Does Imposing Limitations on Rights of Shareholders Influence Company Value and Stock Price?

(1)

M

ASTER

T

HESIS

D

OES IMPOSING LIMITATIONS ON RIGHTS OF

SHAREHOLDERS INFLUENCE COMPANY VALUE

AND STOCK PRICE

?

Student name: Agnieszka Bartosiak

Student number: 11373156

Programe: MSc in Finance

Specialization: Corporate Finance

Supervisor: dhr. Prof. dr. A.W.A. Arnoud Boot

(2)

A

BSTRACT

This paper assesses the sample of about 1000 large firms for a period of 2007-2015 and whether there is a correlation between provisions established to limit shareholder’s rights, measured by the Entrenchment Index, and a company value and an operating performance, measured respectively by the industry-adjusted Tobin’s Q and a net profit margin. Using a year and a firm fixed effects regression no such correlation of a corporate governance and a company performance has been found. Moreover, the research investigates for possible effect of the corporate governance on stock returns, calculating abnormal returns of long-short high-low Entrenchment Index firm portfolios using the Fama and French four-factor model.

Keywords: corporate governance, governance provisions, shareholders’ protection, shareholders’ rights, the Entrenchment Index, company value, company operating

(3)

S

TATEMENT OF

O

RIGINALITY

This document is written by Agnieszka Bartosiak who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(4)

1. I

NTRODUCTION

In 1988 Milton Friedman has published his article “Stakeholder Theory of the Modern Corporation” in which he presents shareholders as the party mostly incentivized in increasing a company’s value – as they are the owners of the company, they are obviously interested in the value of its shares. However, public listed companies are not run by its shareholders, but by the management. Directors not always have the same incentives as investors, which creates a principal-agent issue. Literature describes how they can be tempted to focus on personal interests and violate the basic purpose of the company – bringing

additional value to its shareholders. Should investors be entitled to participate in making the major corporate decisions? Does it increase or decrease the value of the company, if they do have this right? Most shareholders desire to be able to react if the management is trying to limit execution of their entitlements. After all, they do not want to see management “wasting money from their pockets” and not being able to prevent them.

The policies and practises designed to protect shareholder’s rights are part of the corporate governance. Corporate governance is a system of checks and balances that is put in place in the organization to lessen agency costs, that arises due to agency problem -

difference in interest between shareholders and the management. (Larcker, Tayan, 2011). Berk and DeMarzo (2011) describes corporate governance as the system of controls, regulations, and incentives designed to prevent fraud in a company.

Corporate governance is a very broad term and describes qualities of a firm, that are difficult to be measured. How can companies be rated in terms of their corporate governance structures? How can they be compared with each other in this sense? The possibility of quantifying corporate governance would allow to look for answers to even more interesting

(6)

questions. Does the corporate governance have any significant impact on the company value or its performance?

What makes a better investment strategy – to invest in shares of companies with a good or bad corporate governance? As earlier studies showed, companies with higher quality of corporate governance have better operating performance than the rest. However, it does not always make a better investment, because investors already take this effect into account and include it in the stock price.

Investors see companies that score lower on corporate governance as investments with higher risk – worse corporate governance is aligned with weaker protection of shareholders rights.

There are many publications that try to draw the correlation between the level of shareholders’ rights and the company value and performance. Study of Gompers, Ishii and Metrick (2003), as well as Bebchuk, Cohen and Farell (2004) indicate, that there is a negative relationship between the company value and each additional barrier of shareholders to

execute their rights.

However, most of the studies conducted so far examine the governance provisions and the firm performance until the year 2006, so still before the financial crisis in 2007. One can believe, that results provided by studies a decade ago would differ from results of similar studies based on a current sample, due to increasing popularity of the topic of corporate governance, appearance of sophisticated corporate governance indexes and overall easier access to information. Gompers, Ishii and Metrick (2003) and Bebchuk, Cohen and Farell (2004) have found a market mispricing in terms of a corporate governance, by finding that increases in entrenchment of shareholder’s rights in the company is associated with

significant negative abnormal returns. Also, by using a trading strategy based on short selling stocks of companies with high level of entrenchment, and buying low stocks with low level of entrenchment, they create a portfolio that appears to outperform the market in most time

(7)

periods. Due to the growing awareness of investors of that findings, it might no longer be the case. Finally, the recent financial crisis was a reason for implementation of a new

legislation, which possible had an important impact on governance practises of public firms. This study focuses on trying to account for possible changes in how investors are reacting to corporate governance provisions (barriers of shareholders to execute their rights) and looking for a new evidence, that corporate governance has long-term effects and is correlated with company’s value. But even if corporate governance affects company value, it does not necessary need to affect the stock price in the same way. How is the relationship between a corporate governance and the firm performance incorporated in a share price by the market – is it not fully priced, and therefore a portfolio of stocks of companies with better corporate governance gives higher risk-adjusted investment returns compared to the rest of stock market?

In the study the following hypothesis will be tested:

H1: There is significant effect of corporate governance on the company value and operating performance

Previous studies show evidence of correlation between the Tobin’s Q and a corporate governance. There is no strong indicator, that it should no longer be true.

(8)

H2: There are no abnormal returns, that investors can make if they invest in better governed firms

Abnormal returns for specific portfolios, that picked up investments in firms based on their governance score was proved before. However, it is suspected in this research, that even if the corporate governance affects the company value and the operating performance, there were changes in beliefs of investors in comparison to previous studies, caused by easier access to information and their faster implementation in the stock price.

(9)

2. L

ITERATURE

R

EVIEW

2.1. The Principal-Agent Problem

Several authors have been raising questions about the main purpose of the company. Is it to serve in the interest of its owners or maybe should it take into account third parties affected by it? Moreover, should investors indeed have anything to say when it comes to managing the company and what is the role of management?

Friedman (1970) takes the approach, that business cannot have any responsibilities, as only people are able to have them. People who are working in business, such as corporate executives, are employees of the owners of the company, therefore their only responsibilities are the one to the employers. Usually the main goal of the employers is to maximize the firm’s profit. It is possible, that owners have another objective, like a social care in case of hospitals. In that case managers should also have aligned incentives with shareholders, but they should never prioritize social matters above other aims that are owners’ preference – they should be acting as a “principal” and not as an “agent” (Friedman, 1970). Friedman notices, that in cases that managers decide to focus on social objectives, like fighting with an inflation or an unemployment, that is not in the interest of shareholders, they are basically spending the investors’ money without their prior approval. The owners are the individuals, that should decide about matters of their firm and only they can have a possible responsibility to the society.

According to this point of view shareholder should have an absolute power over the company. He should be able to freely exercise his will and managers have no right to limit his activities.

Willingness of the manager to take care of social matters might not be the only reason why his incentives can diverge from the ones of shareholders. Jensen and Meckkling (1976)

(10)

have provided an extensive study on the agency problem, theory of property rights and theory of ownership structure of the firm. They define the agency relationship as a contract under which individual (a principal) engage another individual (an agent) to perform on his behalf, which is giving the authority to the agent. The principal can establish incentives that can prevent the agent from acting against his will, but it will come with an additional cost. The sum of this costs makes an agency costs, which mostly include monitoring and bonding expenditures and additionally added the residual loss – the loss in the value of the company, that appears as a result of a non-preventable divergence of incentives. An example of the bonding cost is the spending on an audit of the financial accounts – screening costs, that should prevent a manager from committing a fraud.

The separation of an ownership and a control will cause a general problem of

an agency (Jensen and Meckkling, 1976). Again there is brought up a point, that the company is not an individual, it is a legal fiction which serves a purpose of bringing together all the contracting relationships and therefore a firm cannot have an objective or a responsibility. Its outcome is an effect of many processes incurring between those relationships. An ownership structure plays here an important role here. The more shares of the company a manager has, the more his incentives are in line to the shareholder’s incentives. However, if a shareholder appears to be at the same time a manager, he might want to use company resources for his own purposes because the cost made by the company is spread on many shareholders, while the manager will enjoy big part of the benefits. An example can be increasing unnecessary spending on his office or on business trips. By generating those expenditures, he bears only a part of the cost and the rest is covered by all of the other shareholders. On the other hand, as the ownership of a manager falls, his incentive to follow the objections of the rest of the investors instead of following his own aims is becoming weaker. He might be against taking up creative activities that could possibly lead to an increase in a company’s value due to the

(11)

hassle and the higher risk, that could eventually lead to the loss of the managerial position (Jensen and Meckkling, 1976).

Another view on the company purpose takes Freeman (2001). According to him corporations interact with individuals, from whom firms receive benefits. This group of individuals can be also harmed by the company or its rights can be violated, for example by selling faulty product that is dangerous for consumers. As consumers, suppliers, employees and a local community are influenced by the company, they should be able to demand special actions from the management. They are all vital to the success of the company and they create a firm’s universe. For example, suppliers have strong influence over a material price and its availability. If they are harmed by the company, they might decide to end

a cooperation, which increases costs of a production or even makes the production impossible (Freeman, 2001).

Customers generate the revenue to the company by buying its products and services. Paying attention to their needs helps in developing a better selling product, which pleases both, suppliers and owners (Peters and Waterman, 1982).

Although the interests of stakeholders and shareholders can be aligned, they do not have to be. Here the management appears as the agent, who has to find a solution to this issue and takes into account all of the claims of the stakeholders in the name of the welfare of the company (Freeman, 2001). Shareholders, who prioritise maximizing company profit, might not take into account those claims, although in the end they can be harmful for the value of their stock. For instance, under the pressure of shareholders a management of pharmaceutical company can consider releasing a new form of a drug for a so far incurable disease, despite the known side effects. After entering the product to the market it can bring a significant revenue, but within couple of months company might receive complains about the side effects and its reputation will suffer, diminishing sales on not only the newly issued product,

(12)

but possible the whole assortment of the company. Moreover, managerial employees might be brought into a court and accused of selling a medicine that is harmful for the society. Faulty products or services, as well as corporate governance scandals, can make companies go bankrupt.

