Print Reference This Reddit This

Review About Corporate Governance Problems And Solutions Finance Essay

Corporate governance as a topic of interest in academic literature dates back to the work of Berle and Means (1932) and till recently, efforts are being made to understand and minimize conflicts that could affect firms performance. The problem that arises in trying to separate ownership (the shareholders) from control (managers) leads to the concept of corporate governance. A most welcomed view of corporate governance is that given by Shleifer and Vishny (1997) as “ways in which suppliers of finance to corporations assure themselves of getting a return on their investment”. It has been argued that such view is constricted and other stakeholders such as the consumers, suppliers etc are interested in the way the firm is run. A more accepted view was given by Denis and McConell (2003) where corporate governance is “a set of both institutional and market based mechanisms that persuades the self‐interested controllers of a company (who make decisions regarding how the company is operated) to make decisions that maximize the value of the company to its owners (the suppliers of capital)”.

2.2 Corporate Governance Problems and Solutions

Shareholders are interested in increasing the value of their shares and getting vale for money but since they have limited expertise to do so, they employ managers. The managers main interest are normally slightly different from that of the shareholders though they may not be against value maximization and may decide to pursue their interest ahead of the interest of the shareholder termed the value maximization of the firm.

There are three major agency conflicts stated by Denis (2001) that may arise between shareholders and their managers which could affect firm performance. First, is the managers desire to remain in power where the manager needs to be changed to increase the value maximization goal either because they lack the know-how or underperformance is seen. Another aspect of serious conflict is, managers are generally referred to as risk averse while shareholders are known as risk takers knowing that shareholders own diversified portfolios and managers have most of their capital tied to one firm and has more to loose something goes wrong. Thirdly, what is being done with the excess money stands as a major agency conflict, whether he pays it out to shareholders and lenders or holds on to it.

2.2.1 Monitoring Managers/Executives:

In an attempt to increase firm performance agency conflicts have to be reduced, different solutions are proposed to reduce such conflicts of which one is monitoring. Monitoring on behalf of shareholders could be internal or external. It is done internally by the board of directors who possess adequate expertise to confirm if a decision is value maximizing or not. Also, some executives on the board are termed ‘busy’ because they serve on multiple boards has been reported to affect firms value. An external form of monitoring is done by large shareholders who have quite a substantial amount of the firm than the average shareholder and could benefit (loose) more if the firms value is appreciating (depreciating). I review below previous literature considering the size of board, ownership by institutional investors, the presence of a busy director and the effect shown to have on the performance of the firm. Board of Directors and Firm Performance:

The Board of directors is an important determinant of governance in firms and has been argued to reduce agency conflict and affect a firms performance when effectively managed. Some prior quantitative research believes firms with large boards perform worse than those with smaller boards; an inverse relationship exists between board size and a measurement of performance, Jensen (1993), Yermack (1996), Hermalin and Weisbach (2003), and Lipton and Lorsch (1992). Reason being that as the board size increases, there is an increased problem in coordination, communication and decision making leading to more agency conflicts between owners and controllers. Having an efficient board size reduces such conflicts. Jensen (1993) believes that an efficient board size is six while Lipton and Lorsch (1992) approve an adequate board size between eight and ten. Eisenberg et al (1998) also finds a similar result of a smaller board size being more effective but believes that the appropriate size varies across small and medium Finnish firms as compared to large firms.

Contrary, some other studies find other existing relationships; Ferris et al (2002) found a positive relationship between firm performance and board for US firms. Bhagat and Black (2002) find no robust association between firm performance and board size, also that firms with independent board perform worse than other firms. Their reason for a different result was attached to the sensitivity of the regression results to the control for endogeneity or to whether one uses a simultaneous equation approach or an OLS. Ownership by Institutional Investors and Firm Performance:

An associated set of findings have focused on the impact of equity holdings by institutional investors or blockholders have on a firms performance , Shleifer and Vishny (1997) believe that shareholders who hold a significant amount of equity in a particular firm have an important position to play in the governance system of a firm. Institutional investors have an increased interest in activities of corporate governance by introducing proposals put forward mainly when firms are performing poorly such institutional investors aim to improve performance, Morck et al 1988, Karpoff et al 1996 but whether they increase the firm value is the issue. Shleifer and Vishny (1986) believe that large blockholders are better able to put a check on management and help enhance the firms value than the average shareholder whose monitoring will be costly than the envisaged benefit and equity holdings by Institutional investors positively impacts firms market value but sometimes tends to takeovers. Holderness and Sheehan (1985) looked at six large blockholders termed as ‘corporate raiders’ and found shareholdings in firms targeted by ‘corporate raiders’ had statistically improved returns upon announcement because they improve management by monitoring and they are well informed enough to recognize underpriced stocks. Similar studies that find excess returns upon announcement of an institutional investor are Mikkelson and Ruback (1985) and Barclay and Holderness (1991) however they argue that restructuring of the corporation or takeover needs to quickly follow. McConnell and Servaes (1990) report a positive significant relationship between block ownership and firm value but their result was not robust to the inclusion of an alternative measure of performance. In Contrast, there has also been evidence of institutional investor having little or no effect on firm performance. Karpoff et al (1996) finds little evidence proposals by institutional investor prompt improved performance by observing no improvement in the share value of firms. There has rather been little research looking at the direct impact of institutional activism on firm performance bearing in mind that such effect is not easily observable because it has an economic positive impact but may not be easily seen due to noisy factors that may affect performance, Black (1998). Also, most studies fail to consider and state the degree to which blockholders are active or passive and do not consider several different alternatives for measuring ownership by institutional investors. Busy Directors and Firm Performance

The number of boards an executive gets to sit on also effects firm performance. Having directors that serve on multiple boards have reputational effect on the directors, assist in providing quality monitoring and signal the worth of the firm (see Fama and Jensen (1983) and Kaplan and Reishus (1990)). Ferris et al (2003) builds on Fama and Jensen’s (1983) work and they make use a cross-sectional model, they argue that limitations should not be placed on the number of board a director can sit on because serving on multiple boards is positively correlated with firm performance. On the other hand Bhagat and Black (1999) believe the reverse, arguing that better performing firms tend to attract directors that serve on several boards. Some other studies believe that the presence of directors on multiple boards leads to more agency conflicts and adversely affects firm performance knowing they are too busy to effectively monitor the business of several firms. Core et al (1999) argued that outside busy directors get higher CEO compensation which in effect leads to weaker performance by firms. Agrawal and Knoeber (1996) consider the endogenous relationship that exists among governance mechanisms and finds a negative relationship between firm performance and the number of board directors that are outsiders which to them is puzzling, they give a rationale that an addition of an outside director could be due to political reasons. Perry and Peyer (2005) looked at the announcement effect and reports that accepting the appointment of being a director depends on the agency problems that currently exists, if agency problem is critical then a busy director negatively affects performance but if fewer agency problems exists then appointing a busy director enhances firm performance . Fich and Shivdasani (2006) use panel data regression and observed directors that function in at least three boards are associated with weaker corporate governance lower market-to-book and profitability. They also found that busy directors are less likely to be chosen again following poor performance of the firm, their study is robust to the control of potential endogeneity of busy directors with firm performance.

2.2.2 Aligning Managers/Executives:

Another way to reduce agency conflict and encouraging managers to be more interested in increasing the value of the firm is aligning managers by providing incentives in the form of compensations and equity ownership. I consider equity ownership by executives and how it impacts firms performance. Firm Performance and Ownership by executives

A substantial amount of corporate governance literature looks at ownership by executives and the effect it has on the firms value. With the belief that managers can use corporate asset for their own benefit rather than make value maximization their priority. This is known as the Alignment effect proposed by Jensen and Meckling (1976) put forward as a solution for agency problem implying that managerial ownership will persuade managers to focus on value maximization knowing they tend to loose if the firm does not perform well. A contrary view to alignment is the entrenchment effect proposed by Shleifer and Vishny (1989) where they argue that as the equity stake of an executive in a firm increases, it becomes difficult to replace them and they have more power to run the firm. Most studies try to look at the relationship between firm performance and executive shareholding; they find that the existing relationship includes both the alignment and entrenchment effect according to the percentage of executive ownership.

Morck et al (1988) were among the first to look at this relationship, they consider 371 Fortune 500 US firms, estimating using a piecewise OLS regression and find that as insider ownership rises to 5% Tobin’s Q increases, when it further rises to 25% Tobin’s Q declines and when insider ownership is above 25%, Tobin’s Q increases slightly. They controlled for several control variables but their analysis was not robust when profit rate was used as an alternative measure for performance. McConnell and Servaes (1990) consider blockholder ownership and insider ownership for different sample from 1986 to 1976 using a larger sample size; they find a positively significant but diminishing relationship for insider ownership with an inflection point between 0.40 and 0.50. Hermalin and Weisbach (1988) look at the effect of managerial ownership has on accounting measurement of performance and finds no cross-sectional relationship. Some other studies that followed Morck et al (1988) work are Loderer and Martin (1997), Cho (1998) and Holderness et al (1999). The studies discussed above treat managerial ownership as an exogenous variable and did not control for the unobservable. Demsetz and Lehn (1985) control for endogeneity using both Tobin’s Q and accounting profit rate to measure performance, they find a non- monotonic relationship, but that managerial ownership is influenced by how volatile the firm is measured by the volatility of stock prices. Himmelberg et al (1999) extend on Demsetz and Lehn’s (1985) work by including more explanatory variables. They use panel data and instrumental variable as an alternative to test for endogeneity of managerial ownership in the performance regression and could not conclude that managerial ownership has a significant impact on performance. Zhou (2001) argues that it is not necessarily that managerial ownership has no effect on firms performance but that fixed effect may not detect such relationship because managerial ownership changes yearly within a firm and differs across firms.

2.2.3 Financial Structure

According to the capital structure theorem, the value of a firm is affected by the capital structure it adapts and the way a firm is financed can either reduce or intensify the agency conflict. Easterbrook (1984) argues that a way to reduce agency conflict that arises between owners and managers is to pay out dividend with the awareness that it re-aligns managers to maximize the value of the firm. On the other hand, Jensen (1986) believes that leverage is a better way to reduce agency conflict which affects firms performance knowing that managers may or may not pay dividend but are obliged to pay back their lenders. Another aspect is the Modigliani-Miller theorem that says in a perfect market, however a firm is financed either by debt or dividend is irrelevant in the firms value but in reality their theory seems to be quite unreasonable because the capital structure a firm embraces affects is firm value. Agrawal and Knoeber (1996) believe that debt as a means of finance is a form of monitoring by lender that affects a firms value.

2.3 Firm Performance and Governance Index:

A new area of governance is seen in Gompers et al (2003), Bebchuk et al (2009) and Brown and Caylor (2006). They believe that a single governance characteristic cannot effectively measure corporate governance and develop different governance indices argued to better explain the effect corporate governance has on firm performance.

Gompers et al (2003) came up with an index that relates corporate governance to firm valuation in the United States developed, they analysed the relationship between governance and equity returns. They constructed a governance index popularly known as the G-Index based on 24 equal-weighted unique corporate governance factors based on surveys conducted by Investor Responsibility Research Centre (hence known as IRRC); these corporate-governance factors are divided into 5 groups: voting rights, director/officers protection, delaying hostile bidders, state laws and other takeover defences. They hypothesize that well-governed firms perform better than weak governed firms. The highest index number represents the largest number of amendments while the lowest index number shows the smallest antitakeover amendment. Firms with stronger shareholder rights (well-governed firms) have higher firm value; higher equity returns and better operating performance. Lower G-index numbers reflect stronger shareholder rights indicating better performance; a negative correlation between G-index and firm value, therefore they predict a positive relationship between good governance and firm performance.

Bebchuk et al (2009) argued of a better index than that developed by Gompers et al, their index is known as the E-index also developed from the provisions of IRRC. They believe some of the 24 governance factors considered by Gompers et al are positively correlated with firm value and weakly explain the relationship between the governance provision by IRRC and performance. They introduce the entrenchment index with six governance factors - staggered boards, limits to shareholder amendments of the bylaws, supermajority requirements for mergers, supermajority requirements for charter amendments, poison pills and golden parachute arrangements. The first 4 categories limit shareholders voting power and the latter two are relevant measures taken in preparation for a hostile takeover. After controlling for firm fixed effect they observed that increases in the E-index was associated with decreases in the firms value and argue it to be negatively correlated to firm value therefore, the higher the level of entrenchment by executives (less- governed firms), the lower the firms performance.

Brown and Caylor (2006) initiate their own governance index known as (Gov-score) based on provisions by Institutional Shareholders Services (ISS). They argue that it has an advantage over G-index and E-index because it includes important governance measures that IRRC which was used to compute the g-index and e-index does not take into consideration and is based on internal and external governance factors. They construct a fifty-one governance factor based on eight governance categories they are audit (consulting fees paid to auditors are less than audit fees paid to auditors), board of directors (controlled by 50% independent outside directors), charter/bylaws (company has a poison pill that was shareholder approved or has no pill), director education (at least a member of the board participated in ISS-accredited director education programme), executive and director compensation (Directors receive all or a fraction of their fees in stock), Ownership (All directors with more than one year of service own stock), Progressive practices (Mandatory retirement age for directors exists) and State of incorporation (Firm is incorporated in a state without any anti-takeover). Brown and Caylor (2006) argue that Gov- score is better linked to firm performance and being more dynamic. An expected positive relationship is to be observed between Gov-score and firm performance.

Brown and Caylor (2008) repeat their study after observing that only a minimal subset of provision is actually related to firm performance based on Brown and Caylor (2006). Brown and Caylor (2008) look at the relationship between corporate governance and operating performance measured as Return on Asset and Return on Equity. They find a significant positive relationship between Gov-score and firm performance but only about ten factors are positively related with either Return on Asset or Return on Equity and motivation for using two accounting measurement was not stressed. Brown and Caylor (2006, 2008) are not consistent with the relationship between Gov-score and firm performance. Also Bebchuk et al (2009) believes that Brown and Caylor work consists of an enormous amount of governance factors which will affect the quality of the index.

Bhagat and Bolton (2008) look at the relationship that exists between corporate governance and performance; they look at alternative measures of corporate governance and consider the endogeneity problem. They find that better governance measured as g-index, e-index is significant and positively related to better operating performance but not with future stock market performance, no significant relationship was found with Gov-score. Their results are robust to a series of alternatives estimation of equations, diagnostic test and the estimation of the standard error.

The relationship between corporate governance characters and firm performance has been argued to be associated with the problem of endogeneity whereby there is a reverse causality between firm performance and corporate governance. Some studies have considered this endogenous relationship ( see Demsetz and Villalonga, 2001, Fich and Shivdasani (2006), Hermalin and Weisbach (2003)). Researchers find either no significant relationship where one has been argued to exist or the wrong sign/effect. Even though endogeneity problems in empirical research are not serious but they are important (Denis, 2001). Several studies try to solve this problem with the use of panel data to treat joint endogeneity, the variable of concern is lagged or the future year of the dependent variable is considered. Some researchers use simultaneous equation when more than one variable is considered endogenous.

Need help with your literature review?

Our qualified researchers are here to help. Click on the button below to find out more:

Literature Review Service

Related Content

In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: