marching to the beat of different drummers: the …people.tamu.edu/~ltihanyi/inscompamj.pdf ·...

MARCHING TO THE BEAT OF DIFFERENT DRUMMERS:THE INFLUENCE OF INSTITUTIONAL OWNERS ON

COMPETITIVE ACTIONS

BRIAN L. CONNELLYAuburn University

LASZLO TIHANYITexas A&M University

S. TREVIS CERTOArizona State University

MICHAEL A. HITTTexas A&M University

This research extends agency theory by exploring the influence of varied, competing,principal interests on executive actions. Findings reveal that ownership of a firm bydedicated institutional investors, who hold concentrated portfolios over time, is pos-itively associated with firm use of strategic competitive actions. Ownership by tran-sient institutional investors, who hold broad portfolios and make frequent trades basedon current earnings, is negatively associated with strategic competitive actions andpositively associated with tactical ones. Appreciable ownership of the same firm bythese two classes of investors influences both strategic and tactical competitive actions.These results have broad implications for executives, investors, and policy makers.

Agency theory provides the conceptual founda-tion for the vast majority of corporate governanceresearch (Dalton, Daily, Certo, & Roengpitya, 2003;Shleifer & Vishny, 1997). According to this frame-work, the actions of executives may at times devi-ate from the interests of shareholders (Jensen &Meckling, 1976). To minimize these deviations,shareholders may rely on a number of governancemechanisms, such as boards of directors, CEO in-centives, concentrated ownership positions, andthe market for corporate control (Dalton, Hitt,Certo, & Dalton, 2007). Despite the availability ofthese various mechanisms, agency theory is funda-mentally focused on firm ownership structures(Jensen & Meckling, 1976). Given the implicationsof ownership for executive actions, research shouldaccount for the possibility that principals have

varying interests (George & Jones, 2000; Yan &Zhang, 2007).

The issue of heterogeneous investment prefer-ences is particularly germane to a group of princi-pals that have increased steadily in the UnitedStates over the past several decades: institutionalinvestors (Gillan & Starks, 2007). Not surprisingly,this increased presence has been accompanied by aflurry of academic research examining the conse-quences of institutional investor ownership (David,Hitt, & Gimeno, 2001). Nonetheless, researchershave recognized that these institutional investorsadopt different investment strategies (Yan & Zhang,2009). Porter (1992), for example, described funda-mental differences between what he called “dedi-cated” and “transient” institutional investors. Ded-icated institutional investors acquire concentratedequity positions in a few firms and have extendedinvestment horizons. These factors provide dedi-cated institutional investors with extensive knowl-edge of the firms in which they invest and a uniqueability to monitor executive actions over time(Bushee, 2004; Bushee & Noe, 2000). In contrast,transient institutional investors tend to acquire lessconcentrated equity stakes in a dispersed portfolioof firms. These investors also have a shorter invest-ment horizon, which explains in part their ten-

We gratefully acknowledge the role of Len Berry inshaping and developing the ideas underlying this re-search. We are also thankful for important commentsprovided by Wally Ferrier, Cindy Devers, and three anon-ymous reviewers. We would like to thank Brian Busheefor providing data on institutional investor classifica-tions. Lastly, we would also like to thank Gerry Sanders,whose guidance and direction as action editor greatlyimproved our research.

� Academy of Management Journal2010, Vol. 53, No. 4, 723–742.

723

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download or email articles for individual use only.

dency to frequently trade in and out of stocks(Bushee, 1998).

The contrasting intertemporal preferences ofthese different institutional investor types have im-portant implications for understanding executiveactions. Executives frequently make decisions fortheir firm to engage in competitive actions thathave unique consequences for different owners ofthe firm. To better understand these competitiveactions, competitive dynamics researchers havedistinguished between strategic and tactical com-petitive actions, both of which create value forshareholders (Chen, Smith, & Grimm, 1992; Smith,Grimm, Gannon, & Chen, 1991; Yu & Cannella,2007). Strategic actions create value by improvingthe ability of a firm to compete over time (Rappa-port, 2005; Smith, Grimm, Chen, & Gannon, 1989);tactical actions create value in the short term viadirect influence on current earnings and marketshare (Hitt, Ireland, & Hoskisson, 2009; Uotila,Maula, Keil, & Zahra, 2009).

We integrate corporate governance research withstudies of competitive dynamics to examine onemajor research question: how do principals withvarying intertemporal preferences influence thecompetitive actions that executives take? Agencytheorists have at times suggested that shareholdersgenerally have long-term interests (Daily, Dalton, &Cannella, 2003; Laverty, 1996). The preferences oftransient institutional investors, however, violatethis assumption (Bushee, 2001). Therefore, we ex-amine whether these owners influence the actionsof executives to align with their short-term objec-tives, even though doing so could result in thepossible neglect of actions directed toward a firm’slong-term competitive posture (Samuel, 2000). Infact, some have argued that capital markets operateefficiently only with respect to expectations ofshort-term earnings and that the market for strate-gic competitive actions designed to improve long-run value is inefficient (Froot, Scharfstein, & Stein,1992). With their concentrated portfolios held overtime, however, dedicated institutional investorsmay develop competencies in valuing long-termexecutive decisions (Bushee, 2004; Edmans, 2009).Therefore, we consider the extent to which dedi-cated institutional investors influence executivesto engage in competitive actions that are consistentwith their interests. Lastly, we explore the compet-itive actions in which executives engage whenthere are competing governance influences withina firm’s ownership structure (Bushee, 1998; Hsu &Koh, 2005). We propose that shareholders do notoperate in isolation but instead look to each otherfor guidance and at times may attempt to influenceeach other to adopt a common voice with respect to

the executive actions they desire (Gutierrez &Kelley, 2008).

To answer these questions, we examine the com-petitive actions taken by a sample of Fortune 500firms competing in rivalries between 1997 and2006. We consider the effects of transient and ded-icated institutional ownership on the extent towhich executives and their firms engaged in tacti-cal and strategic competitive actions as well as theinteraction between these two types of owners. Indoing so, we extend understanding of agency the-ory by exploring executive actions under differing,and competing, ownership structures. We also addto the literature on competitive dynamics by em-pirically examining both tactical and strategic com-petitive moves and by exploring the extent towhich governance structures potentially restrictthe range of competitive activity executives under-take. Results, which largely support our hypothe-ses, suggest that the competitive actions taken byexecutives are significantly related to firm owner-ship structures.

THEORETICAL BACKGROUND

An agency problem occurs when the actions ofexecutives (agents) diverge from the interests ofshareholders (principals) (Dalton et al., 2007;Jensen & Meckling, 1976). Scholars have high-lighted a number of mechanisms that may helpensure that executives act in accordance with theinterests of shareholders. Internally, boards of di-rectors (Daily et al., 2003), executive compensation(Carpenter & Sanders, 2002), and ownership struc-tures (Dalton et al., 2003) are used to help alignexecutive actions with shareholder interests. Whenthese internal monitoring mechanisms fail, an ex-ternal governance mechanism, the takeover market,becomes active to remedy agency problems (Jensen,1993).

Corporate governance scholars have extensivelyexamined the agency problem as it pertains to ex-ecutives and shareholders, but more recent re-search has described potential differences amongprincipals. Scholars have, for example, highlightedproblems that may occur when governments orfamilies own a large percentage of a firm’s equity(Shleifer & Vishny, 1997). Large shareholders mayencourage executives to engage in behaviors thatbenefit some shareholders at the expense of others;Villalonga and Amit referred to this as the “AgencyProblem II” (2006: 387). We build on this idea toexplore the notion that principals have heteroge-neous interests and disparate abilities to influenceexecutives to comply with those interests.

724 AugustAcademy of Management Journal

Institutional Investors

Heightened academic interest in one particularset of principals—institutional owners—reflectsthe growing economic power of this investor class.Institutional investors are a general class of equityholders that file 13-F Securities and ExchangeCommission (SEC) reports. The SEC requires thatinstitutions managing more that $100 million inequity file a quarterly report listing all holdingsthat are greater than 10,000 shares or $200,000 inmarket value. These investors include mutualfunds, hedge funds, pension funds (public, private,and corporate), banks, insurance companies, foun-dations, and endowments. The economic powerand clout of institutional investors has risensteadily in recent decades; in recent years institu-tions have acquired over 70 percent of U.S. equities(Gillan & Starks, 2007).

Because of their increased presence, institutionalinvestors have unique opportunities to overcomeobstacles to firm governance encountered by othershareholders. Clients of institutional investors signover their voting rights, effectively centralizing thebargaining power of all those clients in a singleentity and avoiding campaign costs (Bogle, 2005).Their large holdings provide institutions with in-centives to monitor firm actions and resources withwhich to do so (Gillan & Starks, 2007). Institutionalinvestors also benefit from membership in dedi-cated coordinating bodies, such as InstitutionalShareholder Services (ISS) and the Council of In-stitutional Investors, giving them access to researchand information not available to other investors(Gillan & Starks, 2007).

The steady increase in institutional investorownership in the United States has important im-plications for understanding the actions of execu-tives. Institutional investors have specific objec-tives and can employ multiple mechanisms toincrease the likelihood that executive actions areconsistent with those objectives. The most funda-mental tool is the threat of exit, meaning institu-tional investors can liquidate their equity positionsin firms. Such liquidations discipline a firm forlack of compliance with owner preferences, be-cause each exit reduces the value of the firm (Par-rino, Sias, & Starks, 2003). For instance, in 2003 EdLampert’s hedge fund, ESL Investments, sold off a20 percent stake in Autozone, causing its stockprice to fall.

Institutional investors can also employ “voice” toinfluence executive actions (Filatotchev & Toms,2006). This mechanism pressures executives to actin accordance with shareholder interests via differ-ent forms of activism. Research suggests that exer-

cising voice can affect the types of strategies firmsundertake (David et al., 2001; Hoskisson, Hitt,Johnson, & Grossman, 2002). The most commonmeans by which institutional investors exercisevoice is through their votes as shareholders. Forexample, institutional investors led by State StreetCorporation and Oppenheimer Funds recentlyvoted to approve a major investment by AdvancedMicro Devices to build a chip-making facility inMalta (Bernstein, 2009). Institutional investors cancombine voting power with the power to initiateshareholder proposals for even more effectivegovernance.

In addition to activism through voting, institu-tional investors can exercise their voice in otherways. They can, for example, use the media topressure a firm’s management to make specificcompetitive moves or implement changes to theorganization’s structure. In contrast, they can alsoparticipate in direct negotiations with managementor even publicly announce their opposition to it(David et al., 2001). Institutional investors at timesengage in behind-the-scenes discussions with man-agement or directors. For example, Fidelity (with10 percent ownership) and Highfields Capital Man-agement (with 5 percent ownership) worked be-hind the scenes to persuade the Mays family toendorse a buyout of Clear Channel Communica-tions, even though the board initially opposedthe deal.

Differences among Institutional Investors

Agency theorists have often described principal-agent relationships wherein principals have a uni-fied voice, pressuring executives toward long-termvalue creation (Dalton et al., 2007). They do so, inpart, because executives are presumably prone tounderinvest in long-term strategic projects, whichare typically associated with business risk (Laverty,1996). We consider, however, how dissimilar prin-cipal interests may influence executive actions.Specifically, we examine how principals withshort-term, long-term, and competing interests in-fluence firm-level competitive actions. To do so,we categorize institutional investors according totheir preferences.

Porter (1992) argued that institutional investorsdiffer in their behavior and incentives. He de-scribed “transient” institutional owners that createpressure for myopic strategies. These owners holdstakes in many different firms and frequently tradein and out of firms on the basis of changes infinancial value proxies. The high likelihood thatthese investors will sell a firm’s stock in the eventof disappointing financial news provides incentive

2010 725Connelly, Tihanyi, Certo, and Hitt

to executives to emphasize short-term financialgains. The combination of short time horizons anddiversified holdings makes it difficult for transientinstitutional investors to appropriately value thelong-term benefits of firms’ strategic investments(Matsumoto, 2002; Schnatterly, Shaw, & Jennings,2007). The short-term holdings of transient ownersalso suggest they will capture only a portion of thebenefits of investments with long-term payouts,making them less likely to pressure executives toensure such actions occur (Bushee, 2000).

In contrast, Porter (1992) also described “dedi-cated” owners that maintain large, long-term hold-ings concentrated in a small number of firms. Theseowners have incentive to monitor executive behav-ior and are able to understand richer and morecomplex information about the firms in which theyinvest. Dedicated institutional owners are morelikely to tolerate short-term earnings disappoint-ments as long as they are comfortable with long-runvalue prospects (Koh, 2007). Dedicated institu-tional investors own fewer firms, allowing them tomore effectively monitor the activities of their firms(Chen, Harford, & Li, 2007), and they can appropri-ate a greater share of the benefits of investmentswith long-term payouts (David, O’Brien, Yo-shikawa, & Delios, 2010).

A third group, “quasi-indexers,” also employbuy-and-hold strategies but are characterized byindexing and high portfolio diversification. Institu-tional investors in this group have abdicated theirmonitoring role, because their investment deci-sions follow broad indexes, regardless of the strat-egies adopted by particular firms within those in-dexes. They are, therefore, of less interest from agovernance perspective (Bushee, 1998).

In empirical research exploring the agency prob-lem, it is generally assumed that principal interestsare static. However, institutional investor behav-iors and preferences can change over time.1 Insti-tutional investors move between categories (see theAppendix for a further detailing of the categoriesemployed for our research). For example, accordingto Porter’s classification system, Brandes Invest-ment Partners was a quasi-indexer until it soldappreciable shares of Chiquita Brands Intl., LearProducts, and Ikon Office Solutions in 2005 be-

cause of their lackluster short-term performance,moving Brandes to the transient category. Othersmay change from dedicated to transient or viceversa (Bushee, 2000). Such movements suggest thatprincipal interests are a moving target. Thus, weconcluded that institutional investor interests areneither homogeneous nor static. Therefore, an in-vestigation of the changing nature of principal in-terests over time may elucidate how agency rela-tionships influence firm-level outcomes.

INSTITUTIONAL INVESTORS AND FIRMCOMPETITIVE ACTIONS

Owners of a firm, regardless of differences intheir trading behavior, may be interested in howexecutives maintain the firm’s competitiveness.When firms compete against each other, executivesundoubtedly consider the potential effectiveness ofa range of competitive alternatives (Chen, Su, &Tsai, 2007; Ferrier, Smith, & Grimm, 1999). Com-petitive dynamics researchers classify these alter-native actions as strategic or tactical (Hambrick,Cho, & Chen, 1996; Miller & Chen, 1996). We con-sidered the extent to which a firm’s levels of dedi-cated and transient institutional ownership influencethe amount of strategic and tactical competitive ac-tions that executives undertake.2 We expectedquasi-indexers to have less influence on firm out-comes, partly because they are less homogeneousin their preferences. Therefore, we did not de-velop hypotheses about this third group of insti-tutional investors.

Owner Influence on StrategicCompetitive Actions

Strategic competitive actions entail significantcommitments of specific, distinctive resources andare difficult to implement and reverse (Smith et al.,1991). These actions include, among others, acqui-sitions, strategic alliances, the establishment orclosing of subsidiaries, investments in technology,and restructuring (Chen et al., 1992). When ownerschoose to support strategic actions, they acceptlonger pay-offs and focus on a firm’s long-termcompetitive position (Hopkins, 2003). Neverthe-less, such actions can be costly in the short term,leading to temporary decreases in the share price ofthe firm (Laverty, 1996).1 Prior research has, in part, neglected these dynamics

by classifying institutional investors at a single point intime. Classifying owners differently as their trading hab-its change holds the potential of representing the dy-namic interests of principals and may contribute to the-ory insofar as it underscores the complexity of the agencyproblem.

2 We use the terms “dedicated institutional owner-ship” and “transient institutional ownership” to refer tothe level of ownership, or number of shares held, by eachclass of owner.


Relational Investors, headed by Ralph Whit-worth, exemplifies the dedicated institutional in-vestor (Laing, 2007). The average market value ofRelational Investors’ holdings totaled around $26billion in 2007, but this entire amount was investedin fewer than a dozen firms at once. Relationalseeks to invest in undervalued firms and then im-plement changes that enhance firm value. Theyimplement actions to turn firms’ performancearound in less than three years, but their invest-ment horizon is often longer. In 2007, Relationalacquired a $1.3 billion stake in Home Depot, a firmwhose stock had remained in the doldrums duringa period in which shares of rival Lowes nearlytripled. Relational urged a wide variety of strate-gic changes that led to the departure of HomeDepot’s CEO.

Dedicated institutional investors, such as Rela-tional, may encourage executives of their portfoliofirms to undertake strategic competitive actions forseveral reasons. First, dedicated investors may doso because they understand the value of long-termstrategic actions (Koh, 2007). Laverty (1996) sug-gested that, unlike the market that describes expec-tations for short-term earnings, the capital marketfor long-term actions is inefficient. Because theholdings of dedicated institutional investors areconcentrated in a small number of firms for longperiods of time, dedicated institutional investorsmay be better attuned to the types of actions thatmay add value to their portfolio firms.

Dedicated institutional investors may also en-dorse strategic competitive actions because theyhave the ability to actively monitor executive activ-ity, given that the outcomes of strategic decisionsbecome apparent over time (David et al., 2001).Hambrick et al. (1996) highlighted the difficultiesin implementing such strategic competitive ac-tions. Dedicated institutional investors can provideexecutives with access to expertise that facilitatesimplementation and can encourage the patience toallow it to occur. For example, billionaire investorEdward Lampert has a reputation for building po-sitions in undervalued companies and workingwith them to improve their performance. Such in-vestors can help executives focus on competitiveactions that produce long-term value (Bushee,2001). When executives make decisions involvingstrategic competitive actions, they may be able toreduce their uncertainty by seeking advice frominformed dedicated investors. Specifically, theymay augment their own private information withthe advice of dedicated institutional investors, aprocess described by the information cascade per-spective (Pollock, Rindova, & Maggitti, 2008). Ourfirst hypothesis follows from these arguments:

Hypothesis 1. The level of dedicated owner-ship is positively associated with the amountof a firm’s strategic competitive actions.

Compared to dedicated institutional investors,transient institutional investors are less likely toappreciate or value strategic competitive actions.Dedicated owners may develop an expertise inevaluating the potential of long-term strategic ac-tions, but transient institutional investors are lesslikely to develop the capabilities to make suchjudgments (Loescher, 1984). These investors favorstock value gains resulting from short-term perfor-mance improvements (Abarbanell, Bushee, &Raedy, 2003; Marginson & McAulay, 2008) andtherefore are wary of the short-term performanceimplications of strategic competitive actions (Ke &Ramalingagowda, 2005; Rappaport, 2005). Unlikethe investment strategies of dedicated owners, tran-sient owners’ investment strategies involve rapidchanges in investments. Because of their short-termholdings, transient institutional investors are un-likely to own a stake in a firm long enough torealize the gains associated with strategic compet-itive actions (Bushee, 2001). Therefore, as strategiccompetitive actions increase, transient investorsare likely to shift their investments into other firmsfocusing on short-term performance measures.

The Home Depot example we mentioned earliercan also illustrate the behaviors of transient own-ers. In the years leading up to the sweeping changesintroduced by Relational Investors, numerous tran-sient institutional investors moved quickly in andout of Home Depot ownership. Transient ownerssuch as Jennison Associates LLC and MFS Invest-ment Management, for example, acquired apprecia-ble stakes in Home Depot but maintained thesepositions for less than one year before selling theirstock holdings in the firm.

In addition, transient institutional investors alsolack the motivation to monitor firm strategies be-cause they invest smaller amounts of resources in alarger number of firms (Bushee, 1998; Chen et al.,2007). Moreover, transient investors are reluctantto support strategic competitive actions, becausethey are unlikely to maintain their equity positionslong enough to monitor the long-term implementa-tion these actions require (Koh, 2007). Transientinstitutional investors are sensitive to currentearnings, so they are more likely to sell a firms’stock when strategic competitive actions causethe firm to fall short of quarterly earnings goals(Bushee, 2000). These arguments suggest the fol-lowing relationship:


Hypothesis 2. The level of transient ownershipis negatively associated with the amount of afirm’s strategic competitive actions.

Owner Influence on Tactical Competitive Actions

Often complementary to strategic competitive ac-tions, tactical actions are “designed to fine-tunestrategy; they involve fewer and more general re-sources than strategic actions, are easier to imple-ment, and are often more reversible” (Smith et al.,1991: 63). These actions include pricing decisions,store openings and closings, and improvements inservice quality (Chen et al., 1992). Tactical actionsalmost universally have shorter time horizons thanstrategic actions (Miller & Chen, 1996); they typi-cally have direct implications for a firm’s quarterlyor annual performance. In some cases, however,improvements in the short term may lead to a de-cline in competitive position over the long term.

Several reasons suggest that transient institu-tional investors are likely to encourage tacticalcompetitive actions. Whereas dedicated institu-tional investors are often internally managed (e.g.,pension funds, endowments, foundations), tran-sient institutional investors are commonly in fiercecompetition for clients (Del Guercio & Tkac, 2002;Tihanyi, Johnson, Hoskisson, & Hitt, 2003). If atransient institutional investor experiences a lossor below-average performance in a given quarter, itrisks losing its clients (Khorana, 1996). These pres-sures are likely passed on to the executives of thefirms in which they own shares (Ali, Durtschi, Lev,& Trombley, 2004). Tactical competitive actionseffectively change the window of expected gains tothe short term because firm resources are commit-ted to achieving quarterly or yearly objectives(Hambrick et al., 1996). When performance im-provements are expected to occur in the short term,fund managers can be more confident that gainswill be realized during their investment horizon.

Further, with their diversified holdings, transientowners rely on financial performance proxies(Bushee & Noe, 2000). Therefore, they are likely tomaintain investments in a firms’ stock when tacti-cal competitive actions allow the firm to meet quar-terly earnings goals (Koh, 2007). Transient institu-tional investors are also expert monitors offinancial indicators across industries because theirfrequent trades require them to understand alterna-tive investment opportunities so they can makerapid changes (Del Guercio & Tkac, 2000). Armedwith information about the financial performanceof a broad set of firms, transient institutional own-ers can readily assess a firm’s probable short-termperformance (Yan & Zhang, 2007).

Transient institutional investors typically have ahistory of high portfolio turnover and therefore arewell versed in identifying firm activity that is likelyto result in quick gains. When they capture anappreciable ownership stake in a firm, they maypressure executives to take actions that will pro-vide rapid returns (Grinstein & Michaely, 2005). Ifexecutives are unwilling to comply, transient insti-tutional investors are more likely to sell and investin another firm than to exercise patience (Bushee,2001).

Hypothesis 3. The level of transient ownershipis positively associated with the amount of afirm’s tactical competitive actions.

Given their experience with their portfolio offirms, dedicated owners are likely to understandthat the long-term strategic competitiveness of thefirms in which they invest is best served by a bal-ance of strategic and tactical actions (Smith et al.,1991). Whereas the buy-and-hold investment strat-egy motivates dedicated owners to focus on thelong-term benefits of strategic competitive actions,these investors also realize the importance of main-taining competitive advantage in the short term bymeans of improvements in service quality, priceadjustments, and changes in product offerings (Fer-rier et al., 1999). If, by chance, an existing compet-itive advantage is lost in pursuing long-term objec-tives, it may be difficult to regain the advantage(Chen, 2009). Thus, dedicated institutional inves-tors are likely to encourage executives to maintaintheir competitive edge using tactical competitiveactions while implementing strategic competitiveactions to ensure long-term success (Hitt et al.,2009).

Recent changes at Home Depot illustrate thecomplementary role of strategic and tactical ac-tions. To reconcile lackluster firm performance,Home Depot recently closed its Home Expo DesignCenters; closing this division was a strategic action.At the same time, Home Depot also engaged intactical actions such as instituting a wage freeze forcompany officers and eliminating some headquar-ters support staff. Home Depot executives tookthese actions to hold down costs in the competitivebattle with Lowes. This example illustrates howtactical and strategic competitive actions can beintegrated to maintain a company’s competitiveposition.

Holding a firm’s shares over a long period of timefacilitates the development of trust between dedi-cated investors and executives of their portfoliofirms (Laverty, 2004). Even if investors in generalare concerned about managerial “short-termism”(Laverty, 1996), the familiarity of dedicated insti-


tutional investors with the decision makers of thefirms in which they invest can produce support fora broader range of competitive actions. The knowl-edge these investors gain through their long-termholdings helps them to better understand the ac-tions of executives. In turn, increasing trust andunderstanding provide the support desired by ex-ecutives, allowing them to focus on an integratedbalance of long-term and short-term objectiveswithout fear of losing their jobs. Taken together,these arguments suggest the following:

Hypothesis 4. The level of dedicated owner-ship is positively associated with the amountof a firm’s tactical competitive actions.

Owner Interactions

In the previous sections, we hypothesized oppos-ing influences of dedicated and transient ownerson the strategic competitive actions of the firms inwhich they invest. These conflicts underscore La-verty’s (1996) assertion that inefficiencies obscurethe capital market for valuing potential strategicactions. There may be segments of investors withinthe capital market, however, that are more efficientthan the market as a whole. With their portfoliosconcentrated on a few firms, dedicated investorsare the most likely to gain the capacity to accuratelyevaluate potential strategic actions (Chen et al.,2007). With their longer investment horizons, ded-icated owners have the patience to wait longenough to enjoy the benefits of strategic actions(Bushee, 1998). In addition, these investors haveexperience with monitoring and implementingstrategic actions in other portfolio firms (Chen etal., 2007). Transient owners do not have as muchexperience with these strategic actions as they fre-quently trade in and out of stocks (Bushee, 1998).

There is some evidence to suggest that sharehold-ers look to other shareholders for guidance andinformation about firms in which they are com-monly invested (George, Wiklund, & Zahra, 2005;Wermers, 1999). Transient owners may rely on in-formation about strategic competitive actions fromdedicated owners and thus imitate their support forfirms’ longer-term actions. Transient owners’ in-creased support of strategic actions could be rootedin the collective use of information and social re-inforcement, mechanisms described by the avail-ability cascade perspective (Pollock et al., 2008).Therefore, transient owners may be influenced bydedicated owners when they both own appreciableshares of the same firm. When dedicated owner-ship is high, transient owners may be less resistantto strategic competitive actions because dedicated

owners have the capacity to more appropriatelyvalue these actions and the incentive to monitortheir implementation. This reasoning is consistentwith the assertion of Chen et al. (2007) that allinvestors benefit from monitoring provided by vig-ilant institutional investors.

Another reason dedicated owners reduce thenegative influence of transient owners on strategiccompetitive actions is based on the relative powerof the two types of investor classes (Ryan & Schnei-der, 2003). Dedicated institutional investors gener-ally have larger ownership stakes in the firms inwhich they invest, giving them a larger number ofvotes on any proposals introduced at annual meet-ings. With their large holdings maintained overtime, dedicated owners offer executives the tacitassurance that they will not be disciplined (e.g., viasell-off) for investing in long-term strategic actions(Grinstein & Michaely, 2005). These factors providededicated institutional owners with a strongervoice and greater reward power than their transientcounterparts (French & Raven, 1959). In keepingwith prior arguments, dedicated owners may alsohave more referent power, or power based on re-spect for their ability to understand a firm and itsindustry, than transient owners. For example, thereferent power of famed dedicated investors suchas Nelson Peltz and Kirk Kerkorian added weight totheir opinions and influence. A firm’s executivesare likely to be more attentive to their largest, mostvocal shareholders because of the power, or poten-tial influence, of those shareholders over the futuredirection of the firm (Ingley & van der Walt, 2004).

Hypothesis 5. The level of dedicated owner-ship moderates the relationship between thelevel of transient ownership and strategic com-petitive actions; the relationship is less nega-tive when dedicated ownership is high.

We proposed that both dedicated and transientownership positively influence tactical competi-tive actions. Transient investors develop an exper-tise focused on short-term objectives (Puckett &Yan, 2008). They frequently trade in and out ofstocks and base these trading patterns on momen-tum strategies (Koh, 2007). As such, they developan understanding of tactical actions that are likelyto improve short-term performance. This rationaleis consistent with the idea that short-term institu-tional investors are better at collecting and process-ing information pertaining to short-term objectives(Yan & Zhang, 2009). The relative power of differ-ent classes of investors is less important becausethe interests of transient and dedicated ownersalign with each other.

Although both dedicated and transient institu-


tional investors support tactical competitive ac-tions, transient ownership may moderate the re-lationship between dedicated ownership and thetactical actions that executives undertake. Giventhe rapid turnover of their investments, transientowners are particularly adept at identifying firmsin which they can influence executives to reducecosts and maximize short-term gains. Althoughdedicated investors support tactical actions insofaras they are synergistic with strategic competitiveactions, transient institutional investors provide anadditional perspective based on superior under-standing about the short-term benefit of such ac-tions (Ke & Petroni, 2004; Wermers, 1999). Thevoice of dedicated owners endorsing tactical com-petitive actions is magnified by the presence oftransient owners, who support those actions fordifferent reasons (Yan & Zhang, 2007). When theinterests of principals are divided, so is managerialattention, but when the interests of principals in afirm’s ownership structure are united, they demandthe full attention of executives (Ocasio, 1997).

Transient investors also bring an additionalthreat of exit. Exit is rarely a viable option fordedicated owners because they commonly holdsuch substantial percentages of a firm’s stock that itis difficult or impossible to liquidate the stock be-fore its value declines. Also, it may be difficult toidentify viable alternatives in which to invest theirmoney (e.g., the California Public Employees Re-tirement System has hundreds of billions of dollarsto invest). Owners with massive total portfoliosmay be forced to remain invested in firms evenwhen they would prefer to exit because they arealready broadly invested in all the better alterna-tives. The presence of transient owners, then, addsthe threat of exit to the common voice of dedicatedand transient owners; together, these pressure ex-ecutives to engage in tactical competitive actions.

Hypothesis 6. The level of transient ownershipmoderates the positive relationship betweenthe level of dedicated ownership and tacticalcompetitive actions; the relationship is morepositive when transient ownership is high.

METHODS

Sample

The sample consists of all dual-firm competitiverivalries between firms in the Fortune 500 duringthe years 1997–2006, inclusive. We chose a sampleof dyadic rivalries because a number of uniqueproperties make them a particularly interesting anduseful subset of competitive relationships betweenfirms (Cool, Roller, & Leleux, 1999; Derfus, Magg-

itti, Grimm, & Smith, 2008; Porac, Thomas, Wilson,Paton, & Kanfer, 1995). For example, strategic com-petitive actions taken by a firm in a dyadic rivalryare highly identifiable and focused because the in-dustry is important to both firms in the rivalry(Chen, 1996; Ferrier et al., 1999). Dyadic rivalriesnaturally isolate strategic competitive actions sothat scholars can study them without the noiseinvolved with a larger group of competitors andcompetitive activity (Ferrier & Lee, 2002). Our sam-ple of dyadic rivalries allows us to examine firmsthat have similar motivations to engage, or not en-gage, in competitive actions because they share thesame competitive environment (Chen, 1996). Also,and perhaps most importantly, the actors in dyadicrivalries are well defined, making it possible tostudy how specific actors’ characteristics affectstrategic competitive actions over time (Chen et al.,2007). Our exploration of the strategic and tacticalactions of primary competitors over time also facil-itates comparison of our results with those of othercompetitive dynamics studies that have adopted asimilar approach (e.g., Derfus et al., 2008; Ferrier,2001; Ferrier & Lee, 2002; Ferrier et al., 1999).

Our sample, therefore, includes publicly tradedFortune 500 firms ranked number 1 or 2 in theirindustry in sales, with specialization ratios greaterthan .70 (Ferrier, 2001; Ferrier & Lee, 2002).3 Weimposed the restriction that the first- and second-ranked firms must each hold at least a 20 percentshare of the total market to ensure the firms werelikely to be leaders in strategic and tactical actionstaken in the industry; this eliminated industries inwhich the top two firms did not hold dominantmarket shares, such as SIC code 1531 (homebuild-ing), where a large number of firms own between2–8 percent of the market. These restrictions en-sured a sample of highly visible firms in whichinstitutional investors would likely have a stronginterest. A data set of 72 firms (i.e., 36 rivalries)representative of a broad variety of industries re-sulted. Ownership data were obtained from theCDA/Spectrum database. We collected data oncompetitive actions through a Lexis-Nexis search,coding news articles using prior competitive dy-namics research as a guide (Derfus et al., 2008;Ferrier & Lee, 2002). A competitive action was con-sidered to be any newsworthy move initiated by afirm to enhance its competitive position. Headlinesearches for the 72 firms in this sample yieldedmore than 60,000 Lexis-Nexis headlines during theyears 1997–2006. Reading each headline, we iden-

3 The specialization ratio is the revenue from a firm’slargest segment over its total revenue.


tified 5,550 articles that described unique compet-itive actions. We found that reading all headlinesfor each firm in each year of the analysis providedgreater confidence about the accuracy of identify-ing competitive actions than a content analysis ap-proach (Yu & Cannella, 2007). To confirm our iden-tification of strategic and tactical competitiveactions, an independent coder read and coded all ofthe headlines for a random sample of 10 percent ofthe firms in the data set; agreement between codersfor this random subsample using the intraclass cor-relation coefficient (ICC) was 96.2 percent (Shrout& Fleiss, 1979).

Measures

Dependent variables. We had two dependentvariables: strategic competitive actions and tacticalcompetitive actions. Tactical actions, which are de-signed to fine-tune strategy, are more common. Ex-amples of tactical actions include pricing changesand incremental service improvements (Ferrier,2001). Following prior competitive dynamics re-search, we measured tactical competitive actions asa count variable of the total annual number of tac-tical competitive actions (Chen et al., 1992; Derfuset al., 2008). For all firms in the sampling windowthere were 4,269 tactical actions, and the averagenumber of tactical actions taken by a firm in a givenyear was 6.3.

A headline was considered to report a strategiccompetitive action when the reported action hadfour components. First, it involved a significantcommitment of specific assets (Galbraith & Kazan-jian, 1986), typically a financial investment (e.g.,more than 5 percent of annual firm sales) or asimilar commitment of fixed assets. Second, theaction was difficult to implement (Hambrick et al.,1996). Actions that are difficult to implement areconsidered strategic because they are more difficultfor rivals to imitate. Third, the action would notyield payback for a long time (Miller & Chen, 1996).Fourth, the action was difficult to reverse (Ham-brick et al., 1996; Smith et al., 1991). An action maybe irreversible if overturning it would involve sig-nificant commitment of resources, disruption ofstaff or processes, negotiations with unions orexternal parties, negative publicity, and/or insti-tutional bureaucracy (Chen & Macmillan, 1992).For all firms in the entire sampling window,there were 1,281 strategic actions, and the aver-age number of strategic actions taken by a firm ina given year was 1.9.

Analyzing strategic actions, however, requiresmore than a count of the number of actions taken bya firm in a year. These actions are unique insofar as

they hold the potential of changing the competitivedynamics of an industry and/or moving the indus-try in a new direction. Therefore, it is important toconsider strategic competitive actions not only interms of their number, but also in terms of theirsignificance. To do so, we coded each action foreach of four characteristics: (1) commitment of re-sources, (2) difficulty of implementation, (3) timehorizon, and (4) irreversibility. Two researchers in-dependently coded each strategic competitive ac-tion, rating all four characteristics on a five-pointscale (1 � “very low/short,” 5 � “very high/long”).Agreement between raters assessed using the ICCwas 82.6 percent. The raters then discussed allactions with a characteristic coded with more thana one-point difference to reach agreement; finalagreement of 90.3 was reached. This procedure al-lowed for creation of a weighted measure of strate-gic competitive actions for each firm in every yearof the sampling window. More specifically, eachindividual strategic action was multiplied by itsweighted significance, and these results weresummed annually by firm. Tactical competitive ac-tions, on the other hand, represent short-term ad-justments to existing strategies. As a result, noweighting of these actions was required.

Independent variables. The independent vari-ables focused on institutional owners with at least1 percent equity, a criterion that removed ownerswith marginal equity positions. This cutoff wasconsistent with prior research on the influence ofinstitutional investors insofar as it included onlythose owners with sufficient holdings to promoteinterest and activism (Johnson & Greening, 1999;Tihanyi et al., 2003). We categorized each memberof this group as a dedicated owner, a transientowner, or a quasi-indexer, in accordance withBushee’s (1998) classification system, described inthe Appendix. By eliminating the middle group,quasi-indexers, we explored the two major types ofinstitutional investors, dedicated and transient,which are clearly differentiated. Furthermore, ourseparate categorization of institutional investors foreach year in the analysis captured changes in theirtrading behavior and thus reflected differences ininstitutional investor preferences over time. Ourclassification of institutional investors on an an-nual basis dynamically captured principal interestsover time, which yielded a more process-orientedmeasure than is typically found in agency theoryresearch.

To test hypotheses regarding the influence ofowners, we measured the number of shares held bydifferent types of institutional investors, calculatedas the percentage of the total number of sharesoutstanding for a focal firm. Thus, our measure of


dedicated institutional ownership was the percent-age yielded by dividing the count of the number ofshares owned by dedicated institutional owners ina given firm-year by the total number of sharesoutstanding for the same firm-year. The mean levelof dedicated institutional ownership in a givenfirm-year was 11.9 percent. This was distributedamong an average of 1.9 dedicated owners per firmin a given year. Therefore, the average holding ofany one dedicated institutional investor was about6.3 percent of the total number of outstandingshares of the firm.

We calculated transient institutional ownershipin the same way, as a count of the number of sharesowned by transient institutional investors dividedby the total number of shares outstanding for agiven firm-year. The mean level of transient own-ership was 10.1 percent, distributed among an av-erage of 4.6 transient owners per firm in a givenyear. Therefore, the average holding of any onetransient owner was about 2.2 percent of the totalnumber of shares of the firm. All ownership vari-ables represent shares owned at the end of the yearprior to which competitive actions occurred, sothat independent variables preceded the dependentvariable.

Control variables. Prior research has indicatedthat a range of firm characteristics beyond thosedescribed in our hypotheses can affect competitiveactivity. For example, large firms often have greaterresources and therefore can be more likely to en-gage in competitive activity. Thus, we controlledfor firm size with the natural logarithm of the totalnumber of each firm’s employees. Additionally, itis important that a firm be able to undertake stra-tegic competitive actions. Therefore, financialslack was used to control for this ability; we mea-sured unabsorbed slack using the quick ratio,which is the ratio of current assets less inventory tocurrent liabilities (Ferrier, 2001). Similarly, strate-gic competitive actions may be a function of howwell, or how poorly, a firm is performing. Perfor-mance was therefore included as a control, mea-sured as each firm’s lagged return on sales (ROS)(Derfus et al., 2008).

We also controlled for other forms of competitiveactivity that might influence our results. For exam-ple, we controlled for a firm’s prior competitiveactivity, a potentially important predictor of firmactions, with a count of each firm’s total competi-tive actions (strategic and tactical) in the year priorto the year of analysis. Our hypotheses distinguishbetween strategic and tactical competitive actions,but the two may be related. Therefore, when em-ploying strategic competitive actions as the depen-dent variable, we controlled for tactical competi-

tive actions in the focal year. When using tacticalcompetitive actions as the dependent variable, wecontrolled for strategic competitive actions in thefocal year. Lastly, we controlled for the influence ofindustry competitive activity by limiting our sam-ple to industries with clear dyadic rivalries. How-ever, there may be differences between industriesthat depend on competitive activity that goes be-yond the top two firms in the industry. Therefore,we controlled for the competitive activity of thethird firm (i.e., the firm with the third largest mar-ket share in an industry), measuring this variable asa count of the total number of the third firm’scompetitive actions in the year of analysis.

Industry-level variables might also confound thedependent variable. Firms may be more likely toengage in strategic competitive actions in indus-tries with slow growth because firms in such indus-tries need to attract new customers from their com-petitors. We controlled for industry growth with therate of the percent change in industry gross salesbetween the focal and the previous period for eachfour-digit SIC category. In addition, we controlledfor industry concentration, which was calculatedusing a Herfindahl index based on each four-digitSIC category in the panel data set (Derfus et al.,2008).

The motivation for this study was to understandthe influence of governance on competitive activ-ity. Agency theory suggests that as CEOs receivemore pay (which in our sample is tantamount toreceiving more stock options), they are likely toengage their firms in competitive actions that aremore oriented toward long-term competitiveness.Therefore, CEO compensation was a control, mea-sured as the total value of salary, bonus, and thegranted value of stock options. CEOs who alsoserve as board chairs help establish strong leader-ship but can simultaneously promote entrench-ment. Therefore, the effect of such CEO duality wasalso included, as a dummy control variable coded 1if a single individuual was both a firm’s CEO andboard chair and 0 otherwise. Ownership is anotherimportant form of corporate governance. As own-ership concentration increases, owners might beexpected to exercise their governance role differ-ently. We accounted for this likelihood with anownership Herfindahl, calculated as the sum of thesquared percentage of ownership of each 1 percent(or higher) institutional investor, a value that in-creases with increasing blockholder concentration(Baysinger, Kosnik, & Turk, 1991). Another impor-tant ownership issue is the extent to which theperformance of an investor’s portfolio depends onthe performance of a focal firm. For each investor ina focal firm, we calculated the percentage of the


investor’s total portfolio represented by that inves-tor’s holdings in the focal firm. We averaged thesefor all institutional investors in each firm-year tocalculate an average portfolio share.

Analysis

Table 1 reports the means, standard deviations,and correlations for all variables in this study. Cor-relation levels between variables suggested noproblems of multicollinearity.

One of our dependent variables, tactical compet-itive actions, was measured as a nonnegative inte-ger count. The use of Poisson models is appropriatefor count outcomes, including those derived frompanel data (e.g., Ahuja & Katila, 2004; Penner-Hahn& Shaver, 2005). The other dependent variable,strategic competitive actions, was a weighted non-negative integer count. Weighted counts follow dis-tributions similar to those of count variables, soPoisson models were the most appropriate estimationtechnique for both dependent variables (Dushnitsky& Lenox, 2005; Lerner, 2005). Likelihood-ratio tests ofoverdispersion indicated both dependent variableswere most closely aligned with the negative binomialdistribution.

We pooled observations over ten years, whichcould have violated negative binomial regression’sassumption of residual independence. To accountfor potential contemporaneous correlation, ourmodels also included time dummy variables (Certo

& Semadeni, 2006). Further, we relied on randomeffects to account for potential firm heterogeneity.The Hausman statistic was not significant for thefinal model with either of the dependent variables,suggesting that differences in the coefficients pro-vided by random- and fixed-effects models werenot systematic. Therefore, random-effects modelswere preferred because they conserve degrees offreedom.

RESULTS

Table 2 reports the results for the first dependentvariable, strategic competitive actions. The firstmodel shows the influence of control variables. It isnot surprising that a firm’s tendency to engage instrategic actions in a prior year is a significantpredictor of strategic competitive actions in a focalyear. Tactical competitive actions also predict stra-tegic competitive actions. As expected, a firm’s pastperformance and available financial slack are im-portant predictors of its strategic competitive ac-tions. Firm size is also a statistically significantpredictor, with large firms being more likely toimplement strategic competitive actions. Consider-ing governance influences, we found that CEOcompensation is positively associated with thenumber of strategic competitive actions; higher-paid CEOs implement a greater number of strate-gic actions. Ownership concentration is anotherimportant governance mechanism, in that block-

TABLE 1Descriptive Statistics and Correlationsa

Variable Mean s.d. 1 2 3 4 5 6 7 8 9 10 11 12 13 14

1. Strategic competitiveactionsb

19.11 20.94

2. Tactical competitiveactions

6.26 7.76 .23

3. Prior competitiveactivity

7.64 8.05 .27 .56

4. Third-firm competitiveactivity

3.12 5.44 .12 .07 .05

5. Financial slack 1.32 0.71 .07 .00 .01 �.046. Firm performance 0.04 0.05 .16 .02 .02 �.02 �.147. Firm size 3.31 0.98 .19 .23 .26 .29 �.08 .018. Average portfolio sharec 2.09 2.17 .07 .01 .04 .04 �.11 .07 .029. Ownership Herfindahl 0.03 0.03 .08 .01 .01 �.03 .12 �.18 �.06 .39

10. CEO compensation 7.73 8.99 .28 .19 .27 .05 �.09 .14 .26 �.02 �.1111. CEO duality 0.72 0.45 .07 �.01 �.02 .03 .02 .01 .05 .02 �.02 .0212. Industry growthc 14.59 21.03 .09 .08 .09 .05 .01 .05 .06 .01 �.04 .18 �.0113. Industry concentration 0.27 0.14 .01 .05 .09 .08 .25 �.06 �.11 �.06 .01 �.03 �.09 .0814. Dedicated ownershipc 11.93 10.50 .21 �.03 �.03 �.01 .10 �.06 .06 .13 .36 �.03 .10 �.03 �.1115. Transient ownershipc 10.07 9.80 �.11 .31 .06 .01 .08 �.16 �.01 �.01 .09 .01 .02 �.06 .03 �.08

a Correlations greater than .08 or less than �.08 are statistically significant at p � .05.b Weighted variable.c Expressed as a percentage.


holders with higher concentrations of ownershipare associated with more strategic competitiveactions. The other control variables are notstatistically significant.

Hypothesis 1 states that a firm’s level of owner-ship by dedicated shareholders is positively asso-ciated with its number of strategic competitive ac-tions. Model 2 in Table 2 reports the regressionresults that examine this relationship. As shown,the coefficient for dedicated institutional owner-ship is positive and statistically significant (p �.001). These results provide support for Hypothesis1. Hypothesis 2 posits a negative relationship be-tween transient shareholders and strategic compet-itive actions. Model 2 in Table 2 also reveals astatistically significant, negative association be-tween the level of a firm’s transient ownership andthe amount of its strategic competitive actions (p �.01). Thus, Hypothesis 2 receives support. Additionof the main effects of dedicated and transient institu-tional ownership into model 2 increases the overallmodel fit to a Wald chi-square of 211.4 (23 df) from avalue of 175.5 (21 df) for the control variable model,which is a statistically significant improvement.

Hypotheses 3 and 4 also examine main effectsbut with tactical competitive actions as the depen-dent variable. Table 3 reports the results for theseactions. In model 1, the control variables are statis-tically significant in a pattern similar to that found

with strategic competitive actions. One differenceis that firm performance and financial slack are notstatistically significant, presumably because firmsengage in tactical actions when they are performingwell or poorly but can only engage in strategicactions when they have the financial resources todo so. The ownership Herfindahl index is also notstatistically significant, indicating that blockholderpresence is more important to a firm’s likelihood ofengaging in strategic actions than it is to engagingin tactical actions. Model 2, then, shows the directeffects of the level of dedicated and transient insti-tutional ownership on tactical competitive actions.Transient institutional owners have a statisticallysignificant and positive effect on tactical competi-tive actions. Although transient owners are nega-tively associated with strategic actions (from Table2), they are positively associated with tactical ac-tions (Table 3), lending support to Hypothesis 3.The coefficient for dedicated institutional ownership,on the other hand, is not statistically significant; so,dedicated owners appear to have an indeterminatedirect effect on tactical actions. Therefore, Hypothesis4 is not supported.

The final two hypotheses focus on interactiveeffects. A negative binomial model is a nonlinearestimator, so the coefficients provided by themodel do not represent the marginal effect. Instead,the marginal effect is of the form e�X� and is there-

TABLE 2Random-Effects Negative Binomial Regression of Strategic Competitive Actions

Variables

Model 1:Controls(n � 682)

Model 2:Main Effects

(n � 682)

Model 3:Low Dedicated

Ownership (n � 380)

Model 4:High Dedicated


Tactical competitive actions 0.02** (0.01) 0.02*** (0.01) 0.02** (0.01) 0.02* (0.01)Prior competitive activity 0.01* (0.01) 0.01 (0.01) 0.01 (0.01) 0.02 (0.01)Third-firm competitive activity 0.01 (0.01) 0.01 (0.01) �0.00 (0.01) 0.02* (0.01)Average portfolio share �2.31 (2.24) �2.08 (2.08) �0.89 (2.92) �2.90 (3.07)Ownership Herfindahl 0.07*** (0.02) 0.03* (0.02) 0.06* (0.03) 0.00 (0.03)Financial slack 0.14* (0.06) 0.15* (0.06) 0.11 (0.08) 0.22* (0.09)Firm performance 5.36*** (0.99) 4.90*** (0.98) 4.20*** (1.28) 7.31*** (1.58)Firm size 0.17*** (0.05) 0.12* (0.05) 0.12 (0.66) 0.10 (0.07)CEO compensation 1.49** (0.09) 1.64*** (0.47) 1.22* (0.66) 2.53*** (0.64)CEO duality 0.16 (0.20) 0.12 (0.09) 0.05 (0.13) 0.30* (0.13)Industry growth 0.10 (0.20) 0.14 (0.19) 0.09 (0.29) 0.33 (0.26)Industry concentration �0.09 (0.34) �0.10 (0.33) 0.19 (0.43) �0.35 (0.48)Dedicated ownership 2.45*** (0.50) 4.03** (1.76) 3.15*** (0.96)Transient ownershipa �1.72** (0.60) �2.77*** (0.77) [�2.07] �0.33 (0.93) [�0.32]

Time dummy variables Included Included Included IncludedWald �2 175.53 211.44 138.25 146.31df 21 23 23 23

a The marginal effect is reported in brackets.* p � .05

** p � .01*** p � .001


fore contingent on the values of the independentvariables (Hilbe, 2007; Shaver, 2006). To test theinteraction hypotheses, we followed the guidanceof Penner-Hahn and Shaver (2005) and split thesample by observation. Specifically, we examinedthe marginal effect of the moderated variable on thedependent variable in two subsamples, testingwhether the effect was higher (lower) when themoderator was above (below) the mean. This ap-proach focuses on differences in the marginal effectrather than the coefficient estimator because ofnonlinearity in the model. Specifically, if the ob-servations in one subsample lie on a different partof the estimated curve than those in the other sub-sample, the coefficients are not comparable.

In the first interaction, we split the sample at themean level of dedicated institutional ownership.We report the marginal effect of the level of tran-sient institutional ownership in square brackets inTable 2 (Penner-Hahn & Shaver, 2005). The mar-ginal effect of transient institutional ownership is�2.07 and statistically significant (p � .001) whendedicated institutional ownership is low; it is�0.32 and not statistically significant when dedi-cated ownership is high. This pattern indicates thatthe relationship between transient institutionalownership and strategic competitive actions is lessnegative when dedicated institutional ownership ishigh, thereby providing support for Hypothesis 5.

In the second interaction, we split the sample at themean level of transient institutional ownership (Ta-ble 3). For tactical actions, the marginal effect ofdedicated institutional ownership is �0.72 and notstatistically significant when transient ownershipis low; it is 3.69 and statistically significant (p �.01) when transient ownership is high. So, the mar-ginal effect of dedicated institutional ownership ontactical competitive actions is greater when tran-sient institutional ownership is high, providingsupport for Hypothesis 6.

DISCUSSION

In this study, to answer our primary researchquestion (To what extent does ownership structureinfluence a firm’s competitive actions?), we exam-ined the relationship between institutional owner-ship and a firm’s tactical and strategic competitiveactions. The results of this study indicate that in-stitutional investors have divergent interests withrespect to executive actions, and their influenceextends beyond broad firm-level strategies and pol-icies (David, Bloom, & Hillman, 2007; Hartzell &Starks, 2003). We found that a firm’s ownership bydedicated institutional investors was positively as-sociated with the extent to which executives un-dertook strategic competitive actions. In contrast,ownership by transient institutional investors was

TABLE 3Random-Effects Negative Binomial Regression of Tactical Competitive Actions

Variables

Model 1:Controls(n � 682)

Model 2:Main Effects

(n � 682)

Model 3:Low Transient


Model 4:High Transient


Strategic competitive actions 0.01*** (0.00) 0.01*** (0.00) 0.01*** (0.00) 0.00 (0.00)Prior competitive activity 0.01** (0.00) 0.01* (0.00) 0.02*** (0.01) 0.03*** (0.01)Third-firm competitive activity 0.01 (0.01) 0.01 (0.01) 0.02* (0.01) �0.01 (0.01)Average portfolio share �2.80 (1.79) �2.63 (1.81) �3.44 (1.90) �0.58 (3.62)Ownership Herfindahl �0.01 (0.02) 0.00 (0.02) �0.01 (0.02) 0.00 (0.03)Financial slack �0.07 (0.07) �0.08 (0.07) �0.02 (0.09) �0.17 (0.09)Firm performance 0.08 (0.84) 0.29 (0.84) 0.25 (1.62) �0.14 (1.20)Firm size 0.20** (0.06) 0.22*** (0.07) 0.16* (0.08) 0.10 (0.06)CEO compensation �1.20** (0.46) �1.15** (0.44) �1.97*** (0.52) 0.59 (0.66)CEO duality 0.01 (0.07) �0.01 (0.07) 0.07 (0.09) �0.15 (0.11)Industry growth 0.15 (0.16) 0.18 (0.15) 0.02 (0.18) 0.27 (0.30)Industry concentration �0.54 (0.41) �0.31 (0.40) �1.04* (0.53) 0.46 (0.36)Dedicated ownershipa �0.11 (0.46) �0.78 (0.56) [�0.72] 2.79** (0.84) [3.69]Transient ownership 2.65*** (0.42) 0.63 (1.21) 4.36*** (0.77)

Time dummy variables Included Included Included IncludedWald �2 272.91 338.93 181.03 187.37df 21 23 23 23

a The marginal effect is reported in brackets.* p � .05

** p � .01*** p � .001


negatively associated with such actions. Transientownership was, however, positively associatedwith the extent to which executives undertook tac-tical competitive actions. We also present evidencesuggesting that institutional owners may look toother institutional owners for guidance when theydecide to endorse strategic or tactical competitiveactions. Results indicate that for high levels of ded-icated ownership, the relationship between tran-sient ownership and strategic competitive actionswas less negative. These two types of owners alsointeract in the way they influence tactical compet-itive actions. For high levels of transient institu-tional ownership, dedicated ownership was posi-tively related to number of tactical competitiveactions. These results offer new insight for bothagency theory and competitive dynamics research.

Agency Theory

The theory discussed herein suggests that, to un-derstand executive actions, one must first examinethe degree to which a firm’s investor base has het-erogeneous objectives. Although some investorsprefer that executives implement actions that pro-duce long-term benefits, others are more concernedabout the annual and quarterly costs of such ac-tions (Stein, 1989). Furthermore, hypotheses abouttactical competitive actions contribute to under-standing of the mechanisms that lead executives topotentially engage in economic short-termism, thusinforming the debate about executive myopia as asource of agency conflict (Laverty, 1996; Marginson &McAulay, 2008). By empirically capturing the effectsof different types of institutional investors, this studyalso helps to explain why reviews of empirical re-search on agency theory have often uncovered incon-sistent results (e.g., Dalton et al., 2003).

A potentially important contribution to agencytheory is our explication of time in agency relation-ships (George & Jones, 2000). Integrating executiveactions with differing intertemporal consequencesand ownership structures with differing prefer-ences uncovers a set of complex agency relation-ships that have previously been unexplored.Whereas agency theorists have typically consid-ered alignment from the standpoint of monitoringand “incentivizing” agents (Dalton et al., 2007;Eisenhardt, 1989), we adopted a principal-side per-spective to explore how agents engage in actions toconform to the objectives of complex ownershipstructures (Chen et al., 2007). Our results suggestthat principals with differing interests can influ-ence the implementation of a range of executiveactions. Additionally, our results demonstrate how

some institutional investors may persuade others toinfluence executives toward specific ends.

This research also addresses a methodologicalshortcoming of extant agency theory research. Spe-cifically, we contribute to an emerging segmentof governance literature examining institutionalowner differences and their potential effects onfirm strategies and performance (e.g., Hoskisson etal., 2002; Johnson & Greening, 1999; Kochhar &David, 1996; Tihanyi et al., 2003). This prior re-search has provided important contributions by ex-amining the actions of different classifications ofowners based on competitive pressures (e.g., pres-sure-resistant, pressure-sensitive) and legal forms(e.g., pension funds, mutual funds). Although thisis an important first step, institutional investorsmay change their objectives over time (Bennet,Sias, & Starks, 2003). We made efforts to capturethe dynamic nature of institutional ownership byannually classifying investors on the basis of theiractual trading behavior (Bushee, 1998). In so doing,this study provides a process/action-oriented per-spective on principal-agent relationships that addsto understanding of principals in agency theory.

Competitive Dynamics

Complementing potential contributions to agencytheory, this study extends understanding of com-petitive dynamics and, in particular, the nature ofcompetition in dyadic rivalries (Chen et al., 2007;Ketchen, Snow, & Hoover, 2004). Perhaps most im-portantly, we develop a framework of competitiveactions through the lens of a firm’s owners, a per-spective that has been rarely used in prior compet-itive dynamics research. Our study provides evi-dence that ownership structures likely play animportant role in explaining competitive actions.

Within the awareness-motivation-capabilityframework that scholars have used to explain com-petitive dynamics, the influence of firm ownerscontributes most to understanding of motivation.Research on competitive behavior has focused pri-marily on factors that help firms to be aware ofcompetitor behavior (e.g., TMT heterogeneity andmultimarket competition) and that enhance their ca-pability to initiate competitive behavior (e.g., pastperformance and financial slack) (Ferrier, 2001).These studies have yielded valuable insights regard-ing certain predictors of competitive behavior, em-phasizing environmental and organizational explana-tions. In contrast, scholars have devoted less attentionto the effect of motivation on competitive behavior.The results of our study reveal an important motiva-tional influence (potentially positive or negative) oncompetition: a firm’s ownership structure.


Our study also provides a number of method-ological contributions to the competitive dynamicsliterature. For example, we include an integratedanalysis of tactical and strategic competitive ac-tions. Although a number of studies have focusedon tactical competitive actions such as pricechanges or new service offerings, only a few havealso addressed strategic competitive actions suchas alliance formation, entering new markets, or ac-quiring other firms (e.g., Chen et al., 1992; Smith etal., 1991). Both strategic and tactical actions, how-ever, are important for firms maintaining their com-petitive advantage. Our study, then, advances com-petitive dynamics research by simultaneously andindependently considering these different forms ofcompetitive actions. In addition, we implement anew measure of the magnitude of strategic compet-itive actions. Our approach builds on prior litera-ture examining the reversibility of competitiveactions (Chen & MacMillan, 1992). By using aweighted measure with four dimensions of actionscharacterized as strategic, we capture importantvariation within the broad category of strategiccompetitive actions.

Examining different competitive actions from theowners’ perspective provides important implica-tions for competitive dynamics research. Our re-sults, for instance, indicate that transient institu-tional owners may discourage strategic competitiveactions, which limits the range of competitive op-tions available to firms (Ferrier & Lee, 2002). Tran-sient investors might use the threat of exit to pres-sure executives to consider only those competitiveactions that will not result in short-term earningsshortfalls (Hsu & Koh, 2005). In this context, con-tinuously striving to meet quarterly earnings expec-tations is likely to constrain the focus of executiveattention (Ocasio, 1997). When executives employa smaller range of competitive actions, firms be-come more predictable in their competition, pro-viding an advantage to rivals that do not operatewith these restrictions (Ferrier & Lee, 2002).

Implications for Practice

The results of this study shed light on how thevaried interests of owners help executives to main-tain a balance between strategic and tactical com-petitive actions. Firms in different competitive sit-uations might benefit from tactical changes that areconsistent with the objectives of transient institu-tional owners. Transient owners appear to provideexpertise on tactical actions (Yan & Zhang, 2007) todedicated owners so that they, too, may be morefavorably disposed to such actions. In competitivesituations that require strategic actions, such as a

major acquisition between two competitors, execu-tives might also benefit from the presence of differ-ent institutional investors. Although dedicatedowners highlight the benefits of long-term re-sponses with their support, transient owners aremore likely to emphasize the costs of such strategicactions. The fact that dedicated institutional inves-tors may be able to convince transient owners toincrease their support of strategic actions may helpensure that executives consider strategic actionsmore carefully (Del Guercio, Seery, & Woidtke,2005). Yet these same executives are likely to care-fully consider the costs of such actions because ofthe influence of transient owners. For example,executives might still go forward with an acquisi-tion, but they may negotiate more strenuously,thereby reducing the premium paid to acquire thetarget firm.

Our results also have implications for publicpolicy. Several SEC regulations limit the meansby which institutional investors can impose pres-sure on firms in which they invest (Gillan &Starks, 2007). This study shows that institutionalinvestors with varying motives toward competi-tive actions can provide the expertise and invest-ment horizons that allow executives to examinethe balance between short-term and long-termgoals. Future regulations might acknowledgethese differences in investor preferences morecarefully and endorse governance mechanisms inwhich owners can provide executives with infor-mation and advice about the cost and benefits ofcompetitive actions.

Limitations and Future Research

Although our study reports associations betweeninstitutional ownership and competitive actions,our work has some limitations that should benoted. Like other studies on institutional investors,our study provides only limited knowledge of theinteractions between these investors and the exec-utives of their portfolio firms. Additional informa-tion on how investors or executives initiate conver-sations about competitive actions and when suchconversations take place might enhance our findings.Future research thus might examine how often exec-utives solicit advice from their major investors. Inaddition, board membership may mediate this rela-tionship in some cases (Westphal, 1999).

Future research might also address the extent towhich different types of shareholders benefit firmsby fostering tactical and strategic actions that addvalue and by helping to appropriately balance tac-tical and strategic competitive actions. Some situ-


ations may require more tactical competitive ac-tions, whereas at other times firms may benefitfrom more strategic competitive actions. Share-holders could provide insight to executives as tohow to manage this balance for the most beneficialcompetitive position vis-a-vis rivals.

The analysis presented here employed tacticaland strategic competitive actions of the focal firmsas the dependent variable. An important extensionof this work might involve examining how rivalfirms respond to such actions. Doing so would con-tribute to understanding how governance struc-tures facilitate or hinder a firm’s ability to deterrival activity. In competitive dynamics research,initiation of strategic competitive actions is impor-tant (Yu & Cannella, 2007), but scholars might pro-vide a more complete picture of the influence ofinstitutional owners on competitive rivalries by ex-amining how they affect the speed and variety ofrival firm responses.

Another extension to this work resides at theintersection of governance mechanisms. We con-sidered the effect of owners on competitive dynam-ics; future studies could also explore how otherdimensions of governance structures motivatefirms to engage, or not engage, in competitive ac-tions. For example, as a control variable CEO com-pensation was a significant predictor of both stra-tegic and tactical competitive actions. Scholarsmight examine in more depth the influence of in-centive pay on the mix of competitive actions,along with the influence of CEO, TMT, and boardownership (Balkin & Gomez-Mejia, 1990; Sanders,2001). Governance structures such as boards, own-ers, and compensation structures may have an in-tegrated or configurative influence on the mix offirm competitive actions.

Institutional investors are likely to be most con-cerned with the actions of the largest firms in theirportfolios. However, rivalries exist in industriesthat do not include Fortune 500 firms, and the levelof involvement of institutional owners may vary inthe case of such rivalries. In fact, there are alsorivalries in very small and emerging industries. Therisks and rewards of engaging in competitive ac-tions in such industries are likely different, andtherefore the role of institutional investors mightvary. Our sample was also based on firms head-quartered in the United States, because data weremore readily available and competitive activity waseasier to identify. Future research in this area, how-ever, should examine the competitive actions ofmultinational firms from countries with differentgovernance systems.

Conclusion

In sum, this work contributes to the managementliterature by examining the influence of differentinstitutional owners on the competitive actionstaken by firms in which they hold shares. Ourstudy integrates theory from both corporate gover-nance and competitive dynamics, showing thatownership structure provides a powerful incentive,or disincentive, to engage in strategic and tacticalcompetitive actions. We also uncover a previouslyunexplored mechanism by showing how differentowners may influence the interests of other ownersregarding the competitive actions taken by firms.We hope this study encourages future research inthis area examining different forms of ownership(e.g., family, private investors), other types of firms(e.g., young firms), and more complex interfirmdynamics (e.g., response likelihood, competitiveaction sequencing).

REFERENCES

Abarbanell, J. S., Bushee, B. J., & Raedy, J. S. 2003. Insti-tutional investor preferences and price pressure: Thecase of corporate spin-offs. Journal of Business, 76:233–262.

Ahuja, G., & Katila, R. 2004. Where do resources comefrom? The role of idiosyncratic situations. StrategicManagement Journal, 25: 887–907.

Ali, A., Durtschi, C., Lev, B., & Trombley, M. 2004.Changes in institutional ownership and subsequentearnings announcement abnormal returns. Journalof Accounting, Auditing, and Finance, 19: 221–248.

Balkin, D. B., & Gomez-Mejia, L. R. 1990. Matching com-pensation and organizational strategies. StrategicManagement Journal, 11: 153–169.

Baysinger, B. D., Kosnik, R. D., & Turk, T. A. 1991. Effectsof board and ownership structure on corporate R&Dstrategy. Academy of Management Journal, 34:205–214.

Bennett, J. A., Sias, R. W., & Starks, L. T. 2003. Greenerpastures and the impact of dynamic institutionalpreferences. Review of Financial Studies, 16: 1203–1238.

Bernstein, A. 2009. AMD feeling chipper. Record News,February 19: E6.

Bogle, J. 2005. The ownership of corporate America—Rights and responsibilities. Remarks at the 20thMeeting of the Council of Institutional Investors,Washington, DC.

Bushee, B. J. 1998. The influence of institutional inves-tors on myopic R&D investment. Accounting Re-view, 73: 305–333.

Bushee, B. J. 2000. A taxonomy of institutional investors.Investor Relations Quarterly, 2(4): 13–18.


Bushee, B. J. 2001. Do institutional investors prefer near-term earnings over long-run value? ContemporaryAccounting Research, 18: 207–246.

Bushee, B. J. 2004. Identifying and attracting the “right”investors: Evidence on the behavior of institutionalinvestors. Journal of Applied Corporate Finance,16(4): 28–35.

Bushee, B. J., & Noe, C. F. 2000. Corporate disclosurepractices, institutional investors, and stock returnvolatility. Journal of Accounting Research, 38: 171–202.

Carpenter, M. A., & Sanders, W. G. 2002. Top manage-ment team composition: The missing link betweenCEO pay and firm performance? Strategic Manage-ment Journal, 23: 367–375.

Certo, S. T., & Semadeni, M. 2006. Strategy research andpanel data: Evidence and implications. Journal ofManagement, 32: 449–471.

Chen, M. J. 1996. Competitor analysis and interfirm ri-valry: Toward a theoretical integration. Academy ofManagement Review, 21: 100–134.

Chen, M. J. 2009. Competitive dynamics research: Aninsider’s odyssey. Asia Pacific Journal of Manage-ment, 26: 5–25.

Chen, M. J., & Macmillan, I. C. 1992. Nonresponse anddelayed-response to competitive moves: The roles ofcompetitor dependence and action irreversibility.Academy of Management Journal, 35: 539–570.

Chen, M. J., Smith, K. G., & Grimm, C. M. 1992. Actioncharacteristics as predictors of competitive re-sponses. Management Science, 38: 439–455.

Chen, M. J., Su, K. H., & Tsai, W. 2007. Competitivetension: The awareness-motivation-capability per-spective. Academy of Management Journal, 50:101–118.

Chen, X., Harford, J., & Li, K. 2007. Monitoring: Whichinstitutions matter? Journal of Financial Econom-ics, 86: 279–305.

Cool, K., Roller, L. H., & Leleux, B. 1999. The relativeimpact of actual and potential rivalry on firm prof-itability in the pharmaceutical industry. StrategicManagement Journal, 20: 1–14.

Daily, C. M., Dalton, D. R., & Cannella, A. A. 2003. Cor-porate governance: Decades of dialogue and data.Academy of Management Review, 28: 371–382.

Dalton, D., Daily, C. M., Certo, S. T., & Roengpitya, R.2003. Meta-analysis of financial performance andequity: Fusion or confusion? Academy of Manage-ment Journal, 46: 13–26.

Dalton, D. R., Hitt, M. A., Certo, S. T., & Dalton, C. M.2007. The fundamental agency problem and its mit-igation: Independence, equity, and the market forcorporate control. Academy of Management An-nals, 1(1): 1–64.

David, P., Bloom, M., & Hillman, A. 2007. Investor activ-

ism, managerial responsiveness, and corporate so-cial performance. Strategic Management Journal,28: 91–100.

David, P., Hitt, M. A., & Gimeno, J. 2001. The influence ofactivism by institutional investors on R&D. Acad-emy of Management Journal, 44: 144–157.

David, P., O’Brien, J. P., Yoshikawa, T., & Delios, A. 2010.Do shareholders or stakeholders appropriate therents from corporate diversification? The influenceof ownership structure. Academy of ManagementJournal, 53: XXX–XXX.

Del Guercio, D., Seery, L., & Woidtke, T. 2005. Do boardmembers pay attention when institutional investors“just vote no”? Journal of Financial Economics, 90:84–103.

Del Guercio, D., & Tkac, P. A. 2002. The determinants ofthe flow of funds of managed portfolios: Mutualfunds versus pension funds. Journal of Financialand Quantitative Analysis, 37: 523–557.

Derfus, P. J., Maggitti, P. G., Grimm, C. M., & Smith, K. G.2008. The Red Queen effect: Competitive actions andfirm performance. Academy of Management Jour-nal, 51: 61–80.

Dushnitsky, G., & Lenox, M. J. 2005. When do firmsundertake R&D by investing in new ventures? Stra-tegic Management Journal, 26: 947–965.

Edmans, A. 2009. Blockholder trading, market efficiency,and managerial myopia. Journal of Finance, 64:2481–2513.

Eisenhardt, K. M. 1989. Agency theory: An assessmentand review. Academy of Management Review, 14:57–74.

Ferrier, W. J. 2001. Navigating the competitive land-scape: The drivers and consequences of competitiveaggressiveness. Academy of Management Journal,44: 858–877.

Ferrier, W. J., & Lee, H. 2002. Strategic aggressiveness,variation, and surprise: How the sequential patternof competitive rivalry influences stock market re-turns. Journal of Managerial Issues, 14(2): 162–180.

Ferrier, W. J., Smith, K. G., & Grimm, C. M. 1999. The roleof competitive action in market share erosion andindustry dethronement: A study of industry leadersand challengers. Academy of Management Journal,42: 372–388.

Filatotchev, I., & Toms, S. 2006. Corporate governanceand financial constraints. Journal of ManagementStudies, 43: 407–433.

French, J. R. P., & Raven, B. 1959. The bases of socialpower. In D. Cartwright (Ed.), Studies in social pow-er: 61–73. Ann Arbor: University of Michigan Press.

Froot, K. A., Scharfstein, D. S., & Stein, J. C. 1992. Herdon the street: Informational inefficiencies in a marketwith short-term speculation. Journal of Finance, 47:1461–1484.


Galbraith, J. R., & Kazanjian, R. K. 1986. Strategy imple-mentation: Structure, systems and process. St.Paul: West Publishing.

George, G., Wiklund, J., & Zahra, S. A. 2005. Ownershipand the internationalization of small firms. Journalof Management, 31(2): 210–233.

George, J. M., & Jones, G. R. 2000. The role of time intheory and theory building. Journal of Manage-ment, 26: 657–684.

Gillan, S. L., & Starks, L. T. 2007. The evolution of share-holder activism in the United States. Journal of Ap-plied Corporate Finance, 19: 55–73.

Grinstein, Y., & Michaely, R. 2005. Institutional holdingsand payout policy. Journal of Finance, 60: 1389–1426.

Gutierrez, R. C., & Kelley, E. K. 2008. Institutional herd-ing: Destabilizing buys, stabilizing sells. Workingpaper no. 1107523, SSRN.

Hambrick, D. C., Cho, T. S., & Chen, M. J. 1996. Theinfluence of top management team heterogeneity onfirms’ competitive moves. Administrative ScienceQuarterly, 41: 659–684.

Hartzell, J. C., & Starks, L. T. 2003. Institutional investorsand executive compensation. Journal of Finance,63: 2351–2374.

Higgins, M. C., & Gulati, R. 2006. Stacking the deck: Theeffects of top management backgrounds on investordecisions. Strategic Management Journal, 27:1–25.

Hilbe, J. M. 2007. Negative binomial regression. Cam-bridge, U.K.: Cambridge University Press.

Hitt, M. A., Ireland, R. D., & Hoskisson, R. E. 2009.Strategic management: Concepts and cases. Ma-son, OH: Southwestern.

Hopkins, H. D. 2003. The response strategies of dominantU.S. firms to Japanese challengers. Journal of Man-agement, 29(1): 5–25.

Hoskisson, R. E., Hitt, M. A., Johnson, R. A., & Grossman,W. 2002. Conflicting voices: The effects of institu-tional ownership heterogeneity and internal gover-nance on corporate innovation strategies. Academyof Management Journal, 45: 697–716.

Hsu, G. C. M., & Koh, P. S. 2005. Does the presence ofinstitutional investors influence accruals manage-ment? Evidence from Australia. Corporate Gover-nance: An International Review, 13: 809–823.

Ingley, C., & van der Walt, N. T. 2004. Corporate gover-nance, institutional investors, and conflicts of inter-est. Corporate Governance: An International Re-view, 12: 534–551.

Jensen, M. C. 1993. The modern industrial revolution,exit, and the failure of internal control systems. Jour-nal of Finance, 48: 831–880.

Jensen, M. C., & Meckling, W. H. 1976. Theory of thefirm: Managerial behavior, agency costs and owner-

ship structure. Journal of Financial Economics, 3:305–360.

Johnson, R. A., & Greening, D. W. 1999. The effects ofcorporate governance and institutional ownershiptypes on corporate social performance. Academy ofManagement Journal, 42: 564–576.

Ke, B., & Petroni, K. 2004. How informed are activelytrading institutional investors? Evidence from theirtrading behavior before a break in a string of consec-utive earnings increases. Journal of Accounting Re-search, 42: 895–927.

Ke, B., & Ramalingegowda, S. 2005. Do institutional in-vestors exploit the post-earnings announcementdrift? Journal of Accounting and Economics, 39:25–53.

Ketchen, D. J., Snow, C. C., & Hoover, V. L. 2004. Re-search on competitive dynamics: Recent accom-plishments and future challenges. Journal of Man-agement, 30: 779–804.

Khorana, A. 1996. Top management turnover: An empir-ical investigation of mutual fund managers. Journalof Financial Economics, 40: 403–427.

Kochhar, R., & David, P. 1996. Institutional investors andfirm innovation: A test of competing hypotheses.Strategic Management Journal, 17: 73–84.

Koh, P. S. 2007. Institutional investor type, earningsmanagement and benchmark beaters. Journal of Ac-counting and Public Policy, 26: 267–299.

Laing, J. R. 2007. Insiders, look out! Barron’s Weekly,February 19: 27–30.

Laverty, K. J. 1996. Economic “short-termism”: The de-bate, the unresolved issues, and the implications formanagement practice and research. Academy ofManagement Review, 21: 825–860.

Laverty, K. J. 2004. Managerial myopia or systemic short-termism? The importance of managerial systems invaluing the long term. Management Decision, 42:949–962.

Lerner, J. 2005. The new new financial thing: The originsof financial innovations. Journal of Financial Eco-nomics, 79: 223–255.

Loescher, S. M. 1984. Bureaucratic measurement, shut-tling stock share, and shortened time horizons: Im-plications for economic growth. Quarterly Reviewof Economics and Business, 24: 1–23.

Marginson, D., & McAulay, L. 2008. Exploring the debateon short-termism: A theoretical and empirical anal-ysis. Strategic Management Journal, 29: 273–292.

Matsumoto, D. A. 2002. Management’s incentives toavoid negative earnings surprises. Accounting Re-view, 77: 483–514.

Miller, D., & Chen, M. J. 1996. The simplicity of compet-itive repertoires: An empirical analysis. StrategicManagement Journal, 17: 419–439.


Ocasio, W. 1997. Towards an attention-based view of thefirm. Strategic Management Journal, 18: 187–206.

Parrino, F., Sias, R., & Starks, L. T. 2003. Voting withtheir feet: Institutional ownership changes aroundCEO turnovers. Journal of Financial Economics, 68:3–46.

Penner-Hahn, J., & Shaver, J. M. 2005. Does internationalresearch and development increase patent output?An analysis of Japanese pharmaceutical firms. Stra-tegic Management Journal, 26: 121–140.

Pollack, T. G., Rindova, V. P., & Maggitti, P. G. 2008.Market watch: Information and availability cascadesamong the media and investors in the U.S. IPO mar-ket. Academy of Management Journal, 51: 335–358.

Porac, J. F., Thomas, H., Wilson, F., Paton, D., & Kanfer,A. 1995. Rivalry and the industry model of Scottishknitwear producers. Administrative Science Quar-terly, 40: 203–227.

Porter, M. E. 1992. Capital choices: Changing the wayAmerica invests in industry. Boston: Harvard Busi-ness School.

Puckett, A., & Yan, X. S. 2008. The interim trading skillsof institutional investors. Working paper no.1107953, SSRN.

Rappaport, A. 2005. The economics of short-term perfor-mance obsession. Financial Analysts Journal, 61(3):65–79.

Ryan, L. V., & Schneider, M. 2003. Institutional investorpower and heterogeneity: Implications for agencyand stakeholder theories. Business and Society, 42:398–429.

Samuel, C. 2000. Does shareholder myopia lead to man-agerial myopia? A first look. Applied Financial Eco-nomics, 10: 493–505.

Sanders, W. G. 2001. Behavioral responses of CEOs tostock ownership and stock option pay. Academy ofManagement Journal, 44: 477–492.

Schleifer, A., & Vishny, R. W. 1997. A survey of corpo-rate governance. Working paper series no. 5554,National Bureau of Economic Research.

Schnatterly, K., Shaw, K. W., & Jennings, W. W. 2007.Information advantages of large institutional owners.Strategic Management Journal, 29: 219–227.

Shaver, J. M. 2006. Interpreting empirical findings. Jour-nal of International Business Studies, 37: 451–452.

Shrout, P. E., & Fleiss, J. L. 1979. Intraclass correlations:Uses in assessing rater reliability. PsychologicalBulletin, 86: 420–428.

Smith, K. G., Grimm, C. M., Chen, M. J., & Gannon, M. J.1989. Predictors of response time to competitivestrategic actions: Preliminary theory and evidence.Journal of Business Research, 18: 245–258.

Smith, K. G., Grimm, C. M., Gannon, M. J., & Chen, M. J.1991. Organizational information processing, com-

petitive responses, and performance in the U.S. do-mestic airline industry. Academy of ManagementJournal, 34: 60–85.

Stein, J. C. 1989. Efficient capital markets, inefficientfirms: A model of myopic corporate behavior. Quar-terly Journal of Economics, 104: 655–669.

Tihanyi, L., Johnson, R. A., Hoskisson, R. E., & Hitt, M. A.2003. Institutional ownership differences and inter-national diversification: The effects of boards of di-rectors and technological opportunity. Academy ofManagement Journal, 46: 195–211.

Uotila, J., Maula, M., Keil, T., & Zahra, S. A. 2009. Ex-ploration, exploitation, and firm performance: Anal-ysis of S&P 500 corporations. Strategic Manage-ment Journal, 30: 221–231.

Villalonga, B., & Amit, R. 2006. How do family owner-ship, control, and management affect firm value?Journal of Financial Economics, 80: 385–417.

Wermers, R. 1999. Mutual fund herding and the impacton stock prices. Journal of Finance, 54: 581–622.

Westphal, J. D. 1999. Collaboration in the boardroom:Behavioral and performance consequences of CEO-board social ties. Academy of Management Jour-nal, 42: 7–24.

Yan, X., & Zhang, Z. 2009. Institutional investors andequity returns: Are short-term institutions better in-formed? Review of Financial Studies, 22: 893–924.

Yu, T., & Cannella, A. A. 2007. Rivalry between multina-tional enterprises: An event history approach. Acad-emy of Management Journal, 50: 665–686.

Yu, T., Subramaniam, M., & Cannella, A. A. 2009. Rivalrydeterrence in international markets: Contingenciesgoverning the mutual forbearance hypothesis. Acad-emy of Management Journal, 52: 127–147.

APPENDIX

Classification of Institutional Investors

Bushee (1998, 2003, 2004) categorized institutionalinvestors in accordance with Porter’s (1992) descrip-tions. Using factor and cluster analysis, this method par-simoniously classifies owners into three types: dedi-cated, transient, and quasi-indexer. These classificationshave been used in the accounting literature (Abarbanell,Bushee, & Raedy, 2003; Bushee & Noe, 2000; Ke &Petroni, 2004) but are only recently being applied in themanagement literature (Higgins & Gulati, 2006). Thesecategories may, however, hold value for strategic man-agement research because they delineate groups of insti-tutional investors whose incentives for influencing firmstrategies are reasonably homogeneous (Bushee & Noe,2000).

This classification begins with nine variables that de-scribe the past investment behavior of institutional own-ers. Four of these variables measure diversification of afocal institutional investor’s portfolio. Concentration is


the average percentage of the institutional investor’s totalequity holdings that resides in each firm in its portfolio.Average percentage holding is the average size of theinstitutional investor’s ownership position in firms in itsportfolio. Large block percentage holding is the percent-age of the institutional investor’s total portfolio that isinvested in firms in which it has greater than 5 percentownership. Herfindahl is the sum of the squared percent-age of ownership in each firm in the institutional inves-tor’s portfolio.

Two variables capture the extent of the institutionalowner’s portfolio turnover. Turnover is the absolutechange in the institutional investor’s ownership posi-tions in each quarter divided by the change in its totalequity for all firms in its portfolio. Stability is the per-centage of the institutional investor’s total portfolio hold-ings that reside in firms it has held continuously for twoyears.

The final three variables measure the institutional in-vestor’s trading sensitivity to current earnings. Earningssensitivity 1 is a ratio of change in the institution’s hold-ings in a given firm in each quarter, divided by that firm’schange in quarterly earnings announced during the quar-ter, for each firm in the institution’s portfolio. Earningssensitivity 2 represents the difference between the aver-age change in the earnings of firms in which the institu-tion increased and decreased its holdings. Earnings sen-sitivity 3 is the difference between the institution’schange in its holdings of firms with positive quarterlyearnings and negative quarterly earnings.

These nine variables condense into three factors: port-folio diversification, portfolio turnover, and trading sen-sitivity. A k-means cluster analysis on these three factorsallows institutional investors to be separated into groupsthat are consistent with Porter’s (1992) descriptions:“Transient” institutional investors have highly diversi-fied portfolios, high portfolio turnover, and high tradingsensitivity. “Dedicated” institutional investors are justthe opposite, with concentrated portfolios (i.e., low di-versification), low turnover, and almost no trading sen-sitivity to current earnings. Quasi-indexers, the majorityof institutional investors, lie somewhere in the middle.They generally exhibit diversified portfolios, moderateturnover, and low trading sensitivity.

An important feature of this classification system isthat all institutional investors are classified annually, so

they can move between groups. This yearly assessmentallows principal interests to change over time. In oursample, dedicated institutional investors moved to thetransient category 17 percent of the time and to the quasi-indexer category 9 percent of the time in year-on-yearanalysis (they remained dedicated 74 percent of thetime). Transient institutional investors moved to the ded-icated category 2 percent of the time and to the quasi-indexer category 28 percent of the time (remaining tran-sient 70 percent of the time). So there is also some changein principal interests over time.

Brian L. Connelly ([email protected]) is an assis-tant professor of management and Lowder Research Fel-low at Auburn University. He received his Ph.D. fromTexas A&M University. His research interests includecorporate governance, international strategy, and com-petitive dynamics.

Laszlo Tihanyi ([email protected]) is an associate pro-fessor of management and Mays Research Fellow in theMays Business School at Texas A&M University. He re-ceived his Ph.D. from Indiana University. His researchinterests include international strategies, corporate gov-ernance in multinational firms, institutional changes,and organizational adaptation in emerging economies.

S. Trevis Certo ([email protected]) is an associate pro-fessor of management and Dean’s Council of 100 Distin-guished Scholar in the W.P. Carey School of Business atArizona State University. He received his Ph.D. fromIndiana University. His research interests include corpo-rate governance, initial public offerings, and researchmethodology.

Michael A. Hitt ([email protected]) is a distin-guished professor of management and holds the Joe B.Foster Chair in Business Leadership at Texas A&M Uni-versity. He received his Ph.D. from the University ofColorado. His research interests include managing re-sources in organizations, international strategy, corpo-rate governance, and strategic entrepreneurship.


Copyright of Academy of Management Journal is the property of Academy of Management 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.

marching to the beat of different drummers: the …people.tamu.edu/~ltihanyi/inscompamj.pdf ·...

Documents