AUTHOR - Fuqua School of Businesswillm/bio/cv/papers/EMR_2009Hollow...The core question is whether...

AUTHOR COPY

The hollow corporation revisited:

Can governance mechanisms

substitute for technical expertise

in managing buyer–supplier

relationships?Anne Parmigiani1, Will Mitchell2

1Lundquist College of Business, University of Oregon, Eugene, OR, USA;2The Fuqua School of Business, Duke University, Durham, NC, USA

Correspondence: Anne Parmigiani, Lundquist College of Business, University of Oregon, 1208, Eugene, OR 97403, USA.Tel: þ 1 (541) 346 3497;Fax: þ 1 (541) 346 3341;E-mail: [email protected]

AbstractThis article considers how a firm’s system of exchange skills including internal technicalexpertise and supplier governance mechanisms influence supplier performance, bothindependently and jointly. The core question is whether inter-firm governance mechan-isms, including both relational and contractual mechanisms, can substitute for a firm’sinternal technical skills in maintaining supplier performance or, alternatively, whether a firmrisks hollowing itself out by de-emphasizing internal expertise when it outsources. Thearguments build on the capabilities, inter-organizational governance, and supply manage-ment literatures. We find that internal technical expertise influences multiple dimensions ofsupplier performance, including cooperation, price, quality, delivery, and communication,while relational governance also affects supplier performance though in a more focusedway. In turn, combinations of technical expertise, relational governance, and contractualagreements jointly affect supplier performance. Thus, firms generate superior supplierperformance if they retain internal technical skills as well as increase their use of externalgovernance mechanisms to manage buyer-supplier relationships.European Management Review (2010) 7, 46–70. doi:10.1057/emr.2009.28Keywords: supply management; purchasing; buyer-supplier relationship; technology management;resource-based view

Introduction

Firms need to ensure that they achieve strong perfor-mance from suppliers when they outsource goods andservices, where outsourcing means using external

vendors to supply goods and services needed for a firm’sproducts. Dell, Nike, Boeing, Embraer, and Toyota relyheavily on relationships with suppliers to help them achieveindustry leadership in price, quality, and responsiveness tochanging demands. It is not clear, however, what skillsfirms require to achieve superior supplier performance.Some scholars emphasize inter-firm governance mechan-isms in helping buyers achieve strong supplier perfor-mance, where inter-firm governance mechanisms include

using contracts to specify terms and align incentives(Williamson, 1975; Macneil, 1978) as well as using rela-tional governance mechanisms, such as sharing informa-tion and providing performance feedback, to increasecommitment and generate common goals (Dyer, 1997;Dyer and Singh, 1998; Holcomb and Hitt, 2007). Otheranalysts, though, point to the importance of a firm’s owntechnical expertise, which is its understanding of the tech-nology underlying its goods and services (Grant, 1996), andquestion whether buyers can rely primarily on inter-firmgovernance mechanisms to ensure satisfactory supplierperformance. As Business Week (1986) asked more than

European Management Review (2010) 7, 46–70& 2010 EURAM Macmillan Publishers Ltd. All rights reserved 1740-4754/10

palgrave-journals.com/emr/

AUTHOR COPY

20 years ago and several scholars have echoed recently,do firms risk hollowing themselves out if they do not retainsufficient internal technical expertise to select, evaluate,and assist their suppliers (Richardson, 1993; Brusoni et al.,2001)?

This study contributes to our understanding of howa firm’s system of exchange skills (Macauley, 1963) contri-butes to effective supplier management. If governancemechanisms allow firms to substitute supplier expertisefor their own skills, then firms might be able to increasetheir outsourcing activities to the point that they primarilybecome assemblers of purchased components and servicesor even pure contractual brokers. If governance mechan-isms are not sufficient, however, firms that emphasize out-sourcing without maintaining their own technical expertisemight find themselves on downward spirals of unsatisfac-tory supplier relationships, becoming hollowed-out failuresin their own end-product markets. Thus, we seek to deter-mine whether firms can rely solely on suppliers’ technicalknowledge and achieve desired outcomes through contrac-tual and relational governance mechanisms or whetherfirms must retain their own internal technical skills to besuccessful buyers.

Our work is the most systematic study to date of howtechnical expertise and governance mechanisms, includingboth relational and contractual governance, independentlyand jointly affect supplier performance. Case studies ofaerospace and chemical firms by Brusoni et al. (2001)suggested that firms often ‘know more than they make’,suggesting that firms’ technical knowledge helps them beeffective buyers, but this work did not go into detail aboutrelational governance or consider contractual governance.Takeishi (2002) discussed the need for automotive buyersto have both architectural and component knowledge,but did not study governance. Dyer’s (1997, 2000) workhighlighted relational governance mechanisms in the autoindustry and alluded to the importance of expertise andcomplementary capabilities with suppliers, but did notconsider contracting. Mayer and Salomon (2006) studiedhow technical expertise can complement contracting byreducing hazards, but omitted relational governance.Several scholars (Cannon et al., 2000; Poppo and Zenger,2002; Carson et al., 2006; Cousins and Menguc, 2006) haveinvestigated the interplay between relational and contrac-tual governance, but none has also considered buyerexpertise.

Dissecting the effects of expertise and governance onsupplier performance is a complex undertaking. To betterunderstand these relationships, we take two steps to lever-age a rich data set based on a large-scale survey of smallfirms. First, we consider five distinct facets of supplierperformance, including price, quality, delivery, coopera-tion, and communication. This multi-dimensional app-roach enables us to understand the distinct effects ofexpertise, contractual, and relational governance mechan-isms on different types of performance, in contrast toprior work that typically studies only individual elementsof performance. By considering these, we gain a nuancedview of the buyer/supplier relationship and uncoverpotential trade-offs in performance. Second, we recognizethe endogeneity problem of sourcing mode choice and per-formance. Buyers with considerable expertise may choose

to produce internally, while buyers with superior skills ingovernance may choose to outsource. Our data allow usto control for this possibility through a two-stage modelingprocess in which we first predict which firms will outsourceand then incorporate this effect into the performance predi-ctions (Hamilton and Nickerson, 2003). This method isatypical for studies of supplier performance, but is vital toaccurately understand this outcome.

Thus, our work contributes to the capabilities and supplymanagement literatures by integrating the discussion ofhow governance and technical expertise can influencemultiple dimensions of supplier performance, while con-trolling for firms’ initial choice to outsource. The nextsection provides illustrations of outsourcing and buildsour theoretical predictions. We then describe our survey,which examines buyer-supplier relationships among 164firms in the US metal forming industry in 2002, and discussmethods of analysis. The results then demonstrate thattechnical expertise and relational governance directly andjointly improve multiple aspects of supplier performance.We conclude with implications and extensions of ourfindings that firms need capabilities in all three areas tomanage suppliers effectively.

Background and predictions

Examples of outsourcing successes and difficultiesTo illustrate how technical expertise may relate to con-tractual and relational governance mechanisms, we beginby discussing how several firms in the auto, computer, andfootwear industries manage their suppliers. Toyota exem-plifies a firm that emphasizes both relational governanceactivities and strong technical expertise. Toyota’s relationalgovernance activities allow the company to share informa-tion among suppliers while building high performance,long-term relationships. Enright (2003), for instance, notedthe importance of both contracts and relational governanceprocesses in Toyota’s supplier governance. In turn, theinternal technical expertise means that the company canboth evaluate suppliers effectively and help them moveforward. As Liker (2004: 208) puts it, ‘Toyota believed thatit needed to truly master any core technology in order tomanage its suppliers effectively’; Liker used the example ofsemiconductor technology for hybrid cars, in which Toyotafirst developed its own technical expertise so that thecompany could outsource more effectively.

General Motors and Ford provide counter-point exam-ples to Toyota. Both companies have high productionexpertise but have historically placed less emphasis on rela-tional governance, especially regarding information sharingand developing collaborative, long-term supply relation-ships. As a result, these companies often can pressure theirsuppliers concerning price and quality, because theyrequire high volumes and use detailed contracts to addresscost and quality demands, but frequently struggle to main-tain cooperative relationships that contribute to longer-term goals such as innovation (Enright, 2003).

Dell is an example of a computer company that relies onexternal suppliers. To support their suppliers, Dell empha-sizes both contractual and relational governance, typicallycontracting with partners and sharing information with

Supplier governance and expertise Anne Parmigiani and Will Mitchell

47

AUTHOR COPY

them. At the same time, Dell historically maintained apool of technical expertise rather than relying solelyon supplier skills. Magretta (1998) notes that Dell’sResearch & Development group focuses on process andquality improvements in manufacturing and works closelywith suppliers, both to evaluate supplier skills and tointegrate them into Dell’s ‘virtual integration’ productionprocess. Tan and Young (2003) also noted that Dellmaintains the skills needed to assemble and test vendorprocured parts and assemblies, again so that the companycan both evaluate suppliers and help improve theirprocesses.

Nike also uses suppliers extensively to generate highend-product performance. While Nike’s attention togoverning its partnerships receives some recognition(though much less recognition than its marketing strength),the importance of the company’s own technical expertisereceives much less attention. We discussed this study witha Nike executive, asking him about the importance ofsupplier governance and Nike’s own technical expertise. Henoted first that Nike undertakes extensive relationalgovernance activities, with extensive quality, delivery, andcost auditing processes. The company both evaluatessuppliers closely and presses them for quality and costperformance. Second, Nike typically uses contracts withits suppliers, often exceeding 15 years in length. Third, Nikehas strong technical expertise in footwear and is respon-sible for advanced R&D, manufacturing R&D, and productdevelopment. Technical groups within the company pro-vide a significant amount of direction to their suppliers.Much of the company’s product and manufacturinginnovation comes from groups at Nike headquarters orlocations in Asia, where they have manufacturing directorsin each location who assist suppliers on site. Nike alsofunds and runs learning centers, at which visiting supplierslearn manufacturing techniques, particularly principles oflean manufacturing. Finally, he noted that relationalgovernance activity and technical expertise reinforce eachother, particularly in generating cooperation.

These examples suggest that all three skills – internaltechnical expertise, contractual governance mechanisms,and relational governance mechanisms – can strengthenoutsourcing performance. The next section discusses howthese factors may influence supplier performance, indivi-dually and in combination.

Supplier performanceFirms commonly produce a subset of the goods andservices that make up their end products internally, whilepurchasing other services and physical components fromexternal suppliers. Supplier performance affects buyeroutcomes on several dimensions. The prices supplierscharge influence buyer profitability; over 60% of a firm’scosts can arise from purchased components (Degraeveand Roodhooft, 2001). The quality of purchased itemsaffects a buyer’s production processes, the quality of its endproducts, and its reputation with customers (Womacket al., 1990; Mascarenhas et al., 1998). Cooperative relation-ships with suppliers affect short-term performance bysmoothing deliveries and reducing tactical coordinationcosts, while also influencing longer-term performance by

helping firms develop new capabilities (Dyer and Nobeoka,2000; Novak and Eppinger, 2001). Research has measuredsupplier performance on dimensions such as quality,cost, responsiveness, improvements in product and processdesign, lead time, and inventory turns (Noordewier et al.,1990; Poppo and Zenger, 1998; Krause et al., 2007). Thecore conclusion is that multiple aspects of supplier perfor-mance strongly influence a buyer’s short and long-termperformance in its end-product markets (Elmaghraby,2000).

In order to understand firm performance, therefore, it isimportant to identify factors that influence different aspectsof the performance of a firm’s supplier relationships. Weconsider the direct effect of the buyer’s own technicalexpertise and its supplier governance activities, as well asinteractions among these factors, and develop predictionsfor each effect. We state our hypotheses in terms of thegeneral concept of supplier performance, and then test theimpact of technical expertise and governance mechanismson five key dimensions, including quality, cost, delivery,communication, and cooperation.

Technical expertiseTechnical expertise is the extent to which a buyer under-stands the production processes and affiliated technologiesunderlying a good or service (Grant, 1996; Parmigiani andMitchell, 2009). Technical expertise is a heterogeneous firmcapability, developed over time, composed of interrelatedroutines (Barney, 1991; Szulanski, 1996). We assume thegood requires a certain degree of knowledge to purchase iteffectively and thus a non-trivial amount of specific invest-ment. Firms can gain technical expertise directly throughproduction of the component or indirectly through pro-ducing related products and conducting relevant R&Dactivities (Cohen and Levinthal, 1990). Firms tend tospecialize in similar activities, as they can use commontechnical expertise across products (Richardson, 1972). Wefocus on technical skills rather than on other functionalexpertise, such as marketing skills, because technicalexpertise is particularly important in facilitating supplierrelationships – a buyer needs to understand the technicalcharacteristics of the inputs to its products (Brusoni et al.,2001; Takeishi, 2002). Firms with greater technical expertisewill better predict how attributes of the product can affectmanufacturing processes and, ultimately, the performanceof the end product.

A buyer’s technical expertise will contribute to supplierperformance in four ways: buyers will be able to identifycompetent suppliers, monitor supplier activities moreknowledgeably, provide greater assistance to suppliers,and attract better suppliers. First, firms that understandcore elements of the development and production processwill be better able to identify strong suppliers. Technicallyproficient buyers will be adept at screening and selectingsuppliers because they can interpret their offerings,comparing them on technical attributes as well as price(Petersen et al., 2005). Second, technically strong buyerswill be better able to understand their suppliers’ activities,so that they can monitor their performance more effecti-vely to ensure that they meet the buyers’ specifications(Wheelwright and Hayes, 1985; Krause et al., 1998). Third,


48

AUTHOR COPY

buyers with superior technical expertise will be better ableto communicate technical details to suppliers. Their exper-tise provides them with a component evaluation capabilityby which they are better able to designate appropriatequality metrics and levels, making these both strict andachievable (Lincoln et al., 1998; Tiwana and Keil, 2007).Thus, they can assist in improving supplier skills (Bradach,1997; Krause et al., 1998; Dyer, 2000). Fourth, technicallystrong buyers will be able to attract better suppliers(Richardson and Roumasset, 1995). Suppliers can userelationships with strong buyers to learn about the latestdevelopments and gain knowledge to trade with customers(Hayes et al., 1988; Powell et al., 1996). Suppliers also oftenwant to be part of an elite group of firms supporting aprominent, knowledgeable lead firm (Dyer and Nobeoka,2000). The fact that strong customers are likely to surviveinto the future also will attract suppliers and encouragethem to cooperate, because they can expect a long rela-tionship (Heide and Miner, 1992). Thus, we propose:

Hypothesis 1: The stronger a buyer’s internal technicalexpertise relevant to a purchased good, the greater theperformance the buyer will achieve in its supplierrelationships for the good.

Governance mechanisms: relational governance and contractsA firm’s governance activities also may influence supplierperformance. We first discuss the concept of relationalgovernance mechanisms in the context of supplier manage-ment and then develop a parallel argument concerningcontractual governance. We assume that purchased goodsrequire active management of suppliers due to a non-trivialdegree of knowledge and investment required to under-stand and procure these goods effectively.

We define relational governance mechanisms for suppli-er management as the processes a firm uses to manage itsconnections with suppliers, such as processes for main-taining goodwill, sharing information, and evaluating supp-liers. For simplicity, we refer to these processes as relationalgovernance mechanisms. This definition draws from ideasin the relational capability literature (e.g., Dyer and Singh,1998), applying them to the context of supply management.

Relational governance mechanisms accumulate overtime, become reinforced through regular routines (c.f.,Nelson and Winter, 1982), and are analogous to alliancemanagement mechanisms in that firms will be moresuccessful if they dedicate resources to this activity (Kaleet al., 2002). While firms purchase a variety of goods, eachwith differing attributes, information requirements, andpotential governance hazards, these relational governancemechanisms represent a firm-level approach or philosophyto holistically managing its supply base. Firms varysubstantially in their use of relational governance mechan-isms, even if they have common institutional or industrialcontexts (Lincoln et al., 1998). The classic example is in theautomotive industry in which some firms (e.g. GM) havehad an adversarial approach while others (e.g., Toyota)emphasize partnerships.

The concept of relational governance derives fromMacneil’s (1978, 1980) work on relational contracting, in

which he contrasts one-off exchanges with ongoingexchanges that are based within and upon the relationshipbetween the parties. Thus, maintaining goodwill, solidarity,integrity, and harmony are important. Information sharingis also vital, as this promotes mutuality within the relation-ship and reduces conflict. While perhaps less emphasized,evaluation is also a key component of relational govern-ance, because it promotes fairness through measurementand sticking to the terms of the agreement. QuotingMacneil (1980: 72), ‘Many a problem occurs in perfor-mance, many an outfit skirts too close the edge ofnonperformance and falls off, and a few evildoers in everycrowd cheat or welsh’. Evaluation will reduce the risk ofnonperformance and strengthen the bonds between buyersand suppliers, establishing the parties’ reputations andpromoting future exchange.

Womack et al. (1990) demonstrated the profound impactof cooperative supplier relationships in their studies ofthe automotive industry, which contrasted Japanese firmsthat compared developmental relationships with US firmsthat took more adversarial approaches. Supply relation-ships can be a self-fulfilling prophecy, because firms thatstrive to create and maintain partnerships with supplierswill be likely to have more harmonious relationships thanthose that emphasize more adversarial roles (Hayes et al.,1988). Firms that build upon past relationships and makecommitments to future exchange will build trust and pro-mote cooperation (Heide and John, 1990; Heide and Miner,1992; Zaheer et al., 1998; Carson et al., 2006; Poppo et al.,2008). Activities that deepen the buyer–supplier relation-ship through richer communication and integration leadto stronger bonds of respect and trust, which then lead toimproved performance (Chen et al., 2004; Lawson et al.,2008). In a study of buyer-supplier dyads, Krause et al.(2007) found that while relationship length was notassociated with improved performance, shared values andhigher buyer commitment did lead to better outcomes.They also found that buyer involvement improved quality,flexibility, and delivery, but not cost, suggesting differentialeffects of relational governance.

In addition to maintaining goodwill, processes forinformation exchange and knowledge transfer are impor-tant to effective relational governance as they promotecoordination, adaptation, and learning (Ring and Van deVen, 1992). Dyer and Singh (1998) investigated knowledgeexchange between buyers and suppliers, noting that a firmthat develops these relational capabilities can gain compe-titive advantage. Hult et al. (2007) suggest that promotinga culture of learning and protecting against knowledgeappropriation facilitates exchange. Holcomb and Hitt(2007) concur with both of these insights and proposethat skills in aligning firm and supplier goals, creatingknowledge sharing routines, and developing cooperativeexperience will lead to more effective outsourcing. Moregenerally, alliance studies suggest that boundary spanningactivities help a firm overcome the constraints of localsearch (Mowery et al., 1996; Rosenkopf and Almeida, 2003).

Supplier evaluation and development are more formalmeans of relational governance. Evaluation can includethird-party certification, audits, and written scorecardscomparing suppliers (Dyer, 2000; Petersen et al., 2005).These tools uncover problems before they become crises


49

AUTHOR COPY

and use reputation to bolster suppliers’ efforts. AsMacauley (1963: 63) states, ‘some industrial buyers go sofar as to formalize this sanction (of preserving reputation)by issuing ‘‘report cards’’ rating the performance of eachsupplier. The supplier rating goes to top management ofthe seller organization and these men can apply internalsanctions’. Thus, relational governance can include bothinterpersonal and documented elements. Some aspects arequite formal and codified but, unlike contracts, thesetechniques are highly customized and third parties cannotprovide enforcement or sanctions. Supplier developmentefforts, including technical consultation, creating supplierassociations, and frequent engineering visits also improvethe richness and relevance of knowledge shared. Theseactivities are the culmination of the aforementionedrelational skills of maintaining goodwill, exchanging infor-mation, and evaluating suppliers (Krause et al., 1998; Dyerand Nobeoka, 2000). This logic leads to H2:

Hypothesis 2: The more extensively a buyer employsrelational governance mechanisms, the greater theperformance the buyer will achieve in its supplierrelationships.

Firms often support supplier relationships throughcontractual agreements, whether to complement or sub-stitute for relational governance mechanisms. Contractsare written agreements that designate the ex-ante featuresof the exchange, involve considerable detail, and mayspan several years (Macneil, 1978). Contracts differ fromstandard purchase orders, which cover a simple, one-timetransfer of goods with no provision for change. Contractdocuments not only dictate product specifications, annualquantities, prices with clauses for price changes, and penal-ties for non-performance, but are also rife with legalconditions. The agreements are typically negotiated bysenior managers from various functional areas, involvestrategically significant goods, and provide blanket agree-ments that create a template for ongoing exchange,often including change processes (Argyres and Mayer,2007). Contracts also assist in subsequent monitoring ofexchanges (Alchian and Demsetz, 1972). Firms can considersupplier incentives and occasions for opportunism, build-ing safeguards into the agreement (Williamson, 1985).As they are legal documents, each party has the optionof invoking the courts and thus involving a third party inthe event of a breach in performance. By agreeing toa contract, both the firm and the supplier agree to dealfairly with each other (Helper and Levine, 1992).

Contracts can improve supplier performance by improv-ing communication, providing investment incentives, high-lighting the importance of a relationship, and threateningsanctions. The process of creating an agreement helpsdevelop communication between a firm and its supplier,clarifies expectations, generates a common understandingof goals, and facilitates coordination of buyer-supplieractivities (Arrow, 1974; Reuer and Arino, 2007; Mayerand Teece, 2008). A contract helps assure a supplier willprofit from investments made to support a series ofexchanges, thereby creating incentives to maintain perfor-mance levels and protect these investments (Williamson,1985). In parallel, Kulwani and Narayandas (1995) argue

that suppliers may be more likely to focus on contractedcustomer relationships, because the contract signifies thatthe purchase is important to both the buyer and supplier.In addition, suppliers will want to avoid sanctions outlinedin the contract as well as court costs and reputationaldamage from legal proceedings and therefore will attendmore closely to performance targets, fixing problems beforethey become serious (Macneil, 1978, 1980). Thus, wepredict:

Hypothesis 3: Buyers that use contractual agreementsto govern supply relationships will achieve greaterperformance in their supplier relationships than buyersthat do not use contracts.

The question of endogeneity arises here. It is possiblethat observed governance mechanisms will have no impacton supplier performance if buyers selectively chose to userelational governance and/or contracts in cases that bestsuit them, such as for specialized components, while usingmore arms length approaches in other cases, such ascommodity components. Our empirical analysis will con-trol for some aspects of such endogeneity. Nonetheless,both relational governance and contract negotiation arelearned skills, so that firms that develop more extensivegovernance activities may benefit across their portfolio ofsourcing relationships, even in the presence of potentialendogeneity. Moreover, endogeneity will be less of an issuefor the interaction effects that we discuss below.

The first three hypotheses test what are sometimes posedas competing views on relationship management, but maywell have complementary influences on supplier perfor-mance. The empirical analysis will assess the degree ofcomplementarity or substitution. In addition to directeffects, there may be positive interactions among thedifferent aspects of relationship management, potentiallyreinforcing each other, which we discuss below.

Joint effects of technical expertise and governance mechanismsWe will consider three possible joint effects among technicalexpertise, relational governance, and contractual agree-ments. First, greater technical expertise may enhance thebenefits of relational governance. One component of rela-tional governance involves performance monitoring andevaluation. A buyer’s technical expertise will assist in deve-loping accurate and detailed specifications to provide thebasis for later evaluation. Rather than relying on a standardscorecard, buyers with more knowledge can create bettertools for assessing quality, resulting in improved perfor-mance. Smarter buyers will also be better able to detectquality slippage and assist their suppliers in modifying theirprocesses. Suppliers also will want to perform well to live upto buyer expectations and not breach their trust in therelationship. In addition to more astute supplier evaluation,greater technical skills will lead to more extensive butfocused development activities. Expertise thus improvesevaluation and development, both important aspects ofmanaging supply relationships and improving performance(Krause et al., 1998; Petersen et al., 2005).

Sharing information is also a critical component ofrelational governance. More knowledgeable buyers, who


50

AUTHOR COPY

can discuss complex technical details, should benefit fromdeeper information exchange with suppliers. The combina-tion of technical expertise and relational governancemechanisms may help the firm develop communication,promoting greater learning and cooperation (Richardson,1972; Tunisi and Zanfei, 1998; Zollo et al., 2002). Suchexchanges will be valuable in managing changes in pro-duction due to variation in volumes and shipment timing.Buyers with a greater technical understanding will also bebetter able to predict the next generation of products andwork with suppliers to develop these, thus extending theirrelationships. To learn from suppliers, firms need botha basic understanding of the technology (Cohen andLevinthal, 1990) and partners who are willing and able toexchange ideas.

Several studies suggest that firms with both strongtechnical expertise and the ability to create close externalrelationships enjoy competitive advantages, because theycan better identify and act upon opportunities. Brusoniet al. (2001) found that aircraft and chemical firms thatact as systems integrators cannot fully outsource all tech-nical understanding but rather need to maintain a cognitiveoverlap with suppliers to manage inter-dependenciesamong purchased components. MacDuffie and Helper(1997) showed that automotive firms that wish to createa competitive advantage through their suppliers needboth a technical understanding and motivated suppliers.Bradach (1997) suggested that restaurant franchisors thatoperate their own outlets and thus gain a technical under-standing of the business benefit from sharing ideas withtheir franchisees. Holweg and Pil (2008) studied threeautomotive supply chains and found that performance wasinfluenced by the combination of social structures andtechnology as established patterns of communication andbehavior affected deployment of new systems, suggestinga synergistic effect between relationships and technology.Thus, we posit:

Hypothesis 4a: Stronger internal technical expertiserelevant to a purchased good and more extensive use ofrelational governance mechanisms will mutually enhancesupplier performance.

Second, a firm’s technical expertise should heighten theeffectiveness of contractual agreements. Firms that have abetter technical understanding of a component can createmore complete and meaningful specifications for suppliersto follow. Rather than following a standard boilerplateagreement, knowledgeable buyers can customize a contractto better meet both their needs and those of their supplier(Argyres and Mayer, 2007). They should be able to nego-tiate better terms, particularly regarding specific qualityrequirements, because they have some understanding oftheir supplier’s production process. While, by their nature,contracts are always incomplete, skilled buyers can betteranticipate technical changes and incorporate contingenciesinto the contract.

Once the contract has been negotiated, technicalexpertise will help the firm monitor and evaluate suppliersto enforce contract terms. Suppliers will recognize thefirm’s expertise and be less likely to engage in opportunisticbehavior, such as substituting inferior materials, because

the firm will be able to recognize infractions. A firm’stechnical reputation based upon frequent trade can alsoreinforce contracting. (Lyons, 1996). Firms with moretechnical expertise tend to craft better contracts and areable to reduce disruption and other hazards (Delios andHenisz, 2000; Mayer and Salomon, 2006). This logicsupports the following:

Hypothesis 4b: Stronger technical expertise relevant to apurchased good and the use of contractual agreementswill mutually enhance supplier performance.

Third, joint use of relational governance mechanismsand contractual agreements also may contribute to superiorsupplier performance. Several scholars have suggesteda complementarity between these aspects of the supplyrelationship, particularly since contracts will always beincomplete. Macauley (1963) argues that contracts areimportant but operate better in the background rather thanthe foreground of the supply relationship so that the giveand take of business can be uninterrupted by lawyers. Hesuggests that contracts are valuable communication devicesand organizing tools which serve to stabilize the relation-ship, but that business professionals ‘welcome a measure ofvagueness’ in contracts so they can ‘handle disputes in theirown way’. Cannon et al. (2000) echo this point, and foundthat cooperative norms complemented legal contractsunder conditions of uncertainty and specific investment.Carson et al. (2006) suggest that the type of uncertaintymatters; ambiguity is better addressed through contracts,while volatility is more effectively managed throughrelationships.

