Corporate Governance und Unternehmenswert

Table of content

1) Introduction

2) Definition of “Corporate Governance” and “Corporate Valuation”

3) Statistical Problems

4) Corporate Governance and Corporate Valuation
4.1) Internal control mechanisms
4.1.1) Board composition and size
4.1.2) Insider ownership
4.1.3) Equity ownership structure
4.2) External control mechanism
4.2.1) Market for corporate control
4.2.3) Legal and Regulatory System
4.3) CG-Index

5) Concluding remarks

Tables

Figures

Literature

1) Introduction

In this paper I outline the relationship between Corporate Governance (CG) and the valuation of a firm. Working through the huge body of empirical research I want to give a clear picture of the mechanisms and their relative importance in influencing the valuation. At first I clearly define what I am speaking about: What is Corporate Governance? What do I mean when I speak about “Corporate Valuation”? Once the terms are clear the main question is: Why are these two things connected? What are the theoretical arguments? To evaluate the empirical findings I shortly describe beforehand the statistical problems and pitfalls that arise within this research. With this stable fundament I describe, comprehend and evaluate the descriptive research that has been done until today. Throughout this part I also pay attention to the used methodology. The aim is to give the reader a comprehensive and clear answer to the question how does Corporate Governance influence the Corporate Valuation.

2) Definition of “Corporate Governance” and “Corporate Valuation”

In this work I answer the question of CG’s influence on “Corporate Valuation”. I use research, which explicitly examines the influence on valuation. This is measured by the shareholder value or mainly by Tobin’s Q^[1], i.e. market value of shares plus book value of debt, altogether divided by the book value of assets. Loosely speaking this gives a good relative measurement of how high the investors value the firm. The argument is: the riskier the share, the higher the demanded return. The same argument holds for debt: it is more expensive for the firm, the more risk the debtor thinks there is. All this would show up in a lower Tobin’s Q (henceforth Q). But where does this risk comes from?

The research about CG started with the seminal paper of Jensen and Meckling in 1976. They developed a new “theory of the firm” (so the title of the work) based on the new framework of information asymmetry, which was most prominently developed by Stigler (1961) and Akerlof (1970). The great finding by Jensen and Meckling was the definition of “the concept of agency costs”^[2] [p.305]. There is an agency problem inside the firm due to the separation of ownership and control. They investigated “the nature of the agency costs” [p. 305] and “demonstrate who bears this costs and why” [p.305]. The agent (Manager) rationally doesn’t maximize the value of the firm and the principals (Share or Debt holders) try to minimize this loss - but they have to pay for this effort. CG mechanisms want to reduce this asymmetry and therefore deal with “how investors get the managers to give them back their money.”^[3] This is the loss the investors face and this is why a good CG – which is minimizing this asymmetry - should lead to a higher valuation.

The latest research is concerned with an additional asymmetry. Due to high ownership concentration in emerging and developing markets there is a conflict of interests between controlling shareholders and minority shareholders^[4]. The early literature examined just US CG which is characterized by disperse shareholding.

I do not address research about relationships that just indirectly (if at all) influence the relationship between CG and valuation (CG’s influence on Return on Assets [Table 1], on the dividend payout^[5] or on cash-holding^[6]). But I want to state that the evidence from this literature is in line with the findings of the following research.

3) Statistical Problems

Endogeneity is the biggest challenge in research on effects of CG: “[…] endogeneity […] presents both a curse and an important set of challenges to researchers in empirical corporate governance.”^[7] Endogeneity means that there is more interaction between the firm valuation (explained variable) and the explaining CG mechanisms (regressors). Econometrically speaking: The regressors are correlated with the error terms. The CG mechanisms influence firm valuation not only direct, but also via other attributes of firm valuation - like firm size^[8] - that are included in the error term, i.e. an interaction term influencing both variables.^[9] The other problem of endogeneity is reverse causality – when both variables are determined inside the model, the direction of causality remains unclear. In both cases the regressor is not exogenously given and as a result a standard OLS-Regression gives an inconsistent estimation.^[10]

Many papers are mainly concerned about the endogeneity between ownership structure and value (Table 2). The theoretical argument is that firms will adopt the ownership structure that is most appropriate - given the characteristics of the firm.^[11]

There are different ways how to deal with endogeneity: One is to find exogenous variables, which are highly correlated with the endogenous regressor^[12] (Instrumental Variables, two-stage least square estimation)^[13]. Another possibility is to develop a system of equations, in which each variable is once explained and in the other equations a regressor.^[14] A last possibility is to look at the reaction of firms during a truly exogenous event. Lemmon and Lins (2003) look at the relationship between ownership structure and valuation during the exogenous shock of the Asian Crisis. Bris and Cabolis (2004) look at firms that are bought in a cross-border merger – also an exogenous event - and have to adapt to the new regulatory CG environment.

A more obvious pitfall lies within the Q. The difference in Q between companies and industries is surely not only due to differences in CG. You can think about the valuation factors for a mature industry compared with the rapidly growing high-tech industry. Therefore researchers control for differences in valuation between industries or – highly correlated with this – control for growth opportunities and other variables like size, leverage or growth.

Throughout my further work I will pay attention to these and other statistical problems within the presented CG literature.

4) Corporate Governance and Corporate Valuation

CG mechanisms can be parted in two big categories: Internal control mechanisms and external control mechanisms. The third part is about the newest research that merges the different mechanisms in an index to check for overall CG’s influence on valuation.

4.1) Internal control mechanisms

The first and most appealing instruments to align the interests of the management with the interests of the shareholders lie within the companies themselves. Here I merge those mechanisms, which a company can organize, create or develop individually.

The research about the influence of single CG mechanisms on firm value was concerned mainly with US companies and later with companies in highly developed countries like Germany or Japan. Denis and McConnel (2002) call it the first generation of research. I describe the influence of the board of directors, of voting rights, of insider ownership and ownership structure on a corporate valuation.

4.1.1) Board composition and size

Hermalin and Weissbach (2003) state in their review of the board-literature that the board lays “at the heart of governance” [p. 7]. Their conclusion about the part of the literature that is concerned with influence of board composition on corporate value is clear: “Overall, there is little to suggest that board composition has any crosssectional relationship to firm performance” [p. 12]. In 1991 they found that result in their own study. Their concluding reasoning for this somewhat surprising result: “In particular, firm performance is a function of so many different factors that it is difficult to imagine that the effect of occasional board meetings, etc., would be detectable” [p.12].

Furthermore they review research about whether or not board size matters for valuation. They state and conclude that there is a negative relationship – the smaller the board, the higher the Q.

4.1.2) Insider ownership

There are two theoretical arguments about the effect of insider ownership on valuation. On the one hand it could have a positive effect because of equalizing the managers’ incentives with the interests of shareholders. On the other hand the insiders have an incentive to exercise private benefits of control and expropriate minority shareholder.

Mork et al. (1988) and McConnel and Servaes (1990) find that with low managerial ownership a rise helps aligning the interests and therefore positively influencing valuation. But a high managerial ownership works exactly the other way around. These studies do not control for endogeneity but research that does confirms the result. E.g. Beiner et al. (2004) find a positive relationship for Switzerland (They do not control for different magnitudes of ownership).

In 1991 Hermalin and Weisbach find a more detailed causality for US-firms: An insider ownership below 1% has a positive impact on Q. Furthermore: “The marginal effect of additional ownership at moderate levels of ownership (between one percent and 20%) on q seems to be negative at lower levels and positive at higher levels, […]” [p. 106]. Lins (2002) confirms that management control in emerging markets between 5% and 20% has a negative impact on Q as long as this group is the largest block-holder. (The causality is depicted in Figure 1) If there is a bigger non-management block holder, the impact becomes insignificant.

A strong negative effect can be found when management has control rights which exceed its cash flow rights. Lins (2002) proves this negative impact on Q.

4.1.3) Equity ownership structure

Jensen and Meckling’s influential theory of the firm assumes a widely dispersed firm as it is typical for the United States. Early CG literature discussed the effects of more concentrated ownership. Under surveillance were the more bank-centred economies in Germany and Japan. In Germany in 1996 e.g. 77% of the voting rights were controlled by large shareholders (holding more than 5%).^[15] There is mixed evidence which system performs better and has a greater impact on valuation.

A change in the literature came in 1985. As mentioned in the statistical part, Demsetz and Lehn (1985) argue that ownership structure is endogenously determined.^[16] Beside inconsistent estimations (when OLS is used) one dangerous thing about this is that it can lead to reverse causality.^[17] A wrong interpretation of the influence of ownership structure can lead to false management decisions. This is most likely a reason for the mixed evidence^[18].

[...]

^[1] Tobin (1969).

^[2] Jensen / Meckling (1976).

^[3] Shleifer / Vishny (1997). [p. 738]

^[4] Explained in Lefort / Walker (2005).

^[5] Lefort / Walker (2005).

^[6] Dittmar et al. (2003).

^[7] Denis (2001). [p. 198]

^[8] “[…] companies belonging to large-size industries perform better if they have stronger corporate governance than companies belonging to small-size industries”. Bruno / Claessens (2006). [p. 12]

^[9] Zimmermann (2004).

^[10] Heij et al. (2004). [p. 397 et seq.]

^[11] Demsetz / Lehn (1985).

^[12] Used e.g. in Black et al. (2005).

^[13] Explained in Greene (2003). [p.74 ff]

^[14] Used e.g. in Beiner et al. (2004).

^[15] A review for Germany can be found in: Boehmer (1999).

^[16] For further discussion and evidence see Cho (1998).

^[17] “I find that corporate value affects ownership structure but not the reverse, thereby reversing the interpretation of the relation between ownership structure and corporate value.”, Cho (1998). [p. 104]

^[18] For surveys see Jiraporn et al. (2006) or Denis / McConnel (2001).