5.1 A Graphical View of our Research Results
The first payment asymmetry index (PAS1) provides an intuitive sense of asymmetry.
However, PAS1 does not take capital gains and future investment opportunities into account. Our second payment asymmetry index (PAS2) is developed to address these concerns. A higher value of PAS1 or PAS2, which means higher board compensation together with lower dividend payouts, indicates a more severe payment asymmetry between the board and the shareholders. We use Figures 1 and 2 to present our preliminary research results.
Figure 1 plots PAS1 against B_SHARE. The payment asymmetry argument suggests
that smaller size of board tend to give higher board compensation and lower dividend payouts. Thus, PAS1 should be negatively related to B_SHARE. As shown in Figure 1, the bold line between PAS1 and B_SHARE is clearly negatively associated, supporting our argument of few-shares hypothesis.
[Insert Figure 1]
By means of a comparison between the slope of the dotted line and the solid line in Figure 1, we delineate the good governance hypothesis. Along with an increase in B_SHARE, firms with good governance (the lower dotted line) have a more negative regression slope, indicating a rapid reduction in PAS. By contrast, firms with poor governance (the upper dotted line) have a less negative slope, i.e., a slow reduction in PAS. In short, the negative relationship between B_SHARE and PAS will be greater (less) for firms with good (poor) governance structures.
The solid line shown in Figure 2 represents the few-shares hypothesis, as in Figure 1.
As regards the good governance effect, which does not follow exactly the line in Figure 1, the asymmetric governance hypothesis predicts different slopes as shown in Figure 2, where the slope is steeper in the low-share regime on the left-hand side of the figure. On the other hand, in the high-share regime on the right-hand side of the figure, the asymmetry is less severe, suggesting that good governance provides less incremental value in mitigating the PAS.
[Insert Figure 2]
In other words, our few shares hypothesis predicts that firms in the low-share regime will have a higher PAS, while firms in the high-share regime will have a lower PAS. The good governance hypothesis suggests that the negative effect on PAS of the board
holding few shares will be reduced for firms with better governance. However, the good governance effect is asymmetric, depending on which share regime a firm is in. It should be noted that, along with the good governance effect, there is the possibility that a poor governance effect also exists.23 Poor governance weakens the effect of the B_SHARE on the PAS, while good governance strengthens such an effect (as shown by a steeper slope in Figure 1).
Like the good governance effect, the poor governance effect is asymmetric regarding the negative effect of board members holding few shares. In particular, poor governance has an aggravating effect on the PAS for firms in the low-share regime, while such an effect is mitigated for firms in the high-share regime.
5.2 Basic Statistics
We choose the Taiwanese companies listed over the 1997-2005 period from the Taiwan Economic Journal Database (TEJ) as our research sample. The original number of observations is 10,306. We exclude firms in the banking and financial industries, firms with insufficient data with regard to the corporate governance variables, and firms that experienced a net loss and distributed no dividends, and thus we arrive at a final sample of 5,354 observations.
The descriptive statistics for the explanatory variables in the analysis are reported in Table 1. The average values for B_SHARE, BLOCK, and IND_SEAT are 29.78%, 15.34%, and 8.68%, respectively. With respect to the two deviation measures, the mean for the voting deviation (V_CF), 6.52%, is not large, compared to the 28.17% for the seat
23 In order to show our few-shares hypothesis and good governance hypothesis more clearly, we do not portray the poor governance effect in Figure 1. We do, however, show the asymmetric poor governance
deviation (ST_V). Around 53% of our observations belong to family-controlled companies. The average board size is 9.51 and the proportion of the observations for which the CEO is the chair of the board is 29%. The average percentage of shares held by the foreign institutional investors is 1.63%. The average seat representation ratio of executive board members is 13%. The average natural logarithm of firm size is 15.06.
[Insert Table 1]
Table 2 provides a preliminary mapping among the PAS1, dividend payout, compensation, and the B_SHARE. The first column classifies the PAS1 into 10 sub-groups in ascending order. The second column reports the corresponding yearly industry-median-centered dividend payout ratio, which exhibits a descending pattern. The third column reports the yearly industry-median-centered compensation, which appears in ascending order. The fourth column reports the percentage of shares held by board members. We observe that the higher the PAS1, the more severe the payment asymmetry is. That is to say, firms in the top group favor the outside shareholders the most (a less severe agency problem) and those in the bottom group favor the board members the most (a more severe agency problem).24 Finally, the percentage of shares held by board members decreases as the number of groups increase, suggesting that the fewer the shares held by board members, the more severe the agency problem is. The findings here are consistent with our few-shares hypothesis.
[Insert Table 2]
Following the same approach described above, we further classify the full sample into 100 subgroups. Figure 3 portrays the scatter plot for PAS1 and B_SHARE. As in our
findings in Table 2, our few-shares hypothesis still holds here.
[Insert Figure 3]
5.3 Empirical Results
Tables 3 and 4 report the estimated results using PAS1 and PAS2 as the respective dependent variables. In each table, we consider 13 model specifications to examine the validity of the few-shares hypothesis and the good governance hypothesis. In the first two specifications we consider only B_SHARE and/or its squared form as the explanatory variable. As to the following nine specifications, we incorporate governance variables one by one to show the specific effect of each. The last two columns report the results of the complete model by including all explanatory variables with and without the SIZE variable. Our final conclusions are based on the complete model (column 13), and the other specifications serve as benchmarks.
[insert Table 3]
Here we first explain the findings for PAS1 in Table 3, and then continue on to PAS2 below. For PAS1, the coefficients of B_SHARE are overwhelmingly negative and significant in all columns. Thus, the few -hares hypothesis is strongly supported. The results indicate that the fewer the shares held by board members, the more severe the payment asymmetry is, thereby indicating an aggravation of the board-shareholder agency problem. To investigate whether there is a second-order effect between PAS1 and B_SHARE, we also consider the squared form of B_SHARE. The negative relation between PAS1 and B_SHARE remains unchanged through all other columns. These conclusions are qualitatively the same in Table 4, which provides the estimated results
using PAS2 as the dependent variable. Therefore, consistent with our few-shares hypothesis, the board-shareholder agency problem increases as the number of shares held by board members decreases.
We next discuss the good governance hypothesis in Table 3, where PAS1 is examined.
If the good governance hypothesis is to be supported in the case of the governance variables that help mitigate the severity of PAS, we expect the coefficients of the interaction terms—the B_SHARE multiplied by governance variables—to be negative.
For governance variables that aggravate the severity of PAS, on the other hand, we expect the coefficients of the interaction terms—the B_SHARE multiplied by these governance variables—to be positive. That is, good (poor) governance variables mitigate (aggravate) the severity of payment asymmetry.
Our empirical results show that three governance variables that interact with B_SHARE, namely, FAMILY (coefficient -0.529, p-value < 1%), DUAL (coefficient -0.819, p-value < 1%), and MANG (coefficient -4.952, p-value < 1%), are significantly and negatively associated with PAS1. Therefore, we find that family-controlled companies have less severe payment asymmetry. In addition, DUAL and MANG are efficient governance mechanisms for mitigating the board-shareholder agency problem. It is worth noting that DUAL is typically thought to be harmful from the point of view of governance. However, in our case, a chair-CEO also represents the labor side, and the responsibilities of this role may reduce the tendency to allow an asymmetric payment.
As to the coefficient of SIZE (-19.088, p-value < 1%), the significantly negative result reveals that larger firms tend to avoid asymmetric payment due to their increased visibility and the resulting social pressure (the so-called “political cost effect”).
By contrast, three more governance variables that interact with B_SHARE, namely, ST_V, IND_SEAT and B_SIZE, are positively associated with PAS1. Interestingly, ST_V (coefficient 0.023, p-value < 1%) is significantly positive, while the traditional measure for entrenching controlling shareholders, V_CF (coefficient -0.001, p-value = 0.795), is insignificantly related to the severity of payment asymmetry. Therefore, in stark contrast to the insignificant effect of voting deviation, seat-control deviation does indeed explain the severity of payment asymmetry.
The significantly positive coefficients of B_SHARE×IND_SEAT (0.085, p-value < 1%) and B_SHARE×B_SIZE (0.116, p-value < 1%) deserve more explanation. Conventional wisdom has it that independent directors are expected to enhance the effectiveness of the board. Nevertheless, in Taiwan, the percentage of shares held by independent directors is less than one percent, so the finding here is in fact consistent with our few-shares hypothesis, which argues that monitors with small equity holdings have less incentive to perform their roles effectively. Thus, the effectiveness of independent directors with small equity holdings merits further study. As to B_SIZE, we find that the size of corporate boards is related to payment asymmetry. Finally, we find no significant effect of B_SHARE×INST on PAS1.
The analysis here leaves us room to re-examine the dual role of a CEO. Traditional wisdom has it that the dual role of a CEO has a negative effect on monitoring. However, a CEO who also has a role on the board will receive both an employee bonus (as an employee) and board remuneration (as a board member). The particular context of this study allows us to examine how the CEO who has a dual role is able to alleviate conflicts among stockholders, board members, and employees.
Most of the findings reported in Table 3 are qualitatively similar to those in Table 4, where PAS2 is examined. The only exception is that the explanatory power of B_SHARE
× FAMILY (coefficient -0.257, p-value < 0.167) is reduced to a one-tailed significance.
Overall, we obtain similar conclusions from Tables 3 and 4.
[insert Table 4]
Table 5 presents the estimated results when sample observations are divided into a low-share and a high-share regime based on the median of B_SHARE. Our asymmetric governance hypothesis leads us to expect that our previous findings on the good governance effect will be more conclusive in the low-share regime than in the high-share regime.
The left and right parts of Table 5 report the results of using PAS1 and PAS2, respectively. To facilitate the discussion, we provide the last column of Table 3 (Table 4) in the whole sample column of PAS1 (PAS2) in Table 5. Since PAS2 controls for capital gains and future investment opportunities, and also since the primary results are similar, we focus our discussion on PAS2. It is interesting to note that the coefficients of B_SHARE are negative for the whole sample (-3.067), the low-share subsample (-2.600) and the high-share subsample (-3.316). However, the significant results for B_SHARE can be found only in the whole sample (p-value<0.01) and the low-share subsample (p-value<0.10). The insignificant evidence on B_SHARE in the high-share subsample (p-value = 0.416) implies that a decrease in shares held by board members does not significantly increase the payment asymmetry when board members hold a sufficiently high level of shares. Alternatively, for firms in the low-share regime, a decrease in the shares held by board members will heighten the severity of payment asymmetry. The
findings for B_SHARE suggest that the good governance effect substantiates itself in the low-share regime only, which is consistent with our asymmetric governance hypothesis.
With respect to the relationship between other governance variables and PAS2, of the evidence reported in the whole sample, it is the observations in the low-share regime that provide the primary support for the good governance hypothesis. For example, the significantly positive effect of B_SHARE×ST_V in the whole sample (0.022, p-value
<0.01) and the low-share subsample (0.023, p-value <0.01) cannot be found in the high-share subsample (0.008, p-value=0.590). The same situation can also be found for B_SHARE×FAMILY (with one-tailed significance), B_SHARE×DUAL, and B_SHARE×MANG. The only exception is B_SHARE×BLOCK. The variables of interest B_SHARE×IND_SEAT and B_SHARE×B_SIZE, as well as the control variable SIZE, are found to be significant for the two sub-samples. In other words, the evidence partially supports our asymmetric governance hypothesis.25
6. Conclusion
On the heels of the OECD’s plea for corporate boards to be responsible for aligning key executive and board remuneration with the longer-term interests of their company and its shareholders (OECD 2004), this study examines how corporate governance affects the fairness of payments between board members and shareholders. By examining issues that pertain to payment asymmetry, this study contributes to the line of research on board effectiveness in the context of minority shareholders.
25 The findings for PAS1 are qualitatively similar to those for PAS2 except for B_SHARE×BLOCK and B_SHARE×V_CF. We cannot provide any explanation except that PAS2 is a better proxy for payment asymmetry. In addition, the coefficient of B_SHARE×FAMILY, which is one-tailed significant in the PAS2
Several major findings emerge from our analysis. First, using Taiwan data, we demonstrate that payment asymmetry indeed exists between board members and external shareholders. Next, we find compelling evidence to support our few-shares hypothesis;
that is, the percentage of shares held by board members is inversely associated with the severity of payment asymmetry.
Third, we provide evidence that the agency cost of the few-shares effect is reduced by good governance attributes, thereby supporting our good governance hypothesis. Fourth, we find partial evidence that both the few-shares effect and the good governance effect are asymmetric depending on the level of shares held by board members.
Finally, we go beyond traditional measures by providing a new measure of the deviation between ownership and control. In the payment asymmetry context, we find that the conventional measure (vote deviation) does not have the ability to explain payment asymmetry, while the new measure (seat-control deviation) has some explanatory power in regard to payment asymmetry.
Our study contributes to our understanding of the effectiveness of corporate boards.
We formally document the factors affecting payment asymmetry, which is one of the core governance principles underscored by the OECD (1999, 2004). The prior literature has mainly focused on how the board interacts with other agents (e.g., executives and auditors), while ignoring the board per se. This paper is unique in large measure because it investigates a situation in which the self-interest of the board predominates, with the consequence that the board’s behavior could become a source of dissension between board members and shareholders.
Some important implications concerning board effectiveness emerge from our findings, but, due to expected institutional differences across countries, caution should be taken before making any generalizations based on our conclusions. To cite a few examples, La Porta et al. (1999, 2000, 2002) document cross-country differences in legal institutions and investor protection, and Shleifer and Wolfenzon (2002) identify differences in investor protection and in equity markets. Moreover, there are reportedly differences with respect to earnings management (Leuz, Nanda and Wysocki 2003; Chih, Shen and Kang 2007) as well as disclosure incentives and their effects on the cost of capital (Francis, Khurana and Pereira 2005) around the world.
In light of such differences, it would be valuable in future research to re-examine issues surrounding payment asymmetry in a cross-country context. In addition, it would be equally enlightening to examine the economic consequences of payment asymmetry, such as the effects on the cost of capital and the impact on analyst ratings.
References
Abdullah, S. N. 2006. Directors’ remuneration, firm’s performance and corporate governance in Malaysia among distressed companies. Corporate Governance 6 (2), 162-174.
Ahmed, A., Duellman, S., 2007. Accounting conservatism and board of director characteristics:
An empirical analysis. Journal of Accounting and Economics 43, 411-437.
Almazan, A., Hartzell, J. C., Starks, L. T., 2005. Active institutional shareholders and costs of monitoring: Evidence from executive compensation. Financial Management 34 (4), 5-34.
Anderson, R. C., Mansi, S. A., Reeb, D. M., 2003. Founding family ownership and the agency cost of debt. Journal of Financial Economics 68, 263–285.
Ashbaugh-Skaife, H., Collins, D. W., LaFond, R., 2006. The effect of corporate governance on firms’ credit ratings. Journal of Accounting and Economics 42, 203-243.
Barrock, L. 2002. Big money: When directors earn but not shareholders. The Edge, January 15, available at:www.theedgedaily.com/article.cfm?id=10035
Bates, T. W., Lemmon, M. L., Linck, J. S., 2006. Shareholder wealth effects and bid negotiation in freeze-out deals: Are minority shareholders left out in the cold? Journal of Financial Economics 81 (3), 681-708.
Bhagat, S., Black, B., 2002.The non-correlation between board independence and long-term firm performance. Journal of Corporation Law 27 (2), 231-273.
Bhojraj, S., and Sengupta, P. 2003. Effect of corporate governance on bond ratings and yields:
The role of institutional investors and the outside directors. The Journal of Business 76:
455-475.
Borokhovich, K. A., Parrino, R., Trapani, T., 1996. Outside directors and CEO selection. Journal of Financial and Quantitative Analysis 31: 337-355.
Brickley, J. A., Coles, J. L., Terry, R.L., 1994. Outside directors and the adoption of poison pills.
Journal of Financial Economics 35, 371-390.
Brown, L. D., Caylor, M. L., 2006. Corporate governance and firm valuation. Journal of Accounting and Public Policy 25, 409-434.
Byard, D., Li, Y., Weintrop, J., 2006. Corporate governance and the quality of financial analysts’
information. Journal of Accounting and Public Policy 25, 609-625.
Byrd, J. W., Hickman, K. A., 1992. Do outside directors monitor managers? Evidence from tender offer bids. Journal of Financial Economics 32, 195-222.
Carcello, J.V., Neal, T. L., 2000. Audit committee composition and auditor reporting. The Accounting Review 75 (3), 453-467.
Carcello, J. V., Neal, T. L., 2003. Audit committee characteristics and auditor dismissals following “New” going–concern reports. The Accounting Review 78 (1), 95-117.
Chih, H., Shen, C. and Kang, F. 2007. Corporate social responsibility, investor protection, and earnings management: Some international evidence. Journal of Business Ethics (forthcoming).
Claessens, S., Djankov, S., Fan, J., Lang, L., 2002. Disentangling the incentive and entrenchment effects of large shareholdings. Journal of Finance 57 (6), 2741-2771.
Claessens, S., Djankov, S., Lang, L., 2000. The separation of ownership and control in East Asian corporations. Journal of Financial Economics 58 (1-2): 81-112.
Economics (forthcoming)
Conyon, M.J., 1997. Corporate governance and executive compensation. International Journal of Industrial Organization 15, 493-509.
Conyon, M.J., Peck, S. I., 1998. Board control, remuneration committee, and top management compensation. Academy of Management Journal 41 (2), 146-157.
Cotter, J., Shivdasani, A., Zenner, M., 1997. Do outside directors enhance target shareholder wealth during tender offer contests? Journal of Financial Economics 43, 195-218.
Dalton, D.R., Daily., C. M., 2001. Director stock compensation: An invitation to conspicuous conflict of interests? Business Ethics Quarterly 2 (1), 89-108.
Demsetz, H., Lehn, K., 1985. The structure of corporate ownership: Causes and consequences, Journal of Political Economy 93, 1155-1177.
Eisenberg, T., Sundgren, S., Wells, M. T., 1998. Larger board size and decreasing firm value in small firms. Journal of Financial Economics 48, 35-54.
Ellis, R. D., 1998. Equity derivatives, executive compensation, and agency costs. Houston Law Review 35, 399-451.
Elson, C., 1993. Executive overcompensation: A board-based solution. Boston College Law Review 34 (5), 937-996.
Fama, E. F., Jensen, M. C., 1983. Separation of ownership and control. Journal of Law and Economics 26, 301-325.
Fan, P. H., Wong, T. J., 2002. Corporate ownership structure and the informativeness of accounting earnings in East Asia. Journal of Accounting and Economics 33, 401-425.
Fan, P. H., Wong, T. J., 2002. Corporate ownership structure and the informativeness of accounting earnings in East Asia. Journal of Accounting and Economics 33, 401-425.