Abstract
The purpose of this study is to explore the influence of supply chain relationships on supply chain (SC) resilience through empirically exploring the effects of trust, communication, commitment and cooperation on SC resilience and its further impact on supply chain performance. We employed a cross-sectional survey through online mode for collecting perceptual responses from supply chain professionals. Using 155 completed responses, the study further analysed using SmartPLS to validate the proposed relationships. Findings suggest communication and cooperation as dominant enablers of SC resilience in the integrated model that explored the inter-relationships also among the proposed relational attributes. Also, SC resilience has positive performance implications. Alternate models were suggested to further explore the inter-play between communication, trust, commitment and cooperation and how the same influences the SC resilience. The study is the first to consider the inter-relationships among the relational attributes and how the same as a whole influences SC resilience.
Introduction
Every supply chain activity has inherent risk that an unexpected disruption might occur resulting in unexpected consequences (Andotra & Gupta, 2016; Patnaik, Das & Bahinipati, 2016). For example, Thai floods and Japan earthquake incidents exemplify the sustained losses on manufacturing supply chains (Banham, 2009; Brennan, 2011). Supply chains must be developed with the capacity to sustain operations even during disruptions. Such a capability is supply chain (SC) resilience (Ponomarov & Holcomb, 2009). However, SC resilience lacks sufficient empirical investigation with some notable exceptions, such as Jüttner and Maklan (2011), Blackhurst, Dunn and Craighead (2011), Wieland and Wallenburg (2013), Brandon-Jones, Squire, Autry & Petersen (2014), Roberta-Pereira et al. (2015).
While there are many possible causes and antecedents for SC resilience, our focus is on the role of relationships among supply chain partners (rather than management capabilities) in determining SC resilience. Our focus in this direction is motivated by the fact that relationships are important determinants of other aspects of supply chain operations and performance. For example, they are of paramount importance in efficient supply chain management. Touboulic et al. (2014) argued that supply chain relationships are crucial for creating and maintaining sustainable supply chains comprising a large organization and its’ small-to-medium suppliers. Knoppen et al. (2015) reviewed 60 empirical papers on supply chain relationships and argued that supply chain relationships are crucial for building learning capabilities for improving products and processes across a supply chain. It is this aspect of relationships that has caught our attention. In a subsequent section, we argue that resilience is a dynamic capability and relationships play a key role in providing such capability.
The aspect of supply chain relationships that we are concerned with is relational attributes such as trust. We adapt the commitment–trust theory from relationship marketing (Morgan & Hunt, 1994) to the supply chain context. In particular, we focus on the relationship among communication, trust and commitment expounded in the theory. Morgan and Hunt (1994) in their key mediating variables (KMV) model of trust and commitment argued that communication enhances trust, which enhances commitment, and that both trust and commitment enhance cooperation among partners. Further, we also explore how the inter-relationships among these relational antecedents are effective in determining effective SC performance through SC resilience (Knoppen et al., 2015; Touboulic et al., 2014).
Review of Literature
Supply Chain Resilience
The SC resilience has received conceptual attention in several disciplines. Ponomarov and Holcomb (2009) attempted to integrate the different perspectives on resilience, namely, psychological, economic, organizational, etc., along with supply chain risk management (SCRM). The SCRM involves the notion of resilience as a post-disruption measure of restored operations. Further, Christopher and Rutherford (2004) argued that resilience in supply chains can be created if there is sufficient flexibility in such chains and that resilient supply chains must be adaptable to changing circumstances. In short, agility, adaptability, flexibility and resilience are allied with important characteristics of recent supply chains (Barroso et al., 2011; Christopher & Holweg, 2011; Falasca et al., 2008; Hohenstein, Feisel, Hartmann & Giunipero, 2015).
Consistent with many of the earlier connotations (Hohenstein et al., 2015), we define SC resilience as the ability to recover and continue normal state of operations in the face of disruptions (Ponomarov & Holcomb, 2009). A disruption can be any unexpected situation having the potential of creating a negative outcome for the associated firm (Wagner & Bode, 2006).