A famous example is the Enron Corporation, an American energy company, and its corporate governance scandal publicized in October 2001(Bratton 2002). From one of the most

successful and innovative company it went bankrupt in less than a year, after it was revealed that some executives were hiding billions of dollars in debt from failed projects. The scandal was named the biggest audit failure at that time and had devastating results for the Enron Corporation. From a stock price of $90 in August 2000 it plummeted to 40 cents per share after the company went into Chapter 11 in December 2001 (Bratton 2002).

The Nomura Holdings stands as another example of how a company can suffer due to lack of

proper corporate governance structures in place (Tabuchi 2011). The company is the part of

the Nomura Group, that is being considered the largest securities company in Japan. The Nomura Group offers wide range of services and products, like securities, an IPO

underwriting, an asset management and a capital investment. In 2010 the Nomura Group was accused by some of its clients of suspicious trading in deals, that company was underwriting

upon issuance of new shares in mentioned companies. As a result of this accusation, the

investigation of those activities started and it lead to the discovery of an insider trading by

Nomura sales officers.

Due to the scandal, the company lost many potential clients, who decided to not enter the stock market using Nomura’s services (Tabuchi 2011).

(13)

2.2. Corporate Governance

Appearance of many relationships between contractors in the company can create principal-agent problem. Corporate governance is a mechanism that deals with the ways in which suppliers of finance to corporations assure themselves of getting a return on their investment (Schleifer and Vishny, 1997). Also, it was specified that “corporate governance concerns the structure of rights and responsibilities among the parties with a stake in the firm" (Aoki 2001). However, possibly entrenchment of shareholder’s rights can also produce beneficial effects by reducing the extent to which the threat of a takeover distorts investments in long-term projects (Bebchuk and Stole, 1993).

In general mechanism of a corporate governance can be divided into country or region focused external mechanisms, or internal specific company related mechanisms. The focus of this study in mainly on the internal mechanisms.

An important right that investors desire to have is to vote on the most relevant decisions in the company, like acquisitions or appointing the members of the board (Shleifer and Vishny, 1997).

Managers try to protect themselves from the shareholders’ power by establishing provisions restricting their rights (Gompers, Ishii and Metrick, 2003). Those provisions are referred as “corporate governance provisions” or “shareholders; right provisions”. In most cases they initially served as anti-takeover provisions that should prevent the hostile takeovers. Whether it is plausible to say that those policies are still useful in terms of takeovers, there are still a relevant limitation for shareholders.

Gompers, Ishii and Metrick (2003) describe in their study the most common provisions and create a Governance Index. This index is based on 24 provisions – for every provision that restricts shareholders’ rights there is added a one point to the index. Companies

(14)

with the highest level of the index are described as “dictatorships” and the ones with the lower level are called “democracies”.

In the method of Gompers, Ishii and Metrick’s all provisions are weighted equally, while in reality some provisions might be more important than others. Therefore, Bebchuk, Cohen and Farell (2004) conducted a study in order to check, which provisions matter the most in a firm. As a result, they come up with an Entrenchment Index, that consist of six essential restrictions: staggered boards, limits to amend bylaws, limits to amend charter, a supermajority, golden parachutes and poison pills. Initially those provisions were chosen to examin the preferences of shareholders in votes on precatory resolutions. Selected provisions not only were submitted the most number of times by investors, but also appeared to be the most significant in empirical studies of a relationship between the corporate governance and the company value.

As the Entrenchment Index is crucial for this study, the explanation of the mentioned provisions can be found in the appendix.

2.3. Shareholder’s Rights and Company Value

This section is providing a background on the correlation between corporate governance provisions and the value of the company. Is it indeed true, that a company that lets their investors exercise their control has the higher value than the one that has a set of policies present to prevent them from it? Do better shareholders’ rights help to create a value?

Gompers, Ishii and Metrick (2003) in their pioneer studies were looking for correlation between the Governance Index and the company value, measured by

the Tobin’s Q. They used the sample of about 1500 U.S. companies for years 1990, 1993, 1995 and 1998. Regression of the Governance Index and control variables on the industry-adjusted Tobin’s Q showed a significant negative and strong correlation - each one-point

(15)

increase in the Governance Index is associated with a decrease in the Tobin’s Q of 4.4

percentage points. They also find the positive correlation for the democracy portfolio variable and the Tobin’s Q. By regressing the Governance Index and control variables on operating performance measures they find that on average there is a significant negative correlation of the index with the net profit margin and an on-year sales growth.

But at was noticed before, not all governance provisions might have the same effect. Bebchuk, Cohen and Farell (2004) contributed to the literature by coming with an

Entrenchment Index, that is constructed based on the six most important provisions. They follow the methodology of Gompers, Ishii and Metrick and using the OLS White robust regression of the Entrenchment Index and control variables on the logged industry-adjusted Tobin’s Q find that the coefficient of the entrenchment provisions index is highly significant and negative with a value of -0,044.

An interesting finding has study of Bhagat and Bolton (2011), which shows negative relationship between board independence and operating performance during period before 2002, and a positive relationship after 2002. A wave of governance regulations that appeared in 2002 and later, after discovery of corporate scandals of companies like Enron or Tyco, is pointed out as a possible cause to reverse of this relationship – mostly the Sarbanes-Oxley Act of 2002 (SOX). As a majority of the studies, even recent ones, research only sample before 2002, the effect of the corporate governance on the firm performance remains ambiguous. The paper investigates the results on sample from period 1998-2007, which is a new contribution to the literature, but does not include post-crisis sample. Also, Bhagat and Bolton come up with the idea of a new governance measure – dollar value of director stock ownership.

(16)

2.4. Shareholder’s Rights and Share Price

It is interesting to observe changes in share price that can be caused by certain corporate governance laws. If shareholders are aware of how significant the corporate governance is, they should monitor it carefully. Also, they could decide to “vote with their money” in case that the company does not meet their requirements and sell owned shares (Friedman, 1970). However, if the investors do not incorporate this information in the stock price immediately, then can be created a mispricing on the market and realized returns on the stock would differ systematically from equivalent securities (Gompers, Ishii and Metrick, 2003).

Gompers, Ishii and Metrick take an effort to examine the relationship of the Corporate Governance Index and subsequent returns. By using a regression of a long-short investment strategy for each portfolio with different Governance Index level for each month on the four-factor model of Carhart (1997), they find that a hedge portfolio long “Democracies” and short “Dictatorships” generates an abnormal return of approximately 8,5% per year throughout the sample period.

Bebchuk, Cohen and Farell (2004) following the same methodology but using the Entrenchment Index as a proxy find similar results – shorting the High-Entrenchment Index portfolios and longing the Low-Entrenchment Index one would generate a yearly abnormal return of 7,4% when portfolios are equally-weighted and 14,8% when portfolios are value-weighted. They also notice a decline in noted abnormal returns as the strategy includes more firms in the middle of a normal distribution. Even after controlling for the industry effects, the results remain robust.

Bhagat and Bolton (2011) find, that none of the governance measures are correlated with the current or future market performance. It is consistent with the Efficient Market

(17)

Hypothesis, which means that after finding out that there is a mispricing on the market, the prices adjust to it and correct for the previously omitted information. In addition, companies that are not compliant with the SOX achieve higher abnormal returns upon becoming compliant.

However, Core, Guay and Rusticus (2006) research the same sample and they find that there is a stock price underperformance for companies with weak shareholders’ rights, but no significant underperformance for this companies in the operating performance,

measured as a return on equity. A proposed possible explanation is that investors anticipating that weaker corporate governance causes a worse operating performance and incorporate that information in the stock price, where in fact there is no evidence supporting this statement. Another possible explanation is that weak-corporate-governance firms have a greater takeover protection, as governance provisions serve atthe same time as anti-takeover provisions. Stock returns of those firms can change as investors are surprised by the decreasing probability of receiving a takeover premium.

2.5. Changes after Financial Crisis

Ten years ago Corporate Governance was still in a concept stage and where scholars,

employees of large corporations or block holders were well acquainted with this term, many small investors were not fully aware what it means and what impact it has on their portfolios. Many financial frauds and articles, written about how limiting shareholder’s rights affect the firm, had proven, that this issue matter.

A series of corporate governance scandals in United States in early 2000s, like the WorldCom and the Enron, lead to introduction in 2002 of the Sarbanes-Oxley act. (Cunningham, Asare, Wright, 2006). The act had its effects on auditors and firms,

(18)

oversight over public companies. Also, it sets reporting requirements for accountants and auditors, introduces codes of conduct for securities analysts, as well as requires Comptroller General and the SEC to perform studies and reports on potentially fraudulent activities, like consolidation of accounting firms or issuance of credit ratings. The SOX also regulates reporting of a corporate tax return and penalties for white-collar crimes and a fraud

(Cunningham, Asare, Wright, 2006). New legislation was supposed to improve an oversight and a control over governance practices of public companies. Moreover, its aim was to restore the public trust towards corporations and their boards of directors, specifically trust of investors in financial reporting.

Despite the US government’s effort in ensuring investors that there is an appropriate control over public firms, a shadow fell upon the issue again with the most recent financial crisis, that begun in 2007. It has brought a lot of attention to the corporate governance practices again, as poor policies were partially blamed for causing it. Mostly weak board oversight of financial institutions appears to be an important issue, that might not be addressed properly in 1990s and early 2000s (Kirkpatrick, 2009). As the result of financial crisis the U.S.

government passed in 2010 a Dodd-Frank US financial regulatory reform act (Ziegler and Woolley, 2016). The purpose of the act was to make the financial system more transparent and accountable. It was granting the Security Exchange Commission right to enable certain shareholders a proxy access to remove and nominee managers (Barry and Kairis, 2011) Also, the act required financial institutions to establish independent compensation committee and to adapt say-on-pay vote, which allowed investors to decide about a firm’s compensation

policy. Another requirement is a complete transparency of companies in terms of golden parachutes and again, shareholders were given right to vote on this form of compensation as well (Barry and Kairis, 2011).

(19)

Another proof of increasing popularity of the corporate governance topic is the growing amount of academic publications related to it. According to Web of Science Thomson Reuters database the amount of publications on the topic “Corporate Governance” in 1990 accounted for only 13 articles, while in 2015 it was already 1388. Published studies regarding “Corporate Provisions” risen from 3 articles in 1993 to 318 articles in 2015.

After so significant changes it is appropriate to make the assumption, that investors learnt from the effects of crisis and their awareness about techniques of expanding their rights increased.

(20)

3. M

ETHODOLOGY

3.1. Measure of Corporate Governance

In order to research a relationship of the corporate governance, the firm value and the performance there has to be chosen a measure of the corporate governance. The Governance Index appears to be a good measure, however Although it has its flaws since it incorporates attributes that do not seem relevant. As many studies have used so far the famous

Entrenchment Index to capture limitations of shareholder’s rights and had find out that it is more effective than the Governance Index, this study adapts the E-index as well. Therefore, the first step in this research was to obtain information about specific governance provisions: staggered boards, limits to amend bylaws, limits to amend charter, a supermajority, golden parachutes and poison pills. After receiving this data, it is possible to calculate the level of the index for each company for each year.

The construction of the index is very straightforward. E-index can take a value from zero to six, as there are six provisions that index is based on. For each present limitation there is a one point of the index added, so companies with the index level of zero have no

provisions present and are the most “democratic”, while companies with the level of six have all provisions that exist taken into account.

As the data for a “supermajority” variable was not reported in 2007, missing values were filled in on basis of the data from next coming year. So the companies that had supermajority provisions in 2008 are assumed to have them already a year before.

(21)

3.2. Corporate Governance and Company Value

To examine the long-term relationship of the corporate governance and the company value, the industry-adjusted Tobin’s Q is selected as an appropriate measure of company’s value. The definition of the Tobin’s Q remains the same as used by Kaplan and Zingales (1997) and Gompers, Ishii and Metrick (2003). According to it, the Tobin’s Q is calculated by dividing a market value of assets by the book value of assets.

𝑇𝑜𝑏𝑖𝑛&_{𝑠 𝑄 =}𝑀𝑎𝑟𝑘𝑒𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠

𝐵𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠

Market value of assets is equal to book value of assets and market value of common stocks, with deduction of sum of book value of common stocks and balance sheet differed taxes.

𝑀𝑎𝑟𝑘𝑒𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠

= 𝐵𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎𝑠𝑠𝑒𝑡𝑠 + 𝑀𝑎𝑟𝑘𝑒𝑡 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑐𝑜𝑚𝑚𝑜𝑛 𝑠𝑡𝑜𝑐𝑘𝑠

− 𝐵𝑜𝑜𝑘 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑐𝑜𝑚𝑚𝑜𝑛 𝑠𝑡𝑜𝑐𝑘𝑎 − 𝐵𝑎𝑙𝑎𝑛𝑐𝑒 𝑠ℎ𝑒𝑒𝑡 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑑 𝑡𝑎𝑥𝑒𝑠

As one can expect, values of the Tobin’s Q might cluster between industry and cannot be cross-compared. To adjust the Tobin’s Q for the effect of the industry, the mean of the Tobin’s Q is calculated for each industry recognized by The Standard Industrial

Classification (SIC) code. In the next step an industry-adjusted Tobin’s Q is obtained by deducting from the Tobin’s Q the industry average of the Tobin’s Q.

𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 − 𝑎𝑑𝑗𝑢𝑠𝑡𝑒𝑑 𝑇𝑜𝑏𝑖𝑛&_{𝑠 𝑄 = 𝑇𝑜𝑏𝑖𝑛}&_{𝑠 𝑄 − 𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑇𝑜𝑏𝑖𝑛}&_{𝑠 Q}

Then can be run a fixed-effects regressions of Entrenchment Index on the industry-adjusted Tobin’s Q:

(22)

Where Qit is the industry-adjusted Tobin’s Q, Index is the vector of the Entrenchment Index for company i in period t and Xit is the vector of other company characteristics, described by control variables. Regressions will be controlling for year and firm fixed effects.

As selected control variables standard proxies are used: the logarithm of assets, return of assets, ratio of capital expenditures to total assets, ratio of capital expenditure for research and development to total assets, logarithm of company age, leverage, logarithm of level of insider ownership. Additional control is Delaware dummy variable, taking on value one if the company is placed in Delaware state and zero if otherwise.

We expect to find a correlation between Tobin’s Q and Entrenchment Index, which would be an additional evidence to earlier findings, described in section 2.2. The correlation between the Entrenchment Index and the Tobin’s Q means that there is a correlation between company value and a firm’s corporate governance, which we believe that exist.

3.3. Corporate Governance and Stock Return

To identify the correlation between the E-index and stock returns, companies will be added to specific portfolios based on the index score. The abnormal return associated with taking a long-position in portfolio of companies with a low index value and a short-position in portfolio of companies with a high index value should show the nature of this correlation. Following Gompers, Ishii and Metrick (2003) methodology there will be run a regression using the four-factor model of Carhart (1997):

(23)

where Rt is the excess return to some asset in month t, EXMKTt is the month t

value-weighted market return minus the risk-free rate. SMBt (small minus big), HMLt (high minus low), and Momentumt are the monthly returns at time t on zero-investment factor-mimicking portfolios designed to capture size, book-to-market, and momentum effects. In next step it is possible to check whether the intercept of this regressions reveals the effect of quality of corporate governance on excess returns. A higher value of alpha for portfolio of a weak governance would suggest that companies in this portfolio are undervalued.

Our expectation is that, unlike proved in the past, no abnormal returns can be observed, because investors possibly account for the effects of corporate governance more than before.

3.4. Corporate Governance and Operating Performance

Completing the study, it is necessary to research the relationship of governance provisions and an operating performance. An operating performance is measured as the net profit margin, described as income divided by sales.

𝑁𝑒𝑡 𝑝𝑟𝑜𝑓𝑖𝑡 𝑚𝑎𝑟𝑔𝑖𝑛 = 𝐼𝑛𝑐𝑜𝑚𝑒 𝑆𝑎𝑙𝑒𝑠

The measure is industry-adjusted in the same way as Tobin’s Q. Then the relationship is evaluated by the regression:

𝑂𝑃𝑖C= 𝛼C+ 𝛽C𝐼𝑛𝑑𝑒𝑥 + 𝛾C𝑋BC+ 𝑒C

where OPt is a measure of operating performance of company i in period t, Index is a value of Entrenchment Index and Xit is the vector of control variable. As a control variable is used logarithm of book-to-market ratio. OP is measured as a company’s net profit margin.

(24)

Similarly, to the correlation between the Entrenchment Index and the Tobin’s Q, we hope to find a relationship between the index and net profit margin.

4. D

ATA

4.1. Database

A governance database of the Institutional Shareholder Services Governance database was used in order to receive an information about appearance of selected corporate

governance provisions in U.S. public companies. The sample period ranges from January 2007 until December 2015.

It is an update in comparison to the database used by Gompers, Metrick and Ishii (2003) and not only because of obvious difference in a time period. Gompers, Metrick and Ishii used the Investor Responsibility Research Center (IRRC) Takeover Defense database listings of corporate governance provisions for years 1990, 1993, 1995 and 1998. The universe of this database was drawn from Standard & Poor 1500 and annual listings of other largest-cap companies1

.

After 2007 the methodology of collecting data changed and the amount of variables ceased significantly, therefore the database was split for period before and after 2007. With the ISS Governance Data (after 2007) it is not possible anymore to calculate the G-index, introduced by studies of Gompers, Metrick and Ishii. However, the new dataset does not have a time break, which occurred in the legacy part. Also, it allows to calculate the Entrenchment Index of Bebchuk, Cohen and Farell.

1

(25)

All the companies with dual class common stocks were removed from the selected sample, as their governance structure differs to a great extent from single class commons stock firms, what makes them incomparable. Also real estate investment funds were excluded for the same reasoning. Firms operating in this industry were located by SIC code 6798.

To obtain data about firm’s financial characteristics we use CRSP/Compustat Merged. Also ExecuComp data is used to calculate the level of insider ownership. The age of

companies is estimated based on date of oldest and most recent observations for selected firms in CRSP.

Calculating abnormal returns requires using Fama-French Portfolios and Liquidity Factors. All databases are accessible through Wharton Research Data Services.

(26)

4.2. Descriptive statistics

Table A. 1 presents the incidence of corporate governance provisions for years 2007-2015. The amount of observations varies for each year, from 950 in 2007 to 1060 in 2007-2015.

Among the provisions that belong to the Entrenchment Index, the most common are golden parachutes (79,31% of incidence in 2015), limits to amend bylaws (89% of incidence in 2015) and limits to amend charters (98,53% of incidence in 2015). There is an increase in percentage of incidence for most Entrenchment Index provisions, except the presence of classified boards and poison pills. There is also a significant increase in presence of supermajority requirement - from 29,05% in 2007 it rises to 60,29% in 2015. Incidence of Other Provisions looks more stable than the one of the Entrenchment Index provisions. It can

12 Appendix A. Tables and Figures

TABLEA.1: INCIDENCE OF CORPORATE GOVERNANCE

PROVI-SIONS FOR YEARS 2007-2015 (as percentage)

2007 2008 2009 2010 2011 2012 2013 2014 2015 Entrenchment Index Provisions Classified Board 56.21 52.71 50.67 49.76 45.44 42.82 39.91 36.51 33.43 Golden Parachutes 53.26 33.46 81.01 84.05 83.77 87.7 84.17 83.77 79.31 Bylaws 88.74 88.45 88.17 88.70 89.53 89.87 88.82 88.21 89.12 Charter 89.05 91.12 90.94 91.83 92.89 95.18 95.83 97.64 98.53 Supermajority 29.05 27.64 33.78 37.80 42.65 47.93 51.75 56.70 60.29 Poison Pill 40.32 35.44 27.19 20.61 16.52 13.40 10.14 9.34 6.67 Other Provisions

Blank Check Preferred 92.03 91.76 93.00 93.24 93.49 93.73 93.88 93.41 93.66 Special Meeting 44.66 45.44 46.11 47.16 48.12 49.22 53.31 54.36 55.53 Written Consent 42.11 41.91 41.35 44.21 58.25 58.33 57.81 58.97 58.91 Cumulative Voting 8.08 7.51 7.60 6.99 5.96 5.80 5.46 5.28 4.81 Unequal Voting 0.55 0.53 0.44 0.51 0.51 0.15 1.52 1.23 0.72 Fair Price 12.48 13.21 15.20 16.30 13.75 13.57 13.08 12.58 12.50 Number of Firms 950 1013 1048 1053 1041 1037 1055 1060 1020

TABLE A.2: INCIDENCE OF E-INDEX LEVELS FOR YEARS

2007-2015 (as percentage) 2007 2008 2009 2010 2011 2012 2013 2014 2015 Entrenchment Index 0 0.42 0.30 0.10 0.00 0.00 0.00 0.09 0.09 0.00 1 3.26 4.74 2.29 2.18 1.44 0.87 0.85 0.75 0.98 2 13.37 20.34 9.73 9.02 8.74 8.00 7.01 6.70 7.55 3 30.84 33.17 27.77 27.83 30.16 29.99 31.18 28.96 29.22 4 30.00 25.77 37.21 38.65 39.29 41.47 43.41 47.55 48.63 5 18.53 13.33 19.56 19.47 17.96 17.84 16.11 14.72 12.65 6 3.58 2.37 3.34 2.85 2.40 1.83 1.33 1.23 0.98 Number of Firms 950 1013 1048 1053 1041 1037 1055 1060 1020

(27)

be a signal, that companies are not changing their Other Provisions policies as often, as they do not matter for a corporation to such extend as provisions of the Entrenchment Index.

Table A. 2 presents the Entrenchment Index levels for years 2007-2015. The sample did not change, so the number of observation remains the same as in table A. 1.

The Entrenchment Index consist of following provisions: classified boards, golden parachutes, limits to amend bylaws, limits to amend charters, a supermajority and poison pill. Explanation of each provision can be found in the appendix.

To calculate the Entrenchment Index, a one point is added to the index for appearance of each provision, so an index level of zero means, that company does not have any of the mentioned provisions. However, in the sample the percentage of those companies is very small and there is none company without any provision in years 2010, 2011, 2012, 2015. Most of the companies have from 3 to 4 provisions - around 60-70% of all companies each year. Also, through the years the sample become less diverse, there is less companies that fall

TABLEA.1: INCIDENCE OF CORPORATE GOVERNANCE

PROVI-SIONS FOR YEARS 2007-2015 (as percentage)

2007 2008 2009 2010 2011 2012 2013 2014 2015 Entrenchment Index Provisions Classified Board 56.21 52.71 50.67 49.76 45.44 42.82 39.91 36.51 33.43 Golden Parachutes 53.26 33.46 81.01 84.05 83.77 87.7 84.17 83.77 79.31 Bylaws 88.74 88.45 88.17 88.70 89.53 89.87 88.82 88.21 89.12 Charter 89.05 91.12 90.94 91.83 92.89 95.18 95.83 97.64 98.53 Supermajority 29.05 27.64 33.78 37.80 42.65 47.93 51.75 56.70 60.29 Poison Pill 40.32 35.44 27.19 20.61 16.52 13.40 10.14 9.34 6.67 Other Provisions

Blank Check Preferred 92.03 91.76 93.00 93.24 93.49 93.73 93.88 93.41 93.66 Special Meeting 44.66 45.44 46.11 47.16 48.12 49.22 53.31 54.36 55.53 Written Consent 42.11 41.91 41.35 44.21 58.25 58.33 57.81 58.97 58.91 Cumulative Voting 8.08 7.51 7.60 6.99 5.96 5.80 5.46 5.28 4.81 Unequal Voting 0.55 0.53 0.44 0.51 0.51 0.15 1.52 1.23 0.72 Fair Price 12.48 13.21 15.20 16.30 13.75 13.57 13.08 12.58 12.50 Number of Firms 950 1013 1048 1053 1041 1037 1055 1060 1020

TABLE A.2: INCIDENCE OF E-INDEX LEVELS FOR YEARS

2007-2015 (as percentage) 2007 2008 2009 2010 2011 2012 2013 2014 2015 Entrenchment Index 0 0.42 0.30 0.10 0.00 0.00 0.00 0.09 0.09 0.00 1 3.26 4.74 2.29 2.18 1.44 0.87 0.85 0.75 0.98 2 13.37 20.34 9.73 9.02 8.74 8.00 7.01 6.70 7.55 3 30.84 33.17 27.77 27.83 30.16 29.99 31.18 28.96 29.22 4 30.00 25.77 37.21 38.65 39.29 41.47 43.41 47.55 48.63 5 18.53 13.33 19.56 19.47 17.96 17.84 16.11 14.72 12.65 6 3.58 2.37 3.34 2.85 2.40 1.83 1.33 1.23 0.98 Number of Firms 950 1013 1048 1053 1041 1037 1055 1060 1020

(28)

index. There is a very significant drop of the number of companies with highest score of the index. In 2007 22,11% of all firms have index level of 5 or 6, while in 2015 it dropped to 13,63%, with only 0,98% of companies with index level 6.

5. T

HE

E

MPIRICAL

A

NALYSIS AND

R

ESULTS

5.1. Tobin’s Q and Net Profit Margin Regression Analysis

Results of the regressions of industry-adjusted Tobin’s Q on the entrenchment index and control variable are presented in Table A.3.

The first column shows results of a year fixed-effects regression, using the Entrenchment Index in aggregate. The second column presents results of a year fixed-effects regression also, but using different levels of the index. The third and the fourth column display results of similar regressions, but taking in account firm fixed-effects as well. Robust standard errors of the coefficients are given in the parenthesis.

In order to comply with the hypothesis specified in this paper, the coefficient of the

Entrenchment Index in aggregate and for the higher levels of the index should be negative. The changing sign of the coefficient of Entrenchment Index in regressions is puzzling. For a regression with year fixed effects only it is negative, while controlling for firm fixed effects as well yields coefficient with a positive sign. However, the results are not statistically significant and there cannot be drawn any correlation between the Entrenchment Index and the industry-adjusted Tobin’s Q.

Also not all control variables appear to be significant. A leverage, a company age and assets are negatively correlated with the industry-adjust Tobin’s Q. A return on assets and spending on Research and Development are positively correlated with the dependent variable. Those findings are similar to the ones of Bebchuk, Cohen and Ferrel (2004).

(29)

We also test the possible correlation between the Entrenchment Index and the industry-adjusted Tobin’s Q only for years 2010-2015, removing from the sample years 2007-2009, suspected to be “noisy” due to the financial crisis in 2007. The results are presented in the table A6. As before, there was used a regression with year-fixed effects and year- and firm fixed-effects. However, any if the coefficients of the Entrenchments Index appears to be statistically significant again.

Table A7. displays results of a regressions after taking the subsample of big and small cap companies. Big cap companies are defined as a firm with total market capitalization above 10 billion USD. Small cap companies are defined as a firm with total market capitalization below 2 billion USD. For the big caps with a year-fixed effects regression there is a significant negative coefficient of -0.051 of the Entrenchment Index variable with a significance level 10 percent. It can be interpreted as that with each additional provision added to the Entrenchment Index, the industry-adjusted Tobin’s Q decreases of -0.051 point. However, taking into account the firm-fixed effects as well, the Tobin’s Q and the

Entrenchment Index remain uncorrelated.

Looking for more evidence about possible correlation between the Entrenchment Index and Tobin’s Q for big caps, we perform similar study on the subsample with time period of 2010-2015 only. Results of this study are presented in table A7. Again, there can be found a coefficient of the entrenchment Index of -0.060 significant at the 10 percent level, but only for a regression with fixed-year effects. Adding firm-fixed effects, we receive almost no significant coefficients, leaving quite weak evidence for any effect of corporate governance on company value of companies with big market capitalization.

Table A4. displays result of the regressions of a net profit margin on the Entrenchment Index and a book-to-market ratio. The net profit margin is calculated as the ratio of a net income and sales. As with the Tobin’s Q, results of the regressions do not appear to be statistically

(30)

significant. Coefficient of the Entrenchment Index in aggregate is insignificant with and without control of company fixed effects. The Entrenchment Indexes separate levels coefficients are also not significant after controlling for both year and firm fixed effects.

5.2. Abnormal Returns Regression Analysis

Table A.5 presents the monthly abnormal returns of investing in the long-short portfolio of companies with a low-high value of the Entrenchment Index for the period of January 2007 to December 2015. Two methods were used to calculate the abnormal returns. In the baseline model, abnormal returns were calculated by regressing a return associated with particular trading strategy on the Fama-French (1993) three factor model and Carhart’s momentum (1997). The exact methodology is described in Section 2.2. In the industry-adjusted model, returns were first adjusted for the industry, by deducting the industry’s median return. The industry was localized by the 4-digit SIC code - it’s a stricter method than the one used by Bebchuk, Cohen, and Ferrell (2003), where a 2-digit identification was used. The long and short portfolios were constructed both by respectively equal weighting and by value-weighting, based on the percentage of a total market capitalization.

Using this method, the returns of long-short portfolio of low-high Entrenchment Index are compared with the standard portfolios, Small minus Big, High minus Low and the value-weighted return in excess of the risk-free rate. According to the literature those portfolