Di

Aa

b

c

d

a

AA

KSOI

1

tiastgbocmd

0h

Journal of Operations Management 31 (2013) 330–344

Contents lists available at ScienceDirect

Journal of Operations Management

jo ur nal ho me pa ge: www.elsev ier .com/ locate / jom

rivers and outcomes of open-standard interorganizationalnformation systems assimilation in high-technology supply chains

nníbal C. Soderoa,∗, Elliot Rabinovichb,c, Rajiv K. Sinhad

Department of Supply Chain Management, Sam M. Walton College of Business, University of Arkansas, Fayetteville, AR 72701, USADepartment of Supply Chain Management, W.P. Carey School of Business, Arizona State University, Tempe, AZ 85287, USAKorea University Business School, anam-5Ga, Seongbuk, Seoul 136-701, Republic of KoreaDepartment of Marketing, W.P. Carey School of Business, Arizona State University, Tempe, AZ 85287, USA

r t i c l e i n f o

rticle history:vailable online 27 July 2013

eywords:upply chain management practicespen standards

nterorganizational information systems

a b s t r a c t

In recent years, firms in high-technology supply chains have established internet-based electroniclinkages with their trading partners. As a result, they have improved their ability to coordinate andsynchronize shared business processes by using more complete, accurate, and timely information. Theseelectronic linkages are based on open-standard interorganizational information systems (OSIOS), whichare fundamentally different from traditional electronic data interchanges. OSIOS capture not only thetechnical specifications for data interchange but also the sequential steps for the execution of shared busi-ness processes. Because OSIOS are still at an early diffusion stage, it remains unclear why firms wouldassimilate such an innovation and whether assimilation provides firms any benefits. In this research,we develop a framework grounded on the economics of standards, institutional theory, and strategicinterorganizational information systems literatures to investigate the drivers and outcomes of OSIOSassimilation in a focused context. In order to test our hypotheses based on this framework, we used datafrom a high-technology supply chain and employed econometrics techniques. We found that both com-petition asymmetry across supply chain echelons and OSIOS assimilation within supply chain echelons

predict individual firms’ OSIOS assimilation. The results also suggest that firms’ supply chain dominanceis both a driver and an outcome of OSIOS assimilation, highlighting a mutually reinforcing process. Inaddition, our study reveals boundary conditions of the hypothesized relationships. The use of multi-ple theoretical perspectives, a unique dataset, and innovative statistical techniques to investigate OSIOSassimilation in high-technology supply chains contributes to the body of knowledge in both the supply

anag

chain management and m

. Introduction

Innovation assimilation – the overall process of invention, adop-ion, and deployment of new technology and related processmprovements (Schumpeter, 1934) – is a key source of competitivedvantage (Abernathy and Utterback, 1978). In high-technologyupply chains characterized by networks of firms (Stuart, 2000),he locus of innovation assimilation lies in a wide array of interor-anizational relationships (Oke and Idiagbon-Oke, 2010) that areeyond the purview of individual firms (Choi et al., 2001). In lightf assertions that pressures in institutional environments influenceompetitive strategy (Dacin et al., 2002) and supply chain manage-

ent (SCM) (Liu et al., 2010; Rogers et al., 2007), understanding the

eterminants and implications of innovation assimilation involving

∗ Corresponding author. Tel.: +1 479 575 7119; fax: +1 479 575 5688.E-mail address: asodero@walton.uark.edu (A.C. Sodero).

272-6963/$ – see front matter. Published by Elsevier B.V.ttp://dx.doi.org/10.1016/j.jom.2013.07.008

ement of information systems disciplines.Published by Elsevier B.V.

multiple trading partners in these supply chains is important forboth theory and practice.

In this research, we investigate the emergent phenomenonof assimilation of open-standard interorganizational informa-tion systems (OSIOS) that integrate information technology (IT)with interorganizational business process standards (Bala andVenkatesh, 2007) in high-technology supply chains. Because theassimilation of OSIOS to support the integration of IT and interor-ganizational business standards in supply chains is a fairly recentphenomenon (Paulraj et al., 2012; Zhou and Benton, 2007), researchhas only identified a few initial drivers and benefits of OSIOS assim-ilation. According to prior studies, OSIOS contribute to an orderlysharing of information between firms (Zhou and Benton, 2007) byclearly defining the structure and format of electronic exchangesthrough a common language. Extant research has also underscored

how OSIOS can support the choreography of these exchangesthrough a sequence of steps required to execute cross-boundarybusiness processes between two or more firms (Zhu et al., 2006b).Thus, from an SCM perspective, research has depicted OSIOS as

ations

epi2(sbs((

SMifttFpopOi

boniatatrposiat

aicts1lsci(wOtewrciOto

Oda

A.C. Sodero et al. / Journal of Oper

nablers of modularized interoperability between supply chainartners (Gosain et al., 2003; Nelson et al., 2005). OSIOS are typ-

cally developed by standard-setting consortia of firms (Zhao et al.,007) and are also known as electronic procurement innovationsRai et al., 2009), IT-enabled interorganizational business processtandards (Bala and Venkatesh, 2007), Internet-based electronicusiness (e-business) (Zhu et al., 2006b), Internet-enabled SCMystems (Liu et al., 2010), standard electronic business interfacesMalhotra et al., 2007), and vertical information systems standardsNelson et al., 2005; Wigand and Steinfield, 2005).

Studies investigating the drivers (Boh et al., 2007; Nelson andhaw, 2003; Zhu et al., 2006b) and outcomes (Gosain et al., 2004;alhotra et al., 2005; Saeed et al., 2005) of the assimilation of IT

nnovations in general and OSIOS in particular have predominantlyocused on the diffusion of innovations literature and used eitherhe technology acceptance model (Davis, 1989; Davis et al., 1989) orhe technology–organization–environment model (Tornatzky andleischer, 1990) to describe, rather than to explain or predict, thishenomenon. Although such frameworks describe the assimilationf these innovations in generalizable settings, they do not incor-orate the salient characteristics of both drivers and outcomes ofSIOS assimilation by individual firms competing within and trad-

ng across supply chain echelons.This research aims to address this deficiency in the literature

y shedding light on three main characteristics in the contextf high-technology supply chains: (1) firm supply chain domi-ance, a firm-level variable that captures a firm’s market power

n relation to other firms in the supply chain; (2) competitionsymmetry across supply chain echelons, an industry-level variablehat reflects the variability between competition levels within andcross echelons of the supply chain; and finally (3) OSIOS assimila-ion within a firm’s supply echelon, an industry-level variable thateflects the aggregate level of OSIOS assimilation by competitorsositioned within the same echelon as the focal firm. Our approachf investigating these characteristics at multiple levels of analy-is provides a more nuanced understanding of OSIOS assimilation,ncluding cases in which firms make plans to adopt OSIOS but fail toctually deploy (i.e., accept, adapt, routinize, and institutionalize)hem (Fichman and Kemerer, 1997).

Moreover, building on the work of Bala and Venkatesh (2007)nd Zhu et al. (2006b), we focus on OSIOS assimilation dur-ng an early diffusion stage of OSIOS in high-technology supplyhains. Focusing on an early diffusion stage of OSIOS within high-echnology supply chains is important for two reasons. First, theseupply chains operate in highly dynamic environments (Eisenhardt,989), and as a result, their management requires intense col-

aboration among firms positioned across different echelons tourvive (Nelson et al., 2005; Stuart, 2000). Second, these supplyhains are subject to strong network externalities (Schilling, 2002),n which investments in innovations entail uncertainty and riskKatz and Shapiro, 1985). Therefore, it is unclear why certain firmsould move from planning the adoption to actually deployingSIOS, especially at an early diffusion stage in the assimilation of

hese standards. Although firms pursuing OSIOS assimilation at anarly diffusion stage would incur high levels of risk and cost, theyould also accrue significant benefits from such a move. Indeed,

esearch has argued that OSIOS have the potential to shape supplyhain structures (Dedrick et al., 2008) and fundamentally transformndustries (Varian, 2001; Wigand and Steinfield, 2005). Therefore,SIOS users stand to gain by being at the forefront of this innova-

ion’s assimilation in their supply chain. However, given the noveltyf OSIOS, this issue remains largely untested empirically.

This study also uses a unique dataset that captures actualSIOS assimilation by firms and objective measurements of therivers as well as the outcomes of the assimilation over time. Thispproach is better suited for capturing OSIOS assimilation than

Management 31 (2013) 330–344 331

the predominant empirical approaches in the literature, whichare based on perceptual measures from cross-sectional surveydata and qualitative measures from case study analyses. Althoughthese empirical approaches capture important managerial beliefs,intentions, and expectations behind assimilation, they are limitedin predicting whether and why firms deploy OSIOS. Surveys andcase studies are also limited in capturing inherently dynamic lon-gitudinal relationships regarding actual OSIOS assimilation (Balaand Venkatesh, 2007; Rai et al., 2009), particularly at an earlydiffusion stage, because they gather restricted timeframe data.Moreover, these empirical approaches are based on path modelsthat have neglected the possibility of a two-way direction of causal-ity between the drivers and the outcomes of OSIOS assimilation.

To develop our model of drivers and outcomes of OSIOS assim-ilation, we first provide a description of OSIOS and their keydifferentiating attributes in SCM. We then introduce the economicsof standards behind OSIOS assimilation and review the literatureon institutional theory and strategic interorganizational informa-tion systems. This forms the theoretical foundation to develop ourhypotheses that link competition asymmetry across supply chainechelons, OSIOS assimilation within a firm’s supply chain eche-lon, firm supply chain dominance, and firm OSIOS assimilation. Totest these hypotheses, we first compile a dataset comprising OSIOSassimilation at an early diffusion stage and both industry and firmdata from a high-technology supply chain. We then analyze thedata by estimating simultaneous equation models based on two-stage least squares and the generalized method of moments (GMM)approaches. We conclude with a discussion of our analysis’ find-ings, theoretical and practical contributions, and opportunities forfurther research.

2. Background

2.1. OSIOS and their key differentiating attributes in SCM

OSIOS assimilation has critical implications for supply chainpartnerships. Fuelled by the impact of the Internet’s open standardsin business operations, the growth of OSIOS assimilation has thepotential to transform how trading partners do business (Johnsonand Whang, 2002; Mukhopadhyay and Kekre, 2002). The prolifera-tion of internet-based IT, especially Extensible Markup Language(XML), has laid the foundation for firms to electronically sharericher information with partners in their supply chains (Johnsonand Whang, 2002; Swaminathan and Tayur, 2003). Such informa-tion sharing can promote efficiency (Rai et al., 2009; Saeed et al.,2005) by facilitating interorganizational collaboration and coor-dination (Balakrishnan and Geunes, 2004; Fawcett et al., 2011),joint knowledge creation and assimilation (Malhotra et al., 2005;Salomon and Martin, 2008), and mutual adaptation (Malhotra et al.,2007). Moreover, OSIOS can provide firms strategic flexibility tointerconnect with multiple partners (Gosain et al., 2004). In envi-ronments punctuated by rapid changes in demand, competition,and products (D‘Aveni, 1994), these benefits can provide firms thatassimilate OSIOS greater market power than other supply chainconstituents (Riggins and Mukhopadhyay, 1994; Saeed et al., 2005).In contrast, the lack of common open standards for linkages usingmore traditional interorganizational information systems, such aselectronic data interchange (EDI), has hindered interconnectiv-ity and information sharing among firms (Frohlich, 2002; Gosainet al., 2003). Moreover, the closed nature of standards in these tra-ditional systems has made it economically infeasible for groups

of firms to promote a wider assimilation of these systems acrossindustries. In order to support their efforts in assimilating OSIOS,firms have joined standard-setting consortia (Zhao et al., 2007),such as RosettaNet. OSIOS supported by these consortia are distinct

3 ations

ffita

dTaodcdclsadatttooipesbpntO(panenoth

2

ecf2ncss2ficetbMlO

pOi

32 A.C. Sodero et al. / Journal of Oper

rom traditional interorganizational information systems becauserms involved in the development of the standards are expectedo become their users, in addition to firms that use the standardsfter they are developed (Nelson et al., 2005; Zhao et al., 2007).

Standard-setting consortia inherently follow a consensus-riven, bottom-up approach in the development of standards.his approach is in direct contrast with the unilateral, top–downpproach instituted by dominant firms to develop standards forther interorganizational information systems (e.g., EDI). Theevelopment process of OSIOS standards through this type ofonsortium involves multiple steps, including documentation,evelopment, testing and reviewing, deployment, certification, andompliance. After these steps are completed, the standards are pub-ished so that firms can continue assimilating OSIOS. Because thistandard-setting process is time consuming, resource intensive,nd expensive, it requires member firms to reach a consensus on theifferent shared business processes they need to prioritize for theirssimilation of OSIOS (Nelson et al., 2005). Only then can they planhe adoption and subsequent deployment of OSIOS to complete thisechnology’s assimilation process. OSIOS also differ from other sys-ems that mainly facilitate business-to-business communicationn transactional data (e.g., EDI) because OSIOS cover many aspectsf SCM such as shared cross-boundary business processes, whichnclude collaborative forecasting, vendor-managed inventory, androduct development (Bala and Venkatesh, 2007). The content ofach shared business process standard is complete with messagingervice standards, business dictionaries, technical dictionaries, andusiness process choreography. These XML-based shared businessrocess standards form point-to-point connections, via the Inter-et, that enable execution of relevant business processes betweenrading partners across the supply chain (Gosain et al., 2003). Thus,SIOS can foster interoperability between supply chain partners

Gosain et al., 2003; Nelson et al., 2005). Ultimately, the sharing ofrocesses with the support of OSIOS requires that firms maintain

close and long-term relationship with one another and presentsew opportunities for firms to jointly improve their operations (Liut al., 2010). Furthermore, OSIOS that are assimilated by a sufficientumber of firms enable them to establish business relationshipsn a real-time basis (Gosain et al., 2004; Malhotra et al., 2007)o respond to the needs of highly dynamic and hypercompetitiveigh-technology industries (D‘Aveni, 1994; Eisenhardt, 1989).

.2. Economic factors in the assimilation of OSIOS

According to Zhao et al. (2007) and Zhu et al. (2006a), threeconomic factors shape the assimilation of OSIOS: (1) switchingosts, (2) lock-in effects, and (3) network effects. Together, theseactors can lead to “winner-takes-all” market structures (Schilling,002) characterized by the prevalence of a single widespread tech-ology (David and Greenstein, 1990). Firms are subject to switchingosts at an early diffusion stage because OSIOS assimilation requiresubstantial customization in the form of investments in hardware,oftware, and training (Bala and Venkatesh, 2007; Markus et al.,006; Zhu et al., 2006a). In addition, OSIOS assimilation requiresrms to develop special technical skills to cope with the standards’omplexity (Subramani, 2004). Moreover, because OSIOS are gen-rally used for long-term, dedicated interorganizational linkages,heir assimilation requires a substantial redesign of strategic cross-oundary processes (Mukhopadhyay and Kekre, 2002; Riggins andukhopadhyay, 1994). Together, these conditions for the assimi-

ation of OSIOS translate into switching costs for firms that changeSIOS partners (Zhu et al., 2006a).

Dominant firms in a supply chain are likely to be in a betterosition than their nondominant trading partners to assimilateSIOS at an early diffusion stage, which, in turn, will make switch-

ng costs work in their favor. In general dominant firms assimilate

Management 31 (2013) 330–344

OSIOS early because of their resource availability and expertise,which nondominant firms typically lack. In many cases, nondomi-nant firms cannot afford to become OSIOS users and seek others tohelp them obtain free technology and training (Bala and Venkatesh,2007). Furthermore, dominant firms often commit to assimilat-ing OSIOS in the hopes that other firms will follow suit (Bala andVenkatesh, 2007), thereby increasing the chance of the standardsto eventually become successful. Because of resource availability(Bala and Venkatesh, 2007) and preexisting partnerships (Markuset al., 2006), dominant firms are also likely to join standard-settingconsortia to influence the nature of the technology (Nelson et al.,2005) and dictate more of the specific technology network char-acteristics (Riggins et al., 1994), so that the technology will bettermatch their own existing technologies and cross-boundary pro-cesses (Zhao et al., 2007).

Switching costs also enable dominant firms in the supply chainto impose OSIOS at an early diffusion stage, which in turn locks inother partners (Chwelos et al., 2001; Hart and Saunders, 1997; Teoet al., 2003; Webster, 1995). Extant literature on diffusion of inno-vations also suggests that widespread diffusion of technology (dueto early entry or some favorable event) might lead to the inabil-ity of subsequent superior technologies to ever become successful(Farrell and Klemperer, 2007; Farrell and Saloner, 1985). Therefore,dominant firms have incentives to promote OSIOS assimilation byother partners in the supply chain at an early diffusion stage. Bydeploying OSIOS, dominant firms are able to (1) lock in relation-ships, thereby precluding competitors from doing business withtheir partners (both suppliers and customers), and (2) increase thelikelihood of their technology becoming widespread in the indus-try, furthering the benefits of assimilation.

OSIOS assimilation is also shaped by direct and indirect networkeffects (Zhao et al., 2007; Zhu et al., 2006a). Direct network effectspertain to the positive impact of the number of firms that haveassimilated the technology on the benefits that any one firm canachieve by enabling the sharing of information through OSIOS withmany partners (Rohlfs and Varian, 2003). Indirect network effectsyield greater levels of compatibility in software and hardware solu-tions as the technology diffuses (Church and Gandal, 1992).

3. Theoretical development

Many economists have explored whether innovations diffuseand become widespread in the presence of switching costs, lock-in effects, and network effects (Economides, 1996; Farrell andKlemperer, 2007; Katz and Shapiro, 1985). Several analytical mod-els have employed these effects to investigate IT assimilation (e.g.,Riggins et al., 1994; Weitzel et al., 2006; Zhao et al., 2007). How-ever, such studies fall short in empirically testing these effects,which would improve our understanding of IT assimilation, espe-cially in the OSIOS context. Notable exceptions include Zhu et al.(2006b), who investigate the impact of technological, organiza-tional, and environmental factors on OSIOS assimilation, and Balaand Venkatesh (2007), who examine the effects of institutionalpressures on OSIOS assimilation. We extend both studies by the-orizing relationships that reflect actual OSIOS assimilation byfirms, rather than relationships that reflect managers’ expectations,intentions, and perceptions of assimilation. We also build on thesestudies by integrating economical, institutional, and strategic per-spectives that help expand our understanding of the drivers of andoutcomes of OSIOS assimilation at an early diffusion stage.

Our research focuses on relationships involving two variables

at the industry level (competition asymmetry across supply chainechelons and OSIOS assimilation within the supply chain echelon)and two variables at the firm level (firm supply chain domi-nance and firm OSIOS assimilation). We ground our theoretical

A.C. Sodero et al. / Journal of Operations Management 31 (2013) 330–344 333

Firm level

Industry level

Firm OSIOS

assimilationH1B+

H1A+

H1C+

Competition asymmetry

across supply chain

echelons

Firm supply chain

dominance

OSIOS assimilation within

supply chain echelon

H2+

Institutional

Pressures

Coercive

Normative

Mimetic

ceptu

fseoSie2tbn1ea

3a

esGlItaRimlemp2

scepe

Fig. 1. Con

ramework (in Fig. 1) on two perspectives. To establish a relation-hip between OSIOS assimilation and its drivers, we examine theconomics of open standards through the lens of institutional the-ry. Institutional theory offers rich insights into firms’ adoption ofCM practices (Ketchen and Hult, 2007) and IT for supply chainnteroperability (Bala and Venkatesh, 2007; Cai et al., 2010; Liangt al., 2007; Liu et al., 2010; Teo et al., 2003; Zhang and Dhaliwal,009). In turn, to establish a relationship between OSIOS assimila-ion and its outcomes, we focus on the strategic benefits broughty this innovation to firms through the lens of strategic interorga-izational information systems (Choudhury, 1988; Clemons et al.,993; Johnston and Vitale, 1988). This enables us to investigate theffect of OSIOS assimilation on firms’ ability to gain competitivedvantage.

.1. Institutional theoretical perspective on the drivers of OSIOSssimilation

According to institutional theory, institutional pressure fromnvironmental factors leads firms to engage in specific activities,uch as innovation assimilation (Mansfield, 1983; Robertson andatignon, 1986; Tornatzky and Fleischer, 1990), in order to gain

egitimacy and market positions (Dacin, 1997; Suchman, 1995).n environments characterized by institutional pressure, organiza-ional practices and policies become readily accepted as rationalnd legitimate means to attain organizational goals (Meyer andowan, 1977). Institutional theory posits that structural and behav-

oral changes are driven less by the desire for efficiency andore by the need for organizational legitimacy. It is this drive for

egitimacy that fosters the process of institutionalization, whichventually makes organizations more similar without necessarilyaking them more efficient, giving rise to institutional isomor-

hism (DiMaggio and Powell, 1983; Meyer and Rowan, 1977; Scott,005; Zucker, 1987).

Institutional theory provides a useful point of view for thetudy of OSIOS assimilation at an early diffusion stage because it

an explain how the institutional environment in which a firm ismbedded influences the firm’s engagement in risky and costlyractices at this diffusion stage. According to institutional theory,xternal isomorphic pressures from competitors, trading partners,

al model.

and customers likely induce firms to engage in certain SCM prac-tices, including OSIOS assimilation. DiMaggio and Powell (1983)distinguish three types of isomorphic pressures: coercive, mimetic,and normative. Coercive pressures normally operate through inter-connected relationships. Coercive isomorphism results when firmsacquiesce to “the formal and external pressures exerted upon themby other [firms] upon which they are dependent” (DiMaggio andPowell, 1983, p. 150). Mimetic pressures act through the prevalenceof certain business conducts in a firm’s competitive environment.Mimetic isomorphism ensues as firms respond to uncertainty bymimicking the conduct of other structurally equivalent firms (i.e.,their competitors positioned in the same supply chain echelon).When innovations are poorly understood, goals are ambiguous,or the environment creates uncertainty, firms may model them-selves after firms perceived as legitimate. Therefore, mimicry isoften associated with a bandwagon effect (Rohlfs and Varian, 2003;Sun, 2013). Finally, normative pressures arise from “professionaliza-tion,” as members of an organizational field strive to define theconditions and methods of their work. Normative isomorphismstems primarily from collective expectations of what consti-tutes legitimate behavior, and these expectations are transferredthrough interorganizational channels (e.g., professional associa-tions, conferences, consortia) and gradually become shared norms.Conforming to these shared norms offers assurance to the firm’scompetitors and stakeholders that it will maintain procedurallegitimacy.

3.2. Coercive isomorphism and OSIOS assimilation

We build on the notion of coercive isomorphism to hypothe-size that competition asymmetry across supply chain echelons hasa positive effect on a firm’s OSIOS assimilation. In the context ofOSIOS, we expect that coercive pressures stem mainly from domi-nant customers or suppliers. As previous studies have shown (Balaand Venkatesh, 2007; Chwelos et al., 2001; Hart and Saunders,1998; Liang et al., 2007; Liu et al., 2010; Teo et al., 2003), dominant

firms can require trading partners facing stiff competition to assim-ilate innovations and engage these partners in relationships basedon these technologies. We predict that such coercive pressures forOSIOS assimilation are significant at an early diffusion stage in the

3 ations

pesda

mimaa2sbtwits

cbusneaioeats

Hi

3

shtifihpembctaas(cai

npiot

34 A.C. Sodero et al. / Journal of Oper

resence of high competition asymmetry across supply chain ech-lons. High asymmetry suggests that there are firms in one of theupply chain echelons with enough market power to require non-ominant firms in other echelons (upstream or downstream) tossimilate OSIOS.

As we argued previously, dominant firms have incentives to pro-ote OSIOS assimilation at an early diffusion stage. Although OSIOS

nvolve lower switching costs and fewer lock-in concerns thanore traditional interorganizational information systems, their

ssimilation still carries considerable cost and risk, especially atn early diffusion stage (Bala and Venkatesh, 2007; Markus et al.,006; Zhu et al., 2006a). Thus, pushing for early assimilation byupply chain partners makes strategic sense for dominant firmsecause, by doing so, they can induce these partners to assimilatehe technology and increase the likelihood that OSIOS standardsill succeed. Moreover, by inducing supply chain partners to assim-

late OSIOS at an early diffusion stage, dominant firms can preemptheir competitors from assimilating OSIOS and, in some cases, con-train them to use these standards at higher costs.

Conversely, under low asymmetry, firms will be unable to exertoercive pressure because the distribution of market power wille fairly even among firms in the supply chain. Moreover, becausender low asymmetry no firms will lead in the development oftandards (Axelrod et al., 1995), firms will avoid investing in a tech-ology that is judged as unlikely to be adopted, promoted, andnforced by dominant firms (Chellappa and Saraf, 2010; Churchnd Gandal, 1992). In addition, firms in supply chains character-zed by low asymmetry in market power and uniformly high levelsf competition will have less available resources to engage in newndeavors. In turn, firms in supply chains characterized by lowsymmetry but uniformly low levels of competition will not havehe incentive to push for a new technology that is likely to alter thetatus quo. These arguments lead to our first hypothesis:

1A. Competition asymmetry across supply chain echelons pos-tively influences a firm’s OSIOS assimilation, ceteris paribus.

.3. Mimetic isomorphism and OSIOS assimilation

We build on the notion of mimetic isomorphism to hypothe-ize that OSIOS assimilation within a firm’s supply chain echelonas a nonlinear, monotonically increasing (“hockey-stick”) effect onhe firm’s OSIOS assimilation. In the context of OSIOS assimilation,nstitutional theory suggests that mimetic pressures stem mainlyrom structurally equivalent firms—that is, competitors positionedn the same supply chain echelon. A growing body of literatureas shown that firms exhibit a tendency to imitate their com-etitors’ assimilation of OSIOS (Bala and Venkatesh, 2007; Liangt al., 2007; Liu et al., 2010; Teo et al., 2003). We predict that suchimetic pressures affect firms for two reasons. First, firms may

elieve that competitors that have already assimilated OSIOS willapture market share through greater integration with commonrading partners that have also assimilated this technology (Baland Venkatesh, 2007; Teo et al., 2003). Second, firms may perceive

greater risk of being trapped in a technology outside the industrytandard and face adverse consequences to their competitivenessKatz and Shapiro, 1985) if they fail to assimilate OSIOS when theirompetitors have already done so. Therefore, the extent of OSIOSssimilation in a particular supply chain echelon has a positivenfluence on OSIOS assimilation by other firms.

This positive influence, however, is not linear; rather, it is expo-ential. A firm will be subject to increasingly greater mimetic

ressure to assimilate OSIOS as the assimilation of this technology

ncreases among competitors in its echelon. Specifically, the effectf mimetic pressures on OSIOS assimilation by a firm is subjecto the accumulation of a critical mass of early movers (Lieberman

Management 31 (2013) 330–344

and Montgomery, 1988), which can help the firm reap the bene-fits of assimilating the technology (Sinha and Noble, 2008). Suchnonlinearity emerges because greater OSIOS assimilation amongcompetitors in the same echelon as the firm will also lead to greaterOSIOS assimilation among the firm’s trading partners in adjacentechelons (upstream or downstream) in the supply chain (Bala andVenkatesh, 2007; Liu et al., 2010). As a result, the firm will beincreasingly better positioned to assimilate OSIOS and use themto interconnect with a wider variety of trading partners that sharethe same cross-boundary processes across its distinct functionalareas. In addition, as OSIOS assimilation increases among the firm’scompetitors and more of the firm’s trading partners follow suit,the firm will find increasingly more opportunities to learn fromthese trading partners’ assimilation experiences and ease its ownassimilation of OSIOS. This effect follows from knowledge transfersand spillovers that commonly occur between customers and sup-pliers (Hora and Klassen, 2013) as part of technology assimilation(Schildt et al., 2012; Scott, 2000). Finally, as more of its competi-tors assimilate OSIOS, the firm will be able to find an increasinglygreater availability of hardware and software with the compatiblestandards needed to facilitate OSIOS assimilation (Zhao et al., 2007;Zhu et al., 2006a).

These arguments are consistent with the network effects liter-ature (Economides, 1996; Katz and Shapiro, 1985) and thresholdmodels of innovation diffusion (Geroski, 2000), which suggest thatthe intensity of assimilation of an innovation, such as OSIOS, bya firm increases exponentially with the growth of assimilation ofsuch an innovation by other firms. Therefore, we propose our sec-ond hypothesis, which purports a hockey-stick effect on a firm’sOSIOS assimilation due to mimetic pressure caused by the inten-sity with which the firm’s competitors assimilate this technology.Formally,

H1B. OSIOS assimilation by a firm increases exponentially withOSIOS assimilation within the firm’s supply chain echelon, ceterisparibus.

3.4. Normative isomorphism and OSIOS assimilation

According to institutional theory, dominant firms face strongnormative pressure to assimilate OSIOS and maintain their legiti-macy in the eyes of other firms. Normative pressure plays a role inOSIOS assimilation at an early diffusion stage in two ways. First,many dominant firms are members of standard-setting consor-tia for OSIOS. As a result, managers of these firms may developa shared perception that OSIOS will fit better with their existingshared cross-boundary processes because of their involvement inthe development process. In addition, participation in consortia islikely to instill a sense of duty on managers to develop and pushfor the assimilation of OSIOS within their industrial group (Nelsonet al., 2005). Second, managers of dominant firms may push forOSIOS assimilation, thereby upholding their firms’ resource avail-ability, if they believe that other firms in the industry expect them todo so (Bala and Venkatesh, 2007). Dominant firms will succumb tosuch normative pressures because, in general, they are well placedto pursue OSIOS assimilation at an early diffusion stage and attemptto erect barriers to future market entry by competitors. Moreover,whenever the results of OSIOS assimilation complement previousinvestments in technology already made by firms with marketpower, these firms will be able to gain more than their competitorsfrom OSIOS assimilation (Geroski and Pomroy, 1990).

These arguments underscore the ability of dominant firms

to assimilate OSIOS while incurring lower switching costs thantheir trading partners and competitors (Riordan, 1998; Salop andScheffman, 1983). To the extent that the new technology mayeventually become an industry standard, competitors risk being

ations

pnthOptfid

HO

3p

aimdstta

ittvM(t1a2cSfibei1brrtcl

Hc

4

4

rwchacks

that we expand upon in Section 4.2.5.Our sample comprises 6137 observations from 821 firms.

In keeping with previous work in this domain (Grover, 1993;

1 We strove to establish a broad representation of primary industries in our dataset by including 14 different 4-digit Standard Industrial Classification (SIC) codes toidentify all firms in our research. The SIC codes are 3559, 3570, 3571, 3572, 3576,3600, 3651, 3663, 3670, 3674, 3678, 3825, 5045, and 5065 (respectively, specialindustry machinery; computer and office equipment; electronic computers; com-puter storage devices; computer communications equipment; household audio andvideo equipment components; radio, television broadcast, communication equip-

A.C. Sodero et al. / Journal of Oper

recluded from doing business with dominant firms’ trading part-ers that have already assimilated OSIOS because of incompatibleechnologies, unless they also assimilate the new technology atigher costs. Moreover, dominant firms are likely to assimilateSIOS across multiple functions, whereas nondominant tradingartners are likely to assimilate OSIOS for only business processeshey share with dominant firms. Thus, we expect that dominantrms will be much more involved in OSIOS assimilation than non-ominant firms. This leads to our third hypothesis:

1C. A firm’s supply chain dominance positively influences itsSIOS assimilation, ceteris paribus.

.5. Strategic interorganizational information systemserspective on the outcomes of OSIOS assimilation

Although a firm’s supply chain dominance should drive its OSIOSssimilation, such innovation is, in turn, also likely to lead to anncrease in the firm’s supply chain dominance. Our theoretical argu-

ent for the causality between OSIOS assimilation and supply chainominance builds on the literature on strategic interorganizationalystems. According to this literature, OSIOS assimilation can allowhe firm to gain a more dominant position with respect to its cus-omers and suppliers (Amit and Zott, 2001; Bakos, 1991; Johnstonnd Vitale, 1988).

Three arguments support this effect. First, greater OSIOS assim-lation by a firm facilitates coordination and cooperation betweenhe firm and its suppliers. As a result, the firm will be better ableo adjust and adapt its sourcing activities rapidly to match demandariations for different product characteristics (Gosain et al., 2004;alhotra et al., 2007) and better serve a broader array of customers

Wang and Wei, 2007). Second, greater OSIOS assimilation enableshe firm to exploit its absorptive capability (Cohen and Levinthal,990) to acquire the market information necessary to broaden itsccess to demand downstream in the supply chain (Malhotra et al.,005). In turn, this will put the firm in a better position to negotiateontracts with suppliers competing for its business (Rai et al., 2009;aeed et al., 2005). Third, greater OSIOS assimilation improves therm’s ability to change suppliers in response to variations in itsusiness environment (Gosain et al., 2004). This will limit the firm’sxposure to opportunistic behavior by its suppliers and enhancets contract negotiation position in relation to these firms (Bakos,991; Johnston and Vitale, 1988). It will also enable the firm toroaden the array of customers it can serve without eroding theevenues it receives from them by needing to reduce prices toemain competitive. On the basis of these arguments, we predicthat increases in OSIOS assimilation will promote a firm’s supplyhain dominance as reflected by increases in its market power, ateast in the short run. Thus, our final hypothesis is as follows:

2. A firm’s OSIOS assimilation positively influences its supplyhain dominance, ceteris paribus.

. Methods

.1. Data collection

To test our hypothesized relationships, we conducted ouresearch in the context of the RosettaNet initiative (seeww.rosettanet.org). RosettaNet is a nonprofit standard-setting

onsortium geared to create and implement OSIOS globally in aigh-technology supply chain comprising suppliers of machinery

nd manufacturing components, semiconductor manufacturers,omputer manufacturers, and wholesalers. RosettaNet’s OSIOS arenown as partner interface processes (PIPs). Each PIP refers to apecific process (e.g., request shipping order and notify of shipping

Management 31 (2013) 330–344 335

order confirmation) within a business process area (e.g., logistics)(see Fig. 2).

The selection of this high-technology supply chain has impor-tant implications for our research. First, its environment ischaracterized by hypercompetition (D‘Aveni, 1994), which enablesus to assess sharp shifts in market structure even within a limitedtime span. Second, the supply chain is global in scope (Stuart, 1998,2000) and therefore offers a rich setting to test our hypothesesacross national boundaries and regions. Third, continuous demandfrom downstream segments in the supply chain has driven itsgrowth (Eisenhardt and Schoonhoven, 1996; Lim, 2004; Macher,2006; Salomon and Martin, 2008), which has created the need toadopt SCM practices to interconnect firms across supply chain ech-elons to serve a broad array of customers. Moreover, our focus onRosettaNet in this supply chain is important because RosettaNet isone of the few consortia dedicated to collaborative OSIOS devel-opment. As a result, OSIOS have achieved a level of diffusion (Bohet al., 2007; BusinessWire, 2003) through RosettaNet that makes itpossible to identify more clearly the actual drivers and outcomesof OSIOS assimilation. Finally, examining OSIOS diffusion in a sin-gle supply chain ensures that all firms in our sample are exposedto the same environment. The components of competitive advan-tage in one supply chain may not be appropriate in other supplychains characterized by differences in capital costs and/or marketstructure (Levin et al., 1987; Mansfield, 1969, 1983). Therefore, wecontrol for these sources of variability in our research framework.

To construct the sample for our analyses, we combined fourdatasets. Three datasets correspond to firm-level data compiled bythe RosettaNet consortium, Compustat North America, and Com-pustat Global. The fourth dataset corresponds to industry-level datafrom the Bureau of Labor Statistics (BLS). From RosettaNet, we col-lected data comprising the first nine years of firm-level assimilationof the consortium’s PIPs, from 2001 to 2009, across all memberfirms reporting assimilation. These data include, for each memberfirm i in each year t (t = 2001, . . ., 2009), (1) the types of the PIPs inuse, (2) the number of partners with which firm i had deployedeach PIP, and (3) the total number of partners with which firmi had deployed PIPs. Compustat provided annual financial infor-mation for both member and nonmember firms of the RosettaNetconsortium. Compustat data are limited to publicly traded firms.In addition, the data correspond to each firm’s primary reportedindustry.1 In order to maintain consistency with previous relatedstudies (Dehning et al., 2007; Modi and Mishra, 2011), we use thisinformation to construct measures of competition asymmetry, firmsupply chain dominance, and some of our control variables. The BLSdatabase provided annual information on industry-level IT capi-tal (i.e., investments in computers, software, communication, andother IT-related assets) and also information on industry-level totaltangible wealth (i.e., capital stock).2 We use this information tomeasure IT expenditure intensity, a control variable in our model

ment components; electronic computer accessories; semiconductor and relateddevices; electronic connectors; electronic measuring and testing instruments; com-puter and software wholesale; and electronic parts and equipment wholesale).

2 BLS information is provided at the 3-digit North American Industry ClassificationSystem level. We mapped this information onto the 4-digit SIC level.

336 A.C. Sodero et al. / Journal of Operations Management 31 (2013) 330–344

d business areas. Source: RosettaNet.

Peatbwoofi

aoeHwefipervUHsdtr8td4e

4

i

o

sd

Table 1Constructs, control variables, and data sources.

Primary construct (label) Data sourcefor measures

Competition asymmetry across supply chain echelons(asymmetry)

Compustat

OSIOS assimilation within echelon–exponential(OSIOS Echelon)

RosettaNet

Firm supply chain dominance (relativemarketshare) CompustatFirm OSIOS assimilation (OSIOS) RosettaNetControl variable (label)

Fig. 2. RosettaNet processes an

remkumar and Ramamurthy, 1995; Premkumar et al., 1997; Zhut al., 2006a), we built our sample so that it included both usersnd nonusers. This is especially important to establish a predic-ive model of drivers and outcomes of assimilation. Moreover,ecause we are interested in dynamic longitudinal relationships,e only entered in our analysis data from firms with two or more

bservations in consecutive years. However, we used all firm-levelbservations to compute measures of competition asymmetry andrm market performance.3

Among the 821 firms included in our sample, 81 were usersnd 740 were nonusers. The relatively low number of users inur sample is consistent with the composition of samples of othermpirical studies on diffusion of innovations (Geroski et al., 1993;itt et al., 1997; Lim, 2004). This low number of users comparedith nonusers in our sample is also reflective of our focus on an

arly period of diffusion for RosettaNet’s OSIOS. The 81 users arerms that reported actual assimilation of OSIOS with at least oneartner. According to Bala and Venkatesh’s (2007, p. 342) and Zhut al.’s (2006a, p. 1559) definition, these are OSIOS that have beenoutinized so that they are widely used as an integral part of a firm’salue chain activities through the support of RosettaNet standards.nfortunately, not all firms report assimilation (Nelson et al., 2005).owever, because the 81 users in our sample are a representative

et of the firms in the industry4, and given that we use financialata of all firms in our dataset in our statistical analysis, our estima-ion procedure should produce a lower bound to the actual effectsegarding drivers and outcomes of OSIOS assimilation. Each of the21 firms in our dataset is positioned within a specific echelon ofhe high-technology supply chain supported by RosettaNet stan-ards, according to its 4-digit SIC industry. As Fig. 3 depicts, the 14-digit SIC industries in our dataset are positioned in four differentchelons in our study’s supply chain.

.2. Measures

Table 1 summarizes the data sources we used to operational-ze the four constructs in our hypotheses and the control variables

3 The 821 firms were not required to participate in all nine years encompassingur dataset time span.4 An additional statistical analysis suggests that we cannot reject the hypothe-

is that the distribution of the values for the financial measures of the 81 users isifferent from the distribution of the values for the measures of the 740 nonusers.

Firm research intensity (ResearchIntensity) CompustatFirm’s industry intensity of IT use (ITIntensity) BLS

we used in our empirical model. Table 2 provides the descrip-tive statistics and the correlation coefficients we obtained fromthe measurements of these constructs and control variables. Wedescribe our measurement methodology next.

4.3. Competition asymmetry across supply chain echelons

We base the measurement of our first construct of interest(competition asymmetry across supply chain echelons) on theHerfindahl–Hirschman index of industry concentration (Tirole,1988), defined as follows:

HHIt =nt∑

i=1

marketshare2it , (1)

where HHIt is the Herfindahl–Hirschman index of industry concen-tration for year t, marketshare2

it is the square of the market shareof firm i in year t, and nt is the number of firm observations in yeart (i = 1, . . ., nt, t = 2001, . . ., 2009). The index can range from 0.0 to1.0, moving from a large number of very small firms to a single

monopolistic firm. In general, it is assumed that increases in theHerfindahl–Hirschman index of industry concentration indicate adecrease in the degree of industry competition.5

5 Additional tests (available on request) suggest that the HHIt measure is highlycorrelated with the four and the five firm market concentration ratios, which areother usual measures of industry structure. Parameter estimates in our hypothesestesting are consistent regardless of the measure that we used as the basis for com-petition asymmetry. We report the results using competition asymmetry based on

A.C. Sodero et al. / Journal of Operations Management 31 (2013) 330–344 337

Table 2Descriptive statistics and correlation coefficients.

Mean S.D. (1) (2) (3) (4) (5) (6) (7) (8)

(1) asymmetryit .007 .005 1.00(2) relativemarketshareit .0007 .003 .166*** 1.00(3) OSIOSit .036 .189 .067*** .47*** 1.00(4) OSIOS Echelonit 2.363 0.622 −.11*** .06*** .01 1.00(5) ResearchIntensityit 0.428 5.693 .020 −.02 −.01 −.04** 1.00(6) ITIntensityit 10.339 6.771 .27*** .03* .03* .02 .03** 1.00(7) Asia† .359 .480 −.057*** −.01 −.02 .06*** −.05*** −.34*** 1.00(8) Europe† .097 .296 −.041** −.01 .07*** −.05*** −.02 .02 −.25*** 1.00

Notes: N = 6137 observations.† Dummy variables denoting firm geographical region of operation.* p < 0.05.

** p < 0.01.*** p < 0.001.

Computer manufac turing

3570: comput er and o ffice equipment

3571: electronic co mputers

Semiconductor manu fac turing

3674: semiconductor a nd r elated devices

Machine ry and components for ma nufac turi ng

3559: spe cial indu stry mach inery

3572: comput er storage devices

3576: comput er communi cations e quip ment

3600: electronic and o ther elec trical equ ipmen t and co mponents

3651: household audio and video e quip men t components

3663: radio, televis ion broadcast, communication equ ipmen t

components

3670: electronic co mputer accessor ies

3678: electronic conn ectors

3825: electronic mea sur ing and testing ins tru men ts

Wholesaling

5045: computer and software wholesale

5065: electronic parts and equ ipmen t wholes ale

Fl

tHeotctttw

tr

own echelon and/or when the degree of competition across othersupply chain echelons increases (i.e., when the average concentra-tion, as measured by the Herfindahl–Hirschman index, decreases).6

ig. 3. Constituent 4-digit SIC industries of the high-technology supply chain eche-ons in our dataset.

We obtained annual measures of concentration within each ofhe four echelons in our supply chain by averaging the annualHIt index of all 4-digit SIC industries within their respective ech-lon. Then, for each firm i in our dataset, we computed a measuref competition asymmetry across supply chain echelons in year, asymmetryit, according to a composite index. To calculate thisomposite index, we computed three variances among concentra-ion levels across echelons in year t and then averaged them. The

hree variances, all referring to year t, are (1) the variance betweenhe concentration within the firm’s echelon and the concentrationithin the immediate upstream and/or downstream echelons; (2)

he Herfindahl–Hirschman index of industry concentration because they are moreobust.

the variance among the concentration within the firm’s echelon,the concentration within the immediate upstream and/or down-stream echelons, and the concentration within the second tierupstream/downstream echelons; and (3) the variance among theconcentration levels across all four echelons in the supply chain.Note that our approach gives more weight to the variation in con-centration levels among echelons that are immediately adjacent inthe supply chain. This enables us to capture the potential of a dom-inant firm to exert coercive pressures on its immediate supplierand customer base to assimilate OSIOS. For example, to measurecompetition asymmetry with respect to firms positioned in thewholesaling echelon in Fig. 3, we computed for each year (1) thevariability between the concentration indices for the wholesalingand computer manufacturing echelons; (2) the variability amongthe concentration indices for the wholesaling, computer manufac-turing, and semiconductor manufacturing echelons; and (3) thevariability among the concentration indices for all four echelons.Finally, we averaged the three variances to obtain our measure ofcompetition asymmetry for firms in the wholesaling echelon in therespective year.

4.4. Firm supply chain dominance

We base the measurement of our second construct of interest(firm supply chain dominance) on each firm’s relative market share(relativemarketshareit). In our operationalization, we first computedfirm i’s share of its corresponding echelon’s total revenue in eachyear t. To compute the annual total revenue within each echelon,we added up the sales of all firms in all 4-digit SIC industries inthat echelon in each year. Then, for each firm i in each year t,we divided its sales by the total echelon revenue in that year toobtain its market share within the echelon. Finally, we divided thefirm market share within the echelon by the mean concentrationindex (as measured by the Herfindahl–Hirschman index) across theother three echelons in the supply chain to obtain our measure ofrelativemarketshareit. Thus, based on this measure, a firm’s supplychain dominance increases as its market share increases within its

6 We thank an anonymous reviewer for suggesting that dominance is created byfactors such as scarce resources (e.g., patents), in addition to its relative market share.Indeed, we gathered United States patent count data from the National Bureau ofEconomic Research database (nber.org) and confirmed, through correlation analysis,that our annual measure of firm supply chain dominance positively correlates (atthe 0.001 level) with the count of patent applications in that given year. We did notenter our measure of patent counts in our empirical analysis because the potentialcollinearity with supply chain dominance would bias our estimates.

3 ations

4

mescnnob

ctaodtm

b

iadct

d

fi(aetbwcm

d

f

O

d

4

afibiIin

absence of any strictly exogenous explanatory variables or instru-ments and that extend easily to models with predetermined (e.g.,lagged dependent variables) or endogenous explanatory variables.

38 A.C. Sodero et al. / Journal of Oper

.5. OSIOS assimilation

We adapted Massetti and Zmud’s (1996) “extent of EDI usage”easure to capture each firm i’s OSIOS assimilation intensity in

ach year t (OSIOSit). Our measurement encompasses three dimen-ions: (1) breadth, or the extent to which a firm has establishedonnections with external trading partners; (2) diversity, or theumber of distinct document types a firm handles through con-ections with its trading partners; and (3) depth, or the degreef electronic consolidation that has been established between theusiness processes of two or more trading partners.

To measure the breadth of OSIOS assimilation, breadthit, weounted the number of partners per deployed PIP of firm i in year

within each of the RosettaNet’s business process areas. We thenveraged these measures to obtain an overall mean of the numberf partners per adopted PIP (AveragePartnersperPIPit). Finally, weivided this average by the total number of partners with whichhe firm deployed PIPs in that year (TotalPartnersit) to obtain our

easure of breadthit:

readthit = AveragePartnersperPIPit

TotalPartnersit(2)

To measure the diversity of OSIOS assimilation, diversityit, wedentified the number of business process areas covered by thedopted PIPs by firm i in year t (BusinessProcessAreasit). We thenivided this number by the total number of business process areasovered by RosettaNet’s PIPs (RosettaNetBusinessProcessAreast) inhat year to obtain our final measure of diversityit:

iversityit = BusinessProcessAreasit

RosettaNetBusinessProcessAreast(3)

To measure the depth of OSIOS assimilation, depthit, we identi-ed the types of deployed PIPs within each business process area pp = 1, . . ., q) by firm i in year t (FirmPIPTypesp,it). We also identifiedll types of PIPs made available by RosettaNet to its members withinach business process area p in year t (RosettaNetPIPTypesp,t). Wehen divided the number of types of deployed PIPs by the total num-er of available types of PIPs in each business process area. Finally,e averaged these measures across all q RosettaNet’s business pro-

ess areas in year t (RosettaNetBusinessProcessAreast) to obtain oureasure of depthit:

epthit =q∑

p

FirmPIPTypesp,it

RosettaNetPipTypesp,t

× 1RosettaNetBusinessProcessAreast

(4)

We then added up the three dimensions of OSIOS assimilationor each firm i in each year t to obtain our final measure of OSIOSit:

SIOSit = breadthit + diversityit + depthit . (5)

Each dimension of the measure (i.e., breadthit, diversityit, andepthit) ranges from 0 to 1. Therefore, OSIOSit ranges from 0 to 3.

.6. OSIOS assimilation within a firm’s supply chain echelon

We use the measure OSIOS Echelonit to capture the level of OSIOSssimilation within each firm i’s echelon in year t, excluding therm. In our operationalization, we computed an aggregate scorey adding up the measures of OSIOS assimilation (as discussed

n Section 4.2.3) by all competitors in firm i’s respective echelon.n doing so, we capture not only individual OSIOS assimilationntensity by competitors but also intensity due to the increasingumber of competitors assimilating OSIOS. Then, to capture the

Management 31 (2013) 330–344

hockey-stick effect hypothesized in H1B, we exponentiated theaggregate score to obtain our measure of OSIOS Echelonit.

4.7. Control variables

We incorporate four control variables in our empirical model.These variables characterize a firm’s inherent ability to innovate,the IT expenditure intensity in a firm’s industry, and each firm’sgeographical region of operation (i.e., Asia and Europe). Our firstcontrol variable (ResearchIntensityit) captures a firm i’s ability toinnovate as a function of its research intensity in year t. Giventhe vast differences in annual research-and-development (R&D)expenditures across firms, we measure a firm’s research inten-sity by dividing each firm’s annual R&D expenditures by its annualsales. This approach is consistent with previous studies investigat-ing firms’ levels of innovation (Katila and Ahuja, 2002). Previousresearch also suggests that a firm’s ability to innovate is positivelyrelated to its annual research intensity (Cohen and Klepper, 1996;Dixon, 1980; Lim, 2004; Mansfield, 1981). Thus, we expect thatResearchIntensityit is positively related to OSIOSit. Our second con-trol variable (ITIntensityit) captures the IT expenditure intensity inthe 4-digit SIC industry of a focal firm. We include this variable tocontrol for influences at the industry level on a firm’s decision toassimilate OSIOS caused by both competitive pressures and R&Dspillovers. Following Stiroh (2002) and Brynjolfsson et al. (2008),we use IT capital as a percentage of total tangible wealth as our mea-sure of ITIntensityit. We expect that as this measurement increasesat the industry level, each firm’s OSIOS assimilation increases aswell.7 Our final two control variables (Asiai and Europei) accountfor each firm’s geographical region of operation. The inclusion ofthese variables builds on the notion of innovation clusters (Baptistaand Swann, 1998; Bröcker et al., 2003; Cooke, 2002; Porter, 2000)related to geographical proximity of firms (Katila and Ahuja, 2002),which influence firm-level innovation (Lahiri, 2010) and thus leadto discrepancies in firm-level outcomes across different regionalclusters (Makino et al., 2004). Firms in our focal high-technologysupply chain have traditionally been classified as belonging to threemain geographical regions: North America, Asia, and Europe. More-over, all member firms of the RosettaNet consortium are basedwithin these three regions. Accordingly, we operationalize Asiai andEuropei as binary variables to control for firms based outside NorthAmerica.

5. Results

To test our hypotheses, we estimate a dynamic panel data modelthat consists of two equations: one for firm-level OSIOS assimila-tion and one for firm-level supply chain dominance. The equationsrepresent autoregressive-distributed lag models from panels witha large number of cross-section units (N = 821 firms), each observedfor a relatively small number of periods (T = 9 years). This situ-ation is typical of the use of panel data on firms and calls forestimation methods that do not require the time dimension to belarge to obtain consistent parameter estimates. Assumptions aboutthe properties of initial conditions also play an important role inthis setting because the influence of initial observations on eachsubsequent observation cannot be safely ignored when the timedimension is short (Wooldridge, 2005).

Our approach focuses on methods that can be used in the

7 We thank an anonymous reviewer for suggesting we assess the impact of ITintensity on OSIOS assimilation and for pointing us to the relevant literature.

ations

Strdpcvd1sca

mm

O

fiaOwitt(iaxiOtcO(oOtte

dsfreipfimaasfimi

r

wm

A.C. Sodero et al. / Journal of Oper

trict exogeneity rules out any feedback from present or past shockso current values of the variables, which is often not a naturalestriction in the context of economic models correlating jointlyetermined outcomes such as OSIOS assimilation and firm sup-ly chain dominance. Identification then depends on limited serialorrelation in the error term of the equations, which leads to a con-enient and widely used class of GMM estimators for this type ofynamic panel data (Ahn and Schmidt, 1995; Arellano and Bond,991; Arellano and Bover, 1995; Hansen, 1982). Given our largeample size, this estimation technique provides consistent and effi-ient estimates, even in the presence of unbalanced panel data suchs ours (Newey and McFadden, 1994; Wooldridge, 2002).

The dependent variable for H1A–H1C (OSIOSit) is a continuouseasure of a focal firm i’s OSIOS assimilation intensity in year t. Weodel the equation for H1A–H1C as follows:

SIOSit = ˛10OSIOSi,t−1+˛11asymmetryi,t−1+˛12OSIOS Echeloni,t−1

+ ˛13relativearketsharei,t−1 + x′1itˇ1 + �1i + ε1it, (6)

where OSIOSit denotes the OSIOS assimilation intensity forrm i at time t, OSIOSi,t−1 is the lagged dependent vari-ble, asymmetryi,t−1 is the measure of competition asymmetry,SIOS Echeloni,t−1 is the exponential measure of OSIOS assimilationithin firm i’s supply chain echelon, and relativemarketsharei,t−1,

s the measure of firm i’s supply chain dominance at time − 1. Furthermore, x′

1it is a matrix of other relevant explana-ory variables, �1i controls for unobservable firm-specific effectsJacobson, 1990) representing heterogeneous ability to deploynterconnections (e.g., technological opportunity and appropri-bility conditions), and ε1it is a random error term. The matrix′1it contains six variables: ResearchIntensityit, ITIntensityit, and the

nteractions between relativemarketsharei,t−1 and asymmetryi,t−1,SIOS Echeloni,t−1, Europei, and Asiai. The inclusion of these interac-

ion terms enables us to investigate how differences in firm supplyhain dominance across multiple environments affect a firm’sSIOS assimilation. Consistent with extant research on innovation

e.g., Blundell et al., 1999; Hannan and McDowell, 1984), we lagne year the explanatory variables of interest (i.e., asymmetryi,t−1,SIOS Echeloni,t−1, and relativemarketsharei,t−1) so that we can cap-

ure firms’ actual conduct with respect to OSIOS assimilation afterhose firms’ managers are able to scan, observe, and assess thenvironment.8

To test H2, we use the model summarized in Eq. (7). Theependent variable in this model (relativemarketshareit) corre-ponds to the measure of the supply chain dominance constructor firm i at time t. The independent variables in the model includeelativemarketsharei,t−1, which corresponds to the lagged depend-nt variable, and OSIOSi,t−1, which captures OSIOS assimilationntensity for firm i at time t − 1. Furthermore, as part of the inde-endent variables in the model, �2i controls for unobservablerm-specific effects, ε2it is a random error term, and x′

2it is aatrix of control variables. The matrix x′

2it contains five vari-bles: ResearchIntensityit and the interactions between OSIOSi,t−1nd asymmetryi,t−1, ResearchIntensityit, Europei, and Asiai. The inclu-ion of these interaction terms enables us to identify which types ofrms benefit more from OSIOS assimilation, as well as the environ-ental circumstances under which firms benefit more from such

nitiatives.

elativemarketshareit = ˛20relativemarketsharei,t−1

+ ˛21OSIOSi,t−1 + x′2itˇ2 + �2i + ε2it (7)

8 Given the dynamic nature of the high-technology industry, the use of longer lagsould artificially create the possibility of other changes (e.g., productivity improve-ents, economic cycles) confounding our results.

Management 31 (2013) 330–344 339

We follow Lee (1978) and assess our simultaneous model byestimating the reduced-form equation and calculating the pre-dicted values of the dependent variable in the first stage. That is,OSIOS is expressed as a function of all explanatory variables in Eq.(6). We then use these predicted values in the estimation of Eq.(7) instead of using the actual OSIOS values. We estimate both Eqs.(6) and (7) using a two-step dynamic panel procedure based onArellano and Bond’s (1991) GMM estimator that differences outthe unobservable effects, using instruments for the lagged depend-ent variable. Arellano and Bond’s estimator makes full use of allpossible orthogonality conditions to generate a more efficient esti-mator than, for example, estimators using (t − 2)-dated levels ordifferences of the dependent variable as valid instruments for a(t − 1)-dated lagged dependent variable in a first-differenced paneldata model. Arellano and Bond’s estimator is more efficient becausethe lengthier the panel in time, the more instruments are availablefor use. In this estimation, we limited the maximum number ofperiod lags to be used to six. Thus, for example, instruments dated2001 through 2006 are valid in year 2008, and instruments dated2002 through 2007 are valid in year 2009. Finally, we correct forheteroskedasticity in our equations by using heteroskedasticity-consistent covariance matrix estimators (MacKinnon and White,1985; Newey and West, 1987; White, 1980).

5.1. Drivers of OSIOS assimilation

Model 1 in Table 3 reports the statistical results for H1A–H1C,in which competition asymmetry across supply chain eche-lons (asymmetryi,t−1), OSIOS assimilation within the supply chainechelon (OSIOS Echeloni,t−1), and firm supply chain dominance(relativemarketsharei,t−1) are the independent variables of inter-est and firm-level OSIOS assimilation (OSIOSit) is the dependentvariable. In line with H1A, our results show that asymmetryi,t−1positively affects OSIOSit (˛11 = 14.685, p < 0.05). That is, firm-levelOSIOS assimilation increases as the degree of competition asymme-try across supply chain echelons increases. In addition, our resultsshow that OSIOS Echeloni,t−1 positively affects OSIOSit (˛12 = 0.101,p < 0.05). This provides support for H1B; thus, the relationshipbetween firm-level OSIOS assimilation and OSIOS assimilationwithin its supply chain echelon is nonlinear and monotonicallyincreasing (i.e., exponential) in a hockey-stick fashion. Finally,our results reveal that relativemarketsharei,t−1 positively affectsOSIOSit (˛13 = 1.281, p < 0.01), providing support for H1C. We usethe Hansen–Sargan test (Hansen, 1982; Sargan, 1958) to checkfor overidentifying restrictions in our model. The test statisticis asymptotically chi-squared under the null hypothesis that theinstruments are truly exogenous (i.e., uncorrelated with the errorterm). Our test statistic, ST, is 15.42 (p = 0.566). Therefore, we haveno evidence to reject the hypothesis that the model is overidenti-fied (i.e., the instruments used in our GMM estimation appear tobe exogenous, leading to consistent/unbiased estimates). As such,this confirms that the results in our model provide support forH1A–H1C.

The interaction terms provide noteworthy insights. Nondom-inant firms operating in environments with high competitionasymmetry across supply chain echelons seem to assimilate OSIOSwith more intensity than dominant firms, as suggested by the neg-ative coefficient for the interaction between asymmetryi,t−1 andrelativemarketsharei,t−1 (ˇ11 = −14.956, p < 0.05). Also, Asian firmsassimilate OSIOS with more intensity than North American firmsas firm supply chain dominance increases. The coefficient for theinteraction effect between Asiai and relativemarketsharei,t−1 on

OSIOSit is positive and significantly different from zero (ˇ14 = 0.387,p < 0.05). However, we found no statistically significant differencesbetween North American and European firms regarding levels ofOSIOS assimilation.

340 A.C. Sodero et al. / Journal of Operations

Table 3Generalized method of moments results.

Hypothesis Predictor variables Dependent variable

Model 1OSIOSit

Model 2relativearketshareit

Intercept 0.027**

(0.009)0.007***

(0.001)Industry level

H1A asymmetryi,t−1 14.685*

(7.405)H1B OSIOS Echeloni,t−1 0.101*

(0.005)ITIntensityit 0.009***

(0.002)Firm level

H1C relativemarketsharei,t−1 1.281**

(0.486)0.008***

(0.001)H2 OSIOSi,t−1 0.094*

(0.046)0.003***

(0.001)ResearchIntensityit 2.967***

(0.427)−0.001(0.003)

Interactionsasymmetryi,t−1

*

relativemarketsharei,t−1

−14.956*

(6.770)OSIOS Echeloni,t−1

*

relativemarketsharei,t−1

−0.153(0.146)

Europei* relativemarketsharei,t−1 −0.289

(0.476)Asiai

* relativemarketsharei,t−1 0.387*

(0.183)asymmetryi,t−1

* OSIOSi,t−1 −0.211***

(0.042)ResearchIntensityit

* OSIOSi,t−1 0.002*

(0.001)Europei

* OSIOSi,t−1 −0.003(0.002)

Asiai* OSIOSi,t−1 0.001

(0.001)Model fitHansen–Sargan test forover-identifying restrictions

15.42 20.34

Notes: Standard errors in parentheses. N = 6137 observations.* p < 0.05.

5

eoiTnotztf

acsaOmtOicO

environments subject to strong network externalities (Schilling,

** p < 0.01.*** p < 0.001.

.2. Outcomes of OSIOS assimilation

Model 2 in Table 3 reports the statistical results for H2, whichxamines the influence of firm-level OSIOS assimilation (OSIOSi,t−1)n firm supply chain dominance (relativemarketshareit). Our models overidentified (Hansen–Sargan test statistic ST = 20.34, p = 0.437).herefore, the instruments used in our GMM estimation are exoge-ous, which enables us to assess the impact of OSIOS assimilationn firm supply chain dominance. Since the coefficient for the fit-ed values for OSIOSi,t−1 is positive, significantly different fromero (˛21 = 0.003, p < 0.001), and quite robust to alterations inhe vector of control variables used, our results provide supportor H2.

According to our results from Model 2 in Table 3, OSIOSssimilation effects on firm supply chain dominance decrease asompetition asymmetry across supply chain echelons increases, asuggested by the negative coefficient for the interaction betweensymmetryi,t−1 and OSIOSi,t−1 (ˇ21 = −0.211, p < 0.001). In contrast,SIOS assimilation effects on firm supply chain dominance becomeore positive as research intensity increases, as suggested by

he coefficient for the interaction between ResearchIntensityit andSIOSi,t−1 (ˇ22 = 0.002, p < 0.05). However, no differences emerge

n firm supply chain dominance gains across different geographi-al regions, because our coefficients for the interactions betweenSIOSi,t−1 and Europei and Asiai are nonsignificant.

Management 31 (2013) 330–344

6. Discussion and conclusions

6.1. Theoretical contributions

With our research, we shed light on the relationships amongfirms’ OSIOS assimilation in high-technology supply chains, theinstitutional pressures they face to assimilate OSIOS, and the com-petitive advantages they realize through OSIOS assimilation. Indoing so, we articulate relationships among industry-level con-structs that capture competition asymmetry across supply chainechelons and OSIOS assimilation within supply chain echelons andfirm-level constructs involving supply chain dominance and OSIOSassimilation.

Our study’s methodology gave us a unique opportunity to delvedeeper into the phenomenon under inquiry. First, we compiled adataset using actual OSIOS assimilation data, as reported by userfirms, and actual financial measures, as reported in the Compustatand the Bureau of Labor Statistics databases. This dataset enabledus to assess the relationships among objective measures that canmore accurately reflect the actual factors surrounding OSIOS assim-ilation. Such an approach overcomes limitations in extant OSIOSassimilation research, which is based on perceptual measures thatcan only reflect managerial beliefs, intentions, and expectationsregarding this phenomenon. Second, we estimated a simultaneousequation model in which firm supply chain dominance entered asboth a driver and an outcome of OSIOS assimilation. As a result, wewere able to capture dynamic relationships between supply chaindominance and OSIOS assimilation simultaneously. To the best ofour knowledge, this study pioneers the use of such techniques inthe context of OSIOS assimilation and diffusion.

Our results reveal that competition asymmetry across supplychain echelons is conducive to OSIOS assimilation. Environmentspunctuated by high levels of competition asymmetry createcoercive pressure on firms to assimilate OSIOS. In contrast, envi-ronments characterized by low levels of asymmetry thwart thesecoercive pressures, because firms in such environments do not havethe resources or incentives to foster the diffusion of a potentiallydisruptive innovation. These findings offer a more nuanced under-standing of the effects of competition on innovation, as reflected bythe levels of market concentration across industry sectors (Blundellet al., 1999; Geroski and Pomroy, 1990; Levin et al., 1987). Theunderlying mechanisms of such coercive pressures are the poten-tial effects that emerge when firms assimilate OSIOS at an earlydiffusion stage. Through coercion, dominant firms, whose pres-ence in the supply chain is implied by high competition asymmetryacross echelons, can push for widespread OSIOS assimilation, whichin turn likely reinforces their privileged position in the supply chain.Our findings suggest that it is under high competition asymmetrythat dominant firms are better able to implement activities, suchas global billing, collaborative forecasting, and advanced shippingnotification, with a wider array of nondominant trading partners.

In addition, our results reveal that OSIOS assimilation by a firmexpands as the level of OSIOS assimilation by other firms positionedwithin its respective supply chain echelon increases. This suggeststhat mimetic pressures are at play when it comes to the assimila-tion of OSIOS at the firm level. Our results add to extant literature byalso showing that such mimicry is nonlinear: OSIOS assimilation bya firm accelerates according to a hockey-stick pattern when OSIOSassimilation expands among the firm’s competitors. Though con-sistent with the network effects literature (Economides, 1996; Katzand Shapiro, 1985), our study is among the first to highlight such anonlinear relationship in the context of OSIOS. In high-technology

2002), early users of a disruptive innovation stand to benefit signif-icantly (Schumpeter, 1934). Research has argued that firms tend tofollow competitors’ strategic moves (Young et al., 1996) to protect

ations

teun(

icp1pioeOoiwai

ifiaTittfOsAtcreoaidifimgbmftV

6

atofiiObObfiiat

A.C. Sodero et al. / Journal of Oper

heir competitive position in the industry (Ferrier et al., 1999). Forxample, OSIOS may foster coopetition (Choi et al., 2002) in sit-ations in which a firm may collaborate with its competitors inew product development but subsequently fight for market shareWilhelm, 2011).

Moreover, our results reveal that firm supply chain dominances a predictor of OSIOS assimilation in high-technology supplyhains. This finding is consistent with extant literature that pur-orts that dominance begets innovation (Hannan and McDowell,984; Mansfield, 1969). Our results, however, shed light on theossible role of normative pressures on dominant firms that assim-

late OSIOS (Bala and Venkatesh, 2007; Liu et al., 2010). Throughur results, we show that firms in high-technology supply chainsxpect dominant firms to initiate and promote OSIOS assimilation.ur results also reveal that firm supply chain dominance is notnly a driver but also an outcome of OSIOS assimilation, highlight-ng a mutually reinforcing process. This finding, albeit consistent

ith extant literature (Romer, 1990), is new in the context of OSIOSssimilation and underscores the strategic benefits of early assim-lation of such a risky and uncertain innovation.

The assessment of boundary conditions provides furthernsights. First, we found that OSIOS assimilation by nondominantrms expands more than OSIOS assimilation by dominant firmss competition asymmetry across supply chain echelons increases.his reinforces our primary findings and suggests that nondom-nant firms are subjected to stronger coercive pressures fromheir dominant trading partners to assimilate OSIOS as compe-ition asymmetry increases across the supply chain. Second, weound that dominant firms in Asia are more likely to assimilateSIOS than their North American and European counterparts. Con-

istent with extant diffusion of innovations literature (Acs andudretsch, 1990; Geroski and Pomroy, 1990), our study offers a bet-

er understanding of OSIOS assimilation in high-technology supplyhains by suggesting that OSIOS follow different diffusion trajecto-ies across geographical clusters (Bell, 2005; Lahiri, 2010; Makinot al., 2004). Third, we found that gains in supply chain dominancebtained by firms from OSIOS assimilation decrease as competitionsymmetry across supply chain echelons increases. This findings consistent with the notion that not all firms (specifically, theominant ones) stand to appropriate these gains from OSIOS assim-

lation (Grover and Kohli, 2012). Finally, we found that gains inrm supply chain dominance due to OSIOS assimilation becomeuch stronger as firms increase their research intensity. This sug-

ests that OSIOS users that invest in technological inputs across theoard will gain much more than users with more limited invest-ents (Levin et al., 1985; Lim, 2004; Mansfield, 1981). This finding

urthers our understanding of the role of firm resources in shapinghe actual structural impacts of OSIOS assimilation (Johnston anditale, 1988; Varian et al., 2004) in high-technology supply chains.

.2. Practical contributions

Our results give practitioners clear evidence that dominant firmsre better positioned to assimilate OSIOS and to benefit more fromhis innovation. Despite the clear assimilation advantages of OSIOSver traditional EDI standards, it is still expensive and difficult forrms to transition from planning the adoption to actually deploy-

ng OSIOS at an early diffusion stage of this innovation. This makesSIOS unaffordable for many nondominant firms, which tend toe either small or medium-sized organizations. Therefore, becauseSIOS assimilation requires mutual and synergistic agreementsetween trading partners (Bala and Venkatesh, 2007), dominant

rms likely have the upper hand when it comes to assimilating this

nnovation. The implication is that dominant firms should facilitatend even promote OSIOS assimilation by nondominant partners, sohat the gains of such an innovation could be shared by all members

Management 31 (2013) 330–344 341

engaged in joint supply chain activities. This collaboration shouldbe carefully crafted to ensure that costs do not outweigh the bene-fits of shared assimilation. In addition, as Johnston and Vitale (1988)caution, dominant firms “need to consider carefully the long-termimpact on their industry before they rush into [OSIOS assimila-tion].”

Moreover, when considering OSIOS assimilation, managers indominant firms should be aware that their ability to influence firmsin other supply chain echelons to assimilate this technology willdepend on the intensity of competition these latter firms face. Aslevels of competition increase, firms are more likely to assimilateOSIOS to gain traction with dominant firms positioned across thesupply chain. The implication is that OSIOS assimilation not onlyspans multiple firms across the supply chain but also depends onthe intensity of competition within these firms’ echelons.

6.3. Opportunities for further research

We focused our inquiry on a high-technology supply chain. Assuch, our research is constrained to a setting punctuated by rapidand discontinuous changes in demand, competition, and products(D‘Aveni, 1994; Eisenhardt, 1989). Moreover, our results are con-tingent on the norms and procedures of the RosettaNet consortiumand the characteristics of the RosettaNet OSIOS implementationsin this supply chain. Therefore, a natural extension of this researchcould focus on OSIOS assimilation in different industries involvingother standard-setting consortia. Future studies could also extendthe inquiry to industries encompassing emerging markets in othergeographical regions (e.g., Latin America) to make the model moregeneralizable. In addition, further research could investigate therelationship between industry- and firm-level factors related toOSIOS assimilation for diversified firms competing in different sup-ply chains across multiple products. Competition for these firmsmakes the hypothesized relationships more complex. The dynamicmodel proposed herein could be extended to consider the effectsof multimarket competition and OSIOS assimilation.

In addition, further research might provide a richer and morenuanced understanding of the impact of OSIOS assimilation onfirm supply chain dominance by investigating not only finan-cially derived market outcomes but also operational outcomes(Venkatraman and Ramanujam, 1986) in the proposed framework.The model could be extended to examine more comprehensively,through qualitative or cross-sectional studies, the outcomes ofOSIOS assimilation in terms of cost, quality, flexibility, delivery, andinnovation.

Finally, an important consideration for further research is theinclusion of technological characteristics that facilitate or inhibitinnovation (Saeed et al., 2011). As part of OSIOS assimilation, firmsmust restructure their processes to better capture the informationthey exchange with other firms within their internal organiza-tional information systems (e.g., ERP systems). This underscoresthe importance of assessing organizational capabilities and orga-nizational readiness for technology assimilation (Malhotra et al.,2005; Rai et al., 2009) as moderators of the relationship betweenOSIOS assimilation and firm supply chain dominance.

7. Conclusion

OSIOS are an emerging innovation that depend on and havethe potential to shape supply chain structures and fundamen-tally transform entire industries. To investigate this assertion, we

developed a theoretical framework and empirically evaluated itin a high-technology supply chain context. Our empirical resultsfrom this evaluation highlighted the role played by OSIOS assim-ilation and competition across the supply chain in shaping OSIOS

3 ations

aetmiiu

A

Ra

R

A

A

A

A

A

A

A

B

B

B

B

B

B

B

B

B

B

C

C

C

C

C

C

C

C

42 A.C. Sodero et al. / Journal of Oper

ssimilation by individual firms. The results also showed the exist-nce of a mutually reinforcing process in which dominant firms inhe supply chain further their dominance by assimilating OSIOS

ore intensely than their nondominant trading partners. Ournvestigation provided guidance for firms engaged in assimilat-ng OSIOS, and opened up important avenues for furthering thenderstanding of OSIOS assimilation by firms.

cknowledgment

We thank Debra Praznik and Hussam El-Leithy from theosettaNet consortium for providing data on RosettaNet OSIOSssimilation.

eferences

bernathy, W.J., Utterback, J.M., 1978. Patterns of industrial innovation. TechnologyReview 80 (7), 40–47.

cs, Z.J., Audretsch, D.B., 1990. Innovation and Small Firms. The MIT Press,Cambridge, MA.

hn, S.C., Schmidt, P., 1995. Efficient estimation of models for dynamic panel data.Journal of Econometrics 68 (1), 5–27.

mit, R., Zott, C., 2001. Value creation in e-business. Strategic Management Journal22 (6), 493–520.

rellano, M., Bond, S., 1991. Some tests of specification for panel data: Monte Carloevidence and an application to employment equations. The Review of EconomicStudies 58 (2), 277–297.

rellano, M., Bover, O., 1995. Another look at the instrumental variable estimationof error-components models. Journal of Econometrics 68 (1), 29–51.

xelrod, R., Mitchell, W., Thomas, R.E., Bennett, D.S., Bruderer, E., 1995. Coalition for-mation in standard-setting alliances. Management Science 41 (9), 1493–1508.

akos, J., 1991. Information links and electronic marketplaces: the role of interor-ganizational information systems in vertical markets. Journal of ManagementInformation Systems 8 (2), 31–52.

ala, H., Venkatesh, V., 2007. Assimilation of interorganizational business processstandards. Information Systems Research 18 (3), 340–362.

alakrishnan, A., Geunes, J., 2004. Collaboration and coordination in supply chainmanagement and e-commerce. Production and Operations Management 13 (1),1–2.

aptista, R., Swann, P., 1998. Do firms in clusters innovate more? Research Policy 27(5), 525–540.

ell, G.G., 2005. Clusters, networks, and firm innovativeness. Strategic ManagementJournal 26 (3), 287–295.

lundell, R., Griffiths, R., Van Reenen, J., 1999. Market share, market value and inno-vation in a panel of British manufacturing firms. Review of Economic studies 66(3), 529–554.

oh, W.F., Soh, C., Yeo, S., 2007. Standards development and diffusion: a case studyof RosettaNet. Communications of the ACM 50 (12), 57–62.

röcker, J., Dohse, D., Soltwedel, R., 2003. Clusters and competition as engines ofinnovation: an introduction. In: Innovation Clusters and Interregional Compe-tition. Springer Verlag, Berlin, pp. 1–5.

rynjolfsson, E., McAfee, A., Sorell, M.R., Zhu, F., 2008. Scale Without Mass: BusinessProcess Replication and Industry Dynamics, Harvard Business School WorkingPaper 07-016.

usinessWire, 2003. RosettaNet Global E-Business Standard Reaches Critical Massin High Technology Sector, Business Wire.

ai, S., Jun, M., Yang, Z., 2010. Implementing supply chain information integra-tion in China: the role of institutional forces and trust. Journal of OperationsManagement 28 (3), 257–268.

hellappa, R.K., Saraf, N., 2010. Alliances, rivalry, and firm performance in enterprisesystems software markets: a social network approach. Information SystemsResearch 21 (4), 849–871.

hoi, T.Y., Dooley, K.J., Rungtusanatham, M., 2001. Supply networks and complexadaptive systems: control versus emergence. Journal of Operations Management19 (3), 351–366.

hoi, T.Y., Wu, Z., Ellram, L., Koka, B.R., 2002. Supplier-supplier relationships andtheir implications for buyer-supplier relationships. IEEE Transactions on Engi-neering Management 49 (2), 119–130.

houdhury, V., 1988. Sustaining competitive advantage with inter-organizationalinformation systems. In: Twenty-First Annual Hawaii International Conferenceon Decision Support and Knowledge Based Systems. IEEE Computer SocietyPress Los Alamitos, CA, USA, Honolulu, pp. 44-51.

hurch, J., Gandal, N., 1992. Network effects, software provision, and standardiza-tion. The Journal of Industrial Economics 40 (1), 85–103.

hwelos, P., Benbasat, I., Dexter, A.S., 2001. Research report: empirical test of an EDIadoption model. Information Systems Research 12 (3), 304–321.

lemons, E.K., Reddi, S.P., Row, M.C., 1993. The impact of information technologyon the organization of economic activity: the ‘Move to the middle’ hypothesis.Journal of Management Information Systems 10 (2), 9–35.

Management 31 (2013) 330–344

Cohen, W.M., Klepper, S., 1996. Firm size and the nature of innovation withinindustries: the case of process and product R&D. The Review of Economics andStatistics 78 (2), 232–243.

Cohen, W.M., Levinthal, D.A., 1990. Absorptive capacity: a new perspective on learn-ing and innovation. Administrative Science Quarterly 35 (1), 128–152.

Cooke, P., 2002. Regional innovation systems: general findings and some new evi-dence from biotechnology clusters. The Journal of Technology Transfer 27 (1),133–145.

D‘Aveni, R.A., 1994. Hypercompetition: Managing the Dynamics of Strategic Man-agement. Free Press, New York.

Dacin, M.T., 1997. Isomorphism in context: the power and prescription of institu-tional norms. Academy of Management Journal 40 (1), 46–81.

Dacin, M.T., Goodstein, J., Scott, W.R., 2002. Institutional theory and institutionalchange: introduction to the special research forum. Academy of ManagementJournal 45 (1), 43–56.

David, P.A., Greenstein, S., 1990. The economics of compatibility standards: an intro-duction to recent research. Economics of Innovation and New Technology 1 (1),3–41.

Davis, F., 1989. Perceived usefulness, perceived ease of use, and user acceptance ofinformation technology. MIS Quarterly 13 (3), 319–340.

Davis, F., Bagozzi, R., Warshaw, P., 1989. User acceptance of computer technol-ogy: a comparison of two theoretical models. Management Science 35 (8),982–1003.

Dedrick, J., Xu, S.X., Zhu, K.X., 2008. How does information technology shape supply-chain structure? Evidence on the number of suppliers. Journal of ManagementInformation Systems 25 (2), 41–72.

Dehning, B., Richardson, V.J., Zmud, R.W., 2007. The financial performance effects ofIT-based supply chain management systems in manufacturing firms. Journal ofOperations Management 25 (4), 806–824.

DiMaggio, P.J., Powell, W.W., 1983. The iron cage revisited: institutional isomor-phism and collective rationality in organizational fields. American SociologicalReview 48, 147–160.

Dixon, R., 1980. Hybrid corn revisited. Econometrica 48 (6), 1451–1461.Economides, N., 1996. The economics of networks. International Journal of Industrial

Organization 14 (6), 673–699.Eisenhardt, K., 1989. Making fast strategic decisions in high-velocity environments.

Academy of Management Journal 32 (3), 543–576.Eisenhardt, K.M., Schoonhoven, C.B., 1996. Resource-based view of strategic alliance

formation: strategic and social effects in entrepreneurial firms. OrganizationScience 7 (2), 136–150.

Farrell, J., Klemperer, P., 2007. Coordination and lock-in: competition with switchingcosts and network effects. Handbook of Industrial Organization 3, 1967–2072.

Farrell, J., Saloner, G., 1985. Standardization, compatibility, and innovation. TheRAND Journal of Economics 16 (1), 70–83.

Fawcett, S.E., Wallin, C., Allred, C., Fawcett, A.M., Magnan, G.M., 2011. Informationtechnology as an enabler of supply chain collaboration: a dynamic capabilitiesperspective. Journal of Supply Chain Management 47 (1), 38–59.

Ferrier, W.J., Smith, K.G., Grimm, C.M., 1999. The role of competitive action in mar-ket share erosion and industry dethronement: a study of industry leaders andchallengers. Academy of Management Journal 42 (4), 372–388.

Fichman, R.G., Kemerer, C.F., 1997. The assimilation of software process innova-tions: an organizational learning perspective. Management Science 43 (10),1345–1363.

Frohlich, M.T., 2002. e-Integration in the supply chain: barriers and performance.Decision Sciences 33 (4), 537–556.

Geroski, P., 2000. Models of technology diffusion. Research Policy 29 (4-5),603–625.

Geroski, P., Machin, S., Van Reenen, J., 1993. The profitability of innovating firms.The RAND Journal of Economics 24 (2), 198–211.

Geroski, P.A., Pomroy, R., 1990. Innovation and the evolution of market structure.The Journal of Industrial Economics 38 (3), 299–314.

Gosain, S., Malhotra, A., El Sawy, O.A., 2004. Coordinating for flexibility in e-businesssupply chains. Journal of Management Information Systems 21 (3), 7–45.

Gosain, S., Malhotra, A., Sawy, O.A.E., Chehade, F., 2003. The impact of commone-business interfaces. Communications of the ACM 46 (12), 186–195.

Grover, V., 1993. An empirically derived model for the adoption of customer-basedinterorganizational systems. Decision Sciences 24 (3), 603–640.

Grover, V., Kohli, R., 2012. Cocreating IT value: new capabilities and metrics formultifirm environments. MIS Quarterly 36 (1), 225–232.

Hannan, T.H., McDowell, J.M., 1984. The determinants of technology adoption: thecase of the banking firm. The RAND Journal of Economics 15 (3), 328–335.

Hansen, L.P., 1982. Large sample properties of generalized method of momentsestimators. Econometrica 50 (4), 1029–1054.

Hart, P., Saunders, C., 1997. Power and trust: critical factors in the adoption and useof electronic data interchange. Organization Science 8 (1), 23–42.

Hart, P.J., Saunders, C.S., 1998. Emerging electronic partnerships: antecedents anddimensions of EDI use from the supplier’s perspective. Journal of ManagementInformation Systems 14 (4), 87–111.

Hitt, M.A., Hoskisson, R.E., Kim, H., 1997. International diversification: effects oninnovation and firm performance in product-diversified firms. Academy of Man-agement Journal 40 (4), 767–798.

Hora, M., Klassen, R.D., 2013. Learning from others’ misfortune: factors influenc-ing knowledge acquisition to reduce operational risk. Journal of OperationsManagement 31 (1-2), 52–61.

Jacobson, R., 1990. Unobservable effects and business performance. Marketing Sci-ence 9 (1), 74–85.

ations

J

J

K

K

K

L

L

L

L

L

L

L

L

M

M

M

M

M

MM

M

M

M

M

M

M

N

N

N

N

O

P

P

A.C. Sodero et al. / Journal of Oper

ohnson, M., Whang, S., 2002. e-Business and supply chain management: anoverview and framework. Production and Operations Management 11 (4),413–423.

ohnston, H.R., Vitale, M.R., 1988. Creating competitive advantage with interorgani-zational information systems. MIS Quarterly 12 (1), 10.

atila, R., Ahuja, G., 2002. Something old, something new: a longitudinal studyof search behavior and new product introduction. Academy of ManagementJournal 45 (6), 1183–1194.

atz, M.L., Shapiro, C., 1985. Network externalities, competition, and compatibility.The American Economic Review 75 (3), 424–440.

etchen, D.J., Hult, G.T.M., 2007. Toward greater integration of insights fromorganization theory and supply chain management. Journal of Operations Man-agement 25 (2), 455–458.

ahiri, N., 2010. Geographic distribution of R&D activity: how does it affect innova-tion quality? Academy of Management Journal 53 (5), 1194–1209.

ee, L.F., 1978. Unionism and wage rates: a simultaneous equations models withqualitative and limited dependent variables. International Economic Review 19(1), 415–433.

evin, R.C., Cohen, W.M., Mowery, D.C., 1985. R & D appropriability, opportunity,and market structure: new evidence on some Schumpeterian hypotheses. TheAmerican Economic Review 75 (2), 20–24.

evin, S.G., Levin, S.L., Meisel, J.B., 1987. A dynamic analysis of the adoption of a newtechnology: the case of optical scanners. The Review of Economics and Statistics69 (1), 12–17.

iang, H., Saraf, N., Hu, Q., Xue, Y., 2007. Assimilation of enterprise systems: theeffect of institutional pressures and the mediating role of top management. MISQuarterly 31 (1), 59–87.

ieberman, M.B., Montgomery, D.B., 1988. First-mover advantages. Strategic Man-agement Journal 9 (S1), 41–58.

im, K., 2004. The relationship between research and innovation in the semiconduc-tor and pharmaceutical industries (1981-1997). Research Policy 33 (2), 287–321.

iu, H., Ke, W., Wei, K.K., Gu, J., Chen, H., 2010. The role of institutional pressuresand organizational culture in the firm’s intention to adopt internet-enabledsupply chain management systems. Journal of Operations Management 28 (5),372–384.

acher, J., 2006. Technological development and the boundaries of the firm: aknowledge-based examination in semiconductor manufacturing. ManagementScience 52 (6), 826–843.

acKinnon, J.G., White, H., 1985. Some heteroskedasticity-consistent covariancematrix estimators with improved finite sample properties. Journal of Econo-metrics 29 (3), 305–325.

akino, S., Isobe, T., Chan, C.M., 2004. Does country matter? Strategic ManagementJournal 25 (10), 1027–1043.

alhotra, A., Gosain, S., El Sawy, O.A., 2005. Absorptive capacity configurations insupply chains: gearing for partner-enabled market knowledge creation. MISQuarterly 29 (1), 145–187.

alhotra, A., Gosain, S., El Sawy, O.A., 2007. Leveraging standard electronic businessinterfaces to enable adaptive supply chain partnerships. Information SystemsResearch 18 (3), 260–279.

ansfield, E., 1969. The Economics of Technological Change. Longmans, London.ansfield, E., 1981. Composition of R&D expenditures: relationship to size of firm,

concentration, and innovative output. The Review of Economics and Statistics63 (4), 610–615.

ansfield, E., 1983. Technological change and market structure: an empirical study.The American Economic Review 73 (2), 205–209.

arkus, L., Steinfield, C.W., Wigand, R.T., Minton, G., 2006. Industry-wide informa-tion systems standardization as collective action: the case of the US residentialmortgage industry. MIS Quarterly 30, 439–465.

assetti, B., Zmud, R.W., 1996. Measuring the extent of EDI usage in complex orga-nizations: strategies and illustrative examples. MIS Quarterly 20 (3), 331.

eyer, J.W., Rowan, B., 1977. Institutionalized organizations: formal structure asmyth and ceremony. American Journal of Sociology 83 (2), 340–363.

odi, S.B., Mishra, S., 2011. What drives financial performance-resource efficiencyor resource slack?: evidence from U.S. based manufacturing firms from 1991 to2006. Journal of Operations Management 29 (3), 254–273.

ukhopadhyay, T., Kekre, S., 2002. Strategic and operational benefits of elec-tronic integration in B2B procurement processes. Management Science 48 (10),1301–1313.

elson, M.L., Shaw, M.J., 2003. The adoption and diffusion of interorganizationalsystems standards and process innovations. In: Workshop on Standard Making:A Critical Research Frontier for Information Systems, Seattle, WA.

elson, M.L., Shaw, M.J., Qualls, W., 2005. Interorganizational system standardsdevelopment in vertical industries. Electronic Markets 15 (4), 378–392.

ewey, W.K., McFadden, D., 1994. Large sample estimation and hypothesis testing.In: Engle, R.F., McFadden, D.L. (Eds.), Handbook of Econometrics. North Holland,New York, pp. 2111–2245.

ewey, W.K., West, K.D., 1987. A simple, positive semi-definite, heteroskedas-ticity and autocorrelation consistent covariance matrix. Econometrica 55 (3),703–708.

ke, A., Idiagbon-Oke, M., 2010. Communication channels, innovation tasks and NPDproject outcomes in innovation-driven horizontal networks. Journal of Opera-

tions Management 28 (5), 442–453.

aulraj, A., Chen, I.J., Lado, A.A., 2012. An empirical taxonomy of supply chain man-agement practices. Journal of Business Logistics 33 (3), 227–244.

orter, M.E., 2000. Location, competition, and economic development: local clustersin a global economy. Economic Development Quarterly 14 (1), 15–31.

Management 31 (2013) 330–344 343

Premkumar, G., Ramamurthy, K., 1995. The role of interorganizational and organiza-tional factors on the decision mode for adoption of interorganizational systems.Decision Sciences 26 (3), 303–336.

Premkumar, G., Ramamurthy, K., Crum, M., 1997. Determinants of EDI adoptionin the transportation industry. European Journal of Information Systems 6 (2),107–121.

Rai, A., Brown, P., Tang, X., 2009. Organizational assimilation of electronic procure-ment innovations. Journal of Management Information Systems 26 (1), 257–296.

Riggins, F., Mukhopadhyay, T., 1994. Interdependent benefits from interorganiza-tional systems: opportunities for business partner reengineering. Journal ofManagement Information Systems 11 (2), 37–57.

Riggins, F.J., Kriebel, C.H., Mukhopadhyay, T., 1994. The growth of interorganizationalsystems in the presence of network externalities. Management Science 40 (8),984–998.

Riordan, M.H., 1998. Anticompetitive vertical integration by a dominant firm. Amer-ican Economic Review, 1232–1248.

Robertson, T.S., Gatignon, H., 1986. Competitive effects on technology diffusion.Journal of Marketing 50 (3), 1–12.

Rogers, K.W., Purdy, L., Safayeni, F., Duimering, P.R., 2007. A supplier developmentprogram: rational process or institutional image construction? Journal of Oper-ations Management 25 (2), 556–572.

Rohlfs, J.H., Varian, H.R., 2003. Bandwagon Effects in High-technology Industries.The MIT Press, Cambridge, MA.

Romer, P.M., 1990. The origins of endogenous growth. Journal of Economic Perspec-tives 8 (1), 3–22.

Saeed, K.A., Malhotra, M.K., Grover, V., 2005. Examining the impact of interorgani-zational systems on process efficiency and sourcing leverage in buyer–supplierdyads. Decision Sciences 36 (3), 365–396.

Saeed, K.A., Malhotra, M.K., Grover, V., 2011. Interorganizational system character-istics and supply chain integration: an empirical assessment. Decision Sciences42 (1), 7–42.

Salomon, R., Martin, X., 2008. Learning, knowledge transfer, and technology imple-mentation performance: a study of time-to-build in the global semiconductorindustry. Management Science 54 (7), 1266–1280.

Salop, S.C., Scheffman, D.T., 1983. Raising rivals’ costs. The American EconomicReview, 267–271.

Sargan, J.D., 1958. The estimation of economic relationships using instrumentalvariables. Econometrica 26 (1), 393–415.

Schildt, H., Keil, T., Maula, M., 2012. The temporal effects of relative and firm-level absorptive capacity on interorganizational learning. Strategic ManagementJournal 33 (10), 1154–1173.

Schilling, M.A., 2002. Technology success and failure in winner-take-all markets:the impact of learning orientation, timing, and network externalities. Academyof Management Journal 45 (2), 387–398.

Schumpeter, J.A., 1934. The Theory of Economic Development: An Inquiry into Pro-fits, Capital, Credit, Interest, and the Business Cycle. Harvard University Press,Boston, MA.

Scott, J.E., 2000. Facilitating interorganizational learning with information technol-ogy. Journal of Management Information Systems 17 (2), 81–113.

Scott, W.R., 2005. Institutional theory: contributing to a theoretical research pro-gram. In: Smith, K.G., Hitt, M.A. (Eds.), Great Minds in Management: The Processof Theory Development. Oxford University Press, Oxford, pp. 460–484.

Sinha, R.K., Noble, C.H., 2008. The adoption of radical manufacturing technologiesand firm survival. Strategic Management Journal 29 (9), 943–962.

Stiroh, K., 2002. Information technology and the U.S. productivity revival:what do the industry data say? The American Economic Review 92 (5),1559–1576.

Stuart, T.E., 1998. Network positions and propensities to collaborate: an inves-tigation of strategic alliance formation in a high-technology industry.Administrative Science Quarterly 43 (3), 668–698.

Stuart, T.E., 2000. Interorganizational alliances and the performance of firms: a studyof growth and innovation rates in a high technology industry. Strategic Manage-ment Journal 21 (8), 791–811.

Subramani, M., 2004. How do suppliers benefit from information technology use insupply chain relationships. MIS Quarterly 28 (1), 45.

Suchman, M.C., 1995. Managing legitimacy: strategic and institutional approaches.Academy of Management Review 20 (3), 571–610.

Sun, H., 2013. A longitudinal study of herd behavior in the adoption and continueduse of technology. MIS Quarterly, in press.

Swaminathan, J.M., Tayur, S.R., 2003. Models for supply chains in e-business. Man-agement Science 49 (10), 1387–1406.

Teo, H.H., Wei, K.K., Benbasat, I., 2003. Predicting intention to adopt interorganiza-tional linkages: an institutional perspective. MIS Quarterly 27 (1), 19–49.

Tirole, J., 1988. The Theory of Industrial Organization. MIT Press, Cambridge, MA.Tornatzky, L., Fleischer, M., 1990. The processes of technological innovation. Lexing-

ton Books, Lexington, MA.Varian, H., 2001. High-technology Industries and Market Structure. University of

California, Berkeley, CA.Varian, H.R., Farrell, J., Shapiro, C., 2004. The Economics of Information Technology:

An Introduction. Cambridge University Press, Cambridge.Venkatraman, N., Ramanujam, V., 1986. Measurement of business performance in

strategy research: a comparison of approaches. The Academy of ManagementReview 11 (4), 801–814.

Wang, E.T.G., Wei, H.L., 2007. Interorganizational governance value creation: coor-dinating for information visibility and flexibility in supply chains. DecisionSciences 38 (4), 647–674.

3 ations

W

W

W

W

W

W

W

44 A.C. Sodero et al. / Journal of Oper

ebster, J., 1995. Networks of collaboration or conflict? Electronic data interchangeand power in the supply chain. The Journal of Strategic Information Systems 4(1), 31–42.

eitzel, T., Beimborn, D., König, W., 2006. A unified economic model of standarddiffusion: the impact of standardization cost, network effects, and networktopology. MIS Quarterly 30 (1), 489–514.

hite, H., 1980. A heteroskedasticity-consistent covariance matrix esti-mator and a direct test for heteroskedasticity. Econometrica 48 (4),817–838.

igand, R., Steinfield, C., 2005. Information technology standards choices and indus-try structure outcomes: the case of the U.S. home mortgage industry. Journal ofManagement Information Systems 22 (2), 165–191.

ilhelm, M.M., 2011. Managing coopetition through horizontal supply chain rela-tions: linking dyadic and network levels of analysis. Journal of OperationsManagement 29 (7), 663–676.

ooldridge, J.M., 2002. Econometric analysis of cross section and panel data. TheMIT press, Cambridge, MA.

ooldridge, J.M., 2005. Simple solutions to the initial conditions problem indynamic, nonlinear panel data models with unobserved heterogeneity. Journalof Applied Econometrics 20 (1), 39–54.

Management 31 (2013) 330–344

Young, G., Smith, K.G., Grimm, C.M., 1996. Austrian and industrial organizationperspectives on firm-level competitive activity and performance. OrganizationScience 7 (3), 243–254.

Zhang, C., Dhaliwal, J., 2009. An investigation of resource-based and institutionaltheoretic factors in technology adoption for operations and supply chain man-agement. International Journal of Production Economics 120 (1), 252–269.

Zhao, K., Xia, M., Shaw, M.J., 2007. An integrated model of consortium-basede-business standardization: collaborative development and adoption withnetwork externalities. Journal of Management Information Systems 23 (4),247–271.

Zhou, H., Benton, W.C., 2007. Supply chain practice and information sharing. Journalof Operations Management 25 (6), 1348–1365.

Zhu, K., Kraemer, K., Gurbaxani, V., Xu, S., 2006a. Migration to open-standard interor-ganizational systems: network effects, switching costs, and path dependency.MIS Quarterly 30 (I), 515–539.

Zhu, K., Kraemer, K., Xu, S., 2006b. The process of innovation assimilation by firmsin different countries: a technology diffusion perspective on e-business. Man-agement Science 52 (10), 1557–1576.

Zucker, L.G., 1987. Institutional theories of organization. Annual Review of Sociology13, 443–464.