What determines SC resilience? The literature features a wide variety of causes. In a conceptual study, Christopher and Peck (2004) argue that collaboration, agility, risk management culture and supply chain re-engineering play important roles. Using an in-depth qualitative case study of a collaborative agency, Scholten, Scott and Fynes (2014) added knowledge management as an antecedent to resilient supply chains, in addition to the four factors given by Christopher and Peck (2004). Ponomarov and Holcomb’s (2009) conceptual integrative study, mentioned earlier, used a resource-based view and dynamic capabilities theory to conclude that certain logistics capabilities when integrated appropriately will lead to SC resilience. Conceptualizing resilience as a balanced state consisting of vulnerabilities and capabilities, Pettit, Fiksel and Croxton (2010) argued that SC resilience increases as capabilities increase and vulnerabilities decrease. In an identical but empirical investigation by Juttner and Maklan (2011) found that supply chain risk effect management and supply chain risk knowledge management seemed to enhance SC resilience by improving the flexibility, visibility, velocity and collaboration capabilities (termed as formative resilience capabilities) of the supply chain. Johnson, Elliott and Drake (2013) further extended the work by Juttner and Maklan (2011) and empirically showed that the cognitive, structural and relational dimensions of social capital do influence flexibility, velocity, visibility and collaboration, that is, the formative capabilities of SC resilience. Brandon-Jones et al. (2014) empirically established supply chain visibility as an effective antecedent of resilience and robustness in supply chains. More recently, Pereira, Christopher and Silva (2014) identified (using a review of allied literature from 2000 to 2013) supply chain design, strategic sourcing, supply chain design and risk management as observed inter-organizational issues that can influence the development of SC resilience. Ambulkar et al. (2015) observed that supply chain disruption–oriented firms require the ability to reconfigure resources or have a risk management resource infrastructure to develop resilience. Tukamuhabwa et al. (2015), through an extensive review of 91 prominent studies on resilience, argued that among different strategies available for developing resilience most attention has been focused on increasing flexibility, creating redundancy, forming collaborative supply chain relationships and improving supply chain agility. Their literature review also underscored that much of the literature on resilience is conceptual, theoretical and normative. More recently, Kim et al. (2015) examined the structural relationships among entities in supply networks through comparing four fundamental supply network structures to help understand supply network disruption and resilience. Their analysis showed that node/arc-level disruptions do not necessarily lead to network-level disruptions; further they also indicated that network structure significantly determines the likelihood of disruption. Todo et al. (2015) with evidence from Great East Japan Earthquake emphasized that expansion of supply chain networks has two opposing effects on resilience of firms to disasters.
Thus, among the great diversity of potential antecedents of SC resilience, it is clear that good supply chain relationships could play a role in developing many supply chain capabilities (Srinivasan, Mukharjee & Gaur, 2011), where the literature seems to lack is an empirical investigation focused on the influences of the relational attributes of communication, trust, commitment and cooperation on SC resilience. In the following sections, we argue that adaptation and rapid responses to environmental changes is a prerequisite for SC resilience. Such adaptation requires cooperation of supply chain partners, which in turn requires commitment, trust and communication. Our goal is to provide empirical evidence for such a model.
SC Resilience as a Dynamic Capability
Responding to rapidly changing environments requires firms to have dynamic capabilities. Dynamic capability theory says that is not just the presence of resources (Barney, 1991, 2001) or capabilities that impact firm performance, but how well the firms adapt and respond to the changing environment. Although extant literature has undersigned different connotations of the concept (Barreto, 2010; Sharma & Bhat, 2016), dynamic capabilities were originally defined as ‘the firm’s ability to integrate, build, and reconfigure internal and external competences to address rapidly changing environments’ (Teece, Pisano & Shuen, 1997, p. 516). Alternately, a dynamic capability is also connoted as the ability of a firm to develop, expand and rearrange its assets to achieve a desired objective (Allred et al., 2011; Helfat et al., 2007). Dynamic capabilities are acquired with sustained behavioural mechanisms through which organizations rearrange its execution policies for achieving maximum efficiency (Ambrosini, Bowman & Collier, 2009; Balodi & Basu, 2015; Zollo & Winter, 2002). Such capabilities often become the enabler of competitive advantage as they are quite difficult to imitate in the short run (Teece et al., 2007).
Consistent with the definition of Ponomarov and Holcomb (2009) (Table 1), we propose that SC resilience can be conceptualized as a dynamic capability. Other reasons that this conceptualization makes sense include: it satisfies the characteristics of a higher degree capability (Winter, 2003), it is successfully utilized for changing existing operational procedures (Zollo & Winter, 2002), it enhances assets utilization and alignment and it aids in identifying opportunities and threats in environment and utilizing the same for gaining competitive advantage (Teece, 2007). This establishes that resilience as a capability is ‘dynamic’ as it enables a firm’s supply chain to adapt or adjust to its changing environments.
SC Resilience and Supply Chain Relationships
A firm’s critical resource may span firm boundaries and may be embedded in inter-firm resources and routines (Dyer & Singh, 1998). Dynamic capabilities lead to other capability-based outcomes, such as resilience, due to greater sharing of resources and information in the presence of uncertainties (Lorenzoni & Lipparini, 1999). Collaborative associations developed on trust and commitment can decrease transaction costs (Zaheer et al., 1998) increasing positive performance implications. Thus, both resilience and supply chain performance will depend on a coordinated response of several supply chain partners to environmental uncertainties. This coordinated response, in turn, depends on the strength of relationships (or relational attributes) among the partners.
If a firm’s relationships with its partners are important to SC resilience, the question arises as which relational attributes are important. As indicated in the introductory section, we take recourse to commitment–trust theory (Morgan & Hunt, 1994). The commitment–trust theory argues that there is absence of an explicit ‘buyer–seller’ relationship in network alliances and they are more like partners to an exchange. The success of the alliance is largely dependent on relational attributes (e.g., trust, commitment, communication, cooperation etc.) leading to efficient relationships (Carter & Ellram, 1994). It must be noted that we do not adopt the entire model of Morgan and Hunt because our focus is on relationship trust and commitment in the context of SC resilience. The commitment–trust theory has been investigated in different contexts (Coote, Forrest & Tam, 2003; Friman et al., 2002; Goo & Nam, 2007; Lancastre & Lages, 2006). A few studies have examined the interface of commitment–trust theory and supply chain management, for example, Chen (2011) and Wu et al. (2012). Chen (2011) found a positive relationship among information sharing, information quality and information availability in the development of trust and commitment in supply chain relationships. More recently, Wu et al. (2012) applied the commitment–trust framework of Morgan and Hunt (1994) in the development of supply chain partnerships and found that higher levels of trust can result in improved interactions and aid in increasing interests of both parties, facilitate constant co-operation and communication, reduce uncertainties and reduce a partner’s propensity to leave.
Although, supply chain studies have adopted several relational attributes from the commitment–trust model, the inter-relationships among them have not been captured (Fynes, de Bu´rca & Marshall, 2004; Fynes, De Burca & Mangan, 2008; Fynes, de Bu´rca & Voss, 2005a; Fynes, Voss & de Bu´rca, 2005b). Of different variables, trust and commitment are the building blocks of supply chain relationships (Brinkhoff et al., 2015; Palmatier et al., 2013). Communication helps in building trust among supply chain partners (Sarker, Ahuja, Sarker & Kirkeby, 2011), and both trust and commitment are required for aligning the interests of supply chain partners, that is, cooperation (Morgan & Hunt, 1994; Sønderskov & Daugbjerg, 2011). Hence, we choose these four dominant relational attributes communication, trust, commitment and cooperation among others in exploring their contribution to SC resilience. Figure 1 represents the research model.
Objectives
The commitment–trust model of Morgan and Hunt (1994) have received prior empirical attention in supply chain management, for example, Chen (2011) and Wu et al. (2012). The importance of relational attributes although documented (Fynes et al., 2005a) their inter-relationships were yet to be explored in the context of SC resilience. Hence, we explore in this study regarding the significant relational attributes for SC resilience. Further, we also explore the inter-relationship among these relational attributes. Finally, we also explore how these inter-plays among the relational attributes affect SC performance through SC resilience.
Rationale of the Study
Communication
Communication is defined as ‘the formal as well as informal sharing of meaningful and timely information between firms’ (Anderson & Narus, 1990, p. 44). Communication is necessary for the exchange of important messages for providing a proactive response and recovery following a disruption. In fact, communication is a prerequisite for supply chain partners to harness their relational benefits (Luc, 2006) to the maximum extent and enable the supply chain to successfully restore operations in the event of a disruption. There are two aspects of communication behaviour that are important in relationships (Mohr & Spekman, 1994). First, the qualities of communication such as accuracy, timeliness, adequacy and credibility are important. Second, the form of information sharing or the extent to which critical, and sometimes proprietary, information is exchanged is also important. The quality of communication, information sharing and participation are all significant predictors of successful supply chain relationships (Mohr & Spekman, 1994). Hence, in the event of a disruption, greater the informal message sharing during post-disruption, higher the effectiveness of supply chains in successfully restoring operations. Accordingly, we have our first proposed relationship: R1: Communication positively influences SC resilience.
Cooperation
Cooperation refers to firms working together to achieve mutual goals (Anderson & Narus, 1990). Cooperation simply does not rule out the possibility of a conflict and it is argued that contradictory situations coexist along with cooperative associations (Frazier & Rody, 1991). During a disruption, partners must work together to achieve mutual goals of sustaining operations and ensuring business survival, even in the presence of conflicts over serious issues. Further, during contingencies, cooperation can help in various procedures, such as exchanging information on production schedules, new products/processes and value analysis, leading to reduced costs and improved processes (Landeros & Monczka, 1989). Unless the partners in a supply chain cooperate and collaborate during an emergency situation, the entire supply chain will not be successful in sustaining and safeguarding its operations (Cao & Zhang, 2011; Gligor & Holcomb, 2012; Ponomarov, 2012). Supply chain partners must act jointly and should be careful in aligning their interests during a disruption and the overarching interest is the survival of the business and the supply chain operations. Hence, cooperation among supply chain members is essential for the development of SC resilience. Accordingly, we propose the relationship: R2: Cooperation positively influences SC resilience.
Commitment
The willingness of trading partners to apply effort towards maintaining the relationship is referred to as commitment (Porter et al., 1974), especially in the face of problems and contingencies (Gundlach, Ravi & Mentzer, 1995). High levels of commitment develop the platform in which both parties to the exchange can realize joint goals without any opportunistic behaviour (Cummings, 1984). Committed parties are willing to invest in transaction-specific assets, demonstrating that they can be relied upon to perform essential functions in the future (Anderson & Weitz, 1992). This investment helps in stabilizing supply chain relationships and eliminating the uncertainty of continually searching and forming new relationships. A positive relationship exists between commitment and relationship success (Mohr & Spekman, 1994). Therefore, the culmination of commitment from all partners in a supply chain is necessary for developing any kind of capability, particularly during uncertainties.
Further, commitment of partners entails the continuation of relationships among supply chain partners and therefore acts as a precursor for the alignment of interests. Morgan and Hunt (1994) in their commitment–trust theory argued that: ‘a partner committed to the relationship will cooperate with another member because of a desire to make the relationship work’ (p. 26). Along similar lines, Deer (2012) undersigned commitment as an essential precursor of cooperation among partners acting in a network. Hence, we frame our next relationship: R3: Commitment positively influences cooperation among SC entities.
Trust
Trust is ‘the firm’s belief that another company will perform actions that will result in positive actions for the firm, as well as not take unexpected actions for the firm, that would result in negative outcomes for the firm’ (Anderson & Narus, 1990, p. 45). Different types of trust exists, namely, contractual trust, competence trust and goodwill trust (Fynes et al., 2005a). Zaheer et al. (1998) further distinguished between interpersonal trust and inter-organizational trust. Moorman, Deshpande and Zaltman (1993) refer to trust as the willingness to rely on an exchange partner in whom one has confidence. Morgan and Hunt (1994) referred to trust as ‘a firm’s belief in its partner’s trustworthiness and integrity’. Pruitt (1981) defines trust as the belief that a party’s word is reliable and that a party will fulfil its obligation in an exchange. This definition indicates a firm’s willingness to collaborate (Shukla & Rai, 2015; Singh & Srivastava, 2016).
Good communication is an important antecedent to mutual trust among the supply chain partners. While some studies indicate that the relationship between communication and trust may be debatable (De Ridder, 2006; Harry & Dennis III, 2006; Rosli & Hussein, 2008), we argue that the exchange of timely information in a network builds trust and supply chain relationships (Klimchak, Sherman, MacKenzie & Ward, 2013; Sarker et al., 2011). Consequently: R4: Communication positively influences trust among SC entities.
Zand (1972) argued that the absence of trust will prohibit information exchange and will hamper joint efforts to solve problems. In exchange relationships, the presence of trust will facilitate better stress management and adaptive capability (Williamson, 1985), and will motivate the members to continue their relationships for a long time (Morgan & Hunt, 1994; Tyler & Doerfel, 2006; Welch & Jackson, 2007). Further, it is argued that the alignment of interest (i.e., cooperation) among the different partners can be achieved more efficiently with the presence of trust among the members (Lundin, 2007). Jones and George (1998) argued trust as a beneficial factor for the development of teams and teamwork to be effective. Marcoz, Mauri, Maggioni and Cantù (2014) in the context of hospitality industry undersigned trust to enhance cooperation among participating parties. Accordingly, we posit trust to positively influence commitment and cooperation. This leads to our next segment of relationships: R5: Trust positively influences commitment among SC entities; and, R6: Trust positively influences cooperation among SC entities.
Supply Chain Resilience
As discussed in the previous section, SC resilience is defined as the adaptive capability of the supply chain to prepare for unexpected events, respond to and recover from disruptions. It leads to maintaining continuity of operations at the desired level of connectedness and control over structure and function (Ponomarov & Holcomb, 2009). Supply chain performance is defined as the performance of the various processes included within a firm’s supply chain (Srinivasan et al., 2011). Supply chain models commonly use two different performance measures: cost and customer responsiveness. Examples of measures used to assess supply chain performance of a firm include supplier performance (Davis, 1993), customer satisfaction (Christopher, 1994), inventory costs, number of on-time deliveries, product availability performance and customer response time (Beamon, 1999). Supply chain disruptions can impact supply chain performance adversely when measured in terms of lost sales and damaged reputations.
In the realm of relationship studies in supply chain, Fynes et al. (2005a) found supply chain relationship quality to have a positive impact on supply chain performance. Accordingly, the definition of resilience itself carries the implication of improving performance in the face of disruptions. Further, resilience being a dynamic capability (Teece, 2007) has the ability to cause supply chains to modify operating routines, reconfigure existing processes and adapt to newer environments. The essence of the dynamic capability framework is that firms must possess capabilities that can enable them to positively respond to the changes in their environment and hence such capabilities are ‘dynamic’ and bear positive performance implications (Eisenhardt & Martin, 2000; Lin & Wu, 2014; Makkonen et al., 2014). Thus, it can be argued that SC resilience (as a dynamic capability) developed based on several relational attributes will have positive performance implications. Consequently, R7: SC resilience positively influences SC Performance.
Methodology
Data Collection and Sample Demographics
The study involved proposed relationships among latent variables. Such variables and associated relationships mainly require perceptual responses for validation. Accordingly, we resorted to survey-based perceptual response collection using online web resources. Considering the supply chain as the main unit of analysis, we considered focal manufacturers and their interfaces with SC entities. The study developed the measurement instrument through appropriate pretesting with a sample of SC managers selected from a contact list. The contact list consisted of 1,500 SC professionals in various posts in India and contained their respective email ids. This was procured from an Indian Marketing Research Firm (the firm wanted to remain anonymous).
The SC professionals were mostly higher officials involved directly or indirectly in manufacturing, transportation, distribution and supply chain operations in various sectors. The pretesting phase resulted in subsequent modification of few of the measurement items. While the contact list consisted of several SC professionals, we selected the final respondents (and also respondents for pre-testing) based on two criteria: (a) the SC professional is having > 5 years of work experience directly or indirectly in SC sector and (b) the current designation work experience of the SC professional is at least 2 years. These two criteria developed a final list of 916 potential contacts. The participating respondents were requested to respond based on their learning and expertise at their respective positions. Table 1 shows the sample profile.
Sample Profile
The survey data were collected in several steps. At first, a formal invitation to participate was mailed to the final list of contacts. Subsequently, we mailed two gentle reminder requests spread over a gap of 2 weeks. Of the total 916 emails sent, 84 emails were returned as undeliverable. For 167, partially complete responses were received, giving a response rate of 20.07 per cent (167/832). However, for the final analysis, we retained only complete responses. Thus, the final sample size was 155 resulting in a final response rate of 18.62 per cent (155/832).
Non-response Bias
Using the guidelines given by Armstrong and Overton (1977), the study was tested for non-response bias. The results showed negative dominant differences between responses received early compared to late ones. The sample responses were subsequently divided into two categories based on their early and late classifications. Subsequent application of Mann–Whitney U-tests revealed negligible differences between the categories (p > 0.05).
Common Method Bias
Because of single response collection from every SC entity, the study attempted a formal examination of common method bias (Podsakoff, MacKenzie, Lee & Podsakoff, 2003) and it revealed seven factors with Eigen values above one, explaining 71.6 per cent of cumulative variance. The first factor explained only 25.9 per cent variance. We performed a second test of common method bias using Harman’s test for unique factor as suggested by Flynn, Huo and Zhao (2010). Subsequently, fit indices of chi-sq/df = 15.6; NFI = 0.57; CFI = 0.59 and RMSEA = 0.18 suggest that the obtained factor model is not desired and accordingly we conclude that there is negligible common method bias.
Survey Instrument
The study involved relationships among latent variables. All the latent variables have proven measurement instruments, and the same were adapted suitably to suit the context. Initially, we adopted all the items for measuring the constructs employed in the study. Before launching the actual survey, we pre-tested the instrument with 24 supply chain managers mainly from automobiles, electrical equipment, chemical manufacturers, plastic products manufacturers and wood products manufacturers. Based on the results of the pre-test, some of the items were removed. For example, communication was supposed to be measured with four items initially but one item was removed (based on pre-test) that targeted to ask the respondents if they share proprietary information with one another in supply chains. A total of 22 final survey items (Table 2) were used to measure independent and dependent variables in the study.
Control Variable
Following earlier studies in management research, our study adopted firm size (natural logarithm of employee number) as a control variable.
Measurement Model Assessment
We deployed Partial Least Squares (PLS) checking the items for their reliability and exploring the truth of the proposed relationships. The PLS as a Structural Equation Modeling (SEM) methodology utilizes an approach focused at components for parameter estimation. The PLS allows researchers to model formative constructs for evaluating the estimated parameters with an optimal sample size. The optimal sample requirement in PLS is 10x that of items intended to measure the largest latent variable. We resorted to bootstrapping procedure (with 500 sub-samples) that comes as a standard package in SmartPLS 2.03m3 to overcome the need for a significance test. This helped in evaluating the path coefficients and other parameters. The study resorted to two step analysis. While the former step executed tests of reliability and validity, the latter evaluated discriminant validity.
Following Chin (1998), we assessed reliability and found that for each and every latent variable it is greater than the recommended minimum, that is, 0.7. Further, we utilized several criteria for evaluating convergent validity: (a) item loading > 0.70 and statistical significance, (b) composite reliability > 0.80 and (c) average variance extracted (AVE) > 0.50 (Fornell & Larcker, 1981). In addition, the study evaluated discriminant validity with another procedure: square root of AVE for each latent variable should be greater than its correlations with other latent variables (Fornell & Larcker, 1981). Referring to Table 3, standardized item loadings range from 0.797 to 0.907; composite reliabilities range from 0.894 to 0.935, and average variance extracted (AVE) values range from 0.712 to 0.791. In Table 4, the square root of AVE for every factor is greater than other inter-factor correlations. Hence, these results show a highly acceptable level of reliability, convergent and discriminant validity. Further, Table 5 demonstrates cross loadings of all items as other quality criteria. Table 4 gives the discriminant validity. Table 5 gives the cross loadings of the items.
Measurement Items
Analysis
We resorted to PLS for exploring the value of the coefficients pertaining to our proposed relationships. A two-stage procedure was adopted for estimation processes (Chin, 1998). In the first stage, the value of path coefficients with their significance was evaluated in PLS. Second, we evaluated R-square for the dependent variables for finding the prediction power of our model. Table 6 and Figure 3 summarizes the results of hypotheses testing.
The empirical findings therefore undersigned communication and cooperation as essential enablers of SC resilience. In addition, it also established that trust and commitment are also necessary for building SC resilience as they ensure that partners in a supply chain cooperate when needed for developing such dynamic capability (e.g., SC resilience). The validated empirical model explained 26.6 per cent of the variance in SC resilience which accounted for 22.1 per cent of the variance in supply chain performance. Communication accounted for explaining 10.2 per cent of the variance in trust while trust alone explained 24.8 per cent of the variance in cooperation and 17.6 per cent of the variance in commitment.
Convergent Validity
Discriminant Validity
Cross Loadings of the Items
Summary of Testing
TR=trust
CMT=commitment
COOP=cooperation
RES= supply chain resilience
PERF=supply chain performance
Post-hoc Analysis of Alternative Models
In the current investigation, we also tested two alternate models in order to lend support to our main model analysed earlier. The first alternate model tests the following additional paths: Trust to SC Resilience and Commitment to SC Resilience. However the path coefficients are negative and not statistically significant (Trust à Resilience: –0.05, t = 0.419; Commitment à Resilience: –0.079, t = 0.712). Further, except the above two paths, all the remaining paths were positive and statistically significant. Figures 4 and 5 show PLS testing of alternate models 1 and 2.
Next, in the second alternate model, we incorporated the above two paths in addition with the earlier proposed paths in the main model. This time too it was found that except all other paths, only the above two paths were negative and not statistically significant (Trustà Resilience: –0.052, t = 0.400; Commitment à Resilience: –0.078, t = 0.683). Hence, these alternate model results showed that the proposed main research model is robust and the corresponding empirical results may hold well in many different contexts. Tables 7 and 8 summarize the path loadings and the t-values for both the alternate models.
The above alternate models gave strong support to the original proposed model and its associated findings. The next section discusses the findings and implications.
Results of Alternate Model 1
Results of Alternate Model 2
Conclusion
Our research has investigated the role of relationships among supply chain partners on the ability of supply chains to recover from disruptive events (resilience). Relationships are not the only important antecedents to resilience; others such as creating redundancies, agility and flexibility are important as well Tukamuhabwa et al. (2015). However, given the empirical nature of our investigation, investigating all factors under a single model might prove to be complicated and challenging. Our research fills the gap on empirical testing of the role of supply chain relationships on SC resilience. It parallels the empirical investigation by Wieland and Wallenburg (2013) who empirically explored the influence of relational competencies, namely, communication, cooperation and integration on agility and robustness of supply chains. Our investigation also explored the influence of SC resilience on supply chain performance. The empirical data provided support that communication and cooperation directly affect SC resilience. It also showcased the inter-relationships among communication, trust, commitment and cooperation consistent with the commitment–trust theory proposed by Morgan and Hunt (1994), thereby, confirming the importance of the indirect effects of trust and commitment. Post-hoc testing of alternate models confirms the suitability of commitment–trust theory for our context. This is a significant contribution of our research in terms of building, integrating and extending existing theoretical frameworks.
Managerial Implications
Our study investigated the direct effect of the communication and cooperation among supply chain partners on SC resilience as well as the indirect effects of communication through trust and commitment. We also investigated the subsequent effect of resilience on supply chain performance. Our results show that both the direct and indirect effects are significant in the model. Post-hoc analysis confirms the superiority of the model over other possible configurations of hypothesized direct and indirect effects. The particular configuration of paths among the relational antecedents of communication, trust, commitment and cooperation follows directly from Morgan and Hunt’s (1994) commitment–trust theory. Our goal was to confirm the role of this theory in the context of SC resilience and hence we remained faithful to this structure. Our results confirm the applicability of the theory.
The path from communication to resilience was positive and significant (0.360; t = 3.041), which argues that exchange of formal and informal messages among the supply chain members is critical during emergencies so that necessary actions can be taken by every member so as to restore supply chain operations at the best possible state. This is the essence of SC resilience. Hence, supply chain managers must create an atmosphere of open participation for its supply chain members. This will also create an opportunity for sharing and exchanging valuable risk management knowledge and expertise. Wisner, Tan and Leong (2015) argued that members in a value chain should make it a standard practice to share information and expertise on a regular and frequent basis. Further communication has been found to have a dominant role in enhancing and improving supply chain relationships (Gligor & Autry, 2012; Wieland & Wallenburg, 2013). In line with earlier investigations, our study also establishes that communication is a precursor to building supply chain relationships through improving mutual trust among supply chain partners.
Supply chain members and practitioners therefore should encourage flow of bi-directional information and exchange of the same for increasing transparency and trust among the supply chain partners. Unless a person can trust another in the network, the benefits of collaborative engagement may not be harnessed (Cao & Zhang, 2011). This is shown by the positive and significant path coefficient of the path between communication and trust (0.320; t = 2.602). Similarly, the path between trust and commitment is positive and significant (0.420; t = 4.052) implying supply chain managers to work with their partners to increase transparency in each and every operation as the same will enhance mutual trust and will make the partners more committed towards their responsibilities (Morgan & Hunt, 1994; Welch & Jackson, 2007). Again, it is also empirically proved that if the members are more committed towards their duties and responsibilities or if they are more willing to sustain their supply chain relationships, it will be easier for the focal firm to align the interests of its supply chain members. This is shown as the path coefficient of the path from commitment to cooperation is positive and significant (0.259; t = 2.611). A similar observation is that the more one supply chain partner trust another, the more easily they will cooperate with each other. This is shown by the positive path coefficient of trust to cooperation path (0.330; t = 2.800). This implied that supply chain managers must understand that supply chain partners should be able to trust each other and then only they will be willing to sustain their relationships which will eventually help them to cooperate with each other. All these will lead to the development of SC resilience as the path from cooperation to resilience is positive and significant (0.281; t = 2.668). This proves that effective alignment of interests of supply chain parties is detrimental for the development of SC resilience. Finally, the path from resilience to supply chain performance is also positive and significant implying that the relational antecedents are effective in developing SC resilience which can be transformed into positive performance implications (Ponomarov, 2012; Wieland & Wallenburg, 2013).
Future Research
Numerous research avenues still exist for future research. As stated earlier, other antecedents to resilience such as existence of redundancies, flexibility, agility, etc., may be progressively added to this model. These require larger empirical studies. Research can also concentrate on investigating other relational antecedents of SC resilience such as power, inter-dependence, reciprocity etc., using a combination of theoretical lenses of social exchange theory, relational view etc. Second, future research should focus on studying how the inter-relationship among these other relational attributes (viz. power, reciprocity etc.) holds good for other supply chain capabilities like agility, flexibility and robustness. Our model can be used to investigate the influence on firm performance, measured along a service perspective as proposed by Stank et al. (1999). Further investigations can test the proposed model with secondary data for analysing supply chain (overall) performance. All such studies face obvious challenges in difficulties in obtaining data and controlling model complexity, but would lead to greater understanding of important issues in supply chains.
Footnotes
Acknowledgements
The authors are grateful to the anonymous referees of the journal for their extremely useful suggestions to improve the quality of the article. Usual disclaimers apply.