Governance Role in Payment Asymmetry

In Taiwan, a proposal for the distribution of annual earnings is put forward by the corporate board. In the proposal, the corporate board suggests the amount of earnings to be allocated and also the fraction of that allocation that will make up bonuses for employees and board members and dividends for shareholders. As staff compensation committees are not a standard feature of the boardrooms of Taiwanese companies, board members set their own compensation, as specified in the firms’ Articles of Incorporation.¹² The proposal will be ratified by the shareholders at the annual meeting, but, due to the prevalent application of proxy votes and weak shareholder activism, the corporate board encounters hardly any difficulty in getting ratification from shareholders regarding their compensation. The distribution of earnings is generally a discretionary decision made by the board. This is made more controversial by the fact that the board members are also the setters of the company’s dividend policy. In other words, because board members are simultaneously the setters and receivers of both compensation and dividends, they face a tradeoff between board compensation and dividends for themselves.

members, and Article 3 stipulates the conditions that disqualify would-be candidates.

12 Some firms specify a ratio that is a fixed number or a minimum or maximum ratio to distributable

Previous studies have demonstrated that corporate governance features related to ownership and board structure are important elements of an examination of agency issues involving shareholders (e.g., Gompers et al. 2003; Bates et al. 2006). Good governance mechanisms can help reduce, if not prevent, hidden actions aimed at procuring board members’ private benefits rather than benefits shared by external shareholders. As the distribution of earnings is a decision made by the board members, board control is by far the most important factor. Conflict of interest arises between the board’s compensation and the dividend payout, with the latter dependent on the level of the board members’

ownership. We focus on shareholdings by the board members and other governance attributes related to ownership and board structure in order to define board control.¹³ To be specific, we argue that board members with a small number of shares tend to favor more generous board compensation, which—in the eyes of external shareholders—constitutes unfair payment. By contrast, board members with a significant proportion of shares are less likely to permit such unfair payment, given that they can receive a considerable amount in dividends.

3.1 The Effect of Shareholdings by Board Members on Payment Asymmetry

Two competing arguments, the entrenchment effect and the alignment effect, are conventionally adopted in studies that explore the effects of ownership on agency problems (Shleifer and Vishny 1997). The entrenchment effect argues that large inside shareholders in firms with concentrated ownership have greater incentives to maximize their own gains at the cost of other shareholders. Against this, the alignment effect

13 Ittner, Larcker and Rajan (1997) used (1) shares held by the board members and shares held by the CEO, and (2) the ratio of the board members who are “hired” by the CEO to independent board members to serve as a proxy for the power of the CEO. We adopt these concepts and apply them to describe the power of the

contends that large inside shareholders monitor management more thoroughly than minority shareholders, and that this careful monitoring carries potential benefits for all shareholders. There is convincing evidence for both effects: for the entrenchment effect, see, for example, Fama and Jensen 1983; Morck, Shleifer, and Vishny 1988; Claessens et al. 2002, and for the alignment effect, see, for example, Demsetz and Lehn 1985.

This study adopts the argument of the alignment effect to explain the relationship between board ownership structure and payment asymmetry. As a general rule, the principal and the agent are distinct roles, but in our setting, the role of the board and that of shareholders could overlap a great deal depending on the number of shares held by board members. On the one hand, if fewer stocks are held by board members, the agency problem in the form of payment asymmetry would be more severe, and the reason for this is simple: board members would have a greater incentive to maximize their own compensation rather than increase the dividends of shareholders. On the other hand, if board members are also large shareholders, it is expected that such behavior would be less likely and, when it does exist, less severe. Since firm value and dividends are more important to board members with a large amount of shares, overpayment to board members will be detrimental to firm value. We maintain, therefore, that the alignment effect is much more applicable than the entrenchment effect in the present study.

Though not exactly using the same setting as ours, Elson (1993) showed that firms with compensation committee members that have high equity ownership are less likely to overcompensate corporate executives; this conclusion is in line with the argument that board members with greater equity investment in the firm develop shareholder-like interests, which reduce the possibility and severity of asymmetric payments. Based on the

alignment effect, we predict that the greater the number of shares held by the board, the less severe will be the payment asymmetry. For convenience, we refer to the effect of board stock ownership on the severity of payment asymmetry as the “few-shares”

hypothesis.

3.2. The Effect of Governance on Payment Asymmetry

Prior studies on corporate governance have tended to focus on a single governance attribute effect, e.g., board size or board independence, intended to protect claims to firms’ resources. As some governance attributes may complement each other whereas others may serve as substitutes, we incorporate other governance attributes into our study to investigate how payment asymmetry is affected by various governance mechanisms other than shareholdings by the board members that are intended to control agency conflicts between management and shareholders.

We hypothesize that good governance helps reduce the agency problem resulting from board members having few shares. By holding the level of shareholdings by board members constant, we predict that good governance can lessen agency problems by reducing the severity of payment asymmetry. For example, Shen and Chang (2005) find that the negative effect of restrictions on banking activities engaged in securities, insurance and real estate is reduced by sound governance.¹⁴ The severity of payment asymmetry, on the other hand, is expected to increase when governance is poor. We hypothesize that, depending upon the level of shares held by the board members, governance attributes intended to mitigate agency conflict would help to reduce the

14 Shen and Lee (2006) also find that governance could lower the unfavorable effect of low financial

severity of payment asymmetry. We refer to this hypothesis as the “good governance”

hypothesis.

3.3 The Asymmetry of the Governance Effect

Since there is a competing view in the literature that suggests that concentrated ownership allows controlling shareholders to exercise undue influence to secure benefits that are detrimental to minority shareholders (Shleifer and Vishny 1997), there is the possibility that the effect of good governance may be conditional upon the level of concentration of shares held by the board members. We expect that the effect of good governance on the severity of payment asymmetry is stronger for firms in the low-share regime, as compared to firms in the high-share regime.¹⁵ In cases where the small amount of shares held by the board members significantly affects the severity of payment asymmetry, we further argue that good governance should provide more value to firms in a less desirable situation (e.g., the low-share regime in this study). In other words, good governance mitigates the negative effect of low shareholdings by board members asymmetrically. We hypothesize that the good governance effect is subject to change depending on the level of shares held by the board members. We refer to this as the “asymmetric governance”

hypothesis.