Because contracts reflect the underlying social context(Macauley, 1963; Macneil, 1980; Baker et al., 2002), firmswith more extensive relational governance skills may createmore customized contracts. Cousins and Menguc (2006)found that greater socialization through regular on-sitevisits and workshops as well as greater integration such aslinked production schedules led to improved contractualconformance, suggesting that these relational activitiespromote a better understanding of buyer expectations.Poppo and Zenger (2002) demonstrated that increases inrelational governance tended to be coupled with morecomplex contracts, suggesting that a close relationshipeases the process of creating the contract by narrowing thedomains for opportunism and allowing the document to bemore customized and specific. Zheng Zhou et al. (2008)found that even in China, where relational ties are vitallyimportant, firms use customized contracts to mitigateopportunism due to uncertainty. Klein and Leffler (1981)posit that contracts include some self-enforcing range suchthat the agreements are based on a combination of privateand joint enforcement. Ryall and Sampson (2009) foundthat relationship history affected the type and degreeof contract detail, suggesting that contracts can serve asa roadmap for enforcement of relational sanctions. In thisway, supplier report cards and other relational governancetools could be explicitly or implicitly incorporated intocontracts, with the potential to invoke contractual andrelational sanctions.

Richer relationships may also lead to more effectivecontract execution with more cooperation and less conflict.


51

AUTHOR COPY

Contracts indicate firms’ intent to maintain a relationshipand support cooperation as they expect the relationship tolast (Macneil, 1978; Heide and Miner, 1992; Helper andLevine, 1992). Sako (1992) identified contracts as providinga form of trust that complements two other forms of trust(competence and goodwill), suggesting that contracts helpstabilize relationships. Inter-organizational trust, presum-ably developed through relational mechanisms, reduces thecosts of contract negotiation and subsequent conflict(Zaheer et al., 1998). Contracts can also act as coordinationdevices, which may be more effective when combined withtrust in the relationship as this reduces risk and facilitateslearning (Mellewigt et al., 2005; Reuer and Arino, 2007).In a sharp example of what can happen to a firm that lacksboth relational and contractual governance, Harrison(2004) described how the courts refused to recognizeWilliam Baird’s 30-year, exclusive relationship with Marksand Spencer as relevant – because there was no writtenagreement, Marks and Spencer could terminate withoutnotice and without compensating Baird for any invest-ments.

These arguments suggest that relational governance andcontractual governance act as complements. Strongerrelationships may enable crafting of more effectivecontracts and help deal with uncertainty. Contracts mayact as broad strategic frameworks, while relationships fill inthe tactical and specific details on how business actuallygets done. This discussion leads to our final hypothesis.

Hypothesis 4c: More extensive use of relational govern-ance mechanisms and the use of contractual agreementswill mutually enhance supplier performance.

Counter to hypothesis 4c, scholars such as Ghoshal andMoran (1996) have argued that contracts may inhibit rela-tional governance activities, especially in contexts wherestandardized contracts are commonplace. However, suchconstraints would only arise if firms systematically allowedcontracts to impose limits on their relational activities,rather than creating a common language from whichrelationship management could proceed. Hypothesis 4cassumes that firms at least attempt to avoid suchirrationalities; our empirical analysis will help assess theviews.

In sum, we predict that a firm’s technical expertise andits governance activities will influence supplier perfor-mance, both individually and jointly. By testing thesehypotheses, we can uncover potential tradeoffs for firmsin managing their supply relationships. For example, iftechnical expertise does not influence supplier perfor-mance, firms may be wasting efforts on developing internalknowledge and would be better off developing their con-tractual and relational governance skills. If so, then thecore capabilities of the outsourcing firm could well begovernance skills rather than technical expertise, such thata seeming ‘hollowing out’ of the firm would not bedetrimental. Likewise, if relational governance does notaffect supplier performance, firms could spend their effortsin other areas and perhaps benefit from more competitivesupply relationships. Moreover, if contractual governanceis ineffective, firms might instead rely on other means ofmanaging their relationships. If we find that technical

expertise, relational governance, and governance mechan-isms are substitutes, then firms could select which of theseto emphasize, potentially conserving resources since thestrength in one area could compensate for a lack of skillsin the others. Conversely, if these skills are complements,then firms would benefit by maintaining a base level ofeach skill.

Data and methodsWe studied sourcing decisions of North American metalstamping and powder metal firms for production toolingand service components. These two sectors of the metalforming industry consist of many independent smallcompanies, most of which are private firms. This choiceenables us to focus on relatively straightforward buyer-supplier relationships, ruling out alliances, joint ventures,and other complex relationships that these firms rarelyundertake. While the firms do not gain press attentionwhen they outsource, they still need to carefully considerhow their expertise and governance skills will enable themto implement this strategy. Given the small size of thesefirms, their choice to outsource a particular activity (diemaking) may have the same relative impact on operationsand employment as a Fortune 500 firm outsourcing anentire function (manufacturing). The firms and theirsuppliers are similar in size, controlling for explanationsbased on size and power differentials. Both sectors sharerelatively homogenous production processes and commonend product markets. Supplier costs, delivery, communica-tion, cooperation, and quality levels are vitally important tothese firms and their OEM customers, many of whom are inthe automotive industry (Womack et al., 1990).

We first undertook exploratory interviews with managersof 11 metal forming firms. The interviews helped usunderstand their production processes and identified fivegoods that firms commonly source: die designs, construc-tion of progressive stamping dies (or, in the powder metalindustry, powder compaction dies), die maintenance,end-part machining services, and end-part surface coatingservices. These inputs were common to all firms, strategi-cally significant, and often outsourced. Note that our datawere at the input level, not at the transaction level, as thelatter is a theoretical archetype and infeasible to collect.Even these small firms have hundreds of individual ordertransactions annually. More importantly, managers do notmake decisions at the transaction level. Rather, theyconsider characteristics of the input such as annual volumeor technical features and use these to determine whetherto outsource and from which suppliers. Individual transac-tions, or purchase orders, merely represent how theyimplement these decisions (van Weele, 1994).

In order to produce their end product, these firms mustsource the designs, construction, and maintenance of thedies they use. Die design is a complex process in whichdesigners must select the proper type of steel, considerpress features, and optimize die layout to maximizeproductivity. Dies involve numerous interrelated compo-nents, including top and bottom shoes, punches, bearings,springs, and guide pins. They are typically produced fromtool-grade steel using computer numerical controlled(CNC) mills and electric-discharge machining (EDM)


52

AUTHOR COPY

centers. A considerable degree of expertise is required fordie design and construction, as suggested in a machininghandbook: ‘y die design and die making are complex artsas well as technologies that require considerable skill,knowledge, and practical experience’ (Walsh, 1994: 792).Die maintenance is somewhat less sophisticated, typicallyinvolving the regrinding and replacement of punches after aset number of press strokes or after a press problem. Firmsgenerally source one design or die at a time, comparingbids from several suppliers. Contracts are relativelyunusual; rather, relationships develop between buyers andsuppliers, since with experience the supplier can under-stand the idiosyncrasies of the buyer’s presses andproducts.

End-part machining and surface coating are downstreamoperations necessary to meet customer specifications. End-part machining involves adding features such as slots orholes that cannot be stamped or molded into the part.Surface coating includes plating, electro-coating, or similarprocesses that are typically used for corrosion resistance.Firms usually obtain bids from numerous suppliers butprefer to source locally, due to relatively high volumes andthe need for rapid turnaround. End-part machining andsurface coating are often governed by contracts specifyingthe approximate annual volume required, price, andextensive specifications detailed in drawings and otherrelated documents. Many of the end products of thesebuyers are, in turn, sold to OEM manufacturers (e.g., auto-mobile producers), so contract specifications or informalrequirements include the OEM’s quality and technicaldemands. Because these demands are often difficult topredict, relational governance mechanisms such as frequentsupplier contact are common.

The following example from our empirical settingillustrates how firms in the industry use combinationsof contracts, relational governance, and technical expertiseto manage their supplier relationships. Windfall Products(now a division of Metaldyne, Inc.) is a powder metal partsfirm that manufactures gears, rod guides, exhaust gasrecirculation components, and similar products for autosector companies such as the Delphi Corporation. Windfallhas strong technical expertise in its product lines. Windfallused several mechanisms to manage its relationship withMonroe Plating, a zinc plating supplier. First, a multi-yearcontract stipulated major terms of the relationship, suchas product specifications, annual volumes, and prices.In addition, Windfall personnel visited Monroe’s facility ona monthly basis to conduct quality audits and to maintainpositive relations. Three people from Windfall frequentlyparticipated in the visits: a purchasing manager discussedprice and product flows; a supplier development coordi-nator worked on quality issues and evaluated Monroe’sperformance; and a metallurgist helped Monroe solvetechnical problems concerning the plating and confirmthat their processes met Windfall’s and the OEM’s specifi-cations. Thus, Windfall used contracts and relationalgovernance in managing its relationship with Monroe, butalso leveraged its own technical expertise.

Based upon our understanding of the firms and the fivegoods, we designed a survey booklet with six sections, onefor data on each good and one for overall firm information.This design resulted in rich, fine grained, multi-level data

appropriate for investigating sourcing decisions andrelationships. It also removed recency, size, and familiaritybiases because we did not ask respondents to nominatea supplier but rather inquired about sourcing of classes ofgoods used by the firms. For firms that used multiplesuppliers, we asked respondents to create a summary ratingbased on their allocation of business to each. We chosescales based on reviews of the literature, refined bydiscussions during the preliminary interviews. AppendixA reports the items and their sources; most items usedseven-point scales (Fowler, 1995). The full survey containedabout 300 items in 24 pages, covering several aspects ofthe sourcing decision in addition to contractual andrelational governance mechanisms and supplier perfor-mance. To ensure face validity, we pre-tested the survey bysoliciting feedback from academic colleagues, managers,and industry association executives. We also performeda pilot test with managers that replicated final surveyconditions. We considered surveying suppliers, but abilateral or dyadic survey was infeasible due to the frag-mented nature of the industry, the difficulty in locatingsuppliers (e.g., a die-making firm can be as small as threepeople, operating out of a garage), and the lack of a supplierindustry association or other entity that could providea sampling frame. Hence, we followed common practicein buyer/supplier studies and used the buyer’s perspectiveto assess relationships.

Fortunately, industry associations exist for metal stamp-ing firms (the Precision Metalforming Association) andpowder metal firms (the Metal Powder Industry Federa-tion). From these groups, we obtained membership liststhat we used as the basis for our sample. We called each ofthe 509 member firms to identify the correct contactperson, typically the general manager, to whom to send thesurvey. In 2002, we sent the survey to the 453 firms thatprovided us with contact information. Following Dillman’s(1978, 2000) tailored design method, we initiated betweentwo and six contacts by various modes with each firm,resulting in 193 usable surveys and a 43% usable responserate. This is significantly higher than the 20% rate that iscommon for firm-level surveys (Paxson et al., 1995).

We adopted a key informant single-respondent approachfor the survey. While in some cases it is preferable to havemultiple survey respondents, we believe that due to thesmall size of these firms and the technical and specializednature of the survey, it was appropriate to requestinformation from one highly knowledgeable respondent(Phillips, 1981; Li and Atuahene-Gima, 2002). In our case,the respondent was typically the president or generalmanager. Senior managers in metal forming firms typicallyhave considerable experience in purchasing productionitems, particularly because this is a mature industry andtechnology, with well-established buying and supplyingfirms. One of the authors worked for several years asa purchasing manager in this industry and can attest thatone senior executive typically has oversight of sourcingdecisions and managing suppliers. Our first-hand knowl-edge of this context and the fact we asked about overallsourcing experience over the past year for several goodsalso gave us confidence in this approach. Thus, we believethe use of a single, well-qualified respondent did not biasour sample and was appropriate in this context.


53

AUTHOR COPY

Nevertheless, we took several precautions against com-mon method bias, because we were obtaining ourdependent and independent variables from the same source(Lindell and Whitney, 2001; Podsakoff et al., 2003). Ourprocedural remedies included separating the items relatingto these variables by several pages, using different scaleanchors, and reverse scaling some items. To reduce itemambiguity, we kept questions simple, avoided double-barreled items, and cognitively pre-tested the instrumentprior to launch (Campanelli, 1997). Social desirability wasminimized by protecting respondent anonymity and byobtaining information about a class of goods, rather thana particular supplier. Once we obtained the data, we alsoconducted the Harman one-factor test (Podsakoff andOrgan, 1986), which indicated that common methodsvariance was not an issue.

The responding firms had similar demographic features.Respondents tended to be small (95% employed fewer than500 people, with a mean of about 75 employees), non-union(86%), and fairly old (mean of 44 years in age). Weanalyzed the data for non-response bias (Armstrong andOverton, 1977), comparing respondents to non-respon-dents by firm type and size and found no significantdifferences. Based on these tests and the starting point thatour sample drew on industry association listings thatrepresent the firm populations, we believe non-responseand sample selection bias are unlikely. We also used severalcontrol variables in the subsequent analysis, which willfurther reduce potential bias (Tomaskovic-Devey et al.,1994). After culling cases with no external supplier, the datarepresent 508 purchases for 182 firms for five goods.

VariablesThe following section describes how we measured ourdependent variable, supplier performance, and our keyindependent variables: technical expertise, relational gov-ernance mechanisms, and contractual agreements. Rela-tional governance is a firm-level variable, based on thepractices and philosophies of supply management thatapply to a firm’s entire supply base, while performance andour other key explanatory variables are good-specific. Wealso describe good and firm-level control variables.

Supplier performanceWe devised five performance variables that executives inour preliminary interviews identified as important: suppliercooperativeness, quality, price, delivery, and communica-tion (see Appendix A). These dimensions align with thosefrequently used in the buyer/supplier literature (Krauseet al., 2007). We chose to look at each dimension separately,rather than group performance into one or two factorsso we could understand the impact of our explanatoryvariables on each specific type of performance; there is noa priori reason to believe that different types of perfor-mance tend to move together. We used seven-point Likertscales for the measures, ranging from ‘absolutely terrible’ to‘absolutely terrific’; Fowler (1995) notes that these anchorterms are more descriptive than the standard agree/disagreeterms and should result in better measurement becausethey are more indicative of a buyer’s perception of supplierperformance.

We asked about multiple performance dimensions foreach of five different goods, while limiting ourselves toa single item for each performance dimension for each typeof component so that we would not overly burden therespondents. Respondents reported no ambiguity in under-standing the performance dimensions during pre-tests ofthe instrument, so that single-item measures were appro-priate. For the focal goods, respondents reported that theytypically dealt with four or fewer suppliers, with a mean oftwo. This small number reduces the possibility of ecologicalfallacy (aggregation bias), allowing us to appropriately useaggregate performance data and the predictions based onthis data, rather than requiring data on individual suppliers(Robinson, 1950). The number of suppliers was similar overthe five good categories, although firms tended to useslightly fewer suppliers for die maintenance and slightlymore for surface coating.

Technical expertiseThe variable for a firm’s technical expertise reflected theextent to which the firm possesses capabilities for produ-cing the focal good. These attributes result from a deepunderstanding of the technology related to the good andfrom experience in production. This variable used fouritems, some of which we adapted from prior work (Walkerand Weber, 1984). All items used seven-point scales. Tocreate the scale, we used a weighted average of the fouritems with the weights based upon a confirmatory factoranalysis. We then mean-differenced the value in theinteraction models (Aiken and West, 1991). The reliabilitycoefficient for the technical expertise measure was 0.86,based on the Cronbach’s alpha, well above recommendedbenchmarks (Nunnally, 1967).

Relational governance mechanismsThis firm-level variable includes 13 items that measuregoodwill, information sharing, and evaluation techniques.We included items that measured both historical andprojected commitment to the relationship. We includeddocumentary aspects of relational governance, such asquality certification, plus interpersonal mechanisms suchas communication channels. We chose items based onprior work (Noordewier et al., 1990; Dyer, 1997) and onfactors that emerged in preliminary interviews. Some itemsinvolve written agreements and formal evaluation, consis-tent with our definition of relational governance. Moreover,none correlate over r¼ 0.18 with our contract variable(see Table 2), suggesting independence and the potentialto confidently interpret the direct and indirect effects ofrelational and contractual governance. We used a weightedaverage of the items to create the variable and used mean-differences in the interaction models. Reliability was 0.71,based on Cronbach’s alpha. This is reasonable, consideringthat this figure underestimates other measures of reliabilitysince alpha assumes equal loadings for all items (Nunnally,1967; Bollen, 1989).

These items measure the extent to which a firm useda particular relational governance mechanism, ratherthan how well the firm used the mechanism, because itis not clear how one would create an objective measureof the quality of the mechanisms. Therefore, our approach


54

AUTHOR COPY

assumes that more extensive usage of relational governancemechanisms associates with higher quality relational mana-gement. Clearly, usage frequency and quality may vary,but intensity will correlate with quality so long as a firmhas been able to learn from its experience. In addition,measuring extent rather than quality of relational govern-ance helps avoid the risk of attribution bias, in which firmsthat are satisfied with supplier performance believe theyhave effective relational governance.

Contractual agreementsWe created a contracts variable relative to each good withan item that asked whether the buyer used a long-termwritten agreement for the purchased component (AppendixA). If the buyer chose either of the options involving a long-term contract, we concluded that the firm used contractsrather than purchase orders; as we noted earlier, contractsare distinct from purchase orders for individual transac-tions. We created a 0–1 binary variable to indicate contractuse, which was affirmative for 71 cases (after dropping 16cases that indicated the use of combinations of options).Our operationalization is similar to that of Carson et al.(2006), who also created a binary variable for contractsbased on responses to a survey item with five options. Wefound similar usage of contracts among the different goods,from 13% to 23%, with a mean of 19%; thus a substantialmajority of firms did not use contracts. While we did notobtain details on contract length or clauses, it is likely thatthese contracts are similar between firms as we focused on afew specific goods in a well-defined industry sector.

Control variablesWe included several control variables in both stages of theanalysis. We controlled for asset specificity for each goodthrough a variable composed of three items measuring theavailability of alternate suppliers; this could affect perfor-mance because buyers with few options may have littlerecourse if supplier performance deteriorates (Williamson,1985). We also controlled for volume, technological, andperformance uncertainty as these may affect the choice ofsourcing mode and performance. We controlled for firmand supplier scope economies, as well as for supplierexpertise and scale economies. For all of these input-levelcontrols, we used multi-item scales, many from prior work(e.g., Anderson and Weitz, 1986; Bensaou and Anderson,1999). Other controls included firm age (years, logged),volume of the good required (annual units, based on a five-point scale), whether the focal firm was unionized, whetherit produced powder metal parts, and dummy variables forcomponent type.

Some controls only applied to either the first or secondstage of analysis. Our instrumental variables, used onlyin the first stage, were the similarity of goods sourced,firm economies of scope, and supplier economies of scope,which have little correlation with the performance measu-res. In the second stage, we controlled for concurrentsourcing, meaning that the firm produced the goodinternally and purchased it from outside suppliers. Thissourcing mode provides a credible threat of vertical inte-gration as well as more intimate knowledge of the procuredgoods (Harrigan, 1986, Parmigiani, 2007). Because we do

not have dyadic data, we could not control for relationshipduration or prior exchanges between a particular buyerand supplier. Given the maturity of this industry and theplayers, most of the relationships are stable and fairly long-lived. We also could not control for the size of the buyer’slegal staff, but these small firms typically do not employinternal legal counsel.

Table 1 reports descriptive statistics and Table 2 providescorrelations between the items and the main constructs.The presence of only moderate correlation among fiveperformance measures suggests that they arise from diff-erent processes. The relatively strong correlations betweenthe constructs and their related items and the relativelyweak correlations between the items and the alternateconstructs suggest matching between the measures andconstructs.

MethodsWe used a two-stage technique in which we first modeledsourcing mode choice, all internal production (make) orpurchase some or all of their requirements (buy orconcurrently source). We used output from this stage asan input to the second stage analyses, in which we estima-ted influences on each of the five types of performance.The two-stage approach has become increasingly commonfor similarly structured data to control for endogeneity(Leiblein et al., 2002; Mayer and Nickerson, 2005; Morrowet al., 2007). In this approach, one uses a common set ofindependent and control variables in both stages, but usesone or more unique variables in the first stage. Ideally,these unique variables, called instrumental variables, arenot highly correlated with the ultimate dependent variable(performance) and have been shown by prior work to affectthe first stage dependent variable (sourcing mode choice).

We relied both on our theoretical arguments as well asprior work to determine our common independent vari-ables and first stage instruments. We included variables forexpertise, relational governance, and contracting in bothstages, as these were our key theoretical constructs. We alsoincluded control variables for asset specificity, supplierexpertise, volume uncertainty, technological uncertainty,and performance uncertainty in both stages as prior worksuggest that these may affect mode choice (Williamson,1985) and subsequent performance (Poppo and Zenger,2002). Other control variables included in both stagesaddressed firm characteristics (age, size, unionization, andsector) and product attributes (scale economies, volumerequirements, and dummy variables for each type of input)which have the potential to affect both mode choice andsupplier performance.

We selected three variables as instruments for the firststage to predict mode choice: input similarity, firmeconomies of scope, and supplier economies of scope.Firms may be more likely to internally produce homo-genous products or those related to other goods theyproduce to enjoy improved coordination or knowledgespillovers (Richardson, 1972), but the effects of these onsupplier performance are ambiguous. In particular, thesemay have varying effects by type of performance, perhapsimproving communication but harming quality or delivery.Prior work using this same data set indicated that these


55

AUTHOR COPY

Tab

le1

Descriptive

sta

tistics

and

corr

ela

tions

Mn

SDR

an

ge1

23

45

67

89

1011

12

1C

oo

per

atio

n5.

461.

051–

71

2Q

ual

ity

5.28

1.08

1–7

0.54

13

Pri

ce4.

991.

021–

70.

430.

461

4D

eliv

ery

5.14

1.24

1–7

0.55

0.63

0.46

15

Co

mm

un

icat

ion

5.23

1.06

2–7

0.60

0.53

0.37

0.63

16

Fir

mte

chn

ical

exp

erti

se0.

001.

80�

3.1�

2.9

0.13

0.13

0.06

0.04

0.10

17

Rel

atio

nal

gove

rnan

ce0.

002.

42�

7.6�

5.9

0.09

0.00

�0.

020.

040.

07�

0.03

18

Co

ntr

acti

ng

0.19

0.39

0–1

0.05

0.00

�0.

07�

0.01

0.01

�0.

100.

121

9A

sset

spec

ific

ity

4.06

1.29

1–7

�0.

060.

010.

01�

0.04

0.00

�0.

050.

04�

0.06

110

Vo

lum

eu

nce

rtai

nty

3.81

1.43

1–7

�0.

060.

030.

05�

0.03

�0.

020.

01�

0.22

�0.

27�

0.02

111

Tec

hn

olo

gica

lu

nce

rtai

nty

4.53

1.20

1–7

�0.

04�

0.01

�0.

040.

000.

000.

030.

11�

0.11

�0.

010.

071

12P

erfo

rman

ceu

nce

rtai

nty

2.92

0.90

1–6

�0.

02�

0.01

0.06

�0.

03�

0.08

�0.

05�

0.22

�0.

140.

030.

29�

0.03

113

Fir

msc

op

eec

on

om

ies

3.73

1.49

1–7

�0.

04�

0.05

�0.

08�

0.06

�0.

060.

500.

010.

000.

000.

020.

01�

0.01

14Su

pp

lier

sco

pe

eco

no

mie

s5.

041.

271–

70.

060.

000.

030.

080.

04�

0.27

0.01

0.09

�0.

06�

0.02

0.00

0.05

15Su

pp

lier

Exp

erti

se4.

031.

001.

4–7

�0.

12�

0.10

�0.

02�

0.01

0.00

�0.

49�

0.04

0.11

0.06

�0.

07�

0.01

0.07

16F

irm

age

(Lo

gged

)3.

580.

630.

7–4.

80.

080.

060.

080.

080.

070.

100.

06�

0.02

�0.

030.

010.

07�

0.03

17F

irm

size

(Em

plo

yees

)2.

461.

071–

50.

06�

0.03

0.04

�0.

02�

0.02

0.04

0.23

0.18

0.09

�0.

080.

09�

0.11

18U

nio

n0.

150.

360–

1�

0.03

�0.

08�

0.04

�0.

02�

0.01

0.05

�0.

09�

0.04

�0.

130.

120.

050.

0719

Po

wd

erm

etal

0.23

0.42

0–1

�0.

07�

0.04

�0.

03�

0.07

�0.

090.

030.

040.

020.

11�

0.04

0.00

�0.

0220

Scal

eec

on

om

ies

3.51

1.64

1–7

�0.

09�

0.06

�0.

03�

0.07

�0.

04�

0.01

0.04

�0.

09�

0.02

0.03

�0.

030.

0321

Vo

lum

ere

qu

ired

2.63

1.48

1–5

�0.

07�

0.04

0.00

�0.

02�

0.06

�0.

06�

0.03

0.15

0.00

�0.

17�

0.03

�0.

0122

Inp

ut

sim

ilar

ity

3.51

1.79

1–7

�0.

03�

0.06

�0.

01�

0.03

�0.

040.

03�

0.11

0.13

�0.

16�

0.19

�0.

190.

0523

Die

des

ign

0.20

0.40

0–1

0.05

0.01

�0.

04�

0.04

0.00

0.25

�0.

04�

0.01

0.00

�0.

020.

03�

0.02

24D

iem

ain

ten

ance

0.25

0.44

0–1

0.06

0.04

0.00

�0.

010.

020.

120.

010.

030.

010.

01�

0.01

�0.

0125

En

d-p

art

mac

hin

ing

0.10

0.30

0–1

0.03

0.04

0.03

0.01

0.01

0.23

�0.

02�

0.06

�0.

050.

020.

01�

0.06

26E

nd

-par

tco

atin

g0.

140.

340–

1�

0.15

�0.

100.

030.

07�

0.03

�0.

630.

040.

050.

07�

0.02

�0.

050.

0727

Co

ncu

rren

tso

urc

e0.

440.

500–

10.

040.

110.

00�

0.02

0.05

0.69

0.02

�0.

110.

050.

030.

03�

0.05

28In

vers

em

ills

rati

o�

0.35

0.37

�1.

7–0

�0.

13�

0.07

�0.

12�

0.10

�0.

10�

0.13

0.01

0.02

0.02

0.00

0.02

0.04


56

AUTHOR COPY

Tab

le1

Continued

1314

1516

1718

1920

2122

2324

2526

2728

1 2 3 4 5 6 7 8 910 11 12 13

114

�0.

141

15�

0.33

0.36

116

�0.

020.

04�

0.05

117

0.04

0.01

�0.

070.

071

180.

060.

030.

010.

160.

041

190.

140.

00�

0.08

�0.

320.

17�

0.11

120

0.06

0.09

0.07

0.00

�0.

100.

000.

051

210.

030.

090.

080.

050.

25�

0.05

0.04

�0.

011

22�

0.01

0.00

�0.

05�

0.17

�0.

07�

0.06

0.00

0.07

0.05

123

0.03

�0.

16�

0.13

0.04

0.03

�0.

010.

08�

0.15

�0.

03�

0.06

124

0.17

�0.

11�

0.17

0.08

�0.

030.

06�

0.11

�0.

06�

0.06

�0.

02�

0.17

125

0.13

�0.

05�

0.17

�0.

070.

00�

0.07

�0.

010.

17�

0.03

0.02

�0.

19�

0.13

126

�0.

270.

200.

41�

0.05

�0.

03�

0.04

0.01

0.14

0.08

0.00

�0.

33�

0.23

�0.

271

270.

41�

0.32

�0.

440.

020.

070.

040.

05�

0.04

�0.

120.

030.

180.

220.

23�

0.49

128

�0.

080.

050.

08�

0.02

0.01

�0.

030.

00�

0.01

0.04

�0.

02�

0.02

�0.

06�

0.04

0.07

0.48

1


57

AUTHOR COPY

Tab

le2

Descriptive

sta

tistics

and

corr

ela

tions

–te

chnic

alexpert

ise,

rela

tional

govern

ance,

and

contr

acting

item

sand

constr

ucts

Mn

SD1

23

45

67

89

1011

1213

1415

1617

1819

20

1E

ngi

nee

rin

gst

aff

can

do

3.91

2.06

1

2W

eu

nd

erst

and

dee

pex

per

tise

4.84

1.84

0.61

1

3W

eh

ave

do

ne

for

year

s4.

192.

410.

710.

541

4Sk

ills

rela

ted

–o

ther

pro

du

cts

3.54

2.00

0.63

0.49

0.63

1

5C

on

stru

ct:

tech

nic

alex

per

tise

0.00

1.80

0.90

0.73

0.90

0.80

1

6R

elat

ion

ship

sla

stfo

rye

ars

6.02

0.84�

0.00�

0.04

0.02�

0.02�

0.00

1

7W

ork

thro

ugh

dif

ficu

ltie

s5.

381.

05�

0.04

0.01

0.00�

0.01�

0.01

0.31

1

8U

seco

nfi

den

tial

ity

4.01

1.97

0.02

0.00

0.00

0.03

0.02�

0.02

0.09

1

9U

sefo

rmal

,w

ritt

enco

ntr

acts

5.08

1.83

0.06

0.04

0.04

0.04

0.05

0.14

0.11

0.19

1

10Im

med

iate

info

rm–

volu

me

5.32

1.45

0.00�

0.03

0.01�

0.03�

0.01

0.19

0.15

0.13

0.33

1

11C

om

mu

nic

ate

dai

ly5.

451.

530.

00�

0.02�

0.01�

0.05�

0.02

0.09

0.07

0.22

0.07

0.32

1

12T

ell

per

form

ance

vso

ther

s4.

691.

68�

0.00

0.02

0.04�

0.04

0.01

0.02

0.11

0.23

0.17

0.02

0.24

1

13E

valu

ate

inte

rnal

sam

eas

exte

rnal

5.20

1.66�

0.08�

0.10�

0.06�

0.12�

0.10

0.23

0.16

0.10

0.29

0.16

0.18

0.26

1

14U

sefo

rmal

sco

reca

rd4.

751.

98�

0.02

0.02

0.00

0.02

0.00

0.02

0.11

0.20

0.21

0.15

0.12

0.49

0.17

1

15Q

ual

ity

cert

ific

ate

hel

d1.

971.

040.

020.

020.

01�

0.03

0.01�

0.11�

0.05

0.09

0.20

0.06

0.26

0.28

0.15

0.28

1

16E

ngi

nee

rstr

avel

tosu

pp

lier

s5.

611.

60�

0.06�

0.06�

0.01�

0.09�

0.06

0.01

0.01

0.13

0.11

0.11

0.15

0.13

0.11

0.02

0.13

1

17H

elp

sup

pli

ers

imp

rove

4.90

1.45

0.00�

0.04�

0.01�

0.05�

0.02

0.09

0.13

0.15

0.28

0.04

0.25

0.26

0.19

0.03

0.13

0.30

1

18Su

pp

lier

sh

elp

us

red

uce

cost

s4.

801.

57�

0.02�

0.07�

0.01�

0.05�

0.03

0.13

0.01

0.11

0.09

0.04

0.00

0.10

0.14

0.17

0.03

0.34

0.27

1

19C

on

stru

ct:

rela

tio

nal

gove

rnan

ce0.

002.

42�

0.02�

0.04

0.01�

0.06�

0.03

0.17

0.21

0.48

0.50

0.35

0.45

0.63

0.46

0.59

0.37

0.49

0.54

0.48

1

20C

on

trac

t(0

/1)

0.19

0.39�

0.09�

0.09�

0.09�

0.06�

0.10

0.03

0.00

0.06

0.16

0.12

0.18�

0.02

0.10

0.03

0.12

0.04

0.06�

0.03

0.12

1

Po0.0

5fo

rr4

0.0

8;

All

item

soth

er

than

‘Contr

acts

’are

based

on

7-p

oin

tscale

s(1¼

low

,7¼

hig

h);

n¼

508.


58

AUTHOR COPY

three variables did significantly affect the sourcing decision(Parmigiani, 2007). In turn, as indicated in Table 1, thecorrelations between these variables and supplier perfor-mance are small, suggesting that they would be appropriateinstruments. The prior connection with mode choice andlow correlations with performance suggest that these threevariables are suitable instruments.

Our approach involved a probit model in the first stagewith OLS regression models for the second stage. In bothstages, we used robust standard errors adjusted for repeatobservations by firm to account for within-firm non-independence (Mizruchi and Stearns, 2001). A probit modelis required in the first stage to create the inverse Mills ratiofor the second stage (Kennedy, 2008). The second stageincluded this ratio and a dummy variable for concurrentsourcing, while omitting the three first stage instruments.After culling cases that did not include the contractualagreement variable, our data resulted in 373 purchases from164 unique firms.

Because it is likely that the within-firm observations willnot be independent, which is common in survey research,we used a clustering adjustment that increases the standarderrors but does not affect the coefficient estimates. Thisapproach is more appropriate than a firm fixed-effectsmodel because we have a small number of observations perfirm relative to the number of firms (only one observationper firm in 37 cases). Had we used a fixed effects model,the structure of our data would have resulted in problemsof insufficient variance, perfect prediction, and the inabi-lity to estimate the firm coefficients, thereby potentiallybiasing the coefficients of the primary explanatory variables(Greene, 1997; Kennedy, 2008). By using the clusteringapproach, along with a control variable for firm age, we canmore clearly identify the effect of our key firm-levelvariable of interest, relational governance.

We considered several approaches to modeling theselection stage of the analysis. The most efficient currenteconometric approach uses one-step algorithms to generatefirst and second stage analyses that include common inde-pendent variables, controls, and first stage instruments.Unfortunately, however, the current one-step algorithmsdo not handle firm-specific clustering of standard errorseffectively. As we discuss above, the ability to include clus-tering when assessing supplier performance is important inour study because the data include up to five cases ofsourcing per firm, so that we need to control unmeasured,firm-specific attributes that might affect which firms out-source. Therefore, we concluded that a two step approachthat first predicts whether firms will use outside suppliersand then incorporates a selection variable into each ofthe different performance equations is more appropriate forthe study.

Likewise, our data and conceptual approach preventedus from using structural equation modeling, a populartechnique for analyzing survey data. Structural equationsare difficult to implement for data that incorporatesmultiple levels (e.g., the firm and the good) and interactionterms. The technique also is best suited for dependentvariables measured by multiple item scales, rather than bysingle items. Moreover, our sample size was also relativelysmall for this technique, as five cases per parameter esti-mate are recommended (Bentler and Chou, 1987).

We chose OLS regression rather than ordered logit orprobit regression for the second stage, for both conceptualand empirical reasons. Conceptually, although we createdseven categories (e.g., 1¼ terrible to 7¼ terrific) for ourrespondents, it is possible that some respondents may havebeen more strict than others such that a ‘2’ from some mayhave been akin to a ‘4’ from others. Given this possibility,we can assume a continuous relationship along the perfor-mance dimension but cannot have complete confidence inthe categories. Empirically, we did not have an even distri-bution among all seven reply options, which may bias theestimation algorithm as it relies on ‘cuts’ for each category.Perhaps not surprisingly, less than 10% of observationswere in the lowest three categories, most likely becausebuyers would switch suppliers who performed poorly.Given this distribution, we did not believe that an orderedestimation method was appropriate. Thus, our approach ofa first stage probit for sourcing mode choice and a secondstage OLS model for performance best suited our data andbest fit our conceptual framework. Nonetheless, we used thealternative techniques in sensitivity analyses that we reportin the results section.

ResultsWe first conducted a probit analysis to determine whichfirms would produce internally vs use outside suppliers.This stage controlled for potential endogeneity of supplierperformance based upon attributes that made the firmmore likely to outsource, such as superior governanceskills. In addition, including the governance mechanismvariables (relational governance and contracts) helps con-trol for endogeneity in the use of these mechanisms byassessing whether firms choice to outsource associates withtheir governance choices. This analysis, which Appendix B1reports, indicated that greater technical expertise promotedinternal production, while relational governance andcontracting did not affect mode choice. Among the othervariables, internal production increased with performanceuncertainty, firm age, and firm scope economies, whiledecreasing with technological uncertainty, scale economies,and supplier scope economies. We created an InverseMills Ratio variable based on predictions from this firstmodel and used this term in the second-stage performancemodels (Shaver, 1998; Leiblein et al., 2002; Hamilton andNickerson, 2003; Morrow et al., 2007).

The models in Table 3 tested hypotheses 1, 2, and 3 byassessing how the firm’s technical expertise and use ofgovernance mechanisms influenced the five supplier perfor-mance outcomes. The results are consistent with hypothesis1 on all five performance dimensions and hypothesis 2 onthree dimensions. Consistent with hypothesis 1, technicalexpertise had a positive and strongly significant impact oncooperation, quality, price, delivery, and communication (allat Po0.01). Consistent with hypothesis 2, use of relationalgovernance mechanisms had at least a moderately significantpositive effect on cooperation (Po0.05), delivery (Po0.10),and communication (Po0.05). Relational governance hadno significant impact on quality or price.

Contrary to hypothesis 3, use of contracts had nosignificant effect on any of the performance variables.The lack of significance for contracting may arise because


59

AUTHOR COPY

firms in this industry tend to use a small number ofsuppliers and rarely use contracts, possibly because indus-try maturity may promote reputational bonds over writtenagreements. In addition, as we noted earlier, if firms areable to make appropriate choices of when and when notto use contracts, depending on the nature of a given setof transactions with a particular supplier, then there will belittle or no impact on performance.

Table 4 then presents models that test the joint effects ofexpertise and governance, finding targeted support for

Hypotheses 4a and 4c. The results in Table 4 for the maineffects of technical expertise, relational governance, andcontracting are similar to those in Table 3. Table 4continues to support H1 for all five performance dimen-sions (Po0.01). The results also continue to support H2for cooperation and communication, although at reducedsignificance (Po0.10), while the positive impact on deli-very performance lost its moderate significance. Thereduction in significance for the relational governanceeffects occurs because the interactions with technical

Table 3 The impact of technical expertise and governance mechanisms on supplier performance, main effects (OLS with robust standard errors adjusted for repeatobservations; positive coefficient indicates superior performance)

1. Cooperation 2. Quality 3. Price 4. Delivery 5. Communication

H1 (+): Firm technical expertise 0.19*** 0.21*** 0.17*** 0.29*** 0.22***(0.06) (0.06) (0.06) (0.07) (0.07)

H2 (+): Relational governance 0.04** 0.00 �0.01 0.04* 0.05**(0.02) (0.03) (0.03) (0.03) (0.02)

H3 (+): Contracting 0.09 0.05 �0.17 �0.04 �0.01(0.15) (0.15) (0.14) (0.17) (0.16)

Asset specificity �0.01 0.01 0.03 0.00 0.02(0.04) (0.04) (0.04) (0.05) (0.04)

Volume uncertainty �0.03 0.02 0.02 0.01 0.01(0.04) (0.05) (0.05) (0.05) (0.05)

Technological uncertainty �0.05 0.00 �0.03 �0.01 �0.03(0.05) (0.05) (0.05) (0.06) (0.05)

Performance uncertainty 0.05 0.03 0.05 �0.02 �0.09(0.06) (0.07) (0.07) (0.08) (0.07)

Supplier expertise �0.10 �0.09 �0.08 �0.09 0.10(0.09) (0.08) (0.08) (0.10) (0.09)

Firm age 0.11 0.10 0.12 0.19* 0.08(0.07) (0.09) (0.10) (0.11) (0.10)

Firm size (Employees) 0.03 �0.06 0.04 �0.08 �0.03(0.06) (0.06) (0.06) (0.06) (0.06)

Union �0.06 �0.37** �0.15 �0.09 �0.11(0.16) (0.18) (0.18) (0.20) (0.16)

Powder metal �0.18 �0.05 �0.06 �0.15 �0.22(0.14) (0.16) (0.14) (0.19) (0.16)

Scale economies �0.02 �0.04 �0.02 �0.08 �0.04(0.04) (0.04) (0.04) (0.05) (0.04)

Volume required �0.05 �0.01 0.00 0.02 �0.02(0.04) (0.04) (0.04) (0.05) (0.04)

Concurrent source �0.42*** �0.07 �0.32** �0.23 �0.21(0.14) (0.16) (0.15) (0.19) (0.15)

Component: die design vs die building �0.03 �0.17 �0.18 �0.04 �0.02(0.13) (0.15) (0.14) (0.18) (0.13)

Component: Die maintenance vs die building 0.24 0.33* 0.24 0.32 0.19(0.19) (0.18) (0.22) (0.24) (0.21)

Component: End-part machining vs die building 0.06 �0.03 0.12 0.20 0.04(0.18) (0.18) (0.16) (0.20) (0.18)

Component: End-part coating vs die building �0.17 �0.05 0.26 0.60*** 0.15(0.17) (0.17) (0.18) (0.18) (0.19)

Inverse mills ratioa 0.43 0.73** 0.26 0.80** 0.35(0.29) (0.34) (0.30) (0.39) (0.32)

Number of cases 371 372 373 370 372R-squared 0.10 0.07 0.05 0.08 0.08

*** Po0.01, ** Po0.05; * Po0.10 (one-tailed tests for hypotheses; two-tailed tests for controls; robust standard errors in parentheses; 164unique firms, constant estimates omitted).aStage 1 of the model is a probit analysis that controls for the decision to out-source (Appendix B1).


60

AUTHOR COPY

expertise and contracting pick up part of the effect, whichwe discuss next.

Table 4 reports focused support for two of hypothesesconcerning joint effects (H4a and H4c).1 Greater combina-tions of technical expertise and relational governance (H4a)

contribute to cooperation performance (Po0.05), but haveno significant impact on the other four performancedimensions. Combinations of technical expertise and con-tracting (H4b) had no significant impact on performance.Combinations of relational governance and contracting

Table 4 The impact of technical expertise and governance mechanisms on supplier performance, with interactions (OLS with robust standard errors adjusted for repeatobservations; positive coefficient indicates superior performance)



H2 (+): Relational governance 0.04* 0.00 �0.02 0.03 0.04*(0.02) (0.03) (0.03) (0.03) (0.03)

H3 (+): Contracting 0.11 0.03 �0.25** �0.10 �0.02(0.14) (0.15) (0.15) (0.19) (0.16)

H4a (+): Technical expertise � relational governance 0.03** 0.01 0.00 0.01 0.01(0.01) (0.01) (0.01) (0.02) (0.01)

H4b (+): Technical expertise � contracting 0.08 �0.05 �0.04 �0.01 0.07(0.08) (0.08) (0.06) (0.10) (0.07)

H4c (+): Relational governance � contracting 0.04 0.02 0.10** 0.10* 0.05(0.06) (0.06) (0.06) (0.08) (0.07)






Firm age 0.11 0.10 0.13 0.20* 0.09(0.07) (0.09) (0.10) (0.12) (0.10)

Firm size (Employees) 0.02 �0.06 0.04 �0.09 �0.04(0.06) (0.07) (0.06) (0.06) (0.06)

Union �0.07 �0.36** �0.12 �0.06 �0.10(0.16) (0.18) (0.18) (0.20) (0.16)

Powder metal �0.19 �0.06 �0.08 �0.16 �0.22(0.14) (0.16) (0.14) (0.19) (0.16)

Scale economies �0.02 �0.04 �0.02 �0.08* �0.04(0.04) (0.04) (0.04) (0.05) (0.04)

Volume required �0.05 �0.01 0.00 0.02 �0.02(0.04) (0.04) (0.04) (0.05) (0.04)

Concurrent source �0.42*** �0.07 �0.32** �0.23 �0.21(0.28) (0.16) (0.15) (0.18) (0.15)

Component: Die design vs die building �0.03 �0.16 �0.18 �0.04 �0.03(0.14) (0.15) (0.14) (0.18) (0.13)

Component: Die maintenance vs die building 0.22 0.33* 0.25 0.33 0.19(0.19) (0.18) (0.23) (0.23) (0.21)


Component: End-part coating vs die building �0.16 �0.04 0.26 0.60*** 0.15(0.17) (0.17) (0.18) (0.18) (0.19)

Inverse mills ratioa 0.41 0.74** 0.27 0.79** 0.32(0.28) (0.34) (0.30) (0.39) (0.32)


*** Po0.01, ** Po0.05; * Po0.10 (one-tailed tests for hypotheses; two-tailed tests for controls; robust standard errors in parentheses; 164unique firms, constant estimates omitted).aStage 1 of the model is a probit analysis that controls for the decision to out-source (Appendix B1).


61

AUTHOR COPY

(H4c) had at least moderately significant impact on price(Po0.05) and delivery (Po0.10). Intriguingly, the maineffect of contracts on price performance became signifi-cantly negative (Po0.05), now that the price model addedthe interaction of relational governance and contracting.

The positive joint impact of contracts and relationalgovernance on price satisfaction in model 3 of Table 4 isintriguing, especially when coupled with the negativeimpact of contracts and the lack of impact of relationalgovernance alone. The presence of the joint effect of rela-tional governance and contracts suggests that firms weremost satisfied with contracted pricing performance whenthey were also practiced at identifying desirable suppliersand managing the flow of information between the twofirms. This result may also reflect a zero-sum game natureof price negotiation, with trust and goodwill amelioratingpotential negative outcomes and improving the perceptionof fairness. Thus, neither contractual nor relational gover-nance alone is sufficient to achieve satisfactory prices – thefirm benefits by having both. Indeed, firms that rely oncontracts while lacking relational governance ability maysuffer weaker pricing satisfaction.

Several control variables influence performance. Union-ized buyers tended to be less satisfied about performance,although the effect was significant only for quality, possiblybecause union contracts limit flexibility in production.Concurrent sourcing reduced cooperation and price perfor-mance, perhaps because the combination of internalproduction and external sourcing made coordination moredifficult, made suppliers more suspicious of buyers, and/orgave buyers more accurate cost data. Delivery performancetended to be superior for end product coating activities,possibly because firms can monitor delivery performanceclosely in this final stage operation and replace under-performing suppliers quickly. The inverse Mills Ratio (theoutput of the first stage model) is significant for quality anddelivery performance, suggesting that some the factors thatinfluence the choice to outsource also affect performance.

Sensitivity analyses examined several other influences.We combined the five individual performance items into anaggregate measure and found support for H1, but not forthe other predictions. This is to be expected becauseaggregating hides nuanced relationships, highlighting theimportance of measuring dependent variables at the correctlevel of specificity (Ray et al., 2004). Second, we foundsimilar results when we added the number of suppliers toour models. This variable was never significant and did notaffect the reported hypothesis tests. Third, we consideredthe possibility that the level of the respondent might affecttheir perception of supplier performance; because we hadthe titles of all respondents, we could determine which werehigher-level executives (e.g., President, Vice President,other officers and directors) and included a binary variablefor this characteristic. The respondent level variable wasnot significant and our results were equivalent when weincluded the variable. Finally, we omitted the ‘formal con-tract’ item from the relational governance measure andreran the models, in case this item conflicted with ourmeasure of contracting, finding similar results. We choseto keep this item in the scale because reliability was betterand the item is consistent with the conceptual definition ofrelational governance.

We assessed several alternative specifications for data,and estimation method. We estimated models that omittedthe contracting variable (n¼ 508; the contracting variablehad a substantial number of missing cases), finding similarresults. We ran structural equation models for hypotheses1, 2, and 3, again with similar results. We also tried usingthe ‘Heckman’ command, which runs the two stagessimultaneously. We obtained the same results for H1 forfour of the five dependent variables and the same results forH3 but many of the runs had convergence problems, likelydue to a small number of observations for certain sourcingmodes for some types of goods (e.g., few cases of firmsinternally conducting surface coating). We also estimatedsingle stage regression, ordered logit, and ordered probitmodels, finding similar results for all hypotheses.

Discussion and conclusionWe started by asking whether inter-firm governancemechanisms, including both contractual and relationalmechanisms, can substitute for a firm’s internal technicalskills in maintaining supplier performance or whethera firm risks hollowing itself out by de-emphasizing inter-nal expertise when it outsources. We considered how tech-nical expertise, relational governance, and contractualgovernance affected five elements of supplier performance(cooperation, quality, price, delivery, and communication)independently and jointly, while controlling for potentialendogeneity of the choice to outsource. This work is themost systematic study to date of how systems of exchangeskills involving technical expertise and the two types ofgovernance mechanisms affect supplier performance.

Our results suggest that relational governance oftencontributes to supplier performance but, nonetheless, thathollowing out will result in poor supplier performance.A buyer’s technical expertise significantly improves all fivedimensions of supplier performance. In addition, relationalgovernance has a significant impact on a more focused setof performance dimensions, including both cooperationand communication. At the same time, we found no evi-dence that contractual agreements directly affected perfor-mance, possibly because our firms are relatively small andtypically can choose when contracts make sense and whenthey do not.

The interactions of technical expertise and the two gov-ernance mechanisms have more focused impact. Strongercombinations of technical expertise and relational govern-ance improve supplier cooperation. Greater combinationsof relational governance and contracting, meanwhile,improve price performance and also provide moderatelysignificant benefits for delivery performance. Notably, aswe discussed above, relying on contracts without applyingcomplementary relational skills has a negative impact onsatisfaction with prices.

Thus, in several nuanced ways, technical expertise andsupplier governance mechanisms have both direct andcomplementary effects on supplier performance. Overall,a firm’s technical expertise was the most general driver ofsupplier performance. But both relational and contractualelements of supplier governance mechanisms also influ-enced supplier performance, most strikingly in the abilityto engender cooperative relationships and, in combination,


62

AUTHOR COPY

on price performance. Hence, firms can neither successfullyneglect technical expertise nor ignore governance mechan-isms, but need both types of skills.

The study has several conceptual implications for thegovernance literature. First, this work highlights the needfor firms to maintain their technical skills even as theyoutsource to maintain adequate performance across multipleelements of activities. This finding highlights Richardson’sconceptual discussion of technical skills being developedboth through internal experience and through cooperationwith suppliers and other firms (Richardson, 1972). It alsoreinforces conclusions from a small set of empirical studiesabout the importance of internal technical skills in main-taining effective supply relationships (e.g., MacDuffie andHelper, 1997; Brusoni, et al., 2001). Perhaps it is thesetechnical skills that offset the limits of effective relationalgovernance in long-term relationships (Poppo et al., 2008).Our results advance this literature by demonstrating animpact across a wider range of performance dimensions thanprior research and by controlling for the initial decision tooutsource, which prior studies do not address. Moreover, thefact that technical skills reinforce the benefits of relationalgovernance for cooperation performance extends the tech-nical skills arguments even further, as prior research has notsystematically examined such interactions.

Second, this work speaks to relational governance andcontractual governance arguments. Relational governancemechanisms are an aspect of what Dyer and Singh (1998)refer to as a relational capability. The study increases ourunderstanding of this relational capability and how itaffects supplier performance, both alone and in conjunctionwith technical expertise. Relational governance influencedcooperation, delivery, and communication more directlythan price or quality. It is possible that price and quality aremore objective criteria that are less directly affected byrelational mechanisms.

In turn, the results have nuanced implications for thecontractual governance literature. The lack of a main effectof contracting may appear to run counter to arguments byscholars such as Reuer and Arino (2007) and Mayer andTeece (2008), while the lack of an interaction with technicalexpertise may appear to conflict with results from Mayerand Salomon (2006). The null results may arise because therelatively small firms in our study can select when touse contracts fairly effectively, as we noted above. Indeed,the fact that the item for appropriate use of contractsloads strongly within the relational governance constructspeaks to this conclusion. Instead, performance differencesarise from the harder to manage activities that underlietechnical expertise and relational governance. At thesame time, the bimodal impact of contracting on priceperformance that we discussed earlier – benefits that arisewhen firms combine contracting with relational governanceas opposed to problems when firms attempt to rely oncontracts alone – helps tease out the conflicting argumentsabout contracts, in which some scholars, such as thosewe cited above, predict benefits while others (e.g., Ghoshaland Moran, 1996) expect problems. A key point from ourstudy is that the benefits of contracts are most likelyto occur in combination with relational governance, whileusing contracts alone may sometimes actually createconstraints that inhibit performance.

The industry context produces insights. Our firms weresmall, with less economic power than those often discussedin outsourcing or systems integration studies. It is possiblethat technical expertise and relational governance areparticularly critical to supplier management for such firms,because they cannot rely on bargaining power to generatepositive results.

The survey has several limits that suggest avenues forcontinuing research. The data are cross-sectional with onlythe buyer as the respondent. Because sourcing is a dynamicprocess, consisting of many stages, and because relation-ships develop over time, it would be interesting todetermine whether the results also arise longitudinally.It would also be informative to gain the perspective of thesupplier to determine whether they view the buyer as beingtechnically or managerially skilled. We were somewhatconstrained by having subjective performance measures.An alternate explanation to our results would be thatsatisfactory supplier performance involves appropriateexpectations, such that buyers with closer supply relation-ships will know more about their suppliers and thus be lesslikely to be dissatisfied with the suppliers’ performance.However, our strong and more general results for technicalexpertise over relational governance as well as our twostage approach suggest that the results do not reflect self-fulfilling prophecies, but rather that technically compe-tent firms appear to be better at selecting and managingsuppliers. Nonetheless, it would be insightful to understandhow both parties develop performance expectations andhow expectations relate to more objective measures ofperformance. Although price, quality, delivery, communi-cation, and cooperation are important to all buyers, itwould be interesting to explore the relationships betweentechnical expertise and governance for other types ofperformance, such as innovativeness.

Several extensions to this work would broaden ourunderstanding of supply relationships. Research couldinvestigate the effect of relationship duration and priorexchanges on performance when also incorporating tech-nical and governance skills. Fine-grained empirical work onthe details of the contracts (Mayer and Weber, 2005; Ryalland Sampson, 2009), along with a better understanding ofthe specific relationship dynamics within a particular dyad,could help uncover what specific features of contractualand relational governance complement or conflict with eachother. Further theory development can consider how speci-fic aspects of relational governance affect different typesof performance, potentially building upon empirical workby Krause et al. (2007). We note that the choice of sourcingmode was significant for some types of performance(quality and delivery), but not others; this suggests thatthere could be different mechanisms at work, one requiringa more complex model (including relational contractualconsiderations) and another simply requiring technicalskills. It would be enlightening to use qualitative or otherfine-grained techniques of data collection to better under-stand these mechanisms and processes.

Research could also assess generalizability to othersettings, such as retail or service industries or sectors withlarger firms or more technological volatility. It wouldbe instructive to consider how firms manage multiplesuppliers, including internal production, to determine how


63

AUTHOR COPY

governance features may differ between suppliers. It wouldalso be fascinating to unpack the expertise construct,determining if different kinds of expertise (e.g., process vsproduct) have contingent effects on performance, howthese may relate to the choice of sourcing mode, and ifdiminishing returns exist.

As firms increasingly outsource key components, theyneed to understand how to manage supplier relationships.This study demonstrates the importance of both technicalexpertise and governance activities for supplier perfor-mance. In sum, firms must maintain a multidimensionalinternally and externally focused exchange system tosuccessfully manage supplier relationships.

AcknowledgementWe appreciate financial support from the Institute for SupplyManagement, the American Production and Inventory ControlSociety, and the University of Michigan Business School.

Note

1 Appendices C1–C3 report models that test each interaction termseparately. As above, these models support H4a for cooperationand H4c for price, while supporting H4b for cooperation.The combined model in Table 4 thus represents the resultsconservatively.

References

Aiken, L. S. and S. G. West, 1991, Multiple regression: Testing and interpreting

interactions. London: Sage Publications.

Alchian, A. A. and H. Demsetz, 1972, ‘‘Production, information costs, and

economic organization’’. American Economic Review, 62: 777–795.

Anderson, E. and B. A. Weitz, 1986, ‘‘Make or buy decisions: Vertical

integration and marketing productivity’’. Sloan Management Review,

28: 3–19.

Argyres, N. S. and K. J. Mayer, 2007, ‘‘Contract design as a firm capability:

An integration of the learning and transaction cost perspectives’’. Academy

of Management Review, 32: 1060–1077.

Armstrong, S. J. and T. S. Overton, 1977, ‘‘Estimating nonresponse bias in

mail surveys’’. Journal of Marketing Research, 14: 396–402.

Arrow, K. J., 1974, The limits of organization. New York: W. W. Norton &

Company.

Baker, G., R. Gibbons and K. J. Murphy, 2002, ‘‘Relational contracts and the

theory of the firm’’. Quarterly Journal of Economics, 117: 39–84.

Barney, J. B., 1991, ‘‘Firm resources and sustained competitive advantage’’.

Journal of Management, 17: 99–120.

Bensaou, M. and E. Anderson, 1999, ‘‘Buyer-supplier relations in industrial

markets: When do buyers risk making idiosyncratic investments?’’

Organization Science, 10: 460–481.

Bentler, P. M. and C. P. Chou, 1987, ‘‘Practical issues in structural modeling’’.

Sociological Methods and Research, 16: 78–117.

Bollen, K. A., 1989, Structural equations with latent variables. New York:

Wiley.

Bradach, J. L., 1997, ‘‘Using the plural form in the management of restaurant

chains’’. Administrative Science Quarterly, 42: 276–303.

Brusoni, S., A. Prencipe and K. Pavitt, 2001, ‘‘Knowledge specialization,

organizational coupling, and the boundaries of the firm: Why do firms

know more than they make?’’ Administrative Science Quarterly, 46:

597–621.

Business Week 1986, ‘‘Special report: The hollow corporation’’. 3 March: 56–85.

Campanelli, P., 1997, ‘‘Testing survey questions: New directions in

cognitive interviewing’’. Bulletin de Methodologie Sociologique, 55: 5–17.

Cannon, J. P., R. Achrol and G. T. Gundlach, 2000, ‘‘Contracts, norms,

and plural form governance’’. Journal of the Academy of Marketing

Science, 28: 180–194.

Carson, S. J., A. Madhok and T. Wu, 2006, ‘‘Uncertainty, opportunism,

and governance: The effects of volatility and ambiguity on formal and

relational contracting’’. Academy of Management Journal, 49: 1058–1077.

Chen, I. J., A. Paulraj and A. A. Lado, 2004, ‘‘Strategic purchasing, supply

management, and firm performance’’. Journal of Operations Management,

22: 505–523.

Cohen, W. M. and D. A. Levinthal, 1990, ‘‘Absorptive capacity: A new

perspective on learning and innovation’’. Administrative Science Quarterly,

35: 128–152.

Cousins, P. D. and B. Menguc, 2006, ‘‘The implications of socialization and

integration in supply chain management’’. Journal of Operations

Management, 24: 604–620.

Degraeve, Z. and F. Roodhooft, 2001, ‘‘A smarter way to buy’’. Harvard

Business Review, 79: 2–3.

Delios, A. and W. J. Henisz, 2000, ‘‘Japanese firms’ investment strategies

in emerging economies’’. Academy of Management Journal, 43: 305–323.

Dillman, D. A., 1978, Mail and telephone surveys: The total design method.

New York: Wiley.

Dillman, D. A., 2000, Mail and internet surveys. New York: Wiley.

Dyer, J. H., 1997, ‘‘Effective inter-firm collaboration: How firms minimize

transaction costs and maximize transaction value’’. Strategic Management

Journal, 17: 535–556.

Dyer, J. H., 2000, Collaborative advantage: Winning through extended

enterprise supplier networks. New York: Oxford University Press.

Dyer, J. H. and K. Nobeoka, 2000, ‘‘Creating and managing a high performance

knowledge-sharing network: The Toyota case’’. Strategic Management

Journal, 21: 345–367.

Dyer, J. H. and H. Singh, 1998, ‘‘The relational view: Cooperative strategy

and sources of inter organizational competitive advantage’’. Academy of

Management Review, 23: 660–679.

Elmaghraby, W. J., 2000, ‘‘Supply contract competition and sourcing policies’’.

Manufacturing and Service Operations Management, 2: 350–371.

Enright, M. J., 2003, Buyer-supplier relationships Case HKU257, University

of Hong Kong, Centre for Asian Business Cases.

Fowler, F. J., 1995, Improving survey questions: Design and evaluation.

Thousand Oaks, CA: Sage Publications.

Ghoshal, S. and P. Moran, 1996, ‘‘Bad for practice: A critique of transaction

cost theory’’. Academy of Management Review, 21: 13–47.

Grant, R. M., 1996, ‘‘Toward a knowledge-based theory of the firm’’.

Strategic Management Journal, 17: 109–122.

Greene, W. H., 1997, Econometric analysis. Upper Saddle River, NJ:

Prentice-Hall.

Hamilton, B. A. and J. A. Nickerson, 2003, ‘‘Correcting for endogeneity

in strategic management research’’. Strategic Organization, 1: 51–78.

Harrigan, K. R., 1986, ‘‘Matching vertical integration strategies to competitive

conditions’’. Strategic Management Journal, 7: 535–555.

Harrison, D., 2004, ‘‘Is a long term business relationship an implied contract?

Two views of relationship development’’. Journal of Management Studies,

41: 107–125.

Hayes, R. H., S. C. Wheelwright and K. B. Clark, 1988, Dynamic manufacturing.

New York: The Free Press.

Heide, J. B. and G. John, 1990, ‘‘Alliances in industrial purchasing: The

determinants of joint action in buyer-supplier relationships’’. Journal of

Marketing Research, 27: 24–36.

Heide, J. B. and A. S. Miner, 1992, ‘‘The shadow of the future: Effects of

anticipated interaction and frequency of contact on buyer-seller

cooperation’’. Academy of Management Journal, 35: 285–291.

Helper, S., 1991, ‘‘Strategy and irreversibility in supplier relations: The case of

the US automobile industry’’. Business History Review, 65(4): 781–824.

Helper, S. and D. I. Levine, 1992, ‘‘Long-term supplier relations and

product-market structure’’. Journal of Law, Economics, and Organization,

8: 561–581.

Holcomb, T. R. and M. A. Hitt, 2007, ‘‘Toward a model of strategic

outsourcing’’. Journal of Operations Management, 25: 464–481.

Holweg, M. and F. K. Pil, 2008, ‘‘Theoretical perspectives on the

coordination of supply chains’’. Journal of Operations Management,

26: 389–406.

Hult, G., M. Tomas, D. J. Ketchen, Jr and M. Arrfelt, 2007, ‘‘Strategic

supply chain management: Improving performance through a culture of

competitiveness and knowledge development’’. Strategic ManagementJournal, 28: 1035–1052.


64

AUTHOR COPY

Kale, P., J. H. Dyer and H. Singh, 2002, ‘‘Alliance capability, stock market

response, and long-term alliance success: The role of the alliance function’’.


Kennedy, P., 2008, A guide to econometrics 6th edn. Malden, MA: Blackwell

Publishing.

Klein, B. and K. B. Leffler, 1981, ‘‘The role of market forces in assuring

contractual performance’’. Journal of Political Economy,

89(4): 615–641.

Krause, D. R., R. B. Handfield and T. V. Scannell, 1998, ‘‘An empirical

investigation of supplier development: reactive and strategic processes’’.

Journal of Operations Management, 17: 39–58.

Krause, D. R., R. B. Handfield and B. B. Tyler, 2007, ‘‘The relationships between

supplier development, commitment, social capital accumulation, and

performance improvement’’. Journal of Operations Management, 25:

528–545.

Kulwani, M. U. and N. Narayandas, 1995, ‘‘Long-term manufacturer-supplier

relationships: Do they pay off for supplier firms?’’ Journal of Marketing,

59: 1–16.

Lawson, B., B. B. Tyler and P. D. Cousins, 2008, ‘‘Antecedents and

consequences of social capital on buyer performance improvement’’. Journal

of Operations Management, 26: 446–460.

Leiblein , M. J., J. J. Reuer and F. Dalsace, 2002, ‘‘Do make or buy decisions

matter? The influence of organizational governance on technological

performance’’. Strategic Management Journal, 23: 817–833.

Li, H. and K. Atuahene-Gima, 2002, ‘‘The adoption of agency business activity,

product innovation, and performance in Chinese technology ventures’’.


Liker, J. K., 2004, The Toyota way. New York: McGraw-Hill.

Lincoln, J. R., C. L. Ahmadjian and E. Mason, 1998, ‘‘Organizational learning

and purchase-supply relations in Japan: Hitachi, Matsushita, and Toyota

compared’’. California Management Review, 40: 241–264.

Lindell, M. K. and D. J. Whitney, 2001, ‘‘Accounting for common method

variance in cross-sectional research designs’’. Journal of Applied Psychology,

86: 114–121.

Lyons, B., 1996, ‘‘Empirical relevance of efficient contract theory: Inter-firm

contracts’’. Oxford Review of Economic Policy, 12: 27–52.

Macaulay, S., 1963, ‘‘Non-contractual relations in business: A preliminary

study’’. American Sociological Review, 28: 55–67.

MacDuffie, J. P. and S. Helper, 1997, ‘‘Creating lean suppliers: Diffusing

lean production through the supply chain’’. California Management Review,

39: 118–151.

Macneil, I. R., 1978, ‘‘Contracts: Adjustment of long-term economic relations

under classical, neoclassical, and relational contract law’’. Northwestern

University Law Review, 72: 854–905.

Macneil, I. R., 1980, The new social contract. Binghamton, NY: The Vail-Ballou

Press.

Magretta, J., 1998, ‘‘The power of virtual integration: An interview with

Dell Computer’s Michael Dell’’. Harvard Business Review, 76: 73–84.

Mascarenhas, B., A. Baveja and M. Javil, 1998, ‘‘Dynamics of core competencies

in leading multinational companies’’. California Management Review,

40: 117–132.

Mayer, K. J. and J. A. Nickerson, 2005, ‘‘Antecedents and performance

implications of contracting for knowledge workers’’. Organization Science,

16: 225–242.

Mayer, K. J. and R. M. Salomon, 2006, ‘‘Capabilities, contractual hazards,

and governance: Integrating resource-based and transaction cost

perspectives’’. Academy of Management Journal, 49: 942–959.

Mayer, K. J. and D. J. Teece, 2008, ‘‘Unpacking strategic alliances: The

structure and purpose of alliance versus supplier relationships’’. Journal

of Economics, Behavior, and Organization, 66: 106–127.

Mayer, K. J. and L. Weber, 2005, Developing contracts that help develop

inter-firm relationships: The use of extendibility provisions in information

technology contracts Paper presented at the annual meeting of the

Academy of Management, August, Honolulu HI.

Mellewigt, T., A. Madhok and A. Weibel, 2005, Trust and formal contracts

in inter organizational relationships: Substitutes and complements Paper

presented at the annual meeting of the Academy of Management,

August, Honolulu HI.

Mizruchi, M. and L. B. Stearns, 2001, ‘‘Getting deals done: The use of

social networks in bank decision-making’’. American Sociological Review,

66: 647–671.

Morrow, Jr, J. L., D. G. Sirmon, M. A. Hitt and T. R. Holcomb, 2007,

‘‘Creating value in the face of declining performance: Firm strategies and

organizational recovery’’. Strategic Management Journal, 28: 271–283.

Mowery, D., J. E. Oxley and B. Silverman, 1996, ‘‘Strategic alliances

and interfirm knowledge transfer’’. Strategic Management Journal,

17: 77–91.

Nelson, R. R. and S. G. Winter, 1982, An evolutionary theory of economic

change. Cambridge, MA: Harvard University Press.

Noordewier, T. G., G. John and J. R. Nevin, 1990, ‘‘Performance outcomes

of purchasing arrangements in industrial buyer-vendor relationships’’.

Journal of Marketing, 54: 80–93.

Novak, S. and S. D. Eppinger, 2001, ‘‘Sourcing by design: Product complexity

and the supply chain’’. Management Science, 47: 189–204.

Nunnally, J. C., 1967, Psychometric theory. New York: McGraw-Hill Book

Company.

Parmigiani, A., 2007, ‘‘Why do firms both make and buy? An investigation

of concurrent sourcing’’. Strategic Management Journal, 28: 285–311.

Parmigiani, A. and W. Mitchell, 2009, ‘‘Complementarity, capabilities, and

the boundaries of the firm: The impact of within firm and interfirm

expertise on the concurrent sourcing of complementary components’’.


Paxson, M. C., D. A. Dillman and J. Tarnai, 1995, ‘‘Improving response

to business mail surveys’’. In B.G. Cox, D.A. Binder, B.N. Chinnappa, A.

Christianson, M.J. Colledge and P.S. Kott (eds) Business survey methods.

New York: John Wiley and Sons.

Petersen, K. J., R. B. Handfield and G. L. Ragatz, 2005, ‘‘Supplier integration

into new product development: Coordinating product, process and

supply chain design’’. Journal of Operations Management, 23: 371–388.

Phillips, L. W., 1981, ‘‘Assessing measurement error in key informant reports:

A methodological note on organizational analysis in marketing’’. Journal

of Marketing Research, 18: 395–415.

Podsakoff, P. M. and D. W. Organ, 1986, ‘‘Self-reports in organizational

research: Problems and prospects’’. Journal of Management, 12: 531–544.

Podsakoff, P. M., S. B. MacKenzie, J. -Y. Lee and N. P. Podsakoff, 2003,

‘‘Common methods biases in behavioral research: A critical review of

the literature and recommended remedies’’. Journal of Applied Psychology,

88: 879–903.

Poppo, L. and T. Zenger, 1998, ‘‘Testing alternative theories of the firm:

Transaction cost, knowledge-based, and measurement explanations for

make-or-buy decisions in information services’’. Strategic Management

Journal, 19: 853–877.

Poppo, L. and T. Zenger, 2002, ‘‘Do formal contracts and relational

governance function as substitutes or complements?’’ Strategic Management

Journal, 23: 707–725.

Poppo, L., K. Z. Zhou and S. Ryu, 2008, ‘‘Alternative origins to

interorganizational trust: An interdependence perspective on the shadow

of the past and the shadow of the future’’. Organization Science,

19: 39–55.

Poppo, L., K. Z. Zhou and T. Zenger, 2008, ‘‘Examining the conditional

limits of relational governance: Specialized assets, performance

ambiguity, and long-standing ties’’. Journal of Management Studies,

45: 1195–1216.

Powell, W. W., K. W. Koput and L. Smith-Doerr, 1996, ‘‘Inter-organizational

collaboration and the locus of innovation: Networks of learning in

biotechnology’’. Administrative Science Quarterly, 41: 116–145.

Ray, G., J. B. Barney and W. A. Muhanna, 2004, ‘‘Capabilities, business

processes, and competitive advantage: Choosing the dependent variable

in empirical tests of the resource-based view’’. Strategic Management

Journal, 25: 23–37.

Reuer, J. J. and A. Arino, 2007, ‘‘Strategic alliance contracts, dimensions

and determinants of contractual complexity’’. Strategic Management

Journal, 28: 313–330.

Richardson, G. B., 1972, ‘‘The organisation of industry’’. The Economic Journal

883–896.

Richardson, J., 1993, ‘‘Parallel sourcing and supplier performance in the

Japanese automobile industry’’. Strategic Management Journal, 14: 339–350.

Richardson, J. and J. Roumasset, 1995, ‘‘Sole sourcing, competitive sourcing,

parallel sourcing: Mechanisms for supplier performance’’. Managerial

and Decision Economics, 16: 71–84.

Ring, P. S. and A. Van De Ven, 1992, ‘‘Structuring cooperative relationships

between organizations’’. Strategic Management Journal, 13: 483–498.


65

AUTHOR COPY

Robinson, W. S., 1950, ‘‘Ecological correlations and the behavior of

individuals’’. American Sociological Review, 15: 351–357.

Rosenkopf, L. and P. Almeida, 2003, ‘‘Overcoming local search through

alliances and mobility’’. Management Science, 49: 751–766.

Ryall, M. and R. Sampson, 2009, ‘‘Formal contracts in the presence of

relational enforcement mechanisms: Evidence from technology development

projects’’. Management Science, 55: 906–925.

Sako, M., 1992, Prices, quality, and trust, inter-firm relations in Britain and

Japan. New York: Cambridge University Press.

Shaver, J. M., 1998, ‘‘Accounting for endogeneity when assessing strategy

performance: Does entry mode choice affect FDI survival?’’ Management

Science, 44: 571–585.

Szulanski, G., 1996, ‘‘Exploring internal stickiness: Impediments to the transfer

of best practice within the firm’’. Strategic Management Journal, 17: 27–43.

Takeishi, A., 2002, ‘‘Knowledge partitioning in the inter-firm division of

labor: The case of automotive product development’’. Organization

Science, 13: 321–338.

Tan, J. and M. N. Young, 2003, Dell computer corporation: Investment in

Malaysia as a global strategic tool. Case 903M19, The University of

Western Ontario, Richard Ivey School of Business.

Tiwana, A. and M. Keil, 2007, ‘‘Does peripheral knowledge complement

control? An empirical test in technology outsourcing alliances’’. Strategic

Management Journal, 28: 623–634.

Tomaskovi-Devey, D., J. Leiter and S. Thompson, 1994, ‘‘Organizational

survey nonresponse’’. Administrative Science Quarterly, 39: 439–457.

Tunisi, A. and A. Zanfei, 1998, ‘‘Exploiting and creating knowledge through

customer-supplier relationships: Lessons from a case study’’. R&D

Management, 28: 111–118.

Van Weele, A. J., 1994, Purchasing management: Analysis, planning, and

practice. London: Chapman and Hall.

Walker, G. and D. Weber, 1984, ‘‘A transaction cost approach to make-or-buy

decisions’’. Administrative Science Quarterly, 29: 373–391.

Walsh, R. A., 1994, McGraw-Hill machining and metalworking handbook.

New York: McGraw-Hill.

Wheelwright, S. C. and R. H. Hayes, 1985, ‘‘Competing through

manufacturing’’. Harvard Business Review, 63: 99–109.

Williamson, O. E., 1975, Markets and hierarchies: Analysis and antitrust

implications. New York: The Free Press.

Williamson, O. E., 1985, The economic institutions of capitalism: Firms,

markets, relational contracting. New York: Free Press.

Womack, J. P., D. T. Jones and D. Roos, 1990, The machine that changed

the world. New York: Harper Perennial.

Zaheer, A., B. McEvily and V. Perrone, 1998, ‘‘Does trust matter? Exploring

the effects of inter-organizational and interpersonal trust on performance’’.

Organization Science, 9: 141–159.

Zheng, K. Z., P. Laura and Z. Yang, 2008, ‘‘Relational ties or customized

contracts? A test of alternative perspectives of governance in China’’.

Journal of International Business Studies, 39: 526–534.

Zollo, M., J. J. Reuer and H. Singh, 2002, ‘‘Interorganizational routines

and performance in strategic alliances’’. Organization Science, 13: 701–713.

Appendix A

Survey Items, by ConstructSupplier Performance (dependent variables)Our current sources’ performance ony. isy. (1–7 scale from terrible to terrific) (Poppo and Zenger, 1998)

1. Price competitiveness and value2. Quality level/defect rates3. Cooperation and dispute resolution4. On-time shipments5. Communication and paperwork

Technical Expertise Items (1–7 scale, from ‘not true’ to ‘true’) (Walker and Weber, 1984)

1. Our manufacturing staff can/could easily produce dies.2. Making dies requires a deep expertise that our firm understands.3. We have internally produced dies for years.4. The skills used to make dies are closely related to those that we use to make other similar products.

Relational Governance Mechanism Items (1–7 scale, from ‘not true’ to ‘true’)

1. Our supplier relationships last for years (Noordewier et al., 1990).2. We will always work through difficulties with a supplier rather than switch to a new one (interviews).3. We regularly use confidentiality agreements with our suppliers (Dyer, 1997).4. We use formal, written contracts whenever possible (Macauley, 1963; Macneil, 1978; Heide and Miner, 1992).5. We immediately inform our suppliers whenever there’s a change in volume requirements (Noordewier et al., 1990).6. We communicate daily with major suppliers (interviews).7. We advise suppliers of their performance in relation to that of other suppliers (Noordewier et al., 1990).8. We evaluate our internal production using the same criteria and strictness as our suppliers (Helper, 1991).9. We have a formal, written scorecard that we always use to evaluate our suppliers (Noordewier et al., 1990).10. Level of quality certifications – highest if all, otherwise in order from high to low ISO/TS16949, QS9000, ISO900x (0 to

4 point scale) (interviews)].11. Our engineers never travel to our supplier’s plants (reversed) (Dyer, 1997).12. We frequently help our suppliers improve their processes by providing them with technical, engineering, quality, or

other assistance Dyer, 1997).13. Our suppliers do not help us in reducing costs and overall problem solving (reversed) (Noordewier et al., 1990).

ContractIf you use external suppliers for progressive dies, which best describes your transactions?

& 1. Long-term, rarely modified contracts for multiple dies& 2. Long-term, frequently modified contracts for multiple dies


66

AUTHOR COPY

& 3. Written purchase orders for each die& 4. Verbal purchase orders for each die& 5. Some combination of these or Other (Please explain _____________________)& 6. Not applicable – we source all of our progressive dies internally

Note: Options 1 & 2 coded as using contracts; options 3 & 4 as not using contracts (to avoid ambiguity, we dropped the16 cases that checked option 5; the results were robust to alternative codings)

Appendix BSee Table B1.

Table B1 Probit analysis, All internal (1) vs use outside supplier (0)

All internal vs use supplier

Firm technical expertise 0.37***(0.06)

Relational governance 0.00(0.03)

Contracting �0.08(0.20)

Asset specificity �0.01(0.06)

Volume uncertainty �0.01(0.06)

Technological uncertainty �0.09*(0.06)

Performance uncertainty 0.19**(0.09)

Supplier expertise �0.06(0.09)

Firm age 0.31***(0.12)

Firm size (Employees) �0.04(0.07)

Union �0.16(0.19)

Powder metal 0.03(0.18)

Scale economies �0.08*(0.05)

Volume required 0.01(0.05)

Input similarity �0.03(0.04)

Firm scope economies 0.23***(0.06)

Supplier scope economies �0.24***(0.06)

Component: Die design vs die building 0.03(0.17)

Component: Die maintenance vs die building 0.76***(0.19)

Component: End-part machining vs die building 0.41**(0.20)

Component: End-part coating vs die building �0.24(0.30)

Number of cases 561Log Likelihood �233.62Pseudo R-squared 0.35

*** Po0.01, ** Po0.05; * Po0.10 (one-tailed tests).Robust standard errors in parentheses; constant estimate omitted.


67

AUTHOR COPY

Appendix CSee Tables C1–C3.

Table C1 The impact of technical expertise and governance mechanisms on supplier performance, with expertise and relational governance interaction (OLS withrobust standard errors adjusted for repeat observations; positive coefficient indicates superior performance)




H3 (+): Contracting 0.12 0.06 �0.18 �0.04 0.00(0.14) (0.15) (0.14) (0.17) (0.16)

H4a (+): Technical expertise� relational governance 0.03*** 0.01 �0.01 �0.01 0.01(0.01) (0.01) (0.01) (0.10) (0.01)

Asset Specificity �0.02 0.00 0.03 0.00 0.02(0.04) (0.04) (0.04) (0.05) (0.04)





Firm age 0.11* 0.10 0.12 0.19** 0.08(0.07) (0.09) (0.10) (0.11) (0.10)

Firm size (Employees) 0.03 �0.06 0.05 �0.08* �0.03(0.05) (0.06) (0.06) (0.06) (0.06)

Union �0.07 �0.37** �0.15 �0.09 �0.11(0.16) (0.18) (0.18) (0.20) (0.16)

Powder metal �0.19* �0.05 �0.06 �0.15 �0.22*(0.14) (0.16) (0.14) (0.19) (0.16)

Scale economies �0.03 �0.04 �0.02 �0.08* �0.04(0.04) (0.04) (0.04) (0.05) (0.04)

Volume required �0.05 �0.01 0.00 0.02 �0.02(0.04) (0.04) (0.04) (0.05) (0.04)

Concurrent source �0.41*** �0.07 �0.32** �0.23 �0.21*(0.14) (0.16) (0.15) (0.19) (0.15)

Component: Die design vs die building �0.02 �0.17 �0.18* �0.03 �0.02(0.14) (0.15) (0.14) (0.18) (0.13)

Component: Die maintenance vs die building 0.23 0.32** 0.24 0.32* 0.19(0.19) (0.18) (0.22) (0.24) (0.21)


Component: End-part coating vs die building �0.15 �0.05 0.26* 0.60*** 0.16(0.16) (0.17) (0.18) (0.18) (0.19)

Inverse mills ratioa 0.43* 0.73** 0.26 0.80** 0.35(0.28) (0.34) (0.30) (0.39) (0.32)


aStage 1 of the model is a probit analysis that controls for the decision to out-source (Appendix B1).


68

AUTHOR COPY

Table C2 The impact of technical expertise and governance mechanisms on supplier performance, with expertise and contracting interaction (OLS with robuststandard errors adjusted for repeat observations; positive coefficient indicates superior performance)




H3 (+): Contracting 0.12 0.04 �0.18 �0.04 0.01(0.13) (0.14) (0.14) (0.17) (0.15)

H4b (+): Technical expertise� contracting 0.11* 0.04 �0.05 �0.01 0.08(0.08) (0.08) (0.06) (0.10) (0.08)






Firm age 0.11* 0.10 0.12 0.19** 0.08(0.07) (0.09) (0.10) (0.11) (0.10)


Union �0.07 �0.36** �0.15 �0.09 �0.11(0.16) (0.18) (0.18) (0.20) (0.16)

Powder metal �0.17 �0.05 �0.07 �0.15 �0.21*(0.14) (0.16) (0.14) (0.19) (0.16)

Scale economies �0.02 �0.04 �0.02 �0.08* �0.04(0.04) (0.04) (0.04) (0.05) (0.04)

Volume required �0.05 �0.01 0.00 0.02 �0.02(0.04) (0.04) (0.04) (0.05) (0.04)


Component: Die design vs die building �0.04 �0.16 �0.17 �0.03 �0.03(0.13) (0.15) (0.14) (0.18) (0.13)








69

AUTHOR COPY

Table C3 The impact of technical expertise and governance mechanisms on supplier performance, with relational governance and contracting interaction (OLS withrobust standard errors adjusted for repeat observations; positive coefficient indicates superior performance)



H2 (+): Relational governance 0.04* 0.00 �0.02 0.03 0.04*(0.02) (0.03) (0.03) (0.03) (0.03)

H3 (+): Contracting 0.07 0.04 �0.24* �0.11 �0.04(0.15) (0.15) (0.15) (0.19) (0.17)

H4c (+): Relational governance � contracting 0.02 0.02 0.10** 0.10 0.04(0.06) (0.06) (0.06) (0.08) (0.07)




Performance uncertainty 0.05 0.03 0.03 �0.03 �0.10*(0.06) (0.07) (0.07) (0.09) (0.07)


Firm age 0.11* 0.10 0.13 0.20** 0.09(0.07) (0.09) (0.10) (0.11) (0.10)


Union �0.06 �0.36** �0.13 �0.06 �0.09(0.16) (0.18) (0.18) (0.20) (0.16)

Powder metal �0.18 �0.05 �0.07 �0.16 �0.22*(0.14) (0.16) (0.14) (0.19) (0.16)

Scale economies �0.02 �0.04 �0.02 �0.08** �0.04(0.04) (0.04) (0.03) (0.05) (0.04)

Volume required �0.05 �0.01 0.00 0.02 �0.02(0.04) (0.04) (0.04) (0.05) (0.04)


Component: Die design vs die Building �0.03 �0.17 �0.19* �0.04 �0.03(0.13) (0.15) (0.14) (0.18) (0.13)








70

AUTHOR - Fuqua School of Businesswillm/bio/cv/papers/EMR_2009Hollow...The core question is whether...

Documents

Transcript of AUTHOR - Fuqua School of Businesswillm/bio/cv/papers/EMR_2009Hollow...The core question is whether...