Strategic Management Journal Strat. Mgmt. J., 30: 595–616 (2009) Published online 12 March 2009 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/smj.760 Received 5 May 2006; Final revision received 12 January 2009

THE IMPACT OF OUTSOURCING NEW TECHNOLOGIES ON INTEGRATIVE CAPABILITIES AND PERFORMANCE CARMEN WEIGELT* A.B. Freeman School of Business, Tulane University, New Orleans, Louisiana, U.S.A.

Outsourcing plays an important role for firms adopting new technologies. Although outsourcing provides access to a new technology, it does not guarantee that a firm can subsequently integrate the technology with existing business processes and leverage it in the marketplace. This distinction, however, has rarely been made in the literature. In the context of business process enhancing technologies, this study builds on the resource-based and knowledge-based views to study the impact of outsourcing on firms’ subsequent performance in the market and their integrative capabilities, that is, a firm’s capacity to use and assimilate a new technology with its business processes and build upon it. The study argues that greater reliance on outsourcing may reduce a firm’s learning by doing, internal investment, and tacit knowledge applications, thereby impeding a firm’s integrative capabilities and performance in the market. The study uses survey and archival data on banks’ outsourcing strategies for Internet adoption to test for the performance consequences of outsourcing, which are found to be negative. However, the findings also show that outsourcing is less detrimental for firms with experience in prior related technology. Copyright  2009 John Wiley & Sons, Ltd.

INTRODUCTION Few firms can stay abreast of all new technology developments through internal efforts alone (Teece, 1986; Contractor and Lorange, 1988). Outsourcing relationships with technology vendors have become widespread and have evolved from outsourcing of repetitive and fairly specialized tasks to outsourcing of more complex technologies and entire business processes (Greco, 1997; Fichman and Kemerer, 1997). In 2000, over half of all information technology (IT) services in North America were outsourced (Progent Research, 2002). As the outsourcing of entire business processes becomes more common, the Keywords: outsourcing; capabilities; technology; performance; banking *Correspondence to: Carmen Weigelt, A.B. Freeman School of Business, Tulane University, 7 McAlister Drive, New Orleans, LA 70118, U.S.A. E-mail: [email protected]

Copyright  2009 John Wiley & Sons, Ltd.

questions arise of whether outsourcing is always beneficial, and whether there are limits to benefits from outsourcing. Prior research is equivocal about the performance implications of outsourcing. On the one hand, prior work views outsourcing as a means to increase efficiency, reduce costs, or foster innovation by gaining access to cutting-edge technologies, specialized resources, and learning opportunities (Hamel, 1991; Powell, Koput, and SmithDoerr, 1996; Mowery, Oxley, and Silverman, 1996; Mitchell and Singh, 1996; Brown and Eisenhardt, 1997; Poppo and Zenger, 1998). On the other hand, researchers argue that outsourcing can lead to the hollowing of corporations, the depreciation of firm capabilities (Bettis, Bradley, and Hamel, 1992), or impaired coordination across activities (Leonard-Barton, 1995; Chesbrough and Teece, 1996). Further, prior research rarely distinguishes between a firm gaining access to a new

596

C. Weigelt

technology and its subsequent ability to internalize and leverage such in the market (Hamel, 1991; Ettlie, 1988). Thus, although external partners may provide access to a technology, such access may not fully substitute for internal learning (Attewell, 1992; Fichman and Kemerer, 1997) or guarantee that a firm can use and deploy the technology in the market (Leonard-Barton, 1988; Steensma and Corley, 2000). Hence, the important question arises: When does outsourcing hurt or benefit firm performance? The distinction between gaining access to a technology and being able to effectively use it is particularly important when a firm adopts a new technology to enhance business processes, such as service delivery. This is because the use of these technologies (which include customer-facing and information-systems applications) often requires that they be assimilated into ongoing firm processes (Attewell, 1992; Purvis, Sambamurthy, and Zmud, 2001) as well as adopted by customers (Meuter et al., 2005). A firm must understand how the new technology interacts with its work processes in order to adapt and reconfigure the technology to fit the idiosyncrasies of its business (Leonard-Barton, 1988; Meyer and Goes, 1988; Attewell, 1992; Swanson, 1994). The extent to which technology assimilation triggers new routines and capabilities inside a firm (Damanpour and Evans, 1984) may explain part of the variance in benefits that firms derive from externally sourced technology. Furthermore, gaining access to a new technology does not automatically ensure that a firm can successfully deploy the technology in the market. For a firm to reap benefits from a new technology, such as cost savings from Internet banking or online reservation systems, its customers need to adopt the technology (Meuter et al., 2005). While the innovation literature focuses on customer traits and technology characteristics as predictors of adoption (Rogers, 1995), the services technology literature stresses that a better customer understanding helps firms with influencing their customers’ technology adoption (Parasuraman, 2000; Bitner, Ostrom, and Meuter, 2002). Outsourcing may impact customer adoption of a new technology by affecting the firmcustomer relationship. This article studies the impact of the degree of new technology outsourcing on a firm’s integrative capabilities which reflect a firm’s capacity to use and assimilate a new technology into Copyright  2009 John Wiley & Sons, Ltd.

its business processes and build upon it (Helfat and Raubitschek, 2000), and its performance in the market, captured as customer adoption of a new technology. A firm’s degree of outsourcing is the extent to which it relies on a third party’s expertise versus efforts of its own staff to adopt a new technology. Building on the capabilities literature (Penrose, 1959; Nelson and Winter, 1982), I argue that greater technology outsourcing reduces a firm’s learning by doing and investment in integrative capabilities. I also argue that outsourcing interferes with firm processes used to raise customers’ perceived value of a technology, thereby hurting customer adoption. Finally, I test whether firms with prior related experience (Cohen and Levinthal, 1990) are less likely to suffer the downsides of outsourcing. This study makes several contributions to the literatures on capabilities and technology sourcing. First, by distinguishing between access to a technology and subsequent capabilities related to that technology, this study addresses whether firms are likely to build integrative capabilities for externally sourced technology. In doing so, this study contributes to an emerging research stream that tries to better understand the sources of firm capabilities (Ethiraj et al., 2005). While the theoretical nature of firm capabilities as valuable, inimitable, and path-dependent processes that enable the deployment of resources and enhance firm performance is well conceptualized and empirically tested (Nelson and Winter, 1982; Barney, 1991; Amit and Schoemaker, 1993; Helfat, 1997; Yeoh and Roth, 1999), the same does not hold for our understanding of the sources of firm capabilities (Ethiraj et al., 2005) and the role outsourcing plays therein. Capabilities evolve through learning by doing (Nelson and Winter, 1982) and deliberate investment in resources and organization structures (Zollo and Winter, 2002), but the question remains whether technology sourcing prevents the development of capabilities necessary for the effective use of that technology inside an adopting firm, and, if so, what firms may be able to do to mitigate the downside of outsourcing. Second, this study provides new insights into how outsourcing impacts customer adoption of a new technology and, hence, a firm’s ability to leverage a new technology in the market. Most prior studies on outsourcing focus on manufacturing industries (e.g., semiconductor, automotive, or computer) (Dyer, 1996; Steensma and Corley Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance 2000; Leiblein, Reuer, and Dalsace, 2002) rather than on customer-facing technologies, where the customer interacts with the technology. This distinction matters because in manufacturing industries process outsourcing can be sheltered from the customer, thereby reducing potential disruptions in the firm-customer relationship due to outsourcing. For instance, despite outsourcing their shoe manufacturing, athletic footwear companies Nike and Reebok isolated their customer relationship from their outsourcing partner by retaining product design and marketing in-house (Rosenzweig, 1994). In contrast, technology outsourcing for the support of service delivery directly affects customers, given their role as ‘co-producers’ of the service (Lovelock and Young, 1979; Damanpour, 1996). For example, when outsourcing service delivery, such as call centers or Internet applications, a firm’s customers directly interact with the outsourcing partner. Customers’ willingness to use outsourced services and their perception of service quality1 affect a firm’s ability to reap benefits from outsourcing. Third, this study examines the role that prior experience with related technology plays in mitigating the potentially negative effects of outsourcing on a firm’s integrative capabilities and performance in the market. Studies building on ideas of the resource-based view and evolutionary economics note that capabilities develop in stages and build on prior related experience (Cohen and Levinthal, 1990; Helfat and Peteraf, 2003). This raises the question of whether firms with prior experience are less likely to encounter potential downsides of outsourcing because their experience enables them to learn more easily (Mowery et al., 1996). Prior experience may also enhance a firm’s understanding of the customer-technology link, thereby achieving a fit between technology attributes and customer needs despite outsourcing, which, in turn, is likely to enhance performance in the market. This study is based on both archival and survey data of 94 U.S. banks that participated in two sequential surveys on Internet outsourcing strategies. The context of Internet banking is well 1 For example, customer dissatisfaction with the service and support provided by its outsourcing partner caused Dell to scale back on its outsourcing of customer service and bring part of the function back in-house. Thus, customer discontent with the outsourced service prevented Dell from realizing all the cost savings that outsourcing promised (Johnson, 2003).

Copyright  2009 John Wiley & Sons, Ltd.

597

suited for this study because outsourcing of new customer-facing technologies is widespread in the financial services industry. Furthermore, Internet banking as a new service delivery channel affects a bank’s business processes, and its success in the market depends on customers’ adoption of the technology. Finally, PC banking, a prior technology related to Internet banking that was first introduced in the late 1980s, provides an opportunity for the analysis of the moderating impact of experience in prior related technology on the relationship between new technology outsourcing and integrative capabilities and performance in the market, respectively.

THEORY AND HYPOTHESES Technology outsourcing The condition under which technology outsourcing enhances or hurts performance is a central question for both managers and strategy researchers, and the answer may depend on the source of valuable capabilities and a firm’s ability to integrate and apply them. The resource-based view (RBV) hints at benefits from outsourcing when other firms are the source of valuable capabilities and outsourcing provides a firm with access to these capabilities (Lavie, 2006; Penrose, 1959), but cautions firms against outsourcing when valuable capabilities require learning by doing and the building of path-dependent knowledge stocks inside the firm. Thus, a firm may benefit from outsourcing if it enables the firm to enrich its knowledge stock, tap into specialized resources, and fill voids in its technology portfolio (Womack, Jones, and Roos, 1990; Powell et al., 1996; Mitchell and Singh, 1996; Mowery et al., 1996; Steensma and Corley, 2000), which, in turn, may enhance performance by increasing product variety and speed to market (Brown and Eisenhardt, 1997) or by reducing production costs (Poppo and Zenger, 1998). Furthermore, by introducing firms to new technologies and know-how, outsourcing can counter pitfalls from local search and competency traps (Levinthal and March, 1993); pitfalls that otherwise may lead to core rigidities or missed opportunities by restricting firms to the realm of their existing capabilities (Leonard-Barton, 1992). Hence, technology outsourcing can provide firms with opportunities to strengthen their capability Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

598

C. Weigelt

base and performance in the market beyond that possible through internal efforts alone. On the other hand, a central tenet of the RBV is that valuable capabilities are firm-specific and evolve within the firm (Barney, 1991; Nelson and Winter, 1982). Valuable capabilities are pathdependent and accumulate over time through learning by doing and deliberate investments in activities to build know-how (Ethiraj et al., 2005; Zollo and Winter, 2002). Building on these ideas, researchers have warned that outsourcing can lead to the hollowing of corporations and the depreciation of internal capabilities (Hamel, 1991; Bettis et al., 1992) and that therefore vertical integration is preferred. These researchers argue that outsourcing shifts investments in knowledge-building activities from the firm to the supplier, thus slowing the process of learning by doing in client firms (Bettis et al., 1992). Firms’ ability to integrate and apply external knowledge may also determine whether technology outsourcing benefits or hurts firm integrative capabilities and performance in the market. This ability, according to the knowledge-based view (KBV), depends on the tacitness of knowledge associated with a new technology and the interdependence of activities carried out between firms (Kogut and Zander, 1992; Grant, 1996). While explicit knowledge is easily exchanged among firms, tacit knowledge requires context-specific understanding to make sense, and, therefore, is ‘sticky’ to its owner and the context in which it has accumulated (Polanyi, 1967). Consistent with these ideas, Borys and Jemison (1989) find that supplier relationships for the transfer of tacit technology knowledge failed more often than those transferring codified technology. Similarly, Demsetz (1988) argues that tacit knowledge can only be obtained through outsourcing if it can be embedded in the technology itself. Thus, if knowledge can be codified, for example in blueprints or prototypes, outsourcing is likely to enable a firm to use external expertise to enhance its performance. In contrast, when technology knowledge is tacit, vertical integration facilitates its transfer through shared experience and language among organization members2 (Arrow, 1974; Monteverde, 1995) and outsourcing is likely to be inferior. 2 A stream of research on buyer-supplier relationships in the Japanese auto industry suggests that firms with contracting, or relational, capabilities are sometimes able to recreate firm-like conditions for coordination of tacit knowledge among partners

Copyright  2009 John Wiley & Sons, Ltd.

Further, the interdependence of activities that are required to integrate a new technology with internal processes affects the potential for benefits from outsourcing (Kogut and Zander, 1992; Grant, 1996). Interdependent activities require ongoing communication, knowledge exchange, and mutual adjustment between actors carrying out various parts of an activity (Thompson, 1967; Gulati and Singh, 1998). Outsourcing is likely to be beneficial if activities have low interdependence, are sequential, and can be easily divided into separate subactivities with well-understood interfaces (Wheelwright and Clark, 1992). For example, Nike and Reebok achieved cost savings from outsourcing their shoe manufacturing. One can argue, according to the KBV, that the outsourcing benefits in this case were attributable to the relative sequential and low interdependence of activities and the wellunderstood process of shoe manufacturing. Once Nike has designed a new shoe, it makes a prototype that embodies much of the tacit knowledge that went into the shoe design. This prototype, in turn, aids in handing over the subtask of shoe manufacturing to a third party. In contrast, vertical integration is more beneficial for activities that are interdependent, little understood ex ante, and require frequent knowledge exchange and learning in situ for their completion (Wheelwright and Clark, 1992; Gulati and Singh, 1998). Overall, RBV and KBV arguments imply that whether technology outsourcing benefits or hurts a firm’s integrative capabilities and performance in the market depends not only on gaining access to a technology, but also on whether a firm can integrate externally sourced technology with internal processes. Moreover, research finds that vertical integration tends to be superior to outsourcing for transferring tacit knowledge and managing interdependent activities (Kogut and Zander, 1992; Leonard-Barton, 1995; Chesbrough and Teece, 1996; Leiblein et al., 2002).

(Womack et al., 1990; Nishiguchi, 1994; Dyer, 1996; Dyer and Nobeoka, 2000). They find that many of the benefits emanating from Japanese buyer-supplier networks are driven by characteristics such as a network identity, common language, interfirm employee transfers and team formations that mimic firm mechanisms for creating, transferring, and recombining knowledge. Womack et al. (1990) note that building close interfirm ties and trust relationships took Toyota nearly 20 years, and that investments in buyer-supplier relationships like the ones made by Toyota are rather atypical in traditional Western buyer-supplier relationships. Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance When outsourcing hurts: technology outsourcing and integrative capabilities Having discussed when outsourcing is likely to benefit or hurt a firm, I turn to the type of technology studied in this article. Technologies designed to enhance business processes, such as customerfacing technologies, tend to ‘have an abstract and demanding scientific base’ that requires specialized knowledge (Fichman and Kemerer, 1997: 1346). Third parties specialize in building this expertise by accumulating knowledge from deploying these technologies in client firms, and by packaging and transmitting lessons learned through ‘generic’ applications of the technology to other clients (Attewell, 1992; Swanson, 1994). Although this makes third parties a valuable source of new technology, the idiosyncrasies of business processes require substantial modification and adjustment of the technology for use by client firms (Leonard-Barton, 1988; Fichman and Kemerer, 1997). Because the technology’s interaction with business processes is context-specific and, thus, difficult to fully anticipate ahead of deployment, user firms often discover the technology de novo as it is being deployed and have to develop new capabilities to use the technology effectively (Mowery and Rosenberg, 1989; Attewell, 1992; Swanson, 1994). Thus, technology adoption is a blend of exploiting a third party’s technology knowledge and creating new capabilities for its integration and use (Mowery and Rosenberg, 1989; Swanson, 1994). These new integrative capabilities reflect a client firm’s skills in tailoring a technology to firmspecific needs to enhance its application (Helfat and Raubitschek, 2000; Iansiti and Clark, 1994). Outsourcing is likely to negatively impact an adopting firm’s subsequent integrative capabilities in the context of business process enhancing technologies. First, as outsourcing increases, it limits a firm’s exposure to a new technology, and hence learning by doing, which plays an important role in the development of integrative capabilities to use and apply a new technology in the firm’s business processes (Rosenberg, 1982; Ethiraj et al., 2005). Learning by doing to develop integrative capabilities is an iterative process of successive trials that occur as the firm experiments with a new technology, responds to updates of the technology, and discerns its best uses depending on the technology’s interactions with its business processes Copyright  2009 John Wiley & Sons, Ltd.

599

(Leonard-Barton, 1988; Attewell, 1992). Managing the interplay between technology and firm processes is likely to benefit from the coordination and free flow of information in firms, whereas challenges arise when activities are split among firms due to outsourcing (Leonard-Barton, 1995). Second, capabilities evolve through not only learning by doing, but also deliberate investment in internal processes (Ethiraj et al., 2005; Zollo and Winter, 2002) that facilitates the building of knowhow through a shared understanding around a new technology. As outsourcing increases, it diverts capital and time investments away from internal infrastructure to the management of external relationships. A firm may spend, even in the absence of opportunism, increasing energy on negotiations and convincing third parties of appropriate actions (Conner and Prahalad, 1996). In addition, the firm is likely to allocate fewer resources and talent to areas that are outsourced and, therefore, often perceived as noncore, which leads to a neglect of capabilities in those areas (Leonard-Barton, 1992; Bettis et al., 1992). This lack of internal resource allocation for the use of a new technology is likely to impair the building of integrative capabilities related to the technology. Third, given the difficulty of transferring tacit knowledge across firm boundaries, outsourcing may limit a firm’s insights into codified components transferable with the technology, thereby potentially causing a firm to make faulty assumptions and conclusions about the technology (Cohen and Bacdayan, 1994; Pisano, 1996). Discerning the often unclear cause-and-effect links between a new technology and its applications to business processes requires transferring not only the technology itself, but also its underlying tacit knowledge (Attewell, 1992; Fichman and Kemerer, 1997). Knowledge-transfer problems may be exacerbated if third parties share less technological knowledge than adopting firms require to effectively use a technology within their business processes. Hence, without significant involvement, adopting firms may be unable to understand the causal ambiguities surrounding a new technology, which, in turn, is likely to limit the integrative capabilities they can develop. Hypothesis 1a: The higher a firm’s degree of outsourcing for a new business process enhancing technology, the lower its subsequent integrative capabilities related to the technology. Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

600

C. Weigelt

When outsourcing hurts: technology outsourcing and performance in the market Access to a new technology via outsourcing does not guarantee that a firm can leverage the technology in the market (Steensma and Corley, 2000). The performance benefits that a firm reaps from customer-facing technologies will greatly depend on the extent to which customers adopt the technology (Meuter et al., 2005).3 However, customer adoption of a technology cannot be taken for granted (Rogers, 1995). Similar to organizational members who often face considerable latitude in using a new technology their firm adopts (LeonardBarton and Deschamps, 1988), customers can often choose whether or not to adopt (Curran and Meuter, 2005). Technologies often fail in the market due to a lack of customer adoption that results partly from an inability of the adopting firm to demonstrate the technology’s value, relevance, and benefit to its customers (Rogers, 1995; Meuter et al., 2005). Literature on technology adoption has isolated perceived ease of use and usefulness as factors that affect customers’ acceptance of a new technology (Davis, 1989; Rogers, 1995). Since learning cost, perceived risk, expected time and cost savings associated with using a technology vary for different customer groups, an adopting firm needs to understand its customers and the reasons why some customers opt out (Parasuraman, 2000; Curran and Meuter, 2005). The technology services literature argues that such understanding enables a firm to target communication, education, customer-friendly instructions or aids to their customers’ specific technology concerns, thereby increasing the value that customers perceive from using the technology (Bitner et al., 2002; Durkin et al., 2003; Meuter et al. 2005). Thus, for new technologies to receive high customer adoption, a firm needs to ensure that the technology’s applications meet its customers’ needs (Utterback, 1974; Rogers, 1995). Therefore, a firm’s performance in the market is assessed as customer adoption, that is, the extent to which a firm’s customers use a new technology. Outsourcing may negatively affect a firm’s performance in the market. First, gaining customer 3 For instance, for a bank to realize the cost savings associated with customers migrating from branch to Internet transactions (costing $1.07 versus 1 cent each) (Dandapani, 2004), its customers need to actually use the Internet.

Copyright  2009 John Wiley & Sons, Ltd.

knowledge to tailor technology attributes and communication to customer needs depends on a set of tacit, interdependent processes, such as collecting information on customers’ technology perceptions, interpreting the information in the context of a firm’s market, disseminating it throughout the organization, and acting upon it (Day, 1994; Kohli and Jaworski, 1990). Executing these processes in an iterative fashion, where a firm’s learning about new technology features alternates with obtaining customer feedback on the technology, is crucial for fitting a new technology with customer needs to ensure customer adoption. As outsourcing increases, the extent to which a firm gathers and applies customer knowledge to fit technology features to customer needs declines, which, in turn, is likely to reduce customer adoption. Second, the process of tailoring technology applications to customer tastes is complicated by customers’ variability in preferences: the service level they request, the learning effort they are willing to make, and the service quality they expect from a new technology (Frei, 2006). This variability across customers makes fitting a technology to customer needs a process of successive approximation and trial-and-error learning that is difficult to plan in advance and standardize (Attewell, 1992). Therefore, increased technology outsourcing is likely to result in frequent updating and renegotiating of contracts as new knowledge about customer preferences is discovered. Renegotiations can be time-consuming (Conner and Prahalad, 1996) and can interfere with timely adaptations of the technology to customer needs. Furthermore, outsourcing introduces an extra layer between the firm and its customers, namely the third party providing the technology, which is likely to reduce the extent to which customer needs are reflected in technology applications (Fornell, 1992). Consequently, as outsourcing increases, performance in the market is likely to decline. Third, since customers interact with a technology during service delivery, they become a key input factor to the ‘production process’ (Lovelock and Young, 1979). This active role of the customer makes it difficult to separate service delivery into independent sub-activities that can be executed independently by holders of technology knowledge and holders of customer knowledge. In order to reduce service failures, a firm needs to stay close to its customers to learn about their interactions with the technology and define the customer’s role in Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance service delivery (Meuter et al. 2005). By increasing a firm’s distance to its customers, outsourcing may cause a rise in service delivery glitches, which, in turn, is likely to reduce performance in the market. Hypothesis 1b: The higher a firm’s degree of outsourcing for a new business process enhancing technology, the lower its subsequent performance in the market. Limits to pitfalls from outsourcing: the role of experience in prior related technology The performance impact of outsourcing may depend on whether a firm can absorb external knowledge. The ease with which knowledge can be transferred is, in part, influenced by the recipient firm’s absorptive capacity to interpret, digest, and assimilate external knowledge (Cohen and Levinthal, 1990). Absorptive capacity has been shown to predict innovative activity (Cohen and Levinthal, 1990), research productivity (Cockburn and Henderson, 1998), and the extent of managerial IT use (Boynton, Zmud, and Jacobs, 1994). Studying the adoption of video banking, Pennings and Harianto (1992) show that experience with prior technology applications facilitated a bank’s adoption of video banking. Rosenberg (1990) further highlights the relevance of prior experience by noting that prior experience ‘requires a substantial research capability to understand, interpret, and appraise knowledge that has been placed upon the shelf’ (Rosenberg, 1990: 171, italics in original). Hence, experience in prior related technology may limit the downside of outsourcing by fostering a firm’s ability to integrate external technology with firm processes. Prior experience provides a firm with the capacity to better understand the cause-and-effect relationships underlying a new technology, make sense of it, and integrate it with firm processes (Fichman and Kemerer, 1997). Drawing on prior related experience for applications of new knowledge decreases the likelihood of errors and false starts (Cohen and Bacdayan, 1994), thereby providing the platform on which to build new capabilities (Nelson and Winter, 1982; Helfat and Peteraf, 2003). Thus, a firm may be able to substitute prior related experience for the tacit knowledge components of a technology that are difficult to transfer, thereby reducing the negative effect of outsourcing on integrative capabilities. Copyright  2009 John Wiley & Sons, Ltd.

601

Also, as mentioned earlier, managers often hold areas whose activities are outsourced in low regard resulting in meager allocation of talent and resources to those areas (Leonard-Barton, 1992), which reduces the likelihood of internal capability development (Bettis et al., 1992). Firms with prior experience may be able to counter this tendency by having built expertise in areas currently affected by outsourcing. With prior experience, firms may be less inclined to discredit the importance of integrative capabilities in areas affected by outsourcing. Hypothesis 2a: Firms with experience in prior related technology exhibit less of a decline in subsequent integrative capabilities as their degree of outsourcing for a new business process enhancing technology increases. Similarly, a firm’s prior related experience may lessen the negative effect of outsourcing on a new technology’s performance in the market by enabling the firm to build on prior customer-related knowledge. Such knowledge may have accumulated by studying customers’ reactions to prior technology, experimenting with prior technology at the customer interface, or by previously applying technology to service delivery. Hence, experience may enable a firm to more easily address customer variability in preferences and apply technology to fit customer needs, even as outsourcing increases. The benefits of prior experience may translate into less need for basic learning by doing to yield customer adoption because the firm has already engaged in those learning activities in the past (Cohen and Levinthal, 1994; Pisano, 1990). In addition, by enhancing a firm’s familiarity with a new technology, prior related experience increases a firm’s ability to split technology-related activities into sub-activities and recombine them (Eisenhardt and Tabrizi, 1995). By understanding the interfaces between a technology’s components and firm processes, a firm may be able to more successfully manage the process of collecting information on customer interactions with an outsourced technology, thereby reducing the negative effect of outsourcing on performance in the market. Hypothesis 2b: Firms with experience in prior related technology exhibit less of a decline in subsequent performance in the market as their Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

602

C. Weigelt

degree of outsourcing for a new business process enhancing technology increases. In summary, as a firm’s technology outsourcing increases, its integrative capabilities and performance in the market decrease. However, prior related experience mitigates this negative effect of outsourcing. It follows that pursuing higher degrees of technology outsourcing is particularly disadvantageous for firms lacking prior related experience.

processing system (Federal Financial Institutions Examination Council [FFIEC], 2003 : 22). Thus, Internet banking is ‘the automated delivery of new and traditional banking products and services directly to customers through electronic, interactive communication channels’ (FFIEC, 2003: 1). Online financial services include such basic services as account inquiry and fund transfers as well as more Web-enabled financial services such as account aggregation and electronic bill payment.

DATA METHODS Internet banking Since Internet banking is a new technology that spread into the service sector via technology vendor relationships, it is a suitable setting for studying the impact of new technology outsourcing on integrative capabilities and performance. In 2002, banks spent an estimated $130 billion on IT, of which at least 35 percent went to outsourcing (McKendrick, 2002). Technology vendors specialize in Internet banking solutions and gain expertise by generalizing lessons learned from installing their solutions at multiple banks and by subsequently embedding those lessons learned into prepacked, ‘off-the-shelf’ software solutions. These solutions tend to be ‘generic’ in that they are tailored to the banking sector, but they lack the particularities of a specific bank’s business processes. Hence, contracting with technology vendors may reduce the necessity for a bank to acquire extensive software programming skills, but it does not eliminate the need for internal capabilities to customize, understand, integrate, and use a new technology with internal business processes (Helfat and Raubitschek, 2000). Building integrative capabilities related to a new technology is likely to ensure better integration of the vendor’s products with those of the bank and, thus, better customer satisfaction and customer online service adoption. Banks vary in the extent to which they source Internet solutions from technology vendors, such as Fiserv, EDS, or Jack Henry & Associates, and employ internal staff to customize and integrate these solutions with their business processes. An Internet banking solution is a front-end system that relies on a programming link to transfer information entered by customers online to the bank’s core Copyright  2009 John Wiley & Sons, Ltd.

I collected data from banks’ annual Reports on Condition and Income and conducted two surveys on U.S. banks’ Internet outsourcing activities during the winter of 2001/2002 and the spring of 2003. The second survey was a follow-up to all banks participating in the first survey. For the first survey I randomly sampled 800 bank holding companies (BHCs) with more than $100 million in assets from the Federal Deposit Insurance Corporation’s (FDIC) list of 2,512 BHCs. The FDIC’s Web site listed 5,065 FDICinsured BHCs as of December 2000. In cases of multibank holding companies, I chose the largest entity within the holding company. I contacted each bank by phone to obtain the name and contact information of the most senior executive in charge of the bank’s Internet initiative. Since I could not gather contact information for 32 banks, most of which had been acquired during 2001, the survey sample includes 768 both privately and publicly held banks. To prepare the survey, I reviewed bank Web sites and press releases and conducted interviews with bank executives in charge of their bank’s Internet activities. I designed the survey using measures from the Office of the Comptroller of the Currency Internet banking studies, market research studies, interviews with industry experts, and prior research on innovation and strategy. Eight experts who were either senior bank executives or industry analysts participated in the pilot testing of the survey. I sent the first survey to informants during the winter of 2001/2002 and administered two followup postcard mailings and follow-up phone calls to banks during February to mid-April 2002. The informants were senior executives in charge of their banks’ Internet initiatives. I received replies from 224 banks, which is a response rate of 29 Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance percent. Accounting for 21 acquired banks in the sample of 768 banks, the response rate rose to 30 percent. In April 2003, I sent a follow-up survey to the 224 senior executives who participated in the first survey or, in cases of job changes, to the senior executive now in charge of the bank’s Internet activities. The survey focused on Internet banking capability and performance measures and repeated some measures from my first survey. I conducted follow-up postcard mailings and follow-up phone calls during May and June 2003. A total of 132 banks responded to the second survey, which is a response rate of 59 percent. Accounting for nine banks in the sample that had been acquired during 2002 and early 2003, the response rate increased to 64 percent. The 224 banks replying to the first survey held assets between $100 million to over $10 billion as of December 2001. I conducted t-tests between respondents and nonrespondents and between earlier and later respondents for size, number of banks in a BHC’s structure, and financial condition (loans/deposits) to examine the data for potential nonrespondent bias. The t-tests were not significant, indicating an absence of a nonrespondent bias between respondents and nonrespondents (size: t = −0.18; BHC’s structure: t = 1.15; financial condition: t = 1.07) and between earlier and later respondents (size: t = −1.03; BHC’s structure: t = −0.71; financial condition: t = −1.08) to the first survey. T-tests for the second survey were also not significant (size: t = −1.61; BHC’s structure: t = −1.62; financial condition: t = −0.57). Further, t-tests comparing time of adoption (t = 0.74), degree of initial outsourcing (t = −1.45), and online retail service offerings (t = −1.32) for respondents and nonrespondents to the second survey were not significant. Gerhart, Wright, McMahan, and Snell (2000) caution against rater reliability issues in survey research. To increase rater reliability, both surveys had global and specific measures to assess informants’ knowledge of their banks’ Internet activities (Kumar, Stern, and Anderson, 1993). About 70 percent of respondents had titles at the vice president level or higher. Titles were in ecommerce (37%), operations (25%), IT (16%), marketing (9%), sales (6%), finance (2%), CEO (2%), and other-not specified (4%), which reflects earlier findings that the area that originally championed Internet banking varies across banks (BAI, Copyright  2009 John Wiley & Sons, Ltd.

603

1999). Respondents to the first survey had, on average, a job tenure of 3.8 years and a bank tenure of 8.3 years. Except for 24 respondents who underwent job changes, all respondents were identical in both surveys. To ensure respondents’ competence, I asked each informant to rate his or her personal involvement in (1) the bank’s online initiatives, (2) the addition of new online functionality, and (3) the selection of Internet supplier/partners on a seven-point Likert-type scale. The three items were reliable for both surveys (α = 0.88 and α = 0.85, respectively) and respondents were highly involved (mean = 6.03, S.D. = 1.23, and m = 5.95; S.D. = 1.26, respectively). Although multiple informants are preferred in surveys (Golden, 1992; Kumar et al., 1993) I used only a single informant because utilizing multiple informants from a single firm when a single informant is most knowledgeable often creates problems (Glick et al., 1990). Gerhart, Wright, and McMahan note that single raters tend to be more reliable in smaller rather than in larger firms due to substantial within-firm variation in larger firms; for example, different policies across business areas are a major source of rater reliability bias at the firm level (Gerhart, Wright, and McMahan, 2000: 867). Given the small size of most banks’ e-business units (71% of e-business units had five or fewer employees) and because I asked informants to assess only Internet activities directly related to their area, I believe rater reliability bias is not a serious concern in this study. Common method variance is a potential shortcoming in survey-based research that collects dependent and independent variables from the same respondent (Podsakoff and Organ, 1986). However, I believe that the validity of my findings is not subject to common method bias since the dependent and independent variables come from two different surveys conducted more than a year apart, which reduces correlation among dependent and independent variables that may be attributable to common method bias (Podsakoff and Organ, 1986; McEvily and Chakravarthy, 2002). Furthermore, although the same executives were targeted by both the first and second survey, 24 of the respondents in the second survey differed from those in the first survey because of job changes. Hence, in 24 cases the independent and dependent variables come from different sources. This provides me with the opportunity to test whether the Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

604

C. Weigelt

responses to the dependent measures differ significantly between new and incumbent respondents. I conducted t-tests for the responses to the three items forming the capability measure (t = 0.98; p = 0.33; t = 1.61; p = 0.12; t = 1.53; p = 0.13) and for customer adoption (t = 0.37; p = 0.72). These t-tests were not significant at the p < 0.10 level, implying that the two groups of respondents did not significantly differ in their responses, further indicating that common method bias is not a concern. Variable Measurements Dependent variables The article studies the impact of outsourcing on two dependent variables collected in the second survey: integrative capabilities and performance in the market (customer adoption). Prior work has measured capabilities as number of patents (Mowery et al., 1996), products (Yeoh and Roth, 1999; Katila and Ahuja, 2002), research and development (R&D) expenses (Cohen and Levinthal, 1990; Helfat, 1997), or latent constructs based on surveys (Steensma and Corley, 2000). I follow Steensma and Corley (2000) by measuring integrative capabilities using a latent construct based on three seven-point Likert scale items. Informants4 rated the degree to which their bank (a) is capable of customizing standardized off-the-shelf technology to its Internet applications, (b) is capable of developing future applications of Internet banking services, and (c) has adequate IT skills to operate Internet banking in-house (Appendix). The latent construct of integrative capabilities captures the extent to which a firm is able to assimilate, enhance, and apply a new technology to its internal processes (Mytelka, 1985; LeonardBarton, 1988; Helfat and Raubitschek, 2000). The standardized Cronbach coefficient alpha for the three items was α = 0.85. I also conducted a factor analysis to ensure that the items loaded on one factor. I combined and averaged the three 4 Huber and Power (1985) raise concerns regarding raters’ potential desirability biases and intentional distortions in surveys. However, although it would be favorable for all respondents to report high integrative capabilities for their bank, the mean for this variable is only 3.82 (S.D. = 1.72). A frequency distribution for the capability variable shows that less than 15 percent of respondents reported a score of six or higher, and that less than 25 percent reported a score of five or higher. Hence, a desirability bias seems unlikely to exist in this survey data.

Copyright  2009 John Wiley & Sons, Ltd.

items to create the variable integrative capabilities. Higher values indicate greater integrative capabilities. Isolating the performance of online banking from that of other service delivery channels is challenging; since the inception of Internet banking, banks have struggled to assess the effectiveness of their online operation and its contribution to overall financial performance. Banks are different from other industries that service their customers through multiple channels such as retailing (e.g., Williams-Sonoma or Barnes & Noble) in that customers utilize their bank’s services through multiple channels (e.g., ATMs, branches, call centers, or online) without being charged on a transactionby-transaction basis.5 It is therefore nearly impossible to trace revenues to specific transactions, and hence specific service delivery channels. Although industry analysts have observed that customers who bank online have more products, lower attrition rates, and, on average, 35 percent higher balances than their offline counterparts (Bielski, 2003), the causal chain of whether online customers are more profitable due to banking online or whether more profitable customers bank online cannot be established. Second, tracing costs to specific channels and specific customers is also difficult. Technology is an integral part of a bank’s operations, and several banking channels (e.g., online banking platform, branch, and ATM networks) utilize and draw on the same back-end data processing for their operation. Faced with these limitations, banks use ‘total number of active online customers’ as a way to assess the performance of their online banking channel. This is a suitable performance measure because, while revenues cannot be traced by delivery channel, it is known that online banking is a

5

A retailer such as Barnes & Noble can measure the profit it earns from an online customer versus a retail store customer because each engages in a specific transaction that is documented. That is, revenues and costs for the transaction are measured (e.g., sales price, costs of goods sold, handling and shipping). In contrast, banks cannot track how profitable one channel is versus another. The reason for this is that a customer uses multiple channels for the same banking products without being charged for each individual transaction. Thus, a customer may open a bank account in a branch, withdraw money from the ATM, request a loan over the phone, and check balances and pay bills online. Given this, customer profitability in banking is notoriously difficult to attribute to a specific channel. Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance lower-cost channel than branch and ATM (Dandapani, 2004; Durkin et al., 2003). Migrating customers to online banking is therefore likely associated with increased cost savings for banks. In addition, in the academic literature, customer adoption and usage of a new technology are viewed as measures of new technology acceptance and success (Rogers, 1995; Leonard-Barton and Deschamps, 1988). Although not directly capturing a firm’s online sales and profits, a higher percentage of customer online adoption signals a firm’s ability to penetrate a new market. Therefore, online customer adoption is an appropriate, and widely used, proxy for the performance of Internet banking operations. Accordingly, I measure performance in the market as the percentage of a bank’s total customer base (demand-deposit households) that regularly checks balances online. Respondents reported their total customer base checking balances online as of the spring of 2003, the time of the second survey. To ensure the measure’s validity, I separately asked respondents the following two questions and then calculated a customer online adoption measure: What is your bank’s total number of active online customers? What is your bank’s total number of retail customers (offline and online)? I reestimated the models with this second performance measure and the results were consistent. The correlation between both measures is r = 0.88. This performance measure considers that technologies adopted in business processes or units may not directly affect overall return on equity (ROE) or return on assets (ROA) (Ray, Barney, and Muhanna, 2004).6 Independent variables All independent variables were collected in the first survey. Degree of outsourcing captures the extent to which a firm relies on external vendors to adopt a set of eight online service areas. Prior literature tends to conceptualize interfirm modes as binary (Leiblein and Miller, 2003) or as an ordinal continuum of interfirm relationships (e.g., 6 The logic is that as the performance of one business unit increases thereby positively contributing to ROE (ROA), another business unit may perform poorly resulting in a negative effect on ROE (ROA). Both effects may cancel each other out leading to no visible effect from e-banking on aggregate performance measures such as ROE or ROA.

Copyright  2009 John Wiley & Sons, Ltd.

605

licensing, equity stakes, joint ventures, and acquisition) (e.g., Steensma and Corley, 2000; NichollsNixon and Woo, 2003), although some of these choices (e.g., licensing agreements) can be continuous. Therefore, following Pisano (1990) and Poppo and Zenger (1998), degree of outsourcing is a continuous variable ranging from zero (all in-house) to 100 (all outsourcing). Using external vendors to acquire new technology is prevalent in the banking industry with only a few banks among the largest relying solely on in-house processing. In the sample, only one bank did so; all others used varying degrees of internal and vendor involvement to implement eight online service areas: account balance inquiry and funds transfer, credit/loan/mortgage, bill payment, bill presentment, investment, insurance, CRM, and nontraditional services. Eight banking experts evaluated the list of service areas to ensure it was representative of online service offerings. Informants reported their bank’s percentage of in-house development versus outsourcing for each service area (Appendix). I calculated degree of outsourcing by summing the percentages of outsourcing across all service areas offered and dividing the sum by the total number of online service areas offered. The measure includes service areas adopted completely in-house and accounts for firms differing in their degree of outsourcing across service areas. Experience in prior related technology may influence a firm’s performance in the market and integrative capabilities by providing learning opportunities that foster absorptive capacity (Cohen and Levinthal, 1990). For over 20 years, banks have attempted to shift their customers from physical branches to remote channels, such as home banking, ATMs, and smart or debit cards (Pennings and Harianto, 1992). I define experience in prior related technology as previous PC banking offerings that are a predecessor technology to Internet banking. PC banking familiarized banks with remote customer self-servicing technology and offered services similar to the basic services offered by Internet banking. Although PC banking included several of today’s basic online services, it required users to install the bank’s software on their PCs to access accounts remotely. Experience in prior related technology is a binary variable of one for banks that had at any point in time a PC direct-dial program for their retail customers and of zero otherwise. Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

606

C. Weigelt

Control variables I control for firm- and relationship-specific factors that may influence a firm’s capabilities and performance in the market. Firm size is an archival measure taken from each bank’s annual Report on Conditions and Income. In keeping with prior research in banking, firm size is measured as the log of assets (Dos Santos and Pfeffers, 1995). Larger firms may have greater capacity, slack, and incentives to acquire new capabilities due to their scale (e.g., broad service portfolio) or fixed-cost spreading advantages (e.g., large customer base) (Cohen, 1995). Technical and marketing investments are measured as composite formative indicators (Bollen and Lennox, 1991). Technical investments consist of technological intensity and IT strength that measure tangible and intangible aspects of resource investments, respectively (Amit and Schoemaker, 1993). Technological intensity is measured as a bank’s IT investments in systems, equipment, and data processing divided by total assets (Pennings and Harianto, 1992). IT strength is a reflective measure from the first survey and comprises the seven-point scale items: (a) a bank’s overall technology/IT knowledge and (b) IT investments/budget upon Web launch, which correlate highly (r = 0.59). I created the formative measure technical investments by summing and averaging the z-scores for technological intensity and IT strength. Similarly, I created the formative measure marketing investments by summing and averaging the z-scores for market scope and marketing intensity. Market scope is the percentage of income derived from nontraditional banking sources and captured as ratio of noninterest income to total income. Marketing intensity is measured as a bank’s advertising expenses divided by revenues. Time of adoption is the number of months since year-end 1995 before a bank started to adopt Internet banking. The variable controls for market learning effects and technological advancement that occur as new technologies diffuse through an industry (Schoenecker and Cooper, 1998). Scope is measured as the number of service areas (Appendix) in which a bank initially adopts online services. Greater scope has been shown to increase the complexity of a new technology and the coordination efforts required to adopt it (Novak and Eppinger, 2001). Copyright  2009 John Wiley & Sons, Ltd.

I measure number of external partners as the average number of vendors that a firm had across eight online service areas at time of adoption (Appendix). Contracting with more external parties may increase a firm’s external coordination efforts and divert attention from cultivating capabilities due to bounded rationality (Simon, 1960). Collaboration with partner is a firm’s average degree of collaboration with its partner across eight online service areas (Appendix). Informants rated on a seven-point scale the extent to which their vendor relationship was, on average, armslength or collaborative, which controls for the extent of interaction and knowledge exchange that may occur in external relationships (Mitchell and Singh, 1996). Higher net worth customers, who tend to be more knowledgeable about the Internet and demand more sophisticated services, may be more likely to adopt Internet services (Rogers, 1995). I measure customer affluence as the ratio of deposit accounts over $100,000 divided by all bank deposits. Finally, I control for the number of employees in a bank’s e-business unit. This variable, measured as the log of a bank’s number of e-business employees, controls for the costs and efforts a firm expands internally to adopt a new technology. At the market level, I control for market concentration, which is the four-firm concentration ratio of deposits in a bank’s market defined by the number of states in which that bank has operations. Literature in economics attests to links between market concentration and innovative activity, arguing that market concentration may negatively affect innovative activity by inhibiting competition and potential returns from innovation (Levin, Levin, and Meisel, 1987; Acs and Audretsch, 1987). Hence, market concentration may limit a firm’s incentive to invest in building capabilities related to a new technology since the returns may not be worth the effort in the presence of limited competition due to high concentration. Analysis Estimating the effect of outsourcing on both integrative capabilities and performance in the market (customer adoption) requires two important considerations. First, since managers tend to choose their degree of technology outsourcing based on their expectation that a certain degree of outsourcing will yield greater returns than another, a firm’s Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance outsourcing decision may introduce endogeneity bias (Hamilton and Nickerson, 2003). Second, outsourcing is likely to have a simultaneous rather than sequential effect on integrative capabilities and performance in the market. The process of capability building is likely to occur at the same time as customer adoption. Firms build capabilities for a new technology as they offer it to customers, as they receive feedback from customers adopting the technology, and as they tweak the technology. Regarding the first consideration, I test for potential endogeneity applying the DavidsonHausman-Wu test of endogeneity (Wooldridge, 2003). The hypothesis for the presence of endogeneity in degree of outsourcing was rejected for both dependent variables (integrative capabilities t = 1.24, p = 0.27 and performance in the market t = 1.10, p = 0.30). Thus, potential endogeneity of degree of outsourcing does not seem to be a problem in this dataset7 . Given the absence of endogeneity of outsourcing in the model, but considering the concern for potential simultaneity of integrative capabilities and performance, I apply a seemingly unrelated regression (SUR) model (Zellner, 1962) using STATA to test the hypotheses. I estimate the following set of equations where Xi and Zi are vectors predicting the respective dependent variable. Integrative capabilitiesi = β0 + β1 degree of outsourcingi + β2 Xi + εi

(1)

Performance in the marketi = β3 + β4 degree of outsourcingi + β5 Zi + ηi

(2)

An SUR is an extension of a linear regression that permits correlated errors between equations. Correlation in error terms between equations with different dependent variables is particularly likely when both equations utilize the same dataset. Further, given the absence of endogeneity of outsourcing in the model, an SUR model is preferable to a 7 Given the absence of endogeneity of outsourcing in this dataset, the use of a two-stage least squares (2SLS) or three-stage least squares (3SLS) approach is not necessary. While Heckman twostage models are commonly used to control for the endogeneity of dichotomous or categorical variables capturing outsourcing, a 2SLS or 3SLS approach is used to account for the potential endogeneity of continuous variables (Hamilton and Nickerson, 2003).

Copyright  2009 John Wiley & Sons, Ltd.

607

3SLS because it does not require instruments, and hence is likely to yield more precise estimates.8 I also estimate two-limit Tobit regressions for integrative capabilities bound between one and seven and performance bound between one percent and 100 percent customer adoption. Tobit is appropriate for dependent variables that have observations that are censored by a lower and upper limit bound. Tobit estimates the likelihood that a dependent variable exceeds a threshold value (zero for no customer adoption) and its value, if it exceeds the threshold. In this study, Tobit estimates both the likelihood that a new technology yields customer adoption and the extent to which it does so.

RESULTS Table 1 presents means, standard deviations, and correlations for the measures. Degree of outsourcing correlates negatively with integrative capabilities (r = −0.43) and performance (r = −0.53). Experience in prior related technology correlates positively with the dependent variables (r = 0.20 and r = 0.12, respectively) and negatively with degree of outsourcing (r = −0.21).9 Integrative capabilities and performance in the market estimates Table 2 presents the results for the SUR testing the relationship between degree of outsourcing and integrative capabilities (Hypothesis 1a) and performance in the market (Hypothesis 1b) in Model 1 of Equations 1 and 2, respectively. Model 1 of Equation 1 predicts the effect of outsourcing on integrative capabilities, which is significant and negative (β = −0.023; p < 0.001), indicating that greater outsourcing of new technology 8 To ensure the robustness of the model results presented in this article, I estimated several additional models. First, even though the Davidson-Hausman-Wu test of endogeneity showed that potential endogeneity resulting from degree of outsourcing was not a concern in this data, I estimated a 3SLS regression model (Wooldridge, 2003). The results for the hypotheses were consistent with those reported in Table 2. Second, I estimated separate 2SLS models for both integrative capabilities and performance in the market, respectively, which account for potential endogeneity of outsourcing, but not for potential simultaneity. Again, the results were consistent with those shown in the study. These tests ensure the robustness of the SUR results shown in this article. 9 Although 132 banks replied to both surveys, some banks had incomplete data reducing observations to 94.

Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

1.000

C. Weigelt

1.000 −0.037 −0.026 −0.053 0.139 1.000 −0.008 −0.015 0.069 −0.001 −0.140

Copyright  2009 John Wiley & Sons, Ltd.

Note: N = 94 Correlations greater than r = 0.19 are significant at p < 0.05 greater than r = 0.25 are significant at p < 0.01

1.000 −0.031 −0.048 −0.235 −0.298 0.289 0.110 1.000 0.245 0.008 0.075 −0.059 −0.268 0.279 0.325 1.000 −0.057 −0.036 −0.035 0.059 −0.023 −0.130 0.206 −0.006 1 Capabilities 2 Performance (customer adoption) 3 Firm size 4 Time of adoption 5 Experience in prior related tech. 6 Technical investments 7 Marketing investments 8 Scope 9 Number of external partners 10 Collaboration with partner 11 Market concentration 12 Degree of outsourcing 13 # of employees in e-business unit 14 Customer affluence

3.82 16.02 14.03 46.61 0.32 −0.01 0.02 3.51 1.26 4.55 0.42 86.99 1.10 0.30

1.72 10.35 1.64 17.70 0.47 0.86 0.71 1.87 0.46 1.86 0.09 22.66 0.82 0.12

1.00 1.00 11.70 0.00 0.00 −1.67 −1.30 0.00 0.00 1.00 0.18 0.00 0.00 0.08

7.00 40.00 19.05 74.00 1.00 6.39 2.22 8.00 4.00 7.00 0.65 100.00 3.69 0.74

1.000 0.493 0.409 −0.309 0.197 0.001 0.263 0.096 −0.171 −0.089 −0.364 −0.429 0.232 0.140

1.000 0.339 −0.279 0.118 0.057 0.136 0.191 −0.008 0.181 −0.203 −0.525 0.199 0.107

1.000 −0.443 0.264 −0.101 0.280 0.440 −0.089 −0.068 −0.453 −0.397 0.404 0.144

1.000 −0.274 0.071 −0.291 −0.215 0.137 −0.074 0.122 0.341 −0.116 −0.133

1.000 −0.066 0.249 0.269 −0.118 0.009 −0.212 −0.207 0.191 0.220

7 Max Min S.D. Mean

Table 1. Measure characteristics and correlations

1

2

3

4

5

6

8

9

10

11

1.000 0.201 −0.099 −0.093

12

1.000 −0.255 −0.114

13

1.000 −0.059

14

608

leads to lower integrative capabilities. This confirms Hypothesis 1a. Among the control variables, scope, number of external partners, and market concentration exhibit a negative, significant effect on integrative capabilities. Model 1 of Equation 2 predicts the effect of outsourcing on performance in the market. Degree of outsourcing is negative and significant (β = −0.197; p < 0.001), indicating that greater outsourcing for a new technology lowers performance in the market, supporting Hypothesis 1b. Among the control variables, collaboration has a positive significant effect on performance in the market. I also estimated Tobit models predicting integrative capabilities and performance in the market (Hypotheses 1a and 1b). The results are consistent and presented in Table 2.

Estimates of the interaction effect Model 2 presents results of the interaction effect between degree of outsourcing and experience in prior related technology, testing Hypotheses 2a and 2b that predict that the negative impact of outsourcing on integrative capabilities (Hypothesis 2a) and performance in the market (Hypothesis 2b) is less for firms with experience in prior related technology. The interaction is positive and significant in Model 2 of both Equations 1 and 2 (β = 0.043; p < 0.001 and β = 0.197; p < 0.01, respectively). Additionally, I estimated Tobit models to test Hypothesis 2b. Consistent with the results in Model 2 of Equation 2, the interaction effect is positive and significant (β = 0.222; p < 0.001). To gain further insights into the interaction between experience in prior related technology and outsourcing, I graphed the interaction for firms with and without prior experience at different degrees of outsourcing in Figure 1a and 1b (Aiken and West, 1991). The graph in Figure 1a shows that as outsourcing increases, firms with prior experience exhibit far less of a decline in integrative capabilities than firms without prior experience. In fact, the line is close to horizontal for firms with prior experience. Furthermore, firms with experience in prior related technology achieve higher integrative capabilities in the presence of outsourcing, a finding consistent with absorptive capacity arguments (Cohen and Levinthal, 1990). Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Copyright  2009 John Wiley & Sons, Ltd.

Note: Positive coefficients indicate greater integrative capabilities or greater performance in the market N = 94 observations; unstandardized coefficients; standard errors in parentheses; ∗ p < 0.10; ∗∗ p < 0.05; ∗∗∗ p < 0.01; ∗∗∗∗ p < 0.001 (one-tailed)

Degree of outsourcing

Model 3 8.757∗∗∗ (2.785) 0.138 (0.138) −0.013 (0.010) −0.027 (0.361) −0.073 (0.190) 0.393∗ (0.242) −0.187∗∗∗ (0.094) −0.600∗∗ (0.340) −0.118∗ (0.084) 0.165 (0.219) −5.731∗∗∗∗ (1.905)

Model 2 −2.967 (11.25) 1.012∗ (0.700) −0.043 (0.055) −0.165 (1.949) −0.031 (1.025) −1.681 (1.355) −0.213 (0.523) 0.819 (1.838) 1.070∗∗∗ (0.468) 1.299 (1.248) 5.332∗∗∗ (2.502) 0.208∗ (0.133) −0.013 (0.010) 0.012 (0.344) −0.041 (0.179) 0.260 (0.232) −0.229∗∗∗ (0.089) −0.649∗∗∗ (0.321) −0.081 (0.080) 0.253 (0.210) −4.744∗∗∗ (1.812)

Model 4

TOBIT

16.246 (13.44) 0.821 (0.754) −0.041 (0.060) −0.312 (2.090) 0.073 (1.099) −1.129 (1.445) −0.032 (0.553) 1.318 (1.968) 0.975∗∗ (0.499) 0.669 (1.312)

Model 5

−2.045 (11.76) 0.908 (0.731) −0.046 (0.058) −0.076 (2.024) 0.124 (1.063) −1.697 (1.415) −0.257 (0.542) 0.933 (1.909) 1.130∗∗∗ (0.486) 1.403 (1.299)

Model 6

TOBIT

2.481 1.936 4.103 2.877 (7.492) (7.418) (8.324) (8.062) −0.197∗∗∗∗ −0.312∗∗∗∗ −0.024∗∗∗∗ −0.057∗∗∗∗ −0.206∗∗∗∗ −0.336∗∗∗∗ (0.043) (0.062) (0.008) (0.012) (0.045) (0.067) 0.197∗∗∗ 0.054∗∗∗∗ 0.222∗∗∗∗ (0.079) (0.015) (0.083) 46.85∗∗∗∗ 56.21∗∗∗∗ 47.19∗∗∗∗ 59.80∗∗∗∗ 37.52∗∗∗∗ 44.45∗∗∗∗ 0.33 0.37 −177.37 −171.06 −327.22 −323.76

14.967 (12.83) 0.923 (0.722) −0.038 (0.057) −0.379 (2.010) −0.014 (1.058) −1.171 (1.383) −0.015 (0.534) 1.132 (1.893) 0.933∗∗ (0.478) 0.647 (1.261)

7.786∗∗∗∗ 5.216∗∗∗ (2.509) (2.257) 0.124 0.156∗ (0.126) (0.119) −0.009 −0.010 (0.009) (0.009) −0.013 0.042 (0.326) (0.308) −0.064 −0.051 (0.172) (0.162) 0.260 0.140 (0.217) (0.208) −0.177∗∗∗ −0.219∗∗∗∗ (0.087) (0.082) −0.518∗∗ −0.460∗ (0.306) (0.289) −0.090 −0.060 (0.077) (0.073) 0.201 0.337∗∗ (0.205) (0.198) −4.489∗∗∗∗ −4.055∗∗∗∗ (1.592) (1.542)

−0.023∗∗∗∗ −0.048∗∗∗∗ (0.007) (0.010) Degree of outsourcing × Experience in prior related technology 0.043∗∗∗∗ (0.013) Chi-square 54.04∗∗∗∗ 72.69∗∗∗∗ R-square 0.36 0.43 Log Likelihood

Customer affluence

Market concentration

Number of employees in bank’s e-business unit

Collaboration with partner

Number of external partners

Scope

Marketing investments

Technical investments

Experience in prior related technology

Model 1

Model 2

Equation 2

Equation 1 Model 1

SUR

SUR

Predicting capabilities Predicting performance Predicting capabilities Predicting performance

Seemingly-unrelated (SUR) and Tobit regression estimates of outsourcing on integrative capabilities and performance

Time of adoption

Firm size

Intercept

Table 2.

Technology Outsourcing, Capabilities, and Performance 609

Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

610

C. Weigelt

(a)

DISCUSSION AND CONCLUSION

6

Capabilities

5 4 PC banking

3

No PC banking

2 1 0

(b)

40

Performance (customer adoption)

x=0 x=100 Degree of outsourcing

35 30 25 20

PC banking

15

No PC banking

10 5 x=0

x=100

Degree of outsourcing

Figure 1. Interaction between degree of outsourcing and experience in prior related technology (PC banking)

In fact, there is no difference in integrative capabilities between firms with and without prior experience at zero outsourcing. Thus, prior experience mitigates the negative effect of outsourcing, supporting Hypothesis 2a. Figure 1b shows that under conditions of vertical integration, firms realize similar levels of performance in the market regardless of their prior experience. However, as firms begin to outsource, the gap in performance realized by firms with and without experience in prior related technology widens. Although both groups of firms show a decline in performance as outsourcing increases, this decline is less for firms with prior experience. Thus, firms with experience in prior related technology achieve higher performance in the market when outsourcing than firms without experience. The finding indicates that prior technological experience can be leveraged in the marketplace and thereby mitigates the negative effect of outsourcing, supporting Hypothesis 2b. In summary, the results show that outsourcing negatively impacts a firm’s integrative capabilities and performance in the market. However, experience in prior related technology reduces the negative effect of outsourcing. Copyright  2009 John Wiley & Sons, Ltd.

This study provides new insights into the impact of new technology outsourcing on a firm’s integrative capabilities and performance in the market. The findings imply that greater outsourcing for business process enhancing technologies (Attewell, 1992; Swanson, 1994) lowers a firm’s integrative capabilities and performance in the market. This negative effect of outsourcing is likely due to two reasons. First, although outsourcing provides firms with access to specialized technologies, the building of integrative capabilities related to these technologies requires learning by doing and investment in internal processes (Ethiraj et al., 2005; Zollo and Winter, 2002). Given that business processes are idiosyncratic, firms are likely to learn in situ about a new technology while using it (Attewell, 1992). Such learning, however, decreases as outsourcing increases. Hence, to build integrative capabilities for a new technology, firms need to be involved in the technology adoption process. Passive capability accumulation is unlikely, and outsourcing cannot simply substitute for internal capabilities (Powell et al., 1996). Thus, although a firm may outsource to obtain a technology, it still needs to understand how the technology relates to its internal processes (Brusoni, Prencipe, and Pavitt, 2001). Second, outsourcing may interfere with firm processes designed to raise customers’ perceived value of a new technology and, hence, customers’ adoption of the technology. Understanding how customers perceive and interact with a new technology depends on a set of interdependent, tacit processes, such as collecting information on customer needs, interpreting it, and disseminating it throughout the firm (Day, 1994). These processes may be interrupted when activities are split into subtasks between a firm and external parties. Moreover, learning about customer preferences is a process of successive approximation that varies across firms due to customers’ variability in preferences (Attewell, 1992; Frei, 2006). As such, this process is difficult to standardize or plan in advance, which may require frequent updating and renegotiation in outsourcing relationships. This study finds that greater vertical integration is superior to outsourcing of business process enhancing technologies. These findings are consistent with RBV and KBV arguments that state that greater vertical integration is preferred Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance for interdependent activities involving tacit knowledge (Kogut and Zander, 1992). Given a common language and shared understanding among employees, vertical integration is superior for the coordination of tacit, context-specific know-how (Leonard-Barton, 1995; Monteverde, 1995; Leiblein et al., 2002). On the other hand, there are conditions under which outsourcing may be superior to vertical integration and enable firms to offer higher quality products and realize cost savings (Rosenzweig, 1994; Dyer, 1996; Mitchell and Singh, 1996). According to the KBV, outsourcing is beneficial for sequential, low interdependent activities where the interface between activities is well known (Kogut and Zander, 1992). Outsourcing may also be suitable for products or services previously performed in-house that can be obtained at lower cost or higher quality outside. In that case, outsourcing affects known services that often have become peripheral to a firm’s core and whose linkages to other system components are well known. In contrast, this is not the case with new business process enhancing technologies whose interactions with firm processes are not fully known until their deployment (Attewell, 1992). Hence, when deciding to outsource, firms should consider the interdependence of activities and the need for tacit knowledge exchange among activities. Overall, the finding of a negative effect of outsourcing emphasizes that although outsourcing gives firms access to a new technology, it does not guarantee that a firm can use and deploy the technology in the market (Hamel, 1991). Therefore, it is important to delineate between offering a new technology to customers via outsourcing and learning how to use it. Outsourcing may endow a firm with new technology applications for its customers, but these should not be equated with better performance in the market. In fact, offering new technology applications through outsourcing can be costly if customers do not adopt the new offerings. This may be the case if outsourcing results in technology applications that are little tailored to a firm’s customers. As outsourcing increases, a firm’s staff may push a new technology less to their customers because they do not feel involved in the technology’s success. Hence, internal staff involvement is advisable when outsourcing for a new technology in order to ensure the building of integrative capabilities related to the technology and the application of firm-specific customer knowledge. Copyright  2009 John Wiley & Sons, Ltd.

611

Moreover, the negative effect of outsourcing is more pronounced for firms without experience in prior related technology. These firms exhibit a drastic decline in integrative capabilities and performance in the market as outsourcing increases. Thus, although prior experience provides absorptive capacity (Cohen and Levinthal, 1990), the findings show limits to benefits from absorptive capacity: prior experience can reduce but neither eliminates nor reverses the downside of outsourcing. However, as outsourcing decreases, the difference in integrative capabilities and performance between firms with and without experience in prior related technology declines. Hence, as outsourcing decreases, prior experience seems not to provide a distinguishing advantage, perhaps because benefits from absorptive capacity are countered by inertia resulting from prior IT investment that functions as a short-run substitute for a new technology. That is, prior experience that enhances a firm’s ability to tailor a new technology to customer needs may not lead to greater customer adoption if there is a prior technology that customers have already adopted and that they perceive as a viable alternative (Rogers, 1995). Prior experience also may not lead to greater new capabilities if there are sunk costs from learning the prior technology (Forman, 2005). However, greater outsourcing widens the gap in integrative capabilities and performance between firms with and without prior experience, implying that firms should not underestimate the role of prior experience in absorbing external know-how. This study has limitations. First, studying a single technology within a single industry avoids the difficulty of controlling for differences across technologies and industries in potential profitability, capital costs, and technology advancement (Levin et al., 1987). However, a single-industry study also has limitations, especially regarding its generalization. This study’s findings are applicable to firms in other industries, such as retail, travel, brokerage, or education, that outsource new technology to enhance business processes. Further, several of the major technology vendors for the banking industry, e.g., EDS or FISERV, also provide technology solutions to the health care, retail, telecommunication, and energy industry. With this in mind, there is reason to believe that this study’s findings can be generalized to other industry contexts. Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

612

C. Weigelt

Second, this study’s findings provide insights into the outsourcing-performance link. Specifically, banks that outsource achieve less customer adoption of a new technology. Thus, while this study shows that outsourcing negatively affects a firm’s performance in the market with customer adoption being a suitable performance measure in the context of customer-facing technologies, it does not provide insights into performance outcomes from a cost minimization standpoint or customer profitability, a performance measure that relates revenues to costs. Third, little besides the degree of outsourcing and collaboration is known about the character of the outsourcing relationship. The impact of outsourcing on integrative capabilities and performance in the market may vary based on the supplier’s skills, the specific tasks performed, and the supplier’s prior contacts with the buyer. Future research could study how partner characteristics affect a firm’s capabilities and performance. In addition, researchers could use a more fine-grained measure of outsourcing, distinguishing between R&D, service/product delivery, consulting services, and support services. Fourth, while integrative capabilities as a measure are based on survey data in this study, future research could try to devise new approaches for collecting archival measures of integrative capabilities. Finally, future research could study how internal efforts, such as coordination across departments and units, affect an adopting firm’s integrative capabilities and performance in the market.

ACKNOWLEDGEMENTS I gratefully acknowledge helpful comments that have substantially improved the work from Margaret Cording, Margarethe Wiersema, SMJ Editor Richard Bettis, and two anonymous reviewers. Financial support for the data collection was provided by a National Science Foundation grant (#332-0043) and a Financial Services Exchange grant.

REFERENCES Acs ZJ, Audretsch DB. 1987. Innovation, market structure, and firm size. Review of Economics and Statistics 69(4): 567–574. Copyright  2009 John Wiley & Sons, Ltd.

Aiken LS, West SG. 1991. Multiple Regression: Testing and Interpreting Interactions. Sage Publications: Newbury Park, CA. Amit R, Schoemaker PJH. 1993. Strategic assets and organizational rent. Strategic Management Journal 14(1): 33–46. Arrow K. 1974. The Limits of Organization. W.W. Norton: New York. Attewell P. 1992. Technology diffusion and organizational learning: the case of business computing. Organization Science 3(1): 1–19. BAI. 1999. Managing Technology Investment Decisions. Banking Administration Institute: Chicago, IL. Barney JB. 1991. Firm resources and sustained competitive advantage. Journal of Management 17: 99–120. Bettis RA, Bradley SP, Hamel G. 1992. Outsourcing and industrial decline. Academy of Management Executive 6: 7–22. Bielski L. 2003. Hard to get the online habit: billpay adoptions gradually increase-especially at a handful of banks-but broad use will take years. ABA Banking Journal 95: 1 February: 79–86. Available at: http://www.allbusiness.com/marketing/marketresearch/458119-1.html. Bitner MJ, Ostrom AL, Meuter ML. 2002. Implementing successful self-service technologies. Academy of Management Executive 16(4): 96–109. Bollen K, Lennox R. 1991. Conventional wisdom on measurement: a structural equation perspective. Psychological Bulletin 110(2): 305–314. Borys B, Jemison DB. 1989. Hybrid arrangements as strategic alliances: theoretical issues in organizational combinations. Academy of Management Review 14: 234–249. Boynton A, Zmud R, Jacobs G. 1994. The influence of IT management practice on IT use in large organizations. MIS Quarterly 18: 299–320. Brown SL, Eisenhardt KM. 1997. The art of continuous change: linking complexity theory and time-paced evolution in relentlessly shifting organizations. Administrative Science Quarterly 42: 1–34. Brusoni S, Prencipe A, Pavitt K. 2001. Knowledge specialization, organization coupling, and the boundaries of the firm: why do firms know more than they make? Administrative Science Quarterly 46(4): 597–625. Chesbrough HW, Teece DJ. 1996. When is virtual virtuous? Organizing for innovation. Harvard Business Review 74(1): 65–73. Cockburn I, Henderson R. 1998. Absorptive capacity, coauthoring behavior, and the organization of research in drug discovery. Journal of Industrial Economics 46: 157–183. Cohen MD, Bacdayan P. 1994. Organizational routines are stored as procedural memory: evidence from a laboratory study. Organization Science 5(4): 554–568. Cohen WM. 1995. Empirical studies of innovative activity. In Handbook of the Economics of Innovation and Technological Change, Stoneman P (ed). Blackwell Publishers: Cambridge, MA; 182–264. Cohen WM, Levinthal DA. 1990. Absorptive capacity: a new perspective on learning and innovation. Administrative Science Quarterly 35: 128–152. Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance Cohen WM, Levinthal DA. 1994. Fortune favors the prepared firm. Management Science 40(2): 227–251. Conner KR, Prahalad CK. 1996. A resource-based theory of the firm: knowledge versus opportunism. Organization Science 7(5): 477–501. Contractor F, Lorange P. 1988. Cooperative Strategies in International Business. Lexington Books: Lexington, MA. Curran JM, Meuter ML. 2005. Self-service technology adoption: comparing three technologies. Journal of Services Marketing 19(2): 103–113. Damanpour F. 1996. Organizational complexity and innovation: developing and testing multiple contingency models. Management Science 42(5): 693–716. Damanpour F, Evan WM. 1984. Organizational innovation and performance: the problem of ‘organizational lag.’ Administrative Science Quarterly 29(3): 392–409. Dandapani K. 2004. Success and failure in Web-based financial services. Communications of the ACM 47(5): 31–33. Davis FD. 1989. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly 13(3): 319–339. Day GS. 1994. The capabilities of market-driven organizations. Journal of Marketing 58: (October): 37–52. Demsetz H. 1988. The theory of the firm revisited. Journal of Law, Economics, and Organization 4(1): 141–162. DosSantos BL, Peffers K. 1995. Rewards to investors in innovative information technology applications: first movers and early followers in ATMs. Organization Science 6(3): 214–259. Durkin M, Howcroft B, O’Donnell A, McCartanQuinn D. 2003. Retail bank customer preferences: personal and remote interactions. International Journal of Retail & Distribution Management 31(4/5): 177–189. Dyer JH. 1996. Specialized supplier networks as a source of competitive advantage: evidence from the auto industry. Strategic Management Journal 17(4): 271–291. Dyer JH, Nobeoka K. 2000. Creating and managing a high-performance knowledge-sharing network: the Toyota case. Strategic Management Journal , March Special Issue 21: 345–367. Eisenhardt KM, Tabrizi BN. 1995. Accelerating adaptive processes: product innovation in the global computer industry. Administrative Science Quarterly 40: 84–110. Ethiraj SK, Kale P, Krishnan MS, Singh JV. 2005. Where do capabilities come from and how do they matter? A study in the software services industry. Strategic Management Journal 26(1): 25–45. Ettlie JE. 1988. Taking Charge of Manufacturing. JosseyBass: San Francisco, CA. FFIEC. 2003. E-banking Booklet. In FFIEC Information Technology Examination Handbook, Federal Financial Institutions Examination Council: Washington, DC. Available at: http://www.ffiec.gov/ffiecinfobase/ Copyright  2009 John Wiley & Sons, Ltd.

613

booklets/e banking/e banking.pdf (accessed 13 July 2008). Fichman RG, Kemerer CF. 1997. The assimilation of software process innovations: an organizational learning perspective. Management Science 43(10): 1345–1363. Forman C. 2005. The corporate digital divide: determinants of Internet adoption. Management Science 51(4): 641–654. Fornell C. 1992. A national customer satisfaction barometer: the Swedish experience. Journal of Marketing 56(1): 6–21. Frei FX. 2006. Breaking the trade-off between efficiency and service. Harvard Business Review 84: (November): 1–10. Gerhart B, Wright PM, McMahan GC. 2000. Measurement error in research on the human resources and firm performance relationship: further evidence and analysis. Personnel Psychology 53(4): 855–872. Gerhart B, Wright PM, McMahan GC, Snell SA. 2000. Measurement error in research on human resources and firm performance: how much error is there and how does it influence effect size estimates? Personnel Psychology 53(4): 803–834. Glick WH, Huber GP, Miller CC, Doty DH, Sutcliffe KM. 1990. Studying changes in organizational design and effectiveness: retrospective event histories and periodic assessments. Organization Science 1(3): 293–312. Golden BR. 1992. The past is the past—or is it? The use of retrospective accounts as indicators of past strategy. Academy of Management Journal 35(4): 848–860. Grant RM. 1996. Prospering in dynamically competitive environments: organizational capability as knowledge integration. Organization Science 7(4): 375–387. Greco J. 1997. Outsourcing: the new partnership. Journal of Business Strategy 18(4): 48–54. Gulati R, Singh H. 1998. The architecture of cooperation: managing coordination costs and appropriation concerns in strategic alliances. Administrative Science Quarterly 43(4): 781–814. Hamel G. 1991. Competition for competence and inter-partner learning within international strategic alliances. Strategic Management Journal , Summer Special Issue 12: 83–103. Hamilton BH, Nickerson JA. 2003. Correcting for endogeneity in strategic management research. Strategic Organization 1(1): 53–80. Helfat CE. 1997. Know-how and asset complementarity and dynamic capability accumulation: the case of R&D. Strategic Management Journal 18(5): 339–360. Helfat CE, Peteraf MA. 2003. The dynamic resourcebased view: capability lifecycles. Strategic Management Journal , October Special Issue 24: 997–1010. Helfat CE, Raubitschek RS. 2000. Product sequencing: co-evolution of knowledge, capabilities and products. Strategic Management Journal , October–November Special Issue 21: 961–979. Huber GP, Power DJ. 1985. Retrospective reports of strategic-level managers: guidelines for increasing accuracy. Strategic Management Journal 6(2): 171–180. Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

614

C. Weigelt

Iansiti M, Clark K. 1994. Integration and dynamic capability: evidence from product development in automobiles and mainframe computers. Industrial and Corporate Change 3: 557–606. Johnson M. 2003. Colliding with customers. Computerworld 15 Dec.: 20. Katila R, Ahuja G. 2002. Something old, something new: a longitudinal study of search behavior and new product introduction. Academy of Management Journal 45(6): 1183–1194. Kogut B, Zander U. 1992. Knowledge of the firm, cominative capabilities, and the replication of technology. Organization Science 3(3): 383–397. Kohli AK, Jaworski BJ. 1990. Market orientation: the construct, research propositions, and managerial implications. Journal of Marketing 54(2): 1–18. Kumar N, Stern LW, Anderson JC. 1993. Conducting interorganizational research using key informants. Academy of Management Journal 36(6): 1633–1651. Lavie D. 2006. The competitive advantage of interconnected firms: an extension of the resource-based view. Academy of Management Review 31(3): 638–658. Leiblein MJ, Miller DJ. 2003. An empirical examination of transaction- and firm-level influences on the vertical boundaries of the firm. Strategic Management Journal 24(9): 839–859. Leiblein MJ, Reuer JJ, Dalsace F. 2002. Do make or buy decisions matter? The influence of organizational governance on technological performance. Strategic Management Journal 23(9): 817–833. Leonard-Barton D. 1988. Implementation as mutual adaptation of technology and organization. Research Policy 17: 251–267. Leonard-Barton D. 1992. Core capabilities and core rigidities: a paradox in managing new product development. Strategic Management Journal , Summer Special Issue 13: 111–125. Leonard-Barton D. 1995. Wellsprings of Knowledge: Building and Sustaining the Sources of Innovation. Harvard Business School Press: Boston, MA. Leonard-Barton D, Deschamps I. 1988. Managerial influence in the implementation of new technology. Management Science 34(10): 1252–1265. Levin SG, Levin SL, Meisel JB. 1987. A dynamic analysis of the adoption of a new technology: the case of optical scanners. Review of Economics and Statistics 69(1): 12–17. Levinthal DA, March JG. 1993. The myopia of learning. Strategic Management Journal , Winter Special Issue 14: 95–112. Lovelock CH, Young RF. 1979. Look to consumers to increase productivity. Harvard Business Review 57: (May-June): 168–178. McEvily SK, Chakravarthy B. 2002. The persistence of knowledge-based advantage: an empirical test for product performance and technological knowledge. Strategic Management Journal 23(4): 285–305. McKendrick J. 2002. Leave computing to us: outsourcing hits its stride in banking. Bank Technology News 1 Feb. Meuter ML, Bitner MJ, Ostrom AL, Brown SW. 2005. Choosing among alternative service delivery modes: Copyright  2009 John Wiley & Sons, Ltd.

an investigation of customer trial of self-service technologies. Journal of Marketing 69: (April): 61–83. Meyer AD, Goes JB. 1988. Organizational assimilation of innovations: a multilevel contextual analysis. Academy of Management Journal 31(4): 897–923. Mitchell W, Singh K. 1996. Survival of business using collaborative relationships to commercialize complex goods. Strategic Management Journal 17(3): 169–195. Monteverde KD. 1995. Technical dialog as an incentive for vertical integration in the semiconductor industry. Management Science 41(10): 1624–1639. Mowery DC, Oxley JE, Silverman BS. 1996. Strategic alliances and interfirm knowledge transfer. Strategic Management Journal , Winter Special Issue 17: 77–91. Mowery DC, Rosenberg N. 1989. Technology and the Pursuit of Economic Growth. Cambridge University Press: Cambridge, UK. Mytelka LK. 1985. Stimulating effective technology transfer: the case of textiles in Africa. In International Technology Transfer: Concepts, Measures, and Comparisons, Rosenberg N, Frischtak C (eds). Praeger: New York; 77–122. Nelson RR, Winter SG. 1982. An Evolutionary Theory of Economic Change. Harvard University Press: Cambridge, MA. Nicholls-Nixon CL, Woo CY. 2003. Technology sourcing and output of established firms in a regime of encompassing technological change. Strategic Management Journal 24(7): 651–666. Nishiguchi T. 1994. Strategic Industrial Sourcing : Tthe Japanese Advantage. Oxford University Press: New York. Novak S, Eppinger SD. 2001. Sourcing by design: product complexity and the supply chain. Management Science 47(1): 189–204. Parasuraman A. 2000. Technology readiness index (TRI): a multiple-item scale to measure readiness to embrace new technologies. Journal of Service Research 2(4): 307–320. Pennings JM, Harianto F. 1992. The diffusion of technological innovation in the commercial banking industry. Strategic Management Journal 13(1): 29–46. Penrose ET. 1959. The Theory of the Growth of the Firm. Oxford University Press: Oxford, UK. Pisano GP. 1990. The R&D boundaries of the firm: an empirical analysis. Administrative Science Quarterly 35(1): 153–176. Pisano GP. 1996. Learning-before-doing in the development of new process technology. Research Policy 25: 1097–1119. Podsakoff PM, Organ DW. 1986. Self-reports in organizational research: problems and prospects. Journal of Management 12(4): 531–544. Polanyi M. 1967. The Tacit Dimension. Doubleday Anchor: New York. Poppo L, Zenger T. 1998. Testing alternative theories of the firm: transaction cost, knowledge-based, and measurement explanations for make-or-buy decisions in Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

Technology Outsourcing, Capabilities, and Performance information services. Strategic Management Journal 19(9): 853–877. Powell WW, Koput KW, Smith-Doerr L. 1996. Interorganizational collaboration and the locus of innovation: networks of learning in biotechnology. Administrative Science Quarterly 41(1): 116–145. Progent Research. 2002. Small business IT outsourcing white paper: outsourcing advantages. How small business can benefit from coutsourcing IT services. Progent Corporation: San Jose, CA. http://www.progent.com/pdf/outsource.pdf (accessed 7 January 2009). Purvis RL, Sambamurthy V, Zmud RW. 2001. The assimilation of knowledge platforms in organizations: an empirical investigation. Organization Science 12(2): 117–135. Ray G, Barney JB, Muhanna WA. 2004. Capabilities, business processes, and competitive advantage: choosing the dependent variable in empirical tests of the resource-based view. Strategic Management Journal 25(1): 23–37. Rogers E. 1995. Diffusion of Innovations. Free Press: New York. Rosenberg N. 1982. Inside the Black Box: Technology and Economics. Oxford University Press: Oxford, UK. Rosenberg N. 1990. Why do firms do basic research (with their own money)? Research Policy 19(2): 165–174. Rosenzweig PM. 1994. International sourcing in athletic footwear: Nike and Reebok. Field case 394189, Harvard Business Publishing, Boston, MA. Schoenecker TS, Cooper AC. 1998. The role of firm resources and organizational attributes in determining entry timing: a cross-industry study. Strategic Management Journal 19(12): 1127–1143. Simon HA. 1960. The New Science of Management Decision. Harper & Row: New York.

Copyright  2009 John Wiley & Sons, Ltd.

615

Steensma HK, Corley KG. 2000. On the performance of technology-sourcing partnerships: the interaction between partner interdependence and technology attributes. Academy of Management Journal 43(6): 1045–1067. Swanson EB. 1994. Information systems innovation among organizations. Management Science 40(9): 1069–1092. Teece DJ. 1986. Profiting from technological innovation: implications for integration, collaboration, licensing and public policy. Research Policy 15: 285–305. Thompson JD. 1967. Organizations in Action. McGrawHill: New York. Utterback JM. 1974. Innovation in industry and the diffusion of technology. Science 15: 658–662. Wheelwright SC, Clark KB. 1992. Revolutionizing Product Development: Quantum Leaps in Speed, Efficiency, and Quality. Free Press: New York. Womack JP, Jones DT, Roos D. 1990. The Machine that Changed the World . Rawson Associates: New York. Wooldridge JM. 2003. Introductory Econometrics: A Modern Approach. South-Western College Publishing: Cincinnati, OH. Yeoh P-L, Roth K. 1999. An empirical analysis of sustained advantage in the U.S. pharmaceutical industry: impact of firm resources and capabilities. Strategic Management Journal 20(7): 637–653. Zellner A. 1962. An efficient method of estimating seemingly unrelated regression equations and tests for aggregation bias. Journal of the American Statistical Association 57: 348–368. Zollo M, Winter SG. 2002. Deliberate learning and the evolution of dynamic capabilities. Organization Science 13(3): 339–351.

Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj

616

C. Weigelt

APPENDIX InExhouse ternal

Key measures from survey Integrative capabilities related to the new technology: seven-point Likert scale (1 = low; 7 = high) (from second survey) Rate the degree to which your bank: 1. Is capable of customizing standardized ‘offthe-shelf’ technology to your bank’s Internet applications 2. Is capable of developing future applications of Internet banking services 3. Has adequate IT skills to operate Internet banking in-house All the measures below are from the first survey: PC banking: Yes, from to (month/year)/ No Has your bank ever, at any time, offered a PC direct-dial program for your retail customers? (PC direct-dial programs require client software installed on your customers’ PC with a direct modem connection either to your bank or to an Internet service provider, such as AOL or AT&T, supporting the service on behalf of your bank.) Outsourcing parties: number of external parties Indicate with how many external parties your bank contracted for each service. Degree of outsourcing: At the time when your bank first adopted each of the following online services, what was the percentage of in-house development versus external arrangements (third party/vendor relationships) that your bank used to implement each service?

Copyright  2009 John Wiley & Sons, Ltd.

Account balance inquiry and funds transfer services Bill payment services Bill presentment services Credit/loan/mortgage services Investment (non-FDIC insured) services Insurance services Nontraditional services (e.g., Web site hosting, account aggregation, virtual mall) CRM (customer relationship management) services in general

# of external parties = 100% = 100% = 100% = 100% = 100% = 100% = 100%

= 100%

Time of adoption: Year: Month: When did your Web site begin to offer transactional capabilities? IT strength: seven-point Likert scale What was your bank’s overall technology/IT knowledge and resources relative to peers at the time when deciding to launch your initial Web presence? 1. Overall technology/IT knowledge 2. IT investments/budget

Strat. Mgmt. J., 30: 595–616 (2009) DOI: 10.1002/smj