Do Reputational Capital Boards Enhance Corporate Reputation

International Research Journal of Finance and Economics ISSN 1450-2887 Issue 23 (2009) © EuroJournals Publishing, Inc. 2009 http://www.eurojournals.com/finance.htm Do Reputational Capital Boards Enhance Corporate Reputation? Hsiu-I Ting Department of Money and Banking, College of Finance and Banking National Kaohsiung First University of Science and Technology Kaohsiung, Taiwan (R.O.C.) E-mail: hiting@ccms.nkfust.edu.tw Tel: +886-7-6011000, Ext. 3126; Fax: +886-7-6011039 Abstract Firms with reputational capital boards - those in which a majority of outside directors hold three or more directorships - show a significantly positive impact on corporate reputation. Firms with a better reputation exhibit a higher market-to-book ratio, see more investment opportunities, and are larger in firm size. Simultaneous equations confirm that the reputation index and reputational capital boards are jointly endogenous. Moreover, firms with better technology applications and more civil responsibility are able to enhance outside directors’ prestige and visibility, giving them more opportunities to serve on other boards. This relation remains robust by different reputational capital measures. Finally, reputational capital exhibits an inverted-U shape as time goes by. Keywords: Corporate Governance, Reputational Capital, Corporate Reputation 1. Introduction Large U.S. manufacturing companies with better reputations for social responsibility outperformed companies with poorer reputations during the six-year period of 1982-1987, and even provided investors better stock market returns and lower risk. The implications of these findings for the information content of accounting systems in a social welfare context and for prescriptions in the business policy literature advocating a proactive social policy have been previously discussed (Herremans et al., 1993). There is a growing body of literature showing that serving on multiple boards can be a source of both valuable experience and reputational benefits for outside directors. Vafeas (1999) suggests that the number of directorships held by a director might proxy for reputational capital. However, there is comparatively little evidence of a discussion on the relation between corporate reputation and outside directors serving on multiple boards. The purpose of this paper is to examine whether there is an association between reputational capital and the level of corporate reputation, and whether outside directors possess more valuable experience and reputational benefits in firms with a better reputation. I find that there is a positive relation between reputational capital and corporate reputation. The average number of directorships per director and the percentage of directors holding three or more directorships are higher in firms with a better reputation. High-reputation firms perform better, as measured by market-to-book ratio, and they have more investment opportunities and are larger in firm size. Following Core et al. (1999), I adopt a dummy variable equal to one if 50% or more of the 90 International Research Journal of Finance and Economics - Issue 23 (2009) board’s outside directors individually hold three or more directorships in order to measure reputational capital. The results suggest that firms exhibit a better reputation when outside directors possess reputational capital. I find consistent evidence of a positive relation between reputational capital and corporate reputation by interaction terms and different reputational capital, performance, and firm size measures. The results imply that outside directors receive prestige by serving on multiple boards, thus helping their business to gain a better reputation. Given these results, it is possible that outsiders with reputational capital tend to be appointed to the boards of better reputation companies, because these directors are viewed as helpful in formulating strategies. Therefore, I develop simultaneous equation systems to provide one solution to resolve the problem. The finding confirms that reputational capital and corporate reputation are jointly endogenous. I also try another way to control for the potential endogeneity. By comparing the board’s reputational capital changing over time, I find that reputational capital exhibits an inverted-U shape as time goes by. The remainder of this paper is organized into five sections. Section 2 reviews the prior empirical literature on corporate reputation and reputational capital. The sample is described and the variables are defined in Section 3. In section 4, I document the association between the level of corporate reputation and reputational capital measures. Sensitivity tests are presented to determine the robustness of the results to alternative specifications. A conclusion is provided in Section 5. 2. Literature Review The importance of maintaining a positive corporate reputation, representing both corporate performance and social performance, has become an increasingly significant goal of company executives and boards of directors over the last decade (Hemphill, 2006). Dowling (2006) suggests that if a board of directors focuses on creating and maintaining better corporate reputations, then it will help that firm’s business to regain the high ground in the debate about the economic and social contributions of business. Fama and Jensen (1983) note that reputational effects can be important incentives for outside directors. Gilson (1990), Kaplan and Reishus (1990), and Vafeas (1999) suggest that the number of directorships held by a director might proxy for reputational capital, with such individuals viewed as high quality directors. Moreover, service on multiple boards can provide a director with a greater diversity of experience. Fama (1980) and Fama and Jensen (1983) argue that the market for outside directorships serves as an important source of incentives for outside directors to develop reputations as monitoring specialists. The relation between reputational capital and firm value has been examined in many prior empirical papers. Loderer and Peyer (2002) present a positive association between firm values, measured as Tobin’s Q, and the number of directorships a chairman holds for a sample of Swiss firms. Mace (1986) suggests that outside directorships are perceived to be valuable, because they provide executives with prestige, visibility, and commercial contracts. Ferris et al. (2003) find that the first appointment of a busy director to a board is good news for shareholders, implying that the enhanced experience or reputation of such directors is beneficial. The decision by an executive to accept a directorship can enhance shareholder value of the primary employer if the executive sitting on an outside board learns about different management styles or strategies used in other firms (Bacon and Brown (1974), Booth and Deli (1996), and Carpenter and Westphal (2001)). Kaplan and Reishus (1990), Gilson (1990), Shivdasani (1993), Brickley et al. (1999), and Ferris et al. (2003) all report evidence consistent with a positive relation between director quality and the number of directorships held by an executive. Some evidence suggests that multiple directors are associated with firm success. Miwa and Ramseyer (2000) find that the presence of directors holding multiple directorships was strongly related to firm performance in the cotton spinning industry in Japan during the first decade of the 20th century. Cotter et al. (1997) suggest that shareholders receive larger premiums in tender offers when the board International Research Journal of Finance and Economics - Issue 23 (2009) 91 has multiple directors. Finally, Brown and Maloney (1999) exhibit that firms enjoy superior returns from acquisitions when they have directors who hold multiple directorships. These studies suggest that multiple directorships are consistent with shareholder interests. Whether the results of these studies can be generalized to corporate reputation, however, is open to question. 3. Sample and Data 3.1. Measure of corporate reputation The main dependent variable of reputation is the combined score obtained from the annual CommonWealth magazine survey. The CommonWealth survey covers most industry groups. The survey asks executives, directors, and analysts in particular industries to rate a company on the following ten key attributes of reputation: (1) Vision; (2) Innovation; (3) Customer orientation; (4) Operational performance; (5) Financial soundness; (6) Ability to attract, develop, and keep talented people; (7) Technology application; (8) Multinational operation; (9) Long-term investment value; (10) Civil responsibility (presented by CR1 to CR10 respectively). Ratings are on a scale of 1 (poor) to 10 (excellent). The average score is perceived by the readers as well as the respondents of the survey as the reputation index. 3.2. Measure of reputational capital Support for the reputational capital view of directorships comes from several studies showing that the number of boards that outside directors sit on is tied to the performance of the firms in which these directors are incumbents, either as CEOs or as outside directors. Accordingly, several studies use the number of board seats held by an outside director as a proxy for the director’s reputation in the external labor market (Shivdasani (1993), Vafeas (1999), Brown and Maloney (1999)). I consider a board to have reputational capital if the majority of outside directors serve on three or more directorships. Although the three-directorship criterion is admittedly somewhat arbitrary, I choose this cutoff for several reasons. First, the third quartile of directorships in the sample is exactly three, resulting in a proper split between a board having and not having reputational capital. Second, it reflects the recommendation by the Council for Institutional Investors that directors should sit on no more than two boards. The corporate governance policies of the Council of Institutional Investors (CII (2003)) suggest that individuals with full-time jobs should not serve on more than two other boards and that a CEO should only serve as a director of one other company and should do so only if the CEO’s own company is in the top half of its peer group. Finally, my definition is consistent with prior work by Core et al. (1999), Ferris et al. (2003) and Fich and Shivdasani (2006) who also use the threedirectorship benchmark to classify executives. Following the definition of Core et al. (1999), I construct a (0, 1) indicator that takes the value of one if 50% or more of the board’s outside directors individually hold three or more directorships. Using this definition, 26.44% of boards in the sample are classified as a reputational capital board. Throughout the paper, I refer to this variable as the “reputational capital” indicator. Table 1 summarizes the list of variables. 92 International Research Journal of Finance and Economics - Issue 23 (2009) Table 1: Variable descriptions Variable CRA IR GR OSR DA D1 D3 SA S1 S3 OA OA2 O0 O1 O2 O2A O3 BO3 BO34 BO36 BOAM BO2A BOAQ1 BOAQ3 MB IO SIZE Description Reputation index, the average score of ten attributes of reputation Percentage of inside directors Percentage of gray directors Percentage of outside directors Average number of directorships per director Percentage of directors with one additional directorship Percentage of directors with three or more directorships Average number of directorships per supervisor Percentage of supervisors with one additional directorship Percentage of supervisors with three or more directorships Average number of directorships per outside director The square of OA, included to allow for nonlinearities Percentage of outside directors with no additional directorships Percentage of outside directors with one additional directorship Percentage of outside directors with two directorships Percentage of outside directors with two or more directorships Percentage of outside directors with three or more directorships A dummy variable equal to one if 50% or more of the board’s outside directors individually hold three or more directorships A dummy variable equal to one if 40% or more of the board’s outside directors individually hold three or more directorships A dummy variable equal to one if 60% or more of the board’s outside directors individually hold three or more directorships A dummy variable equal to one if the average number of directorships per outside director is higher than its mean (2.3464) A dummy variable equal to one if 50% or more of the board’s outside directors individually hold two or more directorships A dummy variable equal to one if the average number of directorships per outside director is lower than its first quartile (0.6667) A dummy variable equal to one if the average number of directorships per outside director is higher than its third quartile (3) Market-to-book ratio The ratio of depreciation expenses to sales The natural logarithm of sales 3.3. Sample The sample consists of 551 observations from 1996 to 2005 of the CommonWealth magazine survey. For each firm, I collect data on board variables from annual reports. As most board variables are not available before 2000, observations were reduced to 315 during the 5-year period from 2001 to 2005. Thereafter, 52 firms with insufficient data to describe board characters and another 55 firms with insufficient data to compute control variables (performance, investment opportunities, and size) were subtracted. These criteria yield a final sample of 208 observations of 81 firms across the 5 years. Descriptive statistics for key variables for the 208 observations 1 are presented in the first column of Table 2. I classify directors as insider, gray, or outsider in a manner consistent with Cotter et al. (1997). Full-time employees of the firm are designated as insiders and they account for the largest proportion of the board (38.61%). Directors associated with a company, former employees, those with existing family or commercial ties with the firm other than their directorship, or those with interlocking directorships with the CEO are designated as “gray” (37.43%). Directors that do not fit the description as an insider or gray directors are classified as outside directors. Outside directors only occupy 23.96%, which is significantly lower than the percentage of insider and gray (t-statistics are 5.955 and 4.801, respectively). On average, outside directors hold 2.35 1 Among 208 observations, one firm does not have any supervisors, and forty-five firms do not have outside directors. Therefore, only 207 and 163 observations, respectively, were adopted to compute the number of supervisors and outside directors serving on other boards International Research Journal of Finance and Economics - Issue 23 (2009) 93 additional directorships (the median is 1.5), and 30.98% of outside directors serve on more than three other boards. Table 2: Summary Statistics The sample consists of 208 annual observations of 81 firms between 2001 and 2005. Among 208 observations, one firm does not have any supervisors; forty-five firms do not have outside directors. Therefore, only 207 and 163 observations were adopted to compute the number of supervisors and outside directors serving on other boards. The corporate reputation index data were obtained from the annual CommonWealth magazine survey. Reputation index is the average score of ten key attributes: (1) Vision; (2) Innovation; (3) Customer oriented; (4) Operational performance; (5) Financial soundness; (6) Ability to attract, develop, and keep talented people; (7) Technology application; (8) Multinational operation; (9) Long-term investment value; (10) Civil responsibility. Ratings were on a scale of 1 (poor) to 10 (excellent). Board composition variables were collected from annual reports. I categorize high and low corporate reputations by the median of the average scores of the corporate reputation index (7.0890). The last column provides the difference of means tests between highreputation and low-reputation firms, and it indicates significance at the 1% (***) level, 5% (**) level, and 10% (*) level. All (N=208) Reputation index (CRA) Percentage of insider (IR) Percentage of gray (GR) Percentage of outsider (OSR) Average number of directorships per director (DA) Average number of directorships per supervisor (SA) Average number of directorships per outsider(OA) Percentage of directors with three or more directorships (D3) Percentage of supervisors with three or more directorships (S3) Percentage of outsiders with three or more directorships (O3) Percentage of directors with one additional directorship (D1) Percentage of supervisors with one additional directorship (S1) Percentage of outsiders with one additional directorship (O1) Market-to-book ratio (MB) Depreciation expenses/sales (IO) Log(sales) (Size) Mean 7.1647 0.3861 0.3743 0.2396 5.3727 3.5206 2.3464 0.4242 0.3199 0.3098 0.2346 0.3620 0.3550 0.0043 0.0556 0.3559 SD 0.4704 0.2402 0.2655 0.2017 7.3161 5.7524 2.4365 0.3328 0.3874 0.4024 0.2610 0.3786 0.4799 0.0049 0.0863 0.6477 High-reputation firms (N=104) Mean SD 7.5483 0.3428 0.3976 0.2484 0.3513 0.2766 0.2501 0.2055 7.0370 8.6275 4.7338 7.3995 2.6658 2.3171 0.4926 0.3321 0.3948 0.4008 0.3872 0.4123 0.2146 0.2953 0.3172 0.3642 0.2814 0.5159 0.0050 0.0061 0.0571 0.0964 0.4662 0.8299 Low-reputation firms (N=104) Mean SD 6.7811 0.1734 0.3746 0.2324 0.3974 0.2531 0.2290 0.1984 3.7084 5.2481 2.3191 3.0050 2.0231 2.5248 0.3558 0.3207 0.2457 0.3603 0.2315 0.3787 0.2546 0.2212 0.4064 0.3890 0.4295 0.4311 0.0036 0.0031 0.0541 0.0753 0.2457 0.3603 Difference t-statistics 20.3660*** 0.6892 -1.2547 0.7504 3.3614*** 3.0707*** 1.6935* 3.0221*** 2.8146*** 2.5104** -1.1080 -1.7037* -1.9890** 2.1519** 0.2559 2.4855** In the second and third columns of Table 2, I categorize high and low corporate reputations by the median of the average score of corporate reputation (7.0890). The average score of high and low reputations are respectively 7.55 and 6.78. Compared with firms with a poorer reputation, directors, supervisors, and outside directors all hold more directorships in firms with a better reputation. Moreover, compared with low-reputation firms, high-reputation firms show a higher percentage of directors, supervisors, and outsiders with three or more directorships, and a lower percentage with one directorship. Firms with a better reputation perform better on a market-to-book ratio, have more investment opportunities (depreciation expenses divided by sales), and are larger in firm size, which is measured by the natural logarithm of sales. These results suggest that outside directors have more directorships in firms with a better reputation. Outside directors possess more reputational capital as their company wins a higher reputation. 3.4. Comparing the number of directorships per outside director Firms with different levels of reputation index exhibit different numbers of directorships held by directors. In panel A of Table 3, I sort the sample into four groups. The first two groups, high- and lowreputation firms, are divided by the median of the average score of the corporate reputation index (7.0890). In the third and fourth columns, I categorize progressing and regressing firms by the corporate reputation rank. It is interesting to find that for low-reputation and regressing firms, the percentage of outside directors with one directorship is higher than that with three or more directorships. On the contrary, the percentage of outsiders with two directorships is lower than that with three or more directorships, and the relation is applied to four groups. In other words, when the 94 International Research Journal of Finance and Economics - Issue 23 (2009) firm is on the list of the reputation survey, outsiders tend to hold three or more directorships, not two directorships. However, I still find a slight difference between high- and low-reputation or progressing and regressing firms. For better reputation and progressing firms, the differences in the means between outsiders with three or more directorships and those with two directorships (0.2235 and 0.2516) are larger than those for poor reputation and regressing firms (0.1605 and 0.1785). The result presents that an outside director generally holds more directorships no matter what the levels are of the reputation index and ranking. International Research Journal of Finance and Economics - Issue 23 (2009) 95 Table 3: Board composition, reputation index, and reputation ranking comparisons Panel A compares the number of directorships per outside director. High- and low-reputation firms are separated by the median of the reputation index, 7.089. I sort progressing and regressing firms by their reputation ranking. The first row shows the difference mean between outside directors with no additional directorships and three or more directorships (“O0” minus “O3”). The differences of mean tests between these two groups, t-statistics, are in parentheses. The negative mean indicates that the percentage of outsiders with no additional directorships is higher than with three or more directorships. Panel B compares the level of reputation index and ranking of firms among boards with reputational capital and those without reputational capital. Following Core et al. (1999), I define reputational capital board as that in which a majority of outside directors hold three or more directorships, BO3. BO34, BO36, BOAM, BO2A, and BOAQ3 are alternative reputational capital measures. Among the sample, eighteen firms are on the reputation list every year. The number of the observations in those firms is 90. The first three rows of panel C compare the average number of directorships per director, supervisor, and outsider between firms on the list every year and not on the list every year. The next three columns compare the percentage of directors, supervisors, and outsiders with three or more directorships and the rest of the rows compare the reputational capital of outside directors between two groups. Variable definitions for the acronyms are given in Table 1. *** denotes significance at the 1% level; ** denotes significance at the 5% level; * denotes significance at the 10% level. Panel A Comparing the numbers of directorships held per outside director High-reputation firms Low-reputation firms Progressing firms Regressing firms Percentage of outsiders with no more -0.1830 (-2.5281)** 0.0613 (0.7967) -0.0791 (-0.7416) -0.0538 (-0.8461) directorships minus that with three or more directorships Percentage of outsiders with one -0.1057 (-1.2419) 0.1980 (2.5554)** -0.1594 (-1.7385)* 0.1421 (1.8726)* directorship minus that with three or more directorships Percentage of outsiders with two -0.2235 (-3.5797)*** -0.1605 (-3.2926)*** -0.2516 (-3.3298)*** -0.1795 (-3.8567)*** directorships minus that with three or more directorships Panel B Comparing the reputation index and reputation ranking between boards with and without reputational capital Reputation index Reputation ranking Mean Standard Deviation Mean Standard Deviation With reputational capital (50%)BO3 7.3967 0.5274 1.8182 0.9830 Without reputational capital 7.0813 0.4196 1.5948 0.8918 Difference (t-statistics) 4.002*** 1.481 With reputational capital (40%)BO34 7.3774 0.5284 1.7895 0.9772 Without reputational capital 7.0844 0.4212 1.6026 0.8950 Difference (t-statistics) 3.760*** 1.258 With reputational capital (60%)BO36 7.4352 0.5481 1.8049 0.9803 Without reputational capital 7.0982 0.4255 1.6168 0.9034 Difference (t-statistics) 3.674*** 1.118 With reputational capital (Mean)BOAM 7.3939 0.5295 1.8136 0.9733 Without reputational capital 7.0739 0.4128 1.5906 0.8930 Difference (t-statistics) 4.167*** 1.524 With reputational capital (50%)BO2A 7.3821 0.5232 1.7600 0.9563 Without reputational capital 7.0420 0.3893 1.5940 0.8964 Difference (t-statistics) 4.914*** 1.229 With reputational capital (75%)BOAQ3 7.3990 0.5279 1.8627 0.9802 Without reputational capital 7.0885 0.4249 1.5860 0.8919 Difference (t-statistics) 3.818*** 1.790* Panel C:Comparing board compositions On the list every year Not on the list every Difference year Mean SD Mean SD t-statistics Average number of directorships Director (DA) 6.8377 8.8221 4.2553 5.7084 2.4177** Supervisor (SA) 5.5674 8.0402 1.9768 2.0124 4.1169*** Outsider (OA) 2.8191 2.6363 1.9336 2.1799 2.3170** With three or more directorships Director (D3) 0.4857 0.3688 0.3772 0.2955 2.2867** Supervisor (S3) 0.4176 0.4287 0.2462 0.3365 3.1170*** Outsider (O3) 0.4002 0.4414 0.2308 0.3486 2.6914*** Reputational capital BO3 0.3667 0.4846 0.1864 0.3911 2.8838*** BO34 0.3667 0.4846 0.2034 0.4042 2.5836*** BO36 0.2889 0.4558 0.1271 0.3345 2.8348*** BOAM 0.4000 0.4926 0.1949 0.3978 3.2274*** BOAQ3 0.3778 0.4875 0.1441 0.3527 3.8447*** BO2A 0.4556 0.5008 0.2881 0.4548 2.4848** 96 International Research Journal of Finance and Economics - Issue 23 (2009) 3.5. Reputational capital comparison Panel B of Table 3 examines the difference of corporate reputation level between boards with and without reputational capital. I first adopt the definition of Core et al. (1999) to measure reputational capital. A dummy variable equals one if 50% or more of the board’s outside directors individually hold three or more directorships, BO3. It is obvious that when the majority of outside directors hold three or more directorships, firms achieve higher reputation scores (7.40 is higher than 7.08). Dividing the groups by 50% could be non-objective, and using 40% and 60% as benchmarks does not influence the results. I use the mean and the third quartile of the average number of directorships held by an outsider to construct the other two dummy variables, BOAM and BOAQ3, respectively. The results are consistent and still statistically significant. Fich and Shivdasani (2006) use the mean and median number of directorships in the sample to separate the groups. The mean and median numbers of directorships in my sample are respectively 2.3464 and 1.5000, both are close to 2. Following their rule, a dummy variable, BO2A, equals one if 50% or more of the board’s outside directors individually hold two or more directorships. The results show that firms gain a higher reputation index as the board possesses reputational capital, and the firm will sustain better corporate reputation. There is no similar finding for the reputation ranking variable. Except for defining reputational capital by the third quartile of the number of directorships held by an outside director, boards with and without reputational capital do not alter their reputation ranking. 3.6. Outstanding prestige Among all the firms in the sample, eighteen firms (22.22% of all firms) are on the reputation survey list every year. I am curious about the reputational capital of these excellent reputation firms. Panel C of Table 3 examines the difference in the number of directorships per outside director. Compared with firms that are not on the list every year, the average number of directorships per director and the percentage of directors with three or more directorships are both higher for firms with an excellent reputation. Outside directors of excellent reputation firms own superior reputational capital as well. The results confirm that outside directors gain a higher reputation when their firm holds outstanding prestige, and when they serve on more boards. 4. Reputational Capital and Corporate Reputation Before estimating the regression model, I calculate the correlation of board characters with the reputation index. The reputation index exhibits a positive correlation with the average number of directorships held by outside directors (0.235), reputational capital (0.296), and the percentage of outside directors with three or more directorships (0.351). The positive correlations are applied to directors and supervisors. The table is available upon request. 4.1. Multivariate analysis of corporate reputation I estimate the regression model using the reputation index as the dependent variable. The regressions control corporate governance and financial characteristics which are likely to affect firm reputation. I include the market-to-book ratio in the test. I control firm size using the natural log of sales as proxy. Board composition is controlled for by the percentages of inside, gray, and outside directors. The regressions include the ratio of depreciation expenditures to sales as a measure of the firm’s investment opportunity. The results of the multivariate models are reported in Table 4. In panel A, Model (1) shows that the coefficient for the average number of directorships per outside director is positive. In Model (2), the square of the average number of directorships per outside director is included to investigate the non-linearity relation with reputation. The coefficient for the average number of directorships per outside director remains positive, and the square term is negative and statistically significant at the 1% level. This recognizes the non-linearity relation with reputation. The results suggest that outside International Research Journal of Finance and Economics - Issue 23 (2009) 97 directors with different numbers of additional directorships have different impacts on reputation. The inflection point where the corporate reputation gains associated with the average number of directorships per outside director begin to taper off is at 4.84. Table 4: Regression of corporate reputation index on the reputational capital of outside directors The sample consists of 208 annual observations of 81 firms between 2001 and 2005. The dependent variable for all regressions is the corporate reputation index (CRA). The average number of directorships per outside director (OA) is used to capture the reputational capital of outside directors in Model (1). Considering the non-linear relation between reputational capital and corporate reputation, the square of OA is included in Model (2). Percentage of outside directors with one and percentage of outside directors with three or more directorships are adopted in Model (3). Two dummy variables, BOAQ1 and BOAQ3, are included in Model (4). BOAQ1 equals one if the average of directorships per outside director is lower than its first quartile, 0.6667. BOAQ3 equals one if the average of directorships per outside director is higher than its third quartile, 3. Fifty percent or more of outside directors individually hold three or more directorships, as with the definition of Core et al. (1999), is used to measure reputational capital in model (5). The parentheses contain t-statistics. Variable definitions for the acronyms are given in Table 1. *** denotes significance at the 1% level; ** denotes significance at the 5% level; * denotes significance at the 10% level. (1) (4.6812)*** (-0.4754) (-0.5333) (-0.5935) (2.3340)** (2) (5.1485)*** (-0.8511) (-0.8500) (-0.9785) (4.6243)*** (4.0470)*** (3) (4.8530)*** (-0.4370) (-0.5168) (-0.4764) (4) (5.7708)*** (-1.0638) (-1.1466) (-1.1755) (5) (5.7765)*** (-1.0561) (-1.1255) (-1.1678) Constant IR GR OSR OA OA2 O1 O3 BOAQ1 BOAQ3 BO3 MB IO SIZE Adjusted R2 F Number of observations 7.7533 -0.7953 -0.8817 -0.9751 0.0344 8.1488 -1.3633 -1.3440 -1.5392 0.1991 -0.0206 7.7282 -0.7034 -0.8216 -0.7531 8.5539 -1.5812 -1.6990 -1.7256 8.5303 -1.5658 -1.6636 -1.7117 -0.2141 (-2.2277)** 0.1929 (1.7723)* -0.0211 (-0.2724) 0.2459 (3.2961)*** 22.1787 (3.3952)*** 1.5103 (3.6540)*** 0.1168 (2.2012)** 15.0% 22.7% 5.093*** 163 17.7459 (2.8051)*** 1.7185 (4.3226)*** 0.0961 (1.8879)* 21.5% 17.4% 6.946*** 163 19.0391 (2.9695)*** 1.4666 (3.6908)*** 0.1621 (2.6662)*** 17.4% 6.530*** 163 0.2413 (3.3600)*** 22.7282 (3.5987)*** 21.1346 (3.3277)*** 1.2227 (3.4766)*** 1.2493 (3.5539)*** 0.1370 (2.8259)*** 0.1331 (2.7243)*** 6.462*** 208 7.210*** 163 I employ two measures for the number of directorships held by outsiders: the first variable is the percentage of outsiders with one additional directorship (O1) and the second one is the percentage of outsiders with three or more directorships (O3) in Model (3). These variables show different directions of the coefficients. Outside directors with one additional directorship will lower the reputation index, while they will enhance corporate reputation when they hold three or more directorships. The result is consistent with Table 3. Firms with a reputational capital board exhibit a higher reputation. Furthermore, I use the first and third quartiles of the average number of directorships per outsider, 0.6667 and 3 individually, to divide the groups. Here, BOAQ1 and BOAQ3 are dummy variables equal to one if the average number of directorships per outsider is lower than 0.6667 and higher than 3. The result is similar to model (3). I adopt reputational capital (BO3) in Model (5) and find a positive and significant coefficient on this variable as well. Therefore, all specifications indicate a positive and statistically significant relation between the reputational capital board and the reputation index. The estimate suggests that the impact of reputational capital on corporate reputation is economically non-trivial. The coefficient estimated in Model (5) indicates that a board with reputational capital enhances the reputation index by about 0.24. These results are consistent with Vafeas (1999). Outside directors holding more 98 International Research Journal of Finance and Economics - Issue 23 (2009) directorships are viewed as high quality directors. Serving on multiple boards could enhance directors’ experience and reputation. The signs of coefficients estimated for the control variables are similar with those reported by other studies. I also obtain a positive and statistically significant association between firm performance and reputation (Landgraf and Riahi-Belkaoui, 2003), and I find that investment opportunity is positively associated with reputation index. Reputation index is also positively related to firm size. Most effects of control variables are significant at the 1% level. Table 5: Robustness tests of the regression of the corporate reputation index The sample consists of 208 annual observations of 81 firms between 2001 and 2005. The dependent variable for all regressions is the corporate reputation index (CRA). Several alternative reputational capital measures, BO34, BO36, and BO2A, are adopted separately in Models (1), (2), and (3). Interaction variables, BO3×OSR and BO2A×OSR, are included individually in Models (4) and (5). The parentheses contain t-statistics. Variable definitions for the acronyms are given in Table 1. *** denotes significance at the 1% level; ** denotes significance at the 5% level; * denotes significance at the 10% level. 8.5055 -1.5323 -1.6398 -1.7021 0.2218 (1) (5.7364)*** (-1.0293) (-1.1050) (-1.1567) (3.1035)*** 8.5674 -1.5885 -1.7218 -1.7109 (2) (5.8124)*** (-1.0734) (-1.1670) (-1.1693) 9.1032 -2.1055 -2.2905 -2.3811 0.2994 (3) (6.2828)*** (-1.4474) (-1.5780) (-1.6515) (4.5478)*** 0.3747 -0.4555 21.766 (3.4245)*** 9 1.2399 (3.5140)*** 0.1371 (2.8030)*** 16.7% 6.929*** 208 22.6755 (3.6270)*** 1.1713 (3.3401)*** 0.1397 (2.8903)*** 17.6% 7.336*** 208 19.7482 1.1610 0.1251 (3.1808)*** 20.7567 1.2484 0.1292 (2.3646)*** (-0.9443) (3.2609)*** -0.5438 19.6058 1.1482 0.1159 (-1.4221) (3.1656)*** 8.5848 -1.6191 -1.7371 -1.7232 (4) (5.8074)*** (-1.0909) (-1.1733) (-1.1753) 9.0815 -2.0939 -2.2956 -2.2709 0.4593 (5) (6.2834)*** (-1.4430) (-1.5856) (-1.5768) (3.5274)*** Constant IR GR OSR BO34 BO36 BO2A BO3 BO3×OSR BO2A×OSR MB IO SIZE Adjusted R2 F Number of observations 0.2657 (3.4689)*** (3.3774)*** (2.6280)*** 20.9% 8.801*** 208 (3.5503)*** (2.6353)*** 17.3% 6.417*** 208 (3.3475)*** (2.4184)*** 21.3% 7.993*** 208 4.2. Robustness Tests 4.2.1. Characterizing reputational capital A board is defined as having reputational capital when 50% or more of its outside directors hold three or more directorships. I construct two (0, 1) indicators using 40% and 60% criterion and perform regressions in Table 5. The coefficients estimated for these independent variables are 0.2218 and 0.2657, respectively, and statistically significant at the 1% level. Following the definition of Fich and Shivdasani (2006), another dummy variable, BO2A, is included in Model (3). This estimate is stronger in magnitude than others. The situation of 50% or more of outside directors individually holding two or more directorships enhances the reputation index by about 0.30. 4.2.2. Interaction terms I examine if the marginal impact of reputational capital depends on the percentage of outside directors. Model (4) includes an interaction term of the percentage of outside directors and reputational capital indicator variable. The interaction term is negative, indicating that when there are more outside directors on the board, the impact of reputational capital on corporate reputation is slightly smaller, but not significant. Model (5) repeats the analysis of Model (4) using another reputational capital proxy (BO2A) and generates similar results. These tests continue to yield a positive and statistically significant association between corporate reputation and reputational capital measures. International Research Journal of Finance and Economics - Issue 23 (2009) 99 4.2.3. Performance and size proxies I repeat the analyses presented in Tables 4 and 5 using different proxies for performance, replacing the Smith and Watts (1992) market-to-book ratio calculation by ROA, EPS, and ROE. These tests also yield a positive association between reputational capital and corporate reputation. My results continue to be robust to different constructions of the control variable. Instead of the natural log of sales, I use both natural log of assets and the natural log of capital. These different constructions of the control variable do not qualitatively alter the results. Tables are available upon request. 4.3. The simultaneous approach Considering inferences based on a single-equation approach could be problematic. Both reputational capital and corporate reputation might be endogenously determined. If reputational capital is endogenous, then corporate reputation and the random error υ may be correlated, resulting in inconsistent estimates in Table 4. While the preceding results support the view that outside directors with reputational capital are associated with higher corporate reputation, it is possible that outsider directors with reputational capital are viewed as helpful in formulating strategies. They tend to be appointed to the board of better reputation companies. Therefore, I develop simultaneous equation systems to provide one solution to resolve the problem. Table 6 presents simultaneous equations using 2SLS method. Following Kelejian (1971), I use all predetermined variables, their square terms, and all distinct cross-products as instruments. The first column of Model (1) reports regression results of the average number of directorships per outside director equation. Some attributes of reputation show statistically significant impacts. Outside directors serving on firms with better technology applications and more civil responsibility are considered as having greater diversity of experience. This provides outside directors with more chances to serve on other boards, because they possess prestige, visibility, and commercial contracts. Column two of Model (1) shows that the average number of directorships per outside director still carries a positive sign and is statistically significant at the 1% level. Most importantly, the square term is no longer significant. It suggests that the reputation index increases by 0.39 for every one more directorship increase per outside director. This result shows consistency with the conclusion presented in the preceding section. Taking both equations together, I therefore conclude that the reputation index and reputational capital of outside director are jointly endogenous. Corporate reputation shows impacts on the reputational capital board, while on the other hand better reputational capital of outside directors inspires higher corporate reputation. The other models of Table 6 present the results of additional regressions to test the robustness of the results of Model (1). I use different proxies for reputational capital, replacing the average number of directorships per outside director by the percentage of outside director with two or more directorships (O2A) and with three or more directorships (O3) as proxies. These examinations still yield a positive coefficient of reputational capital. The signs of reputation attributes remain the same. These different constructions of the reputational capital measures do not qualitatively alter the results. In conclusion, the findings in this section suggest that corporate reputation affects reputational capital, which in turn affects corporate reputation, thereby reversing the interpretation of the results from OLS regressions. These findings also imply that ignoring the endogeneity of reputational capital affects the results of the corporate reputation regressions reported in Table 4. 100 Table 6: International Research Journal of Finance and Economics - Issue 23 (2009) Simultaneous equation A simultaneous equation analysis of reputational capital and corporate reputation for 208 observations uses the two-stage least squares method to estimate the following equations: Reputational capital = f (Corporate reputation, Firm performance, Firm size), Corporate reputation = g (Board composition, Reputational capital, Firm performance, Investment opportunity, Firm size) Reputation index (CRA) is the average score of ten key attributes of reputation: (1) Vision; (2) Innovation; (3) Customer orientation; (4) Operational performance; (5) Financial soundness; (6) Ability to attract, develop, and keep talented people; (7) Technology application; (8) Multinational operation; (9) Long-term investment value; (10) Civil responsibility (presented by CR1 to CR10, respectively). Following Kelejian (1971), I use all predetermined variables, their square terms, and all distinct cross-products as instruments. Reputational capital is measured by OA, O2A, and O3 separately in Models (1), (2), and (3). EPS is used to measure firm performance. The parentheses contain t-statistics. Variable definitions for the acronyms are given in Table 1. *** denotes significance at the 1% level; ** denotes significance at the 5% level; * denotes significance at the 10% level. (1) (2) -2.8933 0.1413 0.0557 0.2742 -0.1818 -0.1589 -0.2998 0.4668 0.0811 0.0003 0.0813 O2A (-5.7820)*** (0.7490) (0.3070) (1.9570)* (-1.0060) (-1.1960) (-2.1130)** (2.5230)** (0.7790) (0.0020) (0.8590) CRA 5.8387 (4.8210)*** -1.9896 0.0480 -0.0133 0.1358 -0.0416 -0.1960 -0.3520 0.4363 0.0974 0.0494 0.1565 O3 (-4.0830)*** (0.2610) (-0.0750) (0.9950) (-0.2370) (-1.5140) (-2.5480)* (2.4220)* (0.9600) (0.3060) (1.6970)* (3) 5.5216 CRA (4.1350)*** OA (-2.2600)** (0.2520) (0.8240) (0.4860) (-0.8690) (-2.0960)** (-1.9830)** (1.9650)* (-0.0200) (0.9330) (1.7290)* CRA 5.1830 (3.1290)*** Constant CR1 CR2 CR3 CR4 CR5 CR6 CR7 CR8 CR9 CR10 IR GR OSR OA OA2 O2A O3 EPS IO SIZE Adjusted R2 F -6.8967 0.2904 0.9133 0.4148 -0.9575 -1.6987 -1.7152 2.2161 -0.0128 0.9443 0.9982 1.2516 0.9550 0.8226 0.3919 -0.0033 0.0142 (0.2740) 0.0242 0.7558 -0.1427 (0.7630) (0.5930) (0.5140) (4.6210)*** (-1.6320) 0.0117 (1.3790) 0.0574 (1.3020) 29.67% 6.6591*** 0.7732 (0.6370) 0.6646 (0.5550) 0.6469 (0.5420) 1.4549 (6.9380)*** 1.1481 1.0753 1.0170 (0.8630) (0.8190) (0.7750) 0.5106 (1.9010)* 15.36% 3.4354*** (3.1790)*** (1.6100)* (-1.8140)* 24.81% 7.6420*** 0.0061 (0.9110) 0.4217 (1.1390) -0.0845 (-1.6880)* 40.31% 16.5328*** 0.0151 (1.8260)* 0.0754 (1.7570)* 20.72% 4.5059*** 1.7312 0.0068 0.8464 -0.1080 (5.0010)*** (0.8650) (2.1890)** (-1.6640)* 32.30% 11.9734*** International Research Journal of Finance and Economics - Issue 23 (2009) Table 7: 101 Board changes over time Panel A compares the reputational capital of outside director between t-1 and t (the year when the firm is on the reputation list). It presents the means of the difference of reputational capital measures at times t-1 and t. Negative means indicate that the reputational capital of outside directors is lower at time t-1 than at t. Panel B compares the means of reputational capital measures of outside directors between t+1 and t. It shows the mean difference of reputational capital measures at times t+1 and t. Negative means indicate that the reputational capital of outside directors is lower at time t+1 than at t. The percentage of fewer directorships held by outside directors is included in Panel B. Positive means represent that an outside director holds fewer directorships at time t+1 than at t. The parentheses contain t-statistics. Variable definitions for the acronyms are given in Table 1. *** denotes significance at the 1% level; ** denotes significance at the 5% level; * denotes significance at the 10% level. ALL High-reputation Progressing firms Regressing firms firms Panel A Comparing a board’s reputational capital between t-1 and t -0.1154 (-3.1238)*** -0.0809 (-2.7305)** -0.0938 (-2.1828)*** -0.5057 (-2.4840)** -0.3213 (-2.3581)** -0.4926 (-1.8737)* -0.0784 (-2.0332)** -0.0697 (-2.6657)*** -0.0250 (-0.5384)* Panel B Comparing a board’s reputational capital between t+1 and t -0.1827 (-2.9341)*** -0.1029 (-2.1902)** -0.2500 (-3.5496)*** 1.1538 (1.3823) 1.7197 (2.2147)** -0.0439 (-0.0389) 0.5648 (2.4564)** 0.4632 (2.6670)*** 0.7022 (2.5098)*** 0.7538 (2.1245)** 0.3633 (1.7356)* 1.0041 (1.7818)* Low-reputation firms -0.0577 -0.2419 -0.0437 -0.0865 1.0769 0.5879 0.4129 (-1.9218)* (-1.9505)* (-1.8095)* (-1.9000)* (1.1830) (3.3110)*** (1.4085)* BO3 -0.0865 OA -0.3748 O3 -0.0611 BO3 -0.1346 OA 1.1173 O1 0.5758 O0 0.5920 (-3.6321)*** (-3.1309)*** (-2.6846)*** (-3.4853)*** (1.8242)* (3.9287)*** (2.5508)** 4.4. Board changes over time Another way to control for the potential endogeneity mentioned earlier, I compare the board’s reputational capital changing over time. I use the full sample and high-reputation, progressing, regressing, and low-reputation firms separately in five columns. Panel A of Table 7 presents the means and t-statistics for the variation of reputational capital from t-1 to t. It indicates that outside directors possess higher reputational capital and serve on more boards at time t (when their firms are on the reputation list). The percentage of them with three or more directorships is also higher when their firms are on the list. The results are robust under all circumstances. However, the difference between time t-1 and time t is larger for high-reputation firms. It is perceived that as the firm wins a high reputation, it helps outside directors to gain more reputational capital as well. I then subtract reputational capital at time t from time t+1. Panel B still presents that the reputational capital of board changes over time. Reputational capital reduces significantly at time t+1, after the firms are on the list. When firms are regressing, reputational capital will diminish dramatically. Outside directors move towards holding fewer directorships or even not serving on any additional board, especially for regressing firms. The results of Table 7 indicate that reputational capital exhibits an inverted-U shape as time goes by. 5. Conclusion The reputational capital of outside directors shows a significantly positive impact on corporate reputation, and including the interaction terms does not alter the results. Using alternative proxies of the control variables also does not influence the relation between reputational capital and corporate reputation. Considering the potential endogeneity, I construct simultaneous equations and compare the board’s reputation capital over time. Simultaneous equations confirm that corporate reputation and the reputational capital of outside director are jointly endogenous. Firms with better technology applications and more civil responsibility are able to enhance outside directors’ prestige and visibility, and give them more opportunities to serve on other boards. The relation continues to be robust under different reputational capital measures. The reputational capital of board does change over time, as it exhibits an inverted-U shape as time goes by. I find that the directorships held by outside directors are increasing and then decreasing during the reputation survey period. 102 International Research Journal of Finance and Economics - Issue 23 (2009) Bacon, J. and J. Brown, 1974. “Corporate Directorship Practice: Role Selection, and Legal Status of the Board”, A Joint Research Report from the Conference Board and the American Society of Corporate Secretaries, Inc. (New York, NY). Booth, J.R. and D.N. Deli, 1996. “Factors Affecting the Number of Outside Directorships Held by CEOs”, Journal of Financial Economics. 40, pp. 81-104. Brickley, J., J. Linck, and J. Coles, 1999. “What Happens to CEOs after They Retire? New Evidence on Career Concerns, Horizon Problems, and CEO Incentives”, Journal of Financial Economics 52, pp. 341-377. Brown, W. and M. Maloney, 1999. “Exit, Voice and the Role of Corporate Directors: Evidence from Acquisition Performance”, Working paper, University of Virginia. Carpenter, M.A. and J.D. Westphal, 2001. “The Strategic Context of External Network Ties: Examining the Impact of Director Appointments on Board Involvement in Strategic Decision Making”, Academy of Management Journal 44, pp. 639-660. Core, J., R. Holthausen, and D. Larcker, 1999. “Corporate Governance, Chief Executive Officer Compensation, and Firm Performance”, Journal of Financial Economics 51, pp. 371406. Cotter, J., Anil S., and M. Zenner, 1997. “Do Independent Directors Enhance Target Shareholder Wealth during Tender Offers?” Journal of Financial Economics 43, pp. 195-218. Council of Institutional Investors, 2003. “Council of Institutional Investors’ Corporate Governance Policies”. Dowling, G., 2006. “Reputation Risk: It Is the Board's Ultimate Responsibility”, The Journal of Business Strategy 27 (2), pp. 59-68. Fama, E., 1980. “Agency Problems and the Theory of the Firm”, Journal of Political Economy 88, pp. 288-307. Fama, E. and M. Jensen, 1983. “Separation of Ownership and Control”, Journal of Law and Economics 26, pp. 301-325. Ferris, S., M. Jagannathan, and A. Pritchard, 2003. “Too Busy to Mind the Business? Monitoring by Directors with Multiple Board Appointment”, Journal of Finance 58 (3), pp. 1087-1111. Fich, E. M. and A. Shivdasani, 2006. “Are Busy Boards Effective Monitors?” Journal of Finance 61 (2), pp. 689-724. Gilson, S., 1990. “Bankruptcy, Boards, Banks and Blockholders: Evidence on Changes on Corporate Ownership and Control when Firms Default”, Journal of Financial Economics 27, pp. 355-387. Hemphill, T. A., 2006. “Corporate Internal Investigations: Balancing Firm Social Reputation with Board Fiduciary Responsibility”, Corporate Governance 6 (5), pp. 635-642. Herremans, I. M., P. Akathaporn, and M. McInnes, 1993. “An Investigation of Corporate Social Responsibility Reputation and Economic Performance”, Accounting, Organizations and Society 18 (7-8), pp. 587-604. Kaplan, S. and D. Reishus, 1990. “Outside Directorships and Corporate Performance”, Journal of Financial Economics 27, pp. 389-410. Kelejian, H., 1971. “Two-Stage Least Square and Economic Systems Linear in Parameters but Non-Linear in the Endogenous Variables”, Journal of the American Statistical Association 66, pp. 373-374. Landgraf, E. and A. Riahi-Belkaoui, 2003. “Corporate Disclosure Quality and Corporate Reputation”, Review of Accounting and Finance 2 (1), pp. 86-95. Loderer, C., and U. C. Peyer, 2002. “Board Overlap, Seat Accumulation, and Share Prices, European Financial Management 8, pp. 165-192. Mace, M., 1986. “Directors: Myth and Reality”, Harvard University Press, Boston, MA. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] International Research Journal of Finance and Economics - Issue 23 (2009) [22] [23] [24] [25] 103 Miwa, Y. and J. M. Ramseyer, 2000. “Corporate Governance in Transitional Economics: Lessons from the Pre-War Japanese Cotton Textile Industry”, Journal of Legal Studies 29, pp. 171-204. Shivdasani, A., 1993. “Board Composition, Ownership Structure and Hostile Takeovers”, Journal of Accounting and Economics 16, pp. 167-198. Smith, C. and R. Watts, 1992. “The Investment Opportunity Set and Corporate Financing, Dividend and Compensation Policies”, Journal of Financial Economics 40, pp. 263-292. Vafeas, N., 1999. “Board Meeting Frequency and Firm Performance”, Journal of Financial Economics 53, pp. 113-142.