Trategicsupply Chainmanagement: Improvingperformancethrougha Culture of Competitivenessand Knowledge Development

Strategic Supply Chain Management: Improving Performance through a Culture of Competitiveness and Knowledge Development Author(s): G. Tomas M.
Hult, David J. Ketchen Jr. and Mathias Arrfelt Source: Strategic Management Journal, Vol. 28, No. 10 (Oct., 2007), pp. 1035-1052 Published by: Wiley Stable URL: http://www.jstor.org/stable/20141968 . Accessed: 19/11/2013 11:07
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Wiley and John Wiley & Sons are collaborating with JSTOR to digitize, preserve and extend access to Strategic Management Journal.
http://www.jstor.org
This content downloaded from 168.156.40.175 on Tue, 19 Nov 2013 11:07:50 AM All use subject to JSTOR Terms and Conditions
Strategie
Published online 17 May 2007 inWiley InterScience Received
Management
J, 28: DOI: received
Journal
(2007) 10.1002/smj.627 20 March 2007
Strat. Mgmt. (www.interscience.wiley.com) 2005; Final revision
1035-1052
14 October
STRATEGICSUPPLY CHAINMANAGEMENT: IMPROVING PERFORMANCETHROUGHA CULTURE OF COMPETITIVENESSAND KNOWLEDGE DEVELOPMENT

G. TOMAS M. HULT,1*DAVID J. KETCHENJR2 and MATHIAS ARRFELT1 1
Eli Broad Graduate
U.S.A. of Business, Auburn University, Auburn, Alabama, U.S.A.
School
of Management,
Michigan
State University,
East Lansing,
Michigan, 2 College
For
a central as competitive element has become chains their supply weapons many firms, using on recent the resource-based view and in the process years. management Drawing strategic of this study uses and information-processing the organizational literatures, learning theory from a sample a to and culture the 201 examine knowledge of competitiveness influence of of firms on supply conditions. We found in varied market turbulence chain performance development that
a culture and knowledge their exist between development: synergies of competitiveness on behavioral In addition, based and has a positive association with performance. interaction a we these market turbulence moderates that theories, having relationships, contingency found on on the knowledge link and a negative influence positive development-performance influence about the level of who are confident link. Managers the culture of competitiveness-performance market either firms over turbulence a culture whose time, they will face can of competitiveness are unlikely managers a focus on both a culture to emphasize use this sense to decide whether developing or knowledge chains. For those their in supply development to be able to predict the degree they will face of turbulence to is critical and knowledge development of competitiveness
ensuring success. Copyright ? 2007 JohnWiley & Sons, Ltd.
INTRODUCTION
The quest to discover the determinants of firm per formance has long been central to the strategic field. Indeed, many management leading schol ars have argued that building knowledge about cor some others is the firms why outperform nerstone and Li, of the field (e.g., Hitt, Boyd, and Teece, Schendel, 1994; Sum 2004; Rumelt, mer et al., 1990). In recent years, the nature of shifted toward 'sup has increasingly competition vs. chain' chain (Handfield struggles supply ply
chain management; perfor Keywords: supply strategic resource-based mance; culture; knowledge development; view to: G. Tomas M. Huit, Eli Broad Graduate ^Correspondence East Lans State University, School of Management, Michigan E-mail: hult@msu.edu U.S.A. ing, MI 48824-1121,
and Nichols, 2002; Slone, 2004). Supply chains are value-adding relations of partially discrete, yet transform raw inter-reliant, units that cooperatively materials into finished products through sequen tial, parallel, and/or network structures (Bowersox, rivals such as UPS Closs, and Stank, 1999). When and FedEx their individual clash, it is not merely the collective but rather capabilities of capabilities, determine the that their respective supply chains,
outcome.
field has the strategic management Historically, to not devoted much attention supply empirical such as marketing chains, while related disciplines have and operations management long empha activities. of operational the performance implications in a review of the opera For example, tions management literature, Anderson, Cleveland, and Schroeder (1989: 134) noted: 'proper strategic sized
?WILEY
Copyright ? 2007 JohnWiley & Sons, Ltd.
litferScience'
1036
G. T. M. Huit,
D.
J. Ketchen
Jr and M.
Arrfelt of research on strategic supply chain manage ment. We build on Huit et al. (2002), who intro duced the concept of 'cultural competitiveness' as a reflection of innovativeness, entrepreneurial, and learning orientations,1 and Huit et al. (2004), who examined the knowledge pro development cess, both within rior performance. both studies but tinct. Taking the cern for learning build on of achieving supe a key element of is Learning the frameworks tested are dis studies' shared con previous as our point of departure, we the context
or aligning of positioning ties can significantly impact
and business performance recent years, a small body of strategic manage ment research has begun to examine 'strategic use of a supply chain supply management'?the chain not merely as a means to get products where they need to be, but also as a tool to enhance and Nichols, key outcomes (e.g., Huit, Ketchen, and Slater, 2004). The value 2002; Huit, Ketchen, of strategic supply chain management is reflected in how firms such as Wal-Mart, Zara, Toyota,
operations capabili competitive strength of an organization.' In
used their supply chains as com to gain advantages over peers. petitive weapons to Meanwhile, failing strategically manage supply chains offers serious negative As consequences. Lee (2004) describes, for example, chain supply difficulties led Cisco to write off $2.25 billion in to lose many inventory in 2001 and led Motorola crucial early camera phone sales in 2003. Given the implications for profits and sales, it is perhaps not surprising that the announcement of a major and Dell have supply chain problem erodes a firm's market value by an average of 10 percent (Hendricks and Sing hal, 2003). Like the Huit et al. studies, we focus on explain ing order fulfillment cycle time?the length of an time between order and delivery of the taking needed product to the customer. As Ray, Barney, and Muhanna the effec (2004) note, measuring tiveness of business processes helps test resource based logic and taps into the competitive advan within activities. tages developed Cycle important time is a key metric for directly assessing sup (Nichols, Retzlaff-Roberts, ply chain functioning and Frolick, 1996). More importantly, cycle time is central to a firm's strategic success. As Hand field and Nichols (2002: 13) note, cycle time not has 'a direct only linkage to profits' at the firm but in cycle time allows firms excellence level, to 'grow faster and earn higher profits relative to in their industry, increase market share through early introduction of new products, con trol overhead and inventory costs, and move to
the resource-based view (Wernerfelt, and from the 1984), theory organizational learning (Huber, 1991) and information-processing (Daft and Weick, 1984) literatures to argue that neither a culture of competitiveness nor knowledge devel itself is sufficient to achieve opment by superior in varied market conditions. Instead, performance these phenomena in tandem to achieve operate desired outcomes. data from 201 firms, Using we apply a sophisticated technique?parsimonious interaction modeling (e.g., Ping, 1995)?to highlight the potential value of two phe nomena that together can facilitate superior cycle
time.
latent-variable
THEORETICAL HYPOTHESES
Recent
FOUNDATION
AND
et al. research and by Ray (2004) and Junttila (2002) highlights Schroeder, Bates, the value of examining resources within a firm's In line with this process. operations management our conceptual work, Figure 1 presents model, which is intended to explain cycle time in supply chains. The model includes two higher-order fac tors?culture of competitiveness and knowledge of seven first-order indi development?composed cators (each of which, in turn, has a set of reflective indicators?see 1), as well as their inter Appendix action. Culture of competitiveness (CC) is defined as the 'degree to which [supply] chains are pre to detect and fill gaps between what disposed
1 and Nichols introduced the concept of Huit, Ketchen, (2002) 'cultural competitiveness.' As an anonymous referee pointed out, the term 'cultural competitiveness' seems to denote a comparison of one firm's competitive characteristics against those of another to see which is more successful. Based on this referee's sugges This better tion, we adopt the term 'culture of competitiveness.' reflects the underlying to which focus on the degree concept's values and beliefs centered on customer service are developed. We appreciate the referee's insights on this issue. Strat. Mgmt. J., 28: 1035-1052 (2007) DOI: 10.1002/smj
other firms
of industry leadership.' In contrast to the focus of the Huit et al. studies, single-organization we examine the supply chains of multiple firms. This design feature allows us to shed new light on the critical issue of why some firms outperform positions
others.
This paper is devoted as a next logical step

Copyright ?
to taking what we view in the emerging stream
2007 John Wiley & Sons, Ltd.
Strategie Supply Chain Management

