AOM JOURNAL6

8/3/2019 AOM JOURNAL6

1/20

EFFECTS OF SOCIAL CAPITAL AND POWER ON SURVIVINGTRANSFORMATIONAL CHANGE: THE CASE OF INITIAL

PUBLIC OFFERINGS

HARALD M. FISCHERUniversity of Connecticut

TIMOTHY G. POLLOCKPennsylvania State University

We examined how social capital and the power of venture capitalists and founder-CEOs affect IPO firm survival. Using data from 218 U.S. initial public offeringsconducted in 1992, we found that average management team tenure and an IPO dealsnetwork embeddedness decreased the likelihood of failure during a firms first fiveyears as a public entity. Founder-CEO presence at the time of an IPO interacted withCEO ownership to decrease the likelihood of failure, and CEO ownership and venturecapitalist ownership concentration also decreased that likelihood.

Evolutionary perspectives on organizations areoften accompanied by the presumption that organ-izations face an increased risk of failure early intheir lives owing to liabilities of newness (Aldrich,1999; Stinchcombe, 1965). As organizational goalsand patterns of activity become routinized overtime, increased reliability in performance and ac-countability for actions taken enhance a firms sur-vival chances (Hannan & Freeman, 1984). Thisprocess, however, also generates strong inertialpressures that not only discourage organizational

change, but also, because of the potential for dis-ruption to existing internal and external routines,make change hazardous. Research findings suggestthat even in older, better-established firms, signifi-cant transformational events during their life cyclescan effectively reset the clock and reintroducerisks associated with the liability of newness asfirms struggle to adapt strategies, internal opera-tional and administrative processes, and/or exter-nal ties and relationships (Amburgey, Kelly, &Barnett, 1993). Thus, transformational change inorganizational operations decreases efficiency and

increases failure rates, at least in the short term, asresources and attention are diverted from normal,routinized operating functions to processes involv-ing adaptation and reorientation (Haveman, 1992).

When organizations do survive initial transitionperiods by successfully managing any disruptiveeffects, transformation may ultimately be consid-ered adaptive and enhance the firms competitivecapabilities (Amburgey et al., 1993). Supportingresearch has found that risks associated with reset-ting the liability of newness clock are minimized

when a transforming firm uses and leverages exist-ing competencies (e.g., Haveman, 1992) or when aspecific type of change occurs so routinely that afirm develops a competency in a given form ofadaptation (e.g., Amburgey & Miner, 1992). Miner,Amburgey, and Stearns (1990) also suggested thatan organization may have certain characteristics orresources at the time of a transformational eventthat serve as shields by providing continued ac-cess to financial resources and/or legitimacy duringa period of reorientation. In their study of thefounding, transformation, and dissolution of Finn-

ish newspapers, Miner, Amburgey, and Stearnsfound that the presence or absence of a link be-tween a newspaper and a powerful political partyinfluenced the newspapers ability to survive corechange.

By focusing on a single interorganizational link,however, Miner and her colleagues were unable toconsider how broader network structures, and thenature of the ties within these networks, could alsoenhance a firms ability to survive a transforma-tional event. In addition, extant research has notyet explored how intraorganizationalties can also

Both authors contributed equally to this work. Wewould like to thank Ted Baker, Mason Carpenter, AnneMiner, Hitoshi Mitsuhashi, Huggy Rao, and ViolinaRindova for their helpful comments on earlier versions.We would also like to thank associate editor MarshallSchminke and three anonymous AMJreviewers for theirinvaluable comments and suggestions, as well as PatMaggitti and Johanna Patrick for assistance in data col-lection. This study was supported by a grant from theUniversity of WisconsinMadison School of BusinessResearch Committee.

Academy of Management Journal

2004, Vol. 47, No. 4, 463481.

463


2/20

serve as potent transformational shielding mecha-nisms. Finally, little research on organizationalchange and survival has considered the importantrole that the involvement of key actors within anorganization plays in helping it through a transfor-mational period, and how the power associatedwith the concentration of ownership in the hands

of these individuals can help protect the organiza-tion from potentially harmful environmental pres-sures. In this article, we begin to address theseissues by introducing a third kind of protection,transformational shields based on sociopoliticalforces, and by demonstrating how the power andcommitment of key actors, as well as an organiza-tions internal and external social capital, enhancea firms ability to survive a transformational eventthat may reset the liability of newness clock (Am-

burgey et al., 1993).Prior to evaluating how and what organizational

characteristics present at the time of an event pro-tect a firm from the potentially deleterious effectsof the transformation process, it is critical to deter-mine what sorts of changes are of sufficient magni-tude and consequence to likely reexpose a firm tothe risks associated with the liability of newness.Using theoretical perspectives drawn primarilyfrom organizational ecology (e.g., Barnett & Carroll,1995; Hannan & Freeman, 1984), researchers haveattempted to differentiate core from peripheralchange, arguing that changes to four structural fea-turesmission, authority structure, technology,and marketing strategyare those most likely toreset the liability of newness clock. This approachis somewhat limiting, however, as what constitutesa core or peripheral change, as well as its degree ofcriticality, can vary among firms and populations(Aldrich, 1999; Dobrev, Kim, & Carroll, 2003).

Aldrich (1999) suggested an alternative, nonordi-nal set of criteria for evaluating the degree to whicha given event constitutes a significant transforma-tion. He argued that events resulting in changes tocertain organizational aspects were especially note-worthy. These key changes include the following:changes in organizational goals (for instance, a ma-

jor change in domain, such as an HMO going fromnonprofit to profit status or entering a new product-market); changes in boundaries (such as expansionthrough merger or contraction through divestiture);and changes in activity systems (for instance, adop-tion of new administrative, human resource, ortechnological systems). These events lead to devel-oping new knowledge, new skills, and/or the cre-ation and implementation of new strategic goalsand objectives. Disruption to existing organiza-tional routines (Amburgey et al., 1993), the need tolearn new roles and skill sets (Aldrich, 1999), and

the potential cascade of additional, unintendedchanges resulting from a triggering transforma-tional event are what prove deleterious and likelyto re-expose firms to short-term increases in failurerates.

Employing the framework explicated by Aldrich,we argue that one highly significant and nonrepeat-

able event during many organizations life cycles isthe transformation from private to public owner-ship. Although an initial public offering (IPO) of-fers a number of benefits to firms that successfullynavigate the transition (see Husick and Arrington[1998] and Price Waterhouse [1995] for discussionsof these benefits), the transformation also bringswith it a number of costs and risks. Like an HMOschange from nonprofit to for-profit status, thechange from private to public status necessitates achange in goals, as management must deal withnew kinds of investors who have different objec-

tives and time horizons than early investors, andthus different needs that must be served. In addi-tion, management must also determine how to de-ploy the new financial resources available to theirfirm by expanding existing strategic initiatives orimplementing new strategies for it that can involvea range of activities, such as engaging in acquisi-tions that change the boundaries of the firm orentering new product or geographic markets withwhich the firm has little experience (Husick & Ar-rington, 1998).

IPO firms also undergo a number of changes totheir activity systems that necessitate new learning.An organization will likely be restructured as newexecutives are hired, more formal governance pro-cedures are put in place, and additional staff isadded to handle the reporting requirements of theSecurities and Exchange Commission and interactwith the companys new investors (Husick & Ar-rington, 1998; Price Waterhouse, 1995). Manage-ment must also learn how to deal with reducedflexibility in managerial discretion, increased over-sight from the firms board of directors, greater de-mands from investors for short-term performance,and less tolerance of negative press and perfor-

mance volatility (Price Waterhouse, 1995). As thecompany grows, management must also deal withsignificant cultural changes, as well as change re-sulting from its employees newfound wealth asstock is sold and options are exercised. Some em-ployees may leave the company, while others may

be less willing to continue making the personalsacrifices that were required to get the firm to theIPO stage. Although IPO firms will vary in theintensity with which they experience each of thesespecific changes associated with the triggering IPOevent, all will face the task of adapting their goals,

464 AugustAcademy of Management Journal


3/20

boundaries, and/or activity systems enough to in-cur some reexposure to the risks associated withliabilities of newness.

Over the last several years, IPOs have received asignificant amount of attention in the finance (e.g.,Ibbotson & Ritter, 1995), strategy (e.g., Certo, Covin,Daily, & Dalton, 2001) and organizational (e.g.,

Welbourne & Andrews, 1996) literatures. The pri-mary focus of this research has been on discoveringfactors that impact the initial pricing of IPO stocksand their market returns over various periods oftime. A topic that has received relatively little at-tention, however, is the subject of IPO firm failures.There are only three studies of which we are awarethat have specifically examined the topic of whyIPO firms fail. Platt (1995) used financial liquidityratios in predicting whether IPO firms would go

bankrupt within three years. Welbourne and An-drews (1996) found that the degrees to which IPO

firms valued human resources and used compen-sation schemes based on organizational perfor-mance were positively associated with IPO firmsurvival. And Jain and Kini (2000) found that firmsthat received venture capitalist (VC) backing weremore likely to survive than non-VC backed compa-nies. Finally, in a related study Foster-Johnson,Lewis, and Seward (2001) suggested that long-rununderperformance of IPOs can be attributed toinvestor overoptimism regarding the likelihoodthat IPO firms will survive.

Taken together, these studies suggest that a firmthat has recently undergone an IPO is at a signifi-cant risk of failure, at least in the short term. Whileall of these studies touch, in different ways, on howfinancial, human, and social resources impact theability of an IPO firm to survive, they do not inte-grate the different theories discussed with the

broader organizational literature in this area. In thisstudy, we argue that changing from a privately heldto a publicly traded company constitutes a signifi-cant transformational event (Aldrich, 1999) that re-introduces an IPO firm to the risks associated withthe liability of newness. In the following section,we develop theoretical arguments that suggest IPO

firms may possess sociopolitical resources that canhelp protect them from these risks as they completethe transformation from privately to publiclyowned firms. We tested these arguments using asample of 218 U.S. companies that went public in1992.

THEORY AND HYPOTHESES

Miner and her coauthors defined a transforma-tional shield as an organizational trait that insu-lates an organization against the probability of fail-

ure resulting from transformation (1990: 695). Theauthors differentiated transformational shieldsfrom bufferswhich can be defined as factorsthat insulate or protect an organization from gen-eral environmental pressures that result in the needto change (Aldrich, 1979; Thompson, 1967)byfocusing specifically on protections that come into

play only when significant firm-level change oc-curs. Miner and colleagues overall framework andreview of the related literatures suggested two

broad categories of protection involving resource-and/or legitimacy-based factors that are relevant tofirm change. These transformational shields are de-rived primarily from external sources, and theygive a firm access to slack resources and/or en-hance perceptions of its compliance with socialexpectations or affiliations with important legiti-mating actors.

In this study, we suggest that firms can also pos-

sess a third class of protective shield that arisesfrom both external and internal sociopolitical pro-cesses and resources. Specifically, we argue thatthe presence of a company founder as CEO follow-ing a companys initial public offering, the concen-tration of stock ownership in the hands of the CEOand/or venture capitalists post-IPO, and internaland external social capital, as reflected in key intra-and interorganizational networks, can all protectthe IPO firm during a period of change and enhanceits ability to survive the transformation from pri-vately to publicly held firm. Although some or allof these factors may also change during the yearsfollowing the IPO, their availability at the time thetransformation occurs can have significant im-printing effects (Stinchcombe, 1965) that shapethe evolutionary trajectory of the firm. In a study ofthe telecommunications industry, Noda and Collis(2001) found evidence that heterogeneity in thesuccess of firms was the result of positive feedbackloops that arose from small differences in initialconditions. We argue that similar effects can resultfrom the presence of sociopolitical transforma-tional shields that provide firms with necessaryprotections at critical junctures in the firms devel-

opment and that enhance their chances of surviv-ing transformational events.

Founder-CEO Leadership as a SociopoliticalTransformational Shield

The presence of a founder-CEO at the time of acompanys IPO may enhance the firms survivalchances. Prior research has demonstrated how thepersonal assumptions and perceptions of foundersplay a central role in both the formulation of strat-egy and the manner in which influence will be

2004 465Fischer and Pollock


4/20

distributed within a firm (Baron, Hannan, & Bur-ton, 1999; Boeker, 1988). Nelson (2003) providedevidence that the presence of a founder-CEO at thetime of an IPO results in higher firm valuations, aswell less insider selling and greater retained own-ership by the CEO. Since changes in leadership arealso often associated with shifts in the strategic

direction of a firm (Fredrickson & Westphal, 2001),the presence of a founder-CEO at the time of theIPO may also result in a continuation or extensionof the firms existing strategy, thus reducing therisks associated with radical strategic shifts requir-ing skills or knowledge the firm does not possess(Aldrich, 1999).

A founder-CEO may also have more ability than anonfounder CEO to lead a company through itstransformational period because his or her founderstatus reduces conflict and political battles withinthe firm. Recent research on the psychological

bases of trust (e.g., Tyler & Degoey, 1996) has sug-gested that the degree to which individuals trustleaders significantly increases the likelihood theywill accept their decisions and comply with theirdirectives. Even if individuals do not personallyagree with a decision, they will still comply with itif they believe that a leaders motivations are good(Brockner, Siegel, Daly, Tyler, & Martin, 1997).

Recent theorizing also suggests that the interper-sonal trust resulting from belief in the good inten-tions and reliability of others, and the maintenanceof a shared language and pattern of effective com-munication among organization members, lead topositive firm outcomes (Bolino, Turnley, & Blood-good, 2002; Nahapiet & Ghoshal, 1998). Given theenormous personal investments of time, energy,and money a founder typically expends in gettinghis or her company to the IPO stage (Nelson, 2003;Smith & Miner, 1983), a founder-CEO is likely to bemore committed to completing the transformationand seeing the company through the transition pe-riod than an outsider CEO with fewer sunk costs(Moon, 2001; Ross & Staw, 1993). In addition, sincethe founder-CEO has been involved in the growthand success of the company since its inception, her

or his personal identification with the company(Tyler & Degoey, 1996) can be expected to begreater than that of a mercenary brought in fromthe outside. Thus, agency costs may also be lowerfor firms with founder-CEOs. Indeed, Nelson (2003)noted that the extraordinary commitment of someindividuals to their firms can act as an anti-agency cost by making their drain on organiza-tional resources lighter. Finally, employees havehad opportunities to develop social bonds withfounder-CEOs through their history of working to-gether. Such social bonds have also been shown to

increase individuals perceptions of the trustwor-thiness of leaders (Tyler & Degoey, 1996; Tyler &Lind, 1992). Taken together, these arguments sug-gest that the presence of a founder-CEO can reduceconflict and political wrangling within an IPO firm,providing a relatively stable internal environmentthat can enhance the IPO firms ability to survive its

transformational period. Thus, we hypothesizethat:

Hypothesis 1. The presence of a founder-CEOat the time of a firms transformation fromprivate to public status will reduce the likeli-hood of the IPO firms failure.

Ownership as a Sociopolitical TransformationalShield

CEO ownership. The proportion of outstanding

stock owned by a companys CEO upon completionof its IPO can also serve as a valuable transfor-mational shield. Recent research (e.g., Pollock,Fischer, & Wade, 2002; Tosi et al., 1999) has shownthat direct CEO ownership of a companys stock ismore effective than stock options and other formsof incentive compensation in aligning managementand shareholder interests and in reducing self-serv-ing CEO behaviors. Thus, CEOs with large owner-ship stakes may have a greater interest in makingsure the resources of their companies are used pru-dently and efficiently, and in so doing they reduce

the likelihood that newly public firms will fail(Hitt, Bierman, Shimizu, & Kochar, 2001). In addi-tion, research on escalation of commitment hasfound that managers are more likely to maintainunsuccessful courses of action if they have low jobsecurity (Fox & Staw, 1979). Fox and Staw foundthat the more secure managers were in their jobs,the more willing they were to consider changes inpolicy. Since greater ownership provides a CEOwith more discretion in decision making (Finkel-stein & Hambrick, 1990; Tosi et al., 1999) andgreater power within an organization (Finkelstein,1992), the larger a CEOs post-IPO ownership stake,

the less the CEO may be worried about a short-termperformance setback leading to the loss of his or herposition, and the more open the CEO will be toadaptations in existing organizational strategiesand practices that may be necessary to enhance thelong-term viability of the firm. We therefore hy-pothesize that:

Hypothesis 2. The greater the proportion of anIPO firms stock owned by its CEO followingthe transformation from private to public sta-tus, the lower the likelihood of firm failure.



5/20

Venture capitalist ownership. Another organi-zational actor likely to possess substantial stockownership following an IPO and that has also com-mitted substantial time, energy, and financial re-sources to the IPO firm, is the venture capitalist(Bygrave & Timmons, 1992; Jain & Kini, 2000; Lee &Wahal, in press; Sahlman, 1990). The relationship

between IPO firms and VCs goes beyond the simpleprovision of financial resources. A venture capital-ist can serve as a valuable sounding board when afirm is formulating and implementing corporatestrategies and is frequently involved in helping thefirm recruit experienced personnel and acquiringother resources (Jain & Kini, 2000). Thus, venturecapitalists are often instrumental in establishingthe management teams that are in place at the timea company goes public. As a result, they may havepersonal commitments to seeing the company suc-ceed that goes beyond just their financial interest in

it (Moon, 2001; Ross & Staw, 1993).To the extent that a significant proportion of anIPO firms outstanding stock is concentrated in thehands of a single or small group of venture capital-ists following the initial offering, the VCs are in aposition to protect the leaders of the IPO firmfrom external pressures for significant short-termimprovements in performance. This protection en-hances the leaders discretion (Finkelstein & Ham-

brick, 1990) and gives them the time necessary toadapt to their new environment. If ownership isdispersed among a large number of VCs, individualVC ownership is less likely to be effective as atransformational shield, even if the total percentageof stock owned by VCs is high. The individualcommitment of a given venture capitalist to thecompany may be lower than it would be if thatperson held a larger percentage of the shares, com-munication and coordination costs among the own-ers may be greater, the interests of the various ac-tors are more likely to diverge, and thus ownershippower will be diluted (Tosi et al., 1999; Useem,1996). We therefore hypothesize the following:

Hypothesis 3. The greater the concentrated

ownership of stock by venture capitalists fol-lowing a firms transformation from private to public status, the lower the likelihood of theIPO firms failure.

Although the effects of CEO and VC ownershipmay be additive (that is, more of each type of own-ership reduces the probability of failure indepen-dently), it is also possible that ownership by theseactors can have multiplicative effects. If only afirms CEO or only its venture capitalists own asubstantial block of stock, an inequity exists in the

formal power these actors have in determining thestrategic direction of the company, should differ-ences in opinion arise. Such a situation could re-sult in intense political maneuvering and battles forcorporate control (Westphal, 1998). If the CEO andVCs have relatively equivalent and substantialownership power, however, neither can dominatethe other, and they are more likely to negotiate aresolution to differences (Pfeffer, 1992). This argu-ment suggests the following hypothesis:

Hypothesis 4. The greater the concentratedownership of stock by both CEO and venturecapitalists following a firms transformation from private to public status, the lower thelikelihood of the IPO firms failure.

It is also possible that the commitment and trustassociated with the presence of a founder-CEO andthe control afforded by significant stock ownershipmay both be necessary for these factors to providean effective sociopolitical transformational shield.If a founder gives up the majority of a companysstock during early rounds of financing, and/or thefounders retained ownership is diluted by a largeinitial public offering, he or she will have morelimited power and discretion (Finkelstein, 1992;Finkelstein & Hambrick, 1990), even if the formalposition of authority as CEO is maintained. In ad-dition, ownership power may mean little if em-

ployees do not trust a CEO and his or her motiva-tions. If the CEO is not a founder, the relational andcognitive social capital (Bolino et al., 2002) devel-oped prior to a transformation may be destroyed,and political maneuvering within the companymay increase as individuals vie to curry favor withpowerful constituencies and form coalitions to ad-vance their own interests and positions (Tyler &DeGoey, 1996; Westphal, 1998). Cannella and Ham-

brick (1993) suggested that the disruptive effects ofexecutive departure in conjunction with a changein the formal authority structure can be especially

harmful to a firm. The symbolic message sent toboth internal and external stakeholders by the pres-ence of a powerful founder-CEO with a significantownership stake is that the firms vision, culture,and strategic continuity will be preserved. Thus,we hypothesize that:

Hypothesis 5. The presence of a founder-CEOwho owns a significant percentage of a firmsstock following its transformation from privateto public status will reduce the likelihood ofthe IPO firms failure.



6/20

Internal and External Social Capital asSociopolitical Transformational Shields

Although Miner and her coauthors (1990) con-sidered the value of interorganizational linkages astransformational shields, they focused primarily onsingle ties to powerful actors that could provide (or

withdraw) resources and did not consider how thestructure of broader networks and the nature of theties within those networks could protect a firmfrom potentially negative environmental influ-ences. They also did not consider how links insidean organization might shield it from the pressuresassociated with a significant transformation. Weaddress both of these issues by drawing on socialcapital theory and research to demonstrate how

both intra- and interorganizational links can serveas sociopolitical transformational shields.

Social capital has been defined in a variety ofways by numerous researchers in the social andorganizational sciences (see Adler and Kwon[2002] for a review); however, all of these defini-tions share two common elements: (1) social capitalarises from the structure of relations between andamong actors in a network and (2) an actor has theability to access these network, or social-structural,

benefits (Coleman, 1988). In developing our hy-potheses concerning the effects of social capital onIPO firm survival, we explored two aspectsoneinternal, one externalof an organizations socialstructure that can reduce the potential for conflictand political battles. An important but often over-

looked dimension of social capital is whether net-work relations are internal or external to an or-ganization (Adler & Kwon, 2002). Coleman (1988)asserted that social capital can be defined by itsfunction, but no matter how it is expressed it mustfacilitate the action of actors within a social struc-turebe it among individuals within a firm, or

between a firm and its external stakeholders. Tieswithin organizations have been described asbonding or communal forms of social capital,while external ties have been described as bridg-ing or linking forms of social capital (e.g., Adler

& Kwon, 2002). We argue that both forms of socialcapital affect the survival of IPO firms.

Internal social capital. Links within an organi-zation that enhance cohesiveness and facilitate thepursuit of collective goals can appropriately be de-scribed as social capital (Adler & Kwon, 2002: 21).For an IPO firm, a top management team (TMT)whose members have worked with one another atthe firm for a number of years represents such asource of internal social capital. Working togetherfor substantial periods of time provides a manage-ment team with the opportunity to develop work-

ing patterns, routines, and interpersonal relation-ships that allow it to be more effective in handlingthe discontinuities associated with taking its com-pany public.

Eisenhardt and Schoonhoven (1990) showed thatthe joint work experience of a TMTs memberssignificantly contributed to growth among newly

founded semiconductor firms. These effects werealso found to increase over time, suggesting thepossibility that the composition of a top team at thetime of the IPO may have legacy effects that sig-nificantly affect subsequent survival rates. Carpen-ter (2002) found that average TMT tenure alloweddiverse top teams to benefit from their executivesdifferences and enhance firm performance. Otherresearch on groups (e.g., Bantel & Jackson, 1989)has also suggested that team members who are to-gether longer tend to perform their tasks better,

because they have had the time to develop working

routines and understandings that allow them toleverage the distinctive benefits of their varied backgrounds. This issue can be of special impor-tance to entrepreneurial firms that do not havedeep reservoirs of managerial talent and thereforehave to rely on key individuals to handle majoractivities more or less single-handed. Indeed,Rindova and Kotha (2001), in their study of theability of firms to develop dynamic capabilities,cite the following from a e-mail received from JerryYang, one of the cofounders of Yahoo!: Whilethere are many factors that have enabled us to be inour current leadership position, I would say our

biggest reason is our ability to have put together amanagement team thats stuck together for a longtime (2001: 1274). These arguments are consistentwith the critical elements of the social capital con-struct, and they suggest the following hypothesis:

Hypothesis 6. The greater the average tenure ofa firms top management team at the time ofthe transformation from private to public sta-tus, the lower the likelihood of the IPO firmsfailure.

External social capital. In addition to the com-

munal, or internal, form of social capital that afirms top managers can bring to bear in helping anIPO firm survive the transition from private to pub-lic status, the company also has access to valuableexternal social capital that can be used to enhanceits chance of survival. One of the most importantexternal links an IPO firm has when it goes publicis its tie to the underwriter leading the IPO (Carter& Manaster, 1990; Pollock, Porac, & Wade, 2004).The lead underwriter works closely with the topmanagement of an IPO firm, as well as with itslawyers, its auditors, and the Securities and Ex-



7/20

change Commission (SEC), to guide the firmthrough the IPO process (Husick & Arrington,1998). One of the key functions of the lead under-writer in an IPO is to create a network of investorswho become the IPO firms initial shareholders(Pollock et al., 2004). The relationships that theunderwriter has with these investors can affect the

stability of these networks and the subsequent per-formance and survival potential of the IPO firm.More stable investor networks limit stock price andtrading volatility in the secondary market (Carter &Dark, 1993; Ellis, Michaely, & OHara, 2000); rela-tive stock stability can create positive impressionsof a company that allow it to access resources (Pol-lock, Gulati, & Sadler, 2002). Stable investor net-works also allow a firm to pursue longer-term stra-tegic growth initiatives because they minimizemanagements concerns over investor reactions toshort-term fluctuations in performance (Bushee,

1998).Underwriters can enhance network stability byincluding investors with whom they have embed-ded relationships in the initial IPO deal network(Pollock et al., 2004). Granovetter (1985) arguedthat market relationships vary along a continuum,ranging from those that are instrumental and armslength to those that are embedded in a dense net-work of social ties that promote trust and coopera-tion. The degree to which relationships are eitherembedded or arms length depends on the fre-quency (e.g., Baker, 1984; Granovetter, 1985) andthe concentration (e.g., Baker, 1990; Larson, 1992;Uzzi, 1996) of transactions between the actors in-volved. The more frequent transactions among a setof actors are, and the more the transactions involvethe same set of actors, the more embedded thetransactional relationships. Evidence suggests thatembedded relationships based on prior transactionhistories decrease opportunistic behavior (Uzzi,1996), facilitate information transfer (Larson, 1992;Uzzi, 1996), influence the acquisition and use ofpower (Baker, 1990), build trust between the trans-action partners (Uzzi, 1996), and reduce marketvolatility (Baker, 1984). Embedded relationships

with regular market participants may therefore alsobe helpful to an underwriter as it attempts to createa stable network of investors for an IPO firm.

Underwriters place the vast majority of shares inIPOs with institutional investors (Hanley & Wil-helm, 1995). Many institutional investors partici-pate regularly in the IPO market and thus have thepotential to develop embedded relationships withunderwriters who also participate in this marketregularly (Hanley & Wilhelm, 1995). Institutionalinvestors will be more likely to believe an under-writers claims if they have interacted repeatedly

with that bank in past offerings, and this credibilityreduces investors uncertainty to some extent andincreases their tolerance for variation in the short-term performance of a company undergoing a pe-riod of reorientation following its IPO. Investorswillingness to hold the companys stock for lengthyperiods provides much-needed stability in the

companys stock price, allowing it to focus on stra-tegic and operational initiatives rather than spendexcessive time trying to manage investor and ana-lyst concerns about its performance (Rao & Sivaku-mar, 1999). To the extent that an IPO firms leadunderwriter has the social resources necessary todevelop more embedded deal networks, we hy-pothesize that:

Hypothesis 7. The greater the proportion ofinvestors with embedded ties to the lead un-derwriter in an IPO who are included in thedeal network at the time of the transformation

from private to public status, the lower thelikelihood of the IPO firms failure.

DATA AND METHODS

Sample

Unless otherwise noted, the data for this studywere drawn from the offering prospectuses filed forall U.S. initial public offerings conducted in 1992.As has prior IPO research (e.g., Ritter, 1991; Wel-

bourne & Andrews, 1996), we excluded the follow-ing from the analysis: closed-end mutual funds,real estate investment trusts (REITS), unit offerings,spin-offs, demutualizations of savings banks andinsurance companies, and reverse leveraged buy-outs (LBOs). The final sample contained 245 IPOs.Missing data reduced the sample to 218 IPOs. Theyear 1992 was selected because in this year the IPOmarket was neither overheated nor dormant, andselecting companies from a single year allowed usto control for significant intertemporal fluctuationsin the IPO market (Ibbotson & Ritter, 1995). Amongfirms in our sample, factors such as their averageunderpricing and the proportion delisted within

five years of IPO are consistent with historical av-erages observed in samples drawn from longer timeperiods (e.g., Foster-Johnson, et al., 2001; Ibbotson& Ritter, 1995; Jain & Kini, 2000).

Dependent Variable

IPO firm failure. The dependent variable in thisstudy was a firms failure during the five yearsfollowing its IPO. The five years following an IPOhas been suggested as the time frame during whicha company may be considered a newly public firm.



8/20

After five years IPO firms are considered sea-soned public entities (Foster-Johnson et al., 2001;Loughran & Ritter, 1995; Welbourne & Andrews,1996). As has prior research (Foster-Johnson et al.,2001), we defined IPO firm failure as delisting fromthe primary exchange on which a firm traded witha delisting code between 500 and 585. Delisting

codes indicate the reason an exchange has droppeda firm. Codes between 500 and 585 are associatedwith firm bankruptcy and the inability of a firm tomaintain the minimum size, shareholder number,and stock price requirements for continued listingon the exchange. We constructed a time-varyingdichotomous outcome measure coded 1 during theyear a firm was delisted and 0 otherwise. As wasappropriate for our modeling strategy, a firm wasdropped from the sample after delisting, and theothers remained as right-censored. The data usedto calculate IPO firm failure were drawn from the

CRSP database.

Independent Variables

Founder-CEO presence. This measure was adummy variable coded 1 if one of the originalfounders of a company was the CEO at the time thecompany went public.

CEO ownership. This measure equals the per-centage of shares outstanding that were beneficiallyowned by an IPO firms CEO upon completion ofthe IPO.

VC ownership concentration. Unlike CEO own-ership, which by definition reflects the concentra-tion of stock ownership in the hands of a singleindividual, venture capitalist ownership reflectsownership by multiple entities. In our sample, forfirms with VC backing, the number of VCs thatowned stock in the company following the IPOranged from 1 to 14. To capture the degree to whicha large proportion of an IPO firms shares wereconcentrated in the hands of one or a few VCs, weconstructed a Herfindahl index (Scherer, 1980) bysquaring and summing the percentage of stockownership of each venture capitalist that owned

stock in the company upon completion of the IPO.Average management team tenure. This mea-

sure captured how much experience a top manage-ment teams members had working together at thetime of an IPO. Evidence that a management teamsmembers have worked together for several years isan important indicator that they can function to-gether effectively as a team (Carpenter, 2002). Likeprevious researchers (Carpenter, 2002; Finkelstein& Hambrick, 1990), we calculated average manage-ment team tenure by summing the tenure with anIPO company of each management team member

listed in the prospectus and dividing the sum bythe total number of management team members.

Deal network embeddedness. Institutional in-vestor participation in past deals managed by thelead investment bank for a given IPO was the basisfor determining deal network embeddedness. Fol-lowing prior research that has generated embed-

dedness measures using transaction data (e.g.,Baker, 1990; Uzzi, 1996), we calculated deal net-work embeddedness using a Herfindahl index.First, we identified the lead underwriter for eachIPO in the sample and the institutional investorsthat owned stock in each IPO at the end of thequarter in which the company went public. Wethen determined for each underwriterinstitutionalinvestor pair how frequently the two actors partic-ipated in offerings together. To do so, we collecteddata from IPOs conducted in 1991 on 4,754 suchpairs. For each pair, we identified the number of

offerings (Dij) in which investment bank i partici-pated as a lead manager or comanager and institu-tional investor j owned shares. We also identifiedthe number of deals (Di) that investment bank iparticipated in as a lead or comanager. The concen-tration ratio for each deal network k (DNEk) wasthen calculated with this formula: DNEk (Dijk/Dik)

2. The data used to calculate deal network em-beddedness were drawn from the Compact D SECdatabase.

Control Variables

VC backing. Venture capitalist backing was adummy variable coded 1 if a company receivedventure financing while it was privately held and 0otherwise. Prior research has shown that VC back-ing is negatively associated with IPO firm failure(Jain & Kini, 2000).

Underwriter reputation. When the quality andcharacteristics of an individual, product, or organi-zation are difficult to ascertain, their relationshipswith highly reputable others can provide importantsignals that help reduce these uncertainties in theminds of stakeholders (Podolny, 1994). These types

of relationships are a valuable source of externallegitimacy and thus may also serve as a transforma-tional shield of the type described by Miner and hercoauthors (1990). In keeping with prior research(e.g., Carter & Manaster, 1990; Podolny, 1994), wemeasured underwriter reputation on the basis of aninvestment banks position in the tombstone an-nouncements that officially notify the public of anoffering. A list of underwriters participating in anoffering syndicate appears on every tombstone,listed by status class. Underwriter status classes arecommunally defined by the members of the finan-



9/20

cial community and reinforced over time; thus, aninvestment banks status position in an underwrit-ing syndicate at a given time is the outcome ofsocial ordering processes that have taken place overa long period (Podolny, 1994). We drew the data ontombstone position in 1991 underwriting syndi-cates, which included 3,365 participants represent-

ing 261 investment banks, from Compact Disclo-sures Compact D database. We reverse-coded eachunderwriters status class and divided the resultingnumber by the total number of classes representedin a tombstone. For example, if a tombstone hadthree classes of underwriters, the first class wascoded 1, the second class was coded .67, and thethird class was coded .33. An underwriters repu-tational score equaled the average score across allthe syndicates in which the underwriter partici-pated, multiplied by 100.

Underpricing. The amount of underpricing a

company experiences when its stock begins tradingcan affect stakeholder expectations regarding itspotential, and thus its ability following the IPO toacquire resources (e.g., Jegadeesh, Weinstein, &Welch, 1993; Pollock et al., 2002) that could en-hance its survival chances. Underpricing equaledthe percentage change in a stocks price [(priceendpriceinitial)/(priceinitial)] on the first day the stocktraded on a national exchange. The data used tocalculate underpricing were drawn from CRSP.

Financial condition. All else being equal, thebetter a firms financial condition, the less likely itis to fail. To control for the impact of financialcondition on an IPO firms survival, we includedlagged measures of the firms total sales, net in-come, and annual stock price performance for eachyear it was in the sample. To control for the effectsof extreme values, we transformed total sales intotheir natural logarithms. Because some firms hadno revenues in a given year, a 1 was added to allobservations prior to transformation of the mea-sure. The data used to create these measures weredrawn from the offering prospectuses and COM-PUSTAT. For a few observations (2.6%), we wereonly able to obtain sales and income data for a

portion of the firm-year spells in our sample. Asthese measures varied annually, we used the im-pute command in Stata 7.0 to interpolate valuesfor the missing sales and net income data in theyears data were missing.1

Industry dummy. Companies in different indus-tries could vary systematically on both our inde-

pendent and dependent variables. Prior researchhas suggested that firms involved in more techno-logically complex industries are more prone to fail-ure (Jain & Kini, 2000) and experience higher un-derpricing (Certo et al., 2001). To control for theeffects of industry membership, we constructed adummy variable coded 1 if a firm was in an indus-

try identified as technologically complex by Certoand his colleagues and 0 otherwise.

Industry concentration. Competition in a firmsindustry can also affect its survival (Hannan & Free-man, 1984). To control for the intensity of industrycompetition, we based a Herfindahl concentrationratio for each firm-year observation on the laggedsales for all of the companies in the same two-digitSIC code as an IPO firm. The data used to createthis variable were drawn from COMPUSTAT.

Age of firm at IPO. Firm age at IPO was calcu-lated as the years since incorporation. Younger

firms are subject to a greater likelihood of failure fora variety of reasons (Hannan & Freeman, 1984;Stinchcombe, 1965). However, scholars have alsoargued that the disruptive effects of organizationalchange are the most severe among older firms (e.g.,Amburgey et al., 1993), because the older firms are

better developed and have greater inertia. Such in-ertia could actually make the change from privateto public status more risky for older firms that gopublic. We transformed firm age into its naturallogarithm to reduce the effects of extreme values onthe analysis, adding 1 to all observations beforetransforming the measure.

Offering size. This measure equaled the totalnumber of shares offered during an IPO multiplied

by the offering price. The size of an offering, whichcan signal the market about its relative quality andstability, has been frequently used as a control inIPO research (Ibbotson & Ritter, 1995). Larger offer-ings also provide IPO firms with more cash, whichcan serve as a resource-based transformationalshield following the IPOs. We also transformed thisvariable into its natural logarithm to reduce theeffect of extreme values.

CEO experience. According to human capital

perspectives (e.g., Pennings, Lee, & van Witteloos-tuijn, 1998) the experience of a CEO can affect anorganizations efficiency and its ability to attractcustomers and new investor capital, which can af-fect firm survival. We measured CEO experiencewith two variables. The first, public experience,was a dummy variable coded 1 if a CEO had previ-ously had a senior management role in a publiclytraded company and 0 otherwise. CEOs with priorexperience managing public firms have a specificform of human capital that should enhance theircurrent firms ability to operate effectively under

1 A dummy variable controlling for whether any firm-year spells measure was imputed did not affect ourresults.



10/20

public ownership. We obtained data on executivesprior work histories from the firms offering pro-spectuses, the Dun & Bradstreet Million Dollar da-tabase, the Standard & Poors Net Advantage data-

base, and the Dun & Bradstreet Reference Book ofCorporate Managements.

The second measure, CEO age, indicated the age

of a firms CEO at the time of its IPO. The age ofchief executives has also been discussed as an im-portant element of human capital and has beenshown to yield relatively consistent positive rela-tionships with entrepreneurial firm performance(Daily, McDougall, Covin, & Dalton, 2002; Sapienza& Grimm, 1997). Certo and coauthors (2001) sug-gested investors may value older CEOs in firms thatare going public more highly than younger CEOs

because of the assumed presence of greater matu-rity and experience.

CEO tenure. Independent of a CEOs age and

experience, the amount of time a CEO has beenwith a firm can affect his or her power and discre-tion within the firm (Finkelstein, 1992; Porac,Wade, & Pollock, 1999). CEO tenure was measuredas the number of years a CEO had been with a firmat the time of its IPO.

Deal network size. This measure equaled thenumber of institutional investors who owned stockin a company at the end of the quarter in which thecompany went public. This measure was used as acontrol because the higher the number of investors,the less likely it is that an underwriter will haveembedded relationships with a large proportion ofthe members of the network surrounding a deal.

Management team size. This measure equaledthe number of members on an IPO firms top man-agement team, as identified in the firms offeringprospectus. This measure was included as a control

because management team size can affect averagemanagement team tenure. All else being equal, thelarger a TMT, the more likely it is to have short-tenured members.

Institutional ownership concentration. The con-centration of shares in the hands of institutionalinvestors can affect the discretion of management

(Tosi et al., 1999). We measured institutional inves-tor ownership using a Herfindahl index identical tothe one described for venture capitalist ownership.Ownership levels were based on the number ofshares owned at the end of the quarter in which acompany went public, as reported in the CompactD SEC database.

Years since IPO. The net effects of organiza-tional change are time-dependent, and the amountof time elapsed since going public may affect long-term survival chances. Because the disruptive ef-fects of organizational change have been argued to

decrease with time nonmonotonically (Amburgeyet al., 1993), we also included a squared term.

Method of Analysis

The data were analyzed using discrete-time eventhistory techniques that estimate logit models of

dichotomous outcomes for pooled time series datain which the same units are observed at multipleintervals (Allison, 1984; Yamaguchi, 1991). Covari-ates are allowed, but not required, to vary betweentime periods. Since the data contained multipleobservations of the same IPO firm that were notindependent across spells, we employed a clustercommand using Stata 7.0, which provided a moreconservative test of the hypotheses by using robustestimators of variance.

Using discrete-time event history models offeredseveral advantages. First, although in this study we

focused on the impact of initial conditions on IPOfirm survival, given that we had right censoring inour data, the probabilities of a failure event occur-ring might change as a function of time since thecompany went public. By using discrete-time tech-niques, we could capture potential effects for timeelapsed since an IPO in our analyses. Usingdiscrete-time models also allowed us to includetime-varying financial performance controls in ourmodel. Finally, a number of firms in our sample didnot fail, but were acquired or merged with otherfirms during our period of study. We did not con-sider an acquisition or merger an IPO firm failure,since IPO firms and their investors often see theseas attractive and desirable outcomes (Price Water-house, 1995). However, once a firm is acquired, it isno longer at risk of delisting owing to failure. Usingdiscrete-time models allowed us to keep these com-panies in the sample until they were acquired,thereby providing more complete informationabout the relationship between our independentvariables and firm failure.

RESULTS

Table 1 presents the descriptive statistics and acorrelation matrix for the variables used in ourmodels. Of the final sample of 218 IPO firms, 20that were delisted from their primary exchanges

before the end of the five-year period were countedas firm failures. An additional 51 firms weredelisted because they were acquired by other firms.We dropped these firms from the sample in theyears they were acquired but did not count them asfailures. Thus, 147 of the 218, or 67 percent, of theIPOs successfully managed the five-year transitionfrom newly public to seasoned firms, and these



11/20

TABLE 1Descriptive Statistics and Correlationsa

Variable Mean s.d. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1. IPO firm failure 0.02 0.142. Years since IPO 2.86 1.40 .083. Years since IPO squared 10.12 8.43 .07 .984. High-technology firmb 0.40 0.49 .02 .01 .015. Total salesc, d 10.53 2.28 .02 .16 .16 .356. Net incomed 1,863.75 29,230.19 .09 .02 .02 .08 .247. Annual stock returnd 15.46 82.40 .11 .00 .03 .02 .04 .078. Industry concentrationd 0.07 0.07 .02 .09 .10 .09 .18 .07 .039. Underpricing 12.51 19.23 .06 .01 .01 .05 .02 .06 .01 .04

10. Age of firm at IPO 2.13 0.91 .03 .01 .01 .15 .37 .13 .01 .10 .1211. Offering sizec 16.98 0.85 .04 .00 .00 .13 .36 .11 .08 .07 .09 .2412. TMT size 6.26 2.05 .03 .02 .02 .01 .23 .02 .04 .08 .00 .18 .3613. Deal network size 15.80 11.65 .05 .01 .01 .09 .36 .15 .05 .07 .24 .27 .80 .30

14. CEO public experience 0.48 0.50 .01 .01 .01 .28 .04 .07 .05 .06 .02 .07 .03 .00 .0015. CEO age 47.37 7.51 .02 .00 .00 .08 .12 .11 .03 .06 .07 .06 .10 .02 .13 .0716. CEO job tenure 5.24 5.14 .06 .02 .02 .10 .28 .13 .01 .02 .02 .34 .13 .15 .15 .05 .2217. Venture capitalist

backing0.56 0.50 .03 .00 .00 .36 .21 .18 .01 .16 .03 .25 .01 .11 .06 .32 .10

18. Underwriter reputation 82.12 24.07 .05 .01 .01 .04 .19 .04 .06 .09 .07 .09 .62 .29 .51 .14 .01 19. Institutional investor

ownership concentration0.04 0.04 .04 .01 .00 .10 .01 .13 .00 .03 .02 .02 .18 .19 .16 .17 .04

20. Founder-CEOb 0.75 3.50 .02 .01 .01 .08 .08 .04 .01 .00 .04 .00 .06 .02 .00 .05 .0521. Average TMT tenure 5.52 4.30 .07 .01 .01 .23 .39 .18 .00 .09 .11 .78 .24 .14 .25 .14 .18 22. Deal network

embeddedness1.14 1.07 .08 .02 .03 .04 .21 .13 .01 .02 .23 .19 .57 .21 .68 .08 .05

23. CEO ownership 15.88 18.54 .01 .02 .02 .26 .21 .11 .01 .18 .00 .05 .02 .05 .07 .33 .07 24. Venture capitalist

ownership concentration3.91 7.26 .01 .00 .01 .09 .04 .11 .04 .08 .05 .11 .07 .05 .10 .10 .22

a n 920.b Dummy variable.c Logarithm.d Lagged variable.


12/20

were treated as right-censored in our sample. Thedelisting rate observed in this study is similar tothose observed in other studies on IPO survival(e.g., Foster-Johnson et al., 2001; Jain & Kini, 2000;Welbourne & Andrews, 1996).

Table 2 presents the results of the discrete timemodels testing our hypotheses. Model 1, a baseline

model, includes all the control variables. Model 2includes the main effects of the variables used totest Hypotheses 13. In model 3 we added theinteraction terms, and in model 4, the variablesused to test Hypotheses 6 and 7. Analysis of thelog-likelihoods indicated that, although adding thetheoretical variables to model 2 did not signifi-cantly improve the fit of the model, adding theinteractions in model 3 significantly improvedmodel fit over both model 1 and model 2, andadding the social capital variables resulted in an

additional, significant improvement in model fitover model 3. We therefore used the results ofmodel 4 to test the hypotheses.

Hypotheses 1, 2, and 3, which state that having afounder-CEO, a relatively high level of CEO own-ership, and a relatively high level of venture capi-talist ownership concentration will reduce the like-

lihood of IPO firm failure, were not supported.Although VC ownership concentration was signif-icant, the effect was in the direction opposite thatpredicted by the hypothesis. The other main effectswere not significant. However, our results support

both Hypothesis 4, which states that VC ownershipconcentration will interact with CEO ownership toreduce the likelihood of failure, and Hypothesis 5,which states that having a founder-CEO will inter-act with CEO ownership to reduce the likelihood offailure. Both of the interaction terms were negative

TABLE 2Results of Discrete Time Analyses Predicting IPO Firm Failurea

Variable Model 1 Model 2 Model 3 Model 4

Years since IPO 3.79* (1.91) 3.92* (1.96) 4.07 (2.18) 4.11 (2.43)Years since IPO squared 0.45 (0.28) 0.46 (0.28) 0.48 (0.30) 0.47 (0.34)High-technology firmb 0.80 (0.58) 0.75 (0.58) 0.64 (0.55) 0.98 (0.65)Total salesc, d 0.01 (0.11) 0.03 (0.11) 0.09 (0.13) 0.10 (0.14)Net incomed 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)Annual stock returnd 0.04* (0.02) 0.04* (0.02) 0.04* (0.02) 0.03 (0.02)Industry concentrationd 0.46 (2.53) 0.02 (2.52) 0.41 (2.61) 1.24 (3.06)Underpricing 0.06** (0.02) 0.06** (0.02) 0.05** (0.02) 0.05** (0.02)Firm agec 0.05 (0.23) 0.23 (0.23) 0.37 (0.23) 0.37 (0.35)

Offering sizec 0.18 (0.53) 0.18 (0.64) 0.39 (0.78) 0.04 (0.90)TMT size 0.07 (0.16) 0.07 (0.16) 0.16 (0.15) 0.10 (0.12)Deal network size 0.01 (0.05) 0.01 (0.06) 0.00 (0.06) 0.03 (0.08)CEO public experience 0.44 (0.49) 0.33 (0.58) 0.59 (0.73) 0.79 (0.83)CEO age 0.02 (0.03) 0.02 (0.03) 0.03 (0.04) 0.05 (0.04)CEO job tenure 0.16** (0.06) 0.15* (0.07) 0.14* (0.07) 0.09 (0.08)Venture capitalist backing 0.92 (0.58) 1.46* (0.72) 1.19 (0.75) 1.18 (0.78)Underwriter reputation 0.00 (0.02) 0.00 (0.02) 0.00 (0.02) 0.00 (0.02)Institutional investor ownership concentration 7.79 (7.30) 6.54 (7.43) 9.60 (8.11) 7.40 (8.64)Founder-CEOb 0.32 (0.60) 0.71 (0.76) 0.77 (0.85)Venture capitalist ownership concentration

post-IPO0.02 (0.03) 0.06* (0.03) 0.10* (0.04)

CEO ownership post-IPO 0.02 (0.02) 0.00 (0.21) 0.00 (0.03)Founder-CEO CEO ownership 0.06* (0.03) 0.06* (0.03)Venture capitalist ownership CEO

ownership0.03* (0.02) 0.04* (0.02)

Average TMT tenure 0.31 (0.18)Deal network embeddedness 1.28* (0.63)

Constant 8.61 (10.75) 8.20 (12.65) 6.53 (14.54) 12.81 (16.66)Log-likelihood 58.58 57.53 54.96 50.53

a n 920 for all models.b Dummy variable.c Logarithm.d Lagged variable.

p .10* p .05

** p .01



13/20

and significant (at p .05). Agency theoretic argu-ments would suggest that concentrated ownershipis negatively related to IPO firm failure because itdisciplines management. To address this alterna-tive explanation, in models not reported here wetested interactions between institutional investorownership concentration and both CEO ownership

and the founder-CEO dummy. Neither interactionwas significant. In addition, the positive main ef-fect of venture capitalist ownership on IPO firmfailure also ran counter to an agency-theoretic ex-planation. Finally, Hypotheses 6 and 7 state thatthe internal and external social capital available toan IPO firm can also serve as transformationalshields. Hypothesis 6, which states that the amountof time top management team members have spentworking together prior to an IPO will be negativelyrelated to firm failure, was supported. Hypothesis7, which states that deal network embeddedness

will have a significant, negative relationship withIPO failure, is also supported.An analysis of the coefficients for the variables

with significant relationships provided us withsome additional insight into the impact thesetransformational shields have on IPO firm fail-ure. By multiplying a given coefficient by achange in the independent variable and then ex-ponentiating the product, one can obtain thechange in the odds of an events occurring versusits not occurring. So, for example, if a companyhas a TMT with an average tenure of four years(the median average management team tenure forour sample), the probability of this firms failingin the five years following its IPO is reduced by1 exp[4 (0.3090)], or approximately 71

percent, when it is compared to a companywhose TMT members are all in their first yearof employment with the firm. If a firm in oursample had a founder-CEO and CEO ownershipof 7.9 percent (the median value for our sample),the odds of its failing within the first five yearsafter its IPO are decreased by about 40 percent.This strong and significant interaction is evenmore dramatic when visually depicted by Figure1, which plots CEO ownership against the multi-plier of the failure rate for founder-CEOs versusnonfounders. In the case of CEO ownership of 7.9percent, the multiplier of the failure rate forfounders is approximately 0.60 (where 1.00 0.60 0.40, or a 40 percent reduction in the oddsof failure, all else being equal). At the same levelof CEO ownership, the multiplier for nonfound-ers is about 0.99 (1.00 0.99 0.01, or a 1percent reduction), resulting in an insignificantimpact on the odds of IPO failure. Figure 2 sim-ilarly depicts the multiplier of the failure rateacross the range of CEO ownership levels at the25th, 50th, and 75th percentile of venture capi-talist ownership concentration in our sample.The multiplier of the failure rate, and hence theodds of IPO firm failure, dramatically decreaseas both values approach their maximum rangevalues. Using the same method of interpreting theimpact of a given variable on IPO firm failure, ifa firm in our sample has a median deal networkembeddedness score of 0.94, its odds of failureare decreased by about 70 percent relative toa firm whose network includes no embeddedrelationships.

FIGURE 1Effect of CEO Ownership by Founder Status on IPO Failure Rate



14/20

DISCUSSION

The ability of a firm to survive a significant trans-formation is an important yet understudied area inthe organizations literature. Traditional evolution-ary perspectives have focused on the effects of ex-ternal or environmental conditions and organiza-tional demographic characteristics on firms vitalrates following core change. In this study, we ex-amined how sociopolitical features of a firm itselfcan serve as transformational shields that enhancethe ability to survive the potentially disruptive pe-riod immediately following an IPO. In so doing, wehave brought to bear and extended the originalideas of Miner and her coauthors (1990) by suggest-ing that factors related to an organizations socialcapital and power structure at the time a transfor-mational event is initiated can have long-lastingeffects on a firms life chances. Specifically, weargued that these resources can mitigate the threatsassociated with resetting the liability of newnessclock (Amburgey et al., 1993) and provide a valu-

able social base on which to further develop acompany.

Theoretical Implications

Our findings have significant theoretical impli-cations for perspectives addressing organizationalchange and evolution. The results of our analysisstrongly support our hypotheses regarding the ef-fectiveness of internal and external social capital asa transformational shield, and they begin to addressa gap in organizational research and theory by spe-

cifically exploring how intraorganizational socialcapital can affect a firms survival chances follow-ing a significant transformation. We argued thathigh average management team tenure provides afirm with valuable internal social capital that en-hances management team effectiveness and infor-mation flow among its members, and in so doingcontributes to managements ability to steer thecompany through the disruptive period following

its IPO.We also made and found support for the claim

that characteristics of a companys external owner-ship network structure can protect it from environ-mental pressures. Additionally, this study is amongthe first to explicitly highlight the effects of internalpolitical dynamics on the life chances of a firmexperiencing a major transformation. We suggestedthat the involvement, at the time of a firms IPO, ofkey individuals with both high commitment to theorganization and power based on formal authorityand concentrated ownership could significantly

enhance the chances of the firm surviving the trans-formation. The significant impacts of two interac-tionsthe interaction between having a founder-CEO and CEO ownership, and the interaction

between the CEO and VC ownership variablessuggest that neither commitment nor ownershipalone is sufficient to protect a firm. Rather, bothfactors are necessary to create an effective transfor-mational shield.

Our primary theoretical arguments and findingscan be extended to examine the extent to which afirm is protected from the disruptions and addi-

FIGURE 2Effect of CEO Ownership by Venture Capitalist Ownership

Concentration on IPO Failure Rate



15/20

tional changes that initial firm transformationsfrom private to public ownership may trigger. Bar-nett and Carroll (1995) suggested that core struc-tural change can be defined in terms of the exten-siveness of the concurrent and subsequent changesthat are related to an initial transformation. Cas-cades of associated but unforeseen disruptions to

routines and procedures throughout firms may bethe primary cause of the observed spikes in failurerates associated with organizational change moregenerally (Amburgey et al., 1993; Dobrev et al.,2003).

It is worth reiterating the distinction between buffering mechanisms meant to insulate a firmfrom the need for transformation in the first place,and shielding mechanisms that protect the firmfrom the deleterious effects of a given transforma-tion (Miner et al., 1990). Just as we argued thatsocial and political factors, serving as transforma-

tional shields, may protect a firm from re-exposureto the internal and external threats commonly as-sociated with newness, we speculate that thesesame factors may also ultimately enhance survivalchances by minimizing the cascade effect followinga transformation (Dobrev et al., 2003). Consider theinteraction of CEO-founder status and CEO owner-ship in our study. Only the presence of a founderconcurrently serving as CEO and retaining a highlevel of direct ownership in an IPO firm displayeda beneficial effect. Thus, one might interpret orextend our findings to imply that sociopoliticalshielding, whether by enhancing trust and commu-nication within a firm or by facilitating the devel-opment of external social capital and ties, effec-tively blocks the cascade of associated changescommonly understood to threaten the survival oftransforming firms.

Although agency theory was not central to ouranalysis, our findings also have some interestingimplications for this theoretical perspective. One ofthe primary assumptions of agency theory is thatmanagers are self-interested and risk-avoiding andthat, to the extent a firms ownership and manage-ment are separated, its managers will not act in the

best interests of the firm unless its owners usemonitoring and control mechanisms to discourageself-interested behavior (Tosi et al., 1999). Ourfindings suggest that the simple but powerful logicunderlying this presumption is incomplete, sinceneither CEO ownership, venture capitalist owner-ship concentration, nor institutional investor own-ership concentration alone had a significant effecton IPO firm survival. However, when conditionedon aspects of a firms internal social and politicaldynamics, ownership demonstrated a significantrelationship with survival.

We have suggested that, compared to a CEObrought in from the outside, a founder-CEO willhave greater personal identification with a firm,greater commitment to it, and greater trust from thefirms employees. Substantial retained ownershipprovides founder-CEOs with the power and protec-tion necessary to focus their full attention, capabil-

ities, and resources on leading their companiesthrough their transformations. In terms of agencytheory, these findings parallel the arguments of Leeand ONeill (2003), who suggested that alignmentmechanisms adopted on the basis of agency pre-scriptions may lead to misallocation of resourcesand unnecessary risk seeking. The authors distin-guished between the opportunistic manager andthe steward as CEO and suggested that stewardsmotives are more aligned with the interests of theirprincipals. Our finding that CEO-retained owner-ship benefits a firm as it transforms from private to

public status only when the CEO is also founderlends support to these arguments.In addition, contrary to our expectations, our re-

sults suggested that high retained venture capitalistownership following an IPO actually increasedtheprobability that a firm would fail, unless its CEOalso retained substantial ownership. One explana-tion may be that, after IPO completion, CEO andventure capitalist interests may diverge. Since VCinvestment funds have finite life spans (Sahlman,1990), VCs may be more focused on the short termthan other principals. If a venture capitalist main-tains substantial influence over the strategic direc-tion and operations of a firm through concentratedownership, the VC may be able to force the newlypublic firm to engage in activities that boost short-term performance but are damaging in the longterm. However, if the CEO also retains substantialownership, he or she may be in a strong position toresist these pressures. This finding also suggeststhat agency theory may benefit from a reexamina-tion of its assumption that all managers shareuniform, self-interested motivations and that allinvestors are motivated to enhance long-term per-formance (Wiseman, Gomez-Mejia, & Fugate,

2000). Taken together, our findings and our empha-sis on the impact of sociopolitical factors are lessconsistent with traditional agency explanations ofcontrol than they are with Bolino and coauthors(2002) arguments, which are based on relationaland cognitive elements facilitating trust andcommunication.

Finally, although not hypothesized, the negativerelationship between underpricing and IPO firmfailure is noteworthy and merits some discussion.In the finance literature, underpricing is generallyviewed as a negative consequence of information



16/20

asymmetries (see Ibbotson and Ritter [1995] for areview). Underpricing an IPO results in money be-ing left on the table (Loughran & Ritter, 2002) thatthe IPO firm could presumably have captured if theoffering had been valued more accurately. The re-sults of this study, however, suggest that this per-spective may be somewhat shortsighted; our find-

ings suggest underpricing may lead to outcomesthat enhance a firms ability to survive the transi-tional IPO period. Indeed, recent research (Pollocket al., 2002) has identified such beneficial out-comes as increased analyst coverage and a greaterability to form strategic alliances.

Implications for Management Practice

Our findings also have significant implications both for managers who have recently undergoneIPOs and for those whose firms are at an earlier

stage of development. Our results suggest that ex-ecutives of early-stage firms, especially founders,should carefully consider the amount of stock theygive up to outside investors as they seek financingfor their ventures. Giving up too much stock tooearly in a ventures life can decrease a firmschances of surviving its initial public offering. Ven-ture capitalists should also seriously consider theconsequences of replacing founders with differentCEOs, especially if this change is to be madeshortly before an IPO. Developing stable and trust-ing social and political ties within the organizationprior to the IPO can enhance the firms survivalchances following the offering. Finally, when se-lecting an underwriter to lead their IPO, managersshould carefully consider not just the reputationalcapital of underwriters, but also the social capitalreflected in their networks of relationships withinstitutional investors and the likelihood they willuse their social capital on the firms behalf to con-struct an embedded deal network.

Limitations and Additional Future ResearchDirections

Like any study, this one has limitations that leaveunanswered questions providing the opportunityto develop new research directions. For example,we have focused strictly on the conditions presentat the time of a transition in order to extend thetransformational shield argument (Miner et al.,1990). Subsequent research might instead explorethe ongoing buffering properties of CEO and ven-ture capitalist ownership and of founder-CEO pres-ence and internal and external social capital byfollowing these variables over time and evaluatingtheir ability to insulate firms from the need for

future survival-threatening changes. A second re-search opportunity arises from the dependent vari-able considered in this study. We focused on theimpact of sociopolitical firm characteristics on firmsurvival. Future research might consider their im-pact on the quality of firm survival by pursuingresearch that predicts firm performance.

Another opportunity arises from the limitationsof the archival research method employed in thisstudy. Our approach demanded certain theoreticalassumptions and that constructs such as trust andescalation of commitment be black boxed ratherthan measured directly. Future research could usealternative methodologies, such as surveys, exper-iments, and simulations, to test the theoretical as-sumptions underlying our arguments.

Finally, additional research opportunities relateto the specific context of our study. We developedtheory regarding the impact of a companys socio-

political characteristics in the context of one spe-cific type of organizational transformation, an IPO.Future research could apply and extend our theoryto different transformational contexts, such asmergers and acquisitions and adoptions of newproduction technologies. One limitation of focus-ing on discrete events such as IPOs is that it maynot be the events themselves, but other, underlyingevents occurring around the time of the observedevents that trigger resetting of the liability of new-ness clock. Although we recognize that firms willvary in the amount of organizational trauma thatIPOs generate, we suggest that these events result inextensive changes to organizations goals, bound-aries, and activity systems often enough to allowthem to serve as a proxy for the many disruptiveevents associated with transformational change. In-deed, if IPOs did not constitute a transformationalchange most of the time, we would not expect tosee significant results in our analysis. The signifi-cant findings of this study provide additional sup-port for the notion that an IPO is indeed a transfor-mational change event.

ConclusionOverall, our findings have significant implica-

tions for theory that explores the ways in whichfirms attempt to mitigate the liabilities of newnessthat result from engaging in significant transforma-tional change (Aldrich, 1999; Amburgey et al.,1993). As organizations grow and develop, theyinevitably transform and evolve. Motivations forchange that result in organizational transforma-tions vary from experimentation to a simple need tochange or die (Haveman, 1992). This process isunderstood to increase a firms risk of failure, at



17/20

least in the short term, as it attempts to cope withthe internal and external threats to which changesexpose it. Although prior research has examinedthe implications of environmental conditions forfirm survival in these circumstances, little theoriz-ing and empirical research at an organizationallevel has examined how firms can mitigate these

threats. This study demonstrates how internal andexternal social resources and relationships that ex-ist at the time a transformational change occurs canat least partially neutralize the threats a firm facesand enhance its likelihood of surviving the change.

REFERENCES

Adler, P. S., & Kwon, S. 2002. Social capital: Prospectsfor a new concept. Academy of Management Re-view, 27: 1740.

Aldrich, H. 1979. Organizations and environments.

Englewood Cliffs, NJ: Prentice-Hall.

Aldrich, H. 1999. Organizations evolving. ThousandOaks, CA: Sage.

Allison, P. D. 1984. Event history analysis. NewburyPark, CA: Sage.

Amburgey, T. L., & Miner, A. S. 1992. Strategic momen-tum: The effects of repetitive, positional, and con-textual momentum on merger activity. Strategic

Management Journal, 13: 335348.

Amburgey, T. L., Kelly, D., & Barnett, W. P. 1993. Reset-ting the clock: The dynamics of organizationalchange and failure. Administrative Science Quar-

terly, 38: 5173.Baker, W. E. 1984. The social structure of a national

securities market. American Journal of Sociology,89: 775811.

Baker, W. E. 1990. Market networks and corporate behav-ior. American Journal of Sociology, 96: 589625.

Bantel, K. A., & Jackson, S. E. 1989. Top management andinnovations in banking: Does the composition of thetop team make a difference? Strategic Management

Journal, 13: 338355.

Barnett, W. P., & Carroll, G. R. 1995. Modeling internalorganizational change. In J. Hagan & K. Cook (Eds.),

Annual review of sociology, vol. 21: 217236. PaloAlto, CA: Annual Reviews.

Baron, J. N., Hannan, M. T., & Burton, M. D. 1999. Build-ing the iron cage: Determinants of managerial inten-sity in the early years of organizations. AmericanSociological Review, 64: 527547.

Boeker, W. P. 1988. Organizational origins: Entrepre-neurial and environmental imprinting at the time offounding. In G. Carroll (Ed.), Ecological models oforganizations: 3351. Cambridge, MA: Ballinger.

Bolino, M. C., Turnley, W. H., & Bloodgood, J. M. 2002.Citizenship behavior and the creation of social cap-

ital in organizations. Academy of Management Re-view, 27: 505522.

Brockner, J., Siegel, P. A., Daly, J. P., Tyler, T., & Martin,C. 1997. When trust matters: The moderating effectof outcome favorability. Administrative ScienceQuarterly, 42: 558583.

Bushee, B. J. 1998. The influence of institutional inves-

tors on myopic R&D investment behavior. Account- ing Review, 73: 305333.

Bygrave, W. D., & Timmons, J. A. 1992. Venture capitalat the crossroads. Boston: Harvard Business SchoolPress.

Cannella, A. A., & Hambrick, D. C. 1983. Effects of exec-utive departures and the performance of acquiredfirms. Strategic Management Journal, 14: 137152.

Carpenter, M. A. 2002. The implications of strategy andsocial context for the relationship between top man-agement teams heterogeneity and firm performance.Strategic Management Journal, 23: 275284.

Carter, R. B., & Dark, F. H. 1993. Underwriter reputationand initial public offers: The detrimental effects offlippers. Financial Review, 28: 279301.

Carter, R. B., & Manaster, S. 1990. Initial public offeringsand underwriter reputation. Journal of Finance, 65:10451067.

Certo, S. T., Covin, J. G., Daily, C. M., & Dalton, D. R.2001. Wealth and the effects of founder managementamong IPO-stage new ventures. Strategic Manage-

ment Journal, 22: 641658.

Coleman, J. S. 1988. Social capital in the creation ofhuman capital. American Journal of Sociology, 94:

s95s121.

Daily, C. M., McDougall, P. P., Covin, J. G., & Dalton, D. R.2002. Governance and strategic leadership in entre-preneurial firms. Journal of Management, 155:128.

Dobrev, S. D., Kim, T. Y., & Carroll, G. R. 2003. Shiftinggears, shifting niches: Organizational inertia andchange in the evolution of the U.S. automobile in-dustry, 18851981. Organization Science, 14: 264282.

Eisenhardt, K. M., & Schoonhoven, C. B. 1990. Organiza-tional growth: Linking founding team, strategy, en-

vironment, and growth among United States semi-conductor ventures, 1978 1988. AdministrativeScience Quarterly, 35: 504529.

Ellis, K., Michaely, R., & OHara, M. 2000. When theunderwriter is the market maker: An examination oftrading in the IPO aftermarket. Journal of Finance,55: 10391074.

Finkelstein, S. 1992. Power in top management teams:Dimensions, measurement, and validation. Acad-emy of Management Journal, 35: 505538.

Finkelstein, S., & Hambrick, D. C. 1990. Top managementteam tenure and organizational outcomes: The mod-



18/20

erating role of managerial discretion. Administra-tive Science Quarterly, 35: 484503.

Foster-Johnson, L., Lewis, C. M., & Seward, J. K. 2001. Busted IPOs and windows of misopportunity.Working paper, Dartmouth College, Hanover, NH.

Fox, F. V., & Staw, B. M. 1979. The trapped administra-tor: The effects of job insecurity and policy resis-

tance upon commitment to a course of action. Ad-ministrative Science Quarterly, 24: 449471.

Granovetter, M. S. 1985. Economic action and socialstructure: The problem of embeddedness. American

Journal of Sociology, 91: 481510.

Hanley, K., & Wilhelm, W. J. 1995. Evidence on thestrategic allocation of initial public offerings. Jour-

nal of Financial Economics, 37: 239257.

Hannan, M. T., & Freeman, J. 1984. Structural inertia andorganizational change. American Sociological Re-view, 49: 149164.

Haveman, H. A. 1992. Between a rock and a hard place:Organizational change and performance under con-ditions of fundamental environmental transforma-tion. Administrative Science Quarterly, 37: 48 75.

Hitt, M. A., Bierman, L., Shimizu, K., & Kochar, R. 2001.Direct and moderating effects of human capital onstrategy and performance in professional servicefirms: A resource-based perspective. Academy of


Husick, G. C., & Arrington, J. M. 1998. The initial publicoffering: A practical guide for executives. NewYork: Bowne & Co.

Ibbotson, R. G., & Ritter, J. R. 1995. Initial public offer-

ings. In R. Jarrow et al. (Eds.), Handbook in opera-tions research and management science, vol. 9:9931016. London: Elsevier Sciences.

Jain, B. A., & Kini, O. 2000. Does the presence of venturecapitalists improve the survival profile of IPO firms?

Journal of Business Finance & Accounting, 27:11391176.

Jegadeesh, N., Weinstein, M., & Welch, I. 1993. An em-pirical investigation of IPO returns and subsequentequity offerings. Journal of Financial Economics,34: 153175.

Larson, A. 1992. Network dyads in entrepreneurial set-

tings: A study of the governance of exchange rela-tionships. Administrative Science Quarterly, 37:76104.

Lee, P. M., & ONeill, H. M. 2003. Ownership structuresand R&D investments of U.S. and Japanese firms:Agency and stewardship perspectives. Academy of


Lee, P. M., & Wahal, S. In press. Grandstanding, certifi-cation and underpricing of venture backed IPOs.

Journal of Financial Economics.

Loughran, T., & Ritter, J. R. 1995. The new issues puzzle. Journal of Finance, 50: 2351.

Loughran, T., & Ritter, J. R. 2002. Why dont issuers getupset about leaving money on the table in IPOs?

Review of Financial Studies, 15: 413443.

Miner, A. S., Amburgey, T. L., & Stearns, T. M. 1990.Interorganizational linkages and population dynam-ics: Buffering and transformational shields. Admin-

istrative Science Quarterly, 35: 689713.

Moon, H. 2001. Looking forward and looking back: Inte-grating completion and sunk-cost effects within anescalation-of-commitment progress decision. Jour-

nal of Applied Psychology, 36: 104113.

Nahapiet, J., & Ghoshal, S. 1988. Social capital, intellec-tual capital, and the organizational advantage.Acad-emy of Management Review, 23: 242266.

Nelson, T. 2003. The persistence of founder influence:Management, ownership, and performance effects atinitial public offering. Strategic Management Jour-

nal, 24: 707724.

Noda, T., & Collis, D. 2001. The evolution of intraindus-

try firm heterogeneity: Insights from a process study.Academy of Management Journal, 44: 897925.

Pennings, J. M., Lee, K., & van Witteloostuijn, A. 1998.Human capital, social capital, and firm dissolution.

Academy of Management Journal, 41: 425440.

Pfeffer, J. 1992. Managing with power. Cambridge, MA:Harvard Business Press.

Platt, H. D. 1995. A note on identifying likely IPO bank-ruptcies: A symbolic paradox. Journal of Account-

ing, Auditing & Finance, 10: 7179.

Podolny, J. 1994. Market uncertainty and the social char-acter of economic exchange. Administrative Sci-

ence Quarterly, 39: 458483.Pollock, T. G., Fischer, H. M., & Wade, J. B. 2002. The role

of power and politics in the repricing of executiveoptions. Academy of Management Journal, 45:11721182.

Pollock, T. G., Gulati, R., & Sadler, A. 2002. Relationaland market-based legitimation of internet IPOs.

Academy of Management Best Paper Proceedings,BPS 1116.

Pollock, T. G., Porac, J. F., & Wade, J. B. 2004. Construct-ing deal networks: Brokers as network architectsin the U.S. IPO market and other examples. Acad-

emy of Management Review, 29: 50 72.Porac, J. F., Wade, J. B., & Pollock, T. G. 1999. Industry

categorizations and the politics of the comparablefirm in CEO compensation. Administrative ScienceQuarterly, 44: 112144.

Price Waterhouse. 1995. The going public handbook.New York: Price Waterhouse.

Rao, H., & Sivakumar, K. 1999. Institutional sources ofboundary-spanning structures: The establishment ofinvestor relations departments in the Fortune 500industrials. Organization Science, 10: 2742.

Rindova, V. P., & Kotha, S. 2001. Continuous mor-



19/20

phing: Competing through dynamic capabilities,form, and function. Academy of Management Jour-

nal, 44: 12631280.

Ritter, J. R. 1991. The long-run performance of initialpublic offerings. Journal of Finance, 46: 327.

Ross, J., & Staw, B. M. 1993. Organizational escalationand exit: Lessons from the Shoreham Nuclear Power

Plant. Academy of Management Journal, 36: 701732.

Sahlman, W. A. 1990. The structure and governance ofventure capital organizations. Journal of Financial

Economics, 27: 473521.

Sapienza, H. J., & Grimm, C. M. 1997. Founder charac-teristics, start-up processes, and strategy/structurevariables as predictors of shortline railroad perfor-mance. Entrepreneurship Theory & Practice, 22:524.

Scherer, F. M. 1980. Industrial market structure andeconomic performance (2nd ed.). Boston: Houghton

Mifflin.Smith, N. R., & Miner, J. R. 1983. Type of entrepreneur,

type of firm, and managerial motivation: Implica-tions for organizational life cycle theory. Strategic


Stinchcombe, A. 1965. Social structure and organiza-tions. In J. G. March (Ed.), Handbook of organiza-tions: 42193. Chicago: Rand McNally.

Thompson, J. D. 1967. Organizations in action. NewYork: McGraw-Hill.

Tosi, H. L., Gomez-Mejia, L. R., Loughry, M. L., Werner,S., Banning, K., Katz, J., Harris, R., & Silva, P. 1999.

Managerial discretion, compensation strategy, andfirm performance: The case for the ownership struc-ture. In G. R. Ferris (Ed.), Research in personneland human resource management, vol. 17: 163208. Stamford, CT: JAI Press.

Tyler, T. R., & Degoey, P. 1996. Trust in organizationalauthorities: The influence of motive attributions onwillingness to accept decisions. In R. M. Kramer &T. R. Tyler (Eds.), Trust in organizations: 331356.Thousand Oaks, CA: Sage.

Tyler, T. R., & Lind, E. A. 1992. A relational model ofauthority in groups. In M. Zanna (Ed.), Advances inexperimental social psychology, vol. 25: 115191.New York: Academic Press.

Useem, M. 1996. Investor capitalism: How money man-agers are changing the face of corporate America.New York: Basic Books.

Uzzi, B. 1996. Embeddedness and economic perfor-mance: The network effect. American Sociological

Review, 61: 674698.

Welbourne, T. M., & Andrews, A. O. 1996. Predicting theperformance of initial public offerings: Should hu-man resource management be in the equation? Acad-emy of Management Journal, 39: 891919.

Westphal, J. D. 1998. Board games: How CEOs adapt toincreases in structural board independence frommanagement. Administrative Science Quarterly,43: 511537.

Westphal, J. D., & Fredrickson, J. W. 2001. Who directsstrategic change? Director experience, the selectionof new CEOs, and change in corporate strategy. Stra-tegic Management Journal, 22: 11131137.

Wiseman, R. M., Gomez-Mejia, L. R., & Fugate, M. 2000.Rethinking compensatio

AOM JOURNAL6

Documents

Transcript of AOM JOURNAL6