Supply Chain Capabilities, Risks, and ResilienceI

Xavier Brusseta, Christoph Tellerb

aUniversité de Toulouse, Toulouse Business School, Toulouse, FrancebSurrey Business School, University of Surrey, Guildford, UK

Abstract

Supply chain resilience is an operational capability that enables a disrupted or broken supply chain toreconstruct itself and be stronger than before. This paper examines resilience using the dynamic capabilitiesapproach, grounded in the Resource-Based View of firms. The purpose of this research is to provide insightsfor achieving resilience by mapping the relationships between the practices, resources, and processes overwhich a manager has control. A survey of 171 managers is used to test a conceptual model that proposesrelationship between supply chain capabilities and resilience as well as the moderating role of supply chainrisks. Variance-based structural equation modeling reveals that only tighter integration between echelonsand increasing flexibility lead to added resilience. The perception of supplier risk helps motivate the supplychain manager to enhance integration capabilities and thus achieve higher resilience. Furthermore, theperception of external risks to a supply chain actually reduces the effort of deploying external capabilities toobtain resilience. Overall, the findings strongly support the view that resources, routines, and capabilitiesprovide different results in terms of resilience depending upon supply chain risk factors.

Keywords: Resilience, Supply Chain, dynamic capability, survey

1. Introduction

Supply chain risk management remains a keymanagerial challenge that affects the performanceof organizations (Altay and Ramirez, 2010). Char-acteristics such as tighter collaboration, increasedcomplexity, reduced inventory levels, and ever-wider geographic dispersion have created greatervulnerabilities (Bode et al., 2011). Supply chainmanagement literature is now beginning to explore

IThis paper has been written using material collectedduring the “Baromètre Supply Chain” project led byCapGemini Consulting and in association with the ÉcoleCentrale de Paris and SupplyChainMagazine.fr. We grace-fully acknowledge their support.

Email addresses: x.brusset@tbs-education.fr(Xavier Brusset), c.teller@surrey.ac.uk (ChristophTeller)

how best to build resilience in supply chains, withincreasing attention especially toward value chainfragmentation and geographical extension (Gulatiet al., 2000). All economic disruptions, whethernatural or man-made, carry unforeseen threats tothe performance and profitability of supply net-works (Hindle, 2008; The Economist, 2009).

In sociology and ecology, resilience characterizesan organization or a social body that is able torebuild itself after having been substantially af-fected by an exogenous attack (Berkes et al., 2003).One example from the United States is that ofWalmart’s operations before and after the pas-sage of hurricane Sandy in 2012 (Creighton et al.,2014). Resilience, as defined by Brandon-Joneset al. (2014), page 55, and Christopher and Peck(2004), page 4, is “the ability of a supply chain to

Preprint submitted to International Journal of Production Economics September 13, 2016

return to normal operating performance, within anacceptable period of time, after being disturbed”.

Given the literature and interest in resilience, it issurprising that the management practices requiredto achieve it are approached from so many differentmanagerial viewpoints (e.g., operations, strategy,information systems, marketing, human resources)as exemplified in Li et al. (2008, 2009). In thispaper, we consider the supply chain managers asthe central decision-makers and organizers of themanagement processes within the supply chain. Assuch,they organize, deploy and control all the nec-essary investments, assets, resources, routines, pro-cesses, and systems to achieve the strategic goal ofenabling the supply chain to be resilient. Groundedin the Resource-Based View (RBV) (Wernerfelt,1984), the dynamic capabilities theoretical frame-work introduced in Teece and Pisano (1994) yieldspowerful results which can be brought to bear inthe present setting. Teece (2007) defines a dynamiccapability as the ability to dynamically integrate,build, and reconfigure lower-order competences toachieve congruence with changing business environ-ments.

We apply this framework here to answer the fol-lowing questions: (1) Given the risks being faced,what practices does a supply chain manager deployto obtain such resilience? (2) How do environmen-tal factors related to supply chains influence the ef-fectiveness of these practices? With this dual focus,our research contributes to theory as well as prac-tice by increasing understanding of how to enhanceresilience and providing insights to determine thesupply chain capabilities required to achieve greaterresilience.

We next present a review of the literature and thetheoretical underpinning of our paper, from whichwe derive a conceptual model. Next, we elaborateon the study’s methodology, analytical approach,and the results of our empirical study. The con-cluding sections discuss the theoretical and prac-

tical implications of the findings, highlight limita-tions, and outline directions for future research.

2. Theoretical background and conceptualmodel

The dynamic capabilities approach has gainedwide acceptance as a tool to explain performanceacross competing firms (Barreto, 2010; Teece et al.,1997). According to this perspective, superior per-formance stems from two types of organizationalcapability, namely, dynamic capability and opera-tional capability (Cepeda and Vera, 2007; Helfatand Peteraf, 2003).

The literature has formulated the basic differencebetween dynamic capability and operational capa-bility (Teece, 2007; Winter, 2003). Dynamic capa-bilities are a learned pattern of collective activityand strategic routines through which an organiza-tion can generate and modify operating practicesto achieve a new resource configuration and achieveand sustain a competitive advantage (Teece et al.,1997; Teece, 2007). Barreto (2010) recommendsthat research should focus on the factors that mayhelp (or hinder) firms to achieve the potential rep-resented by their dynamic capabilities. It is impor-tant to recognise that the value of dynamic capabil-ities is context dependent (Wilden et al., 2013) andnot a set recipe or formula for general effectiveness.Organizational response to environmental turbu-lence is faster as well as more effective (Chmielewskiand Paladino, 2007) so ultimately enhances perfor-mance. Attaining competitive advantages requiresefficient and effective sharing and deployment of re-sources between partnering organizations and sup-ply chain partners (Rajaguru and Matanda, 2013).

By contrast, Winter (2003) argues that an oper-ational capability provides the means by which afirm functions or operates to make a living in thepresent. Dynamic capabilities are considered to beof a higher order than operational capabilities, as

2

their role is to contribute to the firm’s higher rel-ative performance over time (Drnevich and Kriau-ciunas, 2011). An operational capability refers toa firm’s ability to execute and coordinate the vari-ous tasks required to perform operational activities,such as distribution logistics and operations plan-ning, which are processes and routines rooted inknowledge (Cepeda and Vera, 2007). This capabil-ity refers to a high-level routine or a collection ofroutines (named organizational routines or compe-tences in Teece et al., 1997) that can be used torespond to unforeseen events affecting the abilityof a supply chain to perform (Barreto, 2010; Eisen-hardt and Martin, 2000). For example, given theincreasing importance of timely and cost-effectiveproduct delivery, supply chain resilience is consid-ered a critical capability to maintain the continuityof operations. Resilience as an operational capabil-ity requires both internal processes as well as thoserelative to the information flow, coordination, andcollaboration with upstream and downstream part-ners.

To build and operate a resilient supply chain, itis helpful to have an in-depth understanding of thelower-order capabilities (or micro-foundations, asdescribed in Teece, 2007) that are required.

Managerial systems, procedures and processesthat undergird each class of capability define whatTeece (2007) call organizational routines or com-petences and what Barreto (2010) sees as a re-quirement for supply chain operations. Henceforthnamed here lower-order capabilities, they are dis-tinct from the capability itself (Teece, 2007). Theselower-order capabilities along the supply chain con-stitute the practices among the different chainmembers using which the supply chain is able toabsorb or recover from disturbances, and still main-tain its ability to deliver value to final customers(Bhamra et al., 2011).

The dynamic capabilities approach makes it pos-sible to characterize the operational capabilities

Organizational

Informational

Relational

Human

Operational

capabilities

Dynamic

capabilitiesLower.order

capabilities

Resources

Firm

Firm

Firm

Sup

ply

.Ch

ain

Resilience

...

External.practices

Integration

Internal.processes

Figure 1: Theoretical model of the Resource-Based-View ofa supply chain

that supply chain managers wish to enhance as wellas the routines, procedures, and processes they ap-ply at their firms and across their supply chains (seeFigure 1). We now switch focus to the relationshipbetween those lower-order capabilities and the op-erational capabilities that characterize resilience.

2.1. Resilience in supply chains

An important aspect for all supply chain man-agers is the capacity of their supply chain to with-stand upheavals, disruptions and unforeseen events(e.g., Brandon-Jones et al., 2014; Bhamra et al.,2011). A supply chain that is still able to performand deliver products and services under such cir-cumstances is characterized as resilient (Blackhurstet al., 2011). This capacity is defined in Fiksel(2006) and in Pettit et al. (2010) as “the capac-ity for an enterprise to survive, adapt, and growin the face of turbulent change”. Resilience hasbroader implications than supply chain risk control.Since supply chains have increased in both lengthand complexity (Blackhurst et al., 2005), naturalcatastrophes, wars, strikes and economic upheavalsseverely impact performance (Chopra and Sodhi,2004; Wagner and Bode, 2008). Hendricks (2005)states that it is critical for firms to enhance the re-siliency (sic) in their supply chains and call for re-search into specific tactics that help firms developsuch capabilities.

Studies are concerned with the ability of the sup-ply chain to return to its original state of opera-tion after being disturbed (Pettit et al., 2010). To-

3

day’s supply chains are more prone to disruptionscaused by natural and man-made events (Wagnerand Bode, 2008). Hence, the ability to recoverquickly has become a topic of concern for practi-tioners and academics.Having described resilience as an operational ca-

pability, we now characterize the lower-order ca-pabilities available to the supply chain manager,which, when deployed across all members of thesupply chain, should generate this capability. Wethen stipulate the corresponding hypotheses.

2.2. Lower-order capabilities and hypothesis devel-opment

We define lower-order capabilities as the set ofphysical, financial, human, technological, and orga-nizational resources (Grant, 1991) coordinated byorganizational routines (Nelson and Winter, 1982)and deployed in an organization and across organi-zations.The literature provides abundant descriptions of

the practical managerial routines and processes de-ployed by a large number of supply chain managers.For the purpose of this study, we centered our at-tention on the organizational, informational, andhuman resources across organizations. Even thoughthe relevant literature often does not mention theRBV, the resources mentioned clearly belong tothe class of lower-order capabilities that we iden-tified above. For example, a fully deployed second-generation material requirements planning (MRPII) system is composed mostly of procedures, infor-mation systems, and skilled operators, as well astangible assets (computers, servers and wide areanetworks). This MRP has to be connected to otherfirms in the supply chain to exchange forecasts, de-livery schedules, and other planning requirements(Akkermans et al., 2003). This lower-order capa-bility, when combined with other similar practices,will contribute to a higher-order operational capa-bility (Su and Yang, 2010). The practices that we

consider as having influence on operational capabil-ities can be grouped as external, integration, andflexibility capabilities, which are described below.

External capabilities: These are the practicesthat in sum represent Efficient Customer Responsepolicies (Skjoett-Larsen et al., 2003). Partnershave to collaborate through systems such as VendorManaged Inventory (VMI) and Collaborative, Plan-ning, Forecasting, and Replenishment (CPFR) withretailers to enhance close cooperation among au-tonomous organizations engaged in joint efforts toeffectively meet end-customer needs (Faisal et al.,2007).

The flow of accurate and real-time informationin the supply chain is considered by many to beas important as the flow of goods. Informationsharing can also provide flexibility and improve theresponsiveness of the supply chain (Gosain et al.,2005; Agarwal et al., 2006). The information sharedmay include: end-customer demand, sales fore-casts, order status, inventory levels, capacity avail-ability, lead times, and quality. Sharing informa-tion can improve transparency, avoid lost sales,speed up payment cycles, create trust, avoid over-production, and reduce inventories (for reviews, seeBhamra et al., 2011; Sahin and Robinson, 2002).Current inter-organizational information systems(IOIS) facilitate the sharing of real-time informa-tion in the supply chain and allow organizationsto be more effectively coordinated throughout thenetwork. These systems are named Advanced Plan-ning and Scheduling (APS), Collaborative Plan-ning, Forecasting and Replenishment (CPFR), andEfficient Customer Response. IOIS also have impli-cations for the way that supply chains are designedand managed. One important example is the useof vendor managed inventory (VMI) systems wherean upstream supplier is able to react directly to theinventory and demand information from a down-stream customer by adjusting the quantity and tim-ing of deliveries (Kotzab, 1999). These practices en-

4

able a supply chain to be reconfigured when facedwith unexpected and disrupting events. As char-acterized in Faisal et al. (2006) and Faisal et al.(2007), supply chains are affected by “informationrisks”.Hendricks (2005) empirically documents how

glitches in supply chains affect operating perfor-mance, naming sources of glitches that run thegamut from parts shortages to reorganizational de-lays and information technology (IT) problems.Klibi et al. (2010) specifies how resilience in sup-ply chains should be assessed in view of the disrup-tions being faced. The solutions they propose toenhance resilience require tighter integration of sup-pliers and distribution networks as well as buildingredundancy and flexibility. Mandal (2012) specifi-cally identifies IT (an important component of ex-ternal capabilities) as one of the sources for in-creased resilience in a supply chain. The resilienceprovided protects the supply chain against the va-garies of the market. Our first hypothesis is thefollowing.

Hypothesis 1. There is a positive relationship be-tween the implementation rate of external capabil-ities (ξ1) and the level of resilience (η1) in supplychains.

Integration capabilities: Supply chain integra-tion has been defined “as the degree to whicha manufacturer strategically collaborates with itssupply chain partners and collaboratively managesintra- and inter-organization processes. The goal isto achieve effective and efficient flows of productsand services, information, money and decisions, toprovide maximum value to the customer at lowcost and high speed” (Naylor et al., 1999; Frohlichand Westbrook, 2001; Flynn et al., 2010). Eventhough the integration of manufacturers and clientshas been studied in the context of China throughthe prism of power relationship commitment theory(Zhao et al., 2008), other literature such as Pagell(2004); Lin et al. (2006); Faisal et al. (2007); Ra-

jaguru and Matanda (2013) view this capability asconsisting of IT systems and practices that employboth information systems and the correspondingmanagerial practices and routines to enhance inter-organizational integration and coordination. Suchintegration of IT with supply chain processes en-hances collaboration in the chain through continu-ous adjustments to the product lineup and invento-ries as well as sharing forecasts, sales data and in-ventory levels (Qrunfleh and Tarafdar, 2013, for anappreciation of the impact of Information Systemson supply chain performance). Collaborative plat-forms provide the possibility of exchanging informa-tion in real time (Boyson et al., 2003). The tech-nologies that enable goods to be tracked and tracedprovide greater control over operations within thechain as well as timely notification and access todetailed information when events occur. This alsocontributes to suppliers’ integration, thus increas-ing efficiency (Danese and Romano, 2011), espe-cially as service levels can be monitored (García-Dastugue and Lambert, 2003).

Integration provides the capability to reduce thecosts and risks of coordination and of transactionsby providing managers an opportunity to under-stand the focal areas that need attention. Hence,they can minimize risks to real-time and free flowof information (Faisal et al., 2007). Mandal (2012)identified the dimensions or antecedents that ITprofessionals perceive as important for achieving re-silience in the Indian context.

Hypothesis 2. There is a positive relationship be-tween the implementation rate of integration capa-bilities (ξ2) and the level of resilience(η1) in supplychains.

Flexibility capability: This last set of practicesincreases the responsiveness of a supply chain tostimuli from end-consumers. It refers to the abil-ity to evaluate and take needs into account quickly(Charles et al., 2011). The forecasting and planningprocesses within the supply chain are scaled up, re-

5

sulting in enhancement of the supply chain’s reac-tive capabilities by enabling it to predict final de-mand changes and adapt to them both in upstreamand downstream operations (Olhager, 2013). Suchpractices, jointly named Sales and Operations Plan-ning (S&Op), provide a vital link between leanmanufacturing operations within the supply chainand responsive distribution and differentiation op-erations (Sauvage, 2003; Faisal et al., 2006).These practices hold important promise in en-

abling risk prevention and recovery (Lavastre et al.,2012). By enabling better control of inventoriesand production schedules, planning and forecastingsystems reduce the risks from both upstream anddownstream events (Stadtler, 2005). These plan-ning systems have long-, medium-, and short-termhorizons and include master planning, requirementsplanning, and demand and distribution planning.Evidence of the use of such systems and routines toprotect the supply chain has been found by Fleis-chmann and Meyr (2003) and Stadtler and Kilger(2005). This leads us to our third hypothesis:

Hypothesis 3. There is a positive relationship be-tween the implementation rate of flexibility capabil-ities (ξ3) and the level of resilience (η1) in supplychains.

Moderating Effects of Supply Chain Risks Us-ing the classification presented in the risk litera-ture review by Heckman et al. (2015), we analyzethe risk sources that might affect the supply chainmanager’s effort to increase the resilience of a sup-ply chain depending on whether the risk sourcelies within or beyond the supply chain boundaries(Wagner and Bode, 2008; Waters, 2007). Internalrisks stem from suppliers and customers. They arereferred to as internal to reflect that they shouldbe within the control of the supply chain manager.External risks are outside her or his control.External Supply Chain Risks: Chopra and Sodhi

(2004) highlight the importance of understandingthe nature and effectiveness of supply chain risks

to be able to set up or strengthen the firm’s ca-pabilities to more effectively manage those risksand thus become more resilient. In terms of risksoutside a firm’s supply chain, Walters (2006) illus-trates the significant impact of external risks—suchas economic, social, and political risks for supplychains—on the performance and qualities of a sup-ply chain. We contend that macro-economic, so-cial, and political risks will counteract the effortsdeployed by the supply chain manager to increasethe resilience of the whole chain. Such external riskscan negatively affect how lower order capabilitieswill develop resilience (Bode et al., 2011; Altay andRamirez, 2010).

Internal Supply Chain Risks: Supply chainsrepresent vertical inter-organizational networks offirms that are closely linked to their up-stream anddown-stream supply chain partners (Carvalho et al.,2012). As such, suppliers as well as customers havean impact on establishing supply chain (manage-ment) capabilities (e.g., Teller et al., 2016) as wellas on resilience. Hwang et al. (2013) highlight theimportance of supplier impact on risks affecting thecapabilities of firms, for example, through a lack ofreliability, lead times, or delivery problems. Tang(2006) and Chopra et al. (2007) explicitly arguethat suppliers represent a source of risks to firmswithin a supply chain. Walters (2006) providescomparable arguments of customers posing poten-tial risks to their up-stream supply chain partners,for example, if the customer goes into administra-tion, generates variable demand, or has orderingproblems. Consequently, risks related to suppliersand customers, that is, risks outside the firm butinside the supply chain affecthow firms will be ableto garner all the benefits from increasing their lowerorder capabilities to develop resilience.

To conclude, we propose that supply chain risksrelated to both external factors and those related toup-stream and down-stream supply chain partnersaffect the relationship between supply chain capa-

6

External-capabilities

Integration-capabilities

Flexibility-capabilities

ξ3

ξ2

ξ1

Resilience

(H2)

�1

(H1)�11

�12

(H3)�13

Control Variables

Size of the firm (c1) Industry affilitation (c2)

Supplier-Chain-Risks

(�1)--External-risks--Internal/Supplier-risks--Internal/Customer-risks

(�2)

(�3)

H43H42

H41

Figure 2: Conceptual model.

bilities and resilience. We thus propose the finalhypothesis H4 as follows:

Hypothesis 4. Supply chain risks (external risks(µ1), supplier risks (µ2), and customer risks (µ3))affect the positive relationship between the imple-mentation rate of (external (ξ1, H41), integration(ξ2, H42), and flexibility (ξ3, H43)) capabilities andthe level of resilience (η1) in supply chains.

Our conceptual model comprises all four hy-potheses, as depicted in Figure 2. Based on thedynamic capabilities approach, our model proposesthe effects of capabilities on the resilience of a sup-ply chain as well as risk factors that influence thoseeffects.We now turn to the empirical test of our concep-

tual model.

3. Methodology

3.1. Research design

The design involves a survey among supply chainmanagers, using a single respondent in each orga-nization as the analysis unit. We considered thesemanagers as key informants (e.g., Campell, 1955)since – due to their role within their organizations– they have the most expertise and access to infor-mation on their organizations’ capabilities, supplychains, and environments.

Our research views the organization as “embed-ded in a network of relationships that impact itsperformance” (Saraf et al., 2007), p. 327. Werecognize that a multiple-respondent survey designwould have been preferable, but chose a single-respondent design to improve acceptable responserate (Saraf et al., 2007), as suggested by Tang andTang (2010) for studying inter-organizational phe-nomena. Even though the study has limited ex-planatory powers owing to the subjective nature ofthe data gathered, the use of subjective data is com-mon in this type of research and considered accept-able (Chan et al., 1997). We opted for a web-basedsurvey approach (Grant et al., 2005) due to the tar-get population size, the number of questions, andthe cost involved in contacting respondents. An-swers were anonymized to allay respondent identi-fication problems.

The population of supply chain managers was ap-proached through an electronic mailing campaignto the 8,000 French tested e-mail addresses of aSupply Chain newsletter. The subscribers are opt-in readers with an interest in general supply-chainmanagement news. Even though 366 replies wererecorded, only 171 were valid for statistical analy-sis, a response rate of 2.1% of the identified popula-tion and 47% of the sample usable (Yu and Cooper,1983). This response rate is comparable to other re-search within the field of supply chain management(e.g., Van der Vaart and Van Donk, 2008; Wagner,2010).

Several economic sectors are represented by thesample, thus increasing the results’ generality.Theusable subset included firms operating in the fol-lowing sectors: food and beverage (17.5%), retail(25.7%), and general manufacturing (24.0%). Thesample reflects a dominant proportion of small tomedium sized firms; 67.3% have less than 1,250 em-ployees.

7

3.2. Common method bias

Since there was a single informant per organiza-tion, the potential for common method bias (CMB)was assessed. There is no single best method avail-able to test CMB (Podsakoff et al., 2012). Further-more, the choice of method is a subject of intensedebate, as is the question of whether CMB can af-fect data (for a critical discussion see Richardsonet al., 2009). We applied the Harman (1967) single-factor test of CMB (Podsakoff and Organ, 1986;Podsakoff et al., 2003), which revealed twelve dis-tinct factors with eigenvalues above or near one thatcumulatively explained 87.6% of total variance. Ac-cording to this test, if common method bias exists,one of the following should be observed: (1) a singlefactor will emerge from a factor analysis of all sur-vey items (Podsakoff and Organ, 1986); or (2) onegeneral factor will emerge that accounts for most ofthe common variance existing in the data. The firstfactor explained 24.32% of the variance, which wasnot the majority of total variance and thus consid-ered to be low enough.

3.3. Non-response bias

Because significant numbers of the targeted pop-ulation failed to respond, we checked for possiblenon-response bias using a “time-trend extrapolationtest” in which “late” versus “early” respondents arecompared along key study variables (first suggestedby Oppenheim, 1966). The assumption behind thistest is that“late” respondents are very similar tonon-respondents, since their responses would nothave been recorded without follow-up efforts (Arm-strong and Overton, 1977). The t-tests conductedshowed no significant differences between “early”and “late” respondents along any of the key studyvariables.

3.4. Measurement

Capabilities and Resilience: The four theoreticalconstructs of our conceptual model—excluding the

moderating and control variables—constitute latentvariables requiring indirect measurement (see Ta-ble 3). We sifted through the nine references in lit-erature that deal with resilience or one of the lower-order capabilities using empirical surveys (Lavas-tre et al., 2012; Mandal, 2012; Qrunfleh and Taraf-dar, 2013, 2014; Richey et al., 2012; Kern et al.,2012; Moon et al., 2012; Hoffmann et al., 2013),and the survey presented in Wilden et al. (2013),which uses dynamic capabilities as second orderconstructs. The focus in each of these surveys isdifferent from ours: often the supply chain manageris not considered to be the decision maker—giventhe questions or items, the respondent could be aproduction manager, a chief executive officer, or anIT chief information officer; or he or she responds tostrategic or policy statements such as “we select thebest quality supplier”. Actual and practical usageof managerial tools and resources are not contem-plated. In fact, there is a decided absence of scalesbased upon the set of physical, financial, human,technological, and organizational resources (Grant,1991), coordinated by organizational routines (Nel-son and Winter, 1982), and deployed in an organi-zation and across organizations. Consequently, wedetermined that our study required a grounding inactual usage of such sets by supply chain managersin their daily work inside their organizations as wellas in the relationships with suppliers and distribu-tors or customers.

So, following Churchill (1979), we started withthe domain specification of each construct and col-lected the relevant measurement items in the liter-ature. However, rather than blindly applying pre-viously utilized measurement items, we used themas a starting point, and revised them based on thefeedback from five experts in the supply chain prac-tice at CapGemini Consulting France. These prac-titioners were aware of the scope and purpose ofour study and thus were able to provide precisefeedback on the measurement items. Using their

8

feedback we were able to tailor each measurementitem to most accurately measure the underlyingconstruct. As such, we employed a grounded ap-proach to make the items as accurate as possible,given our study context. This approach to developthe final items provides for a very high level of faceand content validity, while increasing the practicalrelevance and applicability of the research. After re-ceiving the feedback, we had to accommodate spe-cific changes in the constructs, which the expertshad criticized as being too complicated and havinglimited face validity. The measures for supply chainresilience were thus deductively and inductively de-veloped with the help of practitioners.

A draft version of the survey questionnaire waspre-tested among experts and journalists from sup-plychainmagazine.fr. As a result of this pretest,some inconsistencies and unclear formulations wereaddressed. Given the numerous definitions of re-silience available in the literature and the preva-lent confusion in the minds of practitioners (Kidd,2000), it was expected that a particularly widecross-section would emerge. A broad consensus wasachieved through a general discussion in which eachparticipant described the effect of each practice onthe overall supply chain and how this effect couldbe achieved and measured. In a subsequent pre-test, the questionnaire was presented to five supplychain managers, whose remarks were then incorpo-rated. When questioned about capabilities, the re-spondents were asked to rate their agreement, witha response range from totally disagree (rated 1) tofully agree (rated 5). For each capability, managerswere asked to specify if it was “not applicable totheir particular case” (rated 1), “under considera-tion” (rated 2), partly deployed (3), fully deployedbut still only partially used (4), to fully deployedand in use (rated 5). The result is a list of 9 af-firmations about capabilities found in their supplychains. A final updated list was drawn up thatcaptures both the comments about clarity and sim-

plicity as well as system- or process-related remarksabout their capabilities. The list of all measurementitems underlying the constructs of our conceptualmodel can be found in the appendix.

We consider all constructs in our model to beof a reflective nature. We base this decision onthe notions of Jarvis et al. (2003): The directionof causality goes from the latent construct towardsthe indicators for all of our constructs. This is ofparticular importance for our dependent constructResilience (η1), given Lee and Cadogan (2013) cri-tique on treating formative constructs as being de-pendent.Supply Chain Risks: To measure the three sup-

ply chain risk variables we followed the notions ofWalters (2006) and Heckman et al. (2015), who dis-tinguish between risk that is external to the sup-ply chain, including political, social, environmen-tal and economic risks, and risk that is internal tothe supply chain, which is related to suppliers andcustomers (see Table 3). Our questionnaire asksrespondents to indicate the degree to which theirsupply chain is affected by the various dimensionsof supply chain risk and thus treat these responsesas manifest variables.

3.5. Control variables

We consider two control variable that potentiallyinfluence the proposed effects in our conceptualmodel: company size (c1, operationalised by thenumber of employees) and industry affiliation (c2).The inclusion of the first control is supported by dis-cussion on the different roles and practices of SCMin large as opposed to small organisations, and thusthe notion that the size of a company affects theadvantages gained from SCM (Arend and Wisner,2005).

In terms of the second control Harland (1996)identified that the position of a company in a sup-ply chain (= industry affiliation) affects the man-agement of supply chains. Given the distribution of

9

industry affiliation in our samples we use a dichoto-mous scale to measure our control variable, thatis, the companies –represented by our respondents–are either affiliated to the manufacturing, retail orany other industry.

4. Analyses

4.1. Variance-Based Structural Equation Modeling

This study uses variance-based structural equa-tion modeling (VBSEM) (two main references areWold, 1982, 1985), a technique for component-based structural estimation modeling. Variance-based SEM has distinctive features compared tocovariance-based SEM (SEM-ML). VBSEM hasless restrictive assumptions on characteristics suchas measurement scales, sample size, and distribu-tional assumptions (Chin, 1998b; Tenenhaus et al.,2005). Chin and Newsted (1999) observe that VB-SEM is generally better suited to studies in whichthe objective is prediction, or the phenomenon un-der study is new or changing. Instead of relyingon overall goodness-of-fit tests, variance-based SEMtests the strength and direction of individual pathsby statistical significance (Calantone et al., 1998).The sample size requirement for VBSEM is tentimes the larger value of the following: (a) the blockwith the largest number of indicators, or (b) the de-pendent latent variable with the largest number ofindependent variables impacting it (Chin, 1998b).Tenenhaus et al. (2005), in a more theory-orientedpaper that complements the work of Chin (1998b),compares both SEM-ML and VBSEM. Even thoughit is recognized that these methods give differentresults, for our purpose, VBSEM is more suitablegiven that the theory is still in development. Max-imum Likelihood modeling techniques are bettersuited once confirmatory studies have been made(Lee et al., 2006). VBSEM allows for more ex-ploratory investigations into the links between cer-tain enablers and the traits of supply chains due to

its less rigorous requirement of restrictive assump-tions.

4.2. Evaluation of the measurement and structuralmodel

To systematically evaluate our VBSEM results,we first investigated the measurement model andsubsequently the structural model (Hair et al.,2014). All t-values of the factor loadings are highlysignificant at p < 0.001 (see Table 3). Further,all loadings exceed the suggested size of 0.70 (Hul-land, 1999). The internal consistency is also satis-factory for all factors (Cronbach’s (α > 0.70), andfor all factors the composite reliability (ρ) meetsthe requirement of being above 0.70 (Fornell andLarcker, 1981). The degree of convergent validityproved to be acceptable, with the average variancesextracted (AVE) higher than 0.50 (Bagozzi and Yi,1988). With regard to the constructs’ discriminantvalidity, the AVE is larger than the highest of thesquared inter-correlations with the other factors inthe measurement models (see Table 1). Addition-ally, all factor loadings on the assigned factor arehigher than cross-loadings on the non-assigned fac-tors Chin (1998a). To conclude, all constructs inthe model show sufficient validity.

Table 1: Convergent validity, composite reliability and dis-criminant validity measures for capabilities

Constructs ρ α ξ1 ξ2 ξ3 η1

External (ξ1) .855 .781 (.776)

Integration (ξ2) .814 .701 .479 (.723)

Flexibility (ξ3) .812 .752 .143 .265 (.730)

Resilience (η1) .865 .792 .292 .372 .301 (.785)

Average variance extracted values (AVE) shown on the diagonal;

Squared correlation matrix for constructs below the diagonal;

α, Cronbach’s alpha; ρ, composite reliability;

Structural model: Unlike covariance-based SEM,its variance-based counterpart does not offer com-parable global fit measures (e.g., Henseler andSarstedt, 2013; Hair et al., 2012). Rather than

10

calculating goodness-of-fit measures as proposed byTenenhaus et al. (2005); Hair et al. (2014) suggestinvestigating the coefficient of determination (r2-value) and the significance of the structural pathcoefficient to use as primary evaluation criteria forthe structural model. Our estimation shows an r2-value of .221 which represents a satisfactory value.Furthermore, two out of three structural paths aresignificant and as such, represent medium-size ef-fects according to Cohen (1988).

4.3. Model robustness test

Next, we evaluate the impact of our two controlvariables (c1, c2) on the main associations in ourmodel (see Figure 2), following the procedure ap-plied by Robson et al. (2008). The direct impactof c1 and c2 operationalized by three dummy vari-ables (for the manufacturing, retailing and otherindustries) on the dependent construct ξ2 are all in-significant (t-values � 1.965) and very weak. Whencomparing the structural associations as proposedin our hypotheses by including or excluding the con-trol variables in the model, we see that the coef-ficients change insignificantly on the third decimalplace and the significance of the associations do notchange.These results suggest that the two control vari-

ables do not confound the proposed relationships inour conceptual model. Moreover, we can concludethat the structural associations are independent ofthe industry affiliation and company size. Since thetwo control variables lack explanatory power, wetrimmed our model and excluded them from thefollowing analysis.

5. Results

5.1. Structural effects

The estimation results in Table 2 show thatthe effect of external capabilities on resilience isinsignificant on a 5% level (γ11, .144; p > .05).We therefore cannot support the first hypothesis.

Nevertheless, the other two capability constructs,that is, integration and flexibility, impact resiliencesignificantly (γ12, .246; γ13, .214; p,< .01). Conse-quently, hypotheses H2 and H3 are supported.

5.2. Moderating effectsTo test the proposed moderating effects we ap-

plied the product indicator approach, as suggestedby, among others, Busemeyer and Jones (1983) andKenny and Judd (1984). This means that for eachmoderating effect a product term is calculated us-ing the indicators of a predicting variable (in ourcase, one of the three capability constructs, ξ1, ξ2

or ξ3) and the moderator variables (µ1, µ2 or µ3)(Henseler and Chin, 2010). This term is then in-cluded as a (latent) interaction term and as suchrepresents the moderating effects (see hypothesisH4) in the conceptual model. The impact of theinteraction term on resilience (η1) measures the sig-nificance and the size of the moderating effects.

Henseler and Chin (2010) recommend the prod-uct indicator approach for models such as that pro-posed in this paper, specifically, models where thepurpose of the estimation is to (1) explain impacts,(2) describe interaction effects, and (3) focus on theprediction of endogenous constructs. Furthermore,the product indicator approach is regarded as su-perior to the frequently used multi-group analysiswhen the moderating variable is of a continuous na-ture. Multi-group analysis, and thus the test for in-variance between coefficients, is most appropriate inthe case of dichotomous moderating variables andexperimental designs (Bagozzi et al., 1991).

In terms of external risks (µ1) the resultsshow only a negative significant moderating effect(−.213; p < .05) on the association between Ex-ternal Capabilities on Resilience (γ11). Thus, anincreasing economic risk leads to a weaker impactof external capabilities on resilience. In terms ofsupplier risk (µ2), we found a significant moder-ating effect (.214, p < .01) on the relationship be-tween Integration Capabilities and Resilience. This

11

Table 2: Estimation resultsHypothesis (structural effects) Coeff.

H1 (γ11): External Capabilities (ξ1)→ Resilience (η1) .144ns

H2 (γ12): Integration Capabilities (ξ2)→ Resilience (η1) .246***

H3 (γ13): Flexibility Capabilities (ξ3)→ Resilience (η1) .214**

Hypothesis (moderation effects)

External to the supply chain: External Risks (µ1)‡

H41.µ1: External capabilities (ξ1) x External risks (µ11)→ Resilience (η1) -.213*

H42.µ1: Integration capabilities (ξ2) x External risks (µ11)→ Resilience (η1) .215ns

H43.µ1: Flexibility capabilities (ξ3) x External risks (µ11)→ Resilience (η1) .003ns

Internal to the supply chain: Supplier risks (µ2)

H42.µ2 : External capabilities (ξ1) x Supplier risks (µ12) → Resilience (η1) -.176ns

H42.µ2: Integration capabilities (ξ2) x Supplier risks (µ12) → Resilience (η1) .214**

H43.µ2: Flexibility capabilities (ξ3) x Supplier risks (µ12)→ Resilience (η1) .011ns

Internal to the supply chain: Customer risks (µ3)

H41.µ3 : External capabilities (ξ1) x Customer risks (µ13) → Resilience (η1) -.111ns

H42.µ3: Integration capabilities (ξ2) x Customer risks (µ13) → Resilience (η1) -.091ns

H43.µ3: Flexibility capabilities (ξ3) x Customer risks (µ13)→ Resilience (η1) -.128ns

Notes: t-values calculated by applying a bootstrapping procedure with 1,000 sub-samples

(Chin, 1998b); ns, non-significant; *,p < .05; **,p < .01; ***, p < .001;‡, derived measurement

that combines the rating results related to social, political, economic, and environmental

risks, through the calculation of mean values; caefficients of determination, r2η1 , .221.

12

means that the effect becomes stronger as the sup-plier risk increases. Customer Risk turned out tohave no moderating impact (p > .05) on any ofthe structural effects. We can conclude that Ex-ternal and Supplier Risk represent significant mod-erators as they affect at least one structural pathin the model. Consequently H41 and H42 can besupported.

6. Discussion and concluding remarks

In this paper we provide insights into the lower-order capabilities that help a supply chain toachieve resilience. We provide a research frame-work that builds upon earlier literature about re-silience in supply chains. Within a dynamic capa-bility setting grounded in the Resource-Based View,we describe how lower-order capabilities developedusing cross-functional and inter-organizational rou-tines can provide a supply chain with higher-orderoperational capabilities.Through our research we substantiate the theory-

driven conceptual model of supply chain resilience,which is regarded as a major source of competi-tive advantage (Chang and Grimm, 2006; Li et al.,2008; Wisner, 2003). Starting from theoretical defi-nitions of resilience in supply chains, we have oper-ationalized them with supply chain managers, traitby trait. A conceptual model, embedded in thedynamic capability approach, was developed andtested using data from French supply chain man-agers. In summarizing the contributions of this pa-per, we distinguish between implications for theoryand for practice.Implications for theory

First, we conclude that in the view of supply chainmanagers, resilience is not easily enhanced, eventhough it is a highly desirable trait (Bhamra et al.,2011). In answer to the first question we askedin the Introduction, only integration and flexibil-ity capabilities positively affect the resilience levelof a supply chain. These findings resonate with

those revealed through the Blackhurst et al. (2011)case study research. There, three major categoriesof factors were deemed to enhance resilience: hu-man capital resources, organizational and inter-organizational capital resources, and physical cap-ital resources. Of these, organizational resourceswere said to include defined communication net-works, contingency plans, and supplier relationshipmanagement.

Second, we found that our results deviate fromfindings in the literature. As regards the results re-ported from the empirical investigation in Mandal(2012), the link between external capabilities andresilience cannot be corroborated, while the link be-tween the supply chain infrastructure and integra-tion and resilience is only partially validated. Thismay be due to the fact that the sample selection inMandal (2012) is composed of IT professionals andnot supply chain managers. We are unable to con-firm the notions of Klibi et al. (2010) and see thatthe Efficient Customer Response type of externalcollaboration practices —as exemplified by VendorManaged Inventories and Warehouse ManagementSystems— to streamline inventories across echelonshave no impact on resilience.

Third, we selectively found moderating effects ofsupply chain risks on the relationships between ca-pabilities and the resilience of a supply chain. Morespecifically, the question we asked in the Introduc-tion was: How do environmental factors related tosupply chains influence the effectiveness of thesepractices? The answer we provide here supportsthe notions developed in Walters (2006) in termsof risks external to the supply chain as well as risksinternal to the supply chain related to suppliers. In-terestingly, we find that customer risks do not playa significant role in affecting the proposed relation-ships in our model.

The size of the focal firm as a proxy for the ex-tension of the supply chain has no influence on itsresilience, even though it should be a facilitating

13

factor in the implementation of lower-order capa-bilities. This seems surprising for, as recognized inWaters (2007) several times, larger firms have moresophisticated tools and should thus be better placedto address market as well as environmental risks.

By applying the dynamic capabilities approachas a theoretical underpinning of our research, wehave highlighted the link between specific lower-order capabilities and a supply chain’s operationalcapability, namely, resilience. Additional researchis required along three directions: (a) how to en-hance specific supply chain capabilities; (b) how tocombine those operational capabilities and how bestto add learning capabilities that can be made dy-namic; and (c) how to link such operational capa-bilities for the competitive advantage of a supplychain.

Implications for practiceIn this paper, contrary to most papers dealing withthe subject of resilience, we have positioned our-selves from the point of view of supply chain man-agers to understand how the actions, decisions, andpractices they apply, the routines they set up, thecollaborative and coordination effort and resourcesthat they build upon contribute to the resilience ofthe supply chain to which their firm belongs.

Our results indicate that only some practices andasset and human deployments will provide an in-creased measure of this quality. Managers whocombine and enhance both integration and flexibil-ity capabilities will observe a level of resilience intheir supply chain. This means that they must notonly use information technology tools and routinesto integrate their internal organization (throughtheir ERP) but also use other supply chain man-agement software to integrate their suppliers, cus-tomers, distributors, and logistics service providers.These efforts enhance collaboration by sharing fore-casts and sales data and allowing continuous inven-tory adjustments. In conjunction with logistics ser-vice providers, using track and trace technologies

for goods provides advanced tip-offs about eventsand glitches that affect service levels and quality. Itis a notable result that the supply chains affectedby high supplier risk concomitantly deploy these in-tegration practices and resources.

Our results show that External Capabilities donot influence resilience. When we delve into thetools, practices, and routines involved, the follow-ing interpretation can be suggested: Supply chainmanagers do not have ample experience in apply-ing efficient customer response policies, deployingboth warehouse and transport management sys-tems, streamlining inventories, as well as deployingvendor managed inventories. Hence the resilienceeffects have yet to be observed. The implemen-tation of the routines and processes of IntegrationCapabilities, which involve the deployment of sup-ply chain management software connected to theERP—managing supplier performance, using busi-ness intelligence software to generate reports pro-viding insights into the working of the supply chainas well as tracking goods—are all somewhat re-cent and require additional managerial capacitiesand training to be deployed effectively. Such prac-tices may have not yet been mastered by all supplychains. This view is reinforced by the result that theinfluence of integration capabilities on resilience ishigher when supplier risk is higher. That is, when asupply chain is subject to significant internal risks,the best line of conduct is to foster increased inte-gration of the chain links so as to enhance its overallresilience.

Flexibility capabilities enhance resilience: Re-silience can be augmented through the combina-tion of alternative production and site plans, aswell as by making plans more flexible and versa-tile. The pertinence of the deployment of these re-sources and routines increases with the incidenceof external risks faced by the supply chain, such asraw material price hikes, political upheavals, or reg-ulatory changes. Flexibility functions in several di-

14

mensions. The first dimension is the ability to meetnew demands in terms of product type or quanti-ties. The second is the ability to reconfigure thesupply chain (upstream and downstream) by flex-ibly deciding whether to make or buy, to changelocations, or to implement site specialization, whilekeeping tabs on a pool of suppliers. Unavoidably,such abilities in a supply chain go hand in handwith the supply chain manager’s increased abilityto detect and measure risks. A supply chain man-ager should deploy processes to identify and mon-itor risks and potential areas of trouble as a com-plement to the practices discussed above.When controlling for the impact of the size of

the firm, we found that the size of the focal firm isnot an impediment to resilience, as even small busi-nesses with limited resources can achieve the samelevel of resilience as larger firms. Neither could wedistinguish an effect due to the economic sector. Byextension, the focal firm can occupy any positionin the supply chain (from manufacturer to retailer)without this position affecting its ability to enhanceresilience.

7. Limitations and future research agenda

As with all research, this study has some limita-tions. The respondents to our survey were Frenchmanagers, which results in a bias towards a West-ern European supply chain context. Future stud-ies could be conducted in other country settings.Furthermore, we do not differentiate between dif-ferent industries and supply chain stages (except ascontrol variables), the study of which might yieldadditional insights. We applied a single-informantapproach and thus rely exclusively on the perspec-tive of supply chain managers. Using experts fromother parts of the organization, such as marketingor finance, could complement our findings.Our conceptual model only considers one mod-

erator: supply chain risks. Further analyses of ourdata should include other moderators, such as firm

or supply chain characteristics, to identify particu-lar capability-resilience relationships.

Our approach is quantitative in nature. Qual-itative interviews or focus group discussions withmanagers would help to understand better why ex-ternal capabilities do not affect resilience whereasthe other capabilities do.

Finally, further research is needed about the roleof managerial expertise in building upon informa-tion technology as the means to embed managerialprocesses both within and across organizations soenhancing resilience. Information technology wouldneed to be separated into more nuanced categoriesinvolving lower-order practices and routines (suchas ERP, MRP II, collaborative platforms, tracking& tracing) or higher-order structures, such as in-formation aggregation systems for business intelli-gence. In this way, researchers could better evalu-ate the potential impact of each set of practices onsupply chain resilience.

8. Appendix

15

Table 3: External indicators and loading factorsLatent construct

λ t-value p-valueIndicatorξ1: External Capabilitiesx11: deploying an Efficient Customer Response policy .950 23.944 <.001x12: deploying WMS and TMS .804 13.132 <.001x13: streamlining and resizing inventory in the distribution network .644 7.089 <.001x14: deploying a Vendor Managed Inventory policy .666 6.836 <.001ξ2: Integration Capabilitiesx21:managing the performance of your suppliers in a collaborative way .740 11.135 <.001x22: integrating ERP with other SCM tools .648 6.532 <.001x23: deploying IT-based reporting tools .784 16.870 <.001x24: deploying tracking & tracing tools .715 10.482 <.001ξ3: Flexibility Capabilitiesx31: setting up alternative production contingency plans .903 23.989 <.001x32: developing the versatility and flexibility of your sites .791 11.888 <.001x33:making production sites specialize per technology or product .690 6.645 <.001η1: Resiliencey11: Your supply chain system enables you to evaluate your process vul-

nerabilities constantly.722 11.266 <.001

y12: You deploy alternative plans associated with identified risks .829 21.426 <.001y13: Your firm is able to evaluate the levels of risk facing your supply chain .851 25.749 <.001y14: Your supply chain organization allows you to increase visibility over

all your chain.730 10.081 <.001

Supplier Risksm11: Your supply chain is affected by external political risksm12: Your supply chain is affected by external social risksm13: Your supply chain is affected by external environmental risks N/A N/A N/Am14: Your supply chain is affected by external economic risksm2 : Your supply chain is affected by risks related to your suppliersm3 : Your supply chain is affected by risks related to your customersNotions : All statements based on a five-point Likert scale (1, completely disagree, 5, completely agree)

16

Agarwal, A., Shankar, R., Tiwari, M. K., 2006. Modelingthe metrics of lean, agile and leagile supply chain: anANP-based approach. European Journal of OperationalResearch 173, 211–225. 4

Akkermans, H., Bogerd, P., Yücesan, E., van Wassenhove,L. N., 2003. The impact of ERP on supply chain manage-ment: exploratory findings from a European Delphi study.European Journal of Operational Research 146, 284–301.4

Altay, N., Ramirez, A., 2010. Impact of disasters on firms indifferent sectors: Implications for supply chains. Journalof Supply Chain Management 46 (4), 59–80. 1, 6

Arend, R., Wisner, J. D., May 2005. Small business and sup-ply chain management: is there a fit? Journal of BusinessVenturing 20 (3), 403–436. 9

Armstrong, S. J., Overton, T. S., 1977. Estimating non-response bias in mail surveys. Journal of Marketing 14,396–402. 8

Bagozzi, R. P., Yi, Y., 1988. On the evaluation of structuralequation models. Academy of Marketing Science 16 (1),74–94. 10

Bagozzi, R. P., Yi, Y., Singh, S., June 1991. On the use ofstructural equation models in experimental designs: Twoextensions. International Journal of Research in Market-ing 8 (2), 125–140. 11

Barreto, I., 2010. Dynamic capabilities: a review of past re-search and an agenda for the future. Journal of OperationsManagement 36 (1), 256–280. 2, 3

Berkes, F., Colding, J., Folke, C., 2003. Navigating social-ecological systems; Building resilience for complexity andchange. Cambridge University Press. 1

Bhamra, R., Dan, S., Burnard, K., September 2011. Re-silience: the concept, a literature review and future di-rections. International Journal of Production Research49 (18), 5375–5393. 3, 4, 13

Blackhurst, J., Craighead, C., Elkins, D., Handfield, R.,2005. An empirically derived agenda of critical research is-sues for managing supply-chain disruptions. InternationalJournal of Production Research 43 (19), 4067–81. 3

Blackhurst, J., Dunn, K., Craighead, C., 2011. An empiri-cally derived framework of global supply resiliency. Jour-nal of Business Logistics 32 (4), 374–391. 3, 13

Bode, C., Wagner, S. M., Petersen, K. J., Ellram, L. M.,2011. Understanding responses to supply chain disrup-tions: Insights from information processing and resourcedependence perspectives. Academy of Management Jour-nal 54 (4), 833–856. 1, 6

Boyson, S., Corsi, T., Verbraeck, A., 2003. The e-supplychain portal: a core business model. Transportation Re-search Part E 39, 175–192. 5

Brandon-Jones, E., Squire, B., Autry, C. W., Petersen, K. J.,2014. A contingent resource-based perspective of supply

chain resilience and robustness. Journal of Supply ChainManagement 50 (3), 55–73. 1, 3

Busemeyer, J. R., Jones, L. E., May 1983. Analysis of multi-plicative combination rules when the causal variables aremeasured with error. Psychological Bulletin 93 (3), 549–562. 11

Calantone, R. J., Graham, J. L., Mintu-Wimsatt, A., 1998.Problem-solving approach in an international context: an-tecedents and outcome. International Journal of Researchin Marketing 15 (1), 19–35. 10

Campell, D. T., 1955. The informant in quantitative re-search. American Journal of Sociology 60 (4), 339–342.7

Carvalho, H., Barroso, A. P., Machado, V. H., Azevedo, S.,Cruz-Machado, V., 2012. Supply chain redesign for re-silience using simulation. Computers & Industrial Engi-neering 62, 329–341. 6

Cepeda, G., Vera, D., 2007. Dynamic capabilities and opera-tional capabilities: a knowledge management perspective.Journal of Business Research 60, 426–437. 2, 3

Chan, Y. E., Huff, S. L., Copeland, D. G., 1997. Businessstrategic orientation, information systems strategic ori-entation, and strategic alignment. Information SystemsResearch 8 (2), 125–150. 7

Chang, L., Grimm, C. M., 2006. The application of empiricalstrategic management research to supply chain manage-ment. Production and Operations Management 27, 1–56.13

Charles, A., Lauras, M., Wassenhove, L. V., 2011. A modelto define and assess the agility of supply chains: build-ing on humanitarian experience. International Journal ofPhysical Distribution & Logistics Management 40 (8/9),722–741. 5

Chin, W. W., 1998a. Issues and opinions on structural equa-tions modeling. MIS Quarterly 22 (1), 1–10. 10

Chin, W. W., 1998b. The partial least squares approach forstructural equation modeling. Laurence Erlbaum Asso-ciates, London. 10, 12

Chin, W. W., Newsted, R., 1999. Statistical Strategies forSmall Sample Research. Sage Publications, Ch. Structuralequation modeling analysis with small samples using par-tial least squares, pp. 307–341. 10

Chmielewski, D. A., Paladino, A., 2007. Driving a resourceorientation: reviewing the role of resource and capabilitycharacteristics. Management Decision 45, 462–483. 2

Chopra, S., Reinhardt, G., Mohan, U., 2007. The importanceof decoupling recurrent and disruption risks in a supplychain. Naval Research Logistics 54 (5), 544–555. 6

Chopra, S., Sodhi, M., 2004. Managing risk to avoid supply-chain breakdown. MIT Sloan Management Review 46 (1),53–61. 3, 6

Christopher, M., Peck, H., 2004. Building the resilient sup-

17

ply chain. International Journal of Logistics Management15 (2), 1–13. 1

Churchill, G. A., 1979. A paradigm for developing bettermeasures of marketing constructs. Journal of MarketingResearch 16 (1), 64–73. 8

Cohen, J., 1988. Statistical power analysis for the behavioralsciences, 2nd Edition. L. Erlbaum Associates.URL https://books.google.fr/books?id=Tl0N2lRAO9oC11

Creighton, D. P., Walker, A. C., Mundt, G., 2014. Contin-gency contracting in support of CONUS disasters: a casestudy of the 1994 northridge earthquake, 2005 hurricanekatrina and 2012 hurricane sandy. Ph.D. thesis, CalhounNaval Postgraduate School, Monterey, California, USA.URL http://hdl.handle.net/10945/42603 1

Danese, P., Romano, P., 2011. Supply chain integration andefficiency performance : a study on interaction betweencustomer and supplier integration. Supply Chain Mange-ment: an International Journal 16 (4), 220–230. 5

Drnevich, P. L., Kriauciunas, A. P., 2011. Clarifying the con-ditions and limits of the contributions of ordinary and dy-namic capabilities to relative firm performance. StrategicManagement Journal 32, 254–279. 3

Eisenhardt, K. M., Martin, J. A., 2000. Dynamic capabil-ities: what are they? Strategic Management Journal21 (10-11), 1105–1121. 3

Faisal, M. N., Banwet, D. K., Shankar, R., 2006. Mappingsupply chains on risk and customer sensitivity dimensions.Industrial Management & Data Systems 106 (6), 878–895.5, 6

Faisal, M. N., Banwet, D. K., Shankar, R., 2007. Informationrisks management in supply chains: an assessment andmitigation framework. Journal of Enterprise InformationManagement 6, 677–699. 4, 5

Fiksel, J., 2006. Sustainability and resilience: towards a sys-tems approach. Sustainability: Science, Practice & Policy2 (2), 1–8. 3

Fleischmann, B., Meyr, H., 2003. Supply chain management:Design, coordination and operation. Handbooks in Opera-tions Research and Management Science. Springer, Berlin,Ch. Planning hierarchy, modeling and advanced planningsystems, pp. 455–523. 6

Flynn, B., Huo, B., Zhao, X., 2010. The impact of supplychain integration on performance: a contingency and con-figuration approach. Journal of Operations Management28, 58–71. 5

Fornell, C., Larcker, D. F., 1981. Evaluating structural equa-tion models with unobservable variables and measurementerror. Journal of Marketing Research 18 (1), 39–50. 10

Frohlich, M., Westbrook, R., 2001. Arcs of integration: aninternational study of supply chain strategies. Journal ofOperations Management 19, 185–200. 5

García-Dastugue, S. J., Lambert, S. J., 2003. Internet-enabled coordination in the supply chain. Industrial Mar-keting Management 32 (3), 251–264. 5

Gosain, S., Malhotra, A., El Sawy, O. A., 2005. Coordinat-ing for flexibility in e-business supply chains. Journal ofManagement Information Systems 21, 7–45. 4

Grant, D. B., Teller, C., Teller, W., 2005. Hidden opportuni-ties and benefits in using web-based business-to-businesssurveys. International Journal of Market Research 47 (6),641. 7

Grant, R. M., 1991. The resource-based theory of compet-itive advantage: implications for strategy formulation.California Management Review 33 (3), 114–135. 4, 8

Gulati, R., Nohria, N., Zaheer, A., 2000. Strategic networks,special issue. Strategic Management Journal 21 (3), 203–217. 1

Hair, J. F., Ringle, C. M., Mena, J. A., 2012. An assess-ment of the use of partial least squares structural equationmodeling in marketing research. Journal of the Academyof Marketing Science 40 (3), 414–433. 10

Hair, J. F., Sarstedt, M., Hopkins, L., Kuppelwieser, V. G.,2014. Partial least squares structural equation modeling(PLS-SEM) an emerging tool in business research. Euro-pean Business Review 36 (2), 106–121. 10, 11

Harland, C., December 1996. Supply network strategies thecase of health supplies. European Journal of Purchasing& Supply Management 2 (4), 183–192. 9

Harman, H. H., 1967. Modern factor analysis. University ofChicago Press, Chicago, IL, USA. 8

Heckman, I., Comes, T., Nickel, S., 2015. A critical reviewon supply chain risk – definition, measure and modeling.Omega 52 (119-132). 6, 9

Helfat, C. E., Peteraf, M. A., 2003. The dynamic resource-based view: capability lifecycles. Strategic ManagementJournal 24 (10), 997–1010. 2

Hendricks, K., 2005. Association between supply chainglitches and operating performance. Management Science51 (5), 695–711. 3, 5

Henseler, J., Chin, W. W., 2010. A comparison of approachesfor the analysis of interaction effects between latent vari-ables using partial least squares path modeling. StructuralEquation Modeling: A Multidisciplinary Journal 17 (1),82–109. 11

Henseler, J., Sarstedt, M., 2013. Goodness-of-Fit indicesfor Partial Least Squares path modeling. ComputationalStatistics 28, 565–580. 10

Hindle, T., 2008. Guide to Management Ideas and Gurus.The Economist, Ch. Suply chain management, pp. 177–178. 1

Hoffmann, P., Schiele, H., Krabbendam, K., 2013. Uncer-tainty, supply risk management and their impact on per-formance. Journal of Purchasing & Supply Management

18

19, 199–211. 8Hulland, J., 1999. Use of partial least squares (PLS) in

strategic management research: a review of four recentstudies. Strategic Management Journal 20, 195–204. 10

Hwang, W., Bakshi, N., DeMiguel, V., May 2013. Inducingreliable supply: Simple versus simplistic contracts.Working Paper.URL http://faculty.london.edu/nbakshi/HBD-2013-07-05.pdf 6

Jarvis, C. B., Mackenzie, S. B., Podsakoff, P. M., Mick,D. G., Bearden, W. O., 2003. A critical review of con-struct indicators and measurement model misspecificationin marketing and consumer research. Journal of ConsumerResearch 30 (2), 199–218. 9

Kenny, D. A., Judd, C. M., July 1984. Estimating the non-linear and interactive effects of latent variables. Psycho-logical Bulletin 96 (1), 201–210. 11

Kern, D., Moser, R., Hartmann, E., Moder, M., 2012. Sup-ply risk management: model development and empiricalanalysis. International Journal of Physical Distribution &Logistics Management 42 (1), 60–82. 8

Kidd, P., 2000. Two definitions of agility.URL http://www.cheshirehenbury.com/agility/two-definitions-of-agility.html 9

Klibi, W., Martel, A., Guitouni, A., 2010. The design ofrobust value-creating supply chain networks: a criticalreview. European Journal of Operational Research 203,283–293. 5, 13

Kotzab, H., 1999. Improving supply chain performance byEfficient Customer Response: a critical comparison of ex-isting ECR approaches. Journal of Business & IndustrialMarketing 14 (5/6), 364–377. 4

Lavastre, O., Gunasekaran, A., Spalanzani, A., March 2012.Supply chain risk management in French companies. De-cision Support Systems 52 (4), 828–838. 6, 8

Lee, K. H., Yang, G., Graham, J. L., 2006. Tension andtrust in international business negotiations: American ex-ecutives negotiating with Chinese executives. Journal ofInternational Business Studies 37 (5), 623–641. 10

Lee, N., Cadogan, J. W., 2013. Problems with formativeand higher-order reflective variables. Journal of BusinessResearch 66 (2), 242–247. 9

Li, X., Goldsby, T. J., Holsapple, C. W., 2008. A uni-fied model of supply chain agility: the work design per-spective. International Journal of Logistics Management19 (3), 408–435. 2, 13

Li, X., Goldsby, T. J., Holsapple, C. W., 2009. Supply chainagility: scale development. International Journal of Logis-tics Management 20 (3), 408–424. 2

Lin, C. T., Chiu, H., Chu, P. Y., 2006. Agility index in thesupply chain. International Journal of Production Eco-nomics 100, 285–299. 5

Mandal, S., 2012. An empirical investigation into supplychain resilience. IUP Journal of Supply Chain Manage-ment 9 (4), 46–61. 5, 8, 13

Moon, K. K.-L., Yi, C. Y., Ngai, E. W. T., 2012. An instru-ment for measuring supply chain flexibility for the textileand clothing companies. European Journal of OperationalResearch 222, 191–203. 8

Naylor, J. B., Naim, M., Berry, D., 1999. Leagility: integrat-ing the lean and agile manufacturing in the total supplychain. International Journal of Production Economics 62,107–118. 5

Nelson, R. R., Winter, S. G., 1982. An Evolutionary Theoryof Economic Change. Belknap Press of Harvard Univer-sity, Cambridge, MA, USA. 4, 8

Olhager, J., 2013. Evolution of operations planning andcontrol: from production to supply chains. InternationalJournal of Production Research 51 (23-24), 6836–6843. 6

Oppenheim, A. N., 1966. Questionnaire Design and AttitudeMeasurement. Basic books, NY, NY, USA. 8

Pagell, M., 2004. Understanding the factors that enable andinhibit the integration of operations, purchasing and lo-gistics. Journal of Operations Management 22, 459–487.5

Pettit, T., Fiksel, J., Croxton, K., 2010. Ensuring supplychain resilience: development of a conceptual framework.Journal of Business Logistics 31 (1), 1–21. 3

Podsakoff, P., MacKenzie, S. B., Lee, J. Y., Podsakoff, N. P.,2003. Common method bias in behavioral research: a crit-ical review of the literature and recommended remedies.Journal of Applied Psychology 88 (5), 879–903. 8

Podsakoff, P., MacKenzie, S. B., Podsakoff, N. P., 2012.Sources of method bias in social science research and rec-ommendations on how to control it. Annual Review ofPsychology 63 (1), 539–569. 8

Podsakoff, P., Organ, D. W., 1986. Self-reports of organiza-tional research: problems and prospects. Journal of Man-agement 12 (4), 531–544. 8

Qrunfleh, S., Tarafdar, M., 2013. Lean and agile supplychain strategies and supply chain responsiveness: the roleof strategic supplier partnership and postponement. Sup-ply Chain Management: An International Journal 18 (6),571–582. 5, 8

Qrunfleh, S., Tarafdar, M., 2014. Supply chain informationsystems strategy: impacts on supply chain performanceand firm performance. International Journal of Produc-tion Economics 147, 340–350. 8

Rajaguru, R., Matanda, M. J., May 2013. Effects of inter-organizational compatibility on supply chain capabilities:exploring the mediating role of inter-organizational infor-mation systems (IOIS) integration. Industrial MarketingManagement 42 (4), 620–632. 2, 5

Richardson, H. A., Simmering, M. J., Sturman, M. C., 2009.

19

A tale of three perspectives: Examining post hoc statis-tical techniques for detection and correction of commonmethod variance. Organizational Research Methods.URL http://scholarship.sha.cornell.edu/articles/104 8

Richey, R. G., Adams, F. G., Dalela, V., 2012. Technologyand flexibility: enablers of collaboration and time-basedlogistics quality. Journal of Business Logistics 33 (1), 34–49. 8

Robson, M. J., Katsikeas, C., Bello, D. C., 2008. Drivers andperformance outcomes of trust in international strategicalliances: The role of organizational complexity. Organi-zation Science 19 (4), 647–665. 11

Sahin, F., Robinson, E. P., 2002. Flow coordination andinformation sharing in supply chains: review, implica-tions, and directions for future research. Decision Sciences33 (4), 505–536. 4

Saraf, N., Langdon, C. S., Gosain, S., 2007. IS applicationcapabilities and relational value in interfirm partnerships.Information Systems Research 18 (3), 320–339. 7

Sauvage, T., 2003. The relationship between technology andlogistics third-party providers. International Journal ofPhysical Distribution & Logistics Management 33 (3),236–253. 6

Skjoett-Larsen, T., Thernøe, C., Andersen, C., 2003. Supplychain collaboration, theoretical perspectives and empiricalevidence. International Journal of Physical Distribution &Logistics Management 33 (6), 531–549. 4

Stadtler, H., 2005. Supply chain management and advancedplanning – basics, overview and challenges. EuropeanJournal of Operational Research 163, 575–588. 6

Stadtler, H., Kilger, C., 2005. Supply chain management andadvanced planning, 3rd Edition. Springer, Berlin. 6

Su, Y.-F., Yang, C., 2010. Why are enterprise resource plan-ning systems indispensable to supply chain management.European Journal of Operational Research 203, 81–94. 4

Tang, A., Tang, X., 2010. Leveraging IT capabilities andcompetitive process capabilities for the management ofinter-organizational relationship portfolios. InformationSystems Research 21 (3), 516–542. 7

Tang, C. S., 2006. Robust strategies for mitigating supplychain disruptions. International Journal of Logistics: Re-search and Applications 9, 33–45. 6

Teece, D. J., 2007. Explicating dynamic capabilities: the na-ture and micro foundations of (sustainale) enterprise per-formance. Strategic Management Journal 28 (13), 1319–1350. 2, 3

Teece, D. J., Pisano, G., 1994. The dynamic capabilities offirms: an introduction. Industrial and Corporate Change3 (3), 537–556. 2

Teece, D. J., Pisano, G., Shuen, A., 1997. Dynamic capa-bilities and strategic management. Strategic Management

Journal 18 (7), 509–533. 2, 3Teller, C., Kotzab, H., Grant, D. B., Holweg, C., 2016. The

importance of key supplier relationship management insupply chains. International Journal of Retail and Distri-bution Management 44 (2), 1–16. 6

Tenenhaus, M., Vinzi, V. E., Chatelin, Y.-M., Lauro, C.,2005. PLS path modeling. Computational Statistics &Data Analysis 48, 159–205. 10, 11

The Economist, April6 6th 2009. Idea: supply-chain man-agement. The Economist.URL http://www.economist.com/node/13432670 1

Van der Vaart, T., Van Donk, D., 2008. A critical reviewof survey-based research in supply chain integration. In-ternational Journal of Production Economics 111, 42–55.7

Wagner, S. M., 2010. Indirect and direct supplier develop-ment: performance implications of individual and com-bined effects. IEEE Transactions on Engi 57 (4), 536–546.7

Wagner, S. M., Bode, C., 2008. An empirical examinationof supply chain performance along several dimensions ofrisk. Journal of Business Logistics 29 (1), 307–325. 3, 4, 6

Walters, D., 2006. Demand chain effectiveness – supply chainefficiencies. Journal of Enterprise Information Manage-ment 19 (3), 246–261. 6, 9, 13

Waters, D., 2007. Supply Chain Risk Management: Vulner-ability and Resilience in Logistic. Kogan Page. 6, 14

Wernerfelt, B., 1984. A resource-based view of the firm.Strategic Management Journal 5 (2), 171–180. 2

Wilden, R., Gudergan, S., Nielsen, B. B., Lings, I., 2013.Dynamic capabilities and performance: strategy, struc-ture and environment. Long Range Planning 46, 72–96.2, 8

Winter, S. G., 2003. Understanding the dynamic capabilities.Strategic Management Journal 24 (10), 991–995. 2

Wisner, J. D., 2003. A structural equation model of sup-ply chain management strategies and firm performance.Journal of Business Logistics 24, 1–26. 13

Wold, H., 1982. Systems under Indirect Observation. Vol. 2.North Holland, Amsterdam, Ch. Soft modeling: the basicdesign and some extensions, pp. 1–54. 10

Wold, H., 1985. Encyclopedia of Statistical Sciences. Vol. 6.Wiley, New York, Ch. Partial Least Squares, pp. 581–591.10

Yu, J., Cooper, H., 1983. A quantitative review of researchdesign effects on response rates to questionnaires. Journalof Marketing research, 36–44. 7

Zhao, X., Huo, B., Flynn, B., Yeung, J. H. Y., 2008. The im-pact of power and relationship commitment on the inte-gration between manufacturers and customers in a supplychain. Journal of Operations Management 26, 368–388. 5

20