Market Turbulence Learning Orientation Culture of
1037
Innovativeness Orientation
competitiveness
Firm Age
Entrepreneurial Orientation Interaction
H3 (+)
Knowledge Acquisition Information Distribution
Effect
siCC'Km
Cycle Time Performance ?.
Firm Size Knowledge Development
H2(+)
Shared Meaning
Achieved Memory ? Hypothesized First-Order Control and time Relationship Latent Indicators
Relationship in supply
Figure
1.
A model
of
culture
of
competitiveness,
knowledge chains
development,
cycle
performance
the market desires and what is currently offered' on the resource (Huit et al., 2002: 577). Drawing based view (Wernerfelt, 1984), CC is conceptu latent factor (Godfrey alized as an unobservable and Hill, 1995) that is reflected in three orienta and learn tions?innovativeness, entrepreneurial, orienta The latter affects performance. ing?that the critical element that helps tion?learning?is and CC Specif development. knowledge integrate focuses on the values ically, learning orientation that direct supply chains toward the and beliefs behaviors required for knowledge development. (KD), on the other Knowledge development actions lead to wherein hand, is a phenomenon information distribution, acquisition, knowledge in the sup and achieved memory shared meaning, 1991). As ply chain (Huit et a/., 2004; cf. Huber, is reverberated in a set such, a learning orientation (Baker and Sinkula, reflected by knowledge (e.g., Grant, 1996). Research producing behaviors on organizational 1991) and learning (e.g., Huber, of knowledge-seeking KD 1999) while values is information 1984) (Daft and Weick, processing serve as the primary for the four foundation first-order indicators of KD?knowledge acqui shared meaning, information distribution, sition, rela its higher-order and achieved memory?and The in chains. with tionship supply performance
Copyright ? 2007 JohnWiley & Sons, Ltd.
broader learning literature (i.e., learning orienta tion and organizational learning) is the basis for integrating CC and KD in the model (e.g., Argyris and and Sch?n, 1978; Hedberg, 1981; Nystr?m Starbuck, 1984).
Culture
of competitiveness
in supply
chains
As Barney and Mackey (2005: 5) note, the contin of the resource-based ued theoretical development to not 'simply correlate view requires scholars of resources' at the firm level aggregate measures to the levels but rather to move their investigations of analysis 'where resources reside.' Thus, theory 'at the should be aimed and empirical attention level of the resource, not the level of the firm.' The supply chain offers one such level of analy role at sis where resources reside, and resources' as Huit et al. this level can be prominent. Indeed, do (2002: 580) observe, because chain members not all share 'a common organizational affiliation, ... may be of unique resources the development In this sense, shared sup vital to chain outcomes.' can resources substitute for traditional chain ply features that bind members of a firm, such as struc ture, culture, and strategy (cf. Weick, 1987). on Huit et al. the resource-based view, Building a that culture of (2002) argue competitiveness
Strut. Mgmt. 7., 28: 1035-1052 (2007) DOI: 10.1002/smj
1038
G. T. M. Huit,
D.
J. Ketchen
Jr and M.
Arrfelt and members (Kohli, Jaworski, and Kumar, 1993). or shared meaning, Information is interpretation, common the process by which members develop about data and events (Corner, understandings and Keats, the lack of a Kinicki, 1994). Given in typical supply chains, shared strong culture of supply chain data and events are meanings to harness collective needed action (Huit et al, 2004). Perhaps the most integral component of KD is 'organizational memory' (Huber, 1991), labeled 'achieved memory' for the supply chain context by Huit et al. (2004) based on work by Moorman and Miner is defined as the amount (1997). Memory of knowledge, and familiarity with the experience, chain its process, supply operations, and behaviors; it serves as the mechanism by which knowledge is stored for future strategic use and, as such, is critical as a 'launching' point for future learning behaviors. from the organizational information Theory literature provides the basis for expect processing as a group, the four dimensions should that, ing chain Information enhance supply performance. theory argues that gathering, process processing ing, and interpreting information is the primary job of organized collectivities 1984) (Daft and Weick, et al., 1999). such as supply chains (Bowersox on 'strategic sensemaking' Research has extended to demonstrate this argument that information activities profoundly processing shape the strate firms decisions made within and the resul gic tant outcomes (Meyer, 1982; Thomas, Clark, and Gioia, view (Grant, 1993). The knowledge-based a also 1996) supports development knowledge on the resource-based link. Building performance view's notions of value, rarity, and inimitability, the knowledge-based view centers on the notion that unique abilities to create and exploit wisdom create competitive advantages and thereby enhance et al., 2004). As such, within outcomes Huit (e.g., is that: the supply chain context, our contention Hypothesis positive
performance.
functions as an intangible strategic resource that can be developed by interaction and cooperation CC provides sup among supply chain members. ply chain members with a pattern of shared values and beliefs that assert the importance of certain and drive the chain's As such, CC is rooted approach in the broad phenomenon of 'culture' but is nar rowly focused on a distinct set of cultural orienta and learn innovativeness, tions?entrepreneurial, to fill chains lead strategically ing?that supply future desires and what gaps between customers' is currently offered. is defined as the An entrepreneurial orientation elements (and omit others) to the marketplace. values associated with the pur chain members' and the renewal suit of new market opportunities of existing areas of supply chain activities (e.g., ori Naman and Slevin, 1993). An innovativeness to supply chain members' values new idea generation (i.e., mem to new ideas; Hurley and Huit, bers' openness as mem A is defined orientation 1998). learning bers' values associated with the generation of new entation refers associated with to shape supply insights that have the potential Each of these chain activities Huber, 1991). (cf. is necessary, but individually three orientations insufficient, for the emergence of the higher-order intangible strategic resource of culture of com (Huit et al, 2002). Most importantly, petitiveness rooted in the resource-based view, CC appears to be a valuable, rare, and inimitable strategic resource in supply chains (Barney, 1986; Werner felt, 1984) that can provide a sustainable compet itive advantage and enhanced performance (Huit et al., 2002). Thus, we expect that: Hypothesis a positive
mance.
has 1: Culture of competitiveness with cycle time perfor association
Knowledge
development
in supply
chains
four dimensions that Huber (1991: 90) describes are paramount to learning efforts. Huit et al. (2004) a model to develop of built on these elements dimension is The first development. knowledge which process knowledge by acquisition?the or supply chains, such as organizations entities, is the pro Information distribution obtain wisdom. sources cess by which from different information In is shared. supply chains, this sharing occurs its various nodes throughout the chain, including
Copyright ? 2007 John Wiley & Sons, Ltd.
2: Knowledge association
development with cycle
has a time
culture of competitiveness Synergy between and knowledge development The and (e.g., Argyris and Star Sch?n, 1978; Hedberg, 1981; Nystr?m buck, 1984) serves as the theoretical foundation broader learning literature
Strat. Mgmt. J., 28: 1035-1052 (2007) DOI: 10.1002/smj
Strategie for learning being the key integrator of a cul ture of competitiveness and knowledge develop ment in supply chains. While Huit et al. (2002, both the CC and KD constructs 2004) developed within supply chains, they did not integrate the two concepts. This is unfortunate because the learning is orientation construct within the CC framework on focused the supply chain's knowledge-seeking values 1999) that guide its (Baker and Sinkula, within the KD behaviors knowledge-producing framework (e.g., Grant, 1996; Huber, development link in 1991). As such, learning is both the missing et al. (2002, 2004) the conceptualizations by Huit and the resultant integrator of the two frameworks. In other words, their shared concern for learning suggests that neither CC nor KD is sufficient to maximize Instead, they supplement performance. and reinforce each other for a stronger strategic effect than either alone can provide. For example, Baker and Sinkula (1999: 416) of an organization argue that 'if members [e.g., learning orien supply chain] have an enhanced
Supply
Chain Management
1039
in the composition of change of customers and their preferences and Kohli, (Jaworski 1993)?as one critical element of the environment that the an on has influence the relationships oretically in this research and Beard, (e.g., Dess we In addition, 1984). place particular emphasis on the notion that managerial perceptions, particu larly regarding market uncertainty, shape strategic studied and decision making (Child, 1972; Dun can, 1972; Lawrence and Lorsch, 1967). Similarly, and Dean Sharfman (1991: 682) state that 'the is those parts of the external infor environment mation flow that the firm enacts through attention and belief.' One logical extension is that environ mental and beliefs perceptions shape culture and behavior (Dutton and Jackson, 1987). We that this argument also will hold expect true in supply chains. For example, one of behav mem ioral theory's tenets is that organizational on the conditions in which the ory is dependent choice
tation, they will not only gather and dissemi nate information about markets but also constantly examine functions the quality of their interpretive storage and the validity of the dominant logic that guides the entire process.' At the same time, in the behaviors stressing knowledge-producing com to to the 'culture of chain is lead likely supply
firm operates (Cyert and March, 1963; Levitt and March, 1988). Thompson (1967: 159) considered to be the 'essence of dealing with uncertainty the administrative process.' Accordingly, supply influence chains are likely to realize a positive of market turbulence on the knowledge develop time relationship ment-cycle given the dynamic nature of the behaviors in KD. Indeed, involved applying the concept of requisite variety (Ashby, 1956) suggests that, as the environment's pace of change increases, a premium on developing knowl edge emerges. Requisite variety means that organi zational entities, such as supply chains, must match the environment's complexity with their own inter nal strategies at developing A supply chain adept a greater arse possesses knowledge for overcoming nal of wisdom the complexities created by rapid change than do other supply chains. Thus: Hypothesis 4: Market turbulence has a positive influence on the relationship between knowledge and cycle time performance. development and activities.
infrastructure exemplified by the val petitiveness' ues inherent in a learning orientation (e.g., Slater and Narver, 1995). Applied within supply chains, the expectation of a synergistic interaction between CC and KD is also consistent with Day's (1994) that center on inside-out and outside-in processes the strategic interaction between superiority in pro cess management, and integration of knowledge, on this logic, we diffusion of learning. Based expect that: Hypothesis 3: The interaction between culture
and knowledge development of competitiveness has a positive association with cycle time per
formance.
The moderating Starbuck's (1976)
role of market review
turbulence
task of organizational a wealth of potential provided that can affect firm strategy and oper dimensions ations. In our study, we draw from this liter rate ature to focus on market turbulence?the environments
Structural contingency that the theory suggests value of a resource depends on the context within which it is deployed (Lawrence and Lorsch, 1967). on this Building general tenet, we expect market to suppress the culture of the com turbulence petitiveness-performance relationship. As defined above, CC reflects a supply chain's predisposition to spot and strategically plug gaps between what
Strat. Mgmt. /., 28: 1035-1052 (2007) DOI: 10.1002/smj
1040
G. T. M. Huit,
D.
J. Ketchen
Jr and M. Arrfelt individuals who have responsibilities in supply chain management. ISM is best known for its Index (PMI)?a Purchasing Managers' composite index of purchasing activity among manufacturing firms that is closely monitored insti by financial tutions and economists. We restricted our sample tomanufacturing firms, to focus on the last and instructed respondents order fulfillment process within their supply chains. The sampling frame consisted of a total of 2000 with 201 supply chain management professionals response rate of 10.73 responding for an effective These indi (127 were non-deliverable). percent viduals had been with their firms an average of 11 years, and they represented firms that had existed for an average of 64 years, employed an average of 13,688 people, and had an aver in their supply management age of 38 people unit. The executives who had titles responded such as Director of Purchasing, Director of Pur and Materials Vice Presi chasing Management, dent of Procurement, and Chief Purchasing Offi
cer.
the market desires and what the chain currently offers (Huit et al, 2002). Under low levels of tur these gaps are relatively consistent and bulence, can slow developing, that CC be effec suggesting turbulence tively targeted at filling the gaps. When is high, however, the market's desires shift rapidly and unpredictably, leading the gaps that CC seeks to fill to be fluid and nebulous. Indeed, as Aldrich (1979: 69) stresses, a high turbulence 'leads to externally induced ... that are obscure to administrators and changes difficult to plan for.' Weiss and Heide (1993) also can note that rapid change in the marketplace to already-existing be destructive and detrimental cultural competencies (e.g., a culture of competi level of tiveness) that are deeply ingrained and embedded in the values and belief system of supply chain members. turbulence Thus, while greater market increases the supply chain's knowledge develop ment requirements 1981), (Levinthal and March, also serves greater turbulence in the marketplace as a detriment to a culture of competitiveness. As such, we expect that: 5: Market turbulence has a negative Hypothesis on the between a culture influence relationship and time of competitiveness cycle performance.
and Overton's (1977) assess non-response to extrapolation procedure bias. Table 1 summarizes the results. Although we found a significant difference (p < 0.05) between Armstrong the first and fourth quartiles of the respondents for firm age (with early respondents firms' averaging 55 years and late respondents averaging 74 years), no systematic differences were found between the and late bias early respondents. Thus, non-response is likely not an inhibitor in our analyses.
We
used
METHOD
Data collection
Prior to collecting the data in 1999, we pretested our scale items with eight academics and seven chain executives. Also, we management supply a pilot study with 36 supply manage performed ment executives to assess steps quality. These changes to the instructions to respon being made, mainly dents and the need to keep the responses anony mous to secure study participation (i.e., we opted the surveys for identification pur based on concerns raised in the pretests and (1985) pilot study). Following Huber and Power's on to how data from get guidelines quality key a survey was developed informants, using Dill poses to sup and administered (1978) method chain executives drawn the from management ply of the of Institute membership Supply Manage ment in 1915, ISM is a not-for (ISM). Founded of about 45,000 profit professional organization man's
Measures Tables 2 and 3 present the results of the measure ment assessment. Table 2 summarizes the vari standard deviations, means, correlations, and shared variances. Table 3 reports the average construct reliabilities, variances factor extracted, scales were loadings, and fit indices. Established to measure culture of competitiveness (learn and entrepreneurial orienta ing, innovativeness, tions), knowledge development (knowledge acqui shared meaning, sition, information distribution, and achieved memory), market and turbulence, time firm and size Also, age cycle performance. were included as control variables (e.g., Amburgey used and Rao, 1996). Appendix and their sources. 1 lists the scales used ables'
the research in some resulted
design's
not
to code
Strategie
Table 1. Comparison of early and late respondents S.D.
Supply
Chain Management and RMSEA = 0.07
1041 (X2 =
TLI
Respondents
Mean
all being 0.96, 986.92, d.f. = 491).
Learning
orientation
Early
Late
50
51
5.67
5.92
1.19
0.83
Innovativeness
orientation
Early
Late
50
51
5.13
5.08
1.23
1.29
cultural competitiveness Higher-order (KD) model knowledge development Given CC the theoretical and KD constructs ducted a higher-order structs, arguments in Figure assessment
(CC) and
Entrepreneurial orientation Knowledge acquisition Information distribution Shared meaning Achieved memory Market turbulence Cycle time Firm age* Firm size
Early Late Early Late Early Late Early Late Early Late Early Late Early Late Early Late Early Late
50 51 50 51 50 51 50 51 50 51 50 51 50 51 50 50 50 50
3.97 4.36 4.22 4.19 4.62 4.75 4.78 4.78 5.31 5.51 4.95 4.82 4.35 4.70 55.24 74.29 15158.94 15625.24
1.29 1.31 1.21 1.14 1.32 1.32 1.11 1.33 1.12 0.87 1.21 1.17 0.96 1.09 42.06 45.39 32458.88 25446.49
the underlying 1, we next con of these con
including all purified items, the first-order indicators. The indicators, and the second-order results indicate that, in addition to the item load ings reported in Table 3 for each of three CC and four KD
construct's
dimensions,
higher-order
there
is support
As
for each
learn
structure.
such,
= 0.64, ?-value = 8.26, p < 0.01), ing (loading = 0.88, innovativeness ?-value = 11.25, (loading = 0.81, < and p 0.01), entrepreneurship (loading ?-value = 8.91, /? < 0.01) function as first-order indicators of the higher-order construct of CC (R2$ the first-order range from 40% to 78%), where of the reflective indicators are composed indica in Appendix 1. Likewise, knowledge = 0.80, ?-value = 8.72, p < acquisition (loading = 0.89, information distribution 0.01), (loading ?-value = 8.62, shared p < 0.01), meaning = 0.88, ?-value = 11.82, p < 0.01), and (loading = 0.56, ?-value = 7.11, achieved memory (loading < as first-order indicators of the 0.01) function p construct of KD (R2's range from higher-order 31% to 78%). The model fit for the higher-order structure was 0.96 for each of the DELTA2, RNI, = CFI, and TLI indices, and 0.08 for RMSEA (x2 = 624.85, d.f. 267). tors included
p < 0.05.
measures to were All subjected perceptual and assessments of dimensionality, reliability, validity. The psychometric properties of the nine latent constructs involving 44 items were evaluated in one confirmatory factor analysis simultaneously LISREL 8.80 (J?reskog et al., 2000). (CFA) using we struc the higher-order examined Additionally, ture of CC and KD to provide empirical support, in addition to the theoretical rationale, for the focus on these constructs at the higher-order aggregate
level.
Composite We
reliability
Fit of the measurement The model indices fit was
assessed the latent factors' reliability by cal reliability for each construct culating a composite (Fornell and Larcker, 1981). The formula specifies that
model evaluated using a series of and Anderson by Gerbing
"
recommended
(1992) andHu and Bentler (1999)?the DELTA2,

relative noncentrality (RNI), comparative fit (CFI), Tucker-Lewis (TLI), and the root mean square error of approximation indices. After (RMSEA) items (see Appendix 1), an removing inadequate excellent fit to the data was achieved for the first order based CFA, with DELTA2,
(??K/)2+ (!> )
where CR^ = composite for scale rj\ reliability = standardized for scale item yt, and loading Xyi error for scale item yt. Along Si = measurement with the reliability calculations, we also examined the parameter values as well t estimates and their associated as the average variances extracted
Strut. Mgmt. J., 28: 1035-1052 (2007) DOI: 10.1002/smj
RNI,
CFI,
and
1042
Table 2.
G. T. M. Huit,
Means, standard
D.
J. Ketchen
Jr and M. Arrfelt
and shared variances (n = 201)
deviations, S.D. 0.98 1.13 1.27 1.15
correlations, LO ? IN 0.30 0.55 0.33 0.34 ?
Mean 5.88 5.33 4.43
EO
0.11 0.41 ?
KA 0.12 0.19 0.29

?
ID 0.12 0.24 0.25

0.49
SM 0.14 0.30 0.31

0.37
AM 0.17 0.22 0.20

0.18
MT 0.00 0.00 0.01

0.08
CT 0.11 0.19 0.22

0.18
AGE 0.00 0.01 0.02

0.01
SIZE 0.00 0.04 0.04

0.03
Learning
orientation (LO)
Innovativeness
orientation (10)
Entrepreneurial 0.64 0.44
orientation (EO) Knowledge

acquisition
4.33
0.52
(KA)
Information 4.84 4.99 1.22 1.16 1.00 1.18
0.35 0.37 0.41 0.04 0.33

-0.04 -0.05
0.49 0.55 0.47 0.06 0.44

-0.12 -0.20
0.50 0.56 0.45 0.12 0.47

-0.14 -0.20
0.70 0.61 0.42 0.29 0.42

-0.09 -0.18
? 0.68 0.49 0.22 0.40

-0.06 -0.24
0.46 ? 0.48 0.11 0.37

-0.06 -0.25
0.24 0.23 ? 0.17 0.34

-0.08 -0.15
0.05 0.01 0.03 ? 0.11

-0.02 0.09
0.04 0.14 0.12 0.01 ?

-0.14 -0.24
0.00 0.00 0.01 0.00 0.02

? 0.36
0.06 0.06 0.02 0.01 0.06

0.13 ?
distribution (ID)
Shared meaning
(SM)
Achieved memory 5.55 4.72 4.57
63.62 13,688
(AM)
Market turbulence
(MT)
Cycle time (CT) Firm age (AGE) Firm size (SIZE)
The 1.06 43.27 32,948
variances
are included in the lower triangle of the matrix. correlations are included in the upper triangle of the matrix.
All
correlations
>0.14
are significant
at the p < 0.05
level.
Shared
Table
3.
Summary
statistics
of
the confirmatory variance
factor
analysis
(n
201)
Construct orientation Learning orientation Innovativeness Entrepreneurial Knowledge Information Shared Achieved Market orientation acquisition distribution
Average
extracted
Composite
0.82 0.90 0.86 0.80 0.81 0.92 0.90 0.86 0.74
reliability
Range of loadings 0.47 0.81 0.70 0.61 0.64 0.83

0.80 0.68
meaning memory turbulence
Cycle time = 986.92 X2

d.f. = 491
62.0% 69.8% 60.5% 50.5% 59.3% 74.0% 76.0% 56.2% 45.0%
to 0.93 to 0.89 to 0.83 to 0.79 to 0.89 to 0.89

to 0.91 to 0.81
0.36 to 0.95
DELTA2 = 0.96 RNI = 0.96 CFI = 0.96 TLI = 0.96 RMSEA = 0.07
(Anderson and Gerbing, extracted was calculated
1988). Average as
variance
measurement reliabilities
error for scale from 0.74
loadings the average 76 percent also found where = = average variance extracted for 77; kyi Vn = and item for scale standardized st yi9 loading
ranged ranged from 0.36 variances
item y?. The scales' to 0.92, the factor (p < 0.01), and ranged from 45 to
to 0.95
extracted
(Table 3). The 34 purified items were to be reliable and valid when evaluated item's error variance, modification
based
on each
index, and residual
covariation.
Copyright ?
Strategie Discriminant validity
Supply
Chain Management
1043
the reliability analysis, we established Following discriminant validity by two independent methods. the shared variance between First, we calculated that it was each pair of constructs and verified lower than the variances extracted constructs (Fornell and Larcker,
variance was calculated as
for the involved 1981). Shared
of the hypotheses, within of each tech and constraints strengths et For Shook al., 2004). nique (e.g., example, on allows the the one hand, hierarchical regression direct assessment of change in explanatory power between iterative steps (which we cannot accom using SEM given that our step plish definitively 1 equation, for example, is saturated). Further, as a traditional technique, set it provides a baseline of results for our predictions. On the other hand, the different latent-variable 'parsimonious complex interaction technique' allows for the inclusion of errors and indicators of the higher measurement order factors, and can account for potential CMV et al., 1997; Podsakoff (e.g., Netemeyer problems et al., 2003). Hierarchical moderator the more
for a robust assessment
y2=
l-x/f
where y2 = shared variance between constructs, and with the diagonal element of x// indicating the amount of unexplained variance. Because r\ and s were standardized, y2 was equal to the squared correlation between the two constructs. As shown in Table 3, the average variances extracted were above 50 percent for all but one construct (cycle time, 45%). The shared variances between pairs of all possible scale combinations indicated that the average variances extracted were higher than the shared variance in all cases (Table 2). Second, we examined all possible pairs of con and Phillips structs, as suggested by Bagozzi a in of two-factor CFA models series (1982), using LISREL 8.80. Specifically, each pairwise CFA the 0 coef model was run twice: first, constraining associated to unity; and second, allowing 0 to vary on the results of a x2 difference Based freely. model test, the unconstrained signifi performed the constrained model better than associated cantly > 3.84 was exceeded in when 0 = 1 (i.e., Ax2(d was found between all cases). The lowest Ax2(d knowledge acquisition and information distribution = 32.06). (AX2(1) 2, assessment at Finally, as detailed in Appendix the measurement level found no evidence of com mon method variance. Appendix 2 also describes a more sensitive test conducted at the hypothesis level. Overall, the nine measures and their 34 puri ficient fied indicators were found to be reliable in the context of this study. and valid
regression
analysis
we used As a first step in testing the hypotheses, hierarchical Because three interaction regression. terms were included in the equation, we standard ized all variables to reduce the potential effects of (Cohen et al., 2003). The tech multicollinearity of least squares was used with the control nique variables entered as a block in step 1 (firm age and size), followed by the main effects in step 2 (cul ture of competitiveness, knowledge development, and market and the interaction and turbulence), in step 3. Specifically, moderators the following was in three hierar analyzed regression equation chical steps: Yx = a + ?xXx + ?2X2 + ?3X3 + ?4X4 + ?5X5 + ?6X4X5 + ?7X3X4 + &X3X5 where Yx = + s
= time performance (CT), a cycle = = firm age (AGE), X2 firm size intercept, Xi turbulence X3 = market (SIZE), (MT), of X4 = culture (CC), competitiveness = and X5 (KD), knowledge development s ? random disturbance terms. Consistent with the
ANALYSISAND RESULTS
Table 4 summarizes the results. Hypothesis testing was two via techniques: accomplished (1) hierarchical regression; and (2) a parsimonious latent-variable interaction (e.g., Ping, technique 1995) via LISREL
Copyright ?
literature on simultaneous testing of main effects each along with their interactions, we examined main effect (CC -> CT and KD -> CT) as the effect of a given predictor when the predictor it is at its mean interacts with (Aiken and West, 1991). As such, we discuss the main effect regres sion results of CC conditioned on the notion that KD is at its mean, and vice versa. Within the regression testing, market turbulence was created as a summated index. In addition,
8.80. This
& Sons, Ltd.
dual
testing
allows
2007 JohnWiley
1044
Table 4.
G. T. M. Huit,
Standardized results
D.
of
J. Ketchen
the hypothesis
Jr and M. Arrfelt
testing with cycle time as the criterion variable (n = 201)
Predictor
variables
Three-step hierarchical regression analysis
Three-step
latent-variable parsimonious interaction analysis
Findings
Constrained
Unconstrained
model
model
Step
1: Control
variables -0.06 -0.22 (to assess CMV) 0.06 F = 0.07 6.10" Saturated model -0.76 -2.91* -0.07 -0.23 -0.86 -2.91*** -0.04 -0.54 0.51 0.33 Saturated model -0.16 -2.25*** 7.02***
Firm age (AGE) Firm size (SIZE)

Same-source R2 factor
Model fit Step 2: Main effects Firm age (AGE) Firm size (SIZE) Market turbulence (MT) Culture of competitiveness (CC) Knowledge development (KD)
Same-source R2 factor (to assess CMV)
-0.04 -0.12 0.04 0.35 0.19 0.30 F =
-0.53 -1.69* 0.69 4.21*! 2.17*!
-0.03 -0.12 0.06 0.52 0.07
-0.36 -1.54 0.82 3.25*: 0.39
-0.02 -0.14 -0.11 0.41 -0.19 0.39 0.39
-0.29 -1.78* -1.44 2.14** -0.75 5.04**'
HI supported H2 partially supported
0.39 16.58*
Model
Step
fit
CFI = 0.?
-0.06 -0.07 0.08 0.58 0.11 0.25 0.23 -0.33 0.46 -0.89 -0.98 1.04 3.58*' 0.68 3.07*' 1.94* -2.83*'
CFI = 0.97
-0.05 -0.10 -0.12 0.46 -0.21 0.23 0.22 -0.32 0.45 0.46 -0.72 -1.33 -1.52 2.29** -0.79 2.85*** 1.81* -2.68*** 5.22***
3: Interactions -0.06 -0.09 0.06 -0.92 -1.33 0.91 4.59**= 2.54** 2.36** 2.01** -2.84**=
Firm age (AGE) Firm size (SIZE) Market turbulence (MT) Culture of competitiveness Knowledge CC x KD KD x MT CC x MT
Same-source R2
development
(CC) (KD)
0.39 0.22 0.15 0.17 -0.24
H3 supported H4 supported H5 supported
factor
(to assess
CMV) 0.34 F = 12.28*!
Model
fit
** p < 0.05; * p < 0.10.
CFI
0.98
CFI = 0.97
c* p < 0.01;
based on the higher-order analysis of the measures, empirical justification exists (in addition to the the oretical foundation; Huit et al, 2002; cf. Huit and Ketchen, 2001) to create a summated index of CC of learning, inno and entrepreneurial with vativeness, orientations, at one-third. Likewise, each construct weighted KD was assessed via an index composed of knowl shared information distribution, edge acquisition, and achieved memory weighted meaning, based on both empirical and theoretical 1991; Huit et al, 2004). (e.g., Huber, equally rationale based on the three dimensions
the VIFs were lower than 2.10, model, does not affect the indicating that multicollinearity or of the controls variables weights hypothesized (Mason and Perreault, 1991). To assess Hypotheses 1 and 2, we first examined the results in step 2. In regression when this step, both CC (p < 0.01) and KD (p < 0.05), entered along with market turbulence, are associated with time, significantly cycle providing 1 and Hypothe initial support for both Hypothesis sis 2. The
KD
Together with the regression results, we exam ined the variance inflation factors (VIF) to assess the likelihood results. that multicollinearity affects the In each of the three steps in the hierarchical
inclusion of market turbulence, CC, and in step 2 of the model explained significant in cycle time beyond variance that explained by the control variables in step 1 (p < 0.01), with the a total of R2 = 0.30. step 2 equation explaining The results of Hypotheses 1 and 2 are shown to the full model is specified in step 3
be robust when
Copyright ?
Strut. Mgmt.
/., 28: 1035-1052 (2007) DOI: 10.1002/smj
Strategie Supply Chain Management

main to assess Hypotheses the 3, 4, and 5. Specifically, effects of CC and KD as well as the hypoth esized interaction term of CC x KD (p < 0.05) and the moderators of KD x MT (p < 0.005) and CC x MT (p < 0.01) had significant associations the inclusion of main and interaction
1045
effects is
with cycle time. Age, size, and market turbulence had no direct association with cycle time. Given that the direct relationship between MT -> CT is the results indicate thatMT serves as insignificant, a pure moderator of the CC -? CT (negative) and the KD -> CT (positive) (Sharma, relationships Durand, and Gur-Arie, 1981). The inclusion of the terms (CC x KD, KD x interaction and moderator MT, and CC x MT) in step 3 explained significant variance beyond step 2 (AR2 = 0.04, p < 0.01). The fully specified model (i.e., including steps 1, 2, and 3) resulted in R2 = 0.34 (p < 0.01). Overall, all five hypotheses were supported in the hierar chical regression analysis. latent-variable interaction
(the results for each of the empirically meaningful three steps are included in Table 4). In the full three-step and constrained model, the results indicate that CC (p < 0.01) and the term CC x KD interaction hypothesized (p < 0.01) as well as the two moderators of KD x
MT (p < 0.10) and CC x MT (p < 0.01) had

relations with cycle time (R2 = 0.46; significant = 83.63, d.f. = 48, DELTA2, RNI, CFI, and X2 TLI all = 0.98, RMSEA = 0.06). Likewise, in the unconstrained model, the results indicate that CC (p < 0.05), the interaction term CC x KD (p < the two moderators of KD x MT (p < x MT (p < 0.01), and the 'same (p < 0.01) had a significant asso ciation with cycle time (R2 = 0.46; x2 = 120.20, d.f. = 53, DELTA2, RNI, CFI, and TLI all = 0.97, RMSEA = 0.08). The results for steps 1 and 2 are also included in Table 4 for completeness. 0.01), 0.10) and CC source' factor In comparing steps 2 and 3, using the method devised by McCallum and Mar (1995), the third and unconstrained step in both the constrained an models 7 percent of vari additional explained ance beyond that explained by earlier steps. These Hypotheses results verify that the strengths of the 1, 3,4, and 5 paths were consistent and across the hierarchical and regression
Parsimonious analysis
As a second step in testing the hypotheses, we used a parsimonious latent-variable interaction tech via LISREL 8.80. This devel nique technique, oped by Ping (1995, 1998), is a more parsimo nious estimation technique for latent interaction and quadratic variables than its predecessors by and Judd and (1984) (1987). Our Kenny Hayduk use of this technique to examine the hypothe ses adds to the hierarchical regression analysis in two ways. allows us to the main and in order to mine any the latent-variable technique errors for incorporate measurement interaction effects (Ping, 1995, 1998) assess whether such errors under First,
supported latent-variable interaction analyses. parsimonious 2 was supported in the hier However, Hypothesis archical regression analysis only. Finally, based on an anonymous referee's suggestion, we checked whether our model had greater explanatory than a simpler model wherein all first-order tors (i.e., LO, IO, EO, KA, ID, SM, and along with the controls (e.g., AGE, SIZE, and the moderators (e.g., CC x KD, KD x CC value fac AM) MT)
links within the significant results (Busemeyer and Jones, 1983). Second, we are able to incorporate a test of potential CMV issues at the hypothesis-testing level to determine whether CMV inflates or curtails the magnitude of the obtained et al, effects (e.g., Netermeyer et Podsakoff 2 contains 1997; al, 2003). Appendix details on this analysis. of the parsimonious latent-variable those the hierarchi mirror in analyses cal regression analysis, with the exception that KD interaction or the is not significant in either the unconstrained not 2 constrained models is (i.e., Hypothesis sup ported). Consistent with the hierarchical regression (1997) hierarchi analysis, we followed Ganzach's cal procedure to SEM testing to estimate whether
statistical
The
results
MT, x MT) were allowed to affect cycle time In this model the three mod test, directly. path erators of CC x KD (parameter estimate = 0.14, = 0.17, x MT KD (PE p < 0.01), p<0.05), = -0.24, < and CC x MT p (PE 0.01) along with EO (PE = 0.23, p < 0.01) and KA (PE = 0.15, p < 0.01) were the only significant variables = the higher-order model had 0.35). Overall, (R2 a greater explanatory power than the direct model = 0.46 vs. R2 = (R2 0.35), lending support to our of higher-order structures of at conceptualization least CC and potentially KD (at least in its mod erated format).
1046
G. T. M. Huit,
D.
J. Ketchen
Jr and M.
Arrfelt In terms of substantive issues, this conclusion offers implications for firms, especially those that are interested in gaining the benefits of strate but whose gic supply chain management, supply chains currently rate poorly in terms of CC and KD. Dramatic reorientations of supply chains are difficult to accomplish (Huit et al, 2002), and it is unlikely that the lack of both elements can be remedied quickly. Our results suggest that such firms might benefit by building a culture of com
DISCUSSION
Some caveats apply to our findings. We were to draw on objective indicators of per concerns about the formance due to informants'
unable
sensitivity of that information. Also, the inferences that can be drawn from the results are restricted by use of cross-sectional data drawn from key infor mants. Using multiple informants from each firm over likely enhance the robustness of in light of the fact that studies, especially data are often chain objective performance supply our not available. Also, study tapped into one firm a in supply chain, limiting our ability to fully cap our ture these variables for entire chains. Despite time would future the results offer important steps in about 'strategic supply chain knowledge in general and about why some sup management' limitations, building ply chains outperform others in particular. Below, we discuss the implications of our findings. Specif ically, we focus on (1) the main effects of CC 1 and 2) for cycle time and KD (i.e., Hypotheses (2) the interaction between CC and performance, role KD (Hypothesis 3), and (3) the moderating of market turbulence (Hypotheses 4 and 5). Man that the importance of the agers should recognize we is tied to offer below normative implications their firms compete based on the extent to which cycle time.
petitiveness
first, and then emphasizing knowledge once the cultural elements are estab development lished. This would ensure that the firm enjoys at least some cycle time reduction benefits as soon
as possible. Achieving such benefits is valuable of cycle time's links with because strategically (Handfield and profits and other firm-level metrics Nichols, 2002).
The
interaction
between
culture
of development
competitiveness Building
and knowledge
on two recent works on 'culture of com (Huit et al., 2002) and 'knowledge petitiveness' to (Huit et al., 2004) as vehicles development' cycle time performance the learning component the CC and KD Specifically, and theories in supply chains, that is at the core
improve we used of both elements.
to link these frameworks we drew on the resource
Culture
of competitiveness, knowledge and cycle time performance development, initial predictions examined the potential main effects relating culture of competitiveness (Hypoth esis 1) and knowledge (Hypothesis development Both analyses we 2) to cycle time performance. conducted
Our
from organizational learn a to information offer the and ing processing the concepts oretical delineation that integrates and predicted an interaction effect (Hypothesis 3). The results showed that the CC-KD interaction explains a significant amount of variance in cycle time above findings tiveness, sufficient goals Based we and beyond individual effects. These suggest that neither a culture of competi nor knowledge by itself is development, to achieve on by Huit the results the supply chain performance et al. (2002, 2004). 1 and 2, of Hypotheses
based view
supported the main effect for CC, but the hierarchical regression analysis supported only the main effect for KD. This set of findings can from both a technical perspective be addressed In and from a substantive, conceptual perspective. terms of technical issues, structural equation takes in mea into account the potential error variances it does not. Thus, surement, whereas regression may be that SEM simply offers a more precise 2 that its results for Hypothesis test, suggesting are the ones in which we should have confidence. to conclude that CC has As such, it is reasonable a direct link with cycle time (as shown in the tests of Hypothesis 1) but KD does not.
examined
above that firms that are launch suggested a ing strategic supply chain management approach should focus first on building CC and then pur sue KD initiatives. The results for Hypothesis 3 are once both foundational elements indicate that
in place, potential synergies between CC and KD can be exploited in order to gain additional cycle The results also inform firms time performance.
Strategie whose supply chains are currently good at knowl but that have not yet estab edge development, lished much of a culture across supply chain mem bers. These firms are likely to find that stressing development without a reinforcing cul knowledge On
Supply
Chain Management firms that focus
1047
on develop behaviors development ing appropriate knowledge (i.e., their reap lesser results in stable environments supply chains are not able to take full advantage of their superior learning skills) but are likely to reap the other hand,
ture will not significantly enhance cycle time. Such shared firms should focus attention on developing beliefs and values across their supply chains. We view learning as central to these processes. is the concept that links CC and KD. As Learning such, it may be the linchpin for firms seeking to exploit synergies between CC and KD. Learning in in a set of knowledge the CC context ismanifested 1999), while seeking values (Baker and Sinkula, in a set context is manifested in KD the learning behaviors of knowledge-producing (e.g., Huber, 1991). These values perform a dual role as the learning element of CC while also being the glue and entrepreneurial ori that centers innovativeness in the entations on supply chains' competitiveness
marketplace.
in turbulent markets (because greater advantages seeking and establish they are prone to knowledge ing the requisite variety needed to operate effec tively in turbulent market conditions). The results suggest that managers who are con fident in their sense about the level of market tur bulence they will face can use this sense to decide either CC or KD whether to emphasize developing in their supply chains. For those firms whose man agers are unlikely to be able to predict the degree that will be present in their mar of turbulence over time, learning efforts centered on ketplace suc both CC and KD are critical to sustained cess (cf. Lee, 2004). In fact, our results would tend to suggest that supply chains that develop strong elements of both a culture of competitive ness and knowledge development may be able to offset the effects that the environment has on their (at least with respect to market turbu operations turbulence lence; other aspects of environmental in future research to better need to be investigated the full potential value of the syner understand and effect of a culture of competitiveness gistic a From resource-based development). knowledge a supply chain's unique confluence of perspective, a culture of competitiveness devel and knowledge opment seems likely to provide the high level of that is required to establish a sustain inimitability able competitive advantage (cf. Barney, et al., 2004; Schroeder et al., 2002). 1991; Ray
the learning behaviors within the KD Similarly, and infor framework (i.e., knowledge acquisition also serve in a dual capac mation distribution) they are the main knowledge ity. Specifically, as well as the cultural builders activities producing that facilitates members of a 'common affiliation' stor at and the effective shared meanings arriving in achieved memory (cf. age of new knowledge Daft and Lengel, 1986; Gioia and Thomas, 1996). desire to acquire Thus, supply chain members' it to other mem then distribute and knowledge to achieve a bonding bers provides a mechanism that has been found to be critical in complex (Anderson, H?kansson, ply chain relationships Johanson, 1994). sup and
The moderating
role of market
turbulence
CONCLUSION
One of the central trends in today's economy is that less 'firm vs. firm' and is becoming competition more 'supply chain vs. supply chain.' Indeed, firms such as Dell andWal-Mart have, in essence, rewrit ten the rules of strategy and rivalry in their respec tive industries through using supply chains not just as a means for moving material, but also as a com theories, our petitive weapon. Drawing on multiple stream on research advances the emerging study by shedding strategic supply chain management new light on why some firms outperform others in terms of cycle time. Given that supply chain activi ties shape firms' profits, growth, market share, and
4 and 5 revealed that mar The tests of Hypotheses the effects of both CC ket turbulence moderates and KD, although one caveat is that the results involving KD were weakly supported (p < 0.10) in our SEM analysis. These findings are critical to what drives supply chain success. understanding on the one hand, the results indi For example, cate that a firm that devotes a great deal of effort on developing a supply chain culture focused on the market (i.e., a culture of competi satisfying to in is tiveness) reap positive advantages likely but will fall behind stable market environments, when market turbulence is strong (cf. Slone, 2004).
1048
G. T. M. Huit,
D.
J. Ketchen
Jr and M. Arrfelt
JB. 1986. Barney Organizational source of sustained competitive culture: advantage? and can it be Academy com 17(3): a
other key metrics (Handfield and Nichols, 2002), closing the gap between what we know and what we need to know about the determinants of cycle time across multiple firms is important. While past inquiry suggests competitiveness results highlight independent roles for a culture of our and knowledge development,
ofManagement
Barney JB. 1991. petitive 99-120. Barney
Review
Firm
11(3): 656-665.
resources sustained of Management
advantage.
Journal
the criticality of simultaneously these two concepts and market turbu considering lence in order to minimize cycle time.
TB. 2005. resource-based JB, Mackey Testing In Research in Strategy and theory. Methodology Vol. DD 2, Ketchen DJ, Management, (eds). Bergh Elsevier: New 1-13. York;
Bowersox DJ, Closs DJ, Stank TP. 1999. 21st Century Logistics: Making Supply Chain Integration a Reality. Council of Supply Chain Management Professionals:
Oak tive Brook, IL.
ACKNOWLEDGEMENTS
This research was supported by FedEx, the Insti tute of Supply Management (ISM), the Center for International Business Education at and Research State University and the (MSU-CIBER) Michigan Lowder Center for Family Business and Entre preneurship at Auburn University.
Busemeyer
JR, Jones LE. 1983. Analysis

rules when the causal
of multiplica
variables are
combination
measured with
549-562. Child J. 1972.
error. Psychological
structure, of strategic
Bulletin
93(3):
performance:
Organizational the role
and environment, choice. Sociology
6(1): 1-22. Cohen J, Cohen P, West SG, Aiken LS. 2003. Applied
Multiple Behavioral Regression/Correlation Sciences (3rd edn). and on individual Analysis for Erlbaum: Mahwah, information the NJ.
Corner PD, Kinicki AJ, Keats BW.
1994. Integrating
processing Science 5(3):
REFERENCES
Aiken LS, West SG. 1991. Multiple Regression: Testing
and CA. Aldrich HE. 1979. Organizations and Environments. Prentice-Hall: Amburgey past, Anderson ations Anderson modeling two-step 411-423. Anderson business context. JC, TL, present, Interpreting Interactions. Sage: Newbury Park,
organizational perspectives 294-308. Firm.
choice.
Organization
Cyert RM, March JG. 1963. A Behavioral Theory of the

Prentice-Hall: media New York.
Daft RL, Lengel RH.

requirements, NJ. Cliffs, Englewood Rao H. 1996. Organizational and future directions. Schroeder review. 133-158. 1988. review Structural and equation recommended 103(3): Dess RG. Journal
1986. Organizational
richness, and
information
design.
structural
ecology: Academy 1989. Oper of
Management Science 32(5): 554-571. WeickKE. 1984. Toward DaftRL,

organizations Management 1994. Day GS. systems. interpretation Review 284-295. 9(2): The capabilities of as
a model
Academy market-driven
of
of
Management
strategy: JC,
Journal 39(5): 1265-1286.

of Opera
JC, Cleveland GC, a literature 8(2): DW. Gerbing a in practice: approach.
organizations.
37-52. GG, zational Beard task
Journal of Marketing
DW. 1984. Dimensions
58: (October):
of organi Science
tions Management
environments.
Administrative
Psychological
Bulletin J. 1994.
H?kansson Johanson H, a within relationships Journal 58: of Marketing DA. Action 1978. Perspective. TS. 1977.
business (October):
Dyadic network 1-15.
Quarterly 29(1): 52-73. Dillman DA. 1978. Mail and Telephone Surveys: The Total Design Method. Wiley: New York.
Duncan RB. 1972. and Jackson links to Characteristics perceived SE. ronments Dutton of organizational envi environmental uncertainty. Categorizing action. strategic Academy of
C, Sch?n Argyris A Theory of MA. Reading, Armstrong sponse JS, bias
Organizational
Learning:
Administrative Science Quarterly 17(3): 313-327.

JE, issues: 1987. organizational DF. with 1981.
Addison-Wesley: nonre Estimating Journal of Marketing
O verton in mail
Management
Fornell C, equation measurement
Review 12(1): 76-90.

structural Evaluating and unobservable variables of Marketing Research 18:
surveys.
Larcker models
Research 14: (August): 396-402. Ashby WR. 1956. An Introduction to Cybernetics. Wiley:
New York.
error. Journal
Bagozzi RP,
testing 459-489.
Phillips LW.
1982.
Representing
a holistic
and
organizational
theories:
construal.
Administrative Baker WE,

of market
Science Quarterly 1999. The

and learning Journal
(February): 39-50. Ganzach Y. 1997. Misleading interaction and curvilinear terms. Psychological Methods 2(3): 235-247.
Anderson JC. 1992. Monte DW, Gerbing uations of fit of indices goodness models. Methods equation Sociological Carlo for and eval structural Research
27:
(September): effect
on
Sinkula JM.
orientation
synergistic
orientation of
organizational
performance.
the Academy
ofMarketing Science 27: (Fall): All-All.

21: (November): 132-160. Gioia DA, Thomas JB. 1996. Identity, image, and issue
interpretation: sensemaking during strategic change Strut. Mgmt. J., 28: 1035-1052 (2007) DOI: 10.1002/smj
Strategie
in academia. 370-403. Administrative Science Quarterly 41(3):
Supply
Chain Management
1049
J?reskog KG, S?rbom D, Du Toit S, Du Toit M. 2000. LISREL 8: New Statistical Features (2nd edn).
Scientific Software International: 1984. of D, Judd CM. Kenny effects interactive IL. Chicago, the nonlinear and Estimating latent variables. Psychological
Godfrey PC, Hill CWL.

servables Management Grant RM. of the in strategic Journal 1996.
1995. The problem of unob

research. management 519-533. 16(7): a knowledge-based Journal, Strategic
Toward
firm.
Strategic
Management
theory Winter
Bulletin 96: (July): 201-210. Kohli AK, Jaworski BJ, Kumar A.

measure Research Lawrence ronment: Harvard of market 15: orientation.
1993. MARK?R:
Journal of Marketing and Envi Integration. Harvard
Special Issue 17: 109-122. Handfield RB, Nichols EL. 2002. Supply Chain Redesign:
Value Sys into Integrated Design Transforming tems. Financial Saddle Times-Prentice-Hall: Upper Chains River, LISREL: Press: NJ.
PR,
467-477. (November): JW. 1967. Organization Lorsch and Managing Differentiation Press: triple-a Cambridge, supply MA. chain.
Hayduk LA.
1987. Structural Equation Modeling

Essentials and MD. Advances. Johns
with
Lee
HL.
University The 2004.
Hopkins
Baltimore,
Business Review 83: (October): 102-112. Levinthal D, March JG. 1981. A model of adaptive
organizational Levitt B, March search. Journal of Economic Behavior
learn and unlearn. B. 1981. How Hedberg organizations In Handbook PC, Design, Nystrom of Organizational
and Organization
JG.
2(4): 307-333.
1988. Organizational learning. power, analy (August):
Starbuck WH
York; chain 3-27.
(eds). Oxford University
Press: New
Annual Review of Sociology

Mason CH, and
14: 319-340.
Hendricks KB, Singhai VR. 2003. The effect of supply

glitches on shareholder 21(5): and wealth. 501-522. Journal of Operations Management
Hitt M,
Boyd B, Li D.
2004. The
a vision in Strategy
state of
of the
strategic
future. In
WD Perreault Jr. 1991. Collinearity, of multiple regression interpretation Research 28: sis. Journal of Marketing 268-280.
research management Research Methodology
McCallum R, Mar CM.

moderator and quadratic 1982. C, Miner memory
1995. Distinguishing
effects to in multiple environmental of
between
regression.
and Management,
Vol.
York;
1, Ketchen DJ, Bergh D

1-31.
(eds). Elsevier: New
Psychological
Meyer Moorman tional AD.
Bulletin 118(3): 405-421.

Adapting AS. on new 1997. jolts. organiza and cre
Hu L, Bender PM. 1999. Cutoff criteria for fit indexes

in covariance versus new structure alternatives. analysis: Structural conventional Equation criteria Modeling contribut Science
Administrative Science Quarterly 27(4): 515-536.

The impact performance product
6(1): 1-55.
Huber GP. 1991. ing processes the learning: Organizational and the literatures. Organization DJ. 1985.
ativity. Journal ofMarketing Research 34: (February):

91-106. Naman JL, Sie vin DP. of fit: a model Journal Boles 1993. and Entrepreneurship tests. empirical and Strategic R. of the
2(1): 88-115.
Huber GP, Power Retrospective guidelines Management for strategic-level ing accuracy. 171-180. Huit GTM. sourcing learning Huit GTM, matter? advantage A 1998. managers: Strategic reports increas of
concept Management Netemeyer 1997.
RG, An
137-154. 14(2): DO, JS, McKee into the behaviors
McMurrian antecedents in a
Journal
investigation citizenship
6(2):
organizational
personal
the international strategic Managing as a market-driven function organizational 193-216. Decision Sciences 29(1): system. Ketchen test and of DJ. the 2001. Does market between Strategic orientation positional
selling
85-98. Nichols EL
context. Journal
of Marketing
D, Frolick cycle
61:
MN. time
(July):
1996. in an
Reducing 13-28. Nystr?m tional PC,
Jr, Retzlaff-Roberts order fulfillment
international supply chain. Cycle Time Research 2(1):

Starbuck unlearn. WH. 1984. To avoid organiza 13: for of
relationship
performance.
Management
Journal 22(9): 899-906. Huit GTM, Ketchen DJ, Nichols EL

examination fulfillment of a culture cycle time within
Jr. 2002.
and
An
order Ping
crisis, 1995. A
Organizational
Dynamics
of competitiveness chains. supply
(Spring): 53-65.
RA. parsimonious quadratic interaction and estimating technique latent variables. Journal
Academy
ofManagement Journal 45(3): 577-586. Huit GTM, Ketchen DJ, Slater SF. 2004.
processing, supply RF, chain knowledge performance. GTM. 1998. development, Academy
Information
and strategic
Marketing Research 32(3): 336-347. Ping RA. 1998. EQS and LISREL examples
data. survey in Structural In Interactions and Nonlinear Schumacker Equation Modeling,
using
Effects RE,
of Management market integration orien and
Journal 47(2): 241-253.

Hurley Huit and tation, 42-54. organizational Innovation, an learning:
Marcoulides G A
63-100. Podsakoff zational PM, Organ research:
(eds).
DW. problems
Erlbaum:
1986. and
Mahwah,
NJ;
Self-reports prospects.
empirical examination. Journal of Marketing Jaworski BJ,

antecedents
62(3):
in organi Journal of
Kohli AK.
and
1993. Market
Journal
orientation:
of Marketing
Management 12(4): 531-544. Podsakoff PM, McKenzie SB, Lee J-Y, Podsakoff NP.
2003. Common review method of the biases in behavioral and research: a critical literature recommended
consequences.
52: (July): 53-70.

1050
G. T. M. Huit, Journal
D.
J. Ketchen Psychology
Jr and M. Arrfelt 88(5): undertaken in the supply chain, with a focus on the 'inbound' portion of the chain (i.e., as appli cable to their organization's supply management 'We' and 'participants' refer to the users, activity). in the order buyers, and suppliers that participate fulfillment process. 'Products' refer to both tangi ble (physical goods) and intangible products (ser 'a' at the end of the ques vices). An item with tion was deleted after the measurement purification scales were used process. Seven-point Likert-type for all items ranging from 'strongly disagree' to
'strongly Strategic agree.'
remedies.
879-903. Ray
of Applied
WA. 2004. G, JB, Muhanna Barney Capabilities, business and processes, competitive advantage: the dependent tests variable in empirical choosing of the resource-based view. Management Strategic
Journal 25(1): 23-37.

Rumelt issues RP, Schendel in strategy. DJ DJ. DE, Teece In Fundamental 1994. Issues Fundamental in Strategy:
Research
Agenda,
Rumelt RP,
Schendel DE,
School Press: resource and the
Teece Boston, Schroeder based
Harvard Business (eds). 9-47. MA; Junttila MA. RG, Bates KA, view of manufacturing to manufacturing
2002. strategy
relationship
processes.
Management
Sharfman
Journal 23(2): 105-117. Learning orientation (Huit, 1998)
Dean JW Jr. 1991. and MP, Measuring a multi-dimensional the environment: conceptualizing Journal 681-700. 17(4): approach. of Management Sharma O. 1981. Identification S, Durand RM, Gur-Arie and of moderator variables. Journal analysis of
Marketing Research 18(3): 291-300. Shook CL, Ketchen DJ, Huit GTM, Kacmar KM. 2004.
An assessment of the use of structural equation Strategic and the in strategic management research. modeling Journal 397-404. 25(4): Management Slater JC. orientation 1995. Market SF, Narver
We agree that our ability to learn is the key to improvement in the supply management pro
cess.
learning organization.
63-74. Slone RE. 2004. Harvard Starbuck ments. WH. Leading Business Review 1976.
Journal of Marketing
a chain supply 83: (October):
59(3):
The basic values of this supply management pro cess include learning as a key to improvement. Once we quit learning in the supply management process we endanger our future/ The sense around there is that employee is an investment not an expense. learning
turnaround. 114-121.
and their environ Organizations In Handbook and Organiza of Industrial
tional Psychology, Dunette MD Chicago, IL; 345-354.

Summer Snow C, Bettis of C, the field R, Zeithaml business Duhaime C. 1990. policy DA.
(ed). Rand McNally:

J, Hambrick education Journal sense linkages perfor 36(2): Social D, in of
Innovativeness
orientation
(Hurley
and Huit,
1998)
Technical is readily We
ideas.
I, Grant Doctoral and
strategy. 1993.
Management
Thomas making among mance. 239-270. Thompson Science JD. Bases
16(6): 361-398.
Strategic performance: and action, Journal in Action: Theory.
JB, Clark SM, Gioia and organizational scanning, Academy 1967.
based on research results, in accepted supply management. actively seek innovative supply management innovation, is readily
process.
interpretation, of Management Organizations of Administrative
Innovation
management
accepted for new
in the
supply
McGraw
People
not
are not penalized
ideas that do process is
Hill: New York. Weick K. 1987. Substitutes for strategy. In The Teece DJ (ed). Ballinger: Competitive Challenge,
221-234. MA; Cambridge, Weiss JB. 1993. AM, Heide The nature of organizational
work.a
Innovation
encouraged.
in our supply management
in high technology markets. Marketing Research 30(2): 220-233.

Wernerfeit Strategic B. 1984. A resource-based Journal 5(2): view Management
search
Journal
of
of
Entrepreneurial
orientation
(Naman
and
Slevin, 1993)
the firm.
171-180.
We
emphasize
research
and development
and
APPENDIX
1:MEASUREMENT
SCALES
technological leadership/ We initiate actions to which

respond.
other organizations
The respondents were asked to relate their answers to the last order fulfillment process that they had
We
fast to introduce new administrative techniques and operating technologies.

are
Strategie a strong proclivity for high-risk have projects. are bold in our efforts to maximize We the probability of exploiting opportunities. We
Supply
Chain Management
1051
We develop a shared understanding of the impli cations of a supply management activity. Achieved and Miner,
memory
(Moorman
1997)
Knowledge acquisition 1993) Kumar,
(Kohli,
Jaworski,
and
We meet regularly to find out what products we need in the future. We do a lot of in-house research on products we
may need.
We We
are fast
to detect
changes once a year
in our product to assess the
a great deal of knowledge about the process. supply management We have a great deal of experience with the process. supply management We have a great deal of familiarity with the process. supply management a great deal of research We have invested in the supply management and development We have process/ Market turbulence and Kohli,
preferences.
poll participants services.a quality of our supply management We are fast to detect fundamental shifts in the environment. supply management We review the likely effect of periodically the in management changes supply environment/
(Jaworski
1993)
Information Kumar,
distribution
(Kohli,
Jaworski,
and
1993)
customers' product a over time. bit preferences change quite Our customers tend to look for new products all the time. We have demand for our products from cus tomers who never bought them before. customers New have product needs that are different from our existing customers. We continuously cater to many new customers.
In our kind
of business,
We frequently have interdepartmental meetings to discuss trends in supply management/ We future supply man spend time discussing
agement needs.
and (Huit, Ketchen, Cycle time performance and Slater, 2004) Nichols, 2002; Huit, Ketchen, process is length of the supply management shorter time. every getting We have seen an improvement in the cycle time of the supply management process recently/ are satisfied with We the speediness of the process. supply management the participants in decision making Involving shortens the supply management process. Based on our knowledge of the supply manage ment process, we think it is short and efficient. The length of the supply management process could not be much shorter than today/ The
We immediately know when something tant happens in the supply management

cess.
impor pro
We share data on participant in satisfaction on a the supply management process regular

basis.a
We alert participants when impor something tant happens in the supply management pro
cess.
Shared meaning
(Huit, Ketchen,
and Slater,
2004)
We share supply management information effec the between tively supply management partici
APPENDIX
2: DETAILS ON COMMON
pants.
METHOD VARIANCEANALYSIS
We assessed the potential problem of common method variance inhibiting the analyses at both the measurement and hypothesis-testing levels.
Strut. Mgmt. J., 28: 1035-1052 (2007) DOP. 10.1002/smj
share supply management information effec in the process. tively supply management We develop a shared understanding of the avail able supply management information.
We
1052
G. T. M. Huit,
D.
J. Ketchen
Jr and M. Arrfelt factor. Regarding gle indicant for the same-source the latter, we opted to include a single indicant all the purified (a summated score incorporating indicators) for the same-source factor instead of all applicable factor loadings because of the complex ity of fitting three interaction terms and two higher order factors (CC and KD). This approach also to the level of analysis of the other corresponds indictors. Thus, the following overall equation was tested (i.e., excluding the depiction of the rela tionships of the first-order indicators of CC and
Measurement-level
analysis
To examine common method variance at the mea surement level, we employed Harman's one-factor test within a confirmatory factor analysis setting. If CMV poses a serious threat, a single latent account for all manifest variables worse A fit for the (Podsakoff and Organ, 1986). one-factor model provides support that CMV does not pose a serious threat. The one-factor model resulted in a x2 = 2543.16 with d.f. = 527 (vs. a = 986.92 and d.f. = 491 for the measurement X2 model). Thus, CMV does not appear to be a prob lem at the measurement level. factor would
KD):
m = YnXi + yuX2 + yi3X3 + yl4X4 + yl5X5 rx
Hypothesis-level To examine
analysis
+ yl6X4X5 + yl7X3X4 + ylsX3X5 + yl9X6 +
common method at the variance we used the method described level, hypothesis et al. (1997) and Podsakoff et al. by Netemeyer within the latent-variable (2003) parsimonious interaction 1995). Specif (e.g., Ping, technique to the hypothesized in addition model ically, in Table 4), we tested (labeled 'constrained model' a latent-variable interaction model using structural via LISREL 8.80 that includes equation modeling a 'same-source factor' to the indicators of all con structs. In this analysis, we included the summated
= = where r?x (CT), Xx cycle time performance firm age (AGE), X2 = firm size (SIZE), X3 = = culture of com market turbulence (MT), X4 = (CC), X5 petitiveness knowledge development = 'same-source' and X6 (KD), factor, = disturbance term. A comparison was made of f i the unconstrained model (in which the same-source factor is estimated freely) with a constrained model the same-source factor loading is set to (in which and uncon zero). The results of the constrained are reported in Table 4. While strained models the
'same-source' factor was significant in the uncon
first-order indicators of culture of competitiveness as direct, reflective and knowledge development indicators of those higher-order constructs in con with indicants for each of the latent junction single interaction and moderator terms (CC x KD, KD x MT, and CC x MT), a summated indicator for market turbulence, the two controls, and a sin
strained model, CMV did not have a significant effect on the magnitude of hypothesized relation ships. Thus, CMV does not inhibit the analysis of the hypothesized relationships (e.g., Netemeyer et ai, 1997).
Copyright ?
2007 John Wiley
& Sons, Ltd.

Trategicsupply Chainmanagement: Improvingperformancethrougha Culture of Competitivenessand Knowledge Development

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Trategicsupply Chainmanagement: Improvingperformancethrougha Culture of Competitivenessand Knowledge Development

Uploaded by

Copyright:

Available Formats

Strategic Supply Chain Management: Improving Performance through a Culture of Competitiveness and Knowledge Development Author(s): G. Tomas M.

Strat. Mgmt. (www.interscience.wiley.com) 2005; Final revision

STRATEGICSUPPLY CHAINMANAGEMENT: IMPROVING PERFORMANCETHROUGHA CULTURE OF COMPETITIVENESSAND KNOWLEDGE DEVELOPMENT

ensuring success. Copyright ? 2007 JohnWiley & Sons, Ltd.

Copyright ? 2007 JohnWiley & Sons, Ltd.

or aligning of positioning ties can significantly impact

operations capabili competitive strength of an organization.' In

This paper is devoted as a next logical step

to taking what we view in the emerging stream

2007 John Wiley & Sons, Ltd.

Strategie Supply Chain Management

Entrepreneurial Orientation Interaction

Cycle Time Performance ?.

Firm Size Knowledge Development

has 1: Culture of competitiveness with cycle time perfor association

development with cycle

The moderating Starbuck's (1976)

role of market review

the research in some resulted

Strat. Mgmt. J., 28: 1035-1052 (2007) DOI: 10.1002/smj

Chain Management and RMSEA = 0.07

all being 0.96, 986.92, d.f. = 491).

the underlying 1, we next con of these con

Fit of the measurement The model indices fit was

model evaluated using a series of and Anderson by Gerbing

(1992) andHu and Bentler (1999)?the DELTA2,

deviations, S.D. 0.98 1.13 1.27 1.15

correlations, LO ? IN 0.30 0.55 0.33 0.34 ?

Mean 5.88 5.33 4.43

KA 0.12 0.19 0.29

ID 0.12 0.24 0.25

SM 0.14 0.30 0.31

AM 0.17 0.22 0.20

MT 0.00 0.00 0.01

CT 0.11 0.19 0.22

AGE 0.00 0.01 0.02

SIZE 0.00 0.04 0.04

orientation (EO) Knowledge

0.35 0.37 0.41 0.04 0.33

0.49 0.55 0.47 0.06 0.44

0.50 0.56 0.45 0.12 0.47

0.70 0.61 0.42 0.29 0.42

? 0.68 0.49 0.22 0.40

0.46 ? 0.48 0.11 0.37

0.24 0.23 ? 0.17 0.34

0.05 0.01 0.03 ? 0.11

0.04 0.14 0.12 0.01 ?

0.00 0.00 0.01 0.00 0.02

0.06 0.06 0.02 0.01 0.06

at the p < 0.05

the confirmatory variance

Range of loadings 0.47 0.81 0.70 0.61 0.64 0.83

meaning memory turbulence

Cycle time = 986.92 X2

62.0% 69.8% 60.5% 50.5% 59.3% 74.0% 76.0% 56.2% 45.0%

to 0.93 to 0.89 to 0.83 to 0.79 to 0.89 to 0.89

(Anderson and Gerbing, extracted was calculated

error for scale from 0.74

ranged ranged from 0.36 variances

index, and residual

Strategie Discriminant validity

for the involved 1981). Shared

for a robust assessment

-0.53 -1.69* 0.69 4.21! 2.17!

-0.29 -1.78* -1.44 2.14 -0.75 5.04'