the dynamic interplay of capability strengths

Strategic Management JournalStrat. Mgmt. J., 31: 1386–1409 (2010)

Published online EarlyView in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/smj.893

Received 10 June 2008; Final revision received 1 July 2010

THE DYNAMIC INTERPLAY OF CAPABILITYSTRENGTHS AND WEAKNESSES: INVESTIGATINGTHE BASES OF TEMPORARY COMPETITIVEADVANTAGE

DAVID G. SIRMON,1* MICHAEL A. HITT,1 JEAN-LUC ARREGLE,2and JOANNA TOCHMAN CAMPBELL1

1 Mays Business School, Texas A&M University, College Station, Texas, U.S.A.2 Edhec Business School, Nice, France

Foundational RBV work suggests that firms possess capabilities that represent strengths andothers that represent weaknesses. In contrast, contemporary research has examined capabilitystrengths while largely ignoring weaknesses. Addressing this oversight, we examine the directand integrated effects of sets of capability strengths and capability weaknesses on competitiveadvantage and its empirical correlate—relative performance. Additionally, we explore how envi-ronmental and firm-specific factors influence change in these drivers of competitive advantageover time. Results suggest that weakness sets have a negative effect on relative performance, whilestrength sets have an increasingly positive effect. The integrative effects of strength and weak-ness sets affect relative performance in a complex manner. For example, while high strength/lowweakness firms perform at high levels, firms integrating high strength with high weakness per-form well, but experience considerably more variance in their realized outcomes. Lastly, we findthat the strength and weakness sets change significantly over time in markets where competitionis more intense, thereby undermining the durability of competitive advantage. Our theory andresults indicate that achieving temporary advantage is more difficult than previously thoughtand that the erosion of advantage occurs routinely as a result of dynamic and interactive rivalry.Copyright 2010 John Wiley & Sons, Ltd.

INTRODUCTION

Understanding the sustainability of competitiveadvantage is central to strategic management (Bar-ney, 1991; Porter, 1985; Schendel, 1994). Priorwork—primarily conducted in related fields, suchas economics—suggested that sustained compet-itive advantage is possible in that while the per-formance of most firms converges toward aver-age, some firms are able to resist this trend(e.g., Mueller, 1986; Waring, 1996). However,

Keywords: competitive advantage; capabilities; weak-ness; strength; durability∗ Correspondence to: David G. Sirmon, Mays Business School,4113 TAMU, College Station, Texas 77843, U.S.A.E-mail: [email protected]

more recent work in strategic management demon-strates that such outcomes are rare and oftenonly operative for short periods of time (Wigginsand Ruefli, 2002). Increases in rivalry—extendingeven to the levels of hypercompetition (D’Aveni,1994)—and rapid technological changes (Bettisand Hitt, 1995) undermine the sustainability of acompetitive advantage (Thomas, 1996). For exam-ple, dynamic and aggressive rivals can erode themarket share of industry leaders, eventually lead-ing to their dethronement (Ferrier, Smith, andGrimm, 1999). In fact, Wiggins and Ruefli (2005)concluded that prior research identifying firms withsustained competitive advantage likely identifiedcompanies that achieved a series of temporaryadvantages over time. Thus, evidence suggeststhat achieving sustained competitive advantage

Copyright 2010 John Wiley & Sons, Ltd.

Capability Strengths and Weaknesses 1387

requires managers to understand the bases ofcompetitive advantage as well as the factors thatlead to dynamic changes in these bases thatallow them to concatenate a series of temporaryadvantages.

A primary theory upon which to understand thebases of competitive advantage is the resource-based view (RBV) (Barney, 1991). The RBVsuggests that when a firm resource is both valu-able and rare, a competitive advantage is possi-ble. A recent meta-analysis suggests that, indeed,the basic tenets of the RBV are largely supported(Crook et al., 2008). However, a review of thefoundations upon which the RBV is based suggeststhat there is more to a competitive advantage thanis often considered. The work of Selznick (1957),Penrose (1959), Wernerfelt (1984), and Rumelt(1984) (among others) proposes that the firm isa bundle of resources and capabilities1 whichtogether influence the achievement of a competi-tive advantage. Importantly, these scholars suggestthat this bundle of capabilities is composed of bothstrengths and weaknesses. For example, Werner-felt (1984) defines a resource as ‘anything whichcould be thought of as a strength or weakness of agiven firm’ (1984: 172), Selznick (1957) discussesdistinctive inadequacies in addition to a firm’sdistinctive competencies, and Penrose argues thatdeficient managerial talent can inhibit, at least tem-porarily, organizational objectives.

Despite these foundational positions, the dozensof empirical papers adopting a RBV frameworkhave addressed only the effects of strengths oncompetitive outcomes. In contrast, only one piece,Arend (2008), has attempted to empirically demon-strate that strategic liabilities matter as well.Despite Arend’s important work demonstratingthat weaknesses influence firm turnarounds, muchremains unknown about the influence of capabilityweaknesses on a firm’s competitive advantage. Infact, the omission of weaknesses has been identi-fied as a major shortcoming of the RBV (West andDeCastro, 2001), one that has led to the theoreti-cal and empirical misspecification of competitiveadvantage. This suggests that our current under-standing of the bases of competitive advantage is

1 Resources are the tangible and intangible assets that firmscontrol and capabilities are the ability to perform ‘a coordi-nated set of tasks utilizing organizational resources’ (Helfat andPeteraf, 2003: 999). While for the sake of parsimony we primar-ily employ the terminology of capability here forward, the logicis valid for resources.

likely incomplete and its complex nature under-appreciated. Indeed, this omission has ignited atheoretical debate (see Arend, 2003, 2004; Durand,2002; Powell, 2001, 2002, 2003), along with callsto empirically examine the influence of capabil-ity weaknesses. For example, Arend states that‘the RBV only tells half of the story; the RBVconsiders only factors that positively contribute tosustained performance’ (2004: 1003), while Mont-gomery implores researchers to examine ‘the darkside of the resource spectrum’ and the performanceeffects of firms having ‘liabilities in their resourceinventory’ (1995: 261). Moreover, Armstrong andShimizu (2007: 980) argue that ‘demonstratingnegative effects within the RBV framework willcross-validate the theoretical value of the RBV.’

This study heeds these calls to more fullyexplore the bases of competitive advantage as wellas investigating factors that affect change in them.First, in light of the highly competitive markets inwhich firms must operate to gain any competitiveadvantage, we direct attention to the dynamic inter-dependence among multiple rivals and specificallytheir sets of capabilities. Importantly, in such rival-rous markets, we suggest that it is the relative (tocompetitors) instead of an absolute quality of capa-bilities that matters most for competitive advan-tage. Using this as a frame, we develop theorythat guides our empirical investigation of the directand integrated effects of both capability strengthsand weaknesses on competitive advantage and itsempirical correlate—relative performance (Arend,2003). More specifically, by extending Arend’sconcept of strategic liabilities to incorporate a com-parative lens (Jacobides and Winter, 2005), weidentify in a more fine-grained manner a firm’sindividual capabilities as either a strength or aweakness (two distinct concepts) in relation torivals. Moreover, because firms are conceptualizedas bundles of capabilities, we examine multiplecapabilities simultaneously, allowing us to studythe direct and integrative effects of strength andweakness sets on relative performance. Second,we explore how environmental and firm-specificfactors affect change in a firm’s strength and weak-ness sets, which allows us to better understand thedynamics related to the durability of competitiveadvantage. This is especially important when firmsin highly dynamic and rivalrous landscapes attemptto concatenate a series of temporary advantages.

Our theoretical and empirical examination offersseveral contributions to the literature. First, our

Copyright 2010 John Wiley & Sons, Ltd. Strat. Mgmt. J., 31: 1386–1409 (2010)DOI: 10.1002/smj

1388 D. G. Sirmon et al.

extension of the strategic liabilities concept addsa richer understanding of how a comparativeapproach to value and rarity helps distinguishbetween capability strengths and weaknesses (e.g.,Priem and Butler, 2001). Moreover, this effortcomplements investigations of rivals’ dyadic actionsequences (Ferrier et al., 1999), by reflecting moregenerally how relative strengths and weaknessescreate a dynamic interdependence between multi-ple rivals competing in the same market space.

Second, focusing on capability sets, as opposedto one or two individual capabilities, demonstratesa more complete and, perhaps, accurate view of therelationship between capabilities and competitiveoutcomes. While prior research supports a posi-tive relationship between capability strengths andperformance (e.g., Carmeli and Tishler, 2004), weshow that groups of strengths jointly produce astrong synergistic effect on relative performance.Additionally, we find that sets of weaknesses neg-atively affect relative performance. Most impor-tantly, however, the integrative effects of strengthand weakness sets suggest that the bases for eventemporary competitive advantage are more com-plex than previously thought. For example, ourresults suggest that weaknesses can undermine thepotential competitive advantage of firms possess-ing strengths; however, combining high levels ofstrengths with high levels of weaknesses can actu-ally produce high relative performance, albeit withsignificant variation in realized performance (i.e.,risky combination). Together, these contributionsadvance our understanding of the bases for com-petitive advantage from largely a single channelfocus (i.e., identifying at least one valuable andrare capability) to a multi-channel understandingin which both strengths and weaknesses (and setsthereof) are considered simultaneously.

Third, because capability strengths and weak-nesses affect competitive advantage, exploring howenvironmental and firm-specific factors influencetheir change over time increases our knowledgerelated to the durability of competitive advantage.In support of Wiggins and Ruefli (2005), we con-clude that the durability of a competitive advan-tage is limited because strength and weakness setschange significantly over time in rivalrous mar-kets (D’Aveni, 1994; Ferrier, 2001). As such, thisresearch provides an understanding of some of theendogenous and exogenous factors affecting thetemporal nature of competitive advantage.

Next, we examine prior literature on capabilitystrengths and weaknesses and the theoretical frameprovided by the RBV. This discussion concludeswith the development of the concept of a firm’sstrength and weakness sets, followed by specifictheoretical arguments and hypotheses. We thenpresent the methods and the results of the hypothe-ses tests. A discussion of the findings, emphasizingtheir contributions as well as the study’s limita-tions, concludes the paper.

CONCEPTUAL FRAMEWORK

Although several concepts related to weak-nesses have been suggested, including core rigidi-ties (Leonard-Barton, 1992), resource weaknesses(West and DeCastro, 2001), and competitive dis-advantage (Powell, 2001), Arend’s (2004) concep-tualization of weaknesses as strategic liabilities isthe most formally defined. Stated simply, strate-gic liabilities represent ‘the other side of theledger’ (Arend, 2004: 1006) in resource-basedlogic. Specifically, Arend contrasted strategic as-sets, those capabilities following RBV tenets (e.g.,valuable, rare, etc.) with strategic liabilities, whichare capabilities that are costly (lead to high cost forthe firm), supply restricted and appropriated by thefirm. Supply restricted implies that strategic liabil-ities are not equally distributed among firms andcannot be converted to a benign state with anybenefit to the firm because of the costs involvedto do so. Thus, strategic liabilities represent theworst of capabilities. We argue that additional util-ity can be achieved in extending the treatment ofstrategic liabilities by relaxing the absolute (inter-nal to the firm) perspective and instead utilizing acomparative approach, thereby allowing a broaderconceptualization of weakness.

A comparative lens allows for a more fine-grained differentiation among rivals’ capabilities.As Peteraf and Barney argue ‘competitive advan-tage is the result of having more valuable resourcesthan other firms’ (2003: 317, emphasis added).Thus, ‘it is the strengths relative to competitorsthat matter and not absolute strengths’ (Werner-felt and Karnani, 1987: 192). However, while theapplication of a comparative lens may help iden-tify capability strengths of a firm in a particularcompetitive context (Grimm and Smith, 1997), itsimultaneously suggests a question regarding theeffects of less valuable capabilities.



Traditionally, within the RBV vernacular, valueis based on the ability of the firm to use a capabilityto exploit an opportunity, neutralize a threat, and/orimprove the efficiency or effectiveness of a firm,while rarity represents insufficient supply (Barney,1991). As a result, some RBV work has focusedprimarily on capabilities that have the potentialto produce an absolute level of gain (value) andon those that are in short supply (rare). However,rarity can exist both in terms of a capability’spresence (or lack thereof) or its level. For a spe-cific industry, it is likely that rarity will be moreattuned to levels of effectiveness per capability asopposed to its mere existence. For example, Cas-tanias and Helfat argue that ‘managerial humancapital. . .is scarce if a manager possesses higherquality skills relative to his or her competitors’(2001: 663). In other words, managerial capabilityis not rare per se, but high levels of manage-rial capability—beyond the skills possessed bycompetitors—are rare. Therefore, when using acomparative approach, the variance in the valueof a capability across rivals delineates rarity andis important to competitive outcomes (Grimm andSmith, 1997).

Figures 1a and 1b graphically display the impli-cations of variance in the value of a single capa-bility across competitors. These figures modeldifferent frequency distributions based on the valueof industry members’ research and development(R&D) capability. As shown in Figure 1b, despitea high absolute level of value for all competitors’R&D (perhaps common in some high-technologyindustries), as the variance between competitors’R&D value approaches zero, the capability canoffer only parity for the firms. For these firms,R&D will have little effect on a firm’s competitive

outcomes regardless of its absolute value. Thus, inthis industry, rarity is neither in presence of capa-bility as all competitors have a R&D capability,nor in level, as the variance in the value of com-petitors’ R&D capability is close to zero. In thisexample, the possession of R&D capability by afirm can be thought of as a necessary condition formaintaining a competitive position, but an insuffi-cient one to achieve competitive advantage.

However, as demonstrated in Figure 1a, in-creased variance in the value of competitors’ R&Dcapability gives rise to rarity in the level of thecapability. The distribution of value for the R&Dcapability in the industry specified in Figure 1a hasseveral important competitive implications. First,a R&D capability that is above parity representsa strength for those firms. Moreover, the degreeof strength increases as its relative value continuesto increase beyond parity. Describing strength as acapability that is more valuable than rivals’ corre-sponding capability fits well in the recent lens ofcomparative RBV research (Jacobides and Winter,2005; Peteraf and Barney; 2003, Sirmon, Gove,and Hitt, 2008). Second, while this comparativeapproach allows for more complete and accurateidentification of when a capability is likely to con-tribute to an advantage over competitors, it alsoallows for the modeling of weakness. Specifically,as shown in Figure 1a, a weakness exists whena capability’s relative value is below parity withrivals. Moreover, the degree of weakness increasesas its relative value continues to decline fur-ther below parity. Thus, as the variance increasesamong competitors’ R&D capability value, so doesits potential effect on performance outcomes.

Importantly, this conceptualization of weaknessdoes not necessarily mean that the capability has a

Industry’sfrequencydistribution

Low………...Level of value…….…..High

ParityStrengthWeakness

1a – Heterogeneous 1b – Homogeneous

Low………....Level of value….……High

Figures 1a and 1b. Rarity as a function of a capability’s relative value among competitors



negative value, but instead suggests that its valueis less than similar capabilities held by rivals. Also,if a firm does not possess a requisite capabil-ity, it would represent a weakness. In total then,as increasing capability strengths lead to greatercompetitive advantage, increasing capability weak-nesses are expected to contribute to competitivedisadvantage. Thus, although we build on Arend’sconcept of strategic liabilities, in our conceptual-ization, weaknesses (1) do not have to be abso-lutely costly, only less valuable than competitors’and (2) can be converted from weakness to par-ity or strength over time with a net benefit to thefirm. However, we acknowledge that the specifica-tion of a capability along the continuum betweenweakness and strength is contextually dependentand that converting a capability weakness into astrength involves costs.

Thus, in total, we argue that when the value ofa firm’s capability is below parity, it is a weak-ness and that the disadvantage associated with thatweakness increases as its value decreases. Theparallel and opposite effect operates for strengthsand advantage. Therefore, instead of consideringcompetitive disadvantage as the lack of compet-itive advantage, similar to Powell, we argue thatcompetitive disadvantage exists when an underly-ing capability fails ‘to satisfy the minimum suc-cess requirements. . .required of any firm’ (Powell,2001: 877).

Lastly, focusing on the linkages between indi-vidual capabilities and the firm’s overall compet-itive advantage does not represent the intellectualroots of the RBV well (e.g., Carmeli and Tishler,2004). In fact, Newbert (2008: 751) argues that‘it is unlikely that a firm’s competitive positionis solely attributable to any one specific resourceor capability.’ Likewise, work on complementari-ties is based on the integration of many capabilities(Black and Boal, 1994; Carmeli and Tishler, 2004;Stieglitz and Heine, 2007). Therefore, to examinea more accurate model of competitive advantage,we do not address one or two individual capabil-ities; instead, we explore how capabilities, whichare individually identified as either being a strengthor weakness, form sets of capability strengths andcapability weaknesses that directly and interac-tively affect competitive advantage. Importantly,while any particular capability falls on a continuumfrom weakness to strength, it can be consideredonly one or the other at any particular time, unless

it is found to be at parity where it contributes toneither set.2

THEORETICAL MODEL

Firms operate in dynamic marketplaces in whichcompetitors act and react to each other in orderto exploit any advantage and earn greater relativeperformance (Smith, Ferrier, and Ndofor, 2001). Infact, research shows that these rivalrous dynamicscan lead to the dethronement of market leaders(Ferrier et al., 1999; Ferrier et al., 2002). Whenviewed by the end-user, this rivalrous environ-ment highlights differences among rivals’ marketofferings as supported by their capabilities (Priem,2007). Thus, the differences among rivals’ capa-bility strength and weakness sets are expected toaffect a firm’s relative performance (Grimm andSmith, 1997). We begin by examining the effectof strength sets.

Firm’s strength set and relative performance

The growing literature on the RBV demonstratesthat an individual capability yields positive perfor-mance outcomes when it is valuable and rare (e.g.,Hitt et al., 2001). This relationship holds whenvalue is viewed through a comparative lens; higherlevels of relative value increase positive outcomes(Sirmon et al., 2008). However, this relationshipmay be even more substantial when consideringthe firm’s set of strengths because it can includenumerous capabilities. Thus, enhancing the firm’sset of strengths is likely to differentiate the firmfrom rivals, leading to increasing levels of compet-itive advantage and higher relative performance.

Increasingly positive performance gains are ex-pected for two reasons. First, an enhanced strengthset yields synergy. In other words, the strengthsembedded in a strength set can complement eachother. These complementarities, or enhancing rela-tionships (Black and Boal, 1994), occur whenthe marginal value of a strength is enhancedby increases in other strengths (Milgrom andRoberts, 1995). Complementarities allow firms to

2 Despite not creating advantage, a capability at parity is impor-tant. Capabilities at parity offer more value than capabilitieswhich represent weakness, but not as much as capabilities inpositions of strength. Moreover, they are likely needed by thefirm to be competitive. We thank a thoughtful reviewer for thisimportant idea.



improve both the quality and price of deliveredgoods/services, thereby not only growing marketshare but, perhaps, increasing the size of the mar-ket as well. Second, differences in rivals’ strengthsets allow managers to react to the same marketconditions in unique ways (Chen, 1996). For exam-ple, management’s awareness and motivation toengage in actions to improve their relative per-formance is affected by the firm’s set of capa-bilities (Chen, Su, and Tsai, 2007; Smith et al.,2001). An enhanced strength set allows firms toincrease both their aggressiveness and range ofcompetitive actions (Ferrier, 2001), while likelyreducing the effectiveness of competitors’ retalia-tory actions.

For example, Southwest Airlines’ system ofshort haul, direct flights, along with its operat-ing efficiencies and a highly motivated workforce(representing strengths) allow it not only to offerlow fares, but also to differentiate its service inways that are often superior to most competitors(Sirmon, Hitt, and Ireland, 2007). Southwest’s setof strengths creates synergies that enrich its com-petitiveness by producing greater levels of end-user satisfaction. Moreover, its strengths protectthe firm from competitive actions taken by rivals.Additionally, Southwest’s strength set facilitatesgrowth in the number of customers in the marketby allowing it to effectively compete with substi-tute services (e.g., automobiles, buses, and trains).As Miller, Eisenstat, and Foote (2002: 47) suggest,market share is beneficial because to improve per-formance, a firm must ‘satisfy the needs of a largeenough audience.’

In total, increases in a firm’s strength set dif-ferentiate it from rivals, allowing it to satisfymore end-users. The complementarities among thestrengths multiply the value that can be createdfor the customer as each increase. In addition, theunique actions and responses possible with increas-ing strengths also enhance the value the firm canprovide to customers above and beyond that pro-vided by rivals. Based on these arguments, a firm’sstrength set is likely to have an increasingly posi-tive effect on relative performance.

Hypothesis 1 (H1). There is positive curvilinearrelationship between a firm’s strength set andrelative performance. The relationship growsincreasingly positive as the firm’s strength setincreases.

Next, we consider how the firm’s weakness setaffects relative performance.

Firm’s weakness set and relative performance

Similar to strengths, complementarities can existamong the firm’s set of capability weaknesses,where the marginal value of a weakness decreaseswith increases in other capability weaknesses.Thus, akin to grouping several strengths, groups ofweaknesses together are likely to produce increas-ingly negative performance outcomes. Researchin competitive dynamics suggests that at leastthree factors contribute to a negative curvilin-ear relationship between the firm’s weakness setand relative performance: (1) likelihood of beingattacked, (2) inefficiencies, and (3) inability toexploit opportunities.

A traditional war strategy is to attack rivals inareas of weakness; the same logic is applicablein the current competitive landscape (Bettis andHitt, 1995; Grimm and Smith, 1997). Thus, turn-ing the awareness-motivation-capability perspec-tive around suggests that executives should expectrivals to attack in areas where their firm is weak(Hitt, Ireland, and Hoskisson, 2009). West andDeCastro argue this is true because weaknesses‘lay open the organization to critical strategic vul-nerabilities’ (2001: 418). As a result, when a firmhas weaknesses, the number and intensity of com-petitor attacks are likely to be higher and stronger.As the number and intensity of attacks increase,sales are likely to suffer and the possibility ofrepeat sales dissipates. This leads to increasinglynegative relative performance.

Additionally, the weakness set can increase afirm’s cost structure, which, in turn, impedes thefirm’s ability to profitably satisfy end-users. Forexample, costs increase from assigning employeesto perform tasks for which they are inadequatelytrained. As a result, they perform the task poorly,requiring additional service to correct problemsand/or to mollify dissatisfied customers. Further-more, these weaknesses can be self-reinforcing(e.g., loss of repeat customers), thereby impair-ing future performance. For instance, Leonard-Barton (1992) argues that the most skilled peoplein a functional domain seek employment wheretheir skills and abilities are highly respected andrewarded. Thus, as the best-trained people migrateto the top firms in their functional domain, thefirms with less effective functional skills will



Relative strength setLow High

Low

I Offsetting - undifferentiated

Neutral performance effect (0)

IIRobust advantage

Positive performance effect (+)

Relativeweakness set

High

IIIUndermining

Negative performance effect (-)

IVPrecarious advantage

Positive performance effect (+)

Figure 2. Performance effects of a firm’s integration of relative strength and weakness sets

become increasingly inefficient relative to theirrivals, thus increasing their weakness sets.

Lastly, a weakness set can limit a firm’s abil-ity to pursue new opportunities. For example, ifa firm’s assets, which are often used to facilitateand, indeed, implement strategic actions, are oflower value, it may have to forego certain strate-gic options such as acquisitions, entry into inter-national markets, or major investments in R&D.Excluding these actions from the firm’s reper-toire of strategic alternatives increases the firm’sreliance on a more narrow set of actions; sucha constraint has been shown to harm firm per-formance (Miller and Chen, 1996). Moreover,existing revenue streams are likely to be underincreasing pressure from the attacks of strongerrivals. Therefore, increases in a firm’s weaknessset are expected to differentiate the firm fromits rivals in a negative manner (i.e., rivals’ gainin advantage over the focal firm), thereby pro-ducing a negative curvilinear effect on relativeperformance.

Hypothesis 2 (H2). There is a negative curvi-linear relationship between the firm’s weaknessset and relative performance. The relationshipgrows increasingly negative as the firm’s weak-ness set degrades (i.e., grows larger).

Powell stated that ‘it seems unreasonable toexpect competitive advantage to imply superiorperformance no matter what else the firms may bedoing wrong’ (2001: 877). Therefore, understand-ing the effect of strength/weakness integration onrelative performance is important.

Integrative effect of weakness and strengthsets on relative performance

In order to effectively compete against rivals, firmsmust complete many tasks that, in combination,contribute to the satisfaction of end-users as wellas owners’ wealth (Morrow et al., 2007; Sirmonet al., 2011). To accomplish these outcomes, firmsmust integrate their capabilities in ways that allowthem to cope with external uncertainty (Thompson,1967). Thus, a firm’s strength and weakness setsare interdependent (Powell, 2001). We summarizethe potential combinations of strength and weak-ness sets and their effect on relative performancein Figure 2.

While limited, the few works that have addressedpotential outcomes from the integration of capa-bility strengths and weaknesses suggest that anet-effect logic will prevail. Specifically, Powell(2001) argues that strengths and weaknesses willoffset one another and when an imbalance occurs,performance will tip in favor of the dominant force(positive for strength and negative for weakness).Similar logic was proposed by Ray, Barney andMuhanna (2004). Specifically, they argued thata firm’s performance depends on the net effectof several capabilities. Likewise, Arend (2004)argued that performance is a function of the aggre-gate of firm capabilities. For example, if a hospi-tal’s set of weaknesses is relatively low (e.g., basedperhaps on nursing constraints, configuration andavailability of facilities, etc.), while its strengthsare high (e.g., perhaps based on location, and qual-ity medical doctors) the net-effect logic suggeststhat the set of strengths will overwhelm the set ofweaknesses, thereby producing positive outcomes.This particular combination is represented in Cell



II of Figure 2. Where a firm’s strength set is vastlysuperior to its weakness set, a robust advantageresults and, with it, higher relative performance.

The inverse combination, a low strength setpaired with a high weakness set, represented inCell III of Figure 2, should have a negative effecton relative performance. In this case, the firm’sstrength set exists, but at low levels. The weak-ness set, however, is able to undermine the strengthset’s positive contribution to firm outcomes, result-ing in negative relative performance.

The next combination, a low strength set pairedwith a low weakness set, represented in Cell Iof Figure 2, is likely to yield neutral results. Inthis case, the opposing sets of strengths and weak-nesses offset each other. There is very little todifferentiate firms with this combination fromcompetitors. These firms are close to average. Assuch, no consistent relative performance differ-ences are expected for these undifferentiated firms.

The last combination, a high strength set pairedwith a high weakness set, represented in Cell IV ofFigure 2, is more complex. On the one hand, thenet-effect logic suggests that these firms are notlikely to enjoy a relative performance differentialbecause the strength and weakness sets neutralizeeach other. On the other hand, an opposing andcompelling logic suggests that this combinationis a reasonable approach for managers trying toimprove performance in the presence of highlycompetitive rivals.

To the degree that managers are able to iden-tify weaknesses and strengths, the first alterna-tive they would likely consider is eliminatingweaknesses while enhancing strengths. However,it should be noted that pairing high strengthswith high weaknesses can provide some efficien-cies to the firm. Specifically, selective alloca-tion allows a firm to maximize the impact of itslimited investment resources. Managers can care-fully choose which capabilities to emphasize andfocus their investments in those selected capa-bilities. By focusing their investments, they aremore likely to realize capability strengths. Thisselective investment process forces trade-offs. Theemphasis on certain capabilities requires a de-emphasis of investments in others, likely thosethe managers deem less important for the strate-gies they have formulated. Thus, by maximizinginvestments in capabilities that support a selectiveset of strategic actions (e.g., process innovation,

internationalization, or M&As), managers can bet-ter ensure these capabilities are superior to rivalsand simultaneously acknowledge their weaknesses.Additionally, explicitly considering both strengthsand weaknesses can stimulate positive outcomesby increasing the breadth of the managers’ fieldof vision (Hambrick and Mason, 1984), enhanc-ing the comprehensiveness of their decision mak-ing (Simons, Pelled, and Smith, 1999), reducingstrategic myopia and complacency (Levinthal andMarch, 1993), and leading to strategic consensus(Knight et al., 1999).

Firms pursuing this selective investment patternwill not possess a robust advantage, but insteadmay be thought to have a precarious advantage.While efficient in terms of investment costs, a pre-carious advantage requires managers to contendwith both high strengths and weaknesses. Specifi-cally, a precarious advantage requires managers tooptimize a set of constrained competitive actionsto leverage the firm’s strengths and avoid or bufferagainst its weaknesses. For example, Costco hasdeveloped a focused set of important capabili-ties (e.g., supplier relations, human resources, andprocurement) to offer superior value to a selectedsegment of customers with a limited set of mer-chandise relative to competitors that target a broad-er range of customers using a vast array ofmerchandise (Bell and Leamon, 1998). Researchshows that competitive simplicity can producehighly variable performance outcomes (e.g., Fer-rier and Lyon, 2004; Miller and Chen, 1996).Thus, high levels of strength allow for effective,yet limited, competitive actions, while high levelsof weakness make the firm susceptible to attack.As such, the performance effects of a precariousadvantage are likely to be vulnerable; such anadvantage can lead to positive performance, yetwith increased levels of variation in the realizedpositive outcomes as opposed to firms possessinga robust advantage.

Formally, we list the following hypotheses forfirms represented by Cells II, III, and IV.3

Hypothesis 3a (H3a). The integration of a highstrength set with a low weakness set has a posi-tive effect on relative performance.

3 A formal hypothesis for Cell I is not offered because it is anull hypothesis.



Hypothesis 3b (H3b). The integration of a lowstrength set with a high weakness set has anegative effect on relative performance.

Hypothesis 3c (H3c). The integration of a highstrength set with a high weakness set has apositive effect on relative performance.

Hypothesis 3d (H3d). Precarious advantage pro-duces greater variation in performance out-comes than does a robust advantage.

Because we expect both strength and weaknesssets to affect relative performance, understand-ing factors that influence change in these driversinforms the durability of any advantage.

Change in strength and weakness sets overtime

The drive for profits increases dynamism in thecompetitive landscape (Bettis and Hitt, 1995; Fer-rier, 2001). With such dynamic rivalry, schol-ars argue that competitive advantage is fleeting(D’Aveni, 1994) and empirical results exist thatsupport this contention (Wiggins and Ruefli, 2005).Beyond a general indictment of high levels ofrivalry, our understanding of the dynamics relatedto the durability of competitive advantage is lim-ited. Indeed, if firms’ strength and weakness setschange significantly over time, these changes couldundermine the durability of a firm’s competitiveadvantage (or, in some cases, help build a newcompetitive advantage).

Because rivals, even indirect rivals, work to con-tinuously improve their capabilities in order toattain high relative performance, changes in thestrengths and weakness sets are likely to occur.However, continuous development of new compet-itive advantages requires basic resources, such asinvestment dollars, to maintain or improve the rel-ative strength of the firm’s capabilities (Kor andMahoney, 2005). Two sources of such resourcesare the external environment and the firm’s cur-rent operations. Therefore, we investigate howenvironmental munificence (exogenous) and priorfirm performance (endogenous) influence changesin the firm’s strength and weakness sets overtime.

Environmental munificence refers to ‘thescarcity or abundance of critical resources neededby (one or more) firms operating within an

environment’ (Castrogiovanni, 1991: 542). Whenoperating in munificent environments, the accessi-bility of requisite resources is not a major concern.Investment dollars, employees, and other resourcesare generally available to most or all competi-tors. Thus, while competing in munificent environ-ments reduces the threat of failure, differentiatingthe firm’s capabilities from rivals is more diffi-cult because of the access to plentiful resources.For example, in munificent environments, all firmscould strive to enhance their capability strengths.Furthermore, because capability strengths are crit-ical to firm success (and thereby highly salient),managers are likely aware of rivals’ actions toovertake their capability strengths and are moti-vated to prevent it from happening by taking sim-ilar actions, thereby triggering a Red Queen effectamong rivals (Barnett and McKendrick, 2004).In this case, the availability of critical resourcesserves as a catalyst for firms’ actions and reactionsthat, in fact, will not change their relative positions.Moreover, investment alone may not be adequateto build or increase capability strengths. Instead,managers’ efforts to invest in and bundle resourcesto develop capabilities are needed to create rel-ative strengths (Sirmon et al., 2008; Sirmon andHitt, 2003; 2009). These special managerial skillsare not equally distributed among rivals (Castaniasand Helfat, 2001). Therefore, a munificent envi-ronment alone is insufficient to enhance strengthsets.

The same logic can be applied to reducingcapability weaknesses in munificent environments.Although munificence may allow firms to ignoreweaknesses, there are theoretical reasons whymunificence may allow competitive firms to reducetheir set of weaknesses. First, theory suggests thatapproaching parity is far easier than building acompetitive advantage (Barney, 1991). In otherwords, increasing a capability’s relative value tothe average (i.e., parity) is more likely than over-taking competitors with superior capabilities. Sec-ond, working to improve a capability to parity maynot be as threatening to rivals in munificent envi-ronments. In fact, achieving such parity may notbe adequate to trigger rivals’ awareness nor moti-vate them to react. Such actions are unlikely tounleash a Red Queen effect among rivals. There-fore, while environmental munificence is unlikelyto affect change in a strength set, it allows the firmto reduce its weakness set over time.



Hypothesis 4a (H4a). Higher munificence leadsto a decrease in firms’ weakness sets over time.

Additionally, firms can utilize their own re-sources to change their strength and weaknesssets. For instance, they can allocate flows of firm-specific resources (Dierickx and Cool, 1989) toenrich their stock of capabilities. Moreover, firm-specific resources are largely inaccessible to rivals.For example, financial resources generated by thefirm’s prior performance are available for its pri-vate use. The asymmetric nature of these resourcesallows the firm to potentially differentiate its capa-bilities from those of rivals because managers candirect these private financial resources into firm-specific R&D projects (Ndofor, Sirmon, and He,2011), in-house training sessions, etc. Thus, higherprior performance provides an asymmetric sourceof firm-specific resources with which to enhancethe firm’s set of strengths and/or reduce its set ofweaknesses.

Here, even if rivals are aware and motivated tomatch the firm’s investments, they do not neces-sarily have access to the requisite resources to doso. This, in part, explains why it is especially dif-ficult for firms experiencing poor performance toturn around their fortunes (Morrow et al., 2007);instead they often become less aggressive andslower to act (Ferrier et al., 2002). These argu-ments suggest that higher prior performance pro-vides firms the opportunity to enrich their strengthset as well as reduce their set of weaknesses overtime.

Hypothesis 4b (H4b). Higher prior performanceleads to an increase in firms’ strength sets overtime.

Hypothesis 4c (H4c). Higher prior performanceleads to a decrease in firms’ weakness sets overtime.

METHODS

Sample

The data presented in this study were collectedby the Banque de France4 as part of its Sesame

4 No merchant or power relations exist between the Banque deFrance and the interviewed companies.

project. The Sesame project was a nonprofit ven-ture developed to collect detailed strategic andmanagerial data on French industrial firms to com-plement the financial data that the Banque deFrance already possesses. A random sample of4,169 small and medium sized industrial firms wasselected to participate in the survey (Cool and Hen-derson, 1998). Because of the sample’s size, eachyear econometricians from the Banque de Franceadministered the survey to one-third of the sample,from 2001 to 2003. Thus, the data are not panel-ized. The interviews were conducted by Banquede France agents specially trained in survey tech-niques; they conducted personal interviews withCEOs using a computer-aided questionnaire. Thebusinesses were classified according to the Euro-pean equivalent of the SIC classification system(the NACE) at the three-digit level. We analyzedthose firms that also had complete financial per-formance data, resulting in a final sample of 2,980firms belonging to 78 separate industries. The aver-age number of firms per three-digit industry is38.21 and the average size of these firms is 104employees (please see Appendix 1). This samplewas used to test the performance hypotheses.

From 2004 to 2006, the Banque de Franceadministered a second survey to the initial groupof firms. This survey was identical to the ini-tial instrument. In total, 1,868 firms from the firstround participated in the second survey. Missingdata limited the set of usable observations to 1,578.This sample was used to test hypotheses related tochanges in strength and weakness sets.

Measures

Dependent variables

Three separate dependent variables were used totest the hypotheses. To assess the performancehypotheses, an accounting-based measure was usedbecause many firms in the sample are privatefor which no market-based data are available.Specifically, we measured performance with firmvalue-added. This figure approximates the contri-bution the firm has made in the transformationprocess from inputs to finished output. It mea-sures the full value actually created by the firmthrough its operations before its distribution tothe firm’s stakeholders (e.g., family members andemployees). This measure corresponds to the totaleconomic value created by the capital and labor



employed by a firm and is a fundamental per-formance metric (Lieberman and Dhawan, 2005).Value-added is calculated by subtracting from thefirm’s total revenue any intermediate consump-tion, which includes such items as raw materials,semifinished products, external services used tomanufacture finished products, energy consump-tion, etc. Also, considering the presence of manyprivate firms in the sample, in which owners maytry to limit their declared net profit with tax strate-gies (George, 2005), value-added is a more accu-rate proxy for performance because it is less likelyto reflect such biases. As such, this type of mea-sure is an appropriate metric for empirical studiesof the RBV, especially when private firms are inthe sample. Also, because it is not a ratio, there areno concerns regarding the source of its variance.

Because our theory pertained to competitiveadvantage, a relative construct (Hitt et al., 2009),we utilized a relative performance measure as sug-gested by the literature (Arend, 2003). To calcu-late the relative performance measure, we industrycentered value-added on three-digit industry clas-sifications. Moreover, we lagged this variable tocorrespond to firm performance two years after theyear the survey was completed.5

Importantly, these performance data were notcollected via the Sesame questionnaire, but are partof the Banque de France’s ongoing and regularfinancial data collection efforts for France’s indus-trial firms. Therefore, common method bias is not aconcern with these data. Furthermore, the Banqueaudits the figures where necessary to ensure theiraccuracy, increasing their validity. Finally, perfor-mance measures are not available for the secondsurvey because they are still in the process of beingcentralized and validated by the Banque de France.

To test the hypotheses related to changes instrength and weakness sets, we compared firms’scores for their strength and weakness sets (de-scribed later) from the second survey to the cor-responding scores from the first survey to producethe variables: Strength Set Change and WeaknessSet Change.6 These variables represent the differ-ence between the strength (weakness) set at thetime of the second survey and the strength (weak-ness) set at the time of the first survey. Again,

5 Results with a one-year lag are substantively the same as thosereported.6 Both of these variables satisfy the criteria for representing anormal distribution.

these figures were computed for the 1,578 firmsthat were surveyed twice.

Independent variables

The variables of primary interest to the perfor-mance hypotheses are sets of strengths and weak-nesses. Importantly, these variables represent amuch larger number of capabilities than most priorresearch has examined. Specifically, we obtaineddata from the Sesame questionnaire which mea-sures the relative value of six capabilities. TheCEO (i.e., respondent) was asked to evaluate eachfocal capability against his/her firm’s major com-petitors’ functionally similar capabilities.7 Respon-dents evaluated the focal capability as in a weakerposition (−2), slightly weaker position (−1), aver-age position (0), slightly stronger position (1),or among the strongest positions (2). The indi-vidual capabilities examined included: cost man-agement, product reliability, product innovation,process innovation, brand management, and manu-facturing flexibility. While this is not an exhaustivelist, these capabilities are highly important for cre-ating value in industrial firms, are output driven,and represent the major capabilities of the firmssurveyed.

The strength set variable summarizes firm capa-bilities with values greater than major rivals, whilethe weakness set variable summarizes firm capa-bilities with values less than major rivals. To sepa-rate strengths from weaknesses, we utilized splinefunctions, which model differences in relation-ships above and below a critical threshold level(Greene, 2003; Greve, 2003). Importantly, the crit-ical threshold was set at zero per item response.8

7 Because respondents identified the subset of major competitorsto which to compare their firm, there is a chance of inconsistentscores. However, three different empirical procedures addressthis concern. First, we tested, for each industry, the reliabilityof the six capability scales to measure the consistency amongrespondents. Cronbach’s Alpha scores with a mean score of0.68 indicate acceptable internal consistency. As a robustnesscheck, we also run our analyses only with industries with Alpha>0.70, with no substantive changes in the outcomes. Second,closer examination of the mean industry standard deviation perstrength (0.69–0.80) and weakness (0.41–0.79) suggest thatresponses were consistent. Moreover, the responses per industrydo not deviate from normal distributions. We thank a thoughtfulreviewer for suggesting we examine these attributes.8 To test the robustness of our results, we examined severalalternative operationalizations. First, we kept our spline functionset at zero but instead of following the Deephouse (1999)approach, we simply summed strengths and weaknesses. Second,we explored an empirically, rather than theoretically, set spline



Thus, each capability score above parity was codedas a capability strength, while each capability scorebelow parity was coded as a capability weakness.A value of 0 indicates a lack of strength or weak-ness and, while important to the firm, does notcontribute to either set.

Second, following our theoretical approach em-phasizing firms’ sets of strengths and weaknessesin a competitive market, we used Deephouse’s(1999) approach to compute strength and weak-ness set scores. The following equation illustratesthe calculation of firm f ’s strength set, where Saf

is the strength score for capability a for firm f ,Ave(Sai) is the average strength score for capa-bility a in industry i, and SD(Sai) is its standarddeviation. Industry is based on the three-digit level.The corresponding weakness equation was utilizedto compute each firm’s weakness set score; beforecomputing this variable the absolute value of theraw weakness scores was obtained.

6∑

a=1

[(Saf − Ave(Sai))/SD(Sai)

]

For hypotheses related to changes in strengthand weakness sets, the variables of interest aremunificence and prior performance. Environmentalmunificence refers to the availability of resourcesin the environment (Castrogiovanni, 1991). Fol-lowing Sutcliffe’s (1994) approach, industry saleswere used to create a time trend regression anal-ysis. Munificence is the unstandardized regressioncoefficient in the regression model divided by themean of industry sales. The greater this value, thehigher the industry’s munificence score. Impor-tantly, similar to our performance measure, thedata used for these regressions were not obtainedthrough the Sesame surveys, but were indepen-dently collected by the Banque de France as part oftheir population-level data collection effort. Thus,these figures are not influenced by sample size orby any specific firms in the sample, but reflectthe entire population of firms in the focal indus-try. Lastly, prior performance is proxied by the

location. Specifically, we used the sample mean per industryto establish the spline. For this empirically established spline,we developed two different approaches for the final variables.For one, we simply summed the strengths and weaknesses;for the other, we followed the Deephouse approach. All threeof these alternative approaches produced substantially similarresults (strength and weakness sets directly and interactivelyaffect performance). We thank a thoughtful reviewer for thissuggestion.

firm’s return on assets for the year of the firstsurvey. Again, these data were gathered indepen-dently, thus, common method variance is not aconcern.

Control variables

Several control variables were included in bothanalyses. First, because the initial data were col-lected over three years, we control for the year thequestionnaire was administered. Two dummy vari-ables control for the year effect: Year 1 and Year2. Second, we controlled for the size of the firm,because larger firms are likely to have larger rev-enues, costs and, perhaps, more capabilities, etc.Size is the log of the number of employees inthe firm. Third, we controlled for the potentialeffect of membership in a trade group, becauseprior research suggests that group membership canaffect a firm’s competitive position based on accessto group resources and capabilities beyond thoseheld by the firm (Hoskisson et al., 2004). Group isa dummy variable, with a 1 indicating membershipin a business group. Fourth, to control for poten-tial differences between the firms that are publiclytraded and private firms, we included a dummyvariable for this dichotomy. Public is coded witha 1 indicating public ownership. Fifth, we con-trolled for the age of the firm by including thelogarithm of the number of years since founding.Sixth, we controlled for the firm’s level of productdiversification based on a Herfindahl measure ofrevenues across a potential of ten separate prod-uct categories. All of these control variables wereincluded in both sets of analyses.

Additionally, for the tests of the performancehypotheses, we included both prior performanceand munificence, as discussed earlier, as controls.When testing the hypotheses related to changesin strength and weakness sets, we added threemore controls. These included the number of yearsbetween surveys. This variable is a count of theyears between the initial and the second survey.Also, we controlled for the initial level of strengthsor weaknesses. These initial-level variables capturethe dynamics of rivalry in each market.

Lastly, in both sets of analyses, we controlledfor any other unobserved industry-level effectswith industry dummies entered at the two-digitlevel. Industry clustered robust standard errorswere applied to our regression modeling to provideconservative tests of our hypotheses.



RESULTS

The descriptive statistics and correlations are pre-sented in Table 1, while Table 2 presents theresults of tests for the relative performance hypoth-eses. Table 3 provides the results of tests forthe hypotheses related to changes in strength andweakness sets.

Hypothesis 1 suggests that relative performancewill grow increasingly positive as the firm’sstrength set increases. Models 2 and 3 in Table 2show that both the strength set variable and itssquare are positive and statistically significant.These results provide support for the hypothesizedpositive curvilinear relationship.

Hypothesis 2 proposes that relative performancegrows increasingly negative as the firm’s weaknessset degrades (i.e., grows larger). As seen in Models2 and 3 in Table 2, the weakness set variable isnegative and statistically significant. However, itssquare is not statistically significant. These resultsdo not support Hypothesis 2. Instead of a negativecurvilinear relationship, the results show that afirm’s set of weaknesses has a negative lineareffect on relative performance.9

To test the hypothesized effects of the differentcombinations of strength and weakness sets onrelative performance, we utilized dummy variablesto identify the firms represented by the four cellspresented in Figure 2. While median splits alongthe sample’s strength set score and weakness setscore could be used for the identification of thesegroups,10 to ensure the precision of the test andto reflect the theoretical argument, we identifiedhighs and lows at the 25th percentile level, whichallowed for a strong degree of differentiation, yetoffered a reasonable number of observations pergroup. For example, firms represented by Cell II inFigure 2 (high strength/low weakness) were codedwith a dummy variable when their strength set wasamong the top 25th percentile and their weaknessset was among the bottom 25th percentile. Werepeated this approach until all firms representedby Figure 2 were identified. The comparison group

9 To ensure that the results for H1 and H2 were not drivenby one of the underlining capabilities, we analyzed each splinespecification per capability (before they were operationalized assets). These results show that the outcomes for the sets are notthe result of a single underlying capability, nor were any of theindividual specifications counter to theory.10 The median split approach produced similar results with thelow strength/low weakness group as the comparison.

for these groups of firms are the firms that do notmeet these criteria; those that are near average intheir industries.

The results for the tests of Hypotheses 3a, 3b,and 3c are listed in Table 2, Model 4. Hypothesis3a suggested that the integration of a high strengthset with a low weakness set positively affects rel-ative performance. The positive and statisticallysignificant coefficient for this grouping supportsthis hypothesis. Hypothesis 3b suggested that theintegration of a low strength set with a high weak-ness set negatively affects relative performance.The negative and statistically significant coefficientfor this grouping provides support for this hypoth-esis. Hypothesis 3c suggested that the integrationof a high strength set with a high weakness set hasa positive effect on relative firm performance. Thepositive and statistically significant coefficient forthis grouping supports this hypothesis. Also, it isuseful to note that, as expected, the integration ofa low strength set with a low weakness set has noeffect on relative performance.

Figure 3 provides a rich visual representation ofthe combinative effects strength and weakness setshave on relative performance. The figure is basedon the equation listed in Model 5 and reflects theresults offered in Model 4 as well. The centralregion of the surface is located at the intersectionof the middle strength and middle weakness setvalues. It is not surprising to see that this centralarea relates closely with zero on the relative per-formance scale. Contrasting this central locationwith the four corners of the graph is informativeand reflects the results in Model 4. Moreover, con-sidering the dynamic nature of the graph’s surfacecontours can provide additional understanding.

First, the robust advantage region (high strength/low weakness) generally offers much higherlevels of relative performance. Likewise, the pre-carious advantage region (high strength/high weak-ness) also shows higher levels of performance.Third, the low strength/low weakness region, whatwe term offsetting, does not show much appre-ciable difference in relative performance. Fourth,the low strength/high weakness region, what weterm undermining in Figure 2, shows large lossesin performance. Additionally, the curvilinear sur-face demonstrates the increasingly positive effectof capability strength sets on relative performance.

Comparison of the robust and precariousadvantage regions is important. Graphically,the precarious advantage region seems to offer the



Tabl

e1.

Des

crip

tive

stat

istic

san

dco

rrel

atio

ns

Var

iabl

eM

ean

s.d.

12

34

56

78

910

1112

1314

15

1R

elat

ive

perf

orm

ance

173.

9536

5.65

2Y

ear

10.

340.

470.

033

Yea

r2

0.33

0.47

−0.0

6−0

.51

4Si

ze4.

230.

800.

720.

01−0

.12

5A

ge30

.93

24.8

60.

100.

00−0

.01

0.15

6G

roup

0.39

0.49

0.32

0.00

−0.0

60.

420.

007

Publ

ic0.

010.

120.

090.

00−0

.01

0.11

0.04

0.04

8Pr

ior

perf

orm

ance

20.2

910

7.36

0.00

0.03

−0.0

1−0

.03

−0.0

40.

02−0

.01

9Pr

oduc

tdi

vers

ifica

tion

0.15

0.22

−0.0

1−0

.01

0.01

0.01

0.06

−0.0

50.

02−0

.02

10M

unifi

cenc

e0.

040.

040.

000.

02−0

.16

0.06

−0.0

60.

07−0

.03

0.01

−0.0

511

Stre

ngth

set

0.01

2.74

0.07

−0.0

20.

020.

00−0

.07

0.00

0.01

0.03

−0.0

1−0

.01

12W

eakn

ess

set

−0.0

52.

33−0

.02

0.00

0.00

0.05

0.06

0.03

−0.0

1−0

.03

0.02

0.02

−0.1

013

Low

stre

ngth

&lo

ww

eakn

ess

0.08

0.27

0.00

−0.0

10.

03−0

.02

0.02

0.01

0.00

0.02

−0.0

10.

00−0

.39

−0.3

014

Low

stre

ngth

&hi

ghw

eakn

ess

0.07

0.25

−0.0

20.

020.

000.

050.

040.

040.

03−0

.02

0.00

0.02

−0.3

60.

52−0

.08

15H

igh

stre

ngth

&lo

ww

eakn

ess

0.04

0.19

0.04

0.00

0.01

0.01

−0.0

50.

010.

020.

000.

000.

000.

24−0

.19

−0.0

6−0

.05

16H

igh

stre

ngth

&hi

ghw

eakn

ess

0.04

0.17

0.05

−0.0

10.

000.

04−0

.02

0.05

0.00

−0.0

1−0

.02

−0.0

20.

210.

19−0

.05

−0.0

5−0

.03

Cor

rela

tions

grea

ter

than

0.03

are

sign

ifica

ntat

p<

0.05

;co

rrel

atio

nsgr

eate

rth

an0.

04ar

esi

gnifi

cant

atp

<0.

01;

n=

2,98

0.



Table 2. Strength and weakness sets effects on relative performance

Variable Model 1Relative

performance

Model 2Relative

performance

Model 3Relative

performance

Model 4Relative

performance

Model 5Relative

performance

Constant −21632.940∗∗∗ −21786.210∗∗∗ −21769.360∗∗∗ −21678.540∗∗∗ −21788.770∗∗∗

Year 1 462.152∗∗ 472.008∗∗ 472.882∗∗ 484.637∗∗ 478.197∗∗

Year 2 597.491∗∗ 581.375∗∗ 583.900∗∗ 606.653∗∗ 574.931∗∗

Size 5011.001∗∗∗ 5004.874∗∗∗ 5003.295∗∗∗ 5014.246∗∗∗ 5005.423∗∗∗

Age −3.660 −2.149 −2.140 −2.951 −2.247Group 443.632∗∗ 449.502∗∗ 450.436∗∗ 436.703∗∗ 446.666∗∗

Public 693.932 628.953 629.595 720.225 662.197Prior performance 1.004+ 0.866 0.868 0.958 0.867Product diversification −363.582 −340.470 −337.638 −347.214 −346.385Munificence −2737.559 −2695.303 −2686.905 −2523.158 −2767.025Strength set 109.237∗∗∗ 106.784∗∗∗ 108.816∗∗∗

Weakness set −108.167∗∗ −94.972∗∗ −66.828∗

Strength set squared 14.560∗∗ 14.957∗∗ 16.323∗

Weakness set squared 2.787Strength set × weakness set 17.559+

Low strength & low weakness 120.727Low strength & high weakness −989.144∗∗∗

High strength & low weakness 675.766∗

High strength & high weakness 845.521∗

F 407.70∗∗∗ 289.79∗∗∗ 313.99∗∗∗ 286.16∗∗∗ 290.18∗∗∗

R2 0.53 0.56 0.56 0.56 0.56n 2980 2980 2980 2980 2980

+ p < 0.10, ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001; directional hypotheses use single-tailed tests; industry dummies included but notlisted.

Table 3. Firm and environmental effects on strength and weakness set change

Model 6 Strength set change Model 7 Weakness set change

Constant 0.043 0.042Years between surveys −0.018 −0.072Year 1 −0.148 0.183Year 2 0.077 −0.256Size 0.016 −0.115Age 0.001 0.000Group −0.125 −0.055Public 1.300+ 0.021Product diversification −0.365 0.292Initial strength set −0.487∗∗∗

Initial weakness set −1.111∗∗∗

Munificence 6.340 −9.240∗∗

Prior performance 0.109∗ −0.140∗

F 21.07∗∗∗ 143.98∗∗∗

R2 0.15 0.53n 1.578 1.578

+ p < 0.10, ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001; industry dummies included but not listed.

highest potential performance, but analyses showthat the coefficients for these two groups of firms,found in Model 4, are not statistically different(f = 0.739; ns). Second, we examined thedifferences in the variation between these two

combinations to test Hypothesis 3d. The resultsindicate that firms with a robust advantage (highstrength/low weakness) have far less variance intheir performance outcomes than do the firmswith a precarious advantage (high strength/high



-1500

-1000

-500

0

500

1000

1500

2000

Weakness Set Strength Set

Rel

ativ

e P

erfo

rman

ce

LoLo

HiHi

Figure 3. Response surface of the integration of strengthand weakness sets

weaknesses) (f = 1.873, p < 0.001). More specif-ically, the standard deviation for performance infirms with a robust advantage is approximately 37percent smaller than for firms possessing a precar-ious advantage. These results provide support forHypothesis 3d and have important implications forthe riskiness and durability of these two forms ofadvantage.

The results for the hypotheses related to changein strength and weakness hypotheses are listedin Table 3, Models 6 and 7. Hypothesis 4a sug-gested that environmental munificence helps firmsreduce their weakness sets over time. As depictedin Model 7, munificence has a negative and sta-tistically significant coefficient, providing supportfor this hypothesis. Hypothesis 4b suggested thathigher prior performance would allow firms toincrease their strength sets over time. The posi-tive and statistically significant coefficient listedin Model 6 provides support for this hypothesis.Finally, Hypothesis 4c suggested that higher priorperformance would allow firms to reduce theirweakness sets over time. The negative and statis-tically significant coefficient for prior performancein Model 7 provides support for this hypothe-sis. Additionally, it is worth noting that consistentwith our logic, the results shown in Model 6 indi-cate the lack of a statistically significant relation-ship between munificence and change in a firm’sstrength set.

Lastly, the strong effect for the initial strengthvariable in Model 6 and initial weakness variable

in Model 7 indicate that, indeed, strong competi-tive efforts are being put forth by rivals.

DISCUSSION

Understanding the nature of competitive advantageis at the heart of strategic management (Grimmand Smith, 1997). As such, more fully identi-fying and investigating the bases of competitiveadvantage, as well as the factors that influencetheir change over time, is important to both the-ory and practice. By exploring the role of capa-bility weaknesses along with capability strengths,we more fully describe the bases of competitiveadvantage. Moreover, our investigation of factorsthat affect change in firms’ capability weaknessand strength sets over time adds to the field’slimited understanding of the dynamics related tothe durability of competitive advantage. In total,our work demonstrates that the bases of competi-tive advantage are more complex than previouslythought; strengths are not the only driver of com-petitive advantage—weaknesses matter greatly aswell. Additionally, the durability of any advan-tage is undermined by rivals’, even indirect rivals’,ongoing and dynamic capability investments.

While Chen and Miller (1994) suggested thatconsidering weakness is important to the strate-gic management of firms, research on the RBVhas largely ignored the effects of capability weak-nesses (Montgomery, 1995). Especially absent isempirical research on the effects of weaknesseson competitive advantage. As such, we build onprior conceptual work to theoretically differenti-ate capability strengths and weaknesses by view-ing value and rarity through a comparative lens.Additionally, because firms are conceptualized asbundles of integrated capabilities, we focus on setsof capability strengths and weaknesses. And, asour results show, both strength and weakness setshave important direct and integrated effects oncompetitive advantage via its empirical correlate,relative performance. These results indicate that,indeed, the exclusion of weaknesses has limitedour understanding of competitive advantage. Infact, our results suggest this void in prior researchlikely represents a critical limitation. Addressingthis oversight is important for a more completeand accurate understanding of competitive advan-tage, but especially regarding the temporal natureof competitive advantage. A temporary advantage



does not result only through the developmentor protection of strengths alone; weaknesses alsomatter.

More specifically, our empirical results showthat a firm’s set of capability strengths has anincreasingly positive relationship with relative per-formance. This result is important for at least threereasons. First, this outcome suggests that syner-gistic effects exist among strengths. Second, theresults suggest that continuing to build capabil-ity strengths pays major dividends. Thus, firmsshould not remain static after achieving success;they should continue to develop their capabilities,thereby adding to their competitive advantage andresulting performance. Third, this research is thefirst to specify and support the functional formof the firm’s strength set: strengths complementone another and yield increasingly positive relativeperformance.

Another important finding pertains to the directeffects of weakness sets on competitive advan-tage and relative performance. While we hypothe-sized a negative curvilinear relationship, we founda strong negative linear relationship. Perhaps thecurvilinear effect, which was expected to occurat the highest levels of weaknesses, is not iden-tified because firms exit markets when they havesuch high levels of weakness. Regardless, the neg-ative outcome demonstrates that weaknesses areimportant to relative performance and should beconsidered in future RBV research.

The importance of weaknesses is underscoredby the results of the integration hypotheses. Thelimited theory pertaining to strength/weakness inte-gration suggests that a counterbalancing relation-ship with performance exists; one that favors thegreater force (Powell, 2001; Ray et al., 2004). Thenonsignificant results for the integration of lowstrength and low weakness sets are consistent withthis logic, as are the results for Hypothesis 3b,which show that integrating a low strength setwith a high weakness set negatively affects rel-ative performance. Firms in this latter conditionpossess an overall competitive disadvantage. How-ever, it is important to note that without consider-ation of weaknesses, RBV-based logic would havesuggested that these firms would, perhaps, be ade-quate performers because some level of strengthis possessed; while not the best performers, thesewould not be expected to perform so poorly. Ourresults indicate that modeling weaknesses along

with strengths provides a more accurate picture offirm outcomes.

Next, integrating high strength and low weak-ness sets (robust advantage) and integrating highstrength and high weakness sets (precarious advan-tage) positively affect relative performance. Thelatter finding runs contrary to the net-offset logicand, instead, supports our arguments pertainingto efficient investments. That is, a discriminatinginvestment strategy that maximizes some capa-bilities at the expense of others can be bene-ficial, yet entails risk. While high performanceis possible with both approaches, a precariousadvantage produces much more performance vari-ation than a robust advantage. Moreover, it islikely that a precarious advantage is less durablebecause the high level of weaknesses makes thefirm vulnerable to rivals’ attacks. Research in com-petitive dynamics suggests that rivals are morelikely to attack when they perceive a higherprobability of success (Chen et al., 2007; Hittet al., 2009). Therefore, firms holding a precariousadvantage could benefit by directing profits towardthe elimination of weaknesses, thereby reducingtheir vulnerability to attacks and prolonging theiradvantage.

In total, these results underscore the importanceof having a more complete understanding of thebases of competitive advantage. There is moreto realizing a competitive advantage—even tem-porary advantage—than most previous researchhas suggested. Instead of focusing exclusively onstrengths, our research suggests that the focusshould expand. Specifically, sets of strengths, setsof weaknesses, and their integration should beconsidered. Incorporating weaknesses along withstrengths enhances the efficacy of RBV logicwith respect to competitive advantage. Moreover,as Figure 3 shows, the contours of the responsesurface suggest a dynamic relationship betweenstrength and weakness sets. For example, decreas-ing weakness is positive when a firm possesseslow levels of strength, but negative (or neutral,because the coefficients for the precarious androbust advantages are not statistically different)when possessing high levels of strength. This indi-cates the value of understanding how strength andweakness sets affect performance in a range ofconditions.



Understanding dynamics relatedto the durability of competitive advantage

To understand the dynamics related to the dura-bility of advantages, we also examined how envi-ronmental and firm-based factors affect changesin the capability sets. Through this, we studiedexogenous and endogenous antecedents of tem-porary advantage. Theory suggests that prior per-formance and environmental munificence likelyinfluence change in a firm’s strength and weak-ness sets over time. Understanding these influencesis important because continuous investments byrivals, even indirect rivals, in the developmentof their capabilities results in a dynamic market-place where current competitive advantages can belost.

First, the results show that initial levels ofstrength and weakness sets are strong indicatorsof change in a firm’s strength and weakness setsover time. Thus, a strong parity effect occursamong rivals. Firms put forth significant effortto gain ground on major competitors by increas-ing their own capability strengths and eliminat-ing capability weaknesses. If these variables werenot significant predictors, it would suggest thatan advantage hierarchy is present and resistant tochange, but this is not the case. Also, review-ing the distribution of strength and weakness setsacross time shows that while individual firmsjockey, either losing or gaining advantage as theyadjust their capabilities, the overall distribution ofadvantage in the market space remains stable. Inother words, competing in an age of temporaryadvantage does not mean competitive advantage isimpossible; rather it means maintaining a differen-tiated stock of capabilities is difficult in dynamicmarkets.

In order to create a series of temporarycompetitive advantages, firms need to invest incapabilities that help them maintain/improve theirrelative positions. The results suggest that inmunificent environments, firms are either (1) notaware or (2) not motivated to respond to rivalstaking actions to reduce their weakness sets, butthat they are (1) aware of, (2) motivated to, and(3) capable of matching investments (becauseresources are openly available) in order to increasestrength sets. Thus, munificent environments allowfirms to reduce weaknesses, but are less likelyto facilitate strength gains relative to the com-petitors. As such, a Red Queen effect (Barnett

and McKendrick, 2004) exists for strengths, butnot for weaknesses, when resources are highlyaccessible.

Next, the results show that higher performanceleads to positive changes in both strength andweakness sets over time. Firm-generated resourceshelp firms avoid the erosion of competitive advan-tage. Or for firms with a precarious advantage,directing firm-generated resources to the elimi-nation of weaknesses may be most important.Because these firms are vulnerable to rivals’ at-tacks, reducing weaknesses is important. Missingthis opportunity could lead these firms from a pre-carious advantage to a position where weaknessesundermine their strengths. These results, togetherwith the results pertaining to munificent environ-ments, may help explain why firms get trappedin death spirals (e.g., Arend, 2008; Ferrier et al.,2002; Morrow et al., 2007). Decreasing perfor-mance and low munificence eliminate an abilityto build strengths or eliminate weaknesses. Firmsfacing these conditions are unlikely to overcometheir competitive disadvantages.

Collectively, these results improve our under-standing of factors that influence changes in thebases of competitive advantage and its durability.Moreover, these results offer compelling evidenceof the mechanisms that propel an age of temporaryadvantage.

Limitations and future research

Similar to most research, this study has limita-tions, many of which provide direction for futureresearch. First, our data on capabilities are percep-tual in nature. Although perceptual measures arecommonly used to measure capabilities (e.g., Cooland Henderson, 1998; Danneels, 2008) and priorresearch has shown that they often demonstratesufficient validity (Wall et al., 2004) like ours do,11

we cannot be certain that our results are not influ-enced by perceptual filters. As such, additionalresearch based on objective measures would beuseful. Second, this research represents an exten-sion of prior work that commonly focuses on oneor two capabilities. Nonetheless, the six capabil-ities we measured, while a significant increase,do not represent all firm capabilities. Continuedexpansion of the capabilities considered would beuseful.

11 Please see Footnotes 6 and 7.



Third, our examination of the complementaritiesembedded in the firm’s strength and weakness setsfocused on synergistic relationships. However, itis possible that such relationships could make thefirm susceptible to quick erosion of its competi-tive advantage if a strength becomes obsolete, forexample. In fact, such a situation might mimic theresults reported in Cell III of Figure 2, where capa-bility weaknesses overwhelm the firm’s strengthsand led to poor performance.

In fact, investigating other combinations ofstrengths and weaknesses in more detail mightprovide additional contributions to the literature.Specifically, our results demonstrated that com-bining high strength and high weakness sets isa rewarding, yet risky, option. Future researchcould help us understand the vulnerability cre-ated by this combination. Alternatively, futureresearch could investigate how firm strategy orstructure can leverage this particular combinationfor optimal performance. Specifically, studyingthe effects of various types of interdependence(pooled, sequential, or reciprocal) among strengthsand weaknesses could offer significant contribu-tions to our understanding of how firms opti-mize performance. Additionally, investigating howcapability strengths and weaknesses affect thefirm’s alliance strategy could prove useful. Also,future research that examines how the charac-teristics of the firm’s competitive actions (e.g.,aggressiveness, response speed, competitive reper-toire simplicity, unpredictability, deviance, etc.)mediate the capability-performance relationshipwould be very useful. Or, research might exam-ine how a firm’s strength and weakness setsinfluence the use of these competitive behaviors.These considerations could further our understand-ing of, for example, mutual forbearance, and howcapabilities influence competitive dynamics moregenerally.

Fourth, while we emphasized competitive inter-dependence in our study, we did not have dataon specific competitive actions or dyadic compet-itive interactions. Instead, we directed attentionto the dynamic interdependence among multiplerivals by focusing on their sets of capabilities.Based on prior research, we assume that specificactions are based on the firm’s capability sets,which is consistent with the RBV and compet-itive dynamics literature (Barney, 1991; Grimmand Smith, 1997). Future research could make a

contribution by combining the logics of capabili-ties and competitive dynamics to investigate howsimilarity in firms’ strength and weakness setsaffects competitive tension (Chen et al., 2007).12

In fact, using the awareness-motivation-capabilityframework to specifically contrast how weaknessesaffect interfirm dynamics based on multimarketoverlap or commonality (Chen, 1996; Gimeno,1999; Gimeno and Woo, 1996) could be infor-mative. Our results imply that similar capabilityweaknesses may influence the intensity of com-petitive tension between rivals differently than dosimilar capability strengths.

Fifth, our results pertaining to changes in firms’strength and weakness sets over time suggest thatthe durability of competitive advantage is limited.However, because the linkage in this study is indi-rect, research is needed to show how changes incapability sets over time directly affect the firm’sability to gain or sustain a competitive advantage.Specifically, our aggregation of multiple capabilitystrengths does not allow for fine-grained analy-ses of the influence each specific capability hason a firm’s outcomes. More direct tests of theeffects of changes in the composition of capabilitystrength/weakness sets over time on firm perfor-mance are needed.

Finally, our study examined manufacturingfirms; research should be extended to examinethe capabilities of service firms to determine ifthe results of this study generalize to otherindustries.

Managerial implications

This work provides several implications for man-agers. First, configurations of capabilities must beeffectively managed. In particular, managers mustmanage both capability weaknesses and strengthseffectively. Developing a temporary advantage isnot just about protecting or creating strengths, butalso addressing weaknesses. In fact, reducing oreliminating capability weaknesses may be an expe-dient route to higher firm performance. Theorysuggests that achieving parity with rivals is lessdifficult than surpassing the better competitors inan industry. Thus, to improve performance, man-agers could, perhaps, first concentrate on reduc-ing weaknesses, which might allow their firm

12 We thank a very insightful reviewer for offering us thiscomment.



to achieve performance parity. While the resultsprovide partial empirical support for Powell’s(2001, 2002) suggestion that a primary strategicdifference among firms is not strength, but weak-ness, they also suggest that improving capabilitystrength is important. In fact, the highest perform-ers in our sample possessed high levels of strengthand these strengths provided an increasingly posi-tive effect on performance.

It is also important for managers to recog-nize that current strengths could become tomor-row’s weaknesses if investment is not sustained.In dynamic environments, maintaining a relativeposition requires managers to enhance their firm’scapabilities (Sirmon et al., 2007). Rivals are con-tinuously searching for ways to overcome theircompetitors’ competitive advantage. As such, allfirms must continuously develop their capabilities,requiring ongoing investments (Kor and Mahoney,2005) and emphasis on effective management (Sir-mon et al., 2007).

CONCLUSION

This research endeavored to increase our under-standing of the bases of competitive advantagesas well as the dynamics related to the durabilityof such advantages. The study focused on capa-bility strengths and capability weaknesses. Ourapproach addresses several gaps in current theo-retical approaches, especially prior work on theRBV, and opens promising opportunities for futureresearch on temporary advantage. The results ofthis research enable scholars to more effectivelydetermine how and why some firms achieve com-petitive advantage and superior relative perfor-mance, as well as uncover the factors that affectthe temporal nature of that advantage.

ACKNOWLEDGEMENTS

We acknowledge the helpful comments fromThomas C. Powell, Rodolphe Durand, Richard J.Arend, Erwin Danneels, Ramona L. Paetzold andparticipants at a research seminar held at the Uni-versity of Washington during the development ofthis research. An earlier version of this article waspresented at the Academy of Management’s annualconference. We thank the Banque de France forproviding the data.

REFERENCES

Arend RJ. 2003. Revisiting the logical and researchconsiderations of competitive advantage. StrategicManagement Journal 24(3): 279–284.

Arend RJ. 2004. The definition of strategic liabilities,and their impact on firm performance. Journal ofManagement Studies 41(6): 1003–1027.

Arend RJ. 2008. Differences in RBV strategic factors andthe need to consider opposing factors in turnaroundoutcomes. Managerial and Decision Economics 29(4):337–355.

Armstrong CE, Shimizu K. 2007. A review of approachesto empirical research on the resource-based view of thefirm. Journal of Management 33(6): 959–986.

Barnett WP, McKendrick DG. 2004. Why are someorganizations more competitive than others? Evidencefrom a changing global market. Administrative ScienceQuarterly 49(4): 535–571.

Barney JB. 1991. Firm resources and sustained com-petitive advantage. Journal of Management 17(1):99–120.

Bell DE, Leamon A. 1998. Costco Companies, Inc.Harvard Business School Publishing: Boston, MA.

Bettis RA, Hitt MA. 1995. The new competitivelandscape. Strategic Management Journal 16(S1):7–19.

Black JA, Boal KB. 1994. Strategic resources: traits,configurations and paths to sustainable competitiveadvantage. Strategic Management Journal 15(S2):131–148.

Carmeli A, Tishler A. 2004. The relationship betweenintangible organizational elements and organizationalperformance. Strategic Management Journal 25(13):1257–1278.

Castanias RP, Helfat CE. 2001. The managerial rentsmodel: theory and empirical results. Journal ofManagement 27(6): 661–678.

Castrogiovanni GJ. 1991. Environmental munificence:a theoretical assessment. Academy of ManagementReview 16(3): 542–565.

Chen MJ. 1996. Competitor analysis and interfirmrivalry: toward a theoretical integration, Academy ofManagement Review 21(1): 100–134.

Chen MJ, Miller D. 1994. Competitive attack, retaliationand performance: an expectancy-valence framework.Strategic Management Journal 15(2): 85–93.

Chen MJ, Su KH, Tsai W. 2007. Competitive ten-sion: the awareness-motivation-capability perspective.Academy of Management Journal 50(1): 101–118.

Cool K, Henderson J. 1998. Power and firm profitabilityin supply chains: French manufacturing industryin 1993. Strategic Management Journal 19(10):909–926.

Crook TR, Ketchen DJ, Combs JG, Todd SY. 2008.Strategic resources and performance: a meta-analysis.Strategic Management Journal 29(11): 1141–1154.

Danneels E. 2008. Organizational antecedents of secondorder competences. Strategic Management Journal29(5): 519–543.



D’Aveni RA. 1994. Hypercompetition: Managing theDynamics of Strategic Maneuvering. Free Press: NewYork.

Deephouse DL. 1999. To be different, or to be thesame? It’s a question (and theory) of strategic balance.Strategic Management Journal 20(2): 147–166.

Dierickx I, Cool K. 1989. Asset stock accumulation andsustainability of competitive advantage. ManagementScience 35(12): 1504–1511.

Durand R. 2002. Competitive advantages exist: a critiqueof Powell. Strategic Management Journal 23(9):867–872.

Ferrier W. 2001. Navigating the competitive landscape:the drivers and consequences of competitive aggres-siveness. Academy of Management Journal 44(4):858–877.

Ferrier W, Fhionnlaoich CM, Smith KG, Grimm CM.2002. The impact of performance distress onaggressive competitive behavior: a reconciliationof conflicting views. Managerial and DecisionEconomics 23(4–5): 301–316.

Ferrier W, Lyon D. 2004. Competitive repertoire sim-plicity and firm performance: the moderating role oftop management team heterogeneity. Managerial andDecision Economics 25(6–7): 317–327.

Ferrier W, Smith KG, Grimm CM. 1999. The role ofcompetitive action in market share erosion andindustry dethronement: a study of industry leaders andchallengers. Academy of Management Journal 42(4):372–388.

George G. 2005. Slack resource and the performance ofprivately held firms. Academy of Management Journal48(4): 661–676.

Gimeno J. 1999. Reciprocal threats in multimarketrivalry: staking out ‘spheres of influence’ in the U.S.airline industry. Strategic Management Journal 20(2):101–128.

Gimeno J, Woo C. 1996. Hypercompetition in amultimarket environment: the role of strategicsimilarity and multimarket contact in competitive de-escalation. Organization Science 7(3): 322–341.

Greene WH. 2003. Econometric Analysis (5th edn).Prentice-Hall: Upper Saddle River, NJ.

Greve H. 2003. A behavioral theory of R&D expendituresand innovations: evidence from shipbuilding. Academyof Management Journal 46(6): 685–702.

Grimm CM, Smith KG. 1997. Strategy as Action:Industry Rivalry and Coordination. South-WesternCollege Publishing: Cincinnati, OH.

Hambrick DC, Mason PA. 1984. Upper echelons: theorganization as a reflection of its top managers.Academy of Management Review 9(2): 193–206.

Helfat CE, Peteraf MA. 2003. The dynamic resource-based view: capabilities life cycles. StrategicManagement Journal 24(10): 997–1010.

Hitt MA, Bierman L, Shimizu K, Kochhar R. 2001.Direct and moderating effects of human capital onstrategy and performance in professional servicefirms: a resource-based perspective. Academy ofManagement Journal 44(1): 13–28.

Hitt MA, Ireland RD, Hoskisson RE. 2009. StrategicManagement: Competitiveness and Globalization.Cengage: Mason, OH.

Hoskisson RE, Cannella AA, Tihanyi L, Faraci R. 2004.Asset restructuring and business group affiliation inFrench civil law countries. Strategic ManagementJournal 25(6): 525–539.

Jacobides MG, Winter SG. 2005. The coevolutionof capabilities and transaction costs: explainingthe institutional structure of production. StrategicManagement Journal 26(5): 395–410.

Knight D, Pearce CL, Smith KG, Olian JD, Sims HP,Smith KA, Flood P. 1999. Top management teamdiversity, group process, and strategic consensus.Strategic Management Journal 20(5): 445–465.

Kor YY, Mahoney JT. 2005. How dynamics, manage-ment, and governance of resource deployments influ-ence firm-level performance. Strategic ManagementJournal 26(5): 489–496.

Leonard-Barton D. 1992. Core capabilities and corerigidities: a paradox in managing new productdevelopment. Strategic Management Journal 13(S1):111–125.

Levinthal DA, March JG. 1993. The myopia of learning.Strategic Management Journal 14(S2): 95–112.

Lieberman MV, Dhawan R. 2005. Assessing the resourcebase of Japanese and U.S. auto producers: a stochasticfrontier production approach. Management Science51(7): 1060–1075.

Milgrom P, Roberts J. 1995. Complementarities andfit: strategy, structure, and organizational change inmanufacturing. Journal of Accounting and Economics19(2–3): 179–208.

Miller D, Chen MJ. 1996. The simplicity of competitiverepertoires: an empirical analysis. Strategic Manage-ment Journal 17(6): 419–439.

Miller D, Eisenstat R, Foote N. 2002. Strategy from theinside out: building capability-creating organizations.California Management Review 44(3): 37–54.

Montgomery CA. 1995. Of diamonds and rust: a newlook at resources. In Resource-Based and EvolutionaryTheories of the Firm, Montgomery CA (ed). KluwerAcademic Publishers: Norwell, MA; 251–268.

Morrow JL, Sirmon DG, Hitt MA, Holcomb TR. 2007.Creating value in the face of declining performance:firm strategies and organizational recovery. StrategicManagement Journal 28(3): 271–283.

Mueller DC. 1986. Profit in the Long Run. CambridgeUniversity Press: Cambridge, U.K.

Ndofor HA, Sirmon DG, He X. 2011. Firm resources,competitive actions, and performance: investigatinga mediated model with evidence from the In-vitrodiagnostics industry. Strategic Management Journal .Forthcoming.

Newbert SL. 2008. Value, rareness, competitive advan-tage, and performance: a conceptual-level empiricalinvestigation of the resource-based view of the firm.Strategic Management Journal 29(7): 745–768.

Penrose ET. 1959. The Theory of the Growth of the Firm.John Wiley: New York.


Carolina

Highlight

Carolina

Highlight

Carolina

Highlight

Carolina

Highlight


Peteraf MA, Barney JB. 2003. Unraveling the resource-based tangle. Managerial and Decision Economics24(4): 309–323.

Porter ME. 1985. Competitive Advantage: Creating andSustaining Superior Performance. Free Press: NewYork.

Powell TC. 2001. Competitive advantage: logical andphilosophical considerations. Strategic ManagementJournal 22(9): 875–888.

Powell TC. 2002. The philosophy of strategy. StrategicManagement Journal 23(9): 873–880.

Powell TC. 2003. Strategy without ontology. StrategicManagement Journal 24(3): 285–291.

Priem RL. 2007. A customer perspective on valuecreation. Academy of Management Review 32(1):219–235.

Priem RL, Butler JE. 2001. Is the resource-based view auseful perspective for strategic management research?Academy of Management Review 26(1): 22–40.

Ray G, Barney JB, Muhanna WA. 2004. Capabilities,business processes, and competitive advantage:choosing the dependent variable in empirical testsof the resource-based view. Strategic ManagementJournal 25(1): 23–37.

Rumelt RP. 1984. Toward a strategic theory of the firm.In Competitive Strategic Management , Lamb R (ed).Prentice-Hall: Englewood Cliffs, NJ; 556–570.

Schendel DE 1994. Strategy: search for new paradigms.Strategic Management Journal 15(S2): 1–4.

Selznick P. 1957. Leadership in Administration: ASociological Perspective. Harper and Row: New York.

Simons T, Pelled LH, Smith KA. 1999. Making use ofdifference: diversity, debate and decision comprehen-siveness in top management teams. Academy of Man-agement Journal 42(6): 662–673.

Sirmon DG, Gove S, Hitt MA. 2008. Resource manage-ment in dyadic competitive rivalry: the effects ofresource bundling and deployment. Academy of Man-agement Journal 51(5): 918–935.

Sirmon DG, Hitt MA. 2003. Managing resources: linkingunique resources, management and wealth creation infamily firms. Entrepreneurship Theory and Practice27(4): 339–358.

Sirmon DG, Hitt, MA. 2009. Contingencies withindynamic managerial capabilities: interdependent ef-fects of resource investment and deployment on firmperformance. Strategic Management Journal 30(13):1375–1394.

Sirmon DG, Hitt MA, Ireland RD. 2007. Managing firmresources in dynamic environments to create value:looking inside the black box. Academy of ManagementReview 32(1): 273–292.

Sirmon DG, Hitt MA, Ireland RD, Gilbert BA. 2011.Resource orchestration to create competitive advan-tage: breadth, depth, and life cycle effects. Journal ofManagement. Forthcoming.

Smith KG, Ferrier WJ, Ndofor H. 2001. Competitivedynamics research. In Handbook of Strategic Man-agement , Hitt MA, Freeman RE, Harrison JR (eds).Blackwell Publishers: Oxford, U.K.; 315–361.

Stieglitz N, Heine K. 2007. Innovations and the role ofcomplementarities in a strategic theory of the firm.Strategic Management Journal 28(1): 1–15.

Sutcliffe KM. 1994. What executives notice: accurateperceptions in top management teams. Academy ofManagement Journal 37(5): 1360–1378.

Thomas LG. 1996. Dynamic resourcefulness and thehypercompetitive shift. Organization Science 7(3):221–242.

Thompson JD. 1967. Organizations in Action. McGraw-Hill: New York.

Wall TD, Michie J, Patterson M, Wood SJ, Sheehan M,Clegg CW, West M. 2004. On the validity of sub-jective measures of company performance. PersonnelPsychology 57(1): 95–118.

Waring GF. 1996. Industry differences in the persistenceof firm-specific returns. American Economic Review8(5): 1253–1265.

Wernerfelt B. 1984. A resource-based view of the firm.Strategic Management Journal 5(2): 171–180.

Wernerfelt B, Karnani A. 1987. Competitive strategyunder uncertainty. Strategic Management Journal8(2): 187–194.

West GP, DeCastro J. 2001. The Achilles heel offirm strategy: resource weakness and distinctiveinadequacies. Journal of Management Studies 38(3):417–442.

Wiggins RR, Ruefli TW. 2002. Sustained competitiveadvantage: temporal dynamics and the incidenceand persistence of superior economic performance.Organization Science 13(1): 82–105.

Wiggins RR, Ruefli TW. 2005. Schumpeter’s ghost: ishypercompetition making the best of times shorter?Strategic Management Journal 26(10): 887–911.



APPENDIX 1. DISTRIBUTION OF FIRMS ACROSS INDUSTRIES

NAF Industry Name (2-digit) Industry code(3-digit)

Number ofobservations

Average #of employees

Food and beverage industries 151 105 108.3152 19 117.9153 21 110.0155 28 142.9156 16 69.0157 20 131.9158 98 98.9159 40 78.5

Textile industries 171 6 72.7172 21 114.3173 16 86.9174 27 77.9175 31 94.5177 13 199.4

Manufacture of wearing apparel 182 91 74.0Manufacture of leather products and footwear 191 10 51.8

192 10 59.6193 22 135.2

Manufacture of wood and wood products 201 42 65.2202 14 148.9203 36 91.6204 40 68.9205 11 63.2

Manufacture of pulp, paper, and paperboard 211 14 14.0212 91 91.0

Publishing, printing, and recorded media 221 5 143.2222 128 68.6

Chemical industry 241 44 117.5243 28 99.5244 23 119.7245 36 71.2246 24 84.7

Manufacture of plastic and rubber products 251 21 107.8252 209 101.3

Manufacture of other nonmetallic mineral products 261 29 132.0262 18 108.9263 5 65.2266 42 76.5267 16 59.8268 5 145.0

Metallurgy 271 5 254.6272 8 178.3273 12 68.8274 12 148.4275 31 183.6

Manufacture of fabricated metal products 281 96 83.3282 16 60.3283 78 62.9284 75 112.7285 218 65.5286 59 107.3287 75 106.3



APPENDIX 1. (CONTINUED)

NAF Industry Name (2-digit) Industry code(3-digit)

Number ofobservations

Average # ofemployees

Manufacture of machinery and equipment 291 48 105.9292 80 105.7293 34 75.1294 29 75.4295 96 102.8

Manufacture of electric machineries and equipments 311 22 120.3312 20 106.7313 8 195.9314 5 44.8315 26 53.8316 16 98.1

Manufacture of radio and television equipment 321 38 144.9322 24 118.6

Manufacture of optical and photographic equipment 331 36 93.4332 33 112.6333 20 73.8334 18 93.4

Manufacture of motor vehicles 342 37 107.5343 32 165.8

Manufacture of other transport equipment 351 19 101.6353 15 245.6

Manufacture of furniture 361 106 88.3362 15 82.7364 11 138.9365 9 75.3366 23 82.9

Averages 38.2 104.0


Copyright of Strategic Management Journal is the property of John Wiley & Sons, Inc. and its content may not

be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.

the dynamic interplay of capability strengths

Documents

weakness sets

temporary advantage

high weakness

erosion of advantage

high strengthlowweakness

weakness sets

whilestrength sets

high levels