Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects

DECSUP-12324; No of Pages 15 Decision Support Systems xxx (2013) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Decision Support Systems journal homepage: www.elsevier.com/locate/dss

Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects☆ Hyun-Soo Han a, 1, Jae-Nam Lee b,⁎, Jae Uk Chun b, 2, Yun-Weon Seo c, 3 a b c

School of Business, Hanyang University, 17 Haengdang-Dong, Seongdong-Gu, Seoul 133-791, Republic of Korea Korea University Business School, Anam-Dong, Seongbuk-Gu, Seoul 136-701, Republic of Korea Center for Information Technology Studies, Security Management Institute, Republic of Korea

a r t i c l e

i n f o

Article history: Received 21 November 2012 Received in revised form 7 February 2013 Accepted 18 March 2013 Available online xxxx Keywords: IT outsourcing Client IT capability Vendor IT capability Outsourcing success Resource-based view Complementary perspective PLS

a b s t r a c t This study investigates the direct and complementarity effects of client and vendor Information Technology (IT) capabilities on the success of IT outsourcing. Concretizing the concepts of client and vendor IT capabilities from the resource-based view, we emphasize the importance of complementarity in IT outsourcing. It was hypothesized that the complementarity between client and vendors IT capabilities adds value to outsourcing success beyond their individual effects. We also hypothesized that the increase rate of complementarity is higher when the IT capability levels of both parties are similar (i.e., either high or low) than when those of both parties differ. The proposed hypotheses were tested by using 267 client–vendor matched data, in which client IT capability was measured by members of vendor firms and vice versa. Results indicate that client and vendor IT capabilities are significant factors in outsourcing success and that the complementarity between client and vendor IT capabilities significantly influences outsourcing success. The increase rate of complementarity is higher when the IT capability levels of both parties are similar than when those of both parties differ. The results help deepen our understanding of the bilateral perspective of outsourcing success and support managers in making better outsourcing decisions in selecting clients and vendors. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Over the last decade, geographical growth has become one of the most significant developments in meeting the Information Technology (IT) needs of organizations in IT outsourcing aside from financial growth [24,41]. IT outsourcing, which is defined as handing over the management of IT assets, resources, and activities for required results to one or more external vendors, has become a commonly accepted and growing business practice involving important decisions [27,70]. An emerging stream of IT outsourcing research has focused on managing the relationship between clients and vendors [36,41]. By using various contextual variables from different theoretical perspectives, extensive empirical testing results have reported the importance of outsourcing relationships as the key antecedent of outsourcing success

☆ This work was supported by National Research Foundation of Korea Grant funded by the Korean Government (KRF-2009-327-B00207). ⁎ Corresponding author. Tel.: +82 2 3290 2812; fax: +82 2 922 7220. E-mail addresses: [email protected] (H.-S. Han), [email protected] (J.-N. Lee), [email protected] (J.U. Chun), [email protected] (Y.-W. Seo). 1 Tel.: +82 2 2220 1822; fax: +82 2 2220 1886. 2 Tel.: +82 2 3290 2838; fax: +82 2 922 7220. 3 Tel.: +82 10 5078 9350.

[23,41]. Previous studies support the importance of relationship of both parties in IT outsourcing because successful outcomes are determined not by clients or vendors but by both organizations. However, despite the efforts of prior studies, several weaknesses have been reported. First, client IT capability in outsourcing relationships lacks extensive research. The majority of previous studies examined IT outsourcing relationships and their success only from social and contextual aspects and neglected the importance of client IT-related capability in outsourcing decisions and relationship management [19,36]. Second, most studies overlooked the importance of vendor IT capability, despite the increasing importance of vendors in the outsourcing relationship [43,62]. Most previous studies concentrated mainly on client firms that choose to outsource. Several studies suggested that vendors must have IT competence or capability [43,57], but these studies are conceptual and exploratory without empirical investigations. Finally, although most previous studies emphasize the necessity of understanding both parties to obtain better outsourcing outcomes, client and vendor IT capabilities have never been simultaneously considered in a study [30,36]. For example, Han et al. [29] integrated the IT capability of a firm with social exchange theories to examine the causal structure of the capability, process, and relationship in IT outsourcing. However, their study was conducted from a client perspective only. Thus, the notion of complementarity between client and vendor IT capabilities must be extended to outsourcing

0167-9236/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.dss.2013.03.003

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

2

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

relationships and success. Complementarities between activities can be viewed as existing if, and only if, increasing the level of an activity (i.e., client IT capability) leads to a higher marginal return from increasing the level of other activities (i.e., vendor IT capability) [10]. To fill these research gaps, this study attempts to answer the following questions from the perspective of complementarity: 1) What are the roles of client and vendor IT capabilities in outsourcing success?; 2) Does outsourcing success originate from the complementarity between these capabilities?; and 3) What kind of complementarity between these capabilities demonstrates better performance? To answer these questions, we first conceptualized client and vendor IT capabilities based on prior studies that focused on the Resource-Based View (RBV). We then hypothesized the relationships between these factors and outsourcing success from the perspective of complementarity. This study may be one of the earliest studies to simultaneously consider client and vendor IT capabilities, as well as their interaction effect on outsourcing success from the perspective of complementarity. The proposed hypotheses were then tested by using a sample of 267 matched responses collected from both client and vendor firms in Korea. We believe that this study can contribute to our understanding of the bilateral perspective of outsourcing success and provide useful managerial guidelines to make better outsourcing decisions in selecting clients and vendors. 2. Theoretical development Outsourcing success is defined as the level of fitness between client needs and outsourcing outcomes [41]. Outsourcing is motivated by strategic, economic, and technological benefits; hence outsourcing success can be assessed in terms of the attainment of these benefits [27,41]. Grover et al. [27] defined strategic benefits as the ability of a firm to focus on its core business by outsourcing routine IT activities. Economic benefits refer to the ability of a firm to utilize the expertise and economies of scale of the human and technological resources of the service provider, as well as manage cost structure through unambiguous contractual arrangements. Technological benefits refer to the ability of a firm to obtain cutting-edge IT and avoid the risk of technological obsolescence caused by dynamic changes in IT. Although client firms have high IT capabilities, they can still consider IT outsourcing to enjoy its different benefits (e.g., cost savings and/or new business opportunities). Clients and vendors generally have the same objectives (i.e., making outsourcing projects successful) but have different roles and concerns [57]. Client firms are concerned with the selection of the best vendor, perceived replaceability of the vendor, and perceived switching costs and risks [70]. Replacing a current vendor with another is very difficult for the client firm because of the costs and risks of switching. Therefore, the major concern of client firms is the IT capability of their vendors in successfully conducting outsourcing projects. By contrast, vendor firms believe that outsourcing success is mainly dependent on the ability of clients to effectively acquire, distribute, and leverage IT-related resources [19,43]. To create a successful outsourcing project, client IT capability should be aligned with that of vendor firms and vice versa, even though clients and vendors have different roles in outsourcing projects [43,57]. Capability refers to the ability of organizations to integrate and deploy valued resources [2]. From the perspective of capability, client IT capability can be defined as the ability to which a client firm acquires, distributes, and leverages IT-related resources or assets. By the same token, we consider vendor IT capability as one of the most critical organizational abilities for improving IT outsourcing service performance. Thus, we define vendor IT capability it as the ability to which a vendor firm identifies, responds, and manages IT-related client needs. Prior studies on client and vendor IT capabilities have been mostly conceptual; thus, formal constructs have not been clearly defined. However, their implications on the causal relationship between client and vendor IT capabilities and outsourcing performance are evident

[13,56]. Client and vendor IT capabilities, which include organizationspecific routines, processes, skills, and resources [32], are necessary to achieve outsourcing objectives. Client and vendor IT capabilities are more than specific sets of sophisticated technological functionalities [43,49]. These capabilities are used for managing and leveraging the technologies of the enterprise to sustain and create competitive advantage. Therefore, the IT capabilities of both parties should be considered from the onset of outsourcing project to ensure effective management [37]. Thus, the IT capabilities of both parties play critical roles in the achievement of predefined outsourcing objectives. This study consolidates client and vendor IT capabilities as critical factors in outsourcing success. We assume that client and vendor IT capabilities enable both parties to interact and communicate effectively with each other, thereby leading to better outsourcing performance. On one hand, client IT capability ensures that IT resources are appropriately acquired, distributed, and managed to meet organizational requirements. On the other hand, vendor IT capability affords benefits to clients not only by providing the core skills required to build high-quality information systems but also by helping build good relationships with clients [43]. The complementarity between client and vendor IT capabilities results in a more positive and stronger IT capability effect on outsourcing performance. However, complementarity does not denote the transformation of a negative (or insignificant) relationship to a positive one and vice versa. Therefore, the presence of positive relationships between client IT capability and outsourcing success, as well as between vendor IT capability and outsourcing success, should be prerequisites to understand the complementary relationship between client and vendor IT capabilities. The next section considers the nature and constituents of client and vendor IT capabilities. 2.1. Client IT capability Most prior studies investigated client IT capability primarily from the RBV of firms [8]. According to prior literature, client IT capability consists of IT resources or assets and its management ability. Bharadwaj [13] defined the specific IT resources of client firms as IT infrastructures, human IT resources, and IT-enabled intangibles. Ross et al. [60] classified client IT capability as a reusable technology base (technical asset), competent IT skill (human IT asset), and intimate relationship between the management of IT and business units (relationship asset) of client firms. Lee et al. [42] stated that industries demand a cadre of IT and technical skills in business operations, management, and interpersonal relationships to lead organizational integration and process reengineering activities effectively. They emphasized the capability of IT managers to appreciate business needs, collaborate with functional managers effectively, coordinate IT activities with the support of other functional managers, and anticipate future business needs. Feeny and Willcocks [19] characterized nine distinct IT functions as a set of core capabilities: IT governance, business systems thinking, relationship building, technical architecture design, making technology work, informed buying, contract facilitation, contract monitoring, and vendor development. They also mapped the nine capabilities with the skills and knowledge requirements framework proposed by Lee et al. [42]. Given that client firms possess a variety of IT resources, managers must carefully choose resources that should be prioritized in creating synergetic IT capabilities. We used the RBV, which explicitly recognizes the importance of intangibles [13], to identify measurable constructs that reflect client IT capability necessary for outsourcing success. We identified the three primary strategic areas of IT capability that affect outsourcing success: technology management capability, organizational relationship capability, and vendor management capability. These are the same constructs found in the study of Han et al. [30]. These constructs were conceptualized based on the study of Lee et al. [42], as well as the RBV-grounded studies of Feeny and Willcocks [19] on the operationalization of client IT capability [54]. These constructs were

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

defined by using Valuable, Rare, Inimitable, and Organizational (VRIO) framework to assess the performance implications of resources [14]. Furthermore, we used the resource categorization framework of Mills and Platts [51] to determine whether the three given constructs comprehensively reflect client IT capabilities. First, technology management capability not only refers to the technical knowledge and skills needed to develop IT applications [42] but also to where and how to deploy these knowledge and skills effectively and profitably to meet strategic business objectives [47]. Second, organizational relationship capability reflects the level of relationship between IT and business groups within organizations, which enables businesses to engage constructively in IT issues [12]. Finally, vendor management involves the capability to look beyond existing contractual arrangements and explore long-term potentials to create win–win situations [19]. These capabilities are also evident in corporate-level managerial practices, policies, and processes [25]. Collectively, technology management capability, organizational relationship capability, and vendor management capability complement one another. Based on the categorization of Mills and Platt [51], technology management capability, organizational relationship capability, and vendor management capability closely relate to the overlapping concepts of knowledge resources, network resources, and system and procedural resources. These three capabilities pertain to human and organizational categories, rather than to tangible financial and physical resources; thus, they are not easily imitable. Moreover, they are organization-specific (i.e., they are non-substitutable, valuable, and rare), ensuring IT outsourcing success. Tangible IT resources, such as hardware and network infrastructure, are not quite distinguishable in reflecting client IT capability; hence, these resources are excluded from our research context. Cultural and dynamic capability resources that are related to the IT of firms should also be reflected in the three capabilities. The technology management capability of clients can convince vendors that the goals and concerns of client firms are achievable. This capability also helps clients achieve their goals [12,19]. The knowledge of client firms regarding IT trends can encourage vendors to share valuable information. Firms that have clear IT standardization and blueprints are in a better position to communicate effectively with vendors, thereby ensuring successful outsourcing. Closer inter-organizational relationship can promote more information sharing between clients and vendors not only to reflect business needs correctly but also to achieve business goals. The ability of client firms in vendor management helps vendors successfully implement their services in satisfying client needs [40]. In addition to contractual obligations, formalized outsourcing management processes and work evaluation principles cultivated by effective vendor management are required to create outsourcing benefits [19]. Therefore, the IT capability of a client firm ensures that clients not only manage internal IT resources appropriately but also communicate and share outsourcing-related information and issues with their vendors effectively to obtain better outsourcing performance. That is, the greater client IT capability is associated with higher levels of outsourcing success. The following hypothesis is thus proposed: H1. The higher the level of client IT capability to reflect technology management, organizational relationship, and vendor management capabilities, the higher the level of outsourcing success. 2.2. Vendor IT capability As outsourcing markets grow and the practice of outsourcing matures, most firms increasingly believe that outsourcing vendors will eventually deliver value in the long term [43]. However, variations in outsourcing outcomes call for an in-depth investigation on how vendors can create value for clients in IT outsourcing. Generally, vendors have different predispositions and resources, thus resulting in

3

varied outcomes even in similar outsourcing projects [57]. This concept is called vendor IT capability. IT outsourcing achieves performance excellence by capturing the IT competencies and resources of vendors [42]. Thus, vendor IT capability should be distinct from that of client firms [9]. Despite growing interests in vendor IT capability, prior studies have paid only slight attention to vendor IT capability and how such capability generates value in outsourcing projects. Most studies discuss corporate IT asset and skill requirements for IT personnel. For example, Kim and Chung [37] explained that vendor IT capability includes experience, track record, technical competence, and financial status. Despite numerous studies on IT resources and capabilities that affect business performance, only a few studies have holistically explored vendor IT capability. One exception is the study of Levina and Ross [43], which explored value proposition from the perspective of vendors through the case study approach. They suggested that client satisfaction is achieved when the application of core competencies to outsourcing projects is enabled by a healthy client–vendor relationship, which is primarily influenced by the expertise of vendors in managing the client–vendor interactions. Capabilities are enhanced through the firm-wide experience of vendors in controlling a large number and variety of projects. Moreover, the ability of vendors to satisfy customer needs develops their reputation. Melville et al. [49] posited that IT business value from the complementary resources of client firms improves business processes or process performance, which ultimately influences organizational performance. Meanwhile, these outsourcing processes are influenced by the capability and environment of partner firms. Levina and Ross [43] identified three major areas in the core competence of vendor firms: personnel development, methodology development and dissemination, and customer relationship management. They considered these core competencies as key enablers not only in creating and maintaining the competitiveness of vendors but also in realizing the outsourcing objectives of clients. The present study adopts this classification to conceptualize vendor IT capability as personnel capability, methodology capability, and client management capability, which mostly pertain to human and organizational resource categories rather than physical or financial categories. First, personnel capability reinforces the quality of IT outsourcing results by ensuring that staff members are held accountable for meeting contractual obligations. Thus, personnel capability includes the IT knowledge, ability, and skills of personnel to provide effective IT services in outsourcing contracts. Second, while effective interpersonal skills can help establish proper customer expectations, the methodology capability of vendors is necessary for the consistent delivery of solutions to client problems through standardized outsourcing processes and systematic problem solving. This methodology can also improve the quality of client–vendor relationship by defining and standardizing best practices that lead to significant operational improvements and efficiencies in outsourcing projects. Finally, client management capability helps establish proper customer expectations for reducing project uncertainty by sharing the status of ongoing projects, promoting the understanding of participant roles, and providing valuable comments and feedback, thereby gaining more value from outsourcing projects. Collectively, these three sub-dimensions complement one another and form the overall vendor IT capability in outsourcing services. The personnel, methodology, and client management capabilities of vendor firms in IT outsourcing fulfill the VRIO criteria because they are critical to the successful implementation of IT outsourcing [48]. Moreover, these capabilities are intangible and thus not easily obtainable from the outside. From the resource category perspective of firms [51], personnel and methodology capabilities are related to knowledge as well as to system and procedural resources. Client management capability is associated with network resources as well as with system and procedural resources. Cultural and dynamic

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

4

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

capability resources that are related to vendor IT capability should be reflected on the three capabilities. Therefore, these three capabilities comprehensively reflect vendor IT capability. These capabilities also enable vendor firms to create synergetic capabilities and renew depreciating and obsolete capabilities [43]. Therefore, a vendor that possesses distinct IT capabilities can become the dominant player in the outsourcing business [57]. Many studies on inter-organizational relationships consider capability as a key characteristic in ensuring the success of projects, thus indicating that vendors should have the capability to keep abreast with everchanging technologies and maintain higher quality services and relationships [37]. Client satisfaction in an outsourcing project is influenced by vendor IT capability in identifying client IT needs, understanding client businesses, and managing client relationships [43]. Thus, we propose the following hypothesis: H2. The higher the level of vendor IT capability to reflect personnel, methodology, and client management capabilities, the higher the level of outsourcing success. 2.3. Complementarity between client and vendor IT capabilities The term complementarity was first used to describe the phenomenon of flexible manufacturing. Investments and structural changes in design and manufacturing technology are not made independently at each margin but are closely coordinated [50]. Thus, complementary activities tend to co-occur or be related to one another and have reciprocal positive effects on one another [5]. Complementarity causes the strategic value of the relative magnitude of each asset to increase with the increasing relative magnitude of other strategic assets [2]. Therefore, complementarity is a gestalt phenomenon in which the effect of the whole exceeds the combined effects of the parts [60]. Organizational capabilities and complementarities intersect in two ways. First, organizational capabilities, unlike individual capabilities, automatically entail complementarities because of the pattern of activities they represent [45]. Second, complementarities impart the inimitability necessary for organizational capabilities to become sources of competitive advantage through patterning or interdependence [46]. Thus, organizational capabilities are constituted by the “reciprocal positive effects” of complementary activities. Organizational capabilities also manifest the superadditive effects of complementary activities. Although the concept of complementarity is mainly used to understand the synergetic relationships among activities in an organization (e.g., IT investment with organizational structure and processes [33]), it is also applied to inter-organizational contexts. For example, Saxton [63] suggested that the success of inter-organizational collaborations is a function of partner characteristics. Sarkar et al. [61] examined the effect of inter-firm resource complementarity on the performance of alliances. They conclude that allying with firms with complementary capabilities is likely to ensure the success of the inter-organizational project, but making an effective outsourcing decision to choose the appropriate partner is more important to achieve the benefits of complementarity. In a similar vein, Levina and Ross [43] argued that the benefits of outsourcing projects can be realized through the complementarity between the core capabilities of clients and vendors. Therefore, we believe that the complementarity between client and vendor IT capabilities is one of the main indicators of outsourcing success and an important driver of outsourcing decisions. However, Harrison et al. [31] indicated that the existence of complementary resources does not always ascertain synergy; these resources must be effectively integrated and managed to realize synergy. When the complementary resources of client and vendor IT capabilities are well integrated, a synergetic effect generated from effective and efficient communication, knowledge sharing, co-relationship management, and so forth, can be expected [46]. Accordingly, a joint implementation of several activities

between the client and the vendor may result in better IT outsourcing performance. Positive reinforcement between client and vendor IT capabilities can enhance the abilities of both parties in making their outsourcing projects more successful and mutually beneficial. The effective integration of client and vendor IT capabilities will result in synergistic complementarity with regard to outsourcing success. Thus, we posit the following hypothesis: H3. The complementarity between client and vendor IT capabilities adds value to outsourcing success, in such a way that vendor IT capability augments the positive effect of client IT capability on outsourcing success, and vice versa. The complementarity perspective specifically advocates that the best complementarity of capabilities can be achieved when all parties have the necessary capabilities. Synergy can be maximized if the capability of one party is well balanced with that of the other party. Furthermore, the low capability of one party will hinder the complementarity of the other party's high capability. Prior studies that examined inter-organizational relationships from a consistency perspective indicate that organizations with similar skill levels, cultures, and strategic directions demonstrate greater performance than those with dissimilar ones [26,52]. If we apply the consistency perspective to the context of this study, we can argue that the increase rate of complementarity will be greater when the IT capability levels of both the client and the vendor are either high or low than when client IT capability is high and vendor IT capability is low or vice versa. Hence, we develop the following hypothesis: H4. The increase rate of complementarity between client and vendor IT capabilities is higher when both parties have high or low IT capability levels than when one party has high IT capability level and the other party has low IT capability level. 3. Research design and methodology 3.1. Development of measures The survey instruments for this study were developed either by adapting existing measures to the research context or by converting the definitions of constructs to a questionnaire format. Specifically, variables concerning client IT capability were measured by using the items of Han et al. [30], which were based on those of Lee et al. [42], Bassellier et al. [11], Nelson and Cooprider [53], and Feeny and Willcocks [19]. Variables related to vendor IT capability were assessed by adapting the measures presented by Levina and Ross [45], Lee et al. [42], and Bassellier et al. [11]. Finally, outsourcing success was measured in terms of the strategic, economic, and technological outsourcing gains developed by Grover et al. [27] and used by Han et al. [30]. A five-point Likert-style questionnaire was developed for all measures. However, the concept of complementarity was not directly measured from respondents. Instead, the interaction effect between client and vendor IT capabilities was tested to minimize respondent bias, similar to prior studies (e.g., [65,72]). A pre-test and a pilot test were conducted prior to the main survey. First, the initial version of the questionnaire was pre-tested with ten faculty members and ten practitioners who reviewed each item to improve content and construct validities. After the content of the survey instrument was polished, backward translation (i.e., translation of the material from English to Korean, vice versa, after which versions were compared to resolve discrepancies) was used to ensure consistency between the Korean and original English versions of the instrument [64]. Second, a pilot test was conducted with 35 outsourcing project teams in three global IT outsourcing firms located in Korea. The interviewees included the managers of each outsourcing project. The purpose of this step was to refine diverse questionnaire items that require considerable

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

effort and time to answer. Using the gathered data, factor analysis was conducted, and the reliability of the instrument was assessed using Cronbach's alpha. The resulting alpha values ranged from 0.790 to 0.920, which were acceptable for pilot tests. The research model and variables were refined with reference to the results of the pre- and pilot tests. Multiple phases of instrument development resulted in the significant refinement and restructuring of the survey instrument and the establishment of the initial face validity of the measures [55]. As presented in Table 1 and Appendix A, the final number of items used in this study is 50: 15 items for client IT capability, 13 items for vendor IT capability, 9 items for the outsourcing success of clients firms, and 13 items for general questions related to the respondents and their organizations. To account for extraneous sources of variation in the success of outsourcing benefits, this study incorporated outsourcing project size, project type, and prior outsourcing relationship as control variables. The importance of the outsourcing relationship between clients and vendors may differ depending on the size of the outsourcing project, as measured by the actual monetary amount stated in the contract [4]. Project type was classified into IT consulting service, application development, application maintenance, network installation and maintenance, data center operation, and others [39]. The existence of prior outsourcing relationships between a client firm and a vendor firm may have some influences on the degree to which outsourcing benefits are gained [41]. Thus, controlling the duration of the prior relationship was necessary to eliminate potential spurious time effects in the outsourcing relationship.

3.2. Data collection and sample characteristics Consistent with most prior studies that examined the effects of firm-level resources on outsourcing relationships and outcomes at the project-level (e.g., [24,30,38]), we collected data to represent the project-level measurements and analyses for the following reasons. First, considering that most IT outsourcing contracts and partnerships are made at the project-level between client and vendor firms, IT capability variables are intended to affect the success of IT outsourcing projects rather than the overall performance of the firm. Second, the IT capability of a firm may differently influence various IT outsourcing projects within the firm. Thus, analyses for hypothesis tests were based on the ratings of project managers on the specific firm-level IT capability of focal projects. In the survey instruction for IT capability, we prompted respondents to consider their experiences in IT outsourcing projects, as presented in Appendix A. Therefore, we used the outsourcing projects between client and vendor firms as the unit of analysis of this study and data for the analyses were collected from managers involved in the particular projects. Specific rating sources for the main study variables were as follows. Client IT capability was assessed by the project managers of vendor

Table 1 Question items. Constructs Client IT capability (15)

Variables

Technology management capability Organizational relationship capability Vender management capability Vendor IT capability Personnel capability (13) Methodology capability Client management capability Outsourcing success Strategic, economic, and (9) technological benefits of IT outsourcing General question (13) General questions Total –

Items References 6 4 5 4 5 4 9

[19,42,47]

13 50

– –

5

firms, whereas vendor IT capability was rated by their counterparts. This cross-rating approach was conducted to obtain a neutral and unbiased assessment because evaluation generosity is predominantly found in self-reports [58]. This approach was also intended to tap the reciprocal aspect of outsourcing partnerships. By contrast, the outsourcing success of client firms was evaluated by the project managers of client firms because the ultimate purpose of IT outsourcing is the satisfaction of client firms, not of vendors [71], and the assessments by clients are more conservative than those by vendors as IT service providers. To collect the data noted above, we first contacted the senior executives of five major global IT outsourcing vendors in Korea to ensure their full support in the survey administration and collection processes. The senior executives identified 400 different outsourcing projects that were about to be completed and helped us distribute a total of 800 survey questionnaires to each of the project managers from both clients and vendors (400 client and 400 vendor versions). A sealable return envelope for the confidentiality of responses was enclosed with every questionnaire. Among the initially distributed survey questionnaires, 282 matched responses from 282 client-side and 282 vendor-side project managers were returned (70.5% response rate). We discarded 15 matched responses because of incomplete ratings, yielding a total of 267 usable client–vendor matched reports consisting of 267 client and 267 vendor versions for hypothesis tests (each of the five vendor firms has provided 64, 46, 54, 49, and 54 outsourcing projects, respectively, at the time of data collection). Neither client-side nor vendorside project managers were engaged in multiple outsourcing projects, thus indicating that each of the 267 client–vendor matched reports was about an outsourcing project. The response profile is summarized in Table 2. Prior to hypothesis testing, we conducted two separate tests to assess data quality. First, non-respondent bias was tested. An analysis of non-respondent bias was conducted by comparing the outsourcing project size and period across the sampling frame [6]. For this test, 50 outsourcing projects were randomly selected from the non-respondent projects. The project sizes and periods of these non-respondent projects were compared with those of respondent ones. A series of t-tests indicated no difference in all comparisons at the 0.01 level of significance. Hence, we concluded that respondents could provide adequate answers on the research instrument. Second, given that the unit of analysis in this study was the project-level, we examined whether a significantly greater variance was present in the IT capability across outsourcing projects than between firms. The ANOVA results revealed that significant

Table 2 Profile of the responses. (a) Outsourcing project Outsourcing period (year) Less than 1 year 1–3 years 3–5 years 3–5 years 5 year and above Total

Outsourcing contract amount (million Korean won) 65 83 75 37 7 267

[27,41,62]

14 22 73 89 69 267

(b) Outsourcing types and client firm's industry Client firm's industry

[11,42,43]

Less than 100 100–500 500–1000 1000–5000 5000 and above Total

Manufacturing Government/Public Banking/Finance Construction Transport/Warehousing Service Others Total

Outsourcing type 38 85 25 15 34 45 25 267

IT consulting Application development Application maintenance Network Data center operation Others

25 108 52 30 43 9

Total

267

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

6

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

differences did not exist in the IT capability ratings among the five vendor firms (F = 2.59, p > 0.05) and among the 267 client firms (F = 3.23, p > 0.05); thus, the manner of operation of client and vendor IT capabilities in each outsourcing project differed across the projects. These results supported our project-level analysis. 4. Analysis and results 4.1. Analysis method The Partial Least Squares (PLS) method was selected to examine the proposed hypotheses. The PLS technique is well suited for this study because of several reasons. First, PLS is suitable for assessing theories in their early stages of development [20]. Because this study is an initial attempt to advance a theoretical model by introducing new concepts such as client IT capability, vendor IT capability, and their complementarity, PLS appears appropriate for this study. Second, the PLS requires only a minimal sample size in validating models compared with other structural equation modeling techniques [16]. Although the size of our full sample was more than adequate, we needed to divide our sample into four different sub-groups, 4 specifically in testing H4. Finally, PLS allows for latent constructs to be modeled with formative indicators, which is an important requirement in this study because we modeled two second-order factors (i.e., client and vendor IT capabilities) with formative first-order constructs. This study used PLS-Graph 3.0 in analyzing the measurement and structural models. 4.2. Measurement model Following the recommended two-stage analytical procedure [28], confirmatory factor analysis was conducted to assess the measurement model. Thereafter, the structural relationship was examined. This approach was designed to ensure that the results of the structural relationship would be predicated on an accurate and desirable representation of the indicator reliability in the measurement model. Most constructs were modeled to be reflective in the measurement model. However, based on the rules, suggested by Jarvis et al. [34], to determine whether the measures of the constructs were reflective or formative, the two second-order constructs (i.e., client and vendor IT capabilities) were modeled as formative. Their measures did not only necessarily co-vary, but were also not interchangeable. Furthermore, the direction of causality began from the items to the latent constructs. To validate our measures, three types of instrument validity were assessed: content, convergent, and discriminant. Content validity refers to the representativeness and comprehensiveness of items used to create a scale. It is assessed by examining the generation process of scale items [66]. Content validity is established by ensuring the consistency between measurement items and extant literature through the interviews with IT outsourcing academicians and practitioners and pretesting of the instrument [67]. Convergent validity was then assessed by investigating the composite reliability and extracting the average variance (AVE) of the measures [28]. Although many studies employing PLS use 0.5 as the reliability threshold of the measures, 0.7 is the recommended value

4 The four sub-groups are (1) H–H: high client and high vendor IT capabilities, (2) H–L: high client and low vendor IT capabilities, (3) L–H: low client and high vendor IT capabilities, and (4) L–L: low client and low vendor IT capabilities. The classification was carried out based on median values (i.e., 3.461 for the IT capability of clients and 3.833 for that of vendors). The sample size of the four sub-groups was 92, 42, 47, and 86, respectively.

for a reliable construct [16]. As shown in Table 3, the composite reliability values range from 0.884 to 0.941 for the full sample, 0.793 to 0.937 for the sub-sample of high client and high vendor IT capabilities (H–H), 0.757 to 0.930 for the sub-sample of high client and low vendor IT capabilities (H–L), 0.733 to 0.916 for the sub-sample of low client and high vendor IT capabilities (L–H), and 0.822 to 0.928 for the sub-sample of low client and low vendor IT capabilities (L–L). For AVE, a score of 0.5 is an acceptable level [21]. Table 3 shows that the AVE values of our measures are very satisfactory at 0.620 or above for the full sample, 0.582 or above for the sub-sample of H–H, 0.534 for the sub-sample of H–L, 0.530 for the sub-sample of L–H, and 0.591 for the sub-sample of L–L. Table 3 also shows the loadings and t-values of the measures for the full sample and four sub-samples. All measures are significant on their path loadings at the 0.01 and 0.05 levels. Finally, the discriminant validity of our instrument was verified by examining the square root of the AVE [21]. Table 4 shows that the square root of the AVE for each construct (in the full sample and four sub-samples) is greater than the correlations between that and all the other constructs. Moreover, the results of the inter-construct correlations show that each construct shares a larger variance with its own measures than with other measures. Overall, these results indicate that the measurement models are strongly supported by the gathered data. In addition to validity assessment, the multicollinearity of all variables was assessed by using the Variance Inflation Factor (VIF). The results demonstrate that the VIF values for the constructs are acceptable: 1.512 to 1.876 for the full sample, and 1.102 to 1.745 for the H–H sub-group, 1.105 to 1.465 for the H–L sub-group, 1.128 to 1.708 for the L–H sub-group, and 1.166 to 1.670 for the L– L sub-group. 4.3. Structural model As mentioned earlier, two IT capability constructs (i.e., client and vendor IT capabilities) were conceptualized as second-order factors. Although this study selected PLS as its major analysis approach, we found PLS to be unsuitable for assessing their validity because it does not address the “fit” of the overall model other than the extent to which a criterion variable is explained. Therefore, we used LISREL to test the dimensionality and validity of the two second-order IT capability constructs [68]. As summarized in Appendix B, the results strongly support the efficacy of the second-order constructs proposed in this study. With adequate measurement models and an acceptable level of multi-dimensionality of the second-order constructs, the hypotheses proposed in this study were tested by using the PLS method. We first analyzed the structural model with the full sample to test the first three hypotheses (i.e., H1, H2, and H3) and then tested the structural models by using four sub-group samples to examine H4 (see footnote 4). The significance of all paths in each structural model was assessed by using a bootstrapping resampling procedure on 300 sub-samples. First, we tested the effects of client IT capability (H1) and that of vendor IT capability (H2) on outsourcing success. As expected, the structural links from client IT capability to outsourcing success (β = 0.209; t = 3.387; p b 0.001) and from vendor IT capability to outsourcing success (β = 0.463; t = 7.080; p b 0.001) are all positive and significant. These findings provide empirical support for both H1 and H2. As exemplified by the model, all control variables, such as outsourcing project size, project type, and prior relationship, exhibit insignificant effects on outsourcing success. The two second-order constructs explain approximately 28% of the variance in outsourcing success. Second, we considered the interaction effects between client and vendor IT capabilities on outsourcing success to examine H3

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

Variables

Items

CR/AVE Client's technology management capability

Client's org. relationship capability

Client's vendor management capability

Vendor's personnel capability

Vendor's methodology capability

Vendor's client management capability

Outsourcing success

a b

CTMC1 CTMC2 CTMC3 CTMC4 CTMC5 CTMC6 CORC1 CORC2 CORC3 CORC4 CVMC1 CVMC2 CVMC3 CVMC4 CVMC5 VVPC1 VVPC2 VVPC3 VVPC4 VVMC1 VVMC2 VVMC3 VVMC4 VVMC5 VCMC1 VCMC2 VCMC3 VCMC4 OUTSU1 OUTSU2 OUTSU3 OUTSU4 OUTSU5 OUTSU6 OUTSU7 OUTSU8 OUTSU9

High–Highb (n = 92)

Full (n = 267) a

0.907/0.620

0.900/0.693

0.943/0.768

0.925/0.755

0.924/0.710

0.884/0.656

0.941/0.641

a

Loading

t-value

CR/AVE

0.795 0.843 0.829 0.675 0.778 0.793 0.817 0.760 0.898 0.849 0.850 0.866 0.888 0.886 0.892 0.893 0.881 0.854 0.847 0.816 0.851 0.856 0.807 0.879 0.837 0.841 0.784 0.787 0.809 0.802 0.807 0.737 0.765 0.833 0.852 0.826 0.770

30.808 51.510 41.286 16.507 32.958 30.544 34.052 23.935 72.147 37.979 42.826 49.497 55.238 59.006 51.130 51.807 54.893 40.130 49.575 36.160 48.967 56.430 40.004 62.448 38.688 42.344 30.283 25.706 29.357 28.741 29.480 17.844 24.835 43.994 48.064 35.555 26.748

0.793/0.594

0.883/0.654

0.896/0.634

0.881/0.649

0.822/0.582

0.825/0.542

0.937/0.624

High–Lowb (n = 42) a

Loading

t-value

CR/AVE

0.627 0.643 0.785 0.584 0.583 0.607 0.797 0.740 0.879 0.813 0.765 0.757 0.840 0.823 0.793 0.863 0.772 0.815 0.769 0.758 0.650 0.651 0.619 0.778 0.707 0.754 0.738 0.745 0.776 0.766 0.823 0.749 0.799 0.834 0.818 0.736 0.803

5.368 6.728 22.729 6.780 6.066 6.776 18.874 9.015 36.356 17.749 13.345 15.977 25.137 25.495 13.060 27.041 11.644 17.930 13.670 17.013 7.658 8.975 7.264 17.429 8.363 11.027 15.689 14.603 13.275 13.981 17.140 14.321 15.383 21.832 20.232 13.720 14.197

0.772/0.585

0.757/0.566

0.893/0.627

0.915/0.731

0.816/0.575

0.768/0.534

0.930/0.603

Low–Highb (n = 47) a

Loading

t-value

CR/AVE

0.785 0.786 0.765 0.652 0.601 0.743 0.630 0.633 0.842 0.873 0.732 0.730 0.766 0.867 0.852 0.907 0.904 0.795 0.806 0.689 0.710 0.721 0.675 0.810 0.692 0.723 0.700 0.673 0.648 0.724 0.791 0.845 0.816 0.902 0.880 0.829 0.645

6.686 12.871 10.856 5.226 4.317 6.389 4.388 5.441 16.213 21.251 6.061 4.928 10.765 27.228 14.096 34.057 24.579 6.972 10.695 2.576 2.621 7.478 4.482 7.437 2.058 2.319 2.254 1.993 6.291 8.435 9.955 13.103 10.582 23.015 18.281 10.548 6.211

0.834/0.570

0.799/0.607

0.896/0.632

0.868/0.623

0.827/0.591

0.733/0.530

0.916/0.550

Low–Lowb (n = 86)

Loading

t-value

CR/AVEa

0.692 0.878 0.700 0.646 0.754 0.823 0.731 0.686 0.809 0.777 0.736 0.807 0.805 0.801 0.823 0.850 0.798 0.797 0.707 0.787 0.732 0.710 0.649 0.712 0.786 0.825 0.651 0.642 0.743 0.829 0.713 9.744 0.728 0.773 0.772 0.735 0.623

3.278 20.431 8.780 2.844 9.180 13.857 9.439 4.284 5.068 8.452 7.906 16.071 13.080 13.459 13.669 24.122 13.455 12.734 6.854 9.569 6.709 9.120 5.778 3.209 7.266 10.595 3.006 2.954 10.672 11.323 6.654 7.543 7.621 11.722 11.122 7.904 5.097

0.874/0.640

0.822/0.639

0.913/0.677

0.892/0.674

0.886/0.611

0.831/0.657

0.928/0.591

Loading 0.788 0.798 0.836 0.636 0.730 0.679 0.696 0.649 0.856 0.757 0.788 0.825 0.814 0.850 0.836 0.824 0.880 0.784 0.791 0.662 0.774 0.854 0.712 0.884 0.836 0.834 0.711 0.674 0.794 0.758 0.765 0.673 0.660 0.788 0.828 0.829 0.802

t-value 18.053 19.611 24.650 9.788 13.728 10.276 11.342 9.304 24.499 11.545 17.348 15.448 13.732 23.291 18.229 15.978 39.681 11.028 22.507 8.626 15.838 28.767 11.651 35.499 24.705 17.988 8.456 5.921 14.951 14.926 13.909 7.096 7.150 15.291 21.810 21.703 14.732

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

Table 3 Results of PLS factor analysis.

CR stands for Composite Reliability and AVE is Average Variance Extracted. High–High: high client and high vendor IT capabilities; High–Low: high client and low vendor IT capabilities; Low–High: low client and high vendor IT capabilities; Low–Low: low client and low vendor IT capabilities.

7

8

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

Table 4 Correlations of latent constructs and evidence of discriminant validity. (a) Full sample (n = 267) Construct

Mean

S.D

PRJSIZ

PRIREL

PRJSIZ PRIREL CTMC CORC CVMC VPC VMC VCMC OUTSC

275.02 19.430 3.393 3.380 3.512 3.889 3.362 3.701 3.703

569.73 18.349 0.608 0.630 0.747 0.576 0.673 0.633 0.612

−0.056 0.100 0.085 0.141⁎ 0.026 −0.013 0.032 0.040

0.026 0.001 0.044 0.080 0.045 0.014 −0.036

CTMC

0.787 0.562⁎⁎ 0.586⁎⁎ 0.190⁎⁎ 0.278⁎⁎ 0.112 0.225⁎⁎

CORC

0.832 0.560⁎⁎ 0.363⁎⁎ 0.458⁎⁎ 0.339⁎⁎ 0.357⁎⁎

CVMC

0.876 0.254⁎⁎ 0.379⁎⁎ 0.359⁎⁎ 0.413⁎⁎

VPC

0.869 0.528⁎⁎ 0.517⁎⁎ 0.362⁎⁎

VMC

0.843 0.508⁎⁎ 0.510⁎⁎

VCMC

0.810 0.489⁎⁎

OUTSC

0.800

(b) High–High Sub-sample (n = 92) Construct

Mean

S.D

PRJSIZ

PRIREL

CTMC

PRJSIZ PRIREL CTMC CORC CVMC VPC VMC VCMC OUTSC

345.84 21.980 3.844 3.864 4.037 4.253 4.143 4.163 4.010

670.52 23.132 0.386 0.485 0.499 0.410 0.381 0.436 0.488

CORC

−0.120 −0.029 −0.025 0.175 0.037 0.150 0.156 0.137

−0.094 0.004 0.076 0.155 0.064 0.148 −0.132

0.771 0.001 −0.069 −0.064 −0.242⁎ −0.247⁎

0.809 0.155 0.326⁎⁎ 0.279⁎⁎ 0.229⁎

−0.005

0.080

CVMC

0.796 0.142 0.180 0.500⁎⁎ 0.275⁎⁎

VPC

0.806 0.412⁎⁎ 0.255⁎ 0.120

VMC

0.763 0.469⁎⁎ 0.440⁎⁎

VCMC

0.736 0.347⁎⁎

OUTSC

0.790

(c) High–Low Sub-sample (n = 42) Construct

Mean

S.D

PRJSIZ

PRIREL

CTMC

CORC

CVMC

VPC

VMC

PRJSIZ PRIREL CTMC CORC CVMC VPC VMC VCMC OUTSC

353.94 15.240 3.758 3.649 3.943 3.613 3.243 3.274 3.656

694.11 7.780 0.325 0.371 0.420 0.513 0.418 0.461 0.547

−0.117 0.029 0.185 0.283 0.149 −0.213 0.025 −0.134

−0.144 −0.165 0.030 −0.016 −0.420⁎⁎ −0.314⁎⁎ −0.254

0.765 0.289 0.152 0.156 0.030 −0.015 0.180

0.752 0.032 −0.011 −0.097 0.050 0.021

0.792 −0.196 −0.214 −0.056 0.278

0.855 0.005 0.104 −0.063

0.758 0.501⁎⁎ 0.334⁎

VCMC

0.731 0.320⁎

OUTSC

0.776

(d) Low–High Sub-Sample (n = 47) Construct

Mean

S.D

PRJSIZ

PRIREL

CTMC

CORC

CVMC

VPC

VMC

VCMC

OUTSC

PRJSIZ PRIREL CTMC CORC CVMC VPC VMC VCMC OUTSC

150.04 17.890 2.957 3.043 3.047 4.138 3.979 4.048 3.808

337.64 18.632 0.437 0.449 0.583 0.403 0.426 0.348 0.531

0.138 0.104 0.182 0.206 −0.295⁎ 0.185 −0.059 0.346⁎

−0.047 −0.083 −0.018 0.100 −0.045 0.060 0.102

0.755 0.184 0.596⁎⁎ −0.156 −0.071 −0.278 −0.209

0.779 0.100 −0.281 −0.012 −0.135 0.164

0.795 0.013 −0.164 −0.263 0.076

0.789 −0.229 −0.116 0.042

0.769 0.369⁎ 0.011

0.728 −0.047

0.742

CORC

CVMC

VPC

VMC

VCMC

OUTSC

(e) Low–Low Sub-Sample (n = 86) Construct

Mean

S.D

PRJSIZ

PRIREL

CTMC

PRJSIZ PRIREL CTMC CORC CVMC VPC VMC VCMC OUTSC

229.01 19.590 2.971 2.916 2.993 3.497 3.056 3.227 3.340

470.20 15.706 0.521 0.495 0.656 0.524 0.565 0.515 0.615

−0.016 0.026 −0.111 −0.163 −0.027 −0.256⁎ −0.121 −0.154

0.170 −0.051 0.015 −0.074 0.016 −0.313⁎⁎ −0.050

0.800 0.283⁎⁎ 0.258⁎ −0.051 0.195 −0.238 0.011

0.799 0.328⁎⁎ 0.308⁎⁎ 0.496⁎⁎ 0.139 0.268⁎

0.823 0.118 0.484⁎⁎ 0.346⁎⁎ 0.367⁎⁎

0.821 0.211 0.229⁎ 0.287⁎⁎

0.782 0.358⁎⁎ 0.348⁎⁎

0.810 0.384⁎⁎

0.769

The shade numbers in the diagonal row are square roots of the average variance extracted. PRJSIZ: Project size; PRIREL: Prior relationship; CTMC: Client's technology management capability; CORC: Client's org. relationship capability; CVMC: Client's vendor management capability; VPC: Vendor's personnel capability; VMC: Vendor's methodology capability; VCMC: Vendor's client management capability; OUTSC: Outsourcing success. ⁎ p b 0.05. ⁎⁎ p b 0.01.

[69]. The moderation fit perspective, combined with the complementary moderated regression analysis method, was adopted because the degree of specificity and criterion of the relationship of the variables were high. The results indicate that the interaction between

client and vendor IT capabilities positively and significantly affect outsourcing success (β = 0.365; t = 5.210; p b 0.001). In the model, we found that the effects of the three control variables on outsourcing success are insignificant. The client IT capability, vendor

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

Fig. 1. Complementarity between client and vendor IT capabilities.

IT capability, and their interaction account for approximately 36% of the variance in outsourcing success. To identify the magnitude of the interaction effect between client and vendor IT capabilities, the hierarchical process recommended by Chin et al. [17] was used. We compared the results of the two different models (i.e., non-interaction and interaction models) in terms of their difference in R-squares. Cohen [18] stated that differences in R-squares can be used to assess the overall interaction effect size (ƒ 2) at three different levels: 0.02 to 0.14 for small effects, 0.15 to 0.34 for medium effects, and above 0.35 for large effects. 5 As explained in the testing process of H1 and H2, the main noninteraction model accounts for 28% of the variance, whereas the interaction model explained 36% of the variance in outsourcing success. The results of the hierarchical difference tests reveal an interaction effect size (ƒ 2) of 0.198 (=[0.36 − 0.28] / [1 − 0.28]), which indicates a small interaction effect [17]. Cohen [18] asserted that a low effect size (ƒ 2) does not necessarily imply an unimportant effect. Thus, the interaction model has significantly higher explanatory powers than the main non-interaction model. These findings confirm the interaction effect between client and vendor IT capabilities, thereby providing strong support for H3. Furthermore, we explored the interaction plot, as presented in Fig. 1. This plot was constructed by generating categorical variables for the IT capability of clients and the IT capability of vendors. The median-split method used for transforming continuous variables to dichotomous variables may cause information loss [1]. For example, ratings slightly higher and lower than the median value are categorized into two different groups. Additionally, ratings far and slightly higher (or lower) than the median value are classified into a category. To address this issue in examining the interaction pattern, we excluded the cases in the middle 10% of the two variables (in-between 5% over and below the median value). The resulting marginal mean of outsourcing success for low client and low vendor IT capabilities was 1.760. High vendor and high client IT capabilities increased outsourcing success by 1.060 (2.820 − 1.760) and 0.364 (2.124 − 1.760), respectively, yielding an additive advantage of 1.424 (1.060 + 0.364) to outsourcing success. By contrast, the multiplicative advantage, that is, the complementarity garnered by the combination of high client and high vendor IT capabilities, is 2.451 (4.211 − 1.760), which is significantly higher than 1.424

5 Interaction effect size ƒ2 = [R2 of interaction effect model − R2 of main effect model] / [1 − R2 of main effect model].

9

(t = 2.228, p(t153) = 0.025). These results further support H3, anticipating that the effect of complementarity between client and vendor IT capabilities on outsourcing success is higher than the individual effects. To test H4, we classified the full sample into four sub-groups (see the first footnote), ran the PLS with the respective sub-samples, and compared the results across sub-samples. Results show that the models of both H–H and L–L sub-groups are superior to both H–L and L–H sub-groups. Table 5(a) shows that in the case of H–L and L–H sub-groups, the effects of client and vendor IT capabilities on outsourcing success are not significant except the relationship between vendor IT capability and outsourcing success, though all effects are significant for the H–H and L–L sub-groups. Furthermore, the R-square values of the H–H and L–L sub-groups are higher than those of H–L and L–H sub-groups. We then statistically compared the path coefficients of the interaction between client and vendor IT capabilities to outsourcing success across four different sub-groups. The statistical comparison was conducted by adopting the procedure of Keil et al. [35]. As summarized in Table 5(b), results show stronger path coefficients for the H–H and L–L sub-groups in terms of the interaction between client and vendor IT capabilities than those for the H–L and L–H sub-groups, thereby supporting H4. 5. Discussion and implications This study investigated the effect of client IT capability, vendor IT capability, and their complementarity on outsourcing success. Based on prior studies that focused on RBV, we conceptualized client and vendor IT capabilities as second-order factors. We then hypothesized not only the effects of client and vendor IT capabilities (H1 and H2) separately but also their complementarity, which was expected to be stronger than the individual capabilities of clients and vendors (H3) from the perspective of complementarity. Moreover, we posited that the increase in the rate of complementarity is higher when the IT capability levels of both parties are similar than those of both parties differ (H4). As summarized in Table 6, the results reveal that both client and vendor IT capabilities are significant factors in outsourcing success and that their complementarity significantly influences outsourcing success. Furthermore, the increase rate of complementarity is greater when the IT capability levels of both parties are similar than when those of both parties differ. The results highlight the nature of complementarity in IT outsourcing and suggest that the best complementarity can be achieved when the IT capability of both parties are at a similar level (i.e., either high or low). Roberts [59] found that the working performance of both parties is improved in situations where one party has weak capability in a particular area (bad) and the other party has high capability in the area (good) because the party with weak capability does not try to interfere with the other party in the area. The good/bad contrast suggests a contingent form of complementarity, wherein an organization is unable to elicit value from one activity without undertaking the other. Brynjolfsson et al. [15] also found that complementarity is evident in negative interactions between computerization and existing organizational practices. Contrary to their findings, the result of this study suggests that, in the context of IT outsourcing, a higher degree of complementarity is more attainable when the IT capability levels of both parties are either high or low than when one party's IT capability level is high and the other party's IT capability level is low. However, this finding does not guarantee that the success of IT outsourcing is greater when the IT capability levels of both parties are either low or high. More accurately, this finding denotes that the increase rate of complementarity is higher when the IT capability levels of both parties are either low or high than when one party's IT capability is high and the other party's IT capability is low. These results imply that when the IT capability levels of both parties are at a similar level, they can

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

10

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

Table 5 Complementarity effects across sub-samples. (a) Path Coefficient and R-square of Sub-samples Structural Link

Full sample (=267)

High–High (n = 92)

High–Low (n = 42)

Low–High (n = 47)

Low–Low (n = 86)

Client IT capability → outsourcing success Vendor IT capability → outsourcing success Interaction of client and vendor IT capabilities → outsourcing success R-square Values in each cell are path coefficients (t-statistics)

0.204 (3.458)⁎⁎ 0.452 (7.041)⁎⁎ 0.365 (5.210)⁎⁎ 0.360

0.289 (2.872)⁎ 0.337 (3.971)⁎⁎ 0.297 (3.380)⁎⁎ 0.369

0.213(1.066) 0.332 (2.016)⁎ 0.262 (2.853)⁎⁎ 0.302

0.133 (1.493) 0.173 (1.826) 0.236 (2.475)⁎ 0.257

0.319 (2.120)⁎ 0.362 (2.261)⁎ 0.334 (3.092)⁎⁎ 0.305

(b) Comparison between sub-samples Sub-samples

H–H sub-sample (n = 92)

H–L sub-sample (n = 42)

L–H sub-sample (n = 47)

L–L sub-sample (n = 86)

H–H sub-sample H–L sub-sample L–H sub-sample L–L sub-sample

– t = 10.769⁎⁎ (H–H > H–L) t = 12.353⁎⁎ (H–H > L–H) t = 1.645 (H–H L–L)

– t = 2.591⁎⁎ (H–L > L–H) t = 28.662⁎⁎ (L–L > H–L)

– t = 22.172⁎⁎ (L–L > L–H)

–

H–H: high client and high vendor IT capabilities; H–L: high client and low vendor IT capabilities; L–H: low client and high vendor IT capabilities; L–L: low client and low vendor IT capabilities. ⁎ p b 0.05. ⁎⁎ p b 0.01.

absorb each other's knowledge and experience more efficiently and higher learning is attained in outsourcing projects. The results of this study confirm the existence of a subtle relationship between clients and vendors. Client and vendor IT capabilities are more than simple determinants of IT outsourcing success; they are potential sources of attractive synergetic effects. We not only show that client and vendor IT capabilities are important but also prove that their relationship is a critical determinant of outsourcing success. This result is significant because it extends the perspective of IT outsourcing from individual entities to relationships. Although a single business entity can play its role with a simple, linear effect, we can extract more sophisticated implications from entities if we consider their various relationships and possible dynamic features. With emphasis on complementarity, this study has several theoretical contributions. First, we extended the bilateral view to a more specific concept of complementarity. In IT outsourcing research, obtaining the bilateral views of both clients and vendors is considered necessary and important in relation to business structures [36,62]. However, few studies have actualized and concretized this view. The concept of fit has been applied, but a simple “fit” is insufficient in explaining the sophisticated relationship between clients and vendors. Thus, we introduced the concept of complementarity and concretized it in the context of IT outsourcing. Complementarity is effective because it not only considers both parties simultaneously but also specifies the effect corresponding to the IT capability level of each party. Prior research merely suggests the existence of synergetic effects, whereas this study uses complementarity to predict and explain the effect of synergy in a more scientific manner.

Second, this study conceptualized and empirically validated complementarity in a reciprocal manner by using the cross-rated reports. The value of “reciprocal” data was strengthened in this study because of its theoretical consistency with complementarity. The most important condition for complementarity is the presence of reciprocity between parties. The most common limitation in the validation processes is the difficulty of collecting data that satisfy the balanced perspectives between parties; however, we attempted to incorporate the required reciprocity (or balance) in data analyses by using the cross-rated responses from both sides. This reciprocity in the collected data provides solid support to the concept of complementarity, thereby making the research empirically testable. Finally, this study extended the existing IT capability perspectives of both clients and vendors to IT outsourcing and their empirical validation. Most prior outsourcing studies concentrate mainly on client firms that decide to outsource, thereby overlooking the importance of vendor IT capability. Although understanding the outsourcing relationships of both parties is necessary [36,62], prior studies on the role of vendors are generally conceptual and exploratory, and lack empirical investigation [43,57]. In this study, we conceptualized vendor IT capability and empirically tested its effect, combined with client IT capability, on outsourcing success. Thus, this study, which is an initial empirical attempt, provides a better understanding of the role and complementarity of client and vendor IT capabilities in outsourcing projects. The results of this study have several implications for practitioners. First, the practical importance of this complementarity research will emerge when managers of both client and vendor firms design IT outsourcing projects. In business decision making, the most important

Table 6 Results of hypotheses test. Hypotheses

Results

H1

Supported

H2 H3 H4

The higher the level of client IT capability to reflect technology management, organizational relationship, and vendor management capabilities, the higher the level of outsourcing success. The higher the level of vendor IT capability to reflect personnel, methodology, and client management capabilities, the higher the level of outsourcing success. The complementarity between client and vendor IT capabilities adds value to outsourcing success, in such a way that vendor IT capability augments the positive impact of client IT capability on outsourcing success, and vice versa. The increase rate of complementarity between client and vendor IT capabilities is higher when both parties have high or low IT capability levels than when one party has high IT capability level and the other party has low IT capability level.

Supported Supported Supported

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

factor for managers is the additional, not the expected, value that they can garner through transactions or projects. If the returns on investments are always proportionally predictable without additional benefits, these investments will not be considered very attractive. This study presents the possibility of gaining additional value through IT outsourcing projects in a very detailed manner. This study not only shows the existence of the synergetic effect produced by complementarity but also rationalizes the direction and magnitude of complementarity corresponding to the capability of each party. Therefore, this study provides ideas to managers involved in outsourcing decisions on the expected returns of projects that are under consideration. Second, the result of higher complementarity when the IT capabilities of both clients and vendors are either low or high can be interpreted from the organizational learning perspective. Organizations learn by collaborating with others and by observing and importing the practices of other organizations [44]. In IT outsourcing, our findings suggest that when a client has a low (or high) IT capability level, vendors with low (or high) IT capability levels are effective sources of organizational learning and innovative ideas. If vendor IT capability is high and client IT capability is low, the client tends to follow the basic guidelines provided by the vendor in the outsourcing process without careful consideration to improvements and updates, thereby inhibiting the organizational learning of both parties. This situation occurs because the scarcity of IT capability makes client firms dependent on their vendors for acquiring the knowledge necessary for the realization of their outsourcing objectives [22]. By contrast, if both parties have the similar IT capability levels, they are likely to have more communication and frequent interactions for the development of new ideas and solutions. This situation occurs because clients and vendors perceive mutual benefits from interactions or from negotiations for more stable mechanisms [22]. Therefore, clients with low (or high) IT capabilities have better chances of learning outsourcing knowledge and practices by developing relationships with vendors that have low (or high) IT capabilities. Third, we specified the detailed features of client and vendor IT capabilities to allow managers to improve their IT capability levels in businesses more realistically. Capability has often been regarded as a very important business factor [56,68], but applying this concept is not easy for practitioners because of its highly condensed gist. Thus, we identified the three sub-concepts of client and vendor IT capabilities that influence the outsourcing decisions of managers. Finally, this study offers practical insights on the boundaries of necessary assets or skills that can be cultivated within client firms to leverage the technical expertise of IT outsourcing vendors. Although most IT-related corporate resources can be outsourced and are mandatory in some sense, client IT capability, including technology management, organizational relationship, and vendor management capabilities, should be considered valuable organizationspecific assets. The acceptance of H4 implies that the synergy from IT outsourcing is higher when the IT capabilities of both parties are complementarily leveraged with their maximal capability levels. We can extend this implication further in conjunction with the fundamental enabler of the success of change management from the perspective of client firms. Further research should be conducted to ascertain the causality of the effect of client IT capability on the critical attributes of successful change management such as top management support. The efforts of vendors to change management are increased when the capabilities of such client firms are higher. The limitations of this study offer opportunities for future research. The first limitation of this study is the composition of the sample. The sample lacked randomness because the number of project units was limited. However, we believe that the composition of the sample does not significantly harm the research findings because of the high-quality data collected from both clients and vendors. Second, this study was conducted as a snapshot research without considering

11

the dynamic nature of outsourcing relationships. We can further elaborate the results with additional data and a more in-depth interpretation of the results, especially for the strategic fit, moderation effect, and indirect effect. Third, we believe that using client ratings on vendor IT capability and vendor reports on client IT capability is appropriate. This approach is actually the strength of our study because how one party perceives another in an outsourcing project that aims for a successful partnership would better reflect the relational outcomes of the project compared to one's own capability evaluation. Furthermore, given that IT outsourcing success is assessed by client firms, using vendor ratings of client IT capability is recommended to prevent spurious covariance because of common rater bias [58]. Nonetheless, the self-evaluation of respondents in the context of our study may be more accurate than the ratings of others because IT-related capability is a firm-specific resource [47]. Although these two contrasting views exist, we did not test this effect because self-ratings on IT-related capability were not available. Fourth, similar assessments (either favorable or unfavorable) on outsourcing processes and outcomes between clients and vendors are likely to develop over time; thus the mutual agreement might influence client ratings in outsourcing success. However, among the 267 outsourcing projects examined, the clients and vendors in 89 outsourcing projects (33.3%) reported discrepant ratings on the IT capability of their counterparts. Additionally, we controlled for the potential effect of prior outsourcing relationship duration on outsourcing success [41]. Prior outsourcing relationship duration could serve as a proxy measure for the extent to which clients and vendors develop mutual understanding of outsourcing-related issues. Thus, the inclusion of this control variable in the hypothesis test may partial out the potential effect of mutual agreement on outsourcing success. Nonetheless, the potential effect was not fully considered in the hypothesis tests; thus this issue is a study limitation that warrants further investigations in future research. Fifth, simultaneously understanding the success and risk factors of outsourcing is necessary because they are not mutually exclusive but tightly coupled. This study does not directly investigate risk factors because the focus of this study is the success of outsourcing projects from the perspective of complementarity between client and vendor IT capabilities. However, we believe that the results of this study can provide meaningful implications in terms of both success factors and risk factors. By matching client and vendor IT capabilities, we can find a more effective way to increase the possibility of making an outsourcing project successful and reduce the risk of failure in IT outsourcing. Finally, the results of this study may include some biases because the sample was restricted to Korean firms, even though data were collected from global IT outsourcing firms. Thus, the results may have to be carefully interpreted. The replication of this study in other regions is also needed to improve the generalizability of the findings. 6. Conclusion This study raised the important issue of the bilateral perspective of outsourcing success and the subtle characteristics of the complementarity between the views of clients and vendors. By analyzing data gathered from vendors and clients reciprocally, this study investigated not only the relations among outsourcing factors in terms of client and vendor IT capabilities and their complementarity but also their effects on outsourcing success. It was not surprising to find that client and vendor IT capabilities are significantly associated with outsourcing success, even though the effect of their complementarity shows rather interesting and complicated features. The findings of this study provide useful and salient business implications that can be used in making better outsourcing decisions in selecting clients and vendors and in developing better outsourcing relationships.

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

12

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

Appendix A. The structure of the survey instrument

Variable

Item

Question

Client IT capability (formative second-order factor) Technology management capability Based on the experience in our IT outsourcing project, we think that CTMC1 Our client firm has the ability to standardize information technologies. CTMC2 Our client firm has the ability to integrate various information technologies. CTMC3 Our client firm has the ability to understand IT trends. CTMC4 Our client firm has the ability to identify IT functional requirements. CTMC5 Our client firm has the ability to leverage IT as a strategic competency. CTMC6 Our client firm has the ability to update IT strategy constantly with the changes in the business environment. Organizational relationship capability Based on the experience in our IT outsourcing project, we think that CORC1 In our client firm, the IT department and the management communicate well. CORC2 In our client firm, the IT department's decisions reflect the opinions of operational users. CORC3 In our client firm, the IT department and operational users communicate harmoniously. CORC4 In our client firm, the IT department and operational teams trust each other. Vendor management capability Based on the experience in our IT outsourcing project, we think that CVMC1 Our client firm has a standardized process for vendor selection. CVMC2 Our client firm has the ability to evaluate outsourcing performance. CVMC3 Our client firm has the ability to manage outsourcing processes. CVMC4 Our client firm has a systematic process for contract management. CVMC5 Our client firm has a systematic process for vendor management. Vendor IT capability (formative second-order factor) Personnel capability Based on the experience in our IT outsourcing project, we think that VVPC1 The project team members of our vendor firm have core technical knowledge necessary for our project. VVPC2 The project team members of our vendor firm know a methodology for conducting our project. VVPC3 The project team members of our vendor firm have the ability to apply related technologies to our project. VVPC4 The project team members of our vendor firm generally have competent technical skills. Methodology capability Based on the experience in our IT outsourcing project, we think that VVMC1 In our vendor firm, tasks are performed with systematic methods in knowledge bases. VVMC2 Our vendor firm has standardized outsourcing processes to generate project outputs. VVMC3 Our vendor firm has the ability to solve problems systematically using area experts. VVMC4 Our vendor firm has a systematic education system for project team members. VVMC5 Our vendor firm's methodology for our project is generally acceptable. Client management capability Based on the experience in our IT outsourcing project, we think that VCMC1 Our vendor firm sincerely shares the status of our project with us. VCMC2 Our vendor firm clearly understands each other's roles in our project. VCMC3 Our vendor firm provides us with valuable comments and feedback on our project. VCMC4 Our vendor firm has coordination mechanisms to solve problems with us. Outsourcing success Based on the experience in our IT outsourcing project, we think that OUTSC1 We have been able to refocus on core business. OUTSC2 We have enhanced our IT competence. OUTSC3 We have increased access to skilled personnel. OUTSC4 We have reduced the human resource management cost. OUTSC5 We have increased efficiency in IT expenses. OUTSC6 We have increased efficiency in expenses. OUTSC7 We have reduced the risk of technological obsolescence. OUTSC8 We have increased access to key information technologies. OUTSC9 We are satisfied with our overall benefits from IT outsourcing.

Appendix B. Assessing the two second-order factors Two IT capability constructs (i.e., client and vendor IT capabilities) were conceptualized as second-order factors that were modeled with formative first-order dimensions with reflective second-order measurement items. We tested their dimensionality and validity using LISREL. According to Tanriverdi [68], we need to hypothesize four different models and compare them to validate the reliability and validity of a second-order construct, as depicted in Figs. B-1 and B-2. These four models are as follows: “Model 1 hypothesizes that a unidimensional first-order factor accounts for the variance among all measurement items of the construct; Model 2 hypothesizes that the measurement items group into uncorrelated first-order factors (constrained model); Model 3 hypothesizes that these first-order factors are freely correlated with each other (unconstrained model); and Model 4 hypothesizes a second-order factor that accounts for the patterns of interactions and covariance among the first-order factors” ([68], p.322).

For both second-order factors, we first compared Model 1 with Model 2 and found that Model 2 (χ 2 = 482.33, d.f. = 90 for client IT capability and χ 2 = 424.66, d.f. = 65 for vendor IT capability) was superior to Model 1 (χ 2 = 1191.75, d.f. = 90 for client IT capability and χ 2 = 737.82, d.f. = 65 for vendor IT capability) because of the lower chi-squares with the same degrees of freedom. This indicates that multi-dimensional models for both client and vendor IT capabilities, which consist of three uncorrelated first-order factors respectively, fit better than unidimensional first-order factors. Hence, the multi-dimensionality of both second-order factors was supported. As a next step, we compared between Models 2 and 3, and realized that Model 3 (χ2 = 258.00, d.f. = 87 for client IT capability and χ2 = 284.43, d.f. = 61 for vendor IT capability) shows better fitting than Model 2 (χ2 = 482.33, d.f. = 90 for client IT capability and χ2 = 424.66, d.f. = 65 for vendor IT capability), with significant changes in the chi-square (Δχ2 = 224.33, Δd.f. = 3, p b 0.001 for client IT

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

Model 1

Model 2

Model 3

13

Model 4

Fig. B-1. Four different models for client IT capability.

Model 1

Model 2

Model 3

Model 4

Fig. B-2. Four Different Models for Vendor IT Capability.

capability and Δχ2 = 140.23, Δd.f. = 4 for vendor IT capability). We also found that the standardized factor loadings of measurement items for client and vendor IT capabilities in Model 3 were all significant at the 0.001 level. This result supported the convergent validity of both second-order factors. Furthermore, we can understand the superiority of Model 3 over Model 2 in a way that the measurement items converge on their respective factors, and the factors are distinguished from each other, which support the discriminant validity of the two secondorder constructs [3,7]. Finally, we examined the efficacy of the two second-order constructs in terms of the complementarity among the first-order factors. To do so, we introduced an external criterion variable, outsourcing success,6 to compare Model 3 (testing the direct impacts of three first-order factors on outsourcing success) and Model 4 (testing the collective impact of three first-order factors on outsourcing success) as shown in Model 4 of Figs. B-1 and B-2. As suggested by Tanriverdi [68], this study considered three criteria to examine which model is better. These three criteria were the model statistics of the two specifications, target coefficient (T) statistics, and significance of the parameters reflecting the second-order factor loadings. The testing results based on model statistics show that Model 4 (χ2 = 312.45, d.f. = 98 for client IT capability and χ2 = 376.24, d.f. = 72 for vendor IT capability) was better than Model 3 (χ2 = 305.75, d.f. = 96 for client IT capability and χ2 = 366.59, d.f. = 70 for vendor IT capability). Although the results of both constructs looked similar, we preferred Model 4, i.e., the second-order factor, to Model 3, i.e., the first-order factor, because Model 4 has lesser parameters to be estimated and more degrees of freedom for both client and vendor IT capabilities. In 6 In this test, the average score of the nine outsourcing measure items was used as an external criterion variable.

addition, the target coefficient values (T = 0.979 and T = 0.974 for client IT capability and vendor IT capability, respectively) of the secondorder factors show that they accounted for around 98 and 97%, respectively, of the relations among the first-order factors of client and vendor IT capabilities, indicating that the second-order constructs are better than the first-order constructs [65]. Lastly, the structural links between client IT capability and outsourcing success (t = 3.76, p b 0.01) and between vendor IT capability and outsourcing success (t = 6.31, p b 0.01) as well as all second-order factor loadings were positively significant at the 0.05 and 0.01 levels [69]. Overall, these results provided strong support for the efficacy of the second-order constructs, i.e., client IT capability and vendor IT capability, proposed in this study. References [1] L.S. Aiken, S.G. West, Multiple Regression: Testing and Interpreting Interactions, Sage Publications, London, 1991. [2] R. Amit, P.J.H. Schoemaker, Strategic assets and organizational rent, Strategic Management Journal 18 (1) (1993) 33–46. [3] J.C. Anderson, An approach for confirmatory measurement and structural equation modeling of organizational properties, Management Science 33 (4) (1987) 525–541. [4] S. Ang, D.W. Straub, Production and transaction economics and is outsourcing: a study of the U.S. banking industry, MIS Quarterly 22 (4) (1998) 535–552. [5] N.S. Argyres, J. Bercovitz, K.J. Mayer, Complementarity and evolution of contractual provisions: an empirical study of it services contracts, Organization Science 18 (1) (2007) 3–19. [6] E. Babbie, Survey Research Methods, Wadsworth, Belmont, CA, 1990. [7] R.P. Bagozzi, Y. Yi, L.W. Phillips, Assessing construct validity in organizational research, Administrative Science Quarterly 36 (3) (1991) 421–458. [8] J.B. Barney, Firm resources and sustained competitive advantage, Journal of Management 17 (1) (1991) 99–120. [9] J.B. Barney, How a firm's capabilities affect boundary decision, Sloan Management Review Spring (1999) 137–145. [10] A. Barua, C.H.S. Lee, A.B. Whinston, The calculus of reengineering, Information Systems Research 7 (4) (1996) 409–428.

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

14

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx

[11] G. Bassellier, I. Benbasat, B.H. Reich, The influence of business managers' it competence on championing IT, Information Systems Research 14 (4) (2003) 317–336. [12] G. Bassellier, B.H. Reich, I. Benbasat, Business competence of information technology professional: conceptual development and influence on IT-business partnership, MIS Quarterly 28 (4) (2004) 673–694. [13] A.S. Bharadwaj, A resource-based perspective on information technology capability and firm performance: an empirical investigation, MIS Quarterly 24 (1) (2000) 169–196. [14] J.A. Black, K.B. Boal, Strategic resources: traits, configurations and paths to sustainable competitive advantage, Strategic Management Journal 15 (1994) 131–148. [15] E. Brynjolfsson, A.A. Renshaw, M. Van Alstyne, The matrix of change, Sloan Management Review 38 (2) (1997) 37–54. [16] W.W. Chin, The Partial Least Squares Approach to Structural Equation Modeling in Modern Methods for Business Research, in: G.A. Marcoulides (Ed.), Lawrence Erlbaum Associates, Mahwah, NJ, 1998, pp. 295–336. [17] W.W. Chin, B.L. Marcolin, P.R. Newsted, A partial least squares latent variable approach for measuring interaction effects: results from a Monte Carlo simulation study and an electronic-mail emotion/adoption study, Information Systems Research 14 (2) (2003) 189–217. [18] J. Cohen, Statistical Power Analysis for the Behavioral Sciences, 2nd ed. Lawrence Erlbaum Associates, Hillsdale, NJ, 1998. [19] D.F. Feeny, L.P. Willcocks, Core IS capabilities for exploiting information technology, Sloan Management Review 39 (3) (1998) 9–21. [20] C. Fornell, F.L. Bookstein, Two structural equation models: LISREL and PLS applied to customer exit-voice theory, Journal of Marketing Research 19 (4) (1982) 440–452. [21] C. Fornell, D.F. Larcker, Structural equation models with unobservable variables and measurement errors, Journal of Marketing Research 18 (2) (1981) 39–50. [22] J.B. Goes, S.H. Park, Interorganizational links and innovation: the case of hospital services, The Academy of Management Journal 40 (3) (1997) 673–696. [23] J. Goo, R. Kishore, K. Nam, H.R. Rao, Y.I. Song, An investigation of factors that influence the duration of IT outsourcing relationships, Decision Support Systems 42 (4) (2007) 2107–2125. [24] A. Gopal, K. Sivaramakrishnan, M. Krishnan, T. Mukhopadhyay, Contracts in offshore software development: an empirical analysis, Management Science 49 (12) (2003) 1671–1683. [25] R.M. Grant, The resource-based theory of competitive advantage: implications for strategy formulation, California Management Review 33 (3) (1991) 114–135. [26] D.A. Griffith, M.B. Myers, The performance implications of strategic fit of relational norm governance strategies in global supply chain relationships, Journal of International Business Studies 36 (3) (2005) 254–269. [27] V. Grover, M.J. Cheon, J.T.C. Teng, The effect of service quality and partnership on the outsourcing of information systems functions, Journal of Management Information Systems 12 (4) (1996) 89–116. [28] J.F. Hair, R.E. Anderson, R.L. Tatham, W.C. Black, Multivariate Data Analysis, 5th eds. Prentice Hall, Englewood Cliffs, 1998. [29] J.A. Hall, S.L. Liedtka, Financial performance, CEO compensation, and large-scale information technology outsourcing decisions, Journal of Management Information Systems 22 (1) (2005) 193–221. [30] H.S. Han, J.N. Lee, Y.W. Seo, Analyzing the impact of a firm's capability on outsourcing success: a process perspective, information & management, 45 (1) (2008) 31–42. [31] J.S. Harrison, M.A. Hitt, R.E. Hoskisson, R.D. Ireland, Resource complementarity in business combinations: extending the logic to organizational alliances, Journal of Management 27 (6) (2001) 679–690. [32] H.V.D. Heijden, Measuring IT core capabilities for electronic commerce, Journal of Information Technology 16 (1) (2001) 13–22. [33] L.M. Hitt, E. Brynjolfsson, Information technology and internal firm organization: an exploratory analysis, Journal of Management Information Systems 14 (2) (1997) 81–101. [34] C.B. Jarvis, S.B. Marchenzie, P.M. Podsakoff, A critical review of construct indicators and measurement model misspecification in marketing and consumer research, Journal of Consumer Research 30 (2) (2003) 199–218. [35] M. Keil, B.C.Y. Tan, K.K. Wei, T. Saarinen, V. Tuunainen, A. Wassenaar, A cross-cultural study on escalation of commitment behavior in software projects, MIS Quarterly 24 (2) (2000) 299–325. [36] T. Kern, L. Willcocks, Exploring relationships in information technology outsourcing: the interaction approach, European Journal of Information Systems 11 (1) (2001) 3–19. [37] S. Kim, Y.S. Chung, Critical success factors for is outsourcing implementation from an interorganizational relationship perspective, The Journal of Computer Information Systems 43 (4) (2003) 81–90. [38] C. Koh, S. Ang, D. Straub, IT outsourcing success: a psychological contract perspective, Information Systems Research 15 (4) (2004) 356–373. [39] M.C. Lacity, L.P. Willcocks, An empirical investigation of information technology sourcing practices: lessons from experience, MIS Quarterly 22 (3) (1998) 363–408. [40] M.C. Lacity, L.P. Willcocks, D.F. Feeny, IT outsourcing: maximize flexibility and control, Harvard Business Review 73 (3) (1995) 84–93. [41] J.N. Lee, Y.G. Kim, Effect of partnership quality on is outsourcing success: conceptual framework and empirical validation, Journal of Management Information Systems 15 (4) (1999) 29–61. [42] D.M.S. Lee, E.M. Trauth, D. Farwell, Critical skills and knowledge requirements of IS professionals: a joint academic/industry investigation, MIS Quarterly 19 (3) (1995) 313–340.

[43] N. Levina, J.W. Ross, From the vendor's perspective: exploring the value proposition in information technology outsourcing, MIS Quarterly 27 (3) (2003) 331–364. [44] N.S. Levinson, M. Asahi, Cross-national alliances and interorganizational learning, Organizational Dynamics 24 (2) (1996) 50–63. [45] D. Levinthal, Organizational capabilities in complex worlds, in: G. Dosi, R.R. Nelson, S.G. Winter (Eds.), The Nature and Dynamics of Organizational Capabilities, Oxford University Press, Oxford, 2002. [46] S. Massini, A.M. Pettigrew, Complementarities in organizational innovation and performance, in: A.M. Pettigrew, R. Whittington, L. Melin, C. Sanchez-Runde, F.V.D. Bosch, W. Ruigrok, T. Numagami (Eds.), Innovative Forms of Organizing, Sage Publications, London, 2003, pp. 133–172. [47] F.J. Mata, W.L. Fuerst, J.B. Barney, Information technology and sustained competitive advantage: a resource-based analysis, MIS Quarterly 19 (4) (1995) 487–505. [48] F.W. McFarlan, R.L. Nolan, How to manage an IT Outsourcing alliances, Sloan Management Review 36 (2) (1995) 9–23. [49] N. Melville, K. Kraemer, V. Gurbaxani, Review: information technology and organizational performance: an integrative model of IT business value, MIS Quarterly 28 (2) (2004) 283–322. [50] P. Milgrom, J. Roberts, The economics of modern manufacturing. technology, strategy, and organization, American Economic Review 80 (3) (1990) 511–528. [51] J. Mills, K. Platts, Applying resource-based theory: methods, outcomes and utility for managers, International Journal of Operations and Production Management 23 (2) (2003) 148–166. [52] D. Nath, D. Sudharshan, Measuring strategy coherence through patterns of strategic choices, Strategic Management Journal 15 (1) (1994) 43–61. [53] K.M. Nelson, J.G. Cooprider, The contribution of shared knowledge to IS group performance, MIS Quarterly 20 (4) (1996) 409–429. [54] S.L. Newbert, Empirical research on the resource-based view of the firm: an assessment and suggestions for future research, Strategic Management Journal 28 (2) (2007) 121–146. [55] J.C. Nunnally, Psychometric Theory, McGraw-Hill, NY, 1978. [56] J. Peppard, J. Ward, Beyond strategic information systems — towards an IS capability, The Journal of Strategic Information Systems 13 (2) (2004) 167–194. [57] A. Pinnington, P. Woolcock, The role of vendor companies in IS/IT outsourcing, International Journal of Information Management 17 (3) (1997) 199–210. [58] P.M. Podsakoff, S.B. MacKenzie, J.Y. Lee, N.P. Podsakoff, Common method biases in behavioral research: a critical review of the literature and recommended remedies, Journal of Applied Psychology 88 (5) (2003) 879–903. [59] J. Roberts, The Modern Firm: Organizational Design for Performance and Growth, Oxford University Press, NY, 2004. [60] J.W. Ross, C.M. Beath, D.L. Goodhue, Develop long-term competitiveness through IT assets, Sloan Management Review 38 (1) (1996) 31–42. [61] M.B. Sarkar, R. Echambadi, S.T. Cavusgil, P.S. Aulakh, The influence of complementarity, compatibility, and relationship capital on alliance performance, Journal of the Academy of Marketing Science 29 (4) (2001) 358–373. [62] C. Saunders, M. Gebelt, Q. Hu, Achieving success in information systems outsourcing, California Management Review 39 (2) (1997) 63–79. [63] T. Saxton, The effects of partner and the relationship characteristics on alliance outcomes, The Academy of Management Journal 40 (1997) 443–461. [64] J. Singh, Measurement issues in cross-national research, Journal of International Business Studies 26 (3) (1995) 597–619. [65] M. Song, C. Droge, S. Hanvanich, R. Calantone, Marketing and technology resource complementarity: an analysis of their interaction effect in two environmental contexts, Strategic Management Journal 26 (2005) 259–276. [66] D.W. Straub, Validating instruments in MIS research, MIS Quarterly 13 (2) (1998) 147–170. [67] D. Straub, M. Boudreau, D. Gefen, Validation guidelines for IS positivist research, Communications of the AIS 13 (24) (2004) 380–427. [68] H. Tanriverdi, Information technology relatedness, knowledge management capability, and performance of multibusiness firms, MIS Quarterly 29 (2) (2005) 311–334. [69] N. Venkatraman, Performance implications of strategic. Coalignment: a methodological perspective, Journal of Management Studies 27 (1) (1990) 19–41. [70] D. Whitten, R.L. Wakefield, Measuring switching costs in IT outsourcing services, The Journal of Strategic Information Systems 15 (3) (2006) 219–248. [71] Y. Yoon, K.S. Im, An evaluation system for IT outsourcing customer satisfaction using the analytic hierarchy process, Journal of Global Information Management 13 (4) (2005) 53–75. [72] K. Zhu, The complementarity of information technology infrastructure and E-commerce capability: a resource-based assessment of their business value, Journal of Management Information Systems 21 (1) (2004) 167–202.

Hyun-Soo Han is a Professor in Business School of Hanyang University in Seoul, Korea. He received a B.S. in Industrial Engineering from Seoul National University, and M.S. in Management Science at Korea Advanced Institute of Science and Technologies (KAIST). He earned his Ph.D. in Management from University of Massachusetts Amherst, USA. Before joining as a faculty member of Hanyang University, he had been with Korean IT service company as a director of the IS consulting service department. His recent research interests include IT convergence, operations strategy, IS management, mobile & e-commerce applications, and telecommunications management. His publications appear in various international and domestic journals including Information & Management, International Journal of Technology Management, International Journal of Satellite Communications and Networking, International Journal of Operations and Quantitative Management, European Journal of Operational Research, Annals of Operations Research, and so on.

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003

H.-S. Han et al. / Decision Support Systems xxx (2013) xxx–xxx Jae-Nam Lee is a Professor of MIS in the Business School of Korea University in Seoul, Korea. He was formerly on the faculty of the Department of Information Systems at the City University of Hong Kong. He holds MS and Ph.D. degrees in MIS from the Graduate School of Management of the Korea Advanced Institute of Science and Technology (KAIST). His research interests are IT outsourcing, knowledge management, information security management, e-commerce, and IT deployment and impacts on organizational performance. His published research articles appear in MIS Quarterly, Information Systems Research, Journal of MIS, Journal of the AIS, Decision Support Systems, Communications of the AIS, IEEE Transactions on Engineering Management, European Journal of Information Systems, Communications of the ACM, Information & Management, and others. He is currently serving on the editorial board of Journal of the AIS, Pacific Asia Journal of the AIS, and Electronic Commerce Research and Applications.

15

Yun-Weon Seo is a research fellow in the Center for Information Technology Studies of Security Management Institute (SMI) in Seoul, Korea. Before joining as a research fellow, he had been an officer of Republic of Korea Marine Corps. He graduated from Korea Naval Academy and holds M.S. from Korea Military Academy. He received his Ph.D. degree in Information Technology Management from Hanyang University in Seoul, Korea. His published research articles appear in Information & Management, Research and Practice in Human Resource Management, and others. His research areas are IT outsourcing, project management, information technology management, and electronic commerce.

Jae Uk Chun is an Assistant Professor of Management at Korea University. He received his Ph.D. in Management from the State University of New York at Binghamton. His current research interests include leadership, mentoring relationships, feedback-seeking behavior, and multiple levels-of-analysis incorporation into theory development, measurement, and data analysis. His work has appeared in several journals including Academy of Management Review, The Leadership Quarterly, Journal of Organizational Behavior, Group and Organization Management, and Research in Multi-Level Issues.

Please cite this article as: H.-S. Han, et al., Complementarity between client and vendor IT capabilities: An empirical investigation in IT outsourcing projects, Decision Support Systems (2013), http://dx.doi.org/10.1016/j.dss.2013.03.003