Governance and Performance Implications of Diversification Strategies: Evidence from Large U.S. Firms

The Financial Review 39 (2004) 489--526

Governance and Performance Implicationsof Diversification Strategies: Evidence

from Large U.S. FirmsManohar SinghUniversity of Nevada

Ike Mathur∗Southern Illinois University-Carbondale

Kimberly C. GleasonFlorida Atlantic University

Abstract

Recent research focuses on explaining the diversification discount. However, there is littledirect evidence regarding the relation among ownership structure, corporate governance, andcorporate diversification. The results in this paper suggest that agency issues do not accountfor firms adopting a particular diversification strategy. Also, the performance consequences ofthe shift in the diversification strategy and the subsequent changes in institutional and blockownership structures are not related to agency issues. In fact, investors seem not to avoiddiversified firms per se. We suggest that observed board and ownership differences betweendiversified and focused firms are due to their being at different stages of corporate evolution.

Keywords: diversification discount, diversification strategies, corporate governance, board ofdirectors

JEL Classifications: G32/G34/F23

∗Corresponding author: Department of Finance, Southern Illinois University, Carbondale, IL 62901-4626;Phone: (618) 453-1421; Fax: (618) 453-5626; E-mail: [email protected]

We thank Mary Bange, Diane Denis, Stephen Ferris (the editor), an anonymous referee, and participants atthe 2001 Financial Management Association and the 2002 Australasian Finance and Banking Conferencesfor their helpful comments that have improved the paper.

489

490 M. Singh et al./The Financial Review 39 (2004) 489–526

1. Introduction

Recent evidence on corporate diversification strategy predominantly suggeststhat diversification can be a negative value proposition. Denis, Denis, and Sarin(1997), Servaes (1996), Berger and Ofek (1994), Lang and Stulz (1994), and Billetand Mauer (2000) report direct evidence of corporate value losses attributable to busi-ness diversification. Comment and Jarrell (1994), John and Ofek (1994), Bengtsson(2000), and Gillan, Kensinger, and Martin (2000), among others, report evidence ofvalue gains from corporate refocusing strategies. The literature offers numerous anddiverse explanations of the negative association between corporate value and corpo-rate diversification or scope. One such explanation is the existence of agency conflictbetween owners and managers. Denis, Denis, and Sarin (1997) provide evidence onthe agency hypothesis by analyzing diversification levels and changes therein, as in-fluenced by ownership structure and the market for corporate control. They report anegative relation between the level of diversification and insider ownership, whichthey interpret as evidence of divergent interests of firm managers and shareholders.They also argue that a negative relation between block ownership and the level ofdiversification stems from agency problems that occur when managers are not mon-itored by external shareholders. However, they do not find support for the converseargument, namely, that firms with higher managerial and block ownership undertakemore valuable diversification. More important, they find little evidence that the valueloss from diversification is related to either managerial or block ownership. Further-more, the link between internal corporate governance mechanisms and the value ofdiversification has not been investigated. Thus, there is a lack of evidence on thedirect relation between the negative value consequences of a diversification strategyand the existence or resolution of agency conflicts.

We take this absence of direct evidence on the relation between value-destroyingdiversification and agency conflicts as our starting point and address two basic ques-tions. The first question is whether any observed cross-sectional differences betweenthe ownership and governance characteristics of diversified and focused firms areunambiguous indications of the existence of agency conflict, as reflected in financialperformance measures. Our contribution to the current debate on the issue emanatesfrom our analysis of the differences in internal governance structures to find comple-mentary evidence on the existence, or lack thereof, of conflicts in diversified firmsand the related performance implications. Our findings suggest that there are signif-icant internal corporate governance and ownership differences between focused anddiversified firms. However, agency conflict does not seem to be a valid explanation ofcross-sectional differences in corporate scale and scope for the two groups. We sug-gest an alternative explanation related to corporate lifecycle stage differences and theobserved cross-sectional differences in ownership and governance across diversifiedand focused firms.

The second question we pose is dynamic in nature and relates to the performanceimplications of adopting a diversification strategy. We analyze whether the strategy of

M. Singh et al./The Financial Review 39 (2004) 489–526 491

expanding corporate scope is more likely to lead to performance deterioration ratherthan improvement. We find that diversification per se does not harm average per-formance. To uncover any link between agency structure and performance-reducingdiversification, we analyze diversification-related positive and negative performanceoutcomes as a function of ownership and governance variables. The evidence sup-ports the nonexistence of an agency-conflict impact on the performance outcomes ofcorporate scope expansion.

In summary, this paper makes two contributions. First, we relate the adoptionof diversification or focus strategies to agency variables (i.e., ownership and internalgovernance) in a unified framework. Second, we investigate operating-performanceconsequences of diversification, in contrast to previously analyzed value and stockperformance consequences, and provide evidence of the nonexistence of an agencyinfluence on either the choice or the performance consequences of a diversificationstrategy.

2. Previous literature

2.1. Valuation consequences of corporate scope strategies

The literature on corporate diversification has evolved a logical framework toexplain the costs and benefits of business and geographic diversification strategies.Numerous empirical studies test the theories. On the positive side, product diversifi-cation has been shown to yield multiple benefits ranging from enhanced productivityto financial gains to organizational synergies and strategic advantages.

The logic advanced in favor of diversification includes managerial economies ofscale and efficiency gains through the creation of an additional layer of management,specifically undertaking the coordinating function among various specialized divi-sions. A multisegment corporation with a well-designed organizational structure cancombine the benefits of specialization with those of diversification through a moreefficient decision process.

Diversification can yield financial synergies and increased debt capacity.Lewellen (1971) suggests that by combining projects with uncorrelated cash streams,diversified firms are able to ensure stability of earnings. This coinsurance effect,by allowing greater debt capacity, generates value through an increased interest taxshield and through the smoothening out of losses and gains. Majd and Myers (1987)suggest that as long as some segment of a diversified firm incurs losses in someperiods, the overall tax liability of the multiple segment corporations will be less thanthat of single segment firms. In addition, Williamson (1986) argues that diversifiedfirms reap the benefits of internal capital markets, which are less costly than externalcapital markets due to their ensuring better allocation of resources among compet-ing projects, more efficient information sharing, and more effective post-investmentmonitoring and control. Stulz (1990) suggests that by avoiding the need to go to ex-ternal capital markets for funds, diversified firms can mitigate the underinvestmentproblem (Myers, 1977).


Another argument for diversification has been advanced in terms of long-termperformance and growth potential. Gort, Grabowski, and McGuckin (1985) suggestthat firms diversify to tap the comparative advantage of new businesses relative tothe existing business. Here, diversification is induced more by poor performance andgrowth constraints. Lang and Stulz (1994) support this logic by reporting that firmsdiversify away from low growth to potentially high growth activities.

Regulatory restrictions, primarily antitrust legislation blocking expansion oflarge firms within their own core business areas to prevent market concentration,have been proposed as the catalyst in the 1960s conglomerate mergers wave. Shleiferand Vishny (1991) conjecture that firms with excess cash and experiencing an environ-ment favorable to issuing equity escaped antitrust regulation by acquiring businessesoutside their main line of business. Stigler (1966) argues that the 1950 merger acthad a strong deterrent effect on horizontal mergers by forcing firms to expand in di-verse lines of businesses. However, Servaes (1996) reports that even in the 1960s and1970s diversified firms were not valued at a premium; rather there was, in the 1960s,a substantial discount that declined to zero in the 1970s. His evidence also suggeststhat the closer was the alignment of managers’ interests with those of shareholders,the lesser was the extent of value-destroying diversification. Similarly, Ravenscraftand Scherer (1987) and Wernerfelt and Montgomery (1988) report that even duringthe 1960s merger wave, conglomerate acquisitions were unsuccessful in terms ofpost-merger financial performance and survival.

Event study results comparing acquirers’ returns to diversifying versus nondi-versifying acquisitions suggest that the former acquisitions had negative value conse-quences (Eckbo, 1985; Sicherman and Pettway, 1987; Morck, Schleifer, and Vishny,1990). Although there is no evidence that unrelated acquisitions were less successfulthan the related ones, Kaplan and Weisbach (1992) provided evidence that unrelatedacquisitions are more likely to be divested relative to related ones.

One of the rationalizations of the adverse valuation consequences of diversifica-tion is advanced by Bhide (1990), who suggests that the conditions that precipitatedthe conglomerate wave of the 1960s and early 1970s changed in the 1980s and hencethe current trend toward refocusing. The argument is that previous product and capitalmarket inefficiencies, incompleteness, and information asymmetries made externalmarket interactions costly, making internal market transactions, achieved through di-versification, relatively higher value generating. The current trend of rising capitalmarket sophistication, reduced regulation, better information transparency, and glob-alization has eroded the benefits of diversification (Markides, 1995). Simultaneously,the costs of diversification have gone up, given the increased business environmentuncertainty and volatility resulting in loss of information and control in diversifiedfirm hierarchies (Hill and Hoskisson, 1987).

There is also evidence that diversification adversely affects long-term stockperformance. Comment and Jarrell (1994) report a negative relation between abnormalstock returns and various measures of diversification. Similarly, Lang and Stulz (1994)report a negative relation between the degree of diversification and Tobin’s Q. These


results support the view that diversification reduces value. Indirectly, evidence thatincreasing focus creates value has been interpreted as an indicator of value-destructionby diversification. Berger and Ofek (1994), for example, suggest that the trend towardincrease in focus as reported by such studies as Comment and Jarrell (1994) andLiebeskind and Opler (1993) is a consequence of corporations realizing that unrelateddiversification decreases value. Because increasing focus leads to positive valuationeffects for firms that sell assets (John and Ofek, 1994), it is proposed that the initialdiversification could have been value destroying.

2.2. Agency explanation of diversification strategy

Given that diversification can facilitate creation of internal capital markets andhelp increase debt capacity, Jensen’s (1986) argument that managers with larger debtcapacity and access to free cash flow may undertake nonvalue-maximizing invest-ments could explain value-discounting diversification. Further, due to the interseg-ment transfer of cash, there is a greater possibility of managers undertaking negativevalue projects relative to single-segment firms. A related hypothesis in terms of cross-subsidization predicts that, because even negative net worth subsidiaries survive aspart of a conglomerate, they contribute to value destruction of the diversified firm.On a similar note, Stulz (1990) suggests that diversified firms that have access tointernally generated funds can have a problem of overinvestment. Managers couldactually be investing in negative net present value projects because they are able toavoid market evaluation and monitoring of their projects.

It is argued that managers derive private benefits from diversification in termsof added power and prestige associated with managing large conglomerates (Jensen,1986) and size-related compensation benefits (Jensen and Murphy, 1991). In addi-tion to scale expansion, diversification also helps managers diversify the risk of theirpersonal human capital, which is tied to the firm that they manage (Amihud and Lev,1981). Morck, Schleifer, and Vishny (1990) and Amihud and Lev (1981) providedempirical evidence to this effect. Also, there is a managerial desire for entrench-ment playing its role in a suboptimal diversification strategy. Shleifer and Vishny(1989) argue that managers, in a bid to make themselves indispensable to the firmsthey manage, can opt for diversification at the cost of negative shareholders’ wealthconsequences, a result also suggested by Chen and Ho (2000).

Denis, Denis, and Sarin (1997) provide evidence on the agency hypothesis byanalyzing diversification levels and changes therein, as influenced by ownershipstructure and the market for corporate control. They interpret their finding of a neg-ative relation between the level of diversification and insider ownership as evidenceof divergent interests of managers and shareholders. They also argue that evidenceof a negative relation between block ownership and the level of diversification is amanifestation of agency conflict in terms of managers not being monitored by ex-ternal shareholders. However, they do not find support for the argument that firmswith higher managerial and block ownership undertake more valuable diversification.


More important, they report little evidence that the value loss from diversification isrelated to either managerial or block ownership.

Although the congruence of interest hypothesis suggests that higher insiderownership resolves agency conflict, it is not always the case. Higher insider ownershipcan also mean managerial entrenchment in terms of dominance of voting power. Inthis case, although suboptimal behavior can lead to some managerial wealth loss, theprivate benefits generated from corporate diversification can outweigh that loss. So,there could actually be a positive relation between inside ownership and the adoptionof value and performance-discounting strategies.

It is generally accepted that large ownership stakes by block owners providethem the ability and incentive to monitor firm management. Thus, if diversificationwere agency driven, we would expect that firms with lower block ownership would bethe ones adopting diversification. Also, negative performance consequences wouldmore likely be associated with firms having lower block ownership.

2.3. Governance mechanisms and corporate restructuring

The financial economics literature is generally silent on the role of internalcorporate governance mechanisms in the choice and consequences of diversifica-tion and focus decisions. (See Fields and Keys (2003), for a review of the corpo-rate governance literature in financial economics.) However, in the field of strategicmanagement, studies analyze the corporate restructuring decision as influenced byinternal and external governance structure and dynamics.1 Johnson, Hoskisson, andHitt (1993) integrate financial economics and strategic perspectives to investigateboard involvement in corporate restructuring. Their fundamental premise is that, dueto the predominance of their monitoring role, board members, and particularly out-siders, involve themselves in restructuring only when management’s implementationof strategy is deficient. Johnson, Hoskisson, and Hitt report that although a highermanagerial equity stake makes board involvement in strategic decisions less crucial,outsider ownership and board involvement are positively related. They posit thatmanagerial strategic control and board involvement are substitute mechanisms forexecuting strategic reorientations. According to Johnson, Hoskisson, and Hitt, boardsthat have a higher proportion of outsiders are more liable to initiate restructuring.

Harrison (1987) argues that most boards emphasize monitoring more than thestrategic policy formation. Agency theory proponents argue that the market for cor-porate control disciplines managers (Fama and Jensen, 1983), but when there is poorperformance that requires restructuring action, internal governance systems intervenebefore external market forces activate. Johnson, Hoskisson, and Hitt (1993) posit that

1 For example, Johnson, Hoskisson, and Hitt (1993), Goodstein, Gautam, and Boeker (1994), and Gibbs(1993), among others, study various dimensions of agency and governance dynamics influencing diversi-fication and focus strategies.


boards generally prefer to promote firm efficiency, and hence help shareholder wealthpreservation, before letting the market impose discipline on management.

The size and diversity of the board can affect its ability to initiate strategic re-orientation. In this respect, the resources dependence theory suggests that increasedsize and diversity can yield benefits of creating a network with the external envi-ronment for securing a broader resources base (Pearce and Zahra, 1992; Pfeffer,1993). To adapt to the dynamic external environment, the strategic function of aboard assumes added importance during periods of corporate turbulence and declin-ing performance as argued by Mintzberg (1983). Goodstein, Gautam, and Boeker(1994), however, argue that more diverse boards are more likely to constrain strate-gic change as their preoccupation with governance and institutional role can resultin significant deficiency in their strategic roles. In addition, although larger size isexpected to increase the pool of expertise and resources (Pfeffer, 1993) and to reduceCEO dominance of the board (Singh and Harianto, 1989), it could inhibit strategicchange initiation by the board due to delays and disagreements (Shaw, 1981), lack ofmotivation (Jewell and Reitz, 1981), poor coordination (Gladstein, 1984), and groupadverse dynamics and conflicts (Olson, 1982). Yermack’s (1996) results suggestthat even in their monitoring functions, large boards are not always more effective.In fact, his results indicate that firms with smaller boards show superior financialperformance, have higher Tobin’s Q, and are more efficient in mitigating agencyconflict.

Thus, the size of the board can influence the choice and performance implicationsof a particular strategic option. Although the resources perspective suggests a positiverole for a large board size in choosing an appropriate strategy, large boards can allowCEO dominance, and, combined with a lack of monitoring, it can lead to suboptimalstrategic decisions. If the latter tendency dominates, we would expect firms withsmall boards to be able to restrict diversification as well as achieve positive valueconsequences of their strategic moves.

An important question concerns the market valuation of the agency conflict-resolving role of outsider board members. On the one hand, Demsetz (1983) andHart (1983) suggest that, in an efficient market, boards are irrelevant as the marketitself ensures congruence of interests of agents and principals. On the other hand,Fama (1980) and Fama and Jensen (1983) argue that outsiders, by providing expertknowledge and monitoring services, add value to firms. Although insiders are im-portant providers of operating and strategic guidance, outside directors act more likeguardians of the shareholders’ interests through monitoring. Empirical results supportthe argument that outside directors are more effective monitors and a critical disci-plining device for managers. In this respect, Coughlan and Schmidt (1985) report thatCEO turnover is positively associated with the proportion of outsiders on the board.Weisbach (1988) reports that, following poor performance, more outside directorsare added to the board. Thus, in an agency-theoretic perspective, boards are expectedto be effective monitors and a restraining force on managerial self-serving behavior(Kosnik, 1990).


Rosenstein and Wyatt (1990) report a positive stock-price reaction to firms’announcements of outsiders’ nomination to their boards. Although the existing em-pirical evidence relating board composition to performance is mixed,2 it is also re-ported that outside directors contribute to the value of firms through their evaluationof strategic decisions such as acquisition and mergers (Byrd and Hickman, 1982;Brickley and James, 1987) and through their role in the dismissal of inefficient andpoorly performing management (Weisbach, 1988). The evidence that poorly perform-ing firms seek to add additional outsiders (Hermaelin and Weisbach, 1988) to theirboards is consistent with the argument that outsider directors are value-enhancingagents.

Cochran, Wood, and Jones’s (1985) interpretation of a positive relation betweenthe adoption of golden parachutes and the proportion of outsiders on the board ofdirectors is that outsiders help achieve a better alignment between management andshareholders interests. Similarly, Lee et al. (1992) report that in going private transac-tions, outside directors play a crucial role in ensuring a fair price offer to shareholders.Brickley, Coles, and Terry (1994) report that the stock-price reaction to the adoptionof poison pills is positive in the case of firms with an outsider majority and negative inthe case of firms with a majority of inside directors. Thus, there is substantial evidencethat board composition is a significant determinant of the valuation consequences ofmajor corporate strategic choices.

In this regard, irrespective of whether restructuring involves diversification orfocusing, its consequences should be positive for the firms having lower agencyconflict. If, however, diversification as such is a negative value proposition, drivenpredominantly by agency behavior, then we should see more diversifying moves bythe firms with a higher degree of agency conflict. That is, we should find that firmswith lower outsider, and higher insider and associated outsider proportions, are morelikely to adopt diversification. Additionally, the negative performance consequencesof a particular strategy should be more likely attributable to these firms.

In sum, although previous literature has extensively analyzed the trend, modes,and valuation consequences of corporate diversification strategy, it has failed to tracethe causes and sources of the valuation discount attributable to higher levels of diversi-fication. A few studies that have tried to relate valuation implications of diversificationto the existence of agency cost have not been able to provide conclusive evidence.More important, corporate governance structure and dynamics have not been ana-lyzed in terms of their role as agency conflict deterrents. In addition, although therehave been a number of cross-sectional studies evaluating market value and marketperformance implications of different levels of diversification in a static framework,there is a significant lack of evidence on the accounting performance implications ofcorporate scope and changes therein.

2 For example, whereas Hermaelin and Weisbach (1988) find no significant relation between performanceand outsiders’ proportion on the board of directors, Baysinger and Butler (1985) report a positive relation.


Another emerging strand of diversification related research integrates the inter-national diversification dimension within the framework of business diversification.Denis, Denis, and Yost (2002) report an increasing trend in global diversificationover time. Although global diversification does not constitute a substitute strategyto industrial diversification, they reported that global diversification is also valuediscounting. It is also argued that the diversification discount may not be a univer-sal phenomenon and could be market, legal, and organizational context specific. Forinstance, Fauver, Houston, and Naranjo (2003) suggest that the financial, legal, andregulatory environments all have an important influence on the value of diversifi-cation. Further, in another paper Fauver, Houston, and Naranjo (in press) reportedthat although industrial diversification has a negative effect on firm value in theUnited Kingdom and the United States, there is no significant effect on firm value inGermany. Similarly, Lins and Servaes (2002) report that in emerging markets diver-sification is also value destroying. Although evidence points to negative implicationsof business diversification strategies, the debate is far from being conclusive.

Following the mainstream research on business diversification, we maintain anarrow scope of our analysis by focusing only on business diversification and corpo-rate governance linkage. Specifically, we do not integrate the international diversifi-cation dimension into this analytic framework.

More recent evidence suggests that the reported diversification discount could infact be an artifact of data selection and empirical approaches. For example, Villalonga(2004b), using inputs from a probit model, reports that either there is no diversificationdiscount or there exists a diversification premium. In a subsequent paper (2004a) sheuses new business segment data and finds that diversified firms trade at a premium.Graham, Lemmon, and Wolf (2002) report that when firms add business segmentsdue to reporting changes, there are no changes in the diversification discount. How-ever, some of the discount manifests itself when firms acquire units that are alreadydiscounted. Campa and Kedia (2002) show that the diversification discount could berelated to firm characteristics that lead them to diversify. Whited (2001) argues thatthe diversification discount could be an artifact of measurement errors, whereas Feeand Thomas (2004) state that the discount could be related to greater informationasymmetries associated with diversified firms.

Thus, although the issue of diversification–value relation is being rigorously de-bated, the empirical evidence is far from conclusive. More important, recent researchis highlighting the multidimensionality of the issue in terms of numerous firm andenvironment specific factors influencing the value–diversification relation.

Our contribution to the debate lies in relating simultaneously the firm governanceand ownership structures to the adoption and operating performance implications ofcorporate scope strategies. Specifically, we address the following questions:

1. Are there any significant ownership and governance structure differences be-tween diversified firms and focused firms? Are the differences unambiguouspointers to the existence of agency conflict in firms with a higher degree of


diversification? While Denis, Denis, and Sarin (1997) focus on the role ofownership structure, we extend their work by investigating the differences be-tween the internal governance mechanisms of focused and diversified firms.

2. Is there a significant relation between a higher degree of agency conflictand a greater propensity for firms to diversify? We address this issue withinstatic bivariate and multivariate, as well as dynamic bivariate and multivariateframeworks.

3. Is diversification predominantly a performance-discounting proposition? Ifnot, are the performance outcomes of expanding corporate scope conditionalon corporate governance and ownership structure?

We create an interface between the corporate governance and corporate struc-tural evolution frameworks and analyze how the internal governance mechanismsand structural characteristics relate to corporate scope and its performance implica-tions. We conduct this analysis by relating board and ownership characteristics withthe adoption and performance consequences of corporate diversification and focusstrategies.

Our findings indicate that, contrary to the agency hypothesis, firms with lowermanagerial and block ownership or insider dominated boards are not more likelyto adopt a diversification strategy. In addition, we find evidence suggesting thatdiversification is not performance discounting. Further, performance-discounting di-versification choices are not attributable to the firms having lower insider and blockownership or having insider dominated boards. Finally, our data suggest that agencyconflicts are not the cause of the performance discount in the case of poor perform-ing diversified firms. We offer an alternative explanation of the linkage betweenthe board and ownership characteristics of firms and their choice of diversificationstrategies. We provide supporting evidence that corporate evolutionary stage differ-entials between diversified and focused firms are, at least, partially responsible forthe observed differences between their board and ownership structures.

3. Sample selection and methods

We analyze a sample of large U.S. corporations (i.e., annual sales revenue of$1 billion or more) listed on the NYSE, AMEX, and Nasdaq during the period from1993 to 1998 period. We exclude firms belonging to the financial services industry(SIC 6000–6999) and regulated utilities (SIC 4900–4999). This yields a total of 1,113firms. Of these 1,113 firms, we eliminate 180 American depository receipts. For anadditional 156 firms, we did not have business segment data for our sample period.This leaves a total of 777 firms.

Given that the early 1990s witnessed a wave of corporate downscoping anddownscaling strategies, we chose to create a sample that is a balanced representationof the overall population of large corporations as dynamic entities adopting variousscope adjustment strategies. We select a mix of firms representing adoption of both


diversifying and refocusing strategies.3 We want to test the hypothesized relationsfor a sample of large firms. However, large firms are, in general, more diversified(Denis, Denis, and Sarin, 1997), and there is evidence that diversification destroysvalue (Berger and Ofek, 1994; Servaes, 1996; Denis, Denis, and Sarin, 1997). Byselecting firms that are both diversified and focused and adopting diversifying andfocusing strategies, we have a sample that will not yield results overly influenced bya particular subgroup of firms, namely diversified firms. For purposes of analyzingthe dynamic of scope changes, we measure the changes in the level of corporatediversification over the two-year window between 1995 and 1997. Of the 777 firms,137 firms changed their number of business segments reported over these two years.

Of the 640 firms that chose to keep their number of segments constant over thetwo-year period between 1995 and 1997, 413 firms were single-segment firms in theyear 1995 as well as in the year 1997. Following Lins and Servaes (1999) and Grahamand Rogers (2002) among others, to keep our data size manageable, we focus on arandomly selected (i.e., every ninth firm) subgroup of 42 with complete information(hereafter G5) out of the 413 focused firms for the purpose of analyzing ownershipand governance structure as related to the strategy of maintaining a status quo. Theremaining 227 diversified firms remained diversified without changing the number ofbusiness segments. We randomly select (i.e., every fifth firm) 38 firms with completeinformation from this group (hereafter G3).

Of the total of 137 firms that changed their business scope, there are 38 firmswith complete information that were initially diversified and adopted further diver-sification. We name this group G1. Further, there are 80 firms that were initiallydiversified but chose to refocus. We randomly select (i.e., every second firm) 40 ofthese firms with complete information and group them as G2. We found businessdiversification information on 19 firms with complete information that were focusedin 1995 but became diversified by 1997. This group is named G4. For our purposes,a diversified firm is the one that has business operations spread over more than one4-digit SIC code. A firm that reports a single SIC segment is categorized as a focusedfirm. Our sample, therefore, includes not only diversified and focused firms but alsothose firms that changed their business scope by either diversifying or refocusing.

We obtain our accounting data from Standard & Poor’s Research Insight. Thesource of ownership data is the Compact D database by Primark. We use the percentageof total equity held by the executives and the board members of a firm as a measure ofinsider ownership. We use the proportion of equity held by outside blockholders as aproxy for the incentive and capability of outside equity holders to monitor managers.For this purpose, an outside block equity holder is a stockholder having 5% or moreof the firm’s equity and not linked to firm management in either business or familyrelationships. Board characteristic information is gathered from proxy statements.We measure board composition in the traditional manner as in Baysinger and Butler(1985).

The groupwise breakdown of sample firms at this stage is as follows. Group G1has 42 firms, Group G2 has 19 firms, Group G3 has 38 firms, Group G4 has 40 firms,


and group G5 has 38 firms. Therefore, the total sample size at this stage is 177 firms.See the Appendix for a summary of the sample selection process.

Our analysis is both univariate and multivariate. In our multivariate logisticregression model for predicting scope shift strategies, the dichotomous dependentvariable is measured in terms of change in the diversification level over the periodfrom 1995 to 1997; the independent predictor variables are measured in the base year1995. This approach allows us to take into account the lagged effect of the independentvariables in influencing the dependent variable.

4. Corporate scope-agency characteristics relation

4.1. Static comparative analysis

The proponents of the hypothesis that diversification is agency driven arguethat self-serving managers adopt expanded corporate span to derive personal benefitsin terms of higher perquisites consumption (Jensen, 1986), personal risk reduction(Amihud and Lev, 1981) and securing entrenchment. In this scenario, it is expected thatdiversified firms will have significantly different corporate governance, ownership,and leverage characteristics compared to focused firms. The results of our comparativeanalysis to identify agency characteristic differentials across diversified and focusedfirms are presented in Table 1.

4.1.1. Board of directors

We measure board composition in the traditional manner as in Baysinger andButler (1985). We categorize directors as insiders, affiliated, and outside directors.Inside directors are those board members who are employees of the corporation.Outside directors are further divided into affiliated outsiders and independent out-siders. Whereas the former are those board members that have some type of family ofbusiness relationship with the corporation, the latter categories include those boardmembers that have no such relationship and are only associated with the corporationthrough board membership. In terms of board characteristics, diversified firms, onaverage, do not have a significantly larger board size, or a larger proportion of insidemembers on the board. In fact, focused firms have a significantly higher ratio ofinside directors relative to diversified firms. The difference pertaining to the ratioof independent outsiders is statistically insignificant across the two groups, althoughdiversified firms have a greater proportional representation of outsiders. Althougha larger proportion of independent directors in diversified firms can be due to anoverall larger corporate span of these firms, the evidence certainly does not indicateinsider dominance of the boards of diversified firms that can induce a possible agencyconflict.

The evidence of somewhat larger board size and higher proportion of indepen-dent outsiders in the case of diversified firms supports the resources-dependence

M. Singh et al./The Financial Review 39 (2004) 489–526 501Ta

ble

1

Sam

ple

desc

ript

ion

and

mea

nco

mpa

riso

nac

ross

dive

rsif

ied

and

focu

sed

firm

grou

ps

Thi

sta

ble

cont

ains

mea

nsco

mpa

riso

nte

sts

offi

nanc

ial

and

stru

ctur

alat

trib

utes

ofsu

bgro

ups

ofsa

mpl

efi

rms

for

the

base

year

1995

.Fir

ms

oper

atin

gin

one

busi

ness

segm

ent

inte

rms

of4-

digi

tSI

Cco

des

are

cate

gori

zed

asfo

cuse

dfi

rms.

Firm

sha

ving

oper

atio

nssp

read

over

two

orm

ore

busi

ness

segm

ents

defi

ned

inte

rms

of4-

digi

tSI

Cco

des

are

cate

gori

zed

asdi

vers

ifie

dfi

rms.

Inde

pend

ent

outs

ider

sar

eth

em

embe

rsof

the

boar

dth

atar

ene

ither

man

ager

sno

rlin

ked

toth

efi

rmth

roug

han

ybu

sine

ssor

fam

ilyre

latio

nshi

p.In

side

rson

the

boar

dre

fer

toco

mpa

nyex

ecut

ives

serv

ing

onth

ebo

ard

ofdi

rect

ors.

The

prop

ortio

nof

insi

deow

ners

hip

isde

fine

das

the

perc

enta

geof

equi

tyst

ock

held

byth

em

anag

emen

tand

mem

bers

ofth

ebo

ard

ofdi

rect

ors.

Blo

ckow

ners

hip

isde

fine

das

the

perc

enta

geof

tota

lsto

ckhe

ldby

stak

ehol

ders

havi

ng5%

orm

ore

equi

tyin

the

firm

.Ins

titut

iona

low

ners

hip

mea

sure

sth

epe

rcen

tage

ofa

firm

’seq

uity

held

byin

stitu

tiona

linv

esto

rs.T

otal

asse

tsar

ea

prox

yfo

rfi

rmsi

ze.R

etur

non

asse

tsan

din

dust

ry-a

djus

ted

retu

rnon

asse

tsar

epr

oxie

sfo

rop

erat

ing

perf

orm

ance

.Fre

eca

shfl

owis

cash

from

oper

atin

gac

tiviti

esne

tof

cash

divi

dend

san

dca

pita

lexp

endi

ture

s.T

hem

arke

tto

book

ratio

mea

sure

sgr

owth

oppo

rtun

ities

.Altm

an’s

ZSc

ore

prox

ies

forf

inan

cial

stre

ngth

.The

Her

find

ahli

ndex

mea

sure

sth

ede

gree

ofdi

vers

ific

atio

n.E

xces

sva

lue

refe

rsto

Ber

gera

ndO

fek’

s(1

994)

impu

ted

valu

em

easu

reof

dive

rsif

icat

ion

rela

ted

valu

edi

scou

ntor

prem

ium

.

Gro

upm

ean

Gro

upm

edia

n

Div

ersi

fied

Focu

sed

Dif

fere

nce

Div

ersi

fied

Focu

sed

Dif

fere

nce

Boa

rdsi

ze11

.05

10.2

20.

8311

101

Rat

ioof

insi

der

0.19

0.23

−0.0

4∗0.

190.

20−0

.01

Rat

ioof

inde

pend

ento

utsi

ders

0.70

0.69

0.01

0.73

0.73

0.00

Perc

enta

geof

bloc

kow

ners

hip

24.0

839

.27

−15.

19∗∗

∗16

.74

32.9

0−1

6.16

Perc

enta

geof

insi

deow

ners

hip

12.0

613

.77

−1.7

11.

364.

85−3

.49

Perc

enta

geof

inst

itutio

nalo

wne

rshi

p50

.12

57.9

7−7

.85∗∗

57.6

760

.87

−3.2

Mar

kett

obo

okat

io3.

533.

490.

042.

442.

54−0

.10

Rat

ioof

tota

ldeb

tto

tota

lass

ets

0.30

0.24

0.06

∗0.

200.

27−0

.07

Altm

an’s

Zsc

ore

3.19

5.87

−2.6

8∗∗∗

2.83

3.82

−1.0

1To

tala

sset

s($

MM

)97

2524

3672

88∗∗

∗29

3613

3316

03Sa

les

reve

nue

($M

M)

9057

2489

6568

∗∗∗

3280

1170

2310

Tota

lsha

res

outs

tand

ing

($M

M)

313.

5918

0.44

133.

15∗

104

6836

Ret

urn

onas

sets

4.82

5.70

−0.8

85.

05.

9−0

.9In

dust

ry-a

djus

ted

RO

A−0

.52

0.50

−1.0

2−0

.48

0.87

−1.3

5H

erfi

ndah

lind

ex0.

501.

00−0

.50∗∗

∗0.

501.

00−0

.50

Exc

ess

valu

e0.

030.

18−0

.15

0.03

0.15

−0.1

2∗∗

∗ Ind

icat

esst

atis

tical

sign

ific

ance

at0.

01le

vel.

∗∗In

dica

tes

stat

istic

alsi

gnif

ican

ceat

0.05

leve

l.∗ I

ndic

ates

stat

istic

alsi

gnif

ican

ceat

0.10

leve

l.


argument (Pearce and Zahra, 1992) that boards serve to provide network linkagesbetween firms and their external environment to secure strategic resources. Diversi-fied firms operate in larger and more varied business settings, and their product andcapital market linkages are comparatively more numerous and complex. Although ahigher proportion of independent outside directors may serve external market busi-ness interests on more favorable terms, more insiders on a board provide the necessaryexpertise and guidance in decision making as they are the best informed with regardto corporate operations. Independent outsiders provide valuable monitoring servicesalong with an expanded expert knowledge base. They also impart prestige and legiti-macy to corporate decisions, thus functioning as an important signaling device. Morediversified firms, being large, more visible, and relatively more significant marketplayers, derive strategic market signaling benefits from the larger and more diverseboards. However, the number of key executives that can be inducted to the corporateboard is generally limited to two or three, irrespective of whether a firm is diversifiedor focused. Therefore, on average, in terms of proportional representation, boards ofdiversified firms are relatively less dominated by insiders, a result that is inconsistentwith the predictions of the agency hypothesis of diversification.

The propensity for a firm to be diversified does not appear to depend on thelevel of higher insider ratio on the board. The agency explanation of diversificationpredicts presence of influential management on the board. An alternative explanationcould exist in terms of corporate structural and evolutionary stage differences betweenfocused and diversified units. A possible explanation of higher insider proportion inthe case of focused firms is that there may not be much need for external resourcesnetworking for these firms. In addition, focused firms have structural differencescompared to diversified firms in terms of their having a smaller corporate scale, beingat an earlier stage of corporate evolutionary path, having possible family/founderdominance and intensive firm specific knowledge/expertise that cannot be sharedwith outsiders. In addition, small and focused firms, being less visible, may notbe targeting prestige and legitimacy for their corporate decisions by having moreexternal board members. These structural factors could explain the lower proportionof independents on the boards of directors of focused firms.

4.1.2. Ownership structure

Although the extent of managerial ownership of a firm’s equity indicates thedegree of the congruence of management and shareholders’ interest, the proportionof the outside blockholders’ stake reflects the degree of external monitoring of man-agerial decisions. We use the log of percentage of total equity held by the executivesand the board members of a firm as a measure of inside ownership. Firms with a largeinsider equity ownership stake should have lower agency conflict and lower agencycosts. We use the proportion of equity held by outside blockholders as a proxy for theincentive and capability of outside equity holders to monitor managers. For this pur-pose, an outside block equityholder is a stockholder having 5% or more of the firm’s


equity and not linked to firm management in either business or family relationships.A larger blockholder equity stake would indicate greater incentives and capabilitywith outside blockholders to monitor management.

One of the implications of the agency explanation of diversification is that morediversified firms will have lower insider ownership3 and a lower outsider blockand institutional ownership.4 The evidence on this issue provides some indicationof the existence of agency conflict for our sample firms. Institutional, as well asblock ownership, proportions are lower for the group of diversified firms relative tofocused firms; the differences in both variables are statistically significant. Agencymay appear to be an explanation of these diversified firms, as managers may havemore of a chance to indulge in self-serving behavior in these firms. The differencein insider ownership is not significant.

4.1.3. Leverage

In an agency theoretic framework, higher leverage can either be a consequence ofself-interested managers opting for higher leverage to increase voting power (Harrisand Raviv, 1990) or an agency-conflict deterrent devised to pre-empt free cash flowavailability to managers (Jensen, 1986; Stulz, 1990). Although our results indicate asignificantly higher leverage ratio for diversified firms, we cannot unambiguouslyconclude which of the two situations the higher leverage represents. However, giventhat inside ownership—and, hence, voting power with insiders—is not significantlyhigher for the diversified group, the possibility that agency conflicts drive higherleverage in diversified firms is remote. In fact, possibly due to higher leverage, theyhave a lower Altman’s Z score and, hence, are more prone to bankruptcy, a situationevery manager would like to avoid for the safety of his or her human capital investmentin the firm.

4.1.4. Size and growth

Diversified firms in our sample have, on average, more total assets and net annualsales. The larger average size of diversified firms indicates complimentarity of scaleand scope dimensions of corporate span.

4.1.5. Performance

Our results do not provide unambiguous indications about differential perfor-mance across diversified and focused firms. In terms of operating performance,

3 A weaker alignment of manager-owner interests in terms of Jensen’s “convergence-of-interest”hypothesis.

4 An indicator of lack of effective external decision control and monitoring.


measured as the return on assets (ROA) and the industry-adjusted return on assets,there is no significant difference between the two groups. In general, in our bivariatesetting, we cannot distinguish between the performance implications of being a fo-cused or a diversified firm. However, consistent with previous evidence, Berger andOfek’s (1994) excess value measure is significantly higher for focused firms (0.18)relative to diversified firms (0.03).

4.2. Static multivariate analysis

Although our bivariate analysis does not suggest that governance and ownershipstructure differentials indicate the existence of agency conflict in diversified firms,we pose the question of which firm-specific governance and ownership characteristicsare more likely to result in firms being diversified units. To delineate more clearly therole of agency factors in determining corporate scope, we control for other relevantinfluences on the decision to be a diversified firm.

If agency conflicts drive diversification, firms with a higher representation ofinsiders on the boards would be more likely members of diversified firms group.Similarly, firms with lower managerial ownership, lower block ownership, andlower institutional ownership would be expected to have a higher tendency to bediversified.

In Table 2, we report the results of logit analysis, controlling for leverage, size,growth opportunities, and performance as factors influencing corporate scope. Weestimate five different models with these control variables being common to all themodels and five test variables being introduced separately in each of the models. Ingeneral, the models show that firms with higher leverage, larger firms, firms withhigher market to book, and firms with higher profitability are the ones that have agreater propensity to be diversified firms. The Berger and Ofek (1994) excess valuemeasure has a significant negative coefficient and indicates that firms with lowerexcess value are more likely to be diversified. The test variables, inside ownership,institutional ownership, block ownership, board size, ratio of independent board mem-bers, and ratio of insiders on the board all have insignificant coefficients. Thus, oncewe control for firm specific characteristics, agency conflict in terms of either owner-ship structure or governance mechanisms, does not seem to predict firm membershipin diversified versus focused groups.

An interesting result relates to ownership structure. Although in the bivariatesetup, we do find a significant difference in the institutional and block ownershipof diversified and focused firms, in the multivariate setup, none of the ownershipvariables relate significantly to the likelihood of firms being diversified.

Overall, the results in Tables 1 and 2 provide only limited support for the agencyexplanation of firms’ choice of a larger corporate scope. It appears that differences inownership and board characteristics across diversified and focused firms are relatedmore to the exigencies of their operating in differential business environments andhaving structural differences, and less to their agency characteristics.


Table 2

Logit regression for predicting diversified versus focused firm group membership

This table contains logit regression results for predicting firms’ diversified versus focused group mem-bership. The dichotomous dependent variable is defined = 1 if a firm is diversified and zero otherwise.Firms operating in one business segment in terms of 4-digit SIC codes are categorized as focused firms.Firms having operations spread over two or more business segments defined in terms of 4-digit SIC codesare categorized as diversified firms. Independent outsiders are the members of the board that are neithermanagers nor linked to the firm through any business or family relationship. Insiders on the board referto company executives serving on the board of directors. The proportion of inside ownership is definedas the percentage of equity stock held by the management and members of the board of directors. Blockownership is defined as the percentage of total stock held by stakeholders having 5% or more equity inthe firm. Institutional ownership measures the percentage of a firm’s equity held by institutional investors.Leverage is defined as the ratio of total debt to total assets. Log of total assets is a proxy for firm size.Industry-adjusted return on assets is a proxy for operating performance. The market to book ratio is intro-duced as a control variable to isolate influence of growth opportunities. Excess value refers to Berger andOfek’s (1994) imputed value measure of diversification related value discount or premium.

Model 1 Model 2 Model 3 Model 4 Model 5

Leverage 2.65∗∗ 2.67∗∗ 3.21∗∗∗ 2.09 2.97∗(0.05) (0.04) (0.01) (0.16) (0.06)

Firm size 1.25∗∗∗ 1.16∗∗∗ 1.02∗∗∗ 0.73∗∗ 0.65∗∗0.00 0.00 0.00 0.02 0.03

Market to book ratio 0.18∗ 0.19∗∗ 0.16∗ 0.09 0.080.06 0.04 0.08 0.31 0.40

Industry-adjusted ROA 0.12∗∗ 0.13∗∗ 0.13∗∗ 0.08 0.10∗0.04 0.02 0.02 0.16 0.10

Excess value −1.61∗∗∗ −1.68∗∗∗ −1.63∗∗∗ −1.59∗∗∗ −1.53∗∗∗0.00 0.00 0.00 0.01 0.01

Inside ownership 0.170.20

Institutional ownership −0.020.12

Block ownership −0.010.16

Board size 0.01 −0.010.90 0.93

Board independents’ ratio 0.410.83

Board insiders’ ratio −4.140.13

Constant −10.37 −8.57 −8.01 −5.98 −4.180.00 0.00 0.00 0.02 0.09

∗∗∗ Indicates statistical significance at 0.01 level.∗∗ Indicates statistical significance at 0.05 level.∗ Indicates statistical significance at 0.10 level.

5. Changes in the level of diversification

A logical question relates to the dynamics of corporate span strategy as deter-mined by agency conflict and the ownership and governance mechanisms devisedto alleviate such a conflict. We analyze the dynamics of diversification strategy by


investigating changes in the number of business segments reported by the samplefirms over the period from 1995 to 1997. We use multivariate logit analysis to ex-amine whether firm specific board and ownership characteristics can predict scopeshift strategies. We control for other relevant firm specific influences on a decisionto adopt a strategy of diversification or focusing.

5.1. Multivariate logit analysis

The results of logit analysis aimed at identifying the contribution of variousgovernance and ownership characteristics in precipitating or constraining the decisionto diversify are reported in Table 3. Five different models are estimated for this part ofthe analysis. We control for size, market to book ratio, and performance as measured byindustry-adjusted return on assets. We find that larger firms, and firms with highermarket to book ratio, have a greater propensity to increase their focus. In general,firms with higher excess value have a tendency toward becoming more diversified.The results are consistent with the previous results in terms of diversifying firms beingsimilar to initially focused firms. This also suggests that while focused firms try todiversify, diversified firms try to focus in being dynamic entities adjusting their scaleand scope. Thus, the diversification or focusing adoption strategy may be a functionof the initial configuration of a firm.

The significantly negative coefficient for inside ownership suggests that agencycould be driving the move toward diversification. However, equally plausible is theargument that firms increase their total equity as they grow and diversify, therebypossibly reducing the inside ownership in diversified firms. One can also argue thatlower managerial voting power associated with a higher probability of diversificationprecludes, to a large extent, the possibility of a diversification strategy as an agencyconsequent suboptimal choice. However, we cannot unambiguously conclude whichof the two competing hypotheses, convergence of interest or entrenchment, is driv-ing the results. Although in the case of Denis, Denis, and Sarin’s (1997) results, theexistence of a diversification discount supports the agency argument, in our analysiswe have no such evidence of the negative performance implications of a diversi-fication strategy.5 Therefore, we cannot unambiguously conclude that managerialself-interest drives the adoption of a diversification strategy. We further investigatethe relation between inside ownership and degree of focus later in the paper allowingfor nonlinearity.

The significant positive coefficient of institutional ownership suggests that firmswith higher institutional monitoring are more likely to diversify, thereby suggestingthat diversification is not an agency phenomenon. A significant positive coefficient

5 Later, we provide evidence that diversification does not negatively affect performance. Thus, the weightof our evidence goes to support our argument of the absence of an agency-induced diversificationstrategy.


Table 3

Logit regression for predicting diversifying versus focusing strategy adoption by firms

The table contains logit regression results for predicting firms’ adoption of diversifying versus focusingstrategies. Firms operating in one (more than one) business segment in terms of 4-digit SIC codes arecategorized as focused (diversified) firms. Firms increasing (decreasing) number of business segments arecategorized as diversifying (focusing) firms. For diversifying (focusing) firms the dichotomous dependentdummy takes a value = 1 (zero). Independent outsiders are the members of the board that are neithermanagers nor linked to the firm through any business or family relationship. Insiders on the board refer tocompany executives serving on the board of directors. The proportion of inside ownership is defined as thepercentage of equity stock held by the management and members of the board of directors. Block ownershipis defined as the percentage of total stock held by stakeholders having 5% or more equity in the firm.Institutional ownership measures the percentage of a firm’s equity held by institutional investors. Log oftotal assets is a proxy for firm size. Industry-adjusted return on assets is a proxy for operating performance.The market-to-book ratio is introduced as a control variable to isolate the influence of growth opportunities.Excess value refers to Berger and Ofek’s (1994) imputed value measure of diversification related valuediscount or premium. Whereas the dichotomous dependent variable is measured in terms of change in thediversification level over the 1995–1997 period, the independent predictor variables are measured in thebase year 1995.


Firm size −0.59∗∗ −0.25 −0.18 −0.94∗∗ −0.95∗∗0.03 0.24 0.43 0.03 0.03

Market to book ratio −0.21∗ −0.12 −0.08 −0.20∗ −0.20∗0.06 0.20 0.42 0.09 0.10

Industry-adjusted ROA 0.05 0.01 0.02 −0.08 −0.080.41 0.78 0.58 0.26 0.26

Excess value 1.11∗ 0.91 0.80 2.58∗∗∗ 2.57∗∗∗0.09 0.12 0.17 0.00 0.00

Inside ownership −0.36∗∗0.04

Institutional ownership 0.02∗0.10

Block ownership 0.0030.80

Board size 0.31∗ 0.32∗0.09 0.09

Board independents’ ratio −0.430.85

Board insiders’ ratio 0.770.84

Constant 6.37∗∗∗ 1.71 2.04 5.26∗ 4.810.01 0.34 0.36 0.08 0.18

Model chi-square 10.02 6.71 3.97 14.63 14.64∗∗∗ Indicates statistical significance at 0.01 level.∗∗ Indicates statistical significance at 0.05 level.∗ Indicates statistical significance at 0.10 level.

for board size indicates that firms with larger boards are more likely to expand theircorporate scope. The multivariate results are, to some extent, similar to the previouslyreported bivariate results in that firms with a larger board size, lower inside ownership,and a higher institutional ownership are more likely to adopt diversification.


5.2. Are diversification dynamics conditional on initial performance?

Given that one cannot unambiguously conclude that the adoption of a diver-sification strategy is agency driven, we argue that, conditional on their initial poorperformance, firms will adjust their corporate scope to improve their value. That is,if a diversified firm is discounted initially, it will refocus. Similarly, if a focused firmis discounted, it will increase its scope. In both instances, unless the strategy is illdesigned and poorly executed, both focusing and diversifying firms will gain value.Analogously, if a positive excess value firm changes its strategy, it can actually suffervalue losses irrespective of the strategy it is adopting.

Table 4 reports the changes over time (i.e., 1995–1998) in governance, owner-ship, and performance conditional on initial diversification strategy and excess value,classified by whether firms are focused or diversified and whether they are sellingat a premium or a discount. Consistent with our argument, both the means and themedians indicate that firms that were discounted changed their strategy; diversifiedfirms refocused, and focused firms diversified. The shifts seem to have been aimedat increasing value, as both the focusing and diversifying firms experience gains inexcess value and industry-adjusted return on assets. One implication is that diver-sifying does not harm performance. Diversification by poorly performing focusedcompanies leads to value creation. The second implication is that, given that blockownership and institutional ownership increases for both these groups, institutionaland block investors do not consider one strategy as superior to the other. In fact, firmsthat diversify attract greater increases in block and institutional investment. It appearsthat block and institutional investors have the ability to target expected value gainsand that they do not see a diversification strategy as value destroying. In terms ofgovernance, for initially discounted diversified firms that focus, board size and theratio of insiders go down and the ratio of independents goes up, consistent with ourconjecture.

Looking at firms with an initial premium, we find that although diversifiedfirms diversify further and generate small value gains, they lose in terms of industry-adjusted ROA. Focused firms also adopt diversification, but in that process they losein terms of industry-adjusted ROA as well as excess value. The diversification strategyleads to value loss only for the group of focused firms that were positive excess valuefirms to begin with. For diversified premium firms that diversified further but did notlose value, block and institutional ownership increased, whereas for focused premiumfirms that destroyed value by diversifying, block and institutional ownership wentdown. Thus, we find that it is not the strategy per se that determines investors’ decisionto invest but the performance implications of the strategy. For value-losing focusedfirms board size increases, the ratio of independents on the board goes down, and theratio of insiders goes up. Even here, we have value losses from diversification thatmay not be unambiguously attributable to diversification per se, as we also see thatthese firms actually change their governance structure in a manner that can causeagency conflicts.


ble

4

Dyn

amic

sof

gove

rnan

ce,o

wne

rshi

p,pe

rfor

man

ce,a

nddi

vers

ific

atio

nco

ndit

iona

lupo

nin

itia

lval

ue-d

estr

oyin

gor

valu

e-en

hanc

ing

stra

tegi

es

Thi

sta

ble

desc

ribe

sch

ange

sov

ertim

ein

the

gove

rnan

ce,o

wne

rshi

pst

ruct

ure,

and

fina

ncia

lper

form

ance

ofdi

vers

ifie

dan

dfo

cuse

dfi

rms

cond

ition

alon

thei

rin

itial

stra

tegy

ofdi

vers

ific

atio

nbe

ing

valu

ede

stro

ying

orva

lue

crea

ting.

Firm

sop

erat

ing

inon

ebu

sine

ssse

gmen

tin

term

sof

4-di

gitS

ICco

des

are

cate

gori

zed

asfo

cuse

dfi

rms.

Firm

sha

ving

oper

atio

nssp

read

over

two

orm

ore

busi

ness

segm

ents

defi

ned

inte

rms

of4-

digi

tSIC

code

sar

eca

tego

rize

das

dive

rsif

ied

firm

s.In

depe

nden

tout

side

rsar

eth

em

embe

rsof

the

boar

dth

atar

ene

ither

man

ager

sno

rlin

ked

toth

efi

rmth

roug

han

ybu

sine

ssor

fam

ilyre

latio

nshi

p.In

side

rson

the

boar

dre

fer

toco

mpa

nyex

ecut

ives

serv

ing

onth

ebo

ard

ofdi

rect

ors.

The

prop

ortio

nof

insi

deow

ners

hip

isde

fine

das

the

perc

enta

geof

equi

tyst

ock

held

byth

em

anag

emen

tand

mem

bers

ofth

ebo

ard

ofdi

rect

ors.

Blo

ckow

ners

hip

isde

fine

das

the

perc

enta

geof

tota

lsto

ckhe

ldby

stak

ehol

ders

havi

ng5%

orm

ore

equi

tyin

the

firm

.Ins

titut

iona

low

ners

hip

mea

sure

sth

epe

rcen

tage

ofa

firm

’seq

uity

held

byin

stitu

tiona

lin

vest

ors.

Exc

ess

valu

ere

fers

toB

erge

ran

dO

fek’

s(1

994)

impu

ted

valu

em

easu

reof

dive

rsif

icat

ion

rela

ted

valu

edi

scou

ntor

prem

ium

.The

Her

find

ahli

ndex

mea

sure

sth

ede

gree

offo

cus.

The

figu

res

inth

epa

rent

hese

sar

em

edia

ns.

Div

ersi

fied

Dis

coun

ted

Focu

sed

Dis

coun

ted

Div

ersi

fied

Prem

ium

Focu

sed

Prem

ium

1995

1998

1995

1998

1995

1998

1995

1998

Blo

ckow

ners

hip

26.8

332

.32

41.2

953

.85

22.3

427

.70

39.9

131

.23

(22.

52)

(28.

51)

(34.

96)

(43.

65)

(15.

43)

(15.

10)

(28.

76)

(23.

84)

Inst

itutio

nalo

wne

rshi

p49

.34

55.2

351

.91

61.8

250

.40

55.8

556

.90

55.3

1(5

5.95

)(6

4.08

)(6

0.87

)(7

0.51

)(5

7.60

)(6

0.02

)(5

5.48

)(5

6.75

)In

side

row

ners

hip

17.9

73.

6713

.00

10.2

98.

838.

8013

.59

8.88

(1.5

2)(0

.49)

(4.6

9)(3

.71)

(1.3

0)(0

.56)

(3.2

0)(1

.65)

Boa

rdsi

ze10

.96

10.5

311

.60

10.0

711

.50

11.7

19.

9510

.85

(11.

00)

(10.

50)

(11.

00)

(10.

00)

(11.

00)

(11.

00)

(9.0

0)(1

0.00

)R

atio

ofbo

ard

inde

pend

ents

0.70

0.74

0.62

0.66

0.68

0.71

0.69

0.63

(0.7

1)(0

.78)

(0.5

5)(0

.67)

(0.7

1)(0

.77)

(0.7

2)(0

.67)

Rat

ioof

insi

ders

onth

ebo

ard

0.19

0.18

0.17

0.21

0.19

0.20

0.26

0.28

(0.1

7)(0

.20)

(0.2

0)(0

.17)

(0.2

0)(0

.20)

(0.2

2)(0

.27)

Her

find

ahl’s

inde

x0.

490.

531.

000.

670.

520.

471.

000.

72(0

.50)

(0.4

9)(1

.00)

(0.5

7)(0

.53)

(0.4

3)(1

.00)

(0.7

9)In

dust

ry-a

djus

ted

RO

A−0

.48

−0.0

01−0

.47

0.06

0.49

0.44

0.83

0.37

(−0.

30)

(1.0

4)(0

.99)

(−1.

45)

(0.6

4)(−

0.09

)(0

.08)

(0.2

9)C

hang

ein

exce

ssva

lue

over

0.25

0.28

0.00

3−0

.14

the

peri

od19

95–1

997

(0.2

7)(0

.34)

(0.0

1)(−

0.11

)


Overall, there are two salient results in Table 4. First, firms suffering from poorperformance adopt a strategic shift (i.e., diversifying as well as focusing) to improveperformance. In following this strategy, both diversifying and focusing firms gainvalue. Thus, diversification can and does create value. Second, ownership structurecan very well be determined by expected value gains. If firms increase their value,institutional and block investors increase their holding irrespective of the strategyadopted.

To find further support for our argument that governance and ownership differ-ences across diversifying and focusing firms may not be reflective of agency conflict,in Table 5 we present the base year (1995) initial governance, ownership, and finan-cial characteristics for the five subgroups created on the basis of the scope strategyadopted.

5.2.1. Board of directors

Within the diversified firms, subgroups do not manifest significant differencesin board size or composition. Within the focused subgroups also, the board size andcomposition differences between those diversifying and those remaining focused arestatistically insignificant. The lack of a significant difference in the ratio of insidersand independent outsiders between the two focused subgroups adopting diversifica-tion (G4) and remaining focused (G5) implies that board size and composition arenot related to the decision to remain focused (G5) or to adopt diversification (G4).That is, similar governance structures can lead to the adoption of different strategieswithin focused firms. There could be nongovernance factors explaining adoption of adiversification strategy as opposed to opting for the status quo. Similarly, within thediversified subgroup, a differential strategy of refocusing (G2) and further diversifi-cation (G1) is not influenced by board size and composition.

Comparisons across initially diversified and focused groups reveal that diversi-fied subgroups G1 (diversifying) and G2 (focusing) have significantly lower insidersratio compared to G4, the initially focused subgroup that diversified. The significantdifference in the ratio of insiders between the two diversifying subgroups G4 (initiallyfocused) and G1 (initially diversified) implies that a similar strategic choice (scopeexpansion) could be adopted irrespective of board composition differences. In sum,the results suggest that the adoption of a diversification strategy as such may not beagency induced.

5.2.2. Ownership

Agrawal and Knoeber (1996) argue that firms can adopt different agency de-terrent mechanisms as substitutes. It is possible that, rather than board composition,our sample firms are using ownership structure for alleviating the agency problem.In this scenario we should expect a relation between ownership structure and diver-sification rather than that between governance structure and diversification. In terms


ble

5

Mea

nco

mpa

riso

nof

base

year

(199

5)ch

arac

teri

stic

sof

subg

roup

sba

sed

onsc

ope

stra

tegy

adop

tion

Thi

sta

ble

cont

ains

mea

nsco

mpa

riso

nte

sts

offi

nanc

iala

ndst

ruct

ural

attr

ibut

esof

five

subg

roup

sof

sam

ple

firm

sba

sed

onth

edi

vers

ific

atio

nle

vels

and

chan

ges

ther

ein.

Firm

sop

erat

ing

insi

ngle

busi

ness

segm

enta

refo

cuse

dfi

rms

and

thos

ein

volv

edin

mul

tiseg

men

tbus

ines

ses

are

dive

rsif

ied

firm

s.B

usin

ess

segm

ents

are

defi

ned

inte

rms

of4-

digi

tSIC

code

s.Fi

rms

incr

easi

ng(d

ecre

asin

g)nu

mbe

rofb

usin

ess

segm

ents

over

the

two-

year

peri

odbe

twee

n19

95an

d19

97ar

eca

tego

rize

das

dive

rsif

ying

(foc

usin

g)fi

rms.

Div

ersi

fied

firm

sth

atin

crea

seth

eir

num

ber

ofse

gmen

tsar

epl

aced

ingr

oup

G1,

dive

rsif

ied

firm

sth

atsh

iftt

oop

erat

eas

sing

lese

gmen

tare

plac

edin

grou

pG

2,di

vers

ifie

dfi

rms

that

repo

rtno

chan

gein

num

ber

ofse

gmen

tsfa

llin

grou

pG

3,si

ngle

segm

entf

irm

sth

atin

crea

seth

eir

num

ber

ofse

gmen

tsar

epl

aced

ingr

oup

G4,

and

sing

lese

gmen

tfir

ms

rem

aini

ngas

such

over

the

win

dow

are

grou

ped

into

G5.

Inde

pend

ento

utsi

ders

are

the

mem

bers

ofth

ebo

ard

that

are

neith

erm

anag

ers

nor

linke

dto

the

firm

thro

ugh

any

busi

ness

orfa

mily

rela

tions

hip.

Insi

ders

onth

ebo

ard

refe

rto

com

pany

exec

utiv

esse

rvin

gon

the

boar

dof

dire

ctor

s.T

hepr

opor

tion

ofin

side

owne

rshi

pis

defi

ned

asth

epe

rcen

tage

ofeq

uity

stoc

khe

ldby

the

man

agem

ent

and

mem

bers

ofth

ebo

ard

ofdi

rect

ors.

Blo

ckow

ners

hip

isde

fine

das

the

perc

enta

geof

tota

lst

ock

held

byst

akeh

olde

rsha

ving

5%or

mor

eeq

uity

inth

efi

rm.

Inst

itutio

nal

owne

rshi

pm

easu

res

the

perc

enta

geof

afi

rm’s

equi

tyhe

ldby

inst

itutio

nali

nves

tors

.Tot

alas

sets

are

apr

oxy

for

firm

size

.Ret

urn

onas

sets

and

indu

stry

-adj

uste

dre

turn

onas

sets

are

prox

ies

for

oper

atin

gpe

rfor

man

ce.F

ree

cash

flow

isca

shfr

omop

erat

ing

activ

ities

neto

fca

shdi

vide

nds

and

capi

tale

xpen

ditu

res.

The

mar

kett

obo

okra

tiom

easu

res

grow

thop

port

uniti

es.A

ltman

’sZ

Scor

epr

oxie

sfo

rfi

nanc

ials

tren

gth.

The

Her

find

ahli

ndex

mea

sure

sth

ede

gree

ofdi

vers

ific

atio

n.E

xces

sva

lue

refe

rsto

Ber

ger

and

Ofe

k’s

(199

4)im

pute

dva

lue

mea

sure

ofdi

vers

ific

atio

nre

late

dva

lue

disc

ount

orpr

emiu

m.

G1

G2

G3

G4

G5

G1–

G2

G1–

G3

Inst

itutio

nalo

wne

rshi

ps54

.88

45.9

349

.65

64.2

555

.94

8.95

∗5.

23In

side

row

ners

hip

10.1

12.2

813

.55

10.7

814

.74

−2.1

8−3

.44

Blo

ckow

ners

hip

25.2

723

.46

23.6

44.1

237

.71.

821.

67B

oard

size

11.9

110

.69

10.7

19.

7110

.38

1.22

1.19

Rat

ioof

boar

din

depe

nden

ts0.

680.

70.

710.

620.

72−0

.02

−0.0

2R

atio

ofbo

ard

insi

ders

0.18

0.2

0.2

0.34

0.21

−0.0

2−0

.02

Bet

a0.

860.

930.

911.

090.

89−0

.07

−0.0

5E

xces

sva

lue

0.11

−0.0

70.

060.

260.

160.

190.

05M

arke

tto

book

ratio

3.81

3.12

3.69

3.15

3.65

0.69

0.12

Ret

urn

onas

sets

5.53

3.7

5.3

3.89

6.51

1.83

0.23

Indu

stry

-adj

uste

dR

AO

−0.1

3−1

.42

0.01

−0.9

51.

131.

28−0

.14

Free

cash

flow

−16.

82−3

9.29

31.2

5−3

5.92

2.79

22.4

7−4

8.06

Her

find

ahli

ndex

0.5

0.5

0.52

11

0−0

.02

Sale

sre

venu

e9,

672

9,55

37,

922

1,62

22,

882

119

1,75

0To

tals

hare

sou

tsta

ndin

g27

831

734

675

228

−39

−68

Tota

ldeb

t3,

857

2,99

42,

149

528

729

863

1,70

8To

tala

sset

s11

,625

10,0

347,

500

1,54

62,

840

1,59

14,

126

Altm

an’s

Zsc

ore

3.47

2.67

3.49

6.55

5.57

0.8

−0.0

1

(con

tinu

ed)


Tabl

e5

(con

tinu

ed)

Mea

nco

mpa

riso

nof

base

year

(199

5)ch

arac

teri

stic

sof

subg

roup

sba

sed

onsc

ope

stra

tegy

adop

tion

G1–

G4

G1–

G5

G2–

G3

G2–

G4

G2–

G5

G3–

G4

G3–

G5

G4–

G5

Inst

itutio

nalo

wne

rshi

ps−9

.37

−1.0

7−3

.72

−18.

32∗

−10.

02−1

4.6∗

∗−6

.29

8.3

Insi

der

owne

rshi

p−0

.68

−4.6

4−1

.26

1.5

−2.4

62.

76−1

.2−3

.96

Blo

ckow

ners

hip

−18.

85∗∗

−12.

43∗∗

−0.1

4−2

0.67

∗∗−1

4.24

∗∗−2

0.52

∗∗−1

4.1∗

∗6.

42B

oard

size

2.19

1.53

−0.0

20.

980.

321

0.34

−0.6

6R

atio

ofbo

ard

inde

pend

ents

0.06

−0.0

3−0

.01

0.08

−0.0

20.

08−0

.01

−0.0

9R

atio

ofbo

ard

insi

ders

−0.1

6∗∗∗

−0.0

30

−0.1

4∗∗∗

−0.0

1−0

.14∗

∗∗−0

.01

0.13

Bet

a−0

.23

−0.0

30.

02−0

.15

0.04

−0.1

80.

020.

19E

xces

sva

lue

−0.1

5−0

.04

−0.1

4−0

.34∗

∗−0

.23∗

−0.2

−0.0

90.

1M

arke

tto

book

ratio

0.66

0.16

−0.5

7−0

.03

−0.5

30.

540.

04−0

.5R

etur

non

asse

ts1.

64−0

.98

−1.6

−0.1

9−2

.81∗

∗1.

4−1

.21

−2.6

2In

dust

ry-a

djus

ted

RA

O0.

81−1

.27

−1.4

3−0

.47

−2.5

5∗∗

0.96

−1.1

2−2

.08

Free

cash

flow

19.1

1−1

9.61

−70.

53−3

.37

−42.

0867

.17

28.4

6−3

8.71

Her

find

ahli

ndex

−0.5

∗∗∗

−0.5

−0.0

2−0

.5∗∗

∗−0

.5∗∗

∗−0

.48∗

∗∗−0

.48∗

∗∗0

Sale

sre

venu

e8,

050∗

6,79

01,

630

7,93

0∗∗∗

6,67

0∗∗∗

6,30

0∗5,

040∗

∗−1

,260

Tota

lsha

res

outs

tand

ing

203∗

∗∗50

−28

242∗

∗89

271∗

∗11

7−1

53∗∗

Tota

ldeb

t3,

330

3128

845

2,46

6∗∗

2,26

5∗∗∗

1,62

2∗∗

1,42

0∗∗

−201

Tota

lass

ets

10,0

80∗

8,78

62,

535

8,48

9∗∗

7,19

5∗∗∗

5,95

4∗∗

4,66

0∗−1

,294

Altm

an’s

Zsc

ore

−3.0

8−2

.1∗∗

∗−0

.82∗

∗−3

.88

−2.9

∗∗∗

−3.0

6−2

.09∗

∗∗0.

98

∗∗∗ I

ndic

ates

stat

istic

alsi

gnif

ican

ceat

0.01

leve

l.∗∗

Indi

cate

sst

atis

tical

sign

ific

ance

at0.

05le

vel.

∗ Ind

icat

esst

atis

tical

sign

ific

ance

at0.

10le

vel.


of ownership structure within diversified subgroups, G1 has the highest and G2 hasthe lowest institutional ownership, with the difference being significant. This seemsto be inconsistent with the agency explanation of diversification as the group withthe highest monitoring adopts diversification and the group that refocuses G2 has thelowest institutional ownership. Similar findings are reported in the case of block own-ership. With respect to insider ownership, G3 has the highest (13.55%) and G1 hasthe lowest (10.1%) insider ownership. If diversification is value destroying then weshould see the highest insider ownership in the G2 subgroup followed by G3 and G1subgroups. However, insider ownership differences are not significant. For focusedsubgroups, the group that is diversifying (G4) has lower insider ownership comparedto the one that opts for remaining focused. However, the difference is insignificant.The interesting results pertain to the subgroup comparisons across diversified andfocused groups. G2, the diversified firm subgroup that focuses, has significantlylower block (23.46%) and institutional (45.93%) ownership than G4, the focusedsubgroup that diversifies, with block ownership at 44.12% and institutional owner-ship at 64.25%. These differences are significant and suggest that higher monitoringby large investors does not, on average, lead to firms adopting a focusing strategy.That is, large investors per se do not discourage diversification. The difference ininsider ownership across these two subgroups is insignificant. Overall, the resultsof our analysis of the relation between ownership structure and change in corporatediversification levels do not point to diversification being agency driven.

5.2.3. Financial performance and growth

The groups that change strategy from focused to diversification and from di-versification to focus are the poorest-performing subgroups in terms of ROA andindustry-adjusted ROA, while diversified and focused firms opting for the status quoare the best performing. We conjecture that the strategic shift is aimed at breaking outof poor performance. While focused firms remaining focused (G5) are best perform-ing, initially diversified firms choosing to refocus are the poorest performing, thedifference being statistically significant. The differences are, however, insignificantacross the rest of the five groups. The evidence suggests a link between financialperformance and strategic change irrespective of ownership and governance mech-anisms. For G4, performance improving strategic reorientation takes place despitelower representation of independent outsiders on the boards.

In contrast to focused firms that remain focused (G5), focused firms that diversify(G4) are smaller and perhaps are still at an earlier stage of corporate evolution. Giventhat size and age correlate positively, and younger firms are likely to suffer feweragency conflicts, adoption of diversification by focused firms in G4 could be drivenmore by evolution and growth and less by agency issues.

Across initially diversified and focused groups, we see that the poorest perform-ing groups, G2 and G4, change their strategies irrespective of significant differencesin block and institutional ownership and board composition. However, comparison


of G1 and G2 reveals that the latter group reverses poor-performing diversificationand the former continues with well-performing further diversification despite similarboard and ownership characteristics. Thus, strategy adoption is more related to initialperformance and the need to improve it rather than agency characteristics.

Finally, comparing G3 and G5 reveals that well-performing firms will not changetheir strategy despite differences in ownership and governance structure. Thus, changein strategy or maintaining the status quo relates to initial performance and may notbe influenced significantly by agency variables.

5.3. Changes in governance and ownership structure

Our argument that the adoption of diversification strategies and their perfor-mance implications are not agency related, and that ownership and board characteris-tics are firm structural consequences, is complete once we show that initial governanceand ownership configurations of the diversifying (initially focused) firms evolve tobecome similar to those of diversified firms. For this purpose we compare two sub-groups representing two extreme ends of corporate diversification strategies, namely,those that are diversified initially but refocus (G2) and those that are initially focusedand diversify (G4). In particular, for diversifying firms we expect an increase in boardsize, ratio of outsiders, and a decline in the ratio of insiders on the board. Similarly,growth and diversification financed by external debt or fresh equity issuance wouldlead to a decline in insider and block ownership.

The results in Table 6 partially support this conjecture. For G4 (diversifying)firms, the board size increases. As the boards of diversifying firms expand, moreoutsiders are added to the boards leading to a significant decline in the ratio ofinsiders. Although the ratio of independent outsiders increases for G4 subgroup,the change is statistically insignificant. Although consistent with our argument thatboard structures can change with changes in corporate span, the results other thanthose related to the ratio of insiders on the boards are not statistically significant.

In terms of ownership structure, as expected, although diversifying firms experi-ence a decline in block and insider ownership, the results are insignificant. However,for diversifying firms that were initially diversified, the decline in insider ownershipis significant, whereas focusing firms see a rise in both. A possible explanation forthe decline in insider and block ownership in the case of diversifying firms couldlie in their financing their expansion through external equity issuance in the marketand replacing initial funding provided through managerial or venture capitalist funds.An interesting observation relates to the increase in institutional ownership for theinitially diversified subgroups irrespective of change in strategy. Thus, it appearsthat institutional investors do not ignore firms following a diversification strategyper se. Non-diversification strategy factors, especially expected performance, couldexplain why institutions increase their ownership in these firms. We have alreadynoted similar evidence to this effect earlier in Table 4.


Table 6

Changes in governance, ownership, and financial characteristics of firms

This table contains means comparison tests of financial and structural changes over the scope shift window(1995–1997) for the five subgroups of sample firms based on diversification levels and changes therein. Thetest involves computing the difference in the values of pre-diversification and post-diversification variablesfor each case and testing if the average difference deviates significantly from zero. Firms operating in asingle business segment are focused firms and those involved in multisegment businesses are diversifiedfirms. Firms increasing (decreasing) number of business segments over the two-year period between 1995and 1997 are categorized as diversifying (focusing) firms. Business segments are defined in terms of 4-digitSIC codes. Diversified firms that increase their number of segments are placed in group G1, diversifiedfirms that shift to operate as single segment are placed in group G2, diversified firms that report no changein number of segments fall in group G3, single segment firms that increase their number of segments areplaced in group G4, and single segment firms remaining as such over the window are grouped into G5.Independent outsiders are the members of the board that are neither managers nor linked to the firm throughany business or family relationship. Insiders on the board refer to company executives serving on the boardof directors. The proportion of inside ownership is defined as the percentage of equity stock held by themanagement and members of the board of directors. Block ownership is defined as the percentage of totalstock held by stakeholders having 5% or more equity in the firm. Institutional ownership measures thepercentage of a firm’s equity held by institutional investors. Total assets are a proxy for firm size. Returnon assets and industry-adjusted return on assets are proxies for operating performance. Free cash flow iscash from operating activities net of cash dividends and capital expenditures. The market to book ratiomeasures growth opportunities. Altman’s Z Score proxies for financial strength. The Herfindahl indexmeasures the degree of diversification. Excess value refers to Berger and Ofek’s (1994) imputed valuemeasure of diversification related value discount or premium.

Variable change G1 G2 G3 G4 G5

Industry-adjusted ROA −0.22 1.10 0.48 −1.3 −1.19(0.76) (0.41) (0.63) (0.42) (0.31)

Excess value 0.14 0.15 0.11 −0.35 0.11(0.26) (0.20) (0.25) (0.13) (0.23)

One-year return −18.86 6.81 8.46 52.48∗ 1.34(0.51) (0.30) (0.23) (0.10) (0.90)

Asset turnover ratio −0.02 −0.10∗∗ 0.02 0.23 −0.01(0.74) (0.03) (0.54) (0.44) (0.81)

Insider ownership −7.62∗ −5.70 −7.03 −4.43 −4.83∗∗(0.06) (0.15) (0.17) (0.48) (0.05)

Institutional ownership 7.17∗∗ 8.19∗∗ 1.09 −0.91 3.17(0.01) (0.01) (0.68) (0.91) (0.35)

Block ownership 1.53 5.69 4.18 −7.23 −2.39(0.72) (0.13) (0.21) (0.51) (0.37)

Board size −0.62 0.08 −0.15 0.17 0.57(0.20) (0.84) (0.62) (0.89) (0.19)

Ratio of board independents 0.03 0.02 0.04∗∗ 0.02 −0.01(0.19) (0.25) (0.03) (0.22) (0.54)

Ratio of insiders on the board 0.02 0.00 0.00 −0.07∗∗ −0.01(0.13) (0.81) (0.72) (0.03) (0.66)

Debt to assets ratio −0.02 0.02 −0.03 −0.03 −0.06∗∗∗(0.60) (0.38) (0.18) (0.56) (0.00)

Free cash flow −403 −8 105 22 −59(0.29) (0.92) (0.31) (0.69) (0.61)

(continued )


Table 6 (continued)

Changes in governance, ownership, and financial characteristics of firms

Variable change G1 G2 G3 G4 G5

Market-to-book ratio 0.95 −0.38 0.55 0.26 0.86(0.57) (0.64) (0.69) (0.66) (0.13)

Sales revenue 2408∗ −982 3655∗ 1252∗∗ 1184∗∗∗(0.07) (0.25) (0.10) (0.02) (0.00)

Total assets 3870∗∗∗ −734 4254∗ 1479∗∗∗ 1585∗∗∗(0.00) (0.57) (0.10) (0.01) (0.00)

Herfindahl index −0.06∗∗∗ 0.17∗∗∗ 0.01 −0.35∗∗∗ 0.00(0.00) (0.00) (0.17) (0.00) (1.00)


To directly investigate the performance implication of diversification strategy,we test the proposition that diversification, on average, leads to performance dis-counting and that performance consequences of diversification strategy can be linkedto differences in ownership and governance structure.

5.4. Diversification and performance

The analysis up to this point has provided evidence that adoption of a particularcorporate scope strategy and its performance implications may not be unambiguouslyagency driven. However, direct evidence on the link between ownership and gover-nance characteristics on the one hand and the performance outcome of diversificationstrategy on the other remains to be provided.

We make two points in this regard. First, diversification strategy per se may notbe detrimental to performance, as popularly conceived. Second, the negative perfor-mance consequences of some diversification moves cannot necessarily be attributableto agency conflicts. With respect to the first point regarding diversification being anegative performance proposition, we present our binomial test results in Table 7.The figures in Panel A show that for none of the three measures (i.e., one-year andthree-year excess value and industry-adjusted ROA) is there a significantly highernumber of losing firms than gaining firms from diversification. These results supportour conjecture that diversification is not necessarily value destroying.

In Panel B of Table 7, we present the results related to our second point,namely, that diversification losses is not necessarily attributable to agency conflicts.If agency is an explanation for negative performance-generating diversification, thegroup of firms with low inside ownership should have more performance losers thangainers, and high insider ownership firms should have more gainers than losers.Thus, the group of firms with below (above) median inside ownership should haveproportionally more (less) losers than gainers as suggested by the convergence ofinterest hypothesis. The differences, in the number of losing as well as gaining firms


Table 7

Performance implications of diversification strategy

This table contains results of binomial tests for comparing positive (i.e., increased industry-adjusted returnon assets and excess value: gainers) and negative (i.e., decreased industry-adjusted return on assets andexcess value: losers) consequences of diversification strategy. Panel A reports the number of diversifyingfirms that experienced improved or deteriorated performance over the one-year post and a longer three-yearperiod (one-year post + two-year contemporaneous to shift window) diversification windows. Panel Breports gainer and loser diversifying firms conditional on high versus low inside ownership. The proportionof inside ownership is defined as percentage of equity stock held by the management and members of theboard of directors. Panel C reports gainer and loser diversifying firms conditional on inside dominanceof boards. Insiders on the board refer to company executives serving on the board of directors. Industry-adjusted return on assets is a proxy for operating performance. Excess value refers to Berger and Ofek’s(1994) imputed value measure of diversification related value discount or premium.

Panel A: Binomial test results: Overall performance implications of diversification strategy

Observed Test AsymptoticMeasurement Category Number proportion proportion significance

One-year post excess value Gainers 17 0.50 0.500 1.000Losers 17 0.50Total 34 1.000

1995–1998 excess value Gainers 17 0.55 0.500 0.719Losers 14 0.45Total 31 1.000

1995–1998-adjusted ROA Gainers 29 0.53 0.500 0.787Losers 26 0.47Total 55 1.00

Panel B: Binomial test results: Performance implications of diversification conditionalon ownership structure

Inside Observed Test Asymptoticownership Measurement Category Number proportion proportion significance

Below median One-year post-excess Gainers 7 0.500 0.500 1.000value Losers 7 0.500

Total 14 1.0001995–1998-excess Gainers 6 0.429 0.500 0.791

value Losers 8 0.571Total 14 1.000

1995–1998-adjusted Gainers 11 0.478 0.500 1.000ROA Losers 12 0.522

Total 23 1.000Above median One-year post-excess Gainers 6 0.462 0.500 1.000


1995–1998 excess Gainers 8 0.666 0.500 0.388value Losers 4 0.334

Total 12 1.0001995–1998-adjusted Gainers 11 0.579 0.500 0.648

ROA Losers 8 0.421Total 19 1.000

(continued )


Table 7 (continued)

Performance implications of diversification strategy

Panel C: Binomial test results: Performance implications of diversification conditionalupon board composition

Ratio of insiders Observed Test Asymptoticon the board Measurement Category Number proportion proportion significance

Below median One-year post-excess Gainers 6 0.545 0.500 1.000value Losers 5 0.455

Total 11 1.0001995–1998 excess Gainers 5 0.500 0.500 1.000


1995–1998-adjusted Gainers 11 0.647 0.500 0.332ROA Losers 6 0.353

Total 17 1.000Above median One-year post-excess Gainers 5 0.462 0.500 0.727


1995–1998 excess Gainers 5 0.625 0.500 0.727value Losers 3 0.375

Total 8 1.0001995–1998-adjusted Gainers 7 0.583 0.500 0.774

ROA Losers 5 0.417Total 12 1.000

for both above and below median inside ownership, however, are statistically insignif-icant for both performance change measurement windows. Thus, ownership structuredoes not seem to be related to the performance impact of diversification strategy.

Similarly, results in Panel C indicate that governance characteristics in terms ofboard composition differentials do not indicate any statistically significant differencesbetween proportions of gainers and losers in firms with low (below median) andhigh (above median) ratio of insiders on the board. Thus, managerial control of theboard does not influence the diversification strategy outcome. Overall, we are notable to conclude that agency conflict provides an unambiguous explanation for theadoption and performance implications of a diversification strategy. We find thatcorporate structural, performance, and evolutionary stage differentials are able toexplain linkages between diversification levels and changes therein on the one handand governance and ownership structure on the other.

5.5. Ownership and governance as determinants of diversification

We test if there is a direct relation between degree of focus as measured by theHerfindahl index and corporate ownership and governance structure while controllingfor other firm specific characteristics. Table 8 presents the results for five multipleregressions with the Herfindahl index as the dependent variable. Smaller, more cashconstrained firms and firms with lower leverage have a greater degree of focus.


Table 8

Multiple regression models analyzing ownership and governance characteristics as determinants ofHerfindahl’s index

The dependent variable is the Herfindahl index. Independent outsiders are the members of the board thatare neither managers nor linked to the firm through any business or family relationship. Insiders on theboard refer to company executives serving on the board of directors. The proportion of inside ownership isdefined as the percentage of equity stock held by the management and members of the board of directors.Free cash flow is cash from operating activities net of cash dividends and capital expenditures. Blockownership is defined as the percentage of total stock held by stakeholders having 5% or more equity inthe firm. Institutional ownership measures the percentage of a firm’s equity held by institutional investors.Leverage is defined as the ratio of total debt to total assets. Log of total assets is a proxy for firm size.Industry-adjusted return on assets is a proxy for operating performance. The market to book ratio isintroduced as a control variable to isolate the influence of growth opportunities. Excess value refers toBerger and Ofek’s (1994) imputed value measure of diversification related value discount or premium.


Constant 1.78∗∗∗ 1.42∗∗∗ 1.42∗∗∗ 1.26∗∗∗ 1.00∗∗∗(0.00) (0.00) (0.00) (0.00) (0.00)

Free cash flow −.0001∗∗∗ −0.0001∗ −0.0006 −0.0001∗ −0.0001∗∗(0.01) (0.10) (0.11) (0.07) (0.04)

Excess value 0.13∗∗∗ 0.13∗∗∗ 0.13∗∗∗ 0.14∗∗∗ 0.13∗∗∗(0.00) (0.00) (0.00) (0.00) (0.01)

Leverage −0.18 −0.17 −0.22∗ −0.22 −0.31∗∗(0.15) (0.16) (0.07) (0.15) (0.04)

Firm size −0.12∗∗∗ −0.09∗∗∗ −0.08∗∗∗ −0.05∗∗∗ −0.04∗∗∗(0.00) (0.00) (0.00) (0.03) (0.08)

Industry-adjusted ROA −0.009∗∗ −0.009∗∗ −0.008∗ −0.003 −0.005(0.05) (0.04) (0.06) (0.54) (0.34)

Market-to-book ratio −0.01 −0.01∗ −0.01 −0.009 −0.008(0.17) (0.10) (0.15) (0.36) (0.39)

Insider ownership −0.03∗∗(0.01)

Institutional ownership 0.001∗∗(0.05)

Block ownership 0.001(0.25)

Board independents’ ratio −0.11(0.59)

Board insiders’ ratio 0.55∗∗(0.03)

R2 0.25 0.23 0.21 0.10 0.14F-probability 0.00 0.00 0.00 0.01 0.00


Second, in all five models, excess value is significantly positively related to theHerfindahl index, suggesting that excess value correlates with focus.

Controlling for size, leverage, free cash flow, and performance, we find thathigher levels of insider ownership result in firms being more diversified, a result


that is inconsistent with the agency argument for diversification. As in the bivariateresults, higher institutional ownership is related to more focused firms. One can arguethat high institutional ownership, due to higher monitoring, is the reason behindfirms being focused. However, the argument assumes that diversification is valuedestroying, which as we have seen may not be a valid assumption. For another,our previous evidence suggests that even for initially diversified firms that diversifyfurther, institutional ownership increases. Further, the evidence is also consistentwith the argument that focused firms are on average smaller, and at an earlier stageof corporate evolutionary process, and hence have more institutional financing. Interms of board composition, firms with a higher degree of focus have a higher ratio ofinsiders on the board. This also is more consistent with the explanation that focusedfirms are smaller and at an earlier evolutionary stage than the agency interpretationof diversification.

5.6. Nonlinear relation between diversification and insider ownership

Although in our bivariate results we do not find a significant difference ininsider ownership across diversified and focused firms, with the multivariate testwe find a negative significant relation between insider ownership and degree offocus. It is possible that the relation between insider ownership and the degree offocus is conditional on the level of insider ownership. We investigate the issue inMorck, Schleifer, and Vishny (1990) framework by subgrouping the sample intothree categories of inside ownership, namely, firms with insider ownership between0% and 5%, those between 5% and 25%, and those having insider ownership greaterthan 25%.

Our results (not reported for brevity) suggest that a nonlinear relation existsbetween insider ownership and the Herfindahl index. Insider ownership, other thanin the 5 to 25% range, is negatively related to the Herfindahl index. However, in the5 to 25% range, insider ownership is positively related to the Herfindahl index. InMorck, Schleifer, and Vishny’s (1990) framework, in the midrange in which privatebenefits of suboptimal managerial choices are greater than the private costs, there isa negative relation between insider ownership and performance. If diversification isagency-induced we should expect to see a negative relation between degree of focusand insider ownership. In our case in the midrange—in which, according to Morck,Schleifer, and Vishny, agency conflict dominates—there actually is a positive relationbetween the degree of focus and insider ownership. Thus, if we agree with Morck,Schleifer, and Vishny’s position, our results do not support the agency argument ofdiversification.

5.7. Link between corporate evolutionary stage and ownershipand governance structure

To analyze more closely the link between corporate evolutionary stage and corpo-rate ownership and governance structure, in Table 9 we report the correlations among


ble

9

Cor

rela

tion

amon

ggo

vern

ance

,ow

ners

hip

and

evol

utio

nary

stag

ech

arac

teri

stic

s(y

ear

1995

)

Thi

sta

ble

cont

ains

corr

elat

ion

coef

fici

ents

for

gove

rnan

ce,

owne

rshi

p,an

dev

olut

iona

ryst

age

char

acte

rist

ics

ofth

esa

mpl

efi

rms.

Age

equa

lsye

ars

sinc

ein

corp

orat

ion

ofa

firm

.In

depe

nden

tou

tsid

ers

are

the

mem

bers

ofth

ebo

ard

that

are

neith

erm

anag

ers

nor

linke

dto

the

firm

thro

ugh

any

busi

ness

orfa

mily

rela

tions

hip.

Insi

ders

onth

ebo

ard

refe

rto

com

pany

exec

utiv

esse

rvin

gon

the

boar

dof

dire

ctor

s.T

hepr

opor

tion

ofin

side

owne

rshi

pis

defi

ned

asth

epe

rcen

tage

ofeq

uity

stoc

khe

ldby

the

man

agem

enta

ndm

embe

rsof

the

boar

dof

dire

ctor

s.B

lock

owne

rshi

pis

defi

ned

asth

epe

rcen

tage

ofto

tals

tock

held

byst

akeh

olde

rsha

ving

5%or

mor

eeq

uity

inth

efi

rm.I

nstit

utio

nalo

wne

rshi

pm

easu

res

the

perc

enta

geof

afi

rm’s

equi

tyhe

ldby

inst

itutio

nali

nves

tors

.Tot

alas

sets

are

apr

oxy

forf

irm

size

.The

mar

kett

obo

okra

tiom

easu

res

grow

thop

port

uniti

es.T

heH

erfi

ndah

lind

exm

easu

res

the

degr

eeof

dive

rsif

icat

ion.

Exc

ess

valu

ere

fers

toB

erge

ran

dO

fek’

sim

pute

dva

lue

mea

sure

ofdi

vers

ific

atio

nre

late

dva

lue

disc

ount

orpr

emiu

m.

SG&

AH

erfi

ndah

lIn

stitu

tiona

lIn

side

Blo

ckR

atio

ofR

atio

ofB

oard

Exc

ess

Ass

etex

pens

eM

arke

tto

inde

xow

ners

hip

owne

rshi

pow

ners

hip

insi

ders

inde

p.si

zeva

lue

turn

over

ratio

book

Age

Her

find

ahli

ndex

1.00

0.16

−0.0

50.

26∗∗

0.22

∗−0

.07

−0.1

00.

140.

130.

23∗

0.02

−0.2

4∗∗

Inst

itutio

nal

1.00

−0.3

2∗∗

−0.0

2−0

.22∗

0.30

∗∗0.

070.

05−0

.01

0.08

0.14

0.05

owne

rshi

pIn

side

owne

rshi

p1.

000.

42∗∗

0.31

∗∗−0

.25∗

−0.1

2−0

.06

0.11

−0.0

50.

11−0

.15

Blo

ckow

ners

hip

1.00

0.35

∗∗−0

.34∗

∗−0

.17

0.01

0.05

−0.0

7−0

.09

−0.2

0∗R

atio

ofin

side

rs1.

00−0

.59∗

∗−0

.28∗

∗0.

070.

080.

00−0

.13

−0.1

2R

atio

of1.

000.

160.

030.

040.

24∗

0.28

∗∗0.

08in

depe

nden

tsB

oard

size

1.00

−0.0

8−0

.04

0.02

−0.1

20.

26∗

Exc

ess

valu

e1.

00−0

.28∗

∗0.

30∗∗

0.28

∗∗−0

.03

Ass

ettu

rnov

er1.

00−0

.17∗

−0.1

00.

05SG

&A

expe

nse

1.00

0.26

∗∗−0

.04

ratio

Mar

kett

obo

ok1.

00−0

.17∗

Age

1.00

∗∗In

dica

tes

stat

istic

alsi

gnif

ican

ceat

0.05

leve

l.∗ I

ndic

ates

stat

istic

alsi

gnif

ican

ceat

0.10

leve

l.


our proxies for the evolutionary stage variables (i.e., age, diversity, and growth oppor-tunities) and governance and ownership characteristics. Corporate age is positivelyrelated to board size and the ratio of independent outsiders on the board and neg-atively related to the ratio of insiders on the board. However, the only significantrelation with age is that of board size. In addition, age has positive correlation withthe degree of diversification and has a negative relation with growth opportunities, asexpected. The evidence is consistent with our earlier argument that older, large, morediversified, and slow-growing firms have relatively larger boards, higher independentoutsiders, and lower insider representation on boards. Firms include only a limitednumber of key executives on boards irrespective of firm maturity and size of theboard of directors, thereby yielding an insignificant correlation between corporateage and insiders on the board. Thus, structural and evolutionary stage differentialscan explain, at least partially, the internal governance structure differences betweendiversified and focused firms.

In terms of the relation between the ownership structure and evolutionary stagecharacteristics, firms that are at a higher stage of corporate evolution (i.e., numberof years since incorporation), slower growth, and greater diversity have lower insiderand block ownership. The evidence supports our argument that initially firms rely ontheir insiders’ equity supported by outside block owners, including venture capitalistequity. However, as firms grow and diversify, they rely more on open market equityissuance, thereby reducing the ratio of inside and block ownership. Although thecorrelation between age and block ownership in significant, that between insider andinstitutional ownership and age is insignificant.

6. Conclusions

Recent research (e.g., Berger and Ofek, 1994; Servaes, 1996; Denis, Denis, andSarin, 1997 ) argues that diversified firms sell at a discount when compared to theirfocused counterparts. The literature typically attributes the diversification discountto the presence of agency conflicts between managers and shareholders. Despite thewell-articulated agency argument, the discount runs counter to the argument that acompetitive marketplace would not tolerate the existence of inefficient firms. Morerecent evidence suggests that the reported diversification discount could, in fact, be anartifact of data selection and empirical methods. In general, our results complementthe findings by Anderson et al. (2000) in that differences in corporate governance donot explain the decision to diversify or focus and that the performance implicationsof adopting any particular corporate scope strategy are not explained by variations ingovernance across firms.

We contribute to the debate by examining a variety of issues related to theownership and governance structures of diversified and focused firms. We report thatdiversification is not predominantly a performance-reducing strategy. In this senseour results are consistent with those of Villalonga (2004a, 2004b) that there is nodiversification discount. Further, we provide evidence that the diversification discountis not attributable to agency conflicts. With regard to the ownership structure, our


evidence suggests that firms diversify in the presence of effective external monitoring.The result cannot be attributed to the presence of agency issues. Extending Andersonet al. (2000), we provide evidence that the observed differences in governance andownership structure between focused and diversified firms can be attributable totheir being in different stages of corporate evolution. More important, our evidencesuggests that firms change diversification strategies, either diversifying or focusing,based on managerial perceptions of strategies that can generate gains for stockholders.Thus, poorly performing firms, irrespective of their prior degree of diversificationand adoption of diversifying or focusing strategies, potentially can generate gainswith properly executed strategies. Our results add weight to the argument that thereported diversification discount stems from underlying firm characteristics ratherthan agency conflicts. Finally, the results suggest that governance and ownershipstructures evolve in response to the performance impact of diversification and focusstrategies, rather than influencing these strategies.

Appendix

The following is a brief summary of the sample selection process:

Criterion Sample size

Step 1: All NYSE/Amex/Nasdaq firms with sales = $1 billion 1113Step 2: Elimination of 180 American depositary receipts 933Step 3: Nonavailability of segment data for 156 firms 777Step 4: Firms not changing number of business segments 640

Focused firms remaining focused 413Diversified firms remaining diversified 227Firms reporting change in number of business segments 137

Step 5: Subgroup formation based on business diversification/focus

Group Initial Random selection

G-1 38 38G-2 80 40a

G-3 227 38b

G-4 19 19G-5 413 42c

Total 777 177

a Every second firm selected.b Every fifth firm selected.c Every ninth firm selected.

In summary, we have the following five subgroups, with their descriptive statis-tics summarized in Table 1:

1. Diversified firms diversifying further: G12. Diversified firms refocusing: G23. Diversified firms that do not change: G3


4. Focused firms adopting diversification: G45. Focused firms remaining focused: G5

References

Agrawal, A. and C.R. Knoeber, 1996. Firm performance and mechanisms to control agency problemsbetween managers and shareholders, Journal of Financial and Quantitative Analysis 31, 377–397.

Amihud, Y. and B. Lev, 1981. Risk reduction as a managerial motive for conglomerate mergers, BellJournal of Economics 12, 605–617.

Anderson, R.C., T.W. Bates, J. Bizjak, and M.L. Lemmon, 2000. Corporate governance and firm diversi-fication, Financial Management 21, 5–22.

Baysinger, B. and H. Butler, 1985. Corporate governance and the board of directors: Performance effectof changes in board composition, Journal of Law, Economics and Organization 1, 101–124.

Bengtsson, L., 2000. Corporate strategy in a small open economy: Reducing product diversification whileincreasing international diversification, European Management Journal 18, 444–453.

Berger, P.G. and E. Ofek, 1994. Diversification’s effect on firm value, Journal of Financial Economics37, 39–65.

Bhide, A., 1990. Reversing corporate diversification, Journal of Applied Corporate Finance 5, 70–81.Billet, M.T. and D.C. Mauer, 2000. Diversification and the value of internal capital markets: The case of

tracking stock, Journal of Banking and Finance 24, 1457–1490.Brickley, J.A. and C.M. James, 1987. The takeover market, corporate board composition and ownership

structure: The case of banking? Journal of Law and Economics 30, 161–180.Brickley, J.A., J.L. Coles, and R.L. Terry, 1994. Outside directors and the adoption of poison pills, Journal

of Financial Economics 35, 371–390.Byrd, J.W. and K.A. Hickman, 1982. Do outside directors monitor managers? Evidence from tender offer

bids, Journal of Financial Economics 32, 195–222.Campa, J.M. and S. Kedia, 2002. Explaining the diversification discount, Journal of Finance 57, 1731–

1762.Chen, S.S. and K.W. Ho, 2000. Corporate diversification, ownership structure, and firm value: The

Singapore evidence, International Review of Financial Analysis 9, 315–326.Cochran, P.L., R.A. Wood, and T.B. Jones, 1985. The composition of board of directors and incidence of

golden parachutes, Academy of Management Journal 28, 664–671.Comment, R. and G.A. Jarrell, 1994. Corporate focus and stock returns, Journal of Financial Economics

37, 67–87.Coughlan, A.T. and R.M. Schmidt, 1985. Executive compensation and management turnover, and firm

performance: An empirical investigation, Journal of Accounting and Economics 7, 43–66.Demsetz, H., 1983. The structure of ownership and the theory of the firm, Journal of Law and Economics

26, 375–390.Denis, D.J., D.K. Denis, and A. Sarin, 1997. Agency problems, equity ownership, and corporate diversi-

fication, Journal of Finance 52, 135–160.Denis, D.J., D.K. Denis, and K. Yost, 2002. Global diversification, industrial diversification and firm

value, Journal of Finance 57, 1951–1979.Eckbo, B.E., 1985. Mergers and the market concentration doctrine: Evidence from the capital market,

Journal of Business 58, 325–349.Fama, E., 1980. Agency problems and the theory of the firm, Journal of Political Economy 88, 288–307.Fama, E. and M. Jensen, 1983. Agency problems and residual claims, Journal of Law and Economics 26,

327–350.Fauver, L., J. Houston, and A. Naranjo, 2003. Capital market development, integration, legal systems, and

the value of corporate diversification: A cross-country analysis, Journal of Financial and QuantitativeAnalysis 38, 135–157.


Fauver, L., J.F. Houston, and A. Naranjo, in press. Cross-country evidence on the value of corporateindustrial and international diversification, Journal of Corporate Finance 10, 729–752.

Fee, C.E. and S. Thomas, 2004. Corporate diversification, asymmetric information, and firm value: Evi-dence from stock market trading characteristics, Journal of Corporate Finance 10, 105–129.

Fields, M.A. and P.Y. Keys, 2003. The emergence of corporate governance from Wall St. to Main St.:Outside directors, board diversity, earnings management, and managerial incentives to bear risk, TheFinancial Review 38, 1–24.

Gibbs, P., 1993. Determinants of corporate restructuring: Relative importance of corporate governance,takeover threat, and free cash flow, Strategic Management Journal 14, 51–68.

Gillan, S.L., J.W. Kensinger, and J.D. Martin, 2000. Value creation and corporate diversification: The caseof Sears, Roebuck & Co, Journal of Financial Economics 55, 103–137.

Gladstein, D.G., 1984. Groups in context: A model of task group effectiveness, Administrative ScienceQuarterly 29, 499–517.

Goodstein, J., K. Gautam, and W. Boeker, 1994. The effects of board size and diversity on strategic change,Strategic Management Journal 15, 241–250.

Gort, M., H. Grabowski, and R.H. McGuckin, 1985. Organizational capital and the choice between spe-cialization and diversification, Managerial and Decision Economics 6, 2–9.

Graham, J.R. and D.A. Rogers, 2002. Do firms hedge in response to tax incentives? Journal of Finance57, 815–839.

Graham, J.R., M.L. Lemmon, and J. Wolf, 2002. Does corporate diversification destroy value? Journal ofFinance 57, 695–720.

Harris, M. and A. Raviv, 1990. Capital structure and the informational role of debt, Journal of Finance 45,321–350.

Harrison, J.R., 1987. The strategic use of corporate board committees, California Management Review 30,109–125.

Hart, O.D., 1983. The market mechanism as an incentive scheme, Bell Journal of Economics 14, 42–64.Hermaelin, B.E. and M.S. Weisbach, 1988. The determinants of board composition, Rand Journal of

Economics 19, 589–606.Hill, C.W.L. and R.E. Hoskisson, 1987. Strategy and structure in the multiproduct firm, Academy of

Management Review 12, 331–341.Jensen, M. and K. Murphy, 1991. Performance pay and top-management incentives, Journal of Political

Economy 98, 225–264.Jensen, M.C., 1986. Agency costs of free cash flow, corporate finance and takeover, American Economic

Review 76, 323–329.Jewell, L.N. and H.J. Reitz, 1981. Group effectiveness in organizations (Scott-Foresman, Glenview,

Illinois).John, K. and E. Ofek, 1994. Asset sales and increase in focus, Journal of Financial Economics 37, 105–

126.Johnson, R.A., R.E. Hoskisson, and M.A. Hitt, 1993. Board of director involvement in restructuring: The

effects of board versus managerial controls and characteristics, Strategic Management Journal 14(special summer issue), 33–50.

Kaplan, S.N. and M.S. Weisbach, 1992. The success of acquisitions: Evidence from divestitures, Journalof Finance 47, 107–138.

Kosnik, R., 1990. Effects of board demography and directors? Incentives on corporate greenmail decisions,Academy of Management Journal 33, 129–150.

Lang, L.H.P. and R.M. Stulz, 1994. Tobin’s Q, corporate diversification, and firm performance, Journalof Political Economy 102, 1248–1280.

Lee, C., S. Rosenstein, N. Rangan, and W.N. Davidson, 1992. Board composition and shareholder wealth:The case of management buyouts, Financial Management 21, 58–72.

Lewellen, W.G., 1971. A pure financial rationale for the conglomerate merger, Journal of Finance 26,521–537.


Liebeskind, J. and T.C. Opler, 1993. Corporate Diversification and Agency Cost: Evidence from PrivatelyHeld Firms. Working paper, Ohio State University, Columbus, OH.

Lins, K. and H. Servaes, 1999. International evidence on the value of corporate diversification, Journal ofFinance 54, 2215–2239.

Lins, K. and H. Servaes, 2002. Is corporate diversification beneficial in emerging markets? FinancialManagement 31, 5–31.

Majd, S. and S.C. Myers, 1987. Tax asymmetries and corporate income tax reforms, in: Martin Feldsteined., Effects of Taxation on Capital Accumulation (University of Chicago Press, Chicago, Illinois).

Markides, C.C., 1995. Diversification, restructuring and economic performance, Strategic ManagementJournal 16, 101–118.

Mintzberg, H., 1983. Power in and Around Organizations (Prentice-Hall, Englewood Cliffs, New Jersey).Morck, R., A. Schleifer, and R.W. Vishny, 1990. Do managerial objectives drive bad acquisitions? Journal

of Finance 45, 31–48.Myers, S.C., 1977. The determinants of corporate borrowing, Journal of Financial Economics 5, 147–175.Olson, M., 1982. The Rise and Decline of Nations: Economic Growth, Stagflation and Social Rigidities

(Yale University Press, New Haven, Connecticut).Pearce, J. and S. Zahra, 1992. Board composition from a strategic contingency perspective, Journal of

Management Studies 29, 411–438.Pfeffer, J., 1993. Size composition and function of hospital boards of directors: A study of organization-

environment linkage, Administrative Science Quarterly 18, 349–364.Ravenscraft, D.J. and F.M. Scherer, 1987. Mergers Sell-Offs and Economic Efficiency (Brookings Insti-

tution, Washington, DC).Rosenstein, S. and J.G. Wyatt, 1990. Outside directors, board independence, and shareholder wealth,

Journal of Financial Economics 26, 175–191.Servaes, H., 1996. The value of diversification during the conglomerate merger wave, Journal of Finance

51, 1201–1225.Shaw, M.E., 1981. Group Dynamics: The Psychology of Small Group Behavior (McGraw-Hill, New York).Shleifer, A. and R. Vishny, 1989. Managerial entrenchment: The case of manager specific investments,

Journal of Financial Economics 25, 123–139.Shleifer, A. and R. Vishny, 1991. The takeover wave of the 1980s, Journal of Applied Corporate Finance

4, 49–56.Sicherman, N.W. and R.H. Pettway, 1987. Acquisition of divested assets and shareholder wealth, Journal

of Finance 42, 1261–1271.Singh, H. and F. Harianto, 1989. Management board relationship, takeover risk and the adoption of golden

parachutes, Academy of Management Journal 32, 7–24.Stigler, G.J., 1966. The economic effects of the antitrust laws, Journal of Law and Economics 9, 225–258.Stulz, R., 1990. Managerial discretion and optimal financing policies, Journal of Financial Economics 26,

3–28.Villalonga, B., 2004a. Diversification discount or premium? New evidence from BITS establishment level

data, Journal of Finance 59, 475–502.Villalonga, B., 2004b. Does diversification cause the “diversification discount”? Working paper, Harvard

University, Boston, MA.Weisbach, M., 1988. Outside directors and CEO turnover, Journal of Financial Economics 20, 431–460.Wernerfelt, B. and C.A. Montgomery, 1988. Tobin’s Q and the importance of focus in firm performance,

American Economic Review 78, 246–250.Whited, T., 2001. Is it efficient investment that causes the diversification discount? Journal of Finance 56,

1667–1692.Williamson, O.E., 1986. Economic Organization: Firms, Markets and Policy Control (New York University

Press, New York).Yermack, D., 1996. Higher market value of companies with a small board of directors, Journal of Financial

Economics 40, 185–211.

Governance and Performance Implications of Diversification Strategies: Evidence from Large U.S. Firms

Documents