® Academy of Management Journal 2003, Vol. 46, No. 1, 13-26. META-ANALYSES OF FINANCIAL PERFORMANCE AND EQUITY: FUSION OR CONFUSION? DAN R. DALTON CATHERINE M. DAILY Indiana University S. TREVIS CERTO Texas A&M University RUNGPEN ROENGPITYA Indiana University Agency theory dominates research on equity holdings-firm performance relationships; however, extant studies provide no consensus about the direction and magnitude of such relationships. Consistent linkages have not been demonstrated for firm performance and CEO, officer, director, institutional, or blockholder equity. We conducted a series of meta-analyses of relevant empirical ownership-performance studies. The meta-analyses provide few examples of systematic relationships, lending little support for agency theory. We propose a substitution theory perspective for future ownershipperformance research. bility that the owners and managers of a firm might have divergent, or misaligned, interests. In fact, a central principle of agency theory is that high-ranking corporate officers, acting as the agents of shareholders, can pursue covnses of action inconsistent with the interests of owners (e.g., Eisenhardt, 1989; Jensen & Meckling, 1976; Shleifer & Vishny, 1997). This potential confiict of interests has become a central focus of corporate governance. Macey explained that "corporate governance can be described as the processes by which investors attempt to minimize the transactions costs (Coase, 1937) and agency costs (Jensen & Meckling, 1976) associated with doing business within a firm" (1997: 602). The ability of equity holdings to address the agency problem and enhance firm financial performance was the focus of our study. Extant literature refiects two common themes with regard to mitigating agency costs. The first, which we will refer to as "aligrmient," was nicely captured by Himmelberg, Hubbard, and Palia, who wrote that "it is well known" (1999: 354) that a potential solution to the fundamental agency problem is to provide managers with equity stakes in their firms. Thus, managerial self-interest may be mitigated by aligning the interests of managers and shareholders, and it is presumed firm performance will improve as managers concurrently work for their own and shareholders' benefit (e.g., Jensen & Murphy, 1990; Perry & Zenner, 2000). The second view, illustrated by Agrawal and Knoeber's (1996) work, we will refer to as the "control" approach (also see Bethel, Liebeskind, and Opler Adam Smith, in An Inquiry into the Nature and Causes ofthe Wealth of Nations, provided one of the earliest discussions of the problem of the separation of ownership and control. He suggested that managers of other people's money cannot be expected to "watch over it with the same anxious vigilance" one would expect from owners and that "negligence and profusion, therefore, must always prevail, more or less, in the management of the affairs of such a company" (Smith, 1776/1952: 324). This sentiment presages the essence of agency theory, a theory that has been characterized as "a theory of the ownership (or capital) structure of the firm" (Jensen & Meckling, 1976: 309). Agency theory has been the dominant theme of empirical examinations of the relationship between equity ownership and financial performance (e.g., Thomsen & Pedersen, 2000). Agency theory is largely groimded in the seminal work of Berle and Means (1932), who suggested that a fundamental shift occurred in the early 1900s. At that time, professional managers with little or no equity increasingly gained day-to-day management responsibility for firms that had previously been actively managed by their owners. With this separation of roles arose the possi- The authors wish to thank Professor Frank L. Schmidt for his generous counsel regarding certain technical aspects of our meta-analysis. His was a substantial contribution to this work, and his willingness to share his time and expertise was a model of collegiality. 13 14 Academy of Management Journal [1998] and Dalton, Daily, Ellstrand, and Johnson [1998]). Here, concentrated shareholdings may facilitate the monitoring of firms' managements and lead to improved finn perfonnance. The ownership shares of two types of outside owners, institutions and "blockholders," are typically sufficiently large that these equity owners "are positioned to see to it that management serves their interests" (Demsetz & Lehn, 1985: 1161) and this "ought to yield higher profit rates" (Demsetz & Lehn, 1985: 1174). Although agency theory predicts that equity holdings have important implications for firm performance, the empirical evidence provides no consensus on the relationship between equity holdings by various constituent groups and financial performance. Bothvi^ell (1980: 304), for example, noted that the results of studies examining the relationship of equity with firm performance have been "quite mixed." Hunt concluded that "it is difficult to draw any firm conclusions . .. since no consensus has developed" (1986: 96). More recently. Short noted that "the results of these studies are inconclusive" (1994: 206). Also, Rediker and Seth observed that "there is a singular lack of consistency in the empirical results reported" (1995: 86]. The lack of consistency notwithstanding, McConnell and Servaes observed that "a consensus interpretation is that the allocation of equity ownership matters" (1995:133). In light ofthe inconsistency in the empirical findings, it was this latter observation that guided our research. Given the continuing interest in and empirical attention to equity holdings and their relationship to financial performance, and the dominance of agency theory as a theoretical foundation for these studies, we conducted a series of meta-analyses to examine if some synthesis in these relationships could be demonstrated. Our analysis is based on the equity ownership categories found in extant empirical research (e.g., McConnell & Servaes, 1990, 1995). The discussion of equity categories and their relationship to firm performance as a fiinction of the alignment or control perspective is followed by an overview of our metaanalytical procedures, the results of our analyses, and a discussion of theoretical and practical implications of our study findings. HYPOTHESES Inside Equity Holdings Agency theory provides the rationale for improved firm performance when managers' interests are aligned with those of shareholders through managerial equity holdings. Jensen and Meckling specifically identified "the fraction ofthe equity held by the manager" (1976: 343) as fundamental to ownership struc- February ture. This same rationale has been applied to board members as well. Officers and directors, in various combinations, constitute inside equity holders (e.g.. Bethel & Liebeskind, 1993). Careful review of extant empirical research shows the following inside equity categories: CEO equity holdings, managerial equity holdings, officer and director equity holdings, inside board equity holdings, and outside board equity holdings. These categories guide the following discussion. Agency theory suggests that when insiders hold substantial equity positions in the firms they serve, they are more likely to act in shareholders' interests, given their shared financial interests (Bryan, Hwang, and Lilien [2000] and Perry and Zenner [2000], among others, provide overviews). This is the alignment rationale to which we previously referred. Following this rationale, Jensen and Murphy (1990) noted that stock ownership causes executives' wealth to vary directly with company performance. Absent an equity investment in a firm, executives are more likely to behave opportunistically by supporting projects that further their own interests, at shareholders' expense, and to behave in a manner that further ensures their individual job security (Himmelberg et aL, 1999; Zahra, Neubaum, & Huse, 2000). Eisenhardt (1989) identified this propensity for self-interest as a fundamental element of agency theory. Researchers have also applied this alignment rationale to corporate board members. Many empirical studies, for example, rely on "officer and director equity" to capture insider equity ownership (Jensen, 1993]. Some board members also serve as officers in their firms (inside directors], and the remainder are norunanagement board members (outside directors]. Regardless of the relationship a director has with a firm, the board, as a decision-making body, operates within the boundaries of the corporation. On this basis, then, board members are subject to the same alignment incentives as corporate officers. Jensen (1993], for example, did not distinguish between officers and directors in his observation that agency problems are likely to occur when officers and directors do not directly share in the appreciation of their firms' equity. Substantial equity stakes, however, provide these individuals with enhanced incentives to effectively manage firm performance for the benefit of shareholders. Regardless of whether inside equity holders are CEOs, officers, or directors, the alignment perspective suggested by agency theory principles supports a positive relationship between insider equity holdings and firm performance. The central idea is that equity stakes will resolve conflicts of interest between corporate insiders and shareholders, with a resultant impact on shareholder value. 2003 Dalton, Daily, Certo, and Roengpitya Hypothesis 1. Insider equity holdings will be positively associated with firm financial peiformance. Outside Equity Holdings Agency theory is also informative with regard to the anticipated relationship between outside equity holdings and firm performance. Extant research has identified two central categories of external equity owners: institutional investors and blockholders. Here, the agency theory rationale for a relationship between these equity holders and firm performance is one of control. Institutional investors' and blockholders' equity holdings are sufficiently large to encourage these equity holders to actively monitor firms' decision makers to ensure that the firms are being operated in the best interests of shareholders. Institutional investor equity holdings. Institutional investors have become a powerful force in the corporate landscape, controlling approximately half of the United States equity market (Conference Board, 2000) and accounting for approximately 80 percent of all daily transactions on the U.S. stock exchanges (Zahra et al., 2000). Given their substantial collective equity positions, the institutional investment community has an obvious incentive to vigilantly monitor the officers and directors in whose firms they invest in order to promote these firms' long-term performance (e.g., Alchian & Demsetz, 1972; Shleifer & Vishny, 1997). This oversight is consistent with agency theory and a means of constraining opportunism (e.g., Useem, 1996). We should note that institutional fund managers are themselves agents; nonetheless, fund managers have demonstrated a decided propensity to actively monitor executives in the firms in which their funds invest (e.g.. Black, 1992). Institutional fund managers have been particularly effective in achieving governance changes in the firms they target (e.g., Wahal, 1996). Institutional monitoring is aimed at improving firm performance. As one fund manager noted, "Whatever we do in the area of proxy initiatives or voting proxies, there has to be fundamentally an economic motivation behind it. Before we devote resources to something we really have to be able to say that this is going to leave our participants better off than if we hadn't done it" (Useem, Bowman, Myatt, & Irvine, 1993: 181). Hypothesis 2. Institutional investor equity holdings will be positively associated with firm financial performance. There are those who remain unconvinced about the efficacy of institutional investor monitoring in 15 relation to financial performance. Some inconsistency in findings with regard to institutional investor equity holdings and financial performance may be attributable to the heterogeneity of institutional investors (e.g., Brickley, Lease, & Smith, 1988). It has been demonstrated, for example, that different classes of institutional investors have different objectives (e.g., David, Kochhar, & Levitas, 1998; Kochhar & David, 1996). According to Brickley and his colleagues (1988), the propensity to engage in monitoring differs for "pressure-resistant," "pressure-sensitive," and "pressure-indeterminate" institutional investors; we define and discuss each type below. Pressure-resistant institutional investors include public pension funds, mutual funds, foimdations, and endowments. These investors do not normally have direct business relationships with firms in which they hold equity. As a result, the managers of their holdings are not likely to be subject to influence from the managements of the firms they invest in (Coffee, 1991). They are the most likely among institutional investors to actively monitor organizational decision makers and to consequently be labeled "activist institutional investors" (Brickley et al., 1988). Pressure-sensitive institutional investors include insurance companies, banks, and nonbank trusts. These types of institutional investors typically have ongoing business relationships with many of the firms in which they hold equity positions. These relationships create a dependence that may leave the institutional investors susceptible to managerial influence (Zahra et al., 2000). As a result, they are unlikely to engage in activity that is perceived as actively monitoring the decision makers in the firms in which they invest. Davis and Thompson noted that these funds "have a virtually unblemished history of passivity" (1994: 162). Corporate pension funds typify the pressureindeterminate institutional investor category. Relationships between the funds and the firms in which they invest may exist, but it is less clear than in the case of the pressure-sensitive investor how these relationships will affect propensity to monitor. Corporate pension funds are typically managed by professional investment managers selected by a company's management (Fromson, 1990). As a result of their ties to a firm, these fund managers are unlikely to actively challenge firm decision makers (Barnard, 1991). Their ties are weaker, however, than is typical for pressure-sensitive institutional investors. Given these differences among institutional investors, we propose: 16 Academy of Management Journal Hypothesis 3a. Pressure-resistant institutional investor equity holdings will be positively associated with firm financial performance. Hypothesis 3b. Pressure-sensitive institutional investor equity holdings will be negatively associated with firm financial performance. Hypothesis 3c. Pressure-indeterminate institutional investor equity holdings will be unassociated with firm financial performance. Blockholder equity holdings. Blockholders also have a strong incentive to actively monitor firm management. These equity holders are individuals or groups holding 5 percent or more of a given firm's equity. These owners are easily identified, as section 13(d) of the 1934 Securities and Exchange Act requires that "holders of more than five percent of a class of equity securities held by them are regarded as a 'group' and are required to file a Schedule 13D setting forth considerable information concerning the members of the group" (Sommer, 1990: 368). Any subsequent sales or purchases of the firm's equity by these individuals must also be reported. Agency theory suggests that such large-block shareholders have both the incentive and influence to assure that officers and directors operate in the interests of shareholders [Bethel & Liebeskind, 1993). Demsetz (1983) suggested that the substantial wealth they have at risk implies that the benefits of monitoring will outweigh associated costs. Blockholders have, in fact, demonstrated their ability to effect changes in the composition of boards and corporate constitutions (Pound, 1992). Blockholders typically have a stronger incentive than even activist institutional investors to engage in control activities. The typical institutional investor holds, on average, a 1 percent equity stake in a given firm; by definition, each blockholder in the same firm holds at least a 5 percent stake. Hypothesis 4. Large blockholder equity holdings will be positively associated with firm financial performance. February sary that a simple correlation between an equity and a performance variable be available in a publication or derivable from it (methods for converting other values into correlation coefficients are described by Lipsey and Wilson [2001], Rosenberg, Adams, and Gurevitch [2000], and Rosenthal and DiMatteo [2001]). Using a combination of computer-aided keyword searches and manual searches of pertinent joumals, principally in strategic management, finance, economics, and accounting, we obtained a subset of potentially apphcable research reports. We followed the "ancestry" approach to article identification (e.g.. Cooper, 1998). By working carefully from the more contemporary references, tracking the references on which the articles relied, and iteratively continuing this process, it is possible to determine a set of conmion early references with no published predecessors. The search process yielded 229 empirical studies with 1,880 germane bivariate correlations representing a combined sample of 939,567.^ The relatively large sample-to-study ratio results from the fi-equent use in governance research of multiple measures of equity and financial performance. Statistics from these correlations could only be combined if they reflected similar study characteristics (e.g., Rosenberg et al., 2000; Rosenthal & DiMatteo, 2001). The studies on which we relied do not have that character. Meta-Analytic Procedures The meta-analyses were conducted following guidelines provided by Hunter and Schmidt (1990; see also Hunter & Schmidt, 1994). Meta-analysis is a statistical technique for research synthesis that, while correcting for various statistical artifacts, allows the aggregation of results across separate studies and yields an estimate of the true relationship between two variables in a population. The zeroorder correlations between the variables of interest that a study reports are weighted by the sample size of the study in order to calculate the mean weighted correlation across all of the studies in the analysis. The standard deviation of the observed METHODS Sample We employed multiple search techniques to identify prior empirical research studies that measured equity holdings and firm financial performance. Whether a given indicator of equity or performance was a dependent, independent, or a control variable in a study was unimportant. These variables need not have been a main focus to be included in the meta-analyses. It was only neces- ^ While the total sample size for the combined observations is 939,567, this number is potentially misleading. Although this sample size is used for a single omnibus test to establish the potential existence of moderating influences, it has no utility beyond that as it is artificially inflated. Consider a study of 100 firms with two distinct dependent variables and one independent variable. The study itself would have an n of 100. But it has two correlations, one for each dependent variable with the independent variable. The n for the omnibus test, then, for these two correlations would be 200. Dalton, Daily, Certo, and Roengpitya 2003 correlations is then calculated to estimate their variability. Total variability across studies is comprised of the true population variation, variation due to sampling error, and variation due to other artifacts (that is, reliability and range restriction). Control of these artifacts provides a more accurate estimate of the true variability. To control for such artifacts, we relied on Comprehensive Meta-Analysis (Borenstein, 1997), a software package that employs Hunter and Schmidt's (1990) artifact distribution formulas. Although other meta-analyses in strategic management (e.g., Boyd, 1991; Capon, Farley, & Hoenig, 1990; Rhoades, Rechner, & Sundaramurthy, 2000; Schwenk & Shrader, 1993) have treated observed (not latent) variables as if they were without error (have a presumed reliability of 1.0), we opted for a more conservative 0.80 reliability estimate (e.g., Dalton, Daily, Ellstrand, & Johnson, 1998, Dahon, Daily, Johnson, & Ellstrand, 1999).^ RESULTS The first step in a meta-analysis is to establish a baseline population correlation (Boyd, 1991; Capon et al., 1990; Dalton et al., 1998,1999; Rhoades et al., 2000). More importantly, however, the baseline diagnostics also indicate the potential efficacy of subsequent subgroup analyses. Imagine a hypothetical meta-analysis with several indicators of performance and several independent variables of interest. If the baseline population correlation were very near zero and no moderating influences were indicated (that is, the variance was very small), there would be very little reason to pursue subgroup analysis. That result (modest correlation and little variance) suggests no substantive relationships between the variables of interest. For the data we analyzed, the corrected correlation was a modest .03, but the variance was much larger than would ^ We do not mean to appear critical about the choice of reliability level. In the entire data base (229 studies) on which we relied for this study, equity and financial performance variables were treated as observed in every case, save one in which a six-indicator performance construct was used. With that exception, data were not provided on either the reliability or validity of these measures in any of these studies. It is apparent that the empirical work in this area relies on equity and performance variables as observed and error-free. In the case of the data we used in our metaanalyses, the reliability assumption is robust. As will be demonstrated in Table 1, the corrected correlation for the overall test is .03. We also calculated that value with a .6, .7, and .9 reliability assumption. The largest difference in these compared to the .8 level was an R^ of .0005. 17 be expected if the population correlation were actually near zero. This pattern suggested that subgroup analyses might be productive.^ Using a test of homogeneity, we assessed whether the variability in effect sizes was larger than would be anticipated on the basis of sampling error. If so, our assumption would be that these correlations do not estimate a common population (e.g.. Cooper, 1998; Lipsey & Wilson, 2001).* Hunter and Schmidt (1990) provided one approach, suggesting that heterogeneity (that is, the presence of subgroups) is likely if the sampling error accounts for less than 75 percent of the observed variability. It has also been suggested tbat 90 percent credibility intervals larger than 0.11 imply the presence of subgroups (Kowlowsky & Sagie, 1993). In the baseline test, both of these indicators suggested the presence of moderating variables. In fact, the 90 percent credibility interval was nearly twice (.208) the standard provided by Kowlowsky and Sagie (1993). There are second levels of baseline analysis as well. The omnibus baseline analysis is the best estimate of the corrected correlation for all combinations of ^ A potentially important pretest for meta-analysis involves the time frame of studies. Consider a meta-analysis based on data from 30 to 40 years of empirical work. It is possible that the earlier work, relying on different measures and without the benefit of readily available archival data, would result in different population estimates of r than more contemporary work. Consider a simple example: Early work demonstrates a population estimate of a correlation (r) near zero; more contemporary work results in a much higher estimate ofthe population correlation. If these two groups of work are combined, the estimated correlation may be artificially low. We tested for that effect. The research through 1980 on which we relied yielded an estimated correlation of .01; the period from 1981 through 1990, .04; and from 1991 to the present, .02. In each case, the estimates are very near zero. There is no reason to believe, then, that the time in which the studies were conducted is of consequence to the analyses. •• The expressions "subgroup" and "moderator" are used interchangeably in the meta-analytic literature. This can lead to some confusion. For a large body of research, a moderator is ordinarily "operationalized" as a multiplicative variable. In a regression format, one would determine if the multiplicative term provided marginal variances above that provided by its elements. The analog for this Ln metaanalysis is accomplished through estahlishing subgroups. Two separate meta-analyses would be conducted for the two subgroups. An estimate of the population correlation would be calculated for each. Then, a critical ratio test would be used to determine if the two population correlations were statistically different. Throughout the article, we refer to such testing as "subgroup analysis." 18 Academy of Management Journal February TABLE 1 Meta-Analysis-Corrected Matrix of Population Correlations: Categories of Equity by Financial Performance Indicators" Variable Tobin's Q ROA ROA, IndustryAdjusted'' ROE ROE, IndustryAdjusted*" CEO equity .05* (14) [6,697] .09* (10) [3,749] .02 (4) [2,640] .02 (10) [1,648] Board equity .03 (4) [1,402] .13*" (12) [3,632] .05*" (16) [4,853] .01 (4) [400] Officer and director equity .07* (24) [31,381] .08*'" (7) [6,242] - . 1 1 * " (5) [1,561] -.02" (6) [612] Inside board equity .00 (4) [991] Outside board equity .06* (4) [2,640] Management equity .01 (10) [9,681] Institutional equity .02*" (89) [119,835] .06* (24) [4,213] .09* (5) [947] -.00 (18) [2,874] Blockholder equity .14* " (14) [14,787] -.02" (33) [27,446] .11* (9) [3,485] ROl ROl, IndustryAdjusted - . 1 1 (6) [1,006] -.02* (6) [580] .01 (5) [1,098] .09* (8) [3,307] .04 (8) [1,829] .07 (3) [324] equity measurements and performance indicators. An improved approach would be to examine the indicators of financial performance that are current in the literature (including, Tobin's Q, return on assets [ROA], return on equity [ROE], return on sales [ROS], and the price-to-eamings ratio) with equity holdings. Alternatively, one could examine the individual categories of equity holdings (such as CEO equity, board equity, institutional equity) with financial indicators. The results of both these procedures indicated modest levels of corrected correlations. The homogeneity tests, however, suggested the potential for subgroup analysis.^ H5rpothesis Tests Having determined that subgroup analyses were warranted for individual financial performance and equity categorizations, we turned to individual meta-analyses of specific measurements of equity and performance.*^ To do so, it was necessary to ^ Tables of the second-level base rate analyses would fill several pages. These data in their entirety are available by request from the first author. ^ The indicators of corporate financial performance noted in Table 1 reflect those on which the germane studies rely. There are many other indicators of corporate performance in the data (for instance, specialized metrics for banking performance [repossessed assets/total assets. .00 (11) [3,230] .07* (8) [1,477] -.05* "(19) [3,473] .10* (9) [1,003]-.05 (8) [800] .05 (8) [800] create a matrix of all equity measures by all performance indicators. In a traditional meta-analytical data presentation, such a matrix would require many pages. Instead, we constructed a metaanalysis-corrected correlation matrix (Table 1). In some ways it resembles a correlation matrix. The cell entries, however, are meta-analysis-corrected correlation population estimates. Consider the first entry in Table 1: The best estimate of the actual population correlation between CEO equity and Tobin's Q is .051. Tbe number of samples relied on for this calculation is in parentheses (14 in this case). In brackets is the sample size (6,697). Table 1 contains empty cells for two reasons. The first is a lack of applicable research. For example, we know of no empirical study addressing the relationship between inside board equity and ROS. Empty cells also result from there being a low number of relevant studies from which a meta-analysis can be reasonably interpreted. Clearly, better population estimates can be derived when tiiere are many samples and reasonable sample sizes (Hunter & Sclunidt, 1990). Although specific guidelines do not exist, three studies would seem to be a reasonable minimum. Hypothesis 1 states that insider equity, which was variously measvired in the studies we analyzed as Sheshunoff rankings]; slack; value added/employees; volatility). None of these reach the minimum number of samples (three) to be included in the table. 19 Dalton, Daily, Certo, and Roengpitya 2003 TABLE 1 Continued ROS Shareholder Returns Earnings per Share (EPS) Jensen's Alpha Ahnormal Returns Market-toBook Ratio .07 (3) [728] .03 (3) [598] .03 (13) [4,142] .09* (9) [2,116] .03" (15) [4,795] .17*-" (5) [1,028] .06*'"(61) [25,361] .01 (4) [589] .04* (4) [3,260] .05 (3) [312] .23*'(3) [1,729] -.01 (3) [318] .07* (5) [1,369] -.02 (4) [549] Price-toEarnings Ratio .04" (6) [1,378] .06" (9) [1,329] -.11*-" (7) [717] .04 (10) [2,700] .13* " (13) [1,729] .04* (16) [12,987] .16*'" (14) [1,436] .09* " (7) [1,725] -.00 (3) [3,324] .04* " (19) [24,492] .01 (7) [1,835] .01" (17) [2,641] .02" (6) [1,749] .05* (7) [6,733] -.10* (14) [1,436] ° The number of samples on which a meta-analysis relied is in parentheses; the total sample size for the meta-analysis is in brackets. A superscript asterisk (*) indicates that the corrected correlation (r) from the meta-analysis is statistically significant. A superscript point (") indicates that homogeneity tests suggested the presence of moderating variable(s). Empty cells indicate no relevant empirical studies or less than three observations. Please note that the entries in some cells are not independent from those in others. Consider, for example, a study that examined the relationship of institutional equity to ROA, ROE, and shareholder returns. That would result in three simple correlations, which we would use to calculate the results in three cells (i.e., institutional equity with ROA, with ROE, and with shareholder returns). "^ In the data on which our meta-analysis relied, only one study used something other than contemporaneous data. In this one study, the relationship between the amount of outside director equity was examined with ROA (lagged three years). That correlation was .02. That observation, however, was not included in the analyses appearing in Table 1. There were, however, several cases in which ROA (4 observations) and ROE (12 observations) were measured as multiyear averages that are included in these analyses. We tested the estimated correlations of these studies with those relying on a single year of ROA and ROE (.011 for a single year; .013 for multiple years). equity held by CEOs, managers, inside directors, and/or outside directors, will he positively associated with financial performance. With a single exception (officer and director equity as associated with earnings per share [EPS]), in no instance did an estimated population correlation between measurements of insider equity and financial performance exceed .018 in variance explained. Hypothesis 2 proposes a positive relationship between institutional investor equity and financial performance. In this case, too, the relationship with EPS is tbe largest (r = .129), but none of tbe relationsbips exceeds .017 in variance explained. Hypotbeses 3a-3c address a finer-grained analysis of institutional equity, tbe trichotomy of institutional investor types proposed by Brickley and bis colleagues (1988). We were able to provide a limited test of the proposition tbat pressure-sensitive, pressure-resistant, and pressure-indeterminate institutions will demonstrate different relationsbips witb financial performance. Tbe results are nonsupportive. Corrected correlations were .05, .10, and .02 for pressure-sensitive, pressureresistant, and pressure-indeterminate institutions, respectively. Tbese differences are not statistically significant. Tbese results, bowever, must be carefully interpreted as tbere were only seven samples available witb wbicb to test tbis proposition.'' Hypotbesis 4 predicts a positive relationship between blockholder equity and financial performance. Witb tbe possible exception of EPS, tbere are no substantive relationsbips (.019 variance explained is the largest). Tbere is very little evidence for systematic relationsbips between equity and financial performance in Table 1. Notably, tbe only relationsbips of even marginal consequence involved EPS. Altbougb our rationale for tbose relationsbips is speculative, we note '' Also, the dependent variables are not the same across the seven studies. 20 Academy of Management Journal tbe following interesting discussion about tbe use of EPS as an indicator of corporate financial performance. Meyer and Gupta suggested tbat "managers learned various techniques, including leverage, aimed at increasing EPS witbout otberwise cbanging the performance of the firm . . . EPS became a wrong performance measure as managers adapted tbeir decisions to it" (1994: 323). Perbaps, tben, tbese EPS relationships are somewhat more subject to manipulation. CoUingwood recently noted tbat "earnings management pervades tbe U.S. financial system" (2001: 67). It may also be notable tbat we saw essentially no relationsbips among tbe categories of equity bolders and market returns (sbarebolder returns, Jensen's alpba, and abnormal returns). The bigbest corrected correlation among tbese is .09. Even so, bowever, witb tbese and most of tbe accounting performance measures, there continue to be indications of the presence of subgroups or moderating infiuences. DISCUSSION Tbe universal applicability of agency theory bas recently been questioned (e.g.. Lane, Cannella, & Lubatkin, 1998). Addressing governance studies specifically, Walsb and Kosnik noted tbat researcbers "need to be alert to tbe possibility tbat the bypotbesized effects may be mucb more narrowly circumscribed tban tbe theory's proponents might argue" (1993: 696). Similarly, Davis, Scboorman, and Donaldson cautioned that "exclusive reliance upon agency tbeory is imdesirable because tbe complexities of organizational life are ignored" (1997: 20). Consistent witb tbese admonitions, tbe results of . our meta-analyses do not support agency tbeory's proposed relationsbip between ownersbip and firm performance. Tbis is a significant finding, as agency tbeory bas been cbaracterized as a tbeory of tbe ownersbip structure of tbe firm (e.g., Jensen & Meckling, 1976). Also, as we bave noted, agency tbeory provides tbe tbeoretical foundation for tbe vast majority of researcb conducted in corporate governance (e.g., Sbleifer & Visbny, 1997). Our results illustrate relatively low relationships between various categories of equity and multiple indicators of financial performance. In fairness, we sbould add tbat the question of at what value a correlation becomes important is an interesting one. Rosenthal, Rosnow, and Rubin (2000) provided an illuminating treatment of how relatively small effect sizes can bave critical consequences. Consider tbe Fortune 500. To be placed in tbat group, a company bas to bave at least $3 billion per year in revenues; to be in tbe top 100 requires at least $20 billion in revenues; and tbe top 10 average Fehruary over $130 billion in revenues. Relatively small effect sizes in revenue data from sucb firms could represent many millions of dollars of gain or loss. An Alternative Theoretical Lens: Substitution Theory For us, tbe results of our meta-analyses suggest tbe need to consider alternative tbeoretical lenses for examining the ownership-performance relationship. Other tbeoretical perspectives tbat bave been applied to corporate governance researcb more generally include tbe legalistic perspective (e.g., Sbleifer & Visbny, 1997; Zahra & Pearce, 1989) and stewardsbip tbeory (e.g., Davis et al., 1997). Tbe legalistic perspective addresses tbe legal protections afforded sbareholders (Coffee, 1999; Shleifer & Vishny, 1997). bi tbe face of managerial malfeasance, even minority sbarebolders are free to commence litigation (Coffee, 1999). Legal rights may dominate ownersbip concentration witbin tbis tbeoretical perspective. Stewardsbip tbeory focuses on explanations of why managers' interests would be aligned witb tbose of shareholders (Davis et al., 1997). Within this theory, organizational structures that facilitate managers' power are preferred to those designed to constrain managerial power. In fact, control is viewed as potentially counterproductive (Davis et al., 1997). We propose an additional perspective based on the substitution hypothesis (e.g., Rediker & Seth, 1995; Shleifer & Vishny, 1997). Like agency theory, the substitution bypotbesis focuses on issues surrounding tbe separation of ownership and control in public corporations (Rediker & Setb, 1995). Scbolars taking tbis perspective seek to explain tbe effectiveness of various governance mecbanisms in limiting managerial opportunism. Wbereas agency tbeory addresses the ability of governance mechanisms "to resolve the shareholder-manager agency problem independent of eacb otber," substitution tbeory addresses tbe relationsbips among alternative governance mecbanisms (Rediker & Setb, 1995: 86; see also Agrawal & Knoeber, 1996; Demsetz, 1983). Our proposed substitution theory approach is analogous to contingency theory approaches that have been applied to otber corporate governance issues. Finkelstein and D'Aveni (1994), for example, concluded tbat a contingency approacb was most suitable for explaining tbe adoption of tbe leadersbip structure in wbicb one person is botb CEO and board cbair. Relying on agency tbeory and organization tbeory, tbey empirically establisbed that CEO duality was more or less appropriate depending on tbe presence of certain contingent factors, sucb as CEO power and firm performance. 2003 Dalton, Daily, Certo, and Roengpitya They, too, concluded that agency theory, although applicable, bas inberent limitations. Some support for tbe presence of substitution effects is found in extant researcb (e.g., Agrawal & Knoeber, 1996; Jensen, Solberg, & Zorn, 1992; Kocbbar, 1996; Moyer, Rao, & Sisneros, 1992; Rediker & Seth, 1995). In view of our findings, we propose that ownership structure is one governance mechanism to be considered among a range of governance mecbanisms. Ownersbip categories may effectively substitute for anotber. Also, alternative governance mechanisms may substitute for ownersbip structure. Could, for example, tbe expected relationsbip between CEO equity and financial performance be conditioned on tbe equity beld by other blockholders? A CEO with a 5 percent equity position might bebave very differently depending on wbetber other blockbolders beld, for example, 5 percent or 50 percent of tbe remaining equity. One migbt also consider a simple two-by-two model addressing bigb/low equity beld by officers and directors and bigb/low equity beld by institutional holders (Atkinson and Galaskiewicz [1988] took a similar approach in the context of ownership patterns and corporate contributions). Would one expect differences in financial performance among tbese four cells? Limitations and Future Research Opportunities Tbe inferential logic underlying the results we report is robust. The studies and observations reported in Table 1 concern the largest firms in the United States (drawn, for instance, from tbe Fortune 500, tbe Standard & Poors 500, and tbe Forbes 500).^ Tbis composition provides a distinct metbodological advantage, amounting to our baving a series of samples drawnfiroma discrete population, with replacement. It is tme that the exact elements of the population— the largest corporations—cbange over time. Even so, tbe fundamental nature of tbe population is invariant. In Lykken's (1968) classic formulation, tbese studies amoiuit to an extensive series of constructive replications. Time after time, researcbers, wbile relying on different measurements of equity and performance in varied contexts, bave investigated tbe equity-performance relationsbip. However, sucb a sample also has a downside. Oiu" results should be interpreted with some care as it would not be appropriate to generalize our findings beyond tbis specific sample. Although much of tbe * For the overall test baseline, tbere were some samples comprised of smaller firms and four samples of the largest banking groups. For Table 1, bowever, we omitted these studies. 21 attention in strategic management, accounting, and finance research, for example, is focused on this population of very large corporations, they comprise a very small percentage of total business enterprise. Tbe research we report in Table 1 does not include, for example, initial public offerings (IPOs). Tbis may be an area, however, wherein the equity of various parties may be associated with IPO performance (e.g.. Booth & Cbua, 1996; Mikkelson, Partcb, & Sbah, 1997). One of the limitations of meta-analytical procedures is that causality cannot be tested or imputed. There are exceptions, as meta-analyses of studies with causal experimental protocols can certainly be syntbesized. Corporate governance researcb does not have that character. For the meta-analyses on which our results depend, however, causality is not an issue because there is no indication of systematic relationships. Future research in this domain could easily address these issues. A multiperiod structural equations analysis, for instance, could provide a longitudinal perspective on the results as well as inform the discussion of causality. Consideration of the potential for nonlinear relationships in the ownership-performance area may also be fruitful. Tbe relatively modest empirical literatiu-e examining nonmonotonic relationsbips among equity categorizations and performance measures does not reflect a consensus view of tbe nature of this relationship (e.g., Hermalin & Weisbacb, 1991; McConnell & Servaes, 1990; Morck, Sbleifer, & Visbny, 1988; Tbomsen & Pedersen, 2000). In tbe next generation of research addressing relationships between equity and performance, attention sbould be given to tbe potential for nonlinear relationsbips.^ Conclusion In tbeir review of the corporate governance literature, Shleifer and Vishny concluded tbat tbere are "a variety of still open questions" in tbe field of governance (1997: 774). One of tbe questions tbey identified was wbetber the costs and benefits of ^ We were unable to provide a meta-analysis for a nonlinear relationship between equity and financial performance. The relatively few relevant studies were configured in low, medium, and high categories of equity. In principle, it would have been easy to provide a separate meta-analysis for each, compare them with critical ratios, and determine, in fact, if the estimated population correlations were statistically different. Unfortunately, there was no consensus among the studies on what constituted the inflection points across categories. Also, owing to the small niunber of samples in these categories, it would have been necessary to rely on several financial perfonnance variables in concert. 22 Academy of Management Journal concentrated ownership are significant. Given the centrality of ownership and performance issues in corporate governance research, we are hopeful that the results of the meta-analyses reported here, and our discussion, will encourage further research drawing on alternative theoretical approaches. In particular, we encourage future researchers to consider the potential for governance mechanisms, ownership or otherwise, to effectively substitute for one another and/or operate in concert. REFERENCES February intensity, relative mix, and economic determinants. Joumal ofBusiness, 73: 661-693. Gapon, N., Farley, J. U., & Hoenig, S. 1990. Determinants of financial performance: A meta-analysis. Management Science, 36: 1143-1159. Goase, R. 1937. The nature of the firm. Economica, 4: 386-405. Goffee, J. G. 1991. Liquidity versus control: The institutional investor as corporate monitor. Columhia Law Review, 91: 1277-1368. Goffee, J. G. 1999. The future as history: The prospects for global convergence in corporate governance and its implications. Northwestern University Law Review, 93: 641-707. Agrawal, A., & Knoeber, C. R. 1996. Firm performance and mechanisms to control agency problems between managers and shareholders. Journal of Financial and Quantitative Analysis, 31: 377-397. Gollingwood, H. 2001. The earnings game: Everyone plays, nobody wins. Harvard Business Review, 79(3): 65-74. Alchian, A. A., & Demsetz, H. 1972. Production, information costs, and economic organizations. American Economic Review, 62: 777-795. Gonference Board. 2000. Institutional investment report: Financial assets and equity holdings. New York: Gonference Board. Atkinson, L., & Galaskiewicz, J. 1988. Stock ownership and company contributions to charity. Administrative Science Quarterly, 33: 82-100. Gooper, H. 1998. Synthesizing research: A guide for literature reviews. Thousand Oaks: GA: Sage. Barnard, J. W. 1991. Institutional investors and the new corporate governance. North Carolina LawReview, 69: 1135-1187. Berle, A., & Means, G. 1932. The modem corporation and private property. New York: Macmillan. Bethel, J. E., & Liebeskind, J. 1993. The effects of ownership structure on corporate restructuring. Strategic Management Journal, 14: 15-31. Dalton, D. R., Daily, G. M., Ellstrand, A. E., & Johnson, J. L. 1998. Meta-analytic reviews of board composition, leadership structure, and financial performance. Strategic Management Joumal, 19: 269290. Dalton, D. R., Daily, G. M., Johnson, J. L., & Ellstrand, A. E. 1999. Number of directors on the board and financial performance: A meta-analysis. Academy of Management Joumal, 42: 674-686. Bethel, J. E., Liebeskind, J. P., & Opler, T. 1998. Block share purchases and corporate performance. Journal of Finance, 53: 605-634. David, P., Kochhar, R., & Levitas, E. 1998. The effect of institutional investors on the level and mix of GEO compensation. Academy of Management Joumal, 41: 200-208. Black, B. S. 1992. Agents watching agents: The promise of institutional investor voice. UCLA Law Review, 39: 811-893. Davis, G. F., & Thompson, T. A. 1994. A social movement perspective on corporate control. Administrative Science Quarterly, 39: 141-173. Booth, J. R., & Ghua, L. 1996. Ownership dispersion, costly information, and IPO underpricing. Journal of Financial Economics, 41: 291-310. Davis, J. H., Schoorman, F. D., & Donaldson, L. 1997. Toward a stewardship theory of management. Academy of Management Review, 22: 20-47. Borenstein, M. 1997. Comprehensive Englewood, NJ: Biostat. Demsetz, H. 1983. The structure of ownership and the theory of the firm. Journal of Law & Economics, 26: 375-390. meta-analysis. Bothwell, J. L. 1980. Profitability, risk, and the separation of ownership from control. Joumal of Industrial Economics, 28: 303-311. Boyd, B. K. 1991. Strategic planning and financial performance: A meta-analytical review. Joumal of Management Studies, 28: 353-374. Brickley, J. A., Lease, J. A., & Smith, G. W. 1988. Ownership structure and voting on antitakeover amendments. Journal of Financial Economics, 20: 267291. Bryan, S., Hwang, L., & Lilien, S. 2000. GEO stock-based compensation: An empirical analysis of incentive- Demsetz, H., & Lehn, K. 1985. The structure of corporate ownership: Gauses and consequences. Joumal of Political Economy, 93: 1155-1177. Eisenhardt, K. M. 1989. Agency theory: An assessment and review. Academy of Management Review, 14: 57-74. Finkelstein, S., & D'Aveni, R. A. 1994. GEO duality as a double-edged sword: How boards of directors balance entrenchment avoidance and unity of command. Academy of Management Joumal, 37: 1079-1108. 2003 Dalton, Daily, Certo, and Roengpitya 23 Fromson, B. D. 1990. The big owners roar. Fortune, July 30: 66-78. on equity ownership and corporate values. Joumal of Financial Economics, 27: 595-612. Hermalin, B. E., & Weisbach, M. S. 1991. The effects of board composition and direct incentives on firm performance. Financial Management, 20(4): 101-122. McGonnell, J. J., & Servaes, H. 1995. Equity ownership and the two faces of debt. Journal of Financial Economics, 39: 131-157. Himmelberg, G. P., Hubbard, R. G., & Palia, D. 1999. Understanding the determinants of managerial ownership and the link between ownership and performance. Journal of Financial Economics, 53: 353384. Meyer, M. W., & Gupta, V. 1994. The performance paradox. In B. M. Staw & L. L. Gummings (Eds.), Research in organizational behavior: 309-369. Greenwich, GT: JAI Press. Hunter, J. E., & Schmidt, F. L. 1990. Methods of metaanalysis: Correcting error and bias in research findings. Beverly Hills, GA: Sage. Hunter, J. E., & Schmidt, F. L. 1994. Gorrecting for sources of artificial variation across studies. In H. Gooper & L. V. Hedges (Eds.). The handbook of research synthesis: 324-336. New York: Russell Sage Foundation. Jensen, M. G. 1993. The modern industrial revolution, exit, and the failure of internal control systems. Journal of Finance, 48: 831-880. Jensen, M. G., & Meckling, W. H. 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3: 305-360. Mikkelson, W. H., Partch, M. M., & Shah, K. 1997. Ownership and operating performance of companies that go public. Joumal of Financial Economics, 44: 281307. Morck, R., Shleifer, A., & Vishny, R. W. 1988. Managerial ownership and market valuation. Joumal of Financial Economics, 20: 293-315. Moyer, R. G., Rao, R., & Sisneros, P. M. 1992. Substitutes for voting rights: Evidence from dual class recapitalizations. Financial Management, 21: 35—47. Perry, T., & Zenner, M. 2000. GEO compensation in the 1990s: Shareholder alignment or shareholder expropriation? Wake Forest Law Review, 35: 123-152. Pound, J. 1992. Beyond takeovers: Politics comes to corporate control. Harvard Business Review, 70: 83-93. Jensen, M. G., & Murphy, K. J. 1990. Performance pay and top-management incentives. Joumal of Political Economy, 98: 225-264. Rediker, K. J., & Seth, A. 1995. Boards of directors and substitution effects of alternative governance mechanisms. Strategic Management Joumal, 16: 85—99. Jensen, G. R., Solberg, D. P., & Zorn, T. S. 1992. Simultaneous determination of insider ownership, debt, and dividend policies. Journal of Financial and Quantitative Analysis, 27: 247-263. Rhoades, D. L., Rechner, P. L., & Sundaramurthy, G. 2000. Board composition and financial performance: A meta-analysis ofthe influence of outside directors. Joumal of Managerial Issues, 12: 76-91. Kochhar, R. 1996. Explaining firm capital structure: The role of agency theory vs. transaction cost economics. Strategic Management Joumal, 17: 713-728. Rosenberg, M. S., Adams, D. G., & Gurevitch, J. 2000. MetaWin: Statistical software for meta-analysis. Sunderland, MA: Sinauer Associates, Inc. Kochhar, R., & David, P. 1996. Institutional investors and firm innovation: A test of competing hypotheses. Strategic Management Joumal, 17: 73-84. Rosenthal, R., & DiMatteo, M. R. 2001. Meta-analysis: Recent developments in quantitative methods for literature review. Annual Review of Psychology, 52: 59-82. Kowlowsky, M., & Sagie, A. 1993. On the efficacy of credibility intervals as indicators of moderator effects in meta-analytic research. Journal of Organizational Behavior, 14: 695-699. Lane, P. J., Gannella, A. A., & Lubatkin, M. H. 1998. Agency problems as antecedents to unrelated mergers and diversification: Amihud and Lev reconsidered. Strategic Management Joumal, 19: 555-578. Lipsey, M. W., & Wilson, D. B. 2001. Practical metaanalysis. Thousand Oaks, GA.: Sage Publications. Lykken, D. T. 1968. Statistical significance in psychological research. Psychological Bulletin, 70: 151-159. Macey, J. R. 1997. Institutional investors and corporate monitoring: A demand-side perspective. Managerial and Decision Economics, 18: 601—610. McGonnell, J. J., & Servaes, H. 1990. Additional evidence Rosenthal, R., Rosnow, R. L., & Rubin, D. B. 2000. Contrasts and effect sizes in behavioral research. Gambridge, England: Gambridge University Press. Schwenk, G. R., & Shrader, G. B. 1993. Effects of formal strategic planning on financial performance in small firms: A meta-analysis. Entrepreneurship Theory & Practice, 17: 53-64. Shleifer, A., & Vishny, R. W. 1997. A survey of corporate governance."/ourna7 of Finance, 52: 737-783. Short, H. 1994. Ownership, control, financial structure, and the performance of firms. Journal of Economic Surveys, 8: 203-249. Smith, A. 1776/1952. An inquiry into the nature and causes of the wealth of nations (R. M. Hutchins [Ed.], Great books of the western world). Ghicago: Encyclopedia Britannica. Academy of Management Journal 24 Sommer, A. A. 1990, Corporate governance in the nineties: Managers vs. institutions. University of Cincin- February _A)\ natHaw Review, 59: 357-383. Thomsen, S,, & Pedersen, T, 2000. Ownership structure and economic performance in the largest European companies. Strategic Management Journal, 21: 689-705, Useem, M, 1996. Investor capitalism: How money managers are changing the face of corporate America. York: Basic Books. Useem, M., Bowman, E. H,, Myatt, J,, & Irvine, C, W. 1993, U,S, institutional investors look at corporate governance in the 1990s. European Management Journal, 11: 175-189. Wahal, S, 1996. Pension fund activism and firm performance. Journal of Financial and Quantitative Analysis, 31: 1-23. Walsh, J, P,, & Kosnik, R. D. 1993, Corporate raiders and their disciplinary role in the market for corporate control. Academy of Management Journal, 35: 671-700, Zahra, S. A,, Neubaum, D, O,, & Huse, M. 2000. Entrepreneurship in medium size companies: Exploring the effects of ownership and governance systems. Journal of Management, 26: 947-976. Zahra, S, A., & Pearce, J. A, 1989, Boards of directors and corporate financial performance: A review and integrative model. Journal of Management, 15: 291334, Dan R. Dalton (dalton@indiana.edu) is the dean and the Harold A. Poling Chair of Strategic Management of the Kelley School of Business, Indiana University. He received his Ph,D, at the University of California, Irvine. Professor Dalton's work focuses on corporate governance, particularly option repricing, equity holdings, stock-hased board compensation, and initial public offerings (IPOs). Catherine M. Daily is the David H, Jacohs Chair of Strategic Management in the Kelley School of Business, Indiana University. She received her Ph.D, in strategic management from the Indiana University. Her research interests include corporate governance, strategic leadership, the dynamics of husiness failure, ownership structures, and managerial ethics, S. Trevis Certo received his Ph.D. in strategic management from the Kelley School of Business, Indiana University, and is now on the faculty at the Lowry Mays College and Craduate School of Business, Texas A&M University. His work has focused on corporate governance, with particular emphases on the examination of initial puhlic offerings (IPOs), CEOs and top management teams, and hoards of directors, Rungpen Roengpitya is a doctoral student in strategic management at the Kelley School of Business, Indiana University. Her research interests are corporate equity structures and international governance. 2003 Dalton, Daily, Certo, and Roengpitya 25 APPENDIX Articles Included in the Meta-Analyses" Author Year Publication Agrawal & Knoeber Agrawal & Knoeber Amihud, Lev, & Travlos Ang, Cole, & Lin Baliga, Moyer, & Rao Balkin, Markman, & Gomez-Mejia Barhnhart & Rosenstein Bathaia Bathaia, Moon, & Rao Baysinger, Kosnik, & Turk Beatty Beatty & Zajac Bergh Bergstrom & Rydqvist Berkman & Bradbury Bethel & Liebeskind Bhagat & Black Bhagat, Carey, & Elson Blasi, Conte, & Kruse Boehmer Boeker Boeker Boeker Boeker & Coodstein Bommer & Ellstrand Borokhovich, Brunarski, & Parrino Boyd Brennan & Franks Brook, Henderschott, & Lee Brook, Hendershott, & Lee Brous & Kini Brush, Bromiley, & Hendrickx Bryan, Hwang, & Lilien Bucholtz & Rihbens Byrd & Stammerjohan Campsey & DeMong Carlson & Bathaia Carpenter Carter & Manaster Cebenoyan, Cooperman, & Register Chaganti & Damanpour Chang & Mayers Chen & Jaggi Cho Claessens & Djankov Clyde Cochran, Wood, & Jones Coles & Hesterly Coles, McWilliams, & Sen Conte & Tannenbaum Conyon & Peck Cotter & Zenner Cubbin & Leech Daily Daily Daily & Dalton Daily & Johnson Daily, Jobnson, Ellstrand, & Dalton Datta & Rajagolpalan David, Hitt, & Gimeno David, Kochhar, & Levitas Davis Davis & Stout Demsetz & Lehn Denis, Denis, & Sarin Dennis & Sarin Dhillon & Ramirez Dowen & Bauman Duggal & Millar Duggal & Millar El-Gazzar Elliott Faccio & Lasfer Filbeck 1996 1998 1990 2000 1996 2000 1998 1996 1994 1999 1989 1994 1995 1990 1996 1993 1999 1999 1996 2000 1989 1992 1997 1993 1996 1997 1994 1997 1998 2000 1994 2000 2000 1994 1997 1983 1997 2000 1990 1995 1991 1992 2000 1998 1999 1997 1985 2000 2001 1978 1998 1994 1986 1995 1996 1994 1997 1998 1998 2001 1998 1991 1992 1985 1997 1999 1994 1997 1994 1999 1998 1972 2000 1996 JFQA JFE JF JF SMJ AMJ FR FR FM AMJ AR ASQ SMJ JBF FM SMJ BL BL ILRR JFl ASQ ASQ ASQ AMJ GOM JF SMJ JFE JF JCF FM SMJ JB AMJ FR RBER JBFA JM JF FM SMJ JFE JAPP JFE JCE MDE AMJ JM JM MLR AMJ JFE MDE JM SMJ AMJ JM AMJ SMJ AMJ AMJ ASQ ASQ JPE JF JFE JBF FR QREF JCF AR JFQA JCF QJBE APPENDIX Continued Author Finkelstein & Boyd Finkelstein & D'Aveni Finkelstein & Hambrick Gamble Geczy, Minton, & Scbrand Gedajlovic & Shapiro Glassman & Rhoades Gomez-Mejia, Tosi, & Hinkin Gompers & Lerner Gorton & Rosen Graves Gray & Cannella Had.lock, Houston, & Ryngaert Han, Lee, & Suk Han & Suk Hanson & Song Haushalter Hayward & Boeker Hajrward & Hambrick Heflin & Shaw Hermalin & Weisbach Hill & Snell Hill & Snell Himmelberg, Hubbard, & Palia Hirschey & Zaima Hite & Vetsuypens Hodrick Hoiderness & Sheehan Holl Hoii Hoskisson, Johnson, & Moesel Hoskisson & Johnson Hudson, Jahera, & Lloyd Jain & Kini James & Wier Jarreii & Poulsen Johnson, Hoskisson, & Hitt Jones, Lee, & Thompkins Kamerschen Kamerschen & Paul Ke, Petroni, & Safieddine Keloharju & Kulp Kerr & Kren Kesner Kim, Krinsky, & Lee Kim, Lee, & Francis Kini & Mian Klassen Klein Knopf & Teali Krause Kren & KenLane, Cannella, & Lubatkin Lang, Stulz, & Walkling Lange & Sharpe Leweilen, Loderer, & Rosenfeld Li & Simeriy Lin Livingston & Henry Madden Mahoney, Sundaramurthy, & Mahoney Maksimovic & Unal Mallette & Fowler Mallete & Hogler McBain & Krause McConneil & Servaes McEachern McKean & Kania McWilliams & Sen Mehran Mehran Mikkelson, Partch, & Shah Mishra & Nielsen Year Publication 1998 1994 1989 2000 1997 1998 1980 1987 1999 1995 1988 1997 1999 1999 1998 2000 2000 1998 1997 2000 1991 1988 1989 1999 1989 1989 1999 1988 1975 1977 1994 1992 1992 1994 1990 1988 1993 1997 1968 1971 1999 1996 1992 1987 1997 1988 1995 1997 1998 1996 1986 1997 1998 1991 1995 1989 1998 1996 1980 1982 1997 AMJ AMJ SMJ JBV JF SMJ RES AMJ JLE 1993 1992 1995 1989 1990 1978 1978 1997 1992 1995 1997 2000 JF AMJ JM JBV JFE JIE JB JFQA JFQA JFE JFE FM JF AMJ JM JBF MBR RFE JCF JF ASQ ASQ JFQA FM AMJ AMJ JFE JF JF JFE JFE JIE JIE AMJ SMJ FR JF JFE JFE SMJ MDE AER MIR JAE JBF AMJ JM JAAF FR JFR AR JLE JBF JBE ABR SMJ JFE AFE JFQA SMJ NULR JRI JEB MDE Academy of Management Journal 26 February APPENDIX Continued APPENDIX Continued Publication Author Author Year Mizrucbi & Stearns Moh'd, Perry, & Rimbey Moh'd, Perry, & Rimbey Molz Morck, Nakumura, & Shivadasani Morck, Shieifer, & Vishny Mudambi & Nicosia Muraii & Welch Neun & Santerre O'Reilly, Main, & Crystal Palia & Licbtenbert Palmer, Jennings, & Zhou Peck Peng & Luo Pi & Timme Porac, Wade, & Pollock Pouder & Cantreii Prowse Raad & Reinganum Radice Ramaswamy, Veiiyath, & Gomes Rosenstein & Wyatt Round Ruef & Scott Safieddine & Titman Saiancik & Pfeffer Sanders & Carpenter Santerre & Nuen Schoonhoven, Eisenhardt, & Lyman Short & Keasey Simpson & Gleason Singh & Harianto Singh & Harianto Slovin & Sushka Smith & Amoako-Adu Song & Walkiing Sorenson Stano Steer & Cable Steiner Stigler & Friedland Sundaramurthy Sundaramurthy & Lyon Sundaramurthy, Mahoney, & Mahoney Sundaramurthy & Rechner Sundaramurthy, Rechner, & Wang Szewczyk & Tsetsekos Szewczyk, Tsetsekos, & Varma Thomsen & Pedersen Thonet & Pensgen Tkac Tosi & Gomez-Mejia Trostel & Nichols Tsetsekos & DeFuso Utama & Cready Vafeas Vafeas Vernon Vijh Wade; O'Reilly, & Ghandratat Ware Wedig, Sloan, Hassan, & Morrissey Werner & Tosi Westphai Westphal Westphai & Milton Westphai & Zajac Westphal & Zajac Wruck Yermack Yermack 1994 1995 1998 1988 2000 1988 1998 1989 1986 1988 1999 1993 1996 2000 1993 1999 1999 1992 1995 1971 2000 1997 1976 1998 1999 1980 1998 1993 1990 ASQ FR FR JBR JB JFE AFE Zahra Zahra, Neubaum, & Huse Zajac & Westphai Zajac & Westphai Zajac & Westphal Zeitlin & Norich JBFA 1999 1999 1989a 1989b 1993 1999 1993 1974 1975 1978 1996 1983 1996 1998 1997 JCF " Abbreviations for the journals are as follows: ABR: Accounting and Business Research; AER: American Economic Review; AFE; Applied Financial Economics; AMJ: Academy of Management Journal; AR: Accounting Review; ASQ: Administrative Science Quarterly; BL; Business Lawyer; BS; Business and Society; EI; Economic Inquiry; EJ; Economic Journal; FM; Financial Management; FR; Financial Review; GOM: Group and Organization Management; ILRR; Industrial and Labor Relations Review; IREF; International Review of Economics and Finance; JAAF; Journal of Accounting, Auditing, and Finance; JAE; Journal of Accounting and Economics; JAPP: Journal of Accounting and Public Policy; JB; Journal ofBusiness; JBE; Journal of Behavioral Economics; JBF; Journal of Banking and Finance; JBFA; Journal of Business, Finance, and Accounting; JBR: Journal of Business Research; JBS: Journal of Business Strategies; JBV: Journal of Business Venturing; JCE; Journal of Comparative Economics; JCF; Journal of Corporate Finance; JEB; Journal of Economics and Business; JIE; Journal of Industrial Economics; JF; Journal of Finance; JFE; Journal of Financial Economics; JFl; Journal of Financial Intermediation; JFQA: Journal of Financial and Quantitative Analysis; JFR; Journal of Financial Research; JLE; Journal of Law and Economics; JM: Journal of Management; JMI; Journal of Managerial Issues; JPE; Journal of Political Economy; JPM: Journal of Portfolio Management; JRI; Journal of Risk and Insurance; MBR: Multinational Business Review; MDE; Managerial and Decision Economics; MIR; Management International Review; MLR; Monthly Labor Review; NULR: Northwestern University Law Review; QJBE; Quarterly Journal of Business and Economics; QREB; Quarterly Review of Economics and Business; QREF; Quarterly Review of Economics and Finance; RBER; Review of Business and Economic Research; RES: Review of Economics and Statistics; RFE; Review of Financial Economics; RIDSEC; Rivista Internazionale Di Scienze Economiche e Commercial; RPE; Research in Political Economy; SEJ; Southern Economic Journal; SMJ; Strategic Management Journal. 1997 1996 1993 1992 2000 1979 1999 1994 1982 1990 1997 1999 2000 1971 1999 1990 1975 1988 1995 1998 1999 2000 1995 1997 1989 1996 1995 MDE ASQ JCF ASQ JFE AMJ JBF ASQ JBS JF FM EJ MIR JFE RIDSEC ASQ JF AMJ AMJ EI ASQ IREF AMJ SMJ JF JCF JFQA SEJ SEJ JIE QJBE JLE SMJ JMI SMJ BS JM FR FR SMJ JIE JFQA AMJ AMJ JPM JAE JFE JAPP JFQA JFE ASQ QREB JF AMJ ASQ AMJ ASQ ASQ ASQ JFE JFE JFE Year Puhlication 1996 2000 1994 1995 1996 1979 AMJ JM SMJ ASQ AMJ RPE