strategic resources: traits, configurations and …kimboal.ba.ttu.edu/selected writings/strategic...

Strategic Management Journal, Vol. 15, 131-148 (1994)

STRATEGIC RESOURCES:TRAITS, CONFIGURATIONS AND PATHS TOSUSTAINABLE COMPETITIVE ADVANTAGEJANICE A. BLACK and KIMBERLY B. BOALCollege of Business Administration, Texas Tech University, Lubbock, Texas, U,S.A.

The resource-based view (RBV) of the firm holds that certain assets with certain characteristicswill lead to sustainable competitive advantage. All the traits are required to be present toresult in sustainable competitive advantage. Such a trait approach overlooks the dynamicsof the creation of firm resources especially the strategically important factors as identifiedhv the resource based view theory. We propose that the resources are made up of factornetworks which have specific interfactor and inter-resource relationships that result in thecharacteristic traits being evidenced. These strategic resource factor relationships includenetwork type, available substitutes and cogency relationships (compensatory, enhancing andsuppressing.) Specific configurations that lead to high or very high support of sustainablecompetitive advantage are proposed. Twenty-two specific paths to sustainable competitiveadvantage for a factor, contingent on resource factor traits and relationship configurations,arc proposed. The implications, upon confirmation of these configurations, are discussed.

Positioning the firm for a sustainable competi-tive advantage by utilizing the firm's strengthsto exploit opportunities and neutralize threatswhile avoiding or fixing weaknesses has longserved as the core framework for formulatingthe firm's strategy (Learned et ai. 1965), This'SWOT analysis benefited greatly from theinsights of industrial organizational economics,especially the work of Porter (1980; 1985).Porter's famous 5 forces model emphasizedanalyzing industry structure to assess the rentearning potential of the industry based on entryand exit barriers. While Porter's frameworkhas provided many useful insights to bothpractitioners and researchers by concentratingon the external 'OT" side of the analysis,it nonetheless suffers from several significantproblems.

Key words: Resource-based view of the firm, net-works, systems, strategic configurations, strategicresources

First, it runs the risk of being tautological,i.e., it posits that firms in attractive industriesare successful. They are successful because theyare in attractive industries. A second, but moreimportant limitation of this framework is pointedout by Porter (1991) himself. The framework isconcerned with the cross-sectional problem andnot the longitudinal problem. The cross-sectionalproblem focuses on what makes some industries,and some positions within them more attractive.It docs not address why some firms are able toget into advantageous positions in the first place,and why some firms are able to sustain thesepositions and others are not. A third limitationstems from the implicit advice it gives to managersfor formulating strategy. McWilliams and Smart(1993) point out that it misdirects managers tofocus on industry level characteristics, encourag-ing them to expend resources on influencing theindustry's structure even though their firm willnot uniquely benefit from the changes, thusallowing competitors to free ride on the firm's

CCC 0143-2095/94/090131-18© 1994 by John Wiley & Sons. Ltd.

132 /. A. Black and K. B. Baal

expenditures. One could possibly justify this ifit could be shown that industry structure was thedominant determinant of firm performance.Recent evidence however. (Rumelt. 1991; Roque-bert. Phillips, and Duran, 1993). suggests that,at best, industry structure accounts for 8-15percent of variance in firm performance. Further-more, strategies based on market power can bedisastrous to the firm. For example. Carr (1993)in his analysis of the vehicle components industryfound that firms utilizing a market powerbased strategy significantly underperformed, onmultiple performance measures, their competitorswho followed a re source-based strategy.

These limitations lead many (e.g.. Barney,1986. 1991; Grant. 1991) to argue that strategyformulation starts properly, not with an assess-ment of the organization's external environment,but with an assessment of the organization'sresources, capabilities, and core competencies.This resource based view (RBV) of the firmapproach which emphasizes the internal side of*SWOT' analysis to strategy formulation is gainingin popularity among strategy theorists (Reed andDeFillipi. 1990; Summer et al., 1990; Meyer.1991; Porter, 1991; Peteraf, 1993; Barney, 1991.1992). This emerging framework already contrib-utes some promising insight into conditionsleading to sustainable competitive advantages(Conner. 1991; Grant. 1991; Peteraf. 1993). RBVtheory notes that differences in firm resourceswill lead to differences in sustainable competitiveadvantage (SCA).

Porter argues that RBV theory also runs therisk of being tautological, 'Successful firms aresuccessful because they have unique resources.They should nurture these resources to besuccessful (1991: 108)." Barney (1993) points outthat the independent variables of RBV aredefined at one level of analysis (the levelof resources or bundles of resources) whilecompetitive advantage, not economic rent perse, (the dependent variable) is at a differentlevel of analysis (the level of strategies that thefirm is pursuing). In essence, the independentvariable is at the functional level and thedependent variable is either at the business orcorporate level. This eliminates the charge ofRBV being tautological.

Bromiley (1993) notes that RBV theoryrequires some concrete definitions of resourcesthat is less than 'anything that leads to perform-

ance." There is acknowledged difficulty indetermining a priori what firm resources mightlead to a sustainable competitive advantagegiven the inherent uncertainty of the externalenvironment (Peteraf, 1993; Fiol. 1991). Nonethe-less. Bromiley's (1993) call for the operationali-zation of RBV theory is well taken.

Jay Barney begins to address this issue byidentifying the needed characteristics of firmresources and presents this in the VRIO frame-work (Barney and Griffin. 1992; Barney. 1992).This framework assesses the economic perform-ance imphcations of resources by evaluating theresources for the characteristics of value, rareness,inimitability and organizational orientation(VRIO).

The value of a resource will be dependentupon the firm's combination of resources andthe path that the firm is following. In otherwords, value is the fit of the resource or factorto strategy combined with the fit of the strategyto the external environment. The rareness of aspecific resource depends upon the combinationof physical rareness in the factor market and/orthe rareness of the perceived value of theresource due to a firm's particular resourcecombination. Inimitability is the continuation ofimperfect factor markets via restricted infor-mation, the cost of recreating the specificcombination of resources that give a synergisticresult, or a combination of the two. Substitut-ability rests on the continuation of imperfectfactor markets, the costs involved in the recreationof specific combination, or the cost of finding anew combination of resources that will enablethe firm to compete for the same product market(i.e., a new path with new requirements).Barney's framework combines inimitability andnonsubstitutability into one category by treatingnonsubstitutability as a specialized case of inimita-bility. Finally, the VRIO framework alsoexplicitly considers if the organization is orientedto utilize its strategic resources.

However, the VRIO framework, whileimplicitly acknowledging the importance of adynamic view, treats the evaluation of resourcesfrom a stand alone viewpoint ignoring howresources are nested in and configured with oneanother and the nature of relationships betweenthem. Thus while Barney talks about bundles ofresources, the VRIO framework treats resourcesas singular distinct items.

Strategic Resource Configurations 133

Another internal analysis that also reflects thisbundling problem is the Capital Asset PricingModel (CAPM). When CAPM is used to assessan investment decision, it considers a portfolioof investments as a cross-sectional point-in-timeissue similar to RBV theory and its bundle ofresources. CAPM assessments ignore whetheran investment in one project will affect theprofitability of other ongoing or potential projects(Oviatt, 1989). Rather than investigating theseinternal relationships it assumes away theseproblematic interactions. Furthermore, as Robinsnotes in his evaluation of the CAPM model, theexistence of firm specific capital raises 'seriousproblems in the use of the CAPM to estimate therisk associated with a capital project." (1992: 528)He points out that all firms have collateral assets(X-assets) that are firm specific but do not havedefined opportunity cost and do not exist inisolation from other tradeable assets. A firmsability to generate quasi-rents is a function of theinteraction between these X-assets and othermarketable assets that the firm possesses. Yet. theseX-assets (e.g., organizational routines, companypolicy, culture, etc.) will result in errors in theassessment of risk because of the unknown natureof that interaction. This example not only highlightsthe bundling problem but also points out theinadequacies of financial statements to accuratelydisplay asset value. This latter point is alsosupported by Hall as he too questions 'thesignificance of any quantification of shareholder"funds which does not recognize the value ofintangible assets" (1992: 135).

Although the use of teams or bundles ofresources have been acknowledged in RBVtheory (Grant. 1991; Dierickx and Cool. 1989;Fiol, 1991), most researchers do not address thedynamic aspects of bundling resources and theirimplications to RBV. Thus, while the RBVtheory runs the risk of evaluating and categorizingresources without reference to the system inwhich those resources are embedded, it has nottotally ignored interfactor relationships. Forexample, two such postulated relationships areTeece's (1986) discussion of cospecialized assets,and Amit and Schoemaker's (1993) notion ofcomplementarity relationships affecting the valueof a resource. However, these ideas of howresources within a firm interact with things bothinside and/or outside the firm to create sustainablecompetitive advantage need further development.

In this paper we operationalize RBV theoryby developing the network of relationshipsamong resources that is necessary for thecreation of the characteristic traits identifiedby RBV theorists that are needed to attain asustainable competitive advantage. The paperis organized as follows. The first section givesdefinitions of key RBV terms used in thispaper. The second section looks closer at thecontribution network theory makes to ourunderstanding of firm resources. The thirdsection clarifies the identification of RBVstrategic system resources. The fourth sectiondiscusses the potential strategic relationshipconfigurations in detail. The last section pre-sents implications from this expansion of theRBV theory and our conclusions.

DEFINITIONS OF FIRM RESOURCESAND RELATIONSHIPS

Given the relative youth of RBV theory, weexplain our usage of the key terms and concepts.By accepting resources as the basic unit ofanalysis (Grant, 1991) and constraining theresources of interest to strategic resources whichare those that are theoretically characterized asvaluable, rare, neither imitable and/or substitut-able, and which an organization is orientedtowards using (Barney, 1986, 1991; Dierickx andCool. 1989; Summer, et al.. 1990), we boundour area of interest.

RESOURCE FACTORS ANDCATEGORIES

Several researchers decompose firm resourcesinto combinations of resource "factors' or 'assets'(Barney, 1986; Dierickx and Cool, 1989; Porter,1991; Grant. 1991; Mahoney and Pandian. 1992).Since 'assets' implies something that is owned bya firm and factors include either ownership and/or control, we choose to use factors as theelements making up a resource. While specificrelationships between the factors that make upa particular resource have remained unexplored,general characteristics have been addressed.Bar-ney (1986) points out that resource factors differin their 'tradeability." A tradeable factor is onethat can be specifically identified and its monetary

134 J. A. Black and K. B. Boal

value determined via a 'strategic factor market."Tradeable factors" availability and monetaryvalue in the factor markets will reflect themarket's awareness of those factors' totalrareness (physical and/or particular use). Byimplication, a nontradeable factor will be firmspecific and will not directly have its monetaryvalue determined via that 'strategic factormarket.' As an example. Arrow has earlierbrought attention to the factor of 'trust' which,if purchased, immediately creates doubts in themind of the purchaser about what he/shepurchased. This is a prime example of somethingbuilt up over time that is valuable but nottradeable. He emphasizes this point in thefollowing:

Trust and similar value, loyalty, or truthtellingare examples of what an economist wouldcall "externalities."' They are goods, they arecommodities: they have real practical value;they increase the efficiency of the system, enableyou to produce more goods or more of whatevervalues you hold in high esteem. But they arenot commodities, for which trade on the openmarket is technically possible or even meaningful.(1974: 23).

Dierickx and Cool (1989) suggest that resourcesshould be differentiated as either asset flows orasset stocks. An asset flow is a firm resourcethat can be obtained or adjusted immediately.An asset stock is a firm resource which cannotbe adjusted immediately and which is built upover time from asset flows (Dierickx and Cool.1989). We name this aspect the AcquisitionProcess. Again since a company may utilize bothsomething it directly owns and something that itcontrols, we choose to use 'factor' in referringto the element that helps to create a resource.

The creation of a 2X2 matrix with thetwo dimensions of 'tradeability aspects' and'acquisition process aspects' enables us to lookat four factor types as described in Table I.Note that four general factors types are possible:tradeable asset flows, nontradeable asset flows,tradeable asset stocks, and nontradeable assetstocks. It is the bundling of these four types offactors that results in a particular resource.Thus resources can be viewed as a configurationor network of factors. This in turn implies thatthere will be specific relationships between thefactors.

RESOURCE CATEGORIES

Several resource level categorizations recentlyhave been presented in the literature. Barney(1991) groups all firm resources into threecategories: physical capital resources(Williamson, 1975). human capital resources(Becker, 1964) and organizational capitalresources (Tomer. 1987). Grant (1991) lists sixcategories of firm resources: financial, physical,human, technological, reputation, and organiza-tional. These categorical schemes appear to missthe key issue in the search for sustainablecompetitive advantage (SCA). The key issue forfirm resources, as regards to the creation andmaintenance of SCA. is based on the ability ofthe resource to generate rent. While rentgeneration's economic root is based on thedifferential between the expected rent and theactual rent attained by a resource (Barney, 1993)the bulk of the work rests also on implicationsof scarcity.

Scarcity is related to the ease of identificationof the bundle of factors that creates or is theresource. If the bundle is a relatively simple onewhich can be identified, then the ability toimitate or find substitutes is increased and theopportunity for rent generation is decreased(Grant, 1991) and vice versa. Thus a more usefulcategorization would be one that focuses on thisscarcity issue. Because of the link betweenscarcity and identification, we chose to categorizeresources on the degree to which the factors thatmake up the resource bundle can be identified.This categorization divides firm resources intotwo types: contained resources and systemresources.

Contained resources

A contained resource is comprised of an identifiedsimple network of resource factors that can bemonetarily valued. By network, we mean theconfiguration of factors, as well as, their relation-ships with each other that results in a particularfirm resource. We think that including bothfactors and their relationships is an importantdistinction, for the same reason that a list ofingredients from a recipe is not a cake. A cakerequires the ingredients plus their relationshipsamong them for a successful result. A simplenetwork is one with relatively few, mostly direct.,


links among a small number of factors. 'Simple,"thus, implies that the network has definiteboundaries. It is possible, then, to identifycontained resources and, once identified, tomonetarily value them (Barney. 1986; 1989).This supports Barney"s contention that both assetstocks and asset flows are tradeable either at theresource level or. barring that, at the factorlevel. This implies that if a factor is nontradeable,then, the nontradeable factor must have asubstitute that is tradeable or alternatively bebroken into component parts to allow marketvalue to be imputed to the resource. If thenontradeable factor has no tradeable substituteand can not be broken down into tradeablecomponents then, although the network it isembedded in may appear simple, the nestednontradeable factor implies that the network iscomplex.

When we compare this categorization toBarney's (1991) three categories, it is evidentthat since physical and some human resourcesare 'tradeable' (Barney, 1991; Williamson, 1975).contained resources will typically include them.Given the transparency of the simple network,contained resources are unlikely to directly leadto a SCA.

Two conditions may occur where they mightsupport a SCA. First, the contained resourceis disguised either by being overlooked bycompetitors or by being hidden or kept secretby the firm. In either case, since they are eithertradeable or have substitutes that are tradeable,they will be subject to quick erosion oncediscovered. Second, they might be only onefactor of a complex network which as a wholecreates the competency that supports SCA. Thusonly indirect support is possible in the secondcase.

System resources

A system resource is created by a complexnetwork of firm resource factors. A complexnetwork is one with many direct and indirectlinks between a large number of factors whichare made up of nested system resources,contained resources and other resource factors.'Complex' implies that the network doesn'thave definite boundaries which will makemonetary valuing implausible. Generally speak-ing, a system resource is socially created. As a

social creation, it is endowed with the impliedcreation and recreation of social constructs. Bycontrast, a contained resource is more similarto a discrete artifact. Indeed, it is the attributesof social creation that make the identificationof the complex network of system resourcesdifficult (Barney, 1992; Fiol. 1991). This impliesthat the complete set or even a significantnumber of factors will be neither readily noreasily identifiable. This poses problems inmonetarily valuing such resources. This alsoimplies that fewer of the factors will betradeable and that there are fewer substitutesfor factors or for the resource as a whole.When Barney's categories of resources (physicalresources, human resources and organizationalresources) are again evaluated, it becomesevident that all three categories could beinvolved in the network of a system resource.

In the case of a system resource, organizationsare faced with a resource comprised of acomplex network that makes an unknownamount of contribution to a capability that mayitself be a member of a network of capabilitieswhich lead to a set of competitive advantages.It is clear, that often, while attempting toreplicate themselves through autopoietic pro-cesses (Morgan. 1983; Smith, 1983), the initialidentification of a strategically important systemresource happens only when the resourceis unintentionally destroyed. For example,consider Xerox's initial decision not to developfurther computer hardware and software tech-nologies that later spawned Apple Computersand Microsoft. These firms were participantsin the start of major technological changes inwhich Xerox had given up their ability to createa competitive advantage. Ultimately, it resultedin a loss of competitive advantage in their chosenindustry as well. This inadvertent destruction ofthe support of sustainable competitive advan-tage can be likened to an internal Schumpeterianshock. In this instance the external environmentdidn't change but the internal environmentevolved and discarded the wrong genetic(resource) endowment. From our example, wecan infer that these internal choices may resultin a later external Schumpeterian shock. Tobegin to address the question of identifyingthe networks of competencies, the generalcharacteristics of networks need to be integratedwith RBV theory.


NETWORK THEORY AND THERESOURCE-BASED VIEW OF THEFIRM

An organization's unique set of assets is theresult of the relationships both within and acrossthe levels of factors, resources, and competencies.This results in two types of networks: localnetworks (McCallister and Fischer, 1983) andstructural networks (Berkowitz, 1982). As appliedto RBV theory, a local network is the configu-ration of relationships within a level of analysisas in among the factors, where it is the entirenetwork that results in a resource. The resourceis not merely the listing of its factors but is theinteraction configuration among the resourcefactors. Thus if Resource D is composed offactors A. B, and C; its local network consistsof all the existing relationships among A, B. andC. For example, one can think simplistically ofa unit's performance as a result of the interactionsamong the capacities of unit members (FactorA), the motivations present (Factor B), and theunit's physical and capital resources (Factor C)(Blumberg and Pringle, 1982).

A structural network is the configuration ofrelationships between local networks and betweena factor of a local network and other networksor factors. Again applying social network theoryto RBV theory, this is the configuration ofrelationships between the focal resource andother resources, as well as, the relationshipsbetween other resources and the factors of thefocal resource. For example, if one looks at theresource, a unit"s performance, as a single entity,it will have links to other resources and yet,individually, its factors (people's skills, attitudes,raw materials, etc.) will simultaneously also havelinks between resources and/or factors. It is theconfiguration of both of these sets of links thatcreate the resource's structural network. Thisstructural network will be especially dense fornontradeable factors such as 'trust", given Itamiand Roehi's (1987) observation about the simul-taneous use of intangible assets.

In summary, a resource's internal factor net-work is its local network (McCallister and Fischer,1983). and its relationship outside of its localnetwork is its structural network (Berkowitz,1982). One might think of the product or resourcethat is a department"s end result as the artifactof the local network and its place in a value

chain will reveal the structural network. Noticethat a factor of that product's network, themanager of that department, will also individuallyhave links to other networks (the chain ofcommand relationships). Sayles (1993) notesthat the widespread tactic of downsizing andeliminating middle managers may have a seriousimpact on the firm's ability to retain previouscompetencies. Given that middle managers playa crucial role in integrating and aligning com-petencies, the competency is destroyed in theletting go of the managers (Sayles. 1993). Thishappened due to a lack of understanding of theinter-resource relationships that make up thecompetency and results in further destructionof other competencies due to the structuralrelationships that were involved.

Since a competency includes system resourcesas factors, the competency's local network willinclude all its component resources' local net-works (since a resource is definitionally itslocal network). However, we do not expect acompetency's structural network to be just thesum of its resource's structural networks. Just asthe resource as a distinct entity has relationshipsto other resources, we also expect a competencywill have relationships with other competencies.For example, consider the case where G. M..under Roger Smith, spent 50 billion dollarsretooling and still wound up being the high costproducer. We argue that this occurred becauseplant and equipment are only one resource inthe competency (lean manufacturing) that theywere striving to attain and the weak link wasthe relationships needed with other resources,such as management, HRM systems, the supplychain, engineering, etc. (Womack. Jones, andRoos. 1990).

A competency's structural network will includeall of the factors' structural networks, as well as,the overall competency's external relationships.This hierarchical nesting relationship is believedto exist from the most nested single factorthrough the overall organization as a unit(Berkowitz, 1982). Figure 1 shows a pictorialrepresentation of the hierarchical nestings offactors, resources, and competencies.

Notice the dotted local network relationshiplines in the system resource linking ail thefactors together. Then, in the bottom networkpresentation, note that the system resource isimbedded in the competency's local network. As


NETWORK HIERARCHICALNESTING

SYSTEM NETWORK

COMPETENCY NETWORKwith embedded system network

LOCAL NETWORK-STRUCTURE NETWORK -

- RELATIONSHIPS WITH ANY OTHERFACTOR, RESOURCE OR COMPETENCY

Figure 1. Network hierarchical nesting

implied in Figure 1, the local networks and thestructural networks are not independent. Whilethis creates problems in understanding causalrelationships, it helps us to understand thecreation of organizational synergy. It providesfurther support of the idea of 'emergent powers

being created when some objects or individualsare internally related to each other to form astructure" (Tsoukas, 1989).

Given the potential complexity of the creationof firm resources, the RBV framework providesa useful heuristic for discriminating between

138 / . A. Black and K. B. Boal

situations of competitive parity, temporary com-petitive advantage or sustainable competitiveadvantage (Barney, 1992). Although RBV litera-ture has evaluated a resource as a discrete unit(Barney, 1992; Amit and Schoemaker, 1993;Grant, 1991). in the following section we showhow it also can enable us to look for relationshipsthat are believed to lead to high or very highsupport of sustainable competitive advantage. Inother words, to look for the relationships thatare resulting in the desired characteristics.

RBV implied strategic relationships

In looking for stategic relationships, we notefrom the literature that Schoemaker (1990)suggests that it is necessary to explore howsocially complex resource factors and resourcesmagnify or diminish each other, but he doesn"tspecify what form those relationships may take.In a similar vein. Grant (1991) points out theneed for examining intra and intercapabilityresource relationships. Likwise, Conner (1991)has noted a nested condition in regard to assetstocks and flows. Amit and Schoemaker (1993)suggest that these complexities create nestednessproblems across organizational level. Robins(1992) argues that it is these firm specificrelationships which generate quasi-rents since thetradeable factors (barring market inefficiencies)would have their value bid away.

It has been argued that the needed character-istics of firm resources to generate rents alsomake it unlikely if not impossible to be able todetermine a priori the set of resources neededto gain or sustain a competitive advantage(Barney, 1986, 1991; Peteraf, 1993; Robins.1992). Certainly, uncertainty in the externalenvironment contributes to this a priori determi-nation problem but equally important is theuncertainty in the internal environment due tothe relationships between the factors that makeup a resource, between resources that make upcompetencies and between the competencies thatare needed to follow a firm's strategy. Thenestedness problems may also point to importantrelationships across the levels, as well as. withinthem.

We propose that these relationships are impliedby the characteristics already identified as neededfor SCA. Amit and Schoemaker (1993) expandedupon Barney"s VRIO (1992) base characteristics.

Valuable was expanded to include the subdimen-sions of an external link of 'overlap with strategicindustry factors, and implied internal fit issue of"complementarity".' Rare was expanded to includescarcity and low tradeability. Inimitable wasbroken out into inimitability and limited substitut-ability. Finally Barney's 'O' from 'VRIO\organized to capitalize on the resource, wouldalign with Amit and Schoemaker's appropriabilityand durability characteristics.

The finer grained characteristics provided byAmit and Schoemaker (1993) do provide a betterscreening but the dynamic nature involved in thebundling of the factors and resources intocompetencies remains a mentioned but notintegrated feature. Our application of socialnetwork theory begins to unravel some of theissues of the dynamics. In our presentation, thechoice of a resource to be evaluated is a refiectionof the management's belief in the overlap of theresource with the relevant key industry factors(Amit and Schoemaker, 1993). We do not addressthis external link but turn our attention inwardsto the determination of just what it is that causesa resource to exhibit the rest of the neededcharacteristics. The issues of tradeability. dura-bility and possibly appropriability are reflectionsof the 2X2 Tradeability and Acquisition Processmatrix of factor types. The issues of perceivedscarcity and inimitability are reflections of thecomplexity of the network. While substitutabilityramifications may also be a reflection of thecomplexity of the network, they also may mitigatethat complexity and so substitutability standsas a separate relationship. The internal valuerepresented by complementarity, appropriabilityand possibly durability are the reflections of whatwe term cogency relationships. These additionaldynamic and key relationships are built onSchoemaker (1990). Amit and Schoemaker(1993). and Robins (1992). We suggest thatcogency relationships have three forms: com-pensatory, enhancing. and suppressing/destroying.

A compensatory relationship exists when achange in the level of one resource is offset bya change in the level of another resource. Thisrelationship may be symmetric or asymmetric.Note that here we focus on changes in existingresources and not on replacing the existingresource. Furthermore, compensatory relation-ships are not equivalent to substitutability


relationships. For example, increased effort canmake up for differences in ability but it can'tsubstitute for lack of ability.

An enhancing relationship exists when thepresence of one factor magnifies the impact of adifferent factor. Amit and Schoemaker (1993)refer to this as complementarity. We do notthink that enhancing relationships require abilateral dependence as is implied in Amit andSchoemaker"s discussion of complementarity.That is. an enhancing relationship may also beunidirectional or asymmetric. Changes in Amagnify changes in B but not vice versa.For example. Magic Johnson, a famous U.S.basketball player, was noted for making otherplayers better; other players didn't make himbetter (some might suppress his ability butnot enhance it). Another example arises fromParthasarthy and Sethi's (1992) analysis of flexibleautomation use in manufacturing systems. Theyfound that when both scope and speed flexibilitieswere in place, along with a flexible automationsystem, then there was a positive significanteffect on performance. If there was only thepresence of either speed or scope flexibilitieswith the flexible automation, then there was nosignificant effect on performance. This illustratesthe enhancing relationship between the containedresource (speed and scope flexibilities) and thecontained resource (flexible automation system)but not compensatory relationship that result inthe positive performance differential.

Likewise, suppressing relationships exist whenthe presence of one factor diminishes the impactof another. The lack of ability on the part ofother players suppressing Magic Johnson"s playingwas previously acknowledged as a suppressingrelationship. Again using Parthasarthy and Sethi(1993), they found that a mechanistic structurehad a negative effect on the relationship betweenflexible automation and performance levels. Anextreme case of suppression would be thecomplete destruction ofthe resource. An examplewould be in the clash of cultures in mergingcompanies with the uhimate destructon of thesuppressed culture.

The strategic value of cogency relationships isproposed to be dependent upon specific intraand interresource inherited characteristic traitsand network and substitutability relationships.This view is consistent with the concept ofequifinality in open system theory and supports

the concept of multiple paths leading to SCAfor firms with heterogeneous resources. In otherwords, value is the fit of the factor (and itsrelationships to other factors in the resourcenetwork) to strategy combined with the fit ofthe strategy to the external environment. Indeed,value may not be determined for each resourcefactor individually but for an entire bundle. Forexample, if the ways of equally competing includeResources A, B, C, D, E, F, G, H and I but incertain combinations; then Firm 1 may useA + B + C, while Firm 2 may use A -I- D -I- E.At the same time. Firm 3 may use G + H + Iand Firm 4 may use B -i- E + F. Therefore,resource A has value for Firms 1 and 2, but notfor Firms 3 and 4. The value depends on whatother factors are present or controlled by thespecific firm in question. This implies that aresource factor as such may not be a substitutebut that the entire bundle or configuration ofthe resource factors may act as a substitute foranother and different bundle of resource factors.It is this aspect of configurational use of resourcefactors that allows firms to pursue similarcompetitive strategies with different resources.We believe, this also helps to explain why firmsfollowing different generic competitive strategiescan be equally successful (Conant, Mokwa andVaradarajan, 1990) and why there are noconsistent differences in performances betweenstrategic groups (Cool and Schendel, 1988).We present theoretical combinations of factorcharacteristics (or attributes) and relationships,as well as, potential benefits of these proposedheuristics. Later in this paper, we develop theserelationships more fully (See Figures 3-6). Beforefurther exploring the relationships amongresources, we need to identify the strategicsystem resource.

IDENTIFICATION OF STRATEGICSYSTEM RESOURCES

To summarize, a strategic system resource is asocially created complex network comprised oftradeable and nontradeable factor stocks andflows and their relationships with each other.While complexity may be desirable to confoundcompetitors, complexity makes it difficult for firmsto create, manage, exploit and nurture theirresources. Amit and Schoemaker (1993) highlight


the difficulty of making decisions about resourcedevelopment and deployment in the face ofuncertainty and complexity. Although the specificsof a system resource will be dependent upon itscontext, we develop a conceptual framework toexplain the key dimensions and the relationshipsbetween them in an effort to enlighten a firm asit copes with this issue and to highlight specificallythe implication of a network orientation to thebundling problem noted earlier.

System resource local network dimensions

In considering the composition of the systemresource network, we use the five strategicdimensions presented above. The dimensions arerelated to basic factor characteristics and theirinterfactor relationships. Recall that the fourbasic factor types are derived from the four cellmatrix utilizing Tradeability aspects (Barney,1986) and Acquisition Process aspects (Dierickxand Coo!. 1989). The applicable time dynamismissues for this factor are included in the Acqui-sition Process dimension (Dierickx and Cool.1989; Grant, 1991; Nelson and Winter, 1982;Porter, 1991). Recall also that the relevantrelationships derived from the characteristicsneeded to support SCA include Network Type(Dierickx and Cool, 1989; Grant, 1991), Substitut-

ability (Barney. 1991, 1992; Grant, 1991), andCogency (Schoemaker, 1990). The Cogencyrelationship has three subdimensions: Compensa-tory, Enhancing and Suppressing relationships.We submit that the relevant relationships canonly be considered when a proposed factorhas been identified. The specification of thesecharacteristics will enable a firm to determinehow much and what type of effort it will take tocreate and maintain that factor and uhimatelythe resource and from the resource to thecompetencies that enable it to achieve its SCA.

Based on the preceding, we now presentpotential configurations of firm resource factorsthat we propose are necessary to lead a systemresource to high or very high support ofcompetitive advantage. We intend for these pathsto be useful heuristics enabling both managersand researchers to address the incredible com-plexity and uncertainty that the socially createdresource inherently has.

POTENTIAL CONFIGURATIONS OFSYSTEM RESOURCE LOCALNETWORKS

The support of a sustained competitive advantageis proposed to be the result of the specific

INHERENT

RELATIONSHIPS

VERY HIGH

OR HIGH

SUPPORT OF

SUSTAINABLE

COMPETITIVE

ADVANTAGE

Figure 2. Strategic firm resource factor—inherent traits and relationships


combinations of the listed key dimensions. Thusstrategic resources will have networks with thefollowing configuration attributes. By using theRBV theory as previously presented by Barney(1992), the potential 256 combinations of inferredrelationships are reduced to only 22 strategicconfigurations that theoretically support a SCA.The specific configuration for a particular factorcan be identified by tracing a decision line through

the proposed strategically necessary relationships.While acknowledging that many of these questionshave a range of answers, for the sake of parsimony,these combinations are shown in the form ofdecision trees (See Figures 3-6) and are discussedbelow. To minimize redundancy, the logic for thepath choice is presented with the original pathpresentation. All subsequent presentations of thatdecision path are based on the original logic.

Tradeable asset flow

If a resource factor is a tradeable asset flow, itwill only provide support for SCA in the presenceof X-assets (Robins. 1992) (recall that an X-asset is the firm specific asset that allows theattainment of a quasi-rent). We propose threeconfigurations that can provide high support.They start with the requirements of the tradeableasset flow being a factor of a system resourcethat is in a complex network (Node 1, YES)where substitutes for the element are not available(Node 2, NO). These decisions address the issuesof rareness and time dependency by creatingambiguity in the factor's role in the firm resourceand impacting the amount of time it would takea competitor to imitate it. The firm's path tohigh support of SCA is attained when the factorhas a compensatory cogency relationship onlywith nontradeable factors (Node 3, NO; andNode 4, YES), because decisions 3 and 4 againdiminish the tradeability and fiow aspects byallowing it to be offset with a firm specific assetonly (Path a). The last two paths (b and c) donot have any compensatory relationships (Nodes3 and 4, NO) but have either an enhancing(Node 5, YES) or neutral cogency relationship(Nodes 5 and 6, NO) with any other system factor.We present the enhancing cogency relationship asa path since this type of a relationship mayincrease the particular factor's importance sinceit may be the only possible factor that canprovide that particular enhancing capability.

Nontradeable asset flow

A nontradeable asset flow factor can providehigh support if it is a member of a systemresource with a complex network (Node 1, YES).This context diminishes the problem for SCAfrom the titne dependency aspect of the factor.If it has substitutes (Node 2, YES), but thereare no compensatory relationships either withtradeable or nontradeable network factors (Nodes3 and 4. NO) and there is an enhancingrelationship (Node 5, YES; Path d), the factormay still lead to SCA due to the last relationshipworking to increase its specific importance to thenetwork. Note that this particular configurationmay be capable of only temporary high supportdue to the substitutability relationship, if thesubstitute also is capable of providing an enhanc-ing relationship.

Additional paths are postulated when this factorhas no substitutes (Node 2, NO). Depending uponwhether or not this factor has a compensatoryrelationship with other nontradeable factors, thereare three possible paths to SCA. If it does have acompensatory relationship with a nontradeablefactor (Node 4. YES), this is sufficient to lead tohigh support of SCA (Path e). If, however, it doesnot have any compensatory relationships (Nodes3 and 4, NO), it can lead to very high support ofSCA if it has an enhancing relationship (Node 5,YES; Path f). We believe that if the factor doesnot have an enhancing relationship it will stillsupport SCA if it also does not have any suppressingrelationships (Node 6. NO; Path g). A comparisonof Figures 3 and 4 suggests that nontradeable assetflows provide an additional path to SCA andwhere the paths are similar suggest that in somecases the magnitude of the effects may be greater.

Tradeable asset stock

When the factor under consideration is a tradeableasset stock, it will provide high support when itis a member of a complex network (Node 1,YES). This provides a starting point for assessingthe six paths that lead to high support of SCA.We will first discuss the two paths for whichsubstitutes exist (Node 2, YES). If the factor hasno compensatory relationships with other tradeablenetwork factors (Node 3. NO) but has a compensa-tory relationships with other nontradeable factors(Node 4, YES), that is sufficient to provide support


RESOURCE FACTOR ••TRADEABLE ASSET FLOW

POTENTIAL SUPPORT OFSUSTAINABLE COMPETITIVE

ADVANTAGE

YES

1. Is this factor a member of a complex network?

2. Oo substitutes exist for this factor?

3. Is this lactor in a compensatory relationship with atradeable network factor?

4. Is this factor in a compensatory relationship with anontradeable network factor?

# Decision Path to probable support of SCA

5. Is this factor in an enhancing relationship with another networkfactor?

6. Is this factor in a suppressing relationship with anothernetwork factor?

^ ^ Leads to either competiiive parity or competitivedisadvantage-

* This same pattern is followed for all decision nodes.

Figure 3. System resource local network configuration decision tree

of SCA (Path h). If, however, the factor hasno compensatory relationships with nontradeablenetwork factors (Node A, NO), it can still lead toSCA if it has an enhancing relationship with othernetwork factors (Node 5. YES; Path i). While itis possible that due to the availability of a substitutefor these two configurations to be imitated orentirely substituted for, the overall effect of theenhancing relationship, the time dependency ofthe stock and the complex network causingambiguity about the factor's role make it unlikely.

Returning to Node 2, if no substitutes exist(Node 2. NO) and even if this factor hascompensatory relationships with other networkfactors without regard to tradeability (either Nodes3 or 4, YES), it can still lead to at least highsupport of SCA (Path j). The compensatoryrelationship with a nontradeable network factor(Node 4, YES) magnifies the path effects to veryhigh support due to the complex network creatingambiguity about the factor and the offsettingabilities are tied to firm specific, or very rare,factors (Path k).

If no compensatory relationships exist (Node4. NO), two paths will stilt lead to SCA. In thefirst instance, if this factor has an enhancingrelationship with another network factor (Node5, YES), then it will lead to very high supportof SCA (Path 1). The more of these enhancingrelationships, the stronger the support of SCA.It will still provide support without an enhancingrelationship if it does not have a suppressingrelationship (Node 6, NO; Path m). This ispossible, because given the rest of the factors inthat local network, it is not likely that competitorswill be able to timely duplicate this resource dueto the time constraints in creating this tradeableasset stock and the complex network.

Nontradeabie asset stock

The strongest supporter for sustained competitiveadvantage is the nontradeable asset stock. Com-pared to tradeable asset stocks, nontradeableasset stocks provide the basis for three new pathsand the magnification of some path effects.


RESOURCE I-ACTOR--NON-THADEABLE ASSET FLOW


ADVANTAGE


Z. Do substitutes exist lor this factor?

3, Is this lactor in a compensatory relationship with atradeable network factor?

4. Is this factor in a compensatoiy relationship with anontradeable network lactor?

# Decision Path to suppon of SCA

5. Is this factor in an enhancing relationship with another networkfactor?


Leads to either competitive parity or competitivedisadvantage.

This same pattern is followed for all decision nodes.


Paths m. o, p, q and r follow the same logicas Paths h, i. j . k. and I explained above withrespect to Figure 5. Path t suggests that thepotential for support of SCA is very high whenthe asset stock is nontradeable compared towhen it is tradeable (Path m. Figure 5). Thismagnifying effect is due to the firm specificnature of nontradeable asset stock. Below wediscuss paths s, u, and v.

If a nontradeable asset stock is a complexnetwork member (Node 1, YES) without substi-tutes (Node 2, NO) despite the fact thatthere may be no compensatory nor enhancingrelationships existing (Nodes 3. 4. and 5. NO),even if there is a suppressing relationship withanother network factor (Node 6. YES) providingthe suppressing relationship is not resource orcompetency destroying, high support of SCA(Path s) is still achievable. Note that not allsuppressing relationships are harmful to firmresources either because they do not suppress akey success factor or because they suppress a

core 'incompetency,' i.e.. a core competencywhich hinders the organization's competitiveness.Miller (1990) notes how a firm's focus onits core competencies may ultimately be self-destructive^—a process he refers to as the IcarusParadox.

The last two paths do not require a complexnetwork (Node 1, NO). They do require that nosubstitutes exist (Node 2. NO) but they do notrequire compensatory relationships (Nodes 3 and4, NO). Providing that there is either anenhancing relationship (Node 5, YES) or at leastno suppressing relationships (Node 6, NO), theylead to high support of SCA (Paths u and v).Having no substitutes available will increase theimportance of the firm specific aspects and timedependency aspects. The lack of compensatoryrelationships among the factors does not hinderthe value and the enhancing relationships increasethe firm specific value. The neutral cogencyrelationship demonstrates that it at least doesnot detract from value creation.



ADVANTAGE

RESOURCE FACTOR -TRADEABLE ASSET STOCK


2. Do subsfitutes exist for this lactor?

3. Is this factor in a compensatory relationship with atradeable network (actor?

4. Is this factor in a compensatory relationship with anontradeable network factor?

# Decision Path to support of SCA.

5. Is this factor in an enhancingrelationship with another networkfactor?


Leads to either competitive parity or competitivedisadvantage.

This same pattern is followed tor all decision nodes.


Configuration implications

The logic articulated above suggests that whilethe debate between Barney (1986, 1989) andDierickx and Cool (1989) may be unimportantin the sense that one can achieve SCA nomatter what combination one starts from, it isimportant in the sense that the number ofpotential paths available increases as onemoves from flows to stocks and tradeable tonontradeable. There are nine unique paths thatcan at least potentially lead to high support ofSCA. When these decision paths are constrainedby the resource factor types, the 22 paths justpresented occur.

CONCLUSIONS AND IMPLICATIONS

Practitioner implications

Recall that numerous practitioner problems werenoted earlier in this paper. The problem ofinadvertently destroying a strategic resource bydivestiture or abandonment (Xerox's example),the problem of choosing incorrectly based on arisk assessment that does not include firm specificresources (Robin's 1992 argument), the problemof not understanding the relationships amongbundled things and their importance to theutilization of those bundles are all problems thatpractitioners address with each strategic decisionthey make. This configurational network


RESOURCE FACTOR -•NON-TRADEABLE ASSET STOCK

1. h thia factor a memtMr of a comptax network?

2. Do >ub«titutaB exisi for this factor?

3. Is this factor m a oompansaloiy ralationship with a iradMbta notworkfactor?

4. Is this ladoF in a compensatory ralatlonship with a nontradeable networkactor?

« DBcision Palh to support of SCA,

' Tills saine pattern is followed for all decision nodes.


ADVANTAGE

5, Is thia factor in an enhancingrelationship wilh another network factor?

S, Is Ihis factor in a supproasing relaiiorship with anolhar network facior''

Leads to either competitive parity or compelitive disadvantage.


approach allows practitioners to determine theeffort it will take to create and maintain thesystem asset of interest. By using the six strategicquestions, a practitioner should be able, inaddition to more fully identifying resourcesthat make up their strategic competencies, tounderstand the implications of changing onefactor of one resource on other apparentlyunrelated resources.

While the exact set of resource factors,competencies and distinctive competencies willvary from firm to firm and over time, thisframework gives practitioners a starting pointto more efficiently develop, change and usetheir resources. It also enables the confirmationof an educated guess about a firm resource andto expand a subset of important factors. AsCoff has noted, in many instances a practitionernot only cannot look to a financial statement

or analysis for many of the stratetic resourcesof a firm (since these forms explicitly excludethem) but 'must actually IGNORE informationfrom these sources to be successful in the longrun' (emphasis in the original, 1993:1). Thisanalysis will provide them with somewhere tolook and depending upon their ability toforecast the future, where to invest in systemresources.

Research implications and conclusions

The configurational aspects of a resource bundlesuggest that, while the external environment isimportant in determining and sustaining rentpotential, other elements at either the strategicgroup or firm level are also important (Rumelt,1991). Some may argue that rents are notsustained but are earned all at once though they


may be extracted over time.' We argue thatbeause resources implicitly have different eco-nomic life cycles, they are continually bundled.,unbundled, and rebundied. This results in rentsbeing sustained rather than earned all at once.This also implies that a firm may be using aportfolio of rents instead of attaining a singularsustainable competitive advantage. The degreeto which a firm can keep its currently rent-generating resources from being appropriatedby rivals is the degree to which a firm canmaintain its competitive position.

By closely examining the resource based viewof the firm theoretically implied characteristics,necessary relationships (factor type, networkmembership, substitutability and cogencyrelationships) among factors can be identified.Following this identification of relationships,factor network configurations that lead to highor very high support of sustainable competitiveadvantage are proposed. We present a total of16 projected configurations that should lead tohigh support and six that lead to very highsupport of sustainable competitive advantage.These are the configurations of relationshipsamong the factors needed to create a sytstemresource that can support the attaining of asustainable competitive advantage. In otherwords, the configuration of factors and relation-ships allows the creation of a resource that hasthe needed strategic characteristics of valuablerare, inimitable and organized to utilize.

The specific combination for any firm will bea result of the firm's history (and thus itsexisting set of firm resource factors), a firm'sstrategy, and the degree to which the firm'sstrategy fits the external environment, especiallyin regard to its competitors. The nestingrelationships of factors, resources, competenc-ies and distinctive competencies calls for firmresources comprised of contained resources andsystem resources.

We specifically see the use of substitutionrelationships and cogency relationships as thescreening for strategic resources. The appropri-ate combination of substitution, compensatory,enhancing, neutral and suppressing relation-ships in a system resource makes it possible forall resource factor types to lead to sustainable

Our thanks to Jay Barney for pointing this out.

competitive advantage. If the promise of theseconfigurations holds upon confirmation by fieldresearch, then these configurations and theirbase definitions are an operationalization ofresource based view of the firm theory. In sucha case, then the cross-sectional static view ofthe firm's ability to have a set of resources toattain a sustainable competitive advantage willbe clarified as requested by Porter (1991). Thiswould also be a starting point to address thelongitudinal processes by which a firm createsand/or maintains such a competitive advantage.With this operationaiization, field work andtesting of the firm resource based theoryof sustainable competitive advantage on theintrafirm level is attainable.

ACKNOWLEDGEMENTS

An earlier version of this paper was presentedat the 34th Annual Meeting of the WesternAcademy of Management, San Jose, CA. Theauthors would like to thank Jay Barney.Raphael Amit. anonymous reviewers, partici-pants at the Strategic Management Journal'sconference on New Strategy Paradigms, andthe editors of this special edition, C.K. Prahaladand Gary Hamel for comments on an earlierversion of this paper.

REFERENCES

Amit, R. and P. J. H. Schoemaker (1993). 'Strategicassets and organizational rent'. Strategic Manage-ment Journal, 14(1), pp. 33-45.

Arrow, K. J. (1974). The Limits of Organization. W.W. Norton, New York.

Barney. J. B. (1986). 'Strategic factor markets:Expectations, luck and business strategy". Manage-ment Science, 32, pp. 1231-1241.

Barney, J. B. (1989), 'Asset stocks and sustainedcompetitive advantage: A comment', ManagementScience, 35, pp. 1511-1513.

Barney. J. B. (1991). 'Firm resources and sustainedcompetitive advantage'. Journal of Management,17(1), pp. 99-120.

Barney. J. B. (1992). integrating organizationalbehavior and strategy formulation research: Aresource based view'. Working Paper. Texas A&M,College Station, TX.

Barney, J. B. (September 1993). Public communi-cations in Business Policy and Strategy EmailNetwork.


Barney, J. B. and R. W. Griffin (1992). TheManagement of Organizations: Strategy. Structure.Behavior. Houghton Mifflin Company: Boston,MA.

Becker, G. S. (1964). Human Capital. Columbia, NewYork.

Berkowitz. S. D. (1982). An Introduction to StructuralAnalysis: The Network Approach to Social Research.Butterworth and Co.. Ltd.. Toronto, Canada.

Blumberg, M. and C . D. Pringle (1982). "The missingopportunity in organizational research: Some impli-cations fora theory of work performance". Academyof Management Review. 7(4), pp. 560-569.

Bromiley, P. (September 1993). Public communicationsin Business Policy and Strategy Email Network.

Carr, C. (1993). "Global, national and resource-basedstrategies: An examination of strategic choice andperformance in the vehicle components industry".Strategic Management Journal. 14(7), pp. 551-568.

Coff, R. (September 1993). Public communications inBusiness Policy and Strategy Email Network.

Cool, K. D. and D. E. Schendel (1988). 'Performancedifferences among strategic group members". Stra-tegic management Journal, 9(3), pp. 207-223.

Conant. J. S., M. P. Mokwa and P. R. Varadarajan(1990). "Strategic types, distinctive marketing com-petencies and organizational performance: A mul-tiple measures-based study". Strategic ManagementJourrml. 11(5), pp. 365-383.

Conner, K. (1991). "Theory of the firm: Firmresources and other economic theories'. Journal ofManagement, 17(1), pp. 121-154.

Dierickx, I. and K. Cool (1989). 'Asset stockaccumulation and sustainability of competitiveadvantage". Management Science, 35,pp. 1504^1511.

Fiol. C. M. (1991). "Managing culture as a competitiveresource: An identity based view of sustainablecompetitive advantage". Journal of Management,17. pp. 191-211.

Hall. R. (1992). 'The strategic analysis of intangibleresources'. Strategic Management Journal. 13(2),pp. 135-144.

Itami. H. and T. W. Roehl (1987). Mobilizing InvisibleA.s.sets. Harvard University Press, Cambridge, MA.

Grant, R. M. (1991). 'The resource-based theory ofcompetitive advantage'. California ManagementReview. 33(3), pp. 114-135.

Learned, E. P.. C. R. Christensen, K. E. Andrewsand W. D. Guth (1965). Business Policy: Text andCa.'ies. Irwin, Homewood, IL.

Mahoney, J. T. and J. R. Pandian (1992). 'Theresource-based view within the conversation ofstrategic management". Strategic Management Jour-nal. 13(5), pp. 363-380.

McCallister, L. and C. S. Fischer (1983). 'A procedurefor surveying personal networks'. In R. S. Burtand M. J. Minor (eds.). Applied Network Analysis:A Methodological Introduction. Sage Publications,Beverly Hills, CA, pp. 75-88.

McWilliams, A. and D. L. Smart (1993). 'EfficiencyV. Structure conduct performance: Implications forstrategy research & practice. Journal of Manage-ment, 19, pp. 63-79.

Meyer, A. D. (1991). 'What is strategy's distinctivecompetence?". Journal of Management. 17,pp. 821-833.

Miller, D. (1990). The Icarus Paradox: How Excep-tional Companies Bring About Their Own Downfall.Harper Collins, New York.

Morgan, G. (ed.) (1983). Beyond Method: Strategiesfor Social Research. Sage Publications, BeverlyHills, CA.

Nelson, R. R and S. Winter (1982). An EvolutionaryTheory of Economic Change. Harvard UniversityPress, Cambridge, MA.

Oviatt, R. (1989). 'Caveats concerning application ofthe capital asset pricing model in the strategicmanagement concept". Proceedings of the Academyof Management, 1989 pp. 37-41.

Parthasarthy, R. and S. P. Sethi, (1992). 'The impactof flexible automation on business strategy andorganizational structure". Academy of ManagementReview. 17. pp. 86-111.

Peteraf. M. (1993). "The cornerstones of competitiveadvantage: A resource-based view". Strategic Man-agement Journal. 14(3), pp. 179-192.

Porter, M. E. (1980). Competitive Strategy. Free Press,New York.

Porter, M. E. (1985). Competitive Advantage. FreePress, New York.

Porter, M. E. (1991). 'Towards a dynamic theory ofstrategy". Strategic Management Journal, SummerSpecial Issue, 12, pp. 95-117.

Reed, R. and R. J. DeFillippi (1990), 'Causalambiguity, barriers to imitation, and sustainablecompetitive advantage". Academy of ManagementReview. 15(1), pp. 88-102.

Robins, J. A. (1992). 'Organizational considerationsin the evaluation of capital assets: Toward aresource-based view of strategic investment byfirms'. Organization Science. 3(4). pp. 522-536.

Roquebert. J.. R. Phillips, and C. Duran (August1993). 'How much does strategic managementmatter?'. Presentation at the National Academy ofManagement Meeting. Atlanta, GA.

Rumelt, R. P. (1991). 'How much does industrymatter?" Strategic Management Journal. 12(3).pp. 167-185.

Sayles. L. R. (1993). "Doing things right: A newimperative for middle managers'. OrganizationalDynamics. 21(4), 5-14.

Schoemaker. P. J. H. (1990). 'Strategy, complexityand economic rent". Management Science. ^6(10),pp. 1178-1192.

Smircich, L. (1983). 'Studying organizations as cul-tures'. In G. Morgan (ed.). Beyond Method:Strategies for Social Research. Sage Publications,Beverly Hills, CA, pp. 160-172.

Smith, K. K. (1983). 'Social comparison processes anddynamic conservatism in intergroup relations'.Research in Organizational Behavior. 5,pp. 199-233.

Summer. C. E.. R. A. Bettis. I. H. Duhaime, J. H.Gram. D. C. Hambrick. C. C. Snow. C. P,Zeithaml (1990). '1990 Doctoral education in thefield of business policy and strategy', 1990 YearlyReview of Management of the Journal of Manage-


ment, 12. pp, 167-183.Teece, D. J. (1986). 'Profiting from technological

innovation." Research Policy. 15. pp. 285-305.Tomer. J. F. (1987). Organizational Capital: The

Path to Higher Productivity and Well Being.Praeger. New York.

Tsoukas. M. (1989). The validity of idiographic

research explanations'. Academy of ManagementReview. 14(4), pp. 551-561.

Williamson, O. E. (1975). Markets and Hierarchies.Free Press. New York.

Womack, J. P., D. T. Jones and D. Roos (1990).The Machine That Changed the World. RawsonAssociates, New York.

strategic resources: traits, configurations and …kimboal.ba.ttu.edu/selected writings/strategic...

Documents