Development and analysis of a supply chain strategy taxonomy.

INTRODUCTION

A review of supply chain (SC) literature reveals that SC management is increasingly recognized as a critical component of a firm's strategic plan (e.g., Frohlich and Westbrook 2001; Vickery Jayaram, Droge and Calantone 2003). Numerous studies present different SC strategies for firms in various circumstances (e.g., Fisher 1997; Lee 2002; Christopher, Peck and Towill 2006). Much research focuses on developing typologies that propose "ideal" types of SC strategies, each one presenting a unique combination of organizational attributes (Doty and Glick 1994; Narasimhan, Kim and Tan 2008). However, there has been limited effort in developing a taxonomy of SC strategies that delimits and classifies different strategy groups "a posteriori" through empirical work and in mutually representative groups (Martin-Pena and Diax-Garrido 2008).

There are two primary objectives of this research. The first objective is to derive a numerical taxonomy that classifies manufacturers with similar combinations of SC capabilities into SC Strategy Groups. This research will contribute to the SC management research by providing a taxonomy that complements the existing typologies of SC strategy. Our taxonomy offers a parsimonious description of SC Strategy Groups that provides insights into current SC strategic positions. The second objective of this study is to explore the relationship between the SC Strategy Groups and contextual factors, competitive priorities and performance of group members. The development of a taxonomy of SC Strategy Groups and the comparative analysis of the groups provides important information about the existing state of SC management.

LITERATURE REVIEW

The literature on the configuration approach and the resource-based view (RBV) establishes the theoretical foundation for our research. This review discusses these two theoretical approaches and relates them to our research on SC strategy taxonomy.

Configuration Approach

A configuration approach is widely accepted in the field of strategy and is applied accordingly to a number of operations management studies (Miller and Roth 1994; Boyer, Bozarth and Mcdermott 2000). The conventional econometric research approach statistically isolates the independent effects of each strategic attribute on a firm's performance. However, the configuration approach considers strategic combinations or gestalts as an interrelated bundle (Hambrick 1984; Bozarth and McDermott 1998). The configuration approach is typically divided into the development of typologies and taxonomies. A typology describes ideal types, each of which reflects a particular combination of organizational attributes (Doty and Glick 1994), although no existing firms may fit exactly the suggested ideal type (Venkatraman and Prescott 1990; Bozarth and McDermott 1998). Taxonomies, without defining ideal types, attempt to classify existing organizational phenomena into mutually exclusive and exhaustive groups (Doty and Glick 1994; Miller and Roth 1994; Bozarth and McDermott 1998).

Numerous works focus on developing typologies that specify types of SC strategies that are contingent upon product characteristics (Fisher 1997; Lamming, Johnsen, Zheng and Harland 2000) and supply and demand uncertainty (Lee 2002). Based on conceptual frameworks developed "a priori" these researchers provide a normative suggestion in which performance is a consequence of the congruency between factors: SC strategy, demand characteristics and level of supply uncertainty. More recently, Narasimhan et al. (2008) developed SC strategy typologies based on the degree of supply network dynamics and the degree of focal firm supply network influence and then tested the typologies using a large sample of U.S. firms. Since typologies are often not confirmed empirically (Meyer, Tsui and Hining 1993; Miller 1996), this study made a contribution by validating its proposed typologies through data analysis.

There has been some effort to develop taxonomies in the field of SC study One such effort is made by Harland, Lamming, Zheng and Johnsen (2001) who develop a taxonomy of supply networks. The taxonomy was initially developed based on a combination of interviews and case studies and was later modified based on a phone survey of 50 firms. Another effort is made by Frohlich and Westbrook (2002) classifying firms based on the degree of integration. Our study differs from Harland et al. (2001) and Narasimhan et al. (2008) in its unit of analysis (manufacturing firms versus supply networks) and from Frohlich and Westbrook (2002) in the number of dimensions used for classification (multidimensional SC capabilities versus a single dimension of integration).

This study makes a contribution to the literature by developing a taxonomy that delimits and classifies different types of firms' SC strategies "a posteriori" by using multiple classification variables supported by a large-scale data analysis of manufacturing firms. Our analysis also provides insight into the characteristics of each SC Strategy Group by examining the context, the priorities and the performance levels of each group within the taxonomy. In addition, since the role of SC management in strategic formulation is dynamic in accordance with ever-changing competitive environments (Frohlich and Dixon 2001), the development of our taxonomy based on more recent data is important to understanding the current state of SC strategy.

The RBV in SC Literature

One of the criticisms of taxonomic study is that classification variables are disconnected from existing theory. As a result, the taxa (strategy groups) are viewed as a result of "data dredging" rather than reflecting actual organizational conditions (Doty and Glick 1994; Ketchen and Shook 1996; Bozarth and McDermott 1998). In this study, the approach of defining SC strategy taxonomy is based on the RBV of the firm, which contends that the resources and nonimitable capabilities of firms are the key sources of sustained competitive advantage (Prahalad and Hamel 1990; Oliver 1997).

The literature on the RBV (Wernerfelt 1984; Barney 1991) identifies conditions and firm-specific factors that underlie the competitive advantage enjoyed by a firm. This perspective views a firm as a bundle of resources and capabilities, with firms differing in their endowments of each. While resources are defined as productive factors that a firm uses to achieve its business objectives, capabilities refer to a firm's ability to "deploy these resources to affect a desired end" (Amit and Shoemaker 1993). Since the "competitive advantage" (Porter 1991) tends to vanish when competitors can readily acquire resources from a factor market, one of the main focuses of the RBV is to determine the firm's capabilities that increase the barriers to imitation by bundling various resources (Dierickx and Cool 1989; Peteraf 1993).

The RBV is widely accepted in the strategic management literature but has only more recently been applied to the study of operations management (e.g., Coates and McDermott 2002; Schroeder, Bates and Junttila 2002). However, it has frequently been suggested that operations strategy content embodies the choice of the set of manufacturing capabilities that becomes the source of competitive advantage (Prahalad and Hamel 1990; Miller and Roth 1994). Hunt and Davis (2008) use the "the resource-advantage research tradition," that combines heterogeneous demand theory with an RBV of the organization, to support the conclusion that purchasing strategy, in particular, and SC management, in general, can lead to a long-term competitive advantage. SC strategy, therefore, can be defined as the choice of a set of capabilities that are developed through a pattern of investments over time and cannot be easily imitated or acquired by trade, nor can good substitutes be found (Dierickx and Cool 1989).

Drawing on the RBV, this study develops an SC strategy taxonomy by using six SC capabilities. These capabilities will be examined in more detail in the next section. The development of an SC strategy taxonomy provides a better understanding of the key SC capabilities that firms bundle together.

CONCEPTUAL FRAMEWORK 

In this section, we present our conceptual framework, the variables examined in our research, and our hypotheses. The conceptual framework for our research is shown in Figure 1. Details of the measurement of each variable in the framework will be discussed in our "Research Methodologies" section.

[FIGURE 1 OMITTED]

We analyze the SC capabilities and identify a taxonomy of SC Strategy Groups that deploy similar sets or bundles of SC capabilities. Then, we examine the contextual factors and the competitive priorities that are common to a particular SC Strategy Group. This helps to identify conditions under which firms develop specific SC capabilities. We also examine the performance of particular SC Strategy Groups, identifying differences in the perceptual and actual performance of the SC Strategy Groups.

SC Capabilities and SC Strategy Groups

In this study, we consider six SC capabilities, assessing both SC organizational and SC information technology (IT) dimensions. Our choice of capabilities is based on past research on SC management.

The SC research considers the organizational capabilities of integration with internal and external SC partners as a critical component of managing SCs. Narasimhan and Kim (2002) examine outward-focused capabilities (supplier-based integration and customer-based integration) as well as inward-focused capabilities (that enable systemwide integration) when they evaluate the impact of SC integration on diversification and performance. Similarly, Wisner (2003) considers supplier management strategy, customer relationship strategy and SC management strategy when evaluating the impact of SC management on performance. Germain and Karthik (2006) suggest that external integration alone, without internal integration, will limit the level of performance improvement.

We consider two aspects of organizational capabilities for internal integration--coordination and planning. Coordination is an indicator of a firm's ability to integrate across business processes across the organization and planning is an indicator of a firm's ability to integrate their internal planning process with information from other members of the SC. Our conceptualization of intraorganizational coordination relates to Pagell's (2004) definition of internal integration across operations, purchasing and logistics and Swink, Narasinham and Wang's (2007) concept of horizontal integration. Our planning capability captures the need for integrated planning that incorporates information from customers and suppliers. Aviv (2001) and Fisher, Hammond, Obermeyer and Raman (1994) consider methods for improving planning capabilities. We also consider two aspects of organizational capabilities for external integration--customer involvement and supplier involvement. Vickery et al. (2003) consider two categories of integration--supplier partnering and customer relationships. Similarly, Rungtusanatham, Salvador, Forza and Choi (2003) present an SC management framework that includes integration with suppliers and customers. Drawing on this literature, this study conceptualizes the organizational capabilities of SC integration with four SC organizational capabilities, using measures for Coordination, Planning, Supplier Involvement and Customer Involvement.

The importance of developing IT to manage the SC has also been emphasized in the literature. Forrester (1961) and Sterman (1989) emphasize the need to share information to better coordinate efforts in the SC. Research also suggests that the use of IT in managing the SC needs to align with interorganizational and intraorganizational resources such as the structure of an SC (e.g., Lynagh, Murphy, Poist and Grazer 2001), supplier relationships (Barratt and Rosdahl 2002; Skjott-Larsen, Kotzab and Grieger 2003) and SC strategy (Lee 2002). Other research emphasizes that IT is a necessary but not a sufficient capability for creating sustainable competitiveness (Roberts and Mackay 1998; Strader, Lin and Shaw 1999; Barratt and Rosdahl 2002). The RBV claims that IT can bring competitive advantage when it is bundled with other resources or capabilities (Hammer and Champy 1993; Keen 1993; Powell and Dent-Micallef 1997). Therefore, we consider IT capabilities along with organizational capabilities to identify SC Strategy Groups.

We consider two aspects of IT capabilities--the use of IT for Exploitation and Exploration--as suggested by March (1991). Subramani (2004) also utilizes this classification to categorize IT use for SC management. According to Subramani (2004), Exploitation is the use of IT to improve operational efficiencies (eg., order processing, exchange of information, controlling inventories). These activities or technologies aim to reduce variance or streamline processes. Exploration, on the other hand, is the use of IT to learn about the environment and discover new ways of creating value (e.g., scanning the market, collaborations with suppliers and customers). These activities or technologies aim to understand the variance and develop new methods of dealing with the variance. Exploration typically involves innovations and risk taking while Exploitation brings greater standardization and control Based on the literature, this study examines two aspects of SC IT capabilities based on the use of the Internet for Exploitation and Exploration.

In summary, this study considers six SC capabilities that address issues of SC organizational capabilities and SC IT capabilities. Descriptions of these capabilities are shown in Table 1. SC organizational capabilities refer to the capabilities within the company for coordination (Coordination) and planning (Planning) of the SC and for involvement with suppliers (Supplier Involvement) and customers (Customer Involvement) throughout the SC. SC IT capabilities refer to the use of IT to improve operational efficiencies (Exploitation) and to learn about the environment and discover new ways of creating value (Exploration) in the SC. We examine these six SC capabilities to identify a taxonomy of SC Strategy Groups.

TABLE I

Supply Chain Capabilities

Supply Chain Capability              Defined as Capability to:

Organizational capabilities

  Coordination                 Integrate supply chain activities
                               across the organization

  Planning                     Integrate forecasting and planning
                               of supply chain activities

  Supplier involvement         Integrate and collaborate with suppliers

  Customer involvement         Integrate and collaborate with customers

IT capabilities

  IT for exploitation          Use IT to automate and improve
                               existing supply chain process

  IT for exploration           Use IT to learn about the
                               environment and discover new ways
                               of creating value in the supply chain

SC Strategy Groups and Context

The literature emphasizes the need for developing SC capabilities that are compatible with the context or environment in which the firm operates. We consider several aspects of context: location (country where firm is located), industry (firm's industry type), uncertainty and competitiveness.

The industry and country in which a firm operates has been shown to relate to manufacturing management practices (McKone, Schroeder and Cua 1999; McKone and Schroeder 2002). Lawrence and Lorsch (1967) emphasized that it is important to place the organization in the context of its environment and recognized that an organization must interact with its environment, obtain resources from it and transform them into products in order to survive. For example, Lawrence (1981) indicates that a particular industry (at a specified point in time) can be characterized by its resource constraints and its strategic uncertainty. An organization must adapt to its industry characteristics in order to be competitive in its environment. Similarly, the country in which an organization operates can constrain or enable an organization by the availability of natural resources, the level of government involvement in businesses, the culture and education of employees and other unique characteristics of a country.

In this study, we test the following null hypotheses:

H1:  There is no difference in the industries of the firms across the
     SC Strategy Groups.

H2:  There is no difference in the countries of firms across the SC
     Strategy Groups.

Uncertainty has been recognized as one of the root causes of the difficulty in efficiently coordinating SCs (e.g., Lee, Padmanabhan and Whang 1997b; Ganeshan 1999). Although there has been great effort to improve forecasting performance, the ever-changing competitive environment, such as the proliferation of products, shortened product life cycle and global expansion of the supplier/market, continues to make forecasting a challenge.

Fisher (1997) notes that while innovative products can allow firms to achieve higher profit margins, the very newness of innovative products makes demand unpredictable. As product life cycles shorten, streams of new product innovations continue to lead to demand volatility. Firms that operate in a highly innovative market, therefore, need to have different SC capabilities to minimize the risk of obsolescence and the cost of excess supplies and to meet time-to-market objectives. Our research examines product innovation with measures of the length of the product life cycle and the percentage of sales generated through new products. We test the following null hypotheses:

H3:  There is no difference in the length of the product life cycle
     across the SC Strategy Groups.

H4:  There is no difference in the mean percentage of sales generated
     through new products across the SC Strategy Groups.

In addition to product innovation, there are other factors that contribute to the uncertainty of demand and supply for a product. According to bullwhip effect theory firms that are located in the upper stream of an SC are exposed to a higher level of demand uncertainty because of the amplified demand information aggregated from downstream SC partners (Lee et al. 1997a, b). Our third measure of uncertainty is based on the firm's position within the SC. Therefore, we test the following null hypothesis:

H5:  There is no difference in the position of the firm within
     the supply chain across the SC Strategy Groups.

Competitiveness is the second type of contextual factor that we consider. While most studies focus on recognizing strategic SC practices as a source of competitive advantage (e.g., D'Avanzo, Lewinski and Sassenhove 2003; Cecere, O'Marah and Preslan 2004; Mckone-Sweet, Hamilton and Willis 2005), some identify the relationship between the competitive market and SC strategy. Fein (2006), from his recent data analysis of productivity in the SC, reports that there exists a positive relationship between SC practices and competitive intensity. Randall, Morgan and Morton (2003) report that market growth rate can be an important factor in the choice of SC type (efficient or responsive). They suggest that low growth rate can signal that a market may never develop into a larger market or that a product is at the end of its viable life, both characterized by different forms of competition. Porter (1980), in his early work of developing a competitive strategy framework, identifies five forces that together determine the intensity of industry competition. In his competitive strategy framework, he describes the relationship between forces that drive the competitiveness within an industry and various strategic options for firms. Drawing on these previous studies, our research considers competitiveness as a contextual factor and measures two aspects of this factor with market growth and competitive intensity measures. We test the following null hypotheses:

H6:  There is no difference in the market growth across the SC Strategy
     Groups.

H7:  There is no difference in the competitive intensity across the SC
     Strategy Groups.

SC Strategy Groups and Competitive Priorities

While the research on the fit between operations capabilities and a firm's competitive priorities has been a focal discussion in operations management (e.g., Hayes and Wheelwright 1984; Roth and Velde 1991; Hill 1994), there have been limited efforts to investigate the fit between SC capabilities and competitive priorities. The general consensus within the field of operations management is that the degree of fit between a firm's capabilities and its competitive priorities is related to a firm's performance (Boyer and McDermott 1999).

Frohlich and Westbrook (2001) point out that it is a critical factor for successful firms to carefully link the firm's internal processes to external suppliers and customers in unique SCs. It is a crucial capability for firms to integrate upstream and downstream partners in a way to support the firm's competitive priorities. However, Tamas (2000) reports that an independent survey reveals that only 13% of the 80 respondents believe that their company's SC practices are fully aligned with their priorities.

Our research begins to bridge the gap between theory and practice by examining the relationship between SC capabilities and competitive priorities. We test the following null hypotheses:

H8:   There is no difference in the competitive priorities
      across the SC Strategy Groups in terms of:

H8a:  low price,

H8:   performance quality,

H8c:  fast delivery,

H8d:  on-time delivery and

H8e:  fast response to nonstandard orders.

SC Strategy Groups and Performance

Research provides evidence of the positive relationship between SC capabilities and firm performance. Numerous studies report that there exists a relationship between SC integration and performance (e.g., Frohlich and Westbrook 2001; Narasimhan and Kim 2002; Vickery et al. 2003). Ittner and Larcker (1997) and Wisner (2003) report that strong partnerships/relationships with suppliers and customers improve performance. The literature also suggests that the higher the degree of integration, the better a firm performs (Tan, Kannan and Handfield 1998; Lee, So and Tang 2000; Frohlich and Westbrook 2001). Frohlich and Westbrook (2001) provide evidence that the wider the arc of integration (encompassing suppliers and customers), the stronger the association with performance.

IT is considered to be a core component for increasing the effectiveness of SC transactions (e.g., Croom 2000; Essig and Arnold 2001; Ellinger, Lynch and Hensen 2002; Frohlich and Westbrook 2002). The use of IT for Exploitation and Exploration is expected to lead to different performance outcomes for firms. For instance, benefits associated with Exploitation are definable and concrete--cost reduction, process efficiency, or fast turnaround time (Subramani 2004). Benefits associated with Exploration are harder to evaluate in advance, but may take the form of greater understanding of the operating environment and shared understanding of market demand (Subramani 2004). DeSantis and Poole (1994) use the term appropriation to refer to how firms use IT and use differing patterns of appropriations to explain diverse performance outcomes in the context of group decision support systems. The research suggests that specific patterns of IT use can explain different performance outcomes.

We examine differences in actual performance and perceived performance relative to the competition. We test the following null hypotheses:

H9:   There is no difference in the perceived performance
      relative to the competition across the SC
      Strategy Groups in terms of:

H9a:  manufacturing cost,

H9b:  inventory turns,

H9c:  conformance to product specification,

H9d:  on-time delivery,

H9e:  delivery speed,

H9f:  product mix flexibility and

H9g:  volume flexibility.

H10:  There is no difference in the actual performance
      across the SC Strategy Groups in terms of:

HlOa  manufacturing cost as percentage of sales,

HlOb  inventory turns,

HlOc  percentage of orders shipped on time and

HlOd  length of leadtime.

RESEARCH METHODOLOGIES

Data

This research used data collected by a team of international researchers in 2004 as part of the High Performance Manufacturing Research Project. The data collection constitutes Round 3 of an ongoing research project that began in 1989. The objective of Round 1 of the study was to examine the emergence of Japanese manufacturing practices in the United States. This was done by collecting data from 25 manufacturing plants each, operating in electronic, transportation and machinery industries in the United States. Conducted in 1996, Round 2 of the study extended the original objective to comparing and contrasting the state of manufacturing in five countries. This involved collecting data from 164 plants located in the United States, United Kingdom, Germany, Italy and Japan in electronic, transportation and machinery industries.

In Round 3, referred to as the high-performance manufacturing or HPM project, the study was expanded to include additional countries and to explore more advanced and up-to-date manufacturing practices, including the use of IT and SC practices. This round collected data from 238 manufacturing plants located in Finland, Sweden, Germany, Austria, Italy, Japan, Korea and the United States. This multicountry sample allowed us to examine firms from North America, Europe and Asia. This round also selected 10 firms from three different industries (automotive suppliers, electronics and machinery industries) in each country, for a total of about 30 plants in each country. These three industries provide a sample of firms with a variety of characteristics (different product characteristics, manufacturing practices, levels of competition and SC structures). Table II provides a summary of our data.

TABLE II
                               Description of the Data

Country             AUT  FIN  GER  Italy  Japan  South  Sweden  USA
                                                        Korea

Total                21   30   41     27     35     31      24   29
Electronics          10   14    9     10     10     10       7    9
Machinery             7    6   13     10     12     10      10   11
Transportation        4   10   19      7     13     11       7    9
# Employees (a)     272  350  652    337   1342   4964     344  460
% Exports (a)        78   67   52     59     36     38      55   25
Market share % (a)   19   22   31     24     26     33      23   24

                                   Description of the Data

Country                    Electronics  Machinery  Transportation

Total                           79           79         80
Electronics                     79            0          0
Machinery                        0           79          0
Transportation                   0            0         80
# Employees (a)                534         1148        834
% Exports (a)                   55           54         44
Market share % (a)              21           30         26

(a) Average of sample firms.

The survey instrument was developed by teams of operations management experts, formed for each topic area (i.e., SC management, IT systems, performance measurement). Each team designed survey questions for the specific topic, based on an extensive literature review and expertise in the area. Then the instrument was translated into the native language of the participants by teams from the associated countries. The translations were then translated back into English by a different group of people to check the accuracy of translation. The instrument was also pretested at several manufacturing plants. Necessary modifications to the instrument were made for clarity and consistency across all translations. Care in the development and pretest of the questions provides assurance that the constructs were measured to an acceptable degree of content validity.

Members of the research team contacted the plant manager of the manufacturing plants to request their participation. Sixty-five percent of the plants completed written surveys for inclusion in our research data. This relatively high response rate was assured by communicating with the plants personally and by promising that they would receive a plant profile for comparison with other plants.

Once the plant agreed to participate, the plant manager appointed a coordinator to randomly select the internal participants, distribute the questionnaires, collect the responses and mail the completed surveys back to the research team. The instrument was distributed in 10 questionnaires (each with a different assortment of questions) that were administered to 24 informants ranging from shop floor employees to various functional managers, including the plant manager within a manufacturing plant--Specific questions were assigned to informants on the basis of their job title and expertise in order to increase the probability of getting accurate information. Responses for most questions were collected from multiple informants to provide greater reliability. Different respondents were used for specific measures to avoid common-rater bias. For this study, we used the average response of the informants for each survey item.

Twenty-six firms were eliminated from our data due to one or more missing values for the SC capability measures, resulting in a total of 212 firms being used for the analysis.

Measurement

We used survey questions to measure SC capabilities, contextual factors, competitive priorities and operational performance for each plant. Details of the measures are provided in the Appendices.

SC capabilities. For SC capabilities, we assessed organizational capabilities that involve practices and procedures to collaborate and integrate within the SC and IT capabilities that involve the use of the Internet to leverage information and technology to improve the efficiency of SC processes and to better understand market needs. See Appendix A (SC Organizational capabilities) and Appendix B (SC IT capabilities).

APPENDIX A

Measurement of SC Organizational Capabilities

Please indicate your opinion about the following statements on a
7 point Likert scale.

1 =strongly disagree, 2=disagree, 3=slightly disagree, 4=neutral,
5=slightly agree, 6 =agree, 7=strongly agree
(Respondents were the quality manager, the inventory manager,
the supervisor and the plant superintendent.)

            Measurement Items            Loading       Reliability

              Coordination

Purchasing of common materials is          0.53     [alpha] =0.780
coordinated at the corporate level.

Our corporation implements                 0.82
ordering and
stock management policies, on a
global scale, in order
to coordinate distribution.

Our corporation performs aggregate         0.76
planning for plants,
according to our global
distribution needs.

Managerial innovations are transferred     0.78
among plants within our corporation.

Our corporation transfers technological    0.76
innovations and
know-how between plants

                Planning

We actively plan supply chain activities.   0.79    [alpha] =0.816

We consider our customers'                  0.66
forecasts in our supply chain planning.

We strive to manage each of our             0.75
supply chains as a whole.

We monitor the performance of               0.77
members of our supply chains, in order to
adjust supply chain plans.

We gather indicators of                     0.82
supply chain performance.

          Supplier Involvement

We are comfortable sharing                  0.85   [alpha] =0.770
problems with our suppliers.

In dealing with our                         0.61
suppliers, we are willing
to change assumptions, in order to
find more effective solutions.

We believe that cooperating with            0.83
our suppliers is beneficial.

We emphasize openness                       0.82
of communications
in collaborating with our suppliers.

          Customer Involvement

We frequently are in close                   0.81  [alpha] =0.775
contact with our customers.

Our customers give us feedback on our        0.85
quality and delivery performance.

Our customers are actively involved in       0.64
our product design process.

We strive to be highly responsive            0.75
to our customers' needs.

We regularly survey our customers' needs.    0.60

The SC organizational capabilities included coordination and planning and customer and supplier involvement. The items used to measure Coordination consider the extent of corporatewide cooperation and involvement with SC activities. The items for Planning refer to the importance of using an SC perspective in planning activities. These two capabilities relate to the collaboration within an organization and refer to intraorganizational capabilities. The items for Customer Involvement measure the frequency and type of customer involvement. The items for Supplier Involvement refer to the openness of communication and the level of cooperation between the supplier and the plant. These two capabilities relate to the collaboration with organizations external to the firm and refer to interorganizational capabilities.

The respondents for the survey questions that assessed the SC organizational capabilities were the quality manager, the inventory manager, the supervisor and the plant superintendent. We computed an average rating of the respondents for each of the items. We conducted factor analysis of these four SC capability constructs to verify that the items for each construct were associated with only one single factor. Then, we evaluated item level reliability and convergent validity to assess how each item behaved within the block of items intended to measure the specific SC capability. We assessed item reliability by calculating the Corrected Item to Total Correlation score. One item was dropped from the SC Planning and Coordination constructs since it loaded on a second factor. Anderson and Gerbing (1988) suggest that evidence of convergent validity exists if the manifest variable loads significantly (t-value > 2.58, p < 0.01) on its respective latent variable. All specified parameter estimates were within an acceptable range (see Appendix A) and highly significant, which indicates good convergent validity among the measures of each construct. We assessed construct validity with Cronbach's [alpha]. All Cronbach's [alpha] values exceeded 0.73 (Appendix A).

The SC IT capabilities included IT for Exploitation and IT for Exploration. For this study, we consider Internet use and do not consider all types of IT. We have chosen to examine Internet use since it is becoming a common means for collecting, sharing and analyzing information within the SC (e.g., Gunasekaran, Marri, McGaughey and Nebhwani 2002). Kouvelis, Chambers and Wang (2006) summarize recent research and discuss the importance of the Internet to manufacturers and retailers and how most if not all established firms are expanding their business practices and relationships with some Internet-enabled components. Our measure for Exploitation considers the use of the Internet to improve operational efficiencies and our measure for Exploration considers the use of the Internet to learn about the environment and discover new ways of creating value. The items for Exploitation assess the extent of Internet use for order processing, invoicing, settling accounts, managing inventories and exchanging shipment and delivery information. In contrast, the items for Exploration assess the level of Internet use to support nonroutine, unstructured tasks to understand market trends and customer preferences. Our operationalization for Exploitation and Exploration is similar to Subramani's (2004) categorization of SC management systems along Exploitation and Exploration dimensions and the measures used in Boynton, Zmud and Jocobs (1994).

Six items were used to measure Exploitation and seven items to measure Exploration. The survey respondent for all items was the information systems manager. Responses to all 13 items were binary in nature. We used an additive approach to create the IT capability scales, adding 1 for each use of the Internet. More uses or applications of the Internet, therefore, indicate a higher level of IT capabilities.

Contextual factors. Next we considered seven contextual or environmental factors that might be related to the choice of a firm's SC capabilities. Our research included different countries and industries by design. This research also examined indicators of the uncertainty in the SC (the percent of current products that were introduced within the last 5 years, the average product life cycle, and the firm's position within the SC), and the competitiveness (market growth and competitive intensity) of the industry. Factor analysis and tests for reliability, convergent validity and construct validity were confirmed for the competitive intensity measure (Cronbach's [alpha] was 0.641). Appendix C shows the respondents and measures for each contextual factor.

APPENDIX C

Measurement of Contextual Variables

                                       Uncertainty

Product Life Cycle     What is the average life cycle of
                       your products (years)?
                       (Respondent is plant superintendent.)

Percent New Products   What percent of plant sales is from
                       products introduced in the last five years?
                       (Respondent is process engineer.)

Supply Chain Position  What percent of your sales is to each of the
                       following types of customers? (Respondent is
                       inventory manager.) End Consumers (1), Retailers
                       (2), Wholesalers (3), Distributors (4),
                       Assemblers (5), Manufacturers (6) Calculated by
                       using a weighted average (percent times assigned
                       #). A lower value indicates that the plant is
                       located closer to the end consumers

                                 Competitiveness

Market Growth          What is the annual growth rate in the market
                       which this plant serves? This is not the growth
                       rate in the plant's sales, but rather, the
                       growth rate of the market served. (Respondent is
                       plant manager.)

Competitive Intensity  Please indicate your opinion about the
([alpha]=0.641)        following statements on a 7 point Likert scale.
                       1= strongly disagree, 2 = disagree, 3=slightly
                       disagree, 4= neutral, 5=slightly agree, 6=agree,
                       7 = strongly agree (Respondents include process
                       engineer, plant manager, and plant
                       superintendent.)

                       We are in a highly competitive industry.
                       (loading = 0.809) Our competitive pressures are
                       extremely high. (loading=0.737) There are few
                       significant competitors for our products.
                       (loading =0.532) Competitive moves in our
                       market are slow and deliberate, with long time
                       gaps between different companies' reactions.
                       (Reverse scored.) (loading= -0.700)

Competitive priorities. We also considered the competitive priorities of each firm. We assessed the firm's priorities relative to cost, quality, delivery and flexibility. As shown in Appendix D, we asked the plant manager, plant superintendent and the process engineer to identify the importance of each goal or priority. We computed an average rating of the respondents for each of the items.

APPENDIX D

Measurement of Competitive Priorities

Competitive  Please identify the importance of each goal
Priorities   below. Identify the goal as absolutely crucial only if it
             helps "win the order" from the customer in the marketplace
             relative to the competition. Rate the other goals
             according to their relative importance. On a scale of 1 to
             5 with  1 = least important and 5 = absolutely crucial
             (Respondents include the plant manager, plant
             superintendent and the process engineer.)

Cost         Low price
Quality      High performance quality
Delivery     Fast delivery
             On-time delivery
Flexibility  Fast response to nonstandard orders

Performance. Finally, we considered operational performance measures and assessed multiple dimensions of cost, quality, delivery and flexibility. As shown in Appendix E, we asked the plant managers their opinion about how their plant compares to its competitors in me industry, on a global basis. We also asked the accounting manager for actual performance on cost and delivery. Therefore, we assessed both perceptual and actual performance.

APPENDIX E

Measurement of Performance

                   Relative Competitive Performance Measures

Dimension          Please circle the number that indicates your opinion
of                 about how your plant compares to its competition in
Performance        your industry, on a global basis. 5: superior, 4
                   better than average, 3 average, 2 equivalent to
                   competitors, 1 poor, low end of industry (Respondent
                   is plant manager)

Cost               Unit cost of manufacturing
                   Inventory turnover

Quality            Conformance to product specifications
                   Product capability and performance

Delivery           Cycle time {from raw materials to delivery)
                   On-time delivery performance
                   Fast delivery

Flexibility        Flexibility to change product mix
                   Flexibility to change volume

                      Actual Reported Performance Measures

Dimension of       Calculated based on reported performance
Performance        (Respondent is Accounting Manager)

Cost               Manufacturing cost/sales value of production
                   Total inventory/manufacturing costs (a)

Delivery           Percent of orders shipped on time
                   Average lead time, from the receipt of an order
                   until it is shipped (days) (a)

(a) These measures were transformed logarithmically) for our analysis.

Both actual and perceptual performance measures are important to our study. The actual performance is based on reported accounting data and is therefore an accurate measure of performance. However, the actual performance metrics do not capture differences in "good" performance that may differ by industry as well as by the firm's position within the SC The perceptual measures try to capture a firm's performance relative to its direct competitors.

Analysis Approach

We conducted our data analysis using SPSS statistical software. There were three stages to our analysis: identification of SC Strategy Groups, interpretation of the underlying differences of the strategy groups, and comparison of the context, competitive priorities and performance for the groups. We employed Cluster Analysis to classify the plants based on their SC capability profiles and identify the SC Strategy Groups. We used ANOVA and Scheffe post hoc pair comparison (when necessary assumptions of homogeneity and normality were met) and Kruskal-Wallis comparison (when assumptions for ANOVA were not met) to identify significant differences in SC capabilities for SC Strategy Groups, these analyses helped to interpret the SC strategy for each group.

Next we conducted crosstabs analysis to consider differences in the country and industry of each strategy group or cluster. We also used ANOVA and Scheffe post hoc pair comparison and Kruskal-Wallis comparison to examine the significant differences across the clusters in terms of contextual factors, competitive priorities and performance. The contextual analysis defined the environment of each SC strategy group. The competitive priority analysis highlighted the goals of the SC strategy groups. Finally, the performance analysis identified the competitive positions of the SC strategy groups. These differences enabled us to better understand characteristics of specific SC strategy groups within the taxonomy.

When using the ANOVA and Scheffe post hoc comparisons for identifying significant group mean differences, we designated a statistical significance of 0.05 or less. In our post hoc analysis of the mean differences (to determine which means are different from each other), we used the conservative Scheffe post hoc analysis approach that adjusts the F-critical value for multiple comparisons to minimize type 1 error.

We also tested the assumptions of homogeneity of variance (Levine statistic) and normality (Kolmogorov-Smirnov test). For some variables, the Levine statistic indicated nonhomogeneity and the Kolmogorov-Smirnov test indicated nonnormality of the data. Therefore, we also compared group differences using the Kruskal-Wallis test, a nonparametric test of whether the independent samples that are defined by clusters are from the same population. For this test, we designated a statistical significance of 0.05 or less.

It is important to note that the Kolmogorov-Smirnov test is very sensitive to nonnormality For our analysis, the sample size of 24 or more for each cluster exceeded the 20 degrees of freedom for error suggested lo assure multivariate normality of the sampling distribution of means, even with unequal sample sizes (Tabachnick and Fidell 1989, p. 411). Therefore, for our analysis, the Kolmogorov Smirnov test results were consistent with those from the ANOVA and Scheffe post hoc analysis.

ANALYSIS AND DISCUSSION

For this study, 26 firms were eliminated from our data due to one or more missing values for the SC capability measures, resulting in a total of 212 firms being used for the analysis. In this section, we present our results for each stage of the analysis.

Identification of SC Strategy Groups

Using the six SC capabilities--Coordination, Planning, Supplier Involvement, Customer Involvement, Exploitation and Exploration--we employed Cluster Analysis to identify groups within the SC Strategy taxonomy.

One problem with cluster analysis is the determination of the number of clusters. Three criteria were used. First, we utilized the elbow criteria, a common rule of thumb to determine what number of clusters should be chosen based on the marginal gain in the percentage of variance explained by adding another cluster. Second, we looked for cluster groupings that agreed with existing or expected structures (i.e., that have managerial interpretability).The ANOVA and the Scheffe post hoc tests of mean differences helped guide the interpretability of the results. Third, we were guided by Lehmann's (1979) suggestions that the number of clusters should be limited to between n/30 and n/60 where n is the sample size. Therefore, only models with between three and seven clusters were considered. The three-cluster model best satisfied these three criteria. Table II describes the three resultant SC Strategy Groups in terms of their respective group centroids.

Interpretation of the Strategy Groups

Next we analyzed the mean difference in SC capabilities for each SC Strategy Group. Table III provides the results of the ANOVA and Scheffe post hoc pair comparison as well as the Kruskal-Wallis comparison. The SC Strategy Groups all significantly differed on three of the organizational capabilities (coordination, planning and customer involvement) with Group 1 having low levels, Group 3 having moderate levels and Group 2 having high levels. Group 1 also had significantly lower levels of supplier involvement than Groups 2 and 3. Finally, Group 3 had lower levels of IT capabilities (Exploitation and Exploration) than Groups 1 and 2.

TABLE III
Results of the Cluster Analysis for SC Capabilities

SC                         Test of              Suggested
Capabilities            Assumptions              Test

                   Homogen.     Normality (b)   ANOVA
                  of Var (a)

Coordination     LS-0.363       KS=0.864        ANOVA
                 [rho]=0.686    [rho]=0.444

Planning         LS=2.822       KS=0.835        ANOVA
                 [rho]=0.062    [rho]=0.488

Supplier         LS=1.81        KS=0.516        ANOVA
relationships    [rho]=0.166    [rho]=0.953

Customer         LS = 2.018     KS=0.556        ANOVA
involvement      [rho]=0.135    [rho]=0.917

IT exploitation  LS=5.533       KS=3.590        Kruskal-
                 [rho]=0.000    [rho]=0.000     Wallis

IT exploration   LS = 27.580    KS=3.284        Kruskal-
                 [rho]=0.000    [rho]=0.000     Waliis

Sample size

Summary          Both tests arrive at same conclusions for
of               all capabilities.
results          All clusters differ on SC planning, coordination,
                 and customer involvement (1=low level, 3=mod level
                 and 2=highest level).
                 Cluster 1 has lowest group mean for supplier
                 involvement.
                 Cluster 3 has lower group mean for a II SC IT
                 capabilities (Exploitation and Exploration).

SC               ANOVA and Scheffe Post  Hoc  Analysis
Capabilities

                 SCS Group    SCS       SCS        F=value
                 1 (c),(d)    Group 2   Group 3    [rho]=prob
                                                   N=212

Coordination     4.4611*      5.197*    4.699*     F=32.469
                 (0.0744)     (0.0537)  (0.145)    [rho]=0.000

Planning         4.644*       5.450*    4.986*     F=53.972
                 (0.559)      (0.044)   (0.148)    [rho]=0.000

Supplier         4.774*       5.330     5.126      F=3.029
relationships    (0.0469)     (0.0359)  (0.0108)   [rho]=0.000

Customer         4.972*       5.662*    5.273*     F= 58.722
involvement      (0.052)      (0.036)   (0.107)    [rho]=0.000

IT exploitation  15.203       14.798    6.167*     F=247.622
                 (0.193)      (0.158)   (0.541)    [rho]=0.000

IT exploration   10.281       10.218    5.958*     F=74.548
                 (0.138)      (0.134)   (0.594)    [rho]=0.000

Sample size      64           124       24

Summary          Both tests arrive at same conclusions for
of               all capabilities.
results          All clusters differ on SC planning, coordination,
                 and customer involvement (1=low level, 3=mod level
                 and 2=highest level).
                 Cluster 1 has lowest group mean for supplier
                 involvement.
                 Cluster 3 has lower group mean for a II SC IT
                 capabilities (Exploitation and Exploration).

SC                           Kruskal-Wallis (rank)
Capabilities

                    SCS       SCS       SCS    [x.sup.2] value
                   Group 1  Group 2   Group 3  [rho]=prob.

Coordination       65.93*   130.65*   89.65*   [x.sup.2]=48.97
                                               [rho]=0.000

Planning           53.63*   135.46*   97.85*   [x.sup.2]= 75.66
                                               [rho]=0.000

Supplier           57.20*   131.20*  110.35*   [x.sup.2] = 61.532
relationships                                  [rho]=0.000

Customer           54.13*   136.88*   89.21*   [x.sup.2]=78.956
involvement                                    [rho]=0.000

IT exploitation    127.7    113.8     13.5*    [x.sup.2]=66.379
                                               [rho]=0.000
IT exploration     115.4    113.5     25.9*    [x.sup.2]=40.099
                                               [rho]=0.000

Sample size        64       124       24

Summary            Both tests arrive at same conclusions for
of                 all capabilities.
results            All clusters differ on SC planning, coordination,
                   and customer involvement (1=low level, 3=mod level
                   and 2=highest level).
                   Cluster 1 has lowest group mean for supplier
                   involvement.
                   Cluster 3 has lower group mean for a II SC IT
                   capabilities (Exploitation and Exploration).

(a) Levine statistic used for testing Homogeneity of Variances.

(b) KS is Kolmogorov-Smirnov test of normality.

(c) Top values are mean values for each cluster and bottom values are
standard errors. Numbers with * are significantly different than
the other means using the Scheffe post hoc analysis (0.05 level).

(d) SCS = supply chain strategy.

The results of the cluster analysis suggested that SC Strategy Group 1 has low levels of SC organizational capabilities and high levels of SC IT capabilities. In practice, these firms look for an IT solution to SC problems and do little to change the existing organizational processes within the firm or the SC

Strategy Group 2 has high levels of both SC organizational capabilities and SC IT capabilities. In practice, these firms take an integrative approach to SC management, believing that IT and organizational capabilities are both needed for superior SC performance.

SC Strategy Group 3 has low levels of SC IT capabilities and moderate levels of SC organizational capabilities. In practice, these firms emphasize their SC organizational capabilities but do not develop strong IT capabilities to complement their SC organizational capabilities.

Comparison of the Clusters

After developing and interpreting the clusters, we extended our analysis by systematically examining the factors associated with SG Strategy Group membership. We explored significant differences across clusters for contextual variables, competitive priorities and performance. Tables IV-VIII show our results, including both the ANOVA and Kruskal-Wallis test statistics and significance. Since the conclusions were the same for both tests, we simplified the tables and report the mean and standard error for each group but did not report the rankings generated from the Kruskal-Wallis analysis.

Context: In our data, firms were sampled from three different industries and eight different countries. Table II describes our data sample and identifies differences in the number of employees, the percentage of exports and the market share across industries and countries. Table IV shows the results from a crosstabs analysis for industry. While the results indicated that firms within the transportation industry tended to be in SC Strategy Groups 2 and 3 (develop more SC organizational capabilities) more than those in the machinery industry, these results were not significant at the 0.05 level using the Chi-Square Statistics. Therefore we failed to reject the null hypothesis H1.

TABLE IV
Crosstabs Analysis for Industry

                                       Industry

                      Electronics  Machinery  Transportation  Total

SC Strategy Group

  1
    Count              22           29             13         64
    Expected count     22.3         21.1           20.5       64.0
    Standard residual   0.0          1.7           -1.7

  2
    Count              45           32              47        124
    Expected count     43.3         40.9            39.8      124.0
    Standard residual   0.3         -1.4            1.1

  3
    Count               7            9              8         24
    Expected count      8.4          7.9            7.7       24.0
    Standard residual  -0.5          0.4            0.1

Total
    Count              74           70             68         212
    Expected count     74.0         70.0           68.0       212.0

                       Summary

SC Strategy Group

  1
    Count              Cluster 1 has fewer firms from
    Expected count     transportation industry and more
    Standard residual  from machinery industry than expected

  2
    Count              Cluster 2 has lower number of firms
    Expected count     from machinery industry and more
    Standard residual  from transportation than expected.

  3
    Count              Cluster 3 has approximately the expected number
    Expected count     of firms from each industry.
    Standard residual

  Total
    Count
    Expected count
[x.sup.2]        Value       DF   Asymptotic Significance
Tests                             (2-sided)

Pearson          9.414 (a)    4   0.052
[x.sup.2]

No.of             212
valid cases

Table V shows the crosstabs analysis for country. There were significant differences in the location of firms of SC Strategy Groups; therefore, we rejected the null hypothesis H2. Group 1 had more firms from Japan and fewer firms from Finland and the United Stales than expected. Group 2 had more firms from Finland and fewer from Japan than expected. The Finnish firms in this study appear to develop organizational and IT capabilities together. Group 3 had more firms from the United States than expected and no firms from four countries. It is important for managers to understand that SC strategies differ from one country to another. It provides insight into the SC capabilities of domestic and international competitors. Further research should examine the differences in countries and the nature of the firms within each country (i.e., in our sample, firms in Japan and South Korea were larger on average than firms in other countries and firms in Finland had a higher percentage of exports than many other countries).

TABLE V
Crosstab Analysis by Country

                                      Country

                       AUT   Finland  Germany  Italy  Japan  South
                                                             Korea

SC Strategy Group

  1
    Count              6      3        15       7      13      9
    Expected count     6.0    9.1      12.1     7.8     8.2    6.9
    Standard residua   0.0   -2.0      0.8     -0.3     1.7    0.8

  2
    Count              14     27       25       16      9     13
    Expected count     11.7   17.5     23.4     15.2   15.8   13.5
    Standard residual   0.7   2.3      0.3      0.2   -1.7    -0.1

  3
    Count               0     0        0        3      5       1
    Expected count      2.3   3.4      4.5      2.9    3.1     2.6
    Standard residual  -1.5  -1.8     -2.1      0.0    1.1    -1.0

Total
    Count               20    30       40       26     27      23
    Expected count      20.0  30.0     40.0     26.0   27.0    23.0

                               Country

                         Sweden  United  Total
                                 States
SC Strategy Group

  1
    Count                  9       2       64
    Expected count         5.7     8.2     64.0
    Standard residua       1.4    -2.2

  2
    Count                  8       12      124
    Expected count         11.1    15.8    124.0
    Standard residual      -0.9    -1.0

  3
    Count                  2       13       24
    Expected count         2.2      3.1     24.0
    Standard residual     -0.1      5.7

Total
    Count                  19       27      212
    Expected count         19.0     27.0    212.0

                              Summary

SC Strategy Group

  1
    Count               Cluster 1: More firms from Japan and fewer from
    Expected count      Finland and United States than expected.
    Standard residual

  2
    Count               Cluster 2: More firms from Finland and fewer
                        from
    Expected count      Japan than expected.
    Standard residual

  3
    Count               Cluster 3: Most firms from United States (48% of
    Expected count      Cluster 3 firms) and Japan (20% of cluster 3
                        firms)
    Standard residual   and no firms from AUT, Finland, or Germany.

Total
    Count
    Expected count

    [x.sup.2] Tests         Value       df   Sig. (2-sided)

Pearson [x.sup.2]        70.030 (a)     14      0.000
No. of valid cases      212

Next we examined the differences in the uncertainty of demand and the competitiveness of the market for each SC Strategy Group. Table VI shows the results of the group mean differences.

TABLE VI
SC Strategy Groups and Contextual Factors

Contextual              SC Strategy Groups (a)
Factors

                         1            2         3
Uncertainty

  Percent of           53.565       64.173    58.842
  innovation           (4.702)      (2.811)   (7.610)

  Length of             9.883        8.525      9.291
  product
  life cycle           (0.999)      (0.639    (01.97)

  SC position         347.266      336.385    461.869*
                      (24.477)     (18.411)  (28.294)

Competitiveness

  Competitive           5.325*        5.460      5.783*
  intensity             (.082)      (0.062)   (0.107)

  Market growth       149.593      109.075    109.895
                      (24.22)     (10.012)  (14.998)

Contextual                  Tests of Assumption
Factors

                  Homogeneity (b)   Normality (c)

Uncertainty

Percent of        LS = 1.510           KS = 1.909
innovation       [rho]=0.224          [rho]=0.001

Length of         LS=0.172             KS = 3.142
Product          [rho]=0.842          [rho]=0.000
life cycle

SC position       LS = 3.730           KS = 1.865
                  [rho]=0.026          [rho]=0.002

Competitiveness

Competitive
intensity

                  LS = 0.771           KS=1.104
                  [rho]= 0.464         [rho]=0.175

Market growth     LS = 3.123           KS=2.918
                  [rho]=0.047          [rho]=0.000

Contextual            ANOVA              Kruskal-Wallis
Factors              Results             Results (e)

                     F=Value                CS-value
                 [rho]=probability      [rho]=probability
                  N=sample (d)

Uncertainty

Percent of       F=2.129                  CS=3.486
Innovation       [rho]=0.151              [rho]=0.175
                 N=171

Length of        F=0.726                  CS=2.115
Product          [rho]=0.485              [rho]=0.347
life cycle       N=165

SC position      F=4.932            CS=9.473
                 [rho]=0.008        [rho]=0.009
                 N=170

Competitiveness

Competitive
intensity

                 F=4.373            CS = 8.189
                 [rho]=0.014       [rho]=0.017
                 N=210

Market growth    F=1.914            CS = 3.180
                 [rho]=0.151       [rho]=0.204
                 N=171

Contextual                   Summary
Factors

Uncertainty

Percent of                 No significant difference
Innovation

Length of product          No significant difference
life cycle

SC position                Cluster 3 firms are farther away from
                           Customers than clusters 1 and 2

Competitiveness

Competitive intensity      Cluster 3 has higher competitive intensity
                           than cluster 1

Market growth              No significant difference

(a) Top values are mean values for each cluster and bottom values are
standard errors. Numbers with * are significantly different
than the other means using the Scheffe post hoc analysis (0.05 level)

(b) Levine statistic used for testing homogeneity of variances.

(c) KS is Kolmogorov-Smirnov test of normality.

(d) There were missing values for some contextual factors.
Therefore, the sample size varied for each analysis.

(e) Kruskal-Wallis Results: CS=Chi-Square statistic, [rho]=significance.
Ranks are not reported since same conclusions are made from ANOVA.

Our results showed that there are significant group mean differences in the uncertainty (for the SC position measure) and the competitiveness (for the Competitive Intensity measure) faced by the SC Strategy Groups. In particular, Group 3 (with low IT capabilities) included firms that are further away from the customer and have higher levels of competition. In this context, firms have low levels of IT capability and moderate levels of organizational capabilities. In practice, the relationship between the SC position and the development of SC IT capabilities makes sense. Firms that are upstream in the SC may not have the resources to develop SC IT capabilities or the access to valuable information from their consumers. These firms also may not have direct pressures from large retailers or distributors to develop specific IT systems. However, the relationship between the high levels of competition and the SC strategy group is unexpected. Fein (2006), for example, reported a positive relationship between SC practices and competitive intensity. Furthermore, in practice it is assumed that firms with high levels of competition need to develop SC capabilities to maintain or grow their position in the market.

Although we rejected the null hypotheses H5 and H7, we failed to reject H3 (Length of product life cycle), H4 (% of product sales from new products) and H6 (market growth). These results do not support the research on typologies that recommends different SC strategies based on the innovativeness of the product (Fisher 1997; Lee 2002). This may mean the firms are not adhering to the advice from these typologies and do not differentiate their strategy based on the innovativeness of their products. Of course, this result may also occur because our measures do not fully capture all aspects of product innovativeness from the past studies.

Competitive priorities: Next we examined the group mean differences in the competitive priorities. We expected the groups to exhibit differences which would suggest that there was a link between a firm's priorities and their SC Strategy Group membership. Our results, in Table VII, showed that there were no significant differences in group means for competitive priorities and we failed to reject the null hypothesis H8.

TABLE VII
SC Strategy Groups and Competitive Priorities

Competitive                 SC Strategy Group (a)
Priorities

                          1            2           3

Cost
  Low price              3.804        3.763       3.882
                        (0.095)      (0.071)     (0.138)

Quality
  Performance quality    4.249         4.368      4.347
                        (0.070)      (0.045)     (0.122)

Delivery
  Fast delivery          3.878         3.844        3.785
                        (0.083)       (0.059)      (0.125)

  On-time delivery       4.278         4.186        4.118
                        (0.067)       (0.048)      (0.119)

Flexibility
  Fast response to       3.661         3.493        3.417
  nonstandard orders    (0.088)       (0.069)      (0.158)

Competitive                  Test of Assumptions
Priorities

                          Levine (b)            KS (c)

Cost
  Low price                LS=959               KS=1.941
                        [rho]=0.386          [rho]=0.001

Quality
  Performance quality      LS=2.309             KS=2.131
                        [rho]=0.102          [rho]=0.000

Delivery
  Fast delivery           LS=0.093             KS=2.135
                         [rho]=0.912          [rho]=000

  On-time delivery        LS=0.399             KS=2.337
                        [rho]=0.672          [rho]=0.000

Flexibility
  Fast response to        LS=0.160             KS = 1.818
  nonstandard orders      [rho]=0.852          [rho]=0.003

Competitive             ANOVA            Kruskal-Wallis   Summary
Priorities              Results (d)      Results (e)

Cost
  Low price               F0.256           CS=0.434         No
                        [rho]=0.774        [rho]=0.806      difference
                          N=210

Quality
  Performance quality     F= 1.167         CS= 1.828        No
                        [rho]=0.313        [rho]=0.401      difference
                         N=210

Delivery
  Fast delivery           F=0.181          CS=0.711         No
                        [rho]=0.834        [rho]=0.701      difference
                          N=210

  On-time delivery        F=0.956          CS=1.971         No
                        [rho]=0.386        [rho]=0.373      difference
                        N=210

Flexibility
  Fast response to        F = 1.381         CS = 2.401       No
  nonstandard orders    [rho]=0.254        [rho]=0.122      difference
                         N=210

(a) Top values are mean values for each cluster and bottom values are
standard errors. Numbers in bold are significantly different than the
other means using the Scheffe post hoc analysis (0.05 level).

(b) Levine Statistic used for testing Homogeneity of Variances.

(c) KS is Kolmogorov-Smirnov test of normality.

(d) There were missing values for some competitive priorities.
Therefore, the sample size varied for each analysis.

(e) Kruskal-Wallis Results CS=Chi-Square statistic,
[rho] = significance. Ranks are not reported since same conclusions
are made from ANOVA.

This somewhat unexpected result suggests that firms develop their SC strategies independently from their competitive strategies. We expected firms to align their SC strategies with their competitive priorities in order to use their SC capabilities to achieve performance improvements relative to the priorities. However, our results correspond with Tamas' survey report (2000) that revealed the lack of alignment between strategic priorities and SC strategies.

Manufacturing performance: Finally, we examined the group mean differences in perceived and actual competitive manufacturing performance. Table VIII shows the results from our analysis. Based on the literature review, we expected the group that bundled complementary SC organizational and SC IT capabilities to achieve higher levels of performance.

TABLE VIII
SC Strategy Groups and Performance Relative to the Competition

        Performance              SC Strategy Groups (a)
        Dimensions

                                   1               2        3

Perceived performance relative to the competition

  Cost

    Manufacturing cost             3.123           3.336    3.044
(0.117) (0.088)  (0.172)
(0.118)
    Inventory turnover             3.158           3.460    3.435
                                  (0.117)         (0.087)  (0.152)

  Quality

    Conformance to                 3.710*          4.009*   3.750
    product specs                 (0.092)         (0.065)  (0.150)

  Delivery

    On-time delivery               3.579*          4.035*   3.833
(0.125) (0.075)  (0.173)

    Fast delivery                  3.404           3.541    3.174
                                  (0.103)         (0.073)  (0.015)

  Flexibility

    Flexibility to change          3.875           3.956    3.913
    product mix                   (0.085)         (0.069)  (0.177)

    Flexibility to change          3.638*          4.009*   4.000
    volume                        (0.085)         (0.076)  (0.173)

  Summary                         Cluster 2 has a perceived competitive
                                  advantage in quality, on-time delivery
                                  and flexibility to Clusters 1

Reported performance (f), (g)

  Cost

    Mfg. costs as                   0.764           0.762    0.737
    percentage of sales            (0.027)         (0.023)  (0.061)

    LN (inventory turns)            1.771*          2.257*   2.276
                                   (1.399)         (1.111)  (0.720)

Delivery

    Percentage shipped              90.006           91.084   88.486
    on-time)                       (1.459)         (1.115)  (4.881)

    LN (lead-time)                  3.222          2.7942   2.5619
                                   (1.277)         (1.370)  (1.380)

  Summary                          Cluster 2 has higher performance
                                   on inventory turns than Cluster 1

    Performance Dimensions                 Test of Assumptions

                                          Levine (b)       KS (c)

Perceived performance relative to the competition

  Cost

    Manufacturing cost                     LS=3.259        KS=2.985
                                         [rho]=0.041      [rho]=0.000

    Inventory turnover                     LS=1.169        KS=3.242
                                         [rho]=0.313      [rho]=0.000

  Quality

    Conformance to                         LS=2528         KS=4.348
    product specs                        [rho]=0.082      [rho]=0.000

  Delivery

    On-time delivery                       LS=2.958       KS = 3.538
                                         [rho]=0.054      [rho]=0.000

    Fast delivery                          LS=0.680       KS = 3.506
                                         [rho]=0.508      [rho]=0.000

  Flexibility

    Flexibility to change                  LS=0.301        KS=4.888
    product mix                          [rho]=0.741      [rho]=0.000

    Flexibility to change                 LS=0.536        KS=4.215
    volume                               [rho]=0.586      [rho]=0.000

  Summary                         Cluster 2 has a perceived competitive
                                  advantage in quality, on-time delivery
                                  and flexibility to Clusters 1

Reported performance (f), (g)

  Cost

    Mfg. costs as                          LS=0.506       KS= 1.790
    percentage of sales                  [rho]=0.604      [rho]=0.003

    LN (inventory turns)                   LS=0.325       KS= 1.929
                                         [rho]=0.723      [rho]=0.000

Delivery

    Percentage shipped                     LS=4.370        KS=3.326
    on-time)                             [rho]=0.014      [rho]=0.000

    LN (lead-time)                         LS=0.178       KS=0.590
                                         [rho]=0.878      [rho]=0.878

  Summary                          Cluster 2 has higher performance
                                   on inventory turns than Cluster 1

Performance                         Test of Group Differences
Dimensions

                                        ANOVA       Kruskal-Wallis
                                     Results (d)     Results (e)

Perceived performance relative to the competition

  Cost

    Manufacturing cost                 F= 1.695         CS=4.204
                                     [rho]=0.186      [rho]=0.122
                                       N=192

    Inventory turnover                F=2.271          CS = 5.679
                                     [rho]=0.106      [rho]=0.058
                                       N=190

  Quality

    Conformance to                    F=4.064          CS=7.507
    product specs                     [rho]=0.019      [rho]=0.023
                                        N=195

  Delivery

    On-time delivery                  F=5.339          CS=9.284
                                     [rho]=0.006      [rho]=0.010
                                        N=194

    Fast delivery                      F=2.363          CS=3.708
                                     [rho]=0.097      [rho]=0.157
                                        N=190

  Flexibility

    Flexibility to change             F=0.450          CS=1.375
    product mix                      [rho]=0.638      [rho]=0.503
                                        N=193

    Flexibility to change             F=4.960         CS=12.566
    volume                           [rho]=0.008      [rho]=0.002
                                      M = 195

  Summary                         Cluster 2 has a perceived competitive
                                  advantage in quality, on-time delivery
                                  and flexibility to Clusters 1

Reported performance (f), (g)

  Cost

    Mfg. costs as                     F=0.105          CS=0.197
    percentage of sales              [rho]=0.900      [rho]=0.906
                                        N=179

    LN (inventory turns)               F=3.337        CS = 9.034
                                     [rho]=0.038      [rho]=0.011
                                        N=166

  Delivery

    Percentage shipped                F=3.95           CS=2.687
    on-time)                          [rho]=0.674      [rho]=0.261
                                      N=174

    LN (lead-time)                    F=2.254          CS=3.985
                                     [rho]=0.108      [rho]=0.136
                                       N=168

  Summary                          Cluster 2 has higher performance
                                   on inventory turns than Cluster 1

(a) Top values are mean values for each cluster and bottom values
are standard errors. Numbers with * are significantly different
from each other using the Scheffe post hoc analysis (0.05 level).

(b) Levine Statistic used for testing Homogeneity of Variances.

(c) KS is Kolmogorov-Smirnov test of normality.

(d) There were missing values for some performance factors. Therefore,
the sample size varied for each analysis.

(e) Kruskal-Wallis results: CS = Chi-Square statistic,
[rho]=significance. Ranks are no reported since same conclusions are
made from ANOVA.

(f) Ranks for inventory turns are SCS Group 1 (67.9),
Group 2 (92.2) and Group 3 (89.7).

(g) The logarithm of inventory turns and lead time were used for
this analysis.

Based on the perceived competitive advantage, there were no group mean differences for SC Strategy Groups on the dimension of cost. However, there were differences on the dimensions of quality, delivery and flexibility. SC Strategy Group 2 had higher performance relative to quality, on-time delivery, and flexibility to volume changes than SC Strategy Group 1. Firms in SC Strategy Group 2, by developing higher levels of SC organizational capabilities than SC Strategy Group 1, perceived their performance to be more competitive on these three dimensions. This result supports past research that found that the joint development of IT and organizational capabilities can lead to a competitive advantage.

Based on actual performance measures, the SC Strategy Groups only differed on the dimension of inventory turns, with SC Strategy Groups 2 and 3 (both having higher levels of SC organizational capabilities) having higher inventory turns than Group 1. This result suggests that the development of SC organizational capabilities is associated with improved inventory performance.

These results, showing higher performance for SC Strategy Group 2, indicate that the combination of high SC organizational capabilities and high SC IT capabilities enables firms to perform well on quality, delivery and flexibility. The SC IT and organizational capabilities appear to be complementary and help to achieve competitive advantage in SC responsiveness. This result supports the findings from previous research grounded in the RBV (Clemons and Row 1991; Powell and Dent-Micallef 1997).

Our results suggest that there are some perceived and actual performance benefits from specific SC strategies. However, there are differences in the results for the perceived and actual performance. This may mean that plant managers believe that their capabilities are differentiating their performance but, in reality, they are not. However, the differences could also be due to the difficulty of comparing actual measures across multiple industries, countries and throughout the SC. Good performance in one industry may not be good performance in another. Similarly, a good performance level for a second or third tier supplier may differ from that of a firm delivering final product to the consumer. Future research should examine these differences in more detail.

RESEARCH SUMMARY AND LIMITATIONS

This research identified a taxonomy with three SC Strategy Groups of manufacturers with similar combinations of SC capabilities using a data-based analytical approach. We found that there were significant differences on the dimensions of SC organizational capabilities and SC IT capabilities among the three SC Strategy Groups. One group has high levels of both organizational and IT capabilities. A second group has significantly lower levels of SC organizational capabilities than the other groups. A third group has significantly lower levels of SC Internet capabilities than the other groups.

We also found that firms with high levels of both SC organizational and SC IT capabilities outperformed other firms on at least some of the measures of cost, quality, delivery and flexibility. This result suggests that SC organizational capabilities and SC IT capabilities are complementary and that when bundled together enable firms to improve SC performance. When developing SC capabilities, it is important for managers to recognize the interdependency of organizational and IT capabilities.

This analysis also provides evidence of a significant relationship between groups within the SC strategy taxonomy and uncertainty (measured by SC position) and competitiveness, firms that are further away from the customer and have higher levels of competition have lower levels of SC IT capabilities. These firms may not have access to consumer demand information or may not have the resources to develop SC IT capabilities. Future research should examine these firms in more detail.

Surprisingly, our analysis did not provide support for a relationship between groups within the SC capability taxonomy and competitive priorities. While previous studies suggested that the fit between capabilities and strategy can lead firms to have better performance, our results showed that there still is a missing link between SC capabilities and competitive strategy. The implication of our results is that firms do not yet appear to align their SC capabilities with their competitive priorities. In practice, this mismatch between overall business strategy and SC strategy can occur when the SC management organization is a separate entity within the company and/or operates with different organizational metrics of performance. Given our result that SC strategies are related to overall performance, it is important for managers to match their SC and competitive strategies to best meet their strategic objectives.

This research, involving a large-sample database with a diverse set of companies, makes an important contribution to the SC research by identifying and analyzing an SC capability taxonomy. However, it is important to address the limitations of this research and to expand upon our analysis in the future. First, given the number of measures (context, competitive priorities and performance) examined in this study, the type I error may be inflated. Future research should be done to confirm the findings. Second, this study examined a set of firms at one point in time. It will be interesting to take a more longitudinal approach to the analysis and examine how SC strategies evolve over time. This is particularly important given the rapid changes that occur in II. Additional analysis will be needed to understand whether the taxonomy developed in this research will stand the test of time. Third, this study considered a limited number of contextual factors. It will be important to examine relationships between the SC taxonomy and other contextual factors, such as the complexity of the product and the supplier network and the characteristics of the product and customers. It will be helpful to also consider the interactions between contextual factors, such as between country and uncertainty, country and firm size and the percentage of exports. Finally, our research found no relationship between SC Strategy Groups and competitive priorities and some performance measures. Future research should examine the relationship in more detail. It will be useful to examine the performance benefits of alignment among SC capabilities, competitive priorities and the contextual environment.

REFERENCES

Amit, R. and P.J. Shoemaker. "Strategic Assets and Organizational Rent," Strategic Management journal, (14:1), 1993, pp. 33-46.

Anderson, J.C. and D.W. Gerbing. "Structural Equation Modeling in Practice: A Review and Recommended Two-step Approach," Psychological Bulletin, (103:3), 1988, pp. 411-423.

Aviv, Y. 'The Effect of Collaborative Forecasting in Supply Chain Performance," Management Science, (47:10), 2001, pp. 1326-1343.

Barney, J.S. "Firm Resources and Sustained Competitive Advantage," Journal of Management, (17:1), 1991, pp. 99-120.

Barratt, M. and K. Rosdahl. "Exploring Business-to-business Market Sites," European Journal of Purchasing and Supply Management, (8:2), 2002, pp. 111-122.

Boyer, K.K., C. Bozarth and C, McDermott. "Configurations in Operations: An Emerging Area of study," journal of Operations Management, (18), 2000, pp. 601-604.

Boyer, K.K. and C. McDermott. "Strategic Consensus in Operations Strategy," Journal of Operations Management, (17:3), 1999, pp. 289-306.

Boynton, A.C., R.W. Zmud and G.C. Jocobs. 'The Influence of IT Management Practice on IT Use in Large Organizations," MIS Quarterly, (18:3), 1994, pp. 299-320.

Bozarth, C. and C. McDermott. "Configurations in Manufacturing Strategy: A Review and Directions for Future Research," Journal of Operations Management, (16:4), 1998, pp. 427-439.

Cecere, L., K. O'Marah and L. Preslan. "Driven by Demand," Supply Chain Management Review, (8:8), 2004, pp. 15-17.

Christopher, M. H. Peck and D. Towill."A Taxonomy for Selecting Global Supply Chain Strategies," The International Journal of Logistics Management, (17:2), 2006, pp. 277-287.

Clemons, E.K. and M. Row. "Sustaining IT Advantage: The Role of Structural Differences," Special Issue of MIS Quarterly, (15:3), 1991, pp. 274-292.

Coates, T.T. and C.M. McDermott. "An Exploratory Analysis of New Competencies: A Resource Based View Perspective," Journal of Operations Management, (20), 2002, pp. 435-450.

Croom, S.'The Impact of Web-based Procurement on the Management of Operating Resources Supply," journal of Supply Chain Management, (35:1), 2000, pp. 4-10.

D'Avanzo, R., H. Lewinski and L. Sassenhove.'The Link between Supply Chain and Financial Performance," Supply Chain Management Review, (7:6), 2003, pp. 40-47.

DeSantis, G. and M.S. Poole. "Capturing the Complexity in Advanced Technology Use: Adaptive Structuration theory," Organization Science, (5:2), 1994, pp. 121-148.

Dierickx, I. and K. Cool. "Asset Stock Accumulation and Sustainability of Competitive Advantage," Management Science, (35:12), 1989, pp. 1504-1511.

Doty, D.H. and W.H. Glick. 'Typologies as an Unique form of Theory Building: Toward Improved Understanding and Modeling," Academy of Management Review, (19:2), 1994, pp. 230-251.

Ellinger, A.E., D.F. Lynch and J.D. Hensen. "Firm Size, Web Site Content, and Financial Performance in the Transportation Industry," Industrial Marketing Management, (32:3), 2002, pp. 177-185.

Essig, M. and U. Arnold. "Electronic Procurement in Supply Chain Management: An Information Economics-Based Analysis of Electronic Markets," Journal of Supply Chain Management, (37:4), 2001, pp. 43-49.

Fein, A.J. "Where Productivity is Growing in Wholesale Distribution" Supply House Times, (49:9), 2006, pp. 40-42.

Fisher, M. "What Is the Right Supply Chain for Your Product?" Harvard Business Review, March-April 1997, pp. 105-116.

Fisher, M.L., J.H. Hammond, W.R. Obermeyer and A. Raman. "Making Supply Meet Demand in an Uncertain World," Harvard Business Review, May-June 1994, pp. 83-93.

Forrester, J.W. Industrial Dynamics, Pegasus Communications, Waltham, MA, 1961.

Frohlich, M.T. and J.R. Dixon. "A Taxonomy of Manufacturing Strategies Revisited," Journal of Operations Management, (19), 2001, pp. 541-558.

Frohlich, M.T. and R. Westbrook. "Arcs of Integration: An International Study of Supply Chain Strategies," Journal of Operations Management, (19:2), 2001, pp. 185-200.

Frohlich, M.T. and R. Westbrook. "Demand Chain Management in Manufacturing and Services: Web-based Integration, Drivers and Performance," Journal of Operations Management, (20:6), 2002, pp. 729-745.

Ganeshan, R. "Managing Supply Chain Inventories: A Multiple Retailer, One Warehouse, Multiple Supplier Model," International Journal of Production Economics, (59:1-3), 1999, pp. 341-355.

Germain, R. and N.S. Karthik. 'The Interaction of Internal and Downstream Integration and Its Association with Performance," Journal o] Business Logistics, (27:2), 2006, pp. 29-52.

Gunasekaran, A., H.B. Marri, R.E. McGaughey and M.D. Nebhwani. "E-commerce and Its Impact on Operations Management," International Journal of Production Economy, (75), 2002, pp. 185-197.

Hambrick, D.C. 'Taxonomy Approaches to Studying Strategy: Some Conceptual and Methodological Issues," Journal of Management, (10:1), 1984, pp. 27-41.

Hammer, M. and I. Champy. Re-engineering the Corporation: A Manifesto for Business Revolution, Harper Business, New York, 1993.

Harland, C.M., R.C. Lamming, J.. Zheng and T.E. Johnsen. "A Taxonomy of Supply Networks," Journal of Supply Chain Management, (37:4), 2001, pp. 21-27.

Hayes, R.H. and S.C. Wheelwright. Restoring Our Competitive Edge: Competing through Manufacturing, Wiley, New York, 1984.

Hill, T.J. Manufacturing: Text and Cases, 2nd ed., Irvin, Burr Ridge, IL, 1994.

Hunt, S.D. and D.F. Davis. "Grounding Supply Chain Management in Resource-Advantage Theory," Journal of Supply Chain Management., (44:1), 2008, pp. 10-21.

Ittner, C.D. and D.F. Larcker. "The Performance Effects of Process Management Techniques" Management Science, (43:4), 1997, pp. 522-535.

Keen, P. "Information Technology and the Management Difference: A Fusion Map," IBM Systems Journal, (32:1), 1993, pp. 17-39.

Ketchen, D. and C. Shook. 'The Application of Cluster Analysis in Strategic Management Research: An Analysis and Critique," Strategic Management journal, (17), 1996, pp. 441-459.

Kouvelis, P., C. Chambers and H. Wang. "Supply Chain Management Research and Production and Operations Management: Review, Trends, and Opportunities," Production and Operations Management, (15:3), 2006, pp. 449-469.

Lamming, R., T. Johnsen, 1. Zheng and C. Harland. "An Initial Classification of Supply Networks," International Journal of Operations and Production Management, (20:6), 2000, pp. 675-691.

Lawrence, P.R. "Organization and Environment Perspective: The Harvard Organization and Environment Research Program." In A.H. Van de Ven and W.F. Joyce (Eds.), Perspectives on Organization Design and Behavior, John Wiley and Sons, Inc., New York, 1981, pp. 311-345.

Lawrence, P.R. and J.W. Lorsch. Organization and Environment: Managing Differentiation and Integration, Division of Research, Graduate School of Business Administration, Harvard University, Boston, MA, 1967.

Lee, H. "Aligning Supply Chain Strategies with Product Uncertainty," California Management Review, (44:3), 2002, pp. 105-119.

Lee, H., V. Padmanabhan and S. Whang. "Information Distortion in a Supply Chain: The Bullwhip Effect," Management Science, (43:4), 1997a, pp. 546-558.

Lee, H., V. Padmanabhan and S. Whang. "The Bullwhip Effect in Supply Chains," Sloan Management Review, (38:3), 1997b, pp. 93-102.

Lee, H.L., K.C. So and C.S. Tang. 'The Value of Information Sharing in a Two-level Supply Chain," Management Science, (46:5), 2000, pp. 626-643.

Lehman, D.R. Market Research and Analysis, Richard D. Irwin, Homewood, IL, 1979.

Lynagh, P.M., P.R. Murphy, R.E Poist and W.E Grazer. "Web-based Informational Practices of Logistics Service Providers: An Empirical Assessment," Transportation Journal, (40), 2001, pp. 34-45.

March, J.G. "Exploration and Exploitation in Organizational Learning," Organization Science, (2:1), 1991, pp. 71-87.

Martin-Pena, M.L. and E. Diax-Garrido. 'Typologies and Taxonomies of Operations Strategy: A Literature Review," Management Research News, (31:3), 2008, pp. 200-218.

McKone, K.E., R.S. Schroeder and K. Cua. 'Total Productive Maintenance: A Contextual View," Journal of Operations Management, (17:2), 1999, pp. 123-144.

McKone, K.E. and R.S. Schroeder. "A Plant's Technology Approach and Emphasis: A Contextual View," International Journal of Operations and Production Management, (22:7), 2002, pp. 772-792.

McKone-Sweet, K.E., P. Hamilton and S.B. Willis. 'The Ailing Healthcare Supply Chain: A Prescription for Change," Journal of Supply Chain Management, (41), 2005, pp. 4-17.

Meyer, A.D., A. Tsui and C. Hining. "Guest co-editors' Introduction: Configurations Approaches to Organizational Analysis" Academy of Management Journal, (36:6), 1993, pp. 1175-1195.

Miller, D. "Configurations Revised," Strategic Management Journal (17), 1996, pp. 505-513.

Miller, J.G. and A.V. Roth. "A Taxonomy of Manufacturing Strategies," Management Science, (40:3), 1994, pp. 285-304.

Narasimhan, R. and S.W. Kim. "Effect of Supply Chain Integration on the Relationship between Diversification and Performance: Evidence from Japanese and Korean Firms," Journal of Operations Management, (20:3), 2002, pp. 303-323.

Narasimhan, R., S.W. Kim and K.C. Tan. "An Empirical Investigation of Supply Chain Strategy Typologies and Relationships to Performance," International Journal of Production Research, (46:18), 2008, pp. 5231-5259.

Oliver, C. "Sustainable Competitive Advantage: Combining Institutional and Resource-based View," Strategic Management Journal, (18:9), 1997, pp. 697-713.

Pagell, M. "Understanding the factors that enable and inhibit the integration of operations, purchasing and logistics," Journal of Operations Management, (22:5), 2004, pp. 459-487.

Peteraf, M.A. 'The Cornerstones of Competitive Advantage: A Resource-based View," Strategic Management Journal, (14:3), 1993, pp. 179-191.

Porter, M.E. Competitive Strategy, Free Press, New York, 1980.

Porter, M.E. 'Towards a Dynamic Theory of Strategy" Strategy Management journal, (12), 1991, pp. 95-118.

Powell, T.C. and A. Dent-Micallef. "Information Technology as Competitive Advantage: The Role of Human, Business, and Technology Resources," Strategy Management Journal, (18:5), 1997, pp. 375-405.

Prahalad, C.K. and G. Hamel. 'The Core Competence of the Corporation," Harward Business Review, (68:3), 1990, pp. 79-91.

Randall, T.R., R.M. Morgan and A.R. Morton. "Efficient versus Responsive Supply Chain Choice: An Empirical Examination of Influential Factors," Journal of Product Innovation Management, (20:6), 2003, pp. 430-443.

Roberts, B. and M. Mackay. "IT Supporting Supplier Relationships: The Role of Electronic Commerce," European Journal of Purchasing and Supply Management, (4:2-3), 1998, pp. 175-184.

Roth, A.V. and M. Velde. "Operations as Marketing: A Competitive Service Strategy," Journal oj Operations Management, (10:3), 1991, pp. 303-328.

Rungtusanatham, M, F. Salvador, C. Forza and T.Y. Choi. "Supply-chain Linkages and Operational Performance, a Resource-based-view Perspective," International Journal of Operations and Production Management, (23:9), 2003, pp. 1084-1100.

Schroeder, R.G., K.A. Bates and M.A. Junttila. "A Resource-based View of Manufacturing Strategy and the Relationship to Manufacturing Performance," Strategic Management Journal, (23), 2002, pp. 105-117.

Skjott-Larsen, T., H. Kotzab and M. Grieger. "Electronic Marketplaces and Supply Chain Relationships," Industrial Marketing Management, (32:3), 2003, pp. 173-175.

Sterman, J.D. "Modeling Managerial Behavior: Misperceptions of Feedback in a Dynamic Decision Making Experiment," Management Science, (35), 1989, pp. 321-329.

Strader, T.J., F. Lin and M.J. Shaw. "The Impact of Information Sharing on Order Fulfillment in Divergent Differentiation Supply Chains," Journal of Global information Management, (7:1), 1999, pp. 16-26.

Subramani, M. "How Do Suppliers Benefit from Information Technology Use in Supply Chain Relationships?," MIS Quarterly, (28:1), 2004, pp. 45-74.

Swink, M., R. Narasinham and C. Wang. "Managing Beyond the Factory Walls: Effects of Four Types of Strategic Integration on Manufacturing Plant Performance," Journal of Operations Management, (25:1), 2007, pp. 148-164.

Tabachnick, B. and L. Fidell. Using Multivariate Statistics, Harper and Row Publisher, New York, 1989.

Tamas, M. "Mismatched Strategies: The Weak Link in the Supply Chain?," Supply Chain Management, (5:4), 2000, pp. 171-175.

Tan, K.C., V.F. Kannan and R.B. Handheld. "Supply Chain Management: Supplier Performance and Firm Performance," International Journal of Purchasing and Materials Management, (34:3), 1998, pp. 2-10.

Venkatraman, N. and J..E. Prescott. "Environment-strategy Coalignment: An Impirical Test of Performance Implications," Strategic Management Journal, (11), 1990, pp. 1-23.

Vickery, S.K., J. Jayaram, C. Droge and R. Calantone. 'The Effects of an Integrative Supply Chain Strategy on Customer Service and Financial Performance: An Analysis of Direct versus Indirect Relationships," Journal of Operations Management, (21:5), 2003, pp. 523-539.

Wernerfelt, B. "A Resource-Based View of the Firm," Strategic Management Journal, (5:2), 1984, pp. 171-180.

Wisner, J.D. "A Structure Equation Model of Supply Chain Management Strategies and Firm Performance," Journal of Business Logistics, (25:1), 2003, pp. 1-25.

Kathleen McKone-Sweet (Ph.D., University of Virginia) is an associate professor in the Technology Operations and Information Management Division at Babson College in Babson Park, MA. Dr. McKone-Sweet received her B.S. and M.Eng. in Operations Research and Industrial Engineering at Cornell University. She received her MBA and Ph.D. in Operations Management at the University of Virginia. Her current research is in supply chain management. Her work appears in numerous academic and practitioner journals, including Journal of Operations Management, Production Operations Management, Journal of Supply Chain Management, and Supply Chain Management Review.

Yoo-Taek Lee (DBA, Boston University) is an assistant professor of technology and operations management at Babson College in Babson Park, MA. Professor Lee's research interests include global SC strategies, web-based SC management applications and supplier development strategies from the view of socioeconomics. His current work investigates development strategies of small and medium-sized suppliers in global SCs and complementary resources to the implementation of web-based applications in managing SCs. His work experience includes HR at Samsung, a research associate appointment at Boston University Asian Management Center, and the development of executive education programs for Asian organizations such as Sanyo, Samsung, LG, Daewoo, SK, the Federation of Korean Industries and the Ministry of Commerce in China. He has produced more than 65 best practice reports of Motorola, IBM, Texas Instruments, GE, Federal Express, Intel, Sun Microsystems, Toyota and others.

KATHLEEN MCKONE-SWEET AND YOO-TAEK LEE

Babson College

APPENDIX B

Measurement of SC Internet Capabilities

For which of these activities does your plant use the Internet? Please circle either Yes (1) or No (0). (Respondent is information systems manager.)

Exploitation

Transmitting orders to suppliers

Tracking/tracing supply orders

Real-time integrated scheduling, shipping and warehouse management across supplier network

Providing fixed pricing offers to potential buyers

Online order entry

Customers can check delivery status of their orders

Exploration

Scanning the marketplace for identification of potential sources

Receiving and comparing suppliers' offers

Providing dynamic pricing (negotiations and sellers' bids) for purchased items

Supporting collaborative product design/improvement with suppliers

Supporting collaborative process and technology design/improvement with suppliers

Providing on-line customized customer service, where customers can configure the product within the constraints stated by the plant

Providing dynamic pricing to potential buyers