policies equate to value. Consequently, local merchants need to identify and pursue a distinct customer market or niche. They then need to otTer unique, but fairly priced, merchandise or value-added services that appeal to their market segment. Specifically the competitive methods or tactics which were identified with the higher performing strategies cover three areas (factors): superior quality and service, low prices, and product depth and promotion. To compete against mass merchandisers, discounters, and category killers, the results suggest that rural retailers should: ( I) carry a variety of higher quality and unique merchandise, including merchandise which is recognizable to consumers, (2) carry products which cover a wide range of prices, 74 Journal ofSmall Business Srraiegy Vol. 72, No. 2 FaiiylYinier 2002 (3) provide before and after sales service, (4) promote their establishments, including the use of sales, and, (5) be innovative by making the consumers'hopping experience unique. The results demonstrate that the competitive behaviors of low-price offerings, including sales and sales promotions, are effective. Lower costs support the adoption of such behaviors. Retailers can typically achieve cost advantages through the adoption of a number of activities including inventory control methods, efficient transportation systems, efficient purchasing practices, efficient store staffing, and the use of new technologies including point-of-sales technologies and computers. Additional research is needed, however, to determine if a low- cost/low-price strategy is sustainable for small retailers. Regardless, rural retailers should tighten their belts and develop ways of achieving greater efficiencies. They should concentrate on the basics, which are offen overlooked or not well done by larger retailers. Limitations and Directions for Future Research This study examined the competitive behaviors of retailers in rural communities. The competitive behaviors of retailers in urban or large metropolitan settings may be quite different. Urban retailers may be under different pressures to follow a niche or focused strategy. Future research should study competitive behaviors in these other contexts. While size may effectively inhibit rural retailers from emulating the mass merchandisers and large chains, the development and exploitation of a firm's social capital should be an emerging strategy. Small retailers can join together to make volumes purchases and utilize these discounts to create scale economies. Further small firms can cooperate in joint advertising and other promotional activities (Lowry, 2000). Also, civic involvement and social interactions can increase patronage, especially by older consumers (Miller Bt Kim, 1999). Future research should address the effects of the social capital of small retailers on their performance. This study examined small, not necessarily young, retailers or retail ventures. New ventures oflen possess many of the characteristics of small firms, though they oflen lack the advantages which may accrue through age. However, the results indicate that certain strategies may capture the "essence" of strategic positioning and be appropriate for new ventures. Certain strategies, especially value-oriented or combination strategies, were found to be more successful for small retailers. The same array of strategies could be equally applicable to new ventures (Carter et al., 1994). The issues presented should be examined in the context of new ventures. Further, future studies should examine how the competitive behaviors of small merchants are influenced by the continuing advances of e-commerce and mail order. The retail landscape will face considerable upheaval in the coming years, as more U.S. households shop on-line, potentially producing dramatic revenue growth in web sales. CONCLUSION Historically the advice given to small retailers has been to "provide superior customer service." Superior service will reliably bring customers to, and back to, your store (Buss, 1996).Customer responsiveness is the one building block of competitive advantage which can differentiate small and rural businesses from the competition of the mass merchandisers and category killers. The ways or methods to dilTerentiate are almost unlimited. Following a low- cost/low-price strategy without the benefits of economies of scale enjoyed by many discount merchandisers may be difficult for many small retailers. This is especially true in markets that are undergoing dramatic changes. Following a value-oriented or combination strategy, however, may not be a bad thing. This does not mean having no strategic concentration. A 75 Journal of Small Business S(ra/egy Vol. /2, /f(y. 2 Fa///IV/urer 2002 value-oriented or combination strategy of low-cost/low-price and differentiation can be effective; it need not equate to a lack of strategic focus. While further studies are necessary to determine whether value-oriented strategies create sustainable advantages, and under what environmental conditions such strategies are most effective, this study lends support to the idea that single, discrete generic strategies may not be the most advantageous. Retailers should consider adopting competitive behaviors which constitute and provide perceived value to their customers —they should be effective at both cost reduction and competitive pricing while offering unique products and services. REFERENCES Armstrong, J. & Overton, T.S. (1977). Estimating nonresponse bias in mail surveys. Journal of Marketin Research 16, 396-402. Brennan, D.P. & Lundsten, L. (2000). Impacts of large discount stores on small US towns: Reasons for shopping and retailer strategies. International Journal of Retail & Distribution Mana ement 28 (4/5), 155-161. Broom, H. N. & Longnecker, J.G. (1979). Small Business Mana ement. Cincinnati, OH; Southwestern. Buss, D. D. (1996). The little guys strike back. Nation's Business 84 (7), 18. Cappel, S.D., Wright, PD Wyld, D.CD & Miller, J.H., Jr. (1994). Evaluating strategic effectiveness in the retail sector: A conceptual approach. Journal of Business Il I 21. 299-212.C, NM.. S . TM., R y Md, PD. R Mill, BA. (19942. N «gi Theory development with an empirical base. Strate ic Mana ement Journal 15(l), 21-41. Chaganti, R., Chaganti, R., & Mahajan, V. (1989). Profitable small business strategies under differenttypesofcompetition.Entre reneurshi Theo &Practice S rin,21-35. Conant, J.SD Smart, D.T., & Solano-Mendez, R. (1993). Generic retailing types, distinctive marketing competencies, and competitive advantage. Journal of Retailin 69, 254- 279. Covin, J.G. & Slevin, D.P. (1989). Strategic management of small firms in hostile and benign environments. Strate ic Mana ement Journal 10, 75-87. D'A i.RA.(1994(.~Ni .N Y R,NY:9 P Dess, G.G. & Davis, P.S. (1984). Porter's (1980) generic strategies as determinants of «g g 9 9 kip 4 g 1 1 1 9 f . ~Ad f Mana ement Journal 27,467-488. Dess, G.GD Lumpkin, G. TD & Covin, J.G. (1997). Entrepreneurial strategy making and firm performance: Tests of contingency and configurational models. ~Strate ic Mana ement Journal 18(9), 677-695. Dess, G.G. & Robinson, R.B., Jr. (1984). Measuring organizational performance in the absence of objective measures: The case of the privately-held firms and conglomerate business unit. Strate ic Mana ement Journal 5, 265-273. Dillman, D. A. (1978). Mail and tele hone surve s: The total desi n method. New York, NY: John Wiley and Sons. Droge, CD Vickery, S., & Markland, R.E. (1994). Sources and outcomes of competitivedg:A*9MB dyi d f 1 idn,D(i S( 25(5/6), 669-689. Gaskill, L.R., Van Auken, H.E., & Manning, R.A. (1993). A factor analytic study of the perceived causes of small business failure. Journal of Small Business Mana ement 31(4), 18. Hair, J.FD JrD Anderson, R.E., & Tatham, R.L. (1987). Multivariate data anal sis with ~readin s. New York, NY: Macmillan Publishing Company. 76 Journal ofSmall Business Strategy Vol. 72, No. 2 Fall/Winter 2002 Hair, J. F., Anderson, R E., Tatham, R.L., & Black, W.C. (1995).Multivariate Data Anal sis. Englewood Cliffs, NJ: Prentice HalL Hambrick, D. (1983). An empirical typology of mature industrial-product environments. Academ of Mana ement Journal 26,213-230. Helms, M.M., Haynes, P.J., & Cappel, S.D. (1992). Competitive strategies and business performance within the retail industry. International Journal of Retail & Distribution MMM 20(5), 3-1 ~ . Hill, C.W.L. (1988). Differentiation versus low-cost or differentiation and low-cost: A contingency framework. Academ of Mana ement Review 13, 401-412. Hodgetts, Richard M. (1999). A conversation with Michael E. Porter: Significant extension toward operational improvement and positioning. Or anizational D namics 28(1), 24-33. Hooley, G.J., Lynch, J.E., & Jobber, D. (1993). Generic marketing strategies. International Journal of Research in Marketin 9(1), 75-89. Hosmer, A. (1957).Small manufacturing enterprises. Harvard Business Review 35, 111-122. Kirchhoff, B.A. (1994). Entre reneurshi and d namic ca italism: The economics of business firm formation and rowth. Westport, CT: Praeger. Kotha, S. & Vadlamani, B.L. (1995).Assessing generic strategies: An empirical investigation of two competing typologies in discrete manufacturing industries. ~Strate ic Mana ement Journal 16(1), 75-83. L G, J.R. (2000). TS flg0 I' \ Sy I 6 p 0 Il . ~USA T 0 129(2662):22-25. Mauri, A.J. & Michhels, M.P. (1998). Firm and industry effects within strategic management: An empirical examination. Strate ic Mana ement Journal 19, 211-219. McDougall, P.P. & Robinson, R.B., Jr. (1990). New venture strategies: An empirical identification of eight Marchetypes" of competitive strategies for entry. ~Strate ic Mana ement Journal 11, 447-467. McGee, J.E. (1996). When Wal-Mart comes to town: A look at how local merchants respond to the retailing giant's arrival. Journal of Business & Entre reneurshi 8(1):43-52. McGee, J.E. & Rubach, M. J. (1997). Responding to increased environmental hostility: A study of the competitive behavior of small retailers. Journal of A lied Business R I 13(1):83.94. M G , LE. & S I, C.L. (2000). R p dl g I d n Ild I I 0 g 6 industries: The role of capabilities in small business survival. Journal of Small ~Bi M I 1(2),21-32. Miles, M.P. & Arnold, D.R. (1991). The relationship between marketing orientation and entrepreneurial orientation. Entre reneurshi Theo and Practice 15(4), 49-65. Miller, A. & Dess, G.G. (1993). Assessing Porter's (1980) model in terms of its generalizability, accuracy and simplicity. Journal of Mana ement Studies 30(4), 553-585. Miller, D. (1988).Relating Porter's business strategies to environment and structure: Analysis and performance implications. Academ of Mana ement Journal 31, 280-308. Miller, D. (1992).The generic strategy trap. The Journal of Business Strate 13(l), 37-41. Miller, D. & Friesen, P.H. (1986). Porter's (1980) generic strategies and performance: An empirical examination with American data. Or anization Studies 7(1), 37-55. Miller, N. J. & Kim, S. (1999).The importance of older consumers to small business survival: Evidence from rural Iowa. Journal of Small Business Mana ement 37 (4), 1-15. Mintzberg, H. (1988). Generic strategies: Toward a comprehensive framework. Advances in SSSiM t.G I S,CT: JAIP . 5, 1-63. M y,A.I.(1988).A Ig y fP 'g I gl A~df Mana ement Review 13, 390-400. Pitelis, C. & Taylor, S. (1996). From generic strategies to value for money in hypercompetitive environments. Journal of General Mana ement 21(4), 45-61. 77 Journal ofSmall Business Strategy Vol. /2, No. 2 Fail(IVinter 2002 P, M.E. (19999. ~Cii «: T h 9 I i ~ i 9 9 . ~ Y C,NY:Th 9' Porter, M.E. (1985). ComJTetitive advanta e: Creatin and sustainin su crier erformance. New York, NY: The Free Press. Porter, M.E. (1996).What is strategy? Harvard Business Review 74(6), 61-78. Robinson, R.B., Jr. & Pearce, J.A., II. (1988). Planned patterns of strategic behavior and their relationship to business-unit performance. Strate ic Mana ement Journal 9(1), 43- 60. Statistical Abstract of the United States. (1998), [national data book on line] (Washington, D.Cc U.S. Census Bureau, March 1998); available from http: //www.census.gov/prod/3/98pubs/ 98statab/cc98stab.htm; Internet; accessed 19 October 1999. Thompson, A.A., Jr. & Strickland, A.J., III (2001). Craitin and Executin Strate . 12u Edition. Boston, MA: McGraw-Hill Irwin. Trinh, H. Q. & O'onnor, S.J. (2000). The strategic behavior of U.S. rural hospitals; A longitudinal and path model examination Health Care Mana ement Review 25(4): 48-64 Wagner, B. & Digman, L. (1997). The relationships between generic and time-based strategies and performance. Journal of Mana erial Issues 9(3), 334-354. Woodruff, R.B.(1997).Customer value: The next source of competitive advantage. Journal of the Academ of Marketin Science 25 2), 139-153. Wright, P. (1987). A refinement of Porter's strategies. Strate ic Mana ement Journal 8, 93- 101. Wright, P., Kroll, M., Tu, H., & Helms, M. (1991). Generic strategies and business performance: An empirical study of the screw machine products industry. British Journal of Mana ement 2, 57-65. Michael J. Rubach is an Assistant Professor of Management at the University of Centtal Arkansas with a Ph.D. in Management (Business Policy and Strategy) fiom the University of Nebraska at Lincoln, JD. from Creighton University, and MBA. from the University of Nebraska at Omaha. He teaches courses in strategic management, international business and small business management. His current research interests include small business management and corporate governance. His work has appeared in the Journal of IVorld Business, Journal of Business Ethics, Journal of Applied Business Research, International Journal of Organizational Analysis, and Journal of Business dc Entrepreneurship. Jeffrey E. McGee is an Associate Professor and the Chair of the Management Department of the University of Texas at Arlington. He teaches courses in small business management and entrepreneurship. His research interests include new business development and the strategic management of small businesses. Dr. McGee holds a Ph.D. in entrepreneurship fiom the University of Georgia. His work has been published in the Management Science, Journal of Business Venturing, Sirategic Management Journal, Entrepreneurship Theory and Practice, Journal of Small Business Management, Journal of Applied Business Research, Journal oj Small Business Strategy, and Journal ofBusiness & Entrepreneurshi p. 78 ! Journal of Small Business Strategy Vol. 12, No. 2 Fall/Winter 2002 APP,ENDI X Factor Analysis and Cluster Analysis Procedures The twenty four competitive methods (see table 4) were factor analyzed using a principal components analysis with varimax rotation. A factor loading of .50 was used because factor loadings greater than or equal to .50 can be considered very significant: the larger the absolute size of the loading, the more significant it is in interpreting the matrix (Hair et al., 1987). The Kaiser-Meyer-Olkin measure of sampling adequacy (KMO = 0.921) supported the use of factor analysis because it was substantially greater than the required minimum of .50 (Hair, Anderson, Tatham, k Black, 1995). The factor analysis yielded five patterns of competitive behavior (see table 4). These five factors, each with eigenvalues greater than one, accounted for 65 percent of the total variance. To test the reliability of the constructs, coefficient alphas were calculated for the items loading on each factor. They were as follows: factor one (.81); factor two (.73); factor three (.64); factor four (.58); and factor five (.49). Albeit less than optimal, these reliabilities are encouraging because scale development was not the primary purpose of this study. Moreover, they are higher than those found in other studies that used similar emphasis-anchored items to assess the content of strategic behavior. For example, the coefficient reported by Conant et al. (1993)ranged from .74 to .49. The inherent variability that exists in strategic behavior across firms may be the cause of these less than optimal reliabilities. We labeled each of the patte'ms of competitive behaviors used by the retailers (table 4). Factor one —"superior quality and service" - most closely aligns with a goal of differentiation, where image and service are most important. High quality products and a perception of exclusivity are important. Differentiation is often achieved through innovation and intense customer support (Porter, 1980). The competitive methods loading on the second factor- "low prices" - most closely aligns with an overall goal of cost leadership or low prices, where low cost relative to one's competitors is the dominant theme. The second factor captures the relationship between low-cost leadership and low price (Mintzberg, 1988). Factor three— "image creation" - contains competitive methods that would be associated with methods used to achieve differentiation, the use of advertising and civic involvement, and store identification in an attempt to gain higher market share (Porter, 1980). The small independent retailers emphasizing the fourth factor —"product depth and promotion" - use competitive methods that emphasize depth of products and stocking highly recognized products. This factor is not dominated by either a low cost/low price or differentiation strategy, but has methods that are identifiable with both. Stocking highly recognized products can be both low cost and focused. Maintaining high inventories would most likely be associated with a differentiation strategy, as it could entail incurring higher relative costs to carry the products or inventory. The filth factor —"competitive reconnaissance" - recognizes a method of monitoring the practices of one's competitors, both for pricing and promotions. It is aligned more with a process of adding value than with a specific strategy (Woodruff, 1997). Two competitive methods, maintaining high inventory levels and using computers to monitor sales and inventory levels, did not load. Non-hierarchical cluster analysis was applied to the items comprising each'of the five factors (FASTCLUS in SAS). As previously indicated, cluster analysis has been used extensively to study business level strategies. Further, non-hierarchical cluster analysis is less susceptible to anomalies in the data and, thus, less likely to produce misleading results (Hair et al., 1995). A five-cluster solution (with a pseudo F statistic of 43.2 and a cubic-clustering criterion of 4.2) was produced (table 5):'9 Journal ofSmall Business Strategy Va/. /2, Na. 2 Fa/I/tVinter 2002 Table 2 provides a narrative of the competitive profiles of each cluster. Cluster I, "Variety Discounters," (n=16) includes firms that emphasize low cost while maintaining a wide variety of products. Cluster 2, "Broad Differentiators," (n=45) includes firms that adopt competitive methods of providing customer service, carrying high quality, higher priced merchandise, and maintaining a depth of products. These firms do not emphasize relative low cost, but incur costs to achieve differentiation through service, employee training, advertising, and civic involvement. Firms in Cluster 3, "Focused Differentiators," (n=32) are similar to the "Broad Differentiators" in their primary strategy of differentiation; however, they do not emphasize Table 4 Factor Analysis of Competitive Weapons Used by Small Retailers (n=236) , - Factor Lo'adings Competitive Methods Factor I Factor 2 Factor'I,, Factoi 4 Factor' Superior Quality and Service Before-the-sate service .81 Stocking unique products .77 1ligh quality merchandise .74 Carrying higher priced products .68 Concerted etyort to be innovative .66 After-the-sale service .64 Store layout/merchandise presentation 63 Employee training .59 Lotv Prices Canying lower priced product lines .76 Pricing below competitors .74 Carrying a variety of products .58 Holding sales .52 Sales promotion .50 Image Creation Stocking private label products .69 Use of co-op advenising .66 Civic involvement .61 Advertising - general use .53 Product Depth and Promotion Sales promotion programs .69 Depth of product selection .64 Stocking highly recognized products .63 Competitive Reconnaissance Monitoring competitors'ricing .82 Monitoring competitors'romotions .80 EIGENVALUES 4.68 2.14 1.14 1.08 1.02 Cumulative % of variance 42.7 51.8 56.7 61.0 65.0 explained Cronbach's Alpha .81 .73 .64 .58 .49 advertising, private labels, or civic involvement. Instead, they emphasize monitoring both the pricing and promotions of their competitors. This group does not emphasize relative low cost. Cluster 4, "Muddled Strategists/Straddlers" (n=55) includes firms whose competitive orientation is not clear. These lirms do not seem to emphasize either low cost or differentiation, relying instead upon advertising and monitoring of their competitors. Cluster 5, "Value-Oriented Merchants" (n=88) includes firms that follow both general generic 80 Journal ofSmall Business Strategy Vol. 12, No. 2 Fall/tVinter 2002 strategies: differentiation and low-cost/low-price leadership. They are not concerned with maintaining high inventories or monitoring the activities of their competitors. Table 5 Cluster Means Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 1 -0.048 2.83 -0.68 1.80 -0.02 2 0.44 -0.75 0.28 0.89 -0.19 3 0.99 0.00 -1.01 -0.19 1.05 4 -0.81 -0.12 0.11 -0.33 0.13 5 0.98 0.99 0.20 -0.47 -0.80 81