Printer Friendly

On Audit Committee Appointments.

Policy makers have recently drawn emphasis to the independence of audit committee members as a prime determinant of financial-reporting quality. Yet, ability and willingness to perform well in an audit committee are probably related to other director characteristics, beyond independence, as well. This paper addresses a range of characteristics shared by directors being appointed to the audit committee. To understand these characteristics, I identify 262 nonexecutive directors that are appointed to audit committees of U.S. public firms (both Fortune 500 and others) between 1995 and 1998 and compare them to 262 nonexecutive directors who served on the appointing company's board at that time but who were not appointed to the audit committee.

These comparisons suggest that share ownership in the appointing firm and director experience in other firms (attributes that are likely to enhance director performance) are not different between appointed directors and the rest. In addition, directors who are affiliated with management are only marginally less likely to be chosen for committee service. In contrast, appointed directors serve in significantly fewer other committees and have served for less time on the company's board. These findings suggest that seniority considerations and concerns over the equitable division of committee service among directors drive the appointment decision. They also suggest that, on average, corporate boards overlook important director attributes in making audit committee appointments, and that policy makers should consider attributes beyond director independence in setting guidelines for audit committee appointment. Nevertheless, further research is needed to establish the link between director attributes and the quality of audit committee member decisions.

INTRODUCTION

Both the business and academic press have recently focused on audit committee composition as an important determinant of financial-reporting quality (e.g., Vicknair et al. 1993; Cadbury 1995). In particular, there is widespread support for the notion that audit committees should consist of nonexecutive directors, who are more likely to be independent of management's influence and are thus better suited to oversee the financial-reporting process (Beasley 1996). In this spirit, a recent report by the Blue Ribbon Committee (1999) strongly recommends that committee member directors should be independent of management's influence.

The extent of outsider participation in audit committees is likely to have increased considerably in recent years following public scrutiny of insider appointments on audit committees. On the surface, sparse insider participation in audit committees appears to diminish the importance of concerns over the committee's make-up; i.e., if almost all audit committee members are nonexecutives, one may argue that further concerns over committee composition are not relevant. Importantly, however, prior academic literature has suggested that there are significant differences among the incentives of nonexecutive directors to monitor management because of variation in their degree of independence, equity investment with the firm, and value of reputation capital. Further, the Blue Ribbon Committee (1999) report highlights the importance of not having a fiduciary relationship with the firm, as well as the financial literacy of committee member directors. That is, not all directors, even if nonexecutives, will be equally effective in overseeing the financial-reporting process. Thus, in the context of audit committees, the choice of a new member even among a pool of nonexecutive director candidates is likely to be important. In light of this, the study of factors that determine an audit committee appointment is an important empirical question and will shed light on the efficacy of such decisions.

Accordingly, the present study identifies director characteristics that drive the selection of a director for audit committee service among outside director candidates. Given the ad hoc nature of the characteristics examined, this study is exploratory, contributing to our understanding of determinants of audit committee appointments. Section two of this paper contains a review of related literature and presents the testable propositions, the third section provides a description of the data and method, Section four presents and discusses the results, and the last section provides a summary and concluding remarks.

DETERMINANTS OF COMMITTEE APPOINTMENT

Early research on corporate boards focused on the importance of director affiliation as a significant determinant of a board's monitoring effectiveness. The empirical evidence generally suggests that nonexecutives perform an important monitoring role in specific situations where shareholder interests are at-risk as in the case of takeovers, poison pill adoptions, and management turnover decisions (e.g., Weisbach 1988; Byrd and Hickman 1992; Brickley et al. 1994; Cotter et al. 1997). Further research attempting to measure the importance of outside directors in enhancing performance in a more general, day-to-day sense has provided weak, less convincing evidence on their importance (e.g., Hermalin and Weisbach 1991), clouding the overall conclusions as to the optimal board make-up. In parallel, largely motivated by mixed results on the importance of director affiliation, prior research has suggested several other director attributes that may relate to a director's quality. Below, I discuss these director attributes and their likely relationship to audit committee appointment. Importantly, given the exploratory nature of this study, I do not posit prescriptive expectations on these director attributes. Thus, the relation between director attributes and committee appointment is addressed empirically.

Director Independence

An outside director who develops a fiduciary relationship with the firm is likely to be more dependent on management. Fiduciary relationships are often maintained by consultants, lawyers, and financiers, among others, who often receive compensation for services rendered to the firm other than in their capacity as outside directors. Such a fiduciary relationship may qualify the director's ability to monitor management's decisions in an unbiased manner (see Baysinger and Butler 1985; Byrd and Hickman 1992). Thus, the committee should consist of directors who are, and appear to be, independent of management's influence. Accordingly, subsequent tests in this paper empirically examine the relation between the likelihood of audit committee appointment and an outside director's level of independence.

Director Equity Holdings

Predictions on the value of director equity holdings are mixed. One view is that directors who develop shareholder-like interests become more likely to act to protect shareholder interests (e.g., Hermalin and Weisbach 1991, and Shivdasani 1993). Therefore, shareholder-members of the audit committee will be mindful of their obligation to other shareholders for high-quality financial reports (the incentive effect of ownership). For example, shareholder-members will be motivated to deter managers from manipulating earnings to maximize their compensation, or to manage earnings around equity offerings and share repurchases. In contrast, excessively large equity holdings by audit committee members may induce directors to side with management in manipulating the financial results at the expense of smaller shareholders (the entrenchment effect of ownership). In sum, one may argue for a nonmonotonic relation between the likelihood of an audit committee appointment and an outside director's equity investment in the firm, with the incentive effect prevailing for low ownership levels, and the entrenchment effect dominating thereafter.

Other Directorships Held

Directors with valuable reputation capital are less likely to collude with management because their human capital is at stake. Following prior work (Shivdasani 1993; Kaplan and Reishus 1990; Gilson 1990), the empirical proxy used to measure the value of reputation capital is the number of other directorships held. Shivdasani and Yermack (1999) argue that it is also possible for directors to spread their time commitments too thin by taking on too many directorships, thus compromising their ability to monitor management well. They suggest that directors should hold between one and three additional directorships. Drawing on this work, the value of a director's human capital, as proxied by the number of directorships held, is empirically linked to the likelihood of being appointed to the audit committee.

Length of Board Tenure

If committee service is seen as a vehicle for power in the organization, then seniority may be an important determinant of committee appointment. Further, given that effective monitoring of the financial reports requires a good understanding of the firm's affairs, the nature of its transactions, and management's attitude toward the financial-reporting process, more seasoned directors, and thus more knowledgeable directors, may be preferred for audit committee service. In contrast, if audit committee service is viewed as a burden, or as a task of less importance, new directors will be more likely to become appointed to the committee. Further, newer directors may also be preferred by opportunistic managers if they are less likely to voice their concerns over the firm's financial-reporting practices. Director seniority is approximated by the length of director tenure on the board.

Other Committee Memberships

Another concern about the board's operations is the equal division of labor among directors. Thus, an audit committee appointment is likely to depend on a director's other commitments in the firm. Given limited time availability for committee service, committee membership is likely to be distributed so that there is a just allocation of committee duties. Alternatively, directors who serve on more committees, and are thus more influential, may be more likely to acquire further power by additionally serving on the audit committee.

Compensation Committee Membership

The compensation committee is a major committee of the board of directors. Equitable division of labor and time constraints would suggest that outside directors serve on the audit committee or the compensation committee, but not on both. Alternatively, power considerations would suggest that powerful compensation committee members seek to acquire further power by joining the audit committee. Thus, the likelihood of audit committee appointment is linked to compensation committee membership.(1)

DATA AND METHOD

In order to identify new audit committee appointees, I examine the composition of audit committees from annual proxy statements. A newly appointed director is defined as an audit committee member who is not reported as serving on the audit committee in the previous two years' proxy statements. For this purpose I study the audit committees of two samples of firms over the period between 1994 and 1998: a sample that consists of 310 firms listed on the Fortune 500 list for 1994 (financial firms and utilities are excluded) and a random non-Fortune sample of 238 firms. In both cases, firms had to have at least four out of a maximum of five proxy statements on the EDGAR database between 1994 and 1998. The study of committee appointments longitudinally allows abstracting from endogeneity problems by identifying the specific year of an appointment and by comparing the appointee and control directors at that time. Simply comparing the characteristics of audit committee members to nonmembers irrespective of the appointment year would have confounded the determinants and consequences of committee membership.(2) I study Fortune and non-Fortune firms separately because Fortune firms face a different information and control environment from the smaller, non-Fortune firms.

To isolate "significant" events and to control for wider changes in board structures, firms making three or more new appointments over the sample period were excluded. The reason for this treatment is that multiple appointments are likely to be correlated with broader changes in governance as in the case of CEO replacements and takeovers. This screening criterion restricts applicability of the study's conclusions to firms with a fairly stable control environment, for which audit committee appointments are not discernibly dependent on other contemporaneous changes in board or management structure. Firms were also deleted if they did not add new audit committee members during the sample period. Given the rarity of insider appointments on the audit committee, the tests focus on the determinants of committee appointment among outside directors. The final sample comprises 262 new appointments, 170 appointments by Fortune firms and 92 appointments by non-Fortune firms. Because of the sampling scheme used, the appointments span the period between 1995-1998 (i.e., 1994 is a reference year).

The director variables employed in the empirical tests presented later in the paper are defined as follows. The number of directorships held in other corporations proxies for reputation capital and excludes posts in not-for-profit organizations, fund directorships, and trustee positions. Other committee memberships are defined as the sum of all standing (but not ad hoc) board committees in which a director serves, excluding the audit and compensation committees. A binary compensation committee membership variable is set to 1 for compensation committee members and 0 otherwise. Board tenure is the sum of all board service years. Equity investment is measured by the market value of the shares of common stock that a director holds. A binary "grey" variable captures director independence and is set to 1 if a director is a consultant, bank officer, lawyer, a relative of management, as well as any one that is described as having an economic connection to the firm in the proxy statement, and 0 otherwise. This definition of a "grey" director is stricter than the definition suggested by the Blue Ribbon Committee. The rationale for including all of these parties in the "grey" category is that these individuals have a greater potential for developing a fiduciary relationship to management compared to other outside directors even if such a relationship does not presently exist. This definition is consistent with that used by Baysinger and Butler (1985) and Byrd and Hickman (1992) among others.

To distinguish between appointees and control directors the study employs a standard matched-pairs design. Each new appointee is matched to the nonexecutive director in the same firm and for the same year that is closest to the appointee's age, resulting in 262 matched pairs of appointed and control directors.(3) A director's age and affiliation (inside vs. outside) are likely to be correlated with the director attributes studied here such as the length of board tenure, committee memberships, equity ownership, and the number of other directorships held. Matching appointees and control directors on age and affiliation, thus, attempts to minimize the spurious relations that may be observed between appointees and control directors in the absence of such controls. (For a detailed definition of each of the variables used in the tests see Tables 1 and 2.) In five cases the same control director was matched with two appointed directors because there were no other control director candidates. Excluding these cases did not substantively affect the results.
TABLE 1

Profile for 262 Directors Being Appointed to the Audit Committee
and a Control Group of Directors Matched by Age and Professional
Affiliation(a)

Panel A: Descriptives for the Full Sample of Directors(b)

Variable                                    n     Appointed

Director age                     Mean      262        55.98
                                 Median               56
Market value of shares owned     Mean      239     8,078
(in $thousands)                  Median               43.1
Tenure in years                  Mean      261         3.67
                                 Median                1
Number of other directorships    Mean      262         2.33
                                 Median                2
Other committee memberships      Mean      262         1.11
                                 Median                1
Compensation committee member    Mean      262         0.26
Grey director                    Mean      262         0.29

Panel B: Descriptive Statistics for 170 Committee Appointees and
Control Directors Serving in Fortune 500 Firms(b)

Variable                                    n     Appointed

Director age                     Mean      170        56.54
                                 Median               57
Market value of shares owned     Mean      149    11,850
(in $thousands)                  Median               45.2
Tenure in years                  Mean      170         3.89
                                 Median                1
Number of other directorships    Mean      170         2.79
                                 Median                2
Other committee memberships      Mean      170         1.20
                                 Median                1
Compensation committee member    Mean      170         0.21
Grey director                    Mean      170         0.24

Panel C: Descriptives Statistics for 92 Committee Appointees and
Control Directors from Non-Fortune 500 Firms(b)

          Variable                          n     Appointed

Director age                     Mean      92         54.92
                                 Median               54
Market value of shares owned     Mean      90      1,942
(in $thousands)                  Median               39.4
Tenure in years                  Mean      91          3.24
                                 Median                2
Number of other directorships    Mean      92          1.48
                                 Median                1
Other committee memberships      Mean      92          0.81
                                 Median                1
Compensation committee member    Mean      92          0.35
Grey director                    Mean      92          0.39

Panel A: Descriptives for the Full Sample of Directors(b)

Variable                         Control     Difference

Director age                        57.81        -1.839
                                    58
Market value of shares owned     1,276        6,802
(in $thousands)                     92.7
Tenure in years                      6.76        -3.09
                                     5
Number of other directorships        2.49        -0.16
                                     2
Other committee memberships          1.97        -0.86
                                     1
Compensation committee member        0.55         0.29
Grey director                        0.36        -0.07

Panel B: Descriptive Statistics for 170 Committee Appointees and
Control Directors Serving in Fortune 500 Firms(b)

Variable                         Control     Difference

Director age                        58.49         -1.95
                                    60
Market value of shares owned     1,415        10,435
(in $thousands)                    112.4
Tenure in years                      6.59         -2.70
                                     5
Number of other directorships        2.94         -0.15
                                     2
Other committee memberships          2.13         -0.93
                                     2
Compensation committee member        0.53         -0.32
Grey director                        0.26         -0.02

Panel C: Descriptives Statistics for 92 Committee Appointees and
Control Directors from Non-Fortune 500 Firms(b)

          Variable               Control     Difference

Director age                        56.56         -1.64
                                    56
Market value of shares owned     1,049         893.5
(in $thousands)                     62.8
Tenure in years                      7.06         -3.82
                                     4
Number of other directorships        1.66         -0.18
                                     1
Other committee memberships          1.63         -0.82
                                     1
Compensation committee member        0.59         -0.24
Grey director                        0.54         -0.15

Panel A: Descriptives for the Full Sample of Directors(b)

Variable                          t-value      Wilcoxon-z

Director age                     -3.70(***)    -3.41(***)

Market value of shares owned      0.87         -3.10(***)
(in $thousands)
Tenure in years                  -6.39(***)    -3.81(***)

Number of other directorships    -0.94         -1.10

Other committee memberships      -8.42(***)    -3.81(***)

Compensation committee member    -6.60(***)
Grey director                    -1.84

Panel B: Descriptive Statistics for 170 Committee Appointees and
Control Directors Serving in Fortune 500 Firms(b)

Variable                          t-value      Wilcoxon-z

Director age                     -3.60(***)    -3.82(***)

Market value of shares owned      0.29         -4.62(***)
(in $thousands)
Tenure in years                  -4.39(***)    -4.15(***)

Number of other directorships    -0.66         -0.87

Other committee memberships      -8.48(***)    -4.76(***)

Compensation committee member    -6.06(***)
Grey director                    -0.54

Panel C: Descriptives Statistics for 92 Committee Appointees and
Control Directors from Non-Fortune 500 Firms(b)

          Variable                t-value      Wilcoxon-z

Director age                     -1.63         -1.55

Market value of shares owned      0.40          0.43
(in $thousands)
Tenure in years                  -4.92(***)    -3.85(***)

Number of other directorships    -0.72         -0.76

Other committee memberships      -30.8(***)    -3.75(***)

Compensation committee member    -3.01(***)
Grey director                    -2.20(**)

(*), (**), (***) Significant at the 0.05, 0.01, and 0.01 level,
respectively.

(a) Any age differences found between audit committee appointees and the
control group relate to an inability to match perfectly and have no
other meaning.

(b) For each variable the first and second lines report the sample means
and medians. The right-hand columns report the paired t statistic and
the z-value for the s signed-ranks matched pairs test.
TABLE 2

Logit Regressions of the Probability that a Director Serves on the Audit
Committee

                                             Full Sample
Variables                        Coefficients           Wald Chi-Squared

Intercept                         1.230                       2.275
Director age                      0.001                       0.001
Number of other directorships    -0.004                       0.001
Other committee memberships      -0.441(***)                 11.986
Tenure in years                  -0.067(***)                 10.864
One if compensation committee    -0.762(***)                 11.173
 member, zero otherwise
Value of equity holdings         -0.003                       0.004
1 if grey, 0 otherwise           -0.416(*)                    3.514
Appointed directors                              238
Control directors                                238
-2-Log-likelihood                            78.329(***)
Pseudo-[R.sup.2]                                12.0%

                                             Fortune 500
Variables                        Coefficients           Wald Chi-Squared

Intercept                         1.812                       1.892
Director age                     -0.005                       0.063
Number of other directorships     0.018                       0.111
Other committee memberships      -0.577(***)                 13.805
Tenure in years                  -0.042(*)                    2.731
One if compensation committee    -0.910(***)                  9.737
 member, zero otherwise
Value of equity holdings         -0.030                       0.103
1 if grey, 0 otherwise           -0.281                       0.840
Appointed directors                              149
Control directors                                149
-2-Log-likelihood                            56.021(***)
Pseudo-[R.sup.2]                               13.8 %

                                             Non-Fortune 500
Variables                        Coefficients           Wald Chi-Squared

Intercept                         0.687                       0.303
Director age                      0.010                       0.256
Number of other directorships    -0.048                       0.302
Other committee memberships      -0.162                       0.403
Tenure in years                  -0.115(***)                  9.549
One if compensation committee    -0.694(*)                    2.859
 member, zero otherwise
Value of equity holdings          0.019                       0.030
1 if grey, 0 otherwise           -0.600(*)                    3.112
Appointed directors                              89
Control directors                                89
-2-Log-likelihood                            26.242(***)
Pseudo-[R.sup.2]                                10.6%

(*), (**), (***) Significant at the 0.10, 0.05, and 0.01 level,
respectively.


RESULTS

Table 1 presents results of univariate comparisons of director characteristics across the audit committee appointees and control directors. Panel A presents results from the full sample of firms, while Panels B and C present results for the sub-samples of Fortune and non-Fortune firms respectively. Statistical differences are assessed using the parametric paired t-test and the nonparametric Wilcoxon matched-pairs signed-ranks test. Focusing first on the full sample in Panel A, the median appointee is two years younger than the matched control director. (Age differences between appointees and control directors that are reported in Table 1 represent a failure to match perfectly and should not be used to draw inferences on the age of these directors). Further, the median appointed director owns $43,100 in company stock compared to $92,700 for the median control director. (The mean difference conversely suggests higher equity holdings for committee appointees because of a few large blockholders being appointed to the committee.) Further, appointees have fewer years of service on the board, serve in fewer other committees, and are less likely to serve on the compensation committee. Last, weakly consistent with expectations, "grey" directors who may develop a fiduciary relation to management are marginally less likely to serve on the committee than control directors (29 percent vs. 36 percent). Because univariate relations may be spurious, more conclusive evidence on the determinants of director appointments is provided through logistic regression tests described later in this section.(4)

Panels B and C of Table 1 present descriptive statistics for the subsamples of Fortune 500 and non-Fortune firm appointments respectively. Fortune 500 directors are somewhat older than non-Fortune directors, hold more equity investment in their respective firms, hold more other directorships, serve on more other committees most likely because of the higher use of committees in the larger Fortune boards, and are less likely to serve on the compensation committee probably because more outside directors are available for staffing each committee in Fortune firms. Finally, Fortune directors are less likely to have a "grey" status, probably because of the higher public scrutiny that is exerted on Fortune boards and on the greater degree of managerial entrenchment in the smaller, non-Fortune firms. The only notable difference across the two sub-samples is that in the Fortune sample, appointees and control directors are about as likely to have a "grey" status, while in non-Fortune firms "grey" directors are significantly less likely to be appointed to the audit committee.

Table 2 presents results of logistic regressions on the determinants of audit committee appointments. The model is estimated three times: first, using the full sample of 238 matched pairs of directors (23 pairs are eliminated because of insufficient data on the market value of equity holdings and one because of insufficient data on director tenure); second, by focusing on audit committee appointments by Fortune firms; and third, by focusing on audit committee appointments by non-Fortune firms.(5) Importantly, each of the three models is significant at the 0.01 level as indicated by its -2Log-likelihood ratio, suggesting that the proposed group of determinants of audit committee appointment is meaningful. Focusing first on the full sample of directors, the results suggest that other committee service reduces the likelihood of an audit committee appointment, as expected. Similarly, compensation committee service is inversely and significantly related to the likelihood of an audit committee appointment, reducing the probability of an appointment by 53.33 percent. (By taking the exponent of the variable coefficient [-0.762] = 0.4667 and subtracting 1 we get a -0.5333 or a 53.33 percent reduction in the likelihood of appointment). Interestingly, board tenure term is negative and statistically significant at p [is less than] 0.01. Finally, "grey" directors are less likely to be appointed to the audit committee as advocated by the Blue Ribbon Committee (1999) report (p [is less than] 0.06).(6,7)

An examination of the second model suggests that the results are substantially similar for the subsample of Fortune 500 appointments. First, appointees serve on fewer other committees. Second, committee appointments reduce the likelihood of serving on the compensation committee by 60 percent (by taking the exponent of the variable coefficient exp [-0.91] = 0.40 and subtracting 1 we get -0.60 or a 60 percent reduction in the likelihood of appointment). This reflects the fact Fortune 500 firms usually have the luxury of appointing different outside directors to the audit and compensation committees. Third, appointees are more likely to be appointed early in their board tenure. Director independence in terms of a fiduciary relationship is not, however, related to the likelihood of a committee appointment among Fortune firms. In contrast, among non-Fortune appointments the only significant determinants of appointment are the length of board tenure, compensation committee membership, and director affiliation. The results for non-Fortune firms may be explained by the fact non-Fortune firms have smaller boards with fewer committees, fewer outside directors, and thus fewer candidates for committee membership, making the appointment choice among outside directors less systematic than in the case of large, Fortune firms.(8)

DISCUSSION AND CONCLUSIONS

This study identifies empirical determinants of audit committee appointments by comparing the characteristics of 262 nonexecutive audit committee appointees to age-matched nonexecutive control directors between 1995 and 1998, drawn from 548 public firms. The results suggest that the value of common stock owned by directors and other directorship posts held are statistically unrelated to the likelihood of committee appointment. Also, there is only weak evidence that "grey" directors are less preferred for committee service. In contrast, the likelihood of being appointed to the audit committee is significantly lower for directors serving in many other board committees. Further, new directors are more likely to be appointed to the committee than seasoned directors, possibly because the audit committee is not considered that important, or because the audit committee is considered as a training ground for new directors. Together, this evidence is consistent with a "next-in-line" explanation, where seemingly less important factors such as the equal division of labor among directors and seniority considerations drive the audit committee appointment decision.

In sum, while director independence has received considerable attention by the SEC, the stock exchanges, and the press, potentially important determinants of audit committee effectiveness other than nonexecutive status have received rather limited attention. (For example, DeZoort and Salterio [2001] report that the independence and audit knowledge of audit committee members increase the likelihood that committee members will advocate a "substance" approach in an accounting policy dispute with management). Therefore, it may not be enough for audit committees to consist exclusively of outside directors; attention should be paid to other characteristics such as experience, training, and incentives.

A limitation of this study is that it does not provide direct empirical evidence that these qualities actually lead to improvements in financial reporting. Further, director qualities and committee appointment may have common causes, so that these qualities are associated with, but do not actually determine committee appointment. Nevertheless, this study highlights a potential explanation for the wide level of variation in financial-reporting quality across committees that are seemingly homogeneous in outside director participation. This conclusion points to the need for further research on finer director attributes as factors affecting the audit committee's ability to enhance the quality of financial reporting.

Several helpful comments and suggestions were provided by Arnie Wright (the editor) and three anonymous reviewers. The skillful research assistance of Elena Theodorou and Effie Chandriotou is acknowledged. This project was partly funded by the Institute of Certified Public Accountants of Cyprus and by the Cyprus Stock Exchange.

(1) Additionally, an anonymous referee has suggested that ethnic origin, race, sex, financial literacy, and religion could influence the director appointment process. Although these are potentially important factors, relevant data is not readily available in annual proxy statements, rendering their empirical examination very difficult. Therefore, this study is best seen as an exploratory examination of several, but not all, factors leading to committee appointment.

(2) For example, a positive association between audit committee membership and director tenure could have been interpreted (1) as evidence that more seasoned directors are more likely to be selected for audit committee service, or (2) that committee membership increases the probability that a director survives the director attrition process, and thus results in longer tenure lengths. The time-series approach adopted here helps to identify differentiating factors that precede the appointment.

(3) To address the possibility that committee appointees are not drawn from the sample of outside directors for each firm, but from among a much larger population of decision makers outside the firm that should be used as a benchmark, empirical tests separate directors who completed two or more years on the board from all others, aiming at a distinction between directors being appointed to the board so that they specifically serve on the committee, from directors being appointed to the committee after serving on the board for some time. The empirical results are similar in the two cases.

(4) To probe further into the importance of director affiliation, outside directors are partitioned into finer affiliation categories as follows: (1) executives of nonfinancial firms, (2) executives of financial firms, (3) consultants, (4) lawyers, (5) retired decision makers and professional directors, (6) public directors, and (7) private investors. The results suggest that 48.4 percent of the appointees are executives in industrial firms; categories 2-6 are represented fairly evenly; and category 7 is represented by 2.2 percent of all appointees. Most importantly, a [chi square] test discerns no difference among appointed and control directors for any affiliation subcategory.

(5) Although some of the Spearman and Pearson correlations between the director attributes are statistically significant, multicollinearity, as evidenced by the VIF, is generally not a problem here. When both director tenure and tenure squared are included in the model, their VIF are 6.14 and 6.68, respectively. No other VIF exceeds 1.28.

(6) To check whether multiple appointments are determined differently than single appointments, the sample was split into 140 firms with one committee appointment over the sample period and 49 firms (98 director pairs) with two appointments. The results are generally stronger for the sample having made only one appointment. For the second subsample, other committee memberships and tenure length are the significant appointment determinants, while compensation committee membership is not, possibly because firms lose the flexibility of assigning different directors to the audit and compensation committees as the number of required committee appointments rises.

(7) To examine whether these relationships remained stable over the sample period, the sample was split into 104 appointments made between 1995-96, and 134 appointments made between 1997-98 with qualitatively similar results across periods.

(8) The full model was re-estimated after introducing squared terms for the variables to examine whether the documented relations are monotonic. While the coefficients for ownership and directorships remain insignificant, the coefficient for tenure squared is positive and significant; the inflection point, however, is 16.6 tenure years, suggesting there is no reversal within a practical tenure range. Further, the coefficient for committee memberships squared is positive with an inflection point of 2.6 (i.e., the relation between the likelihood of appointment and the number of committee memberships turns from negative to positive for directors serving in more than 2.6 committees), suggesting that directors with high committee responsibilities are more likely to be appointed than directors with more moderate committee responsibilities.

REFERENCES

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

Beasley, M. 1996. An empirical analysis of the relation between the board of director composition and financial statement fraud. The Accounting Review 71: 443-465.

Blue Ribbon Committee. 1999. Report on improving the effectiveness of corporate audit committees. Business Lawyer 54: 1067-1096.

Brickley, J., L. Coles, and R. Terry. 1994. Outside directors and the adoption of poison pills. Journal of Financial Economics 35: 371-390.

Byrd, J., and A. Hickman. 1992. Do outside directors monitor managers? Evidence from tender offer bids. Journal of Financial Economics 32: 195-221.

Cadbury, A. 1995. Committee on the financial aspects of corporate governance: Compliance with the code of best practice. London, U.K.: Gee Publishing.

Cotter, J., A. Shivdasani, and M. Zenner. 1997. Do independent directors enhance target shareholder wealth during tender offers? Journal of Financial Economics 43: 195-218.

DeZoort, F. T., and S. Salterio. 2001. The effects of corporate governance experience and audit knowledge on audit committee members' judgments. AUDITING: A Journal of Practice & Theory (forthcoming).

Gilson, S. 1990. Bankruptcy, boards, banks, and blockholders: Evidence on changes on corporate ownership and control when firms default. Journal of Financial Economics 27: 355-387.

Hermalin, B., and M. Weisbach. 1991. The effects of board composition and direct incentives on firm performance. Financial Management 20:101-112.

Kaplan, S., and D. Reishus. 1990. Outside directorships and corporate performance. Journal of Financial Economics 27: 389-410.

Shivdasani, A. 1993. Board composition, ownership structure, and hostile takeovers. Journal of Accounting and Economics 16: 167-198.

--, and D. Yermack. 1999. CEO involvement in the selection of new board members: An empirical analysis. Journal of Finance 54: 1829-1853.

Vicknair, D., K. Hickman, and K. Carnes. 1993. A note on audit committee independence: Evidence from NYSE on "grey" area directors. Accounting Horizons 7: 53-57.

Weisbach, M. 1988. Outside directors and CEO turnover. Journal of Financial Economics 20: 431-460.

Submitted April 1999

Accepted September 2000

Nikos Vafeas is an Associate Professor at the University of Cyprus.
COPYRIGHT 2001 American Accounting Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Vafeas, Nikos
Publication:Auditing: A Journal of Practice & Theory
Article Type:Statistical Data Included
Geographic Code:1USA
Date:Mar 1, 2001
Words:5595
Previous Article:Bankruptcies, Audit Reports, and the Reform Act.
Next Article:The Impact of Electronic Commerce Assurance on Financial Analysts' Earnings Forecasts and Stock Price Estimates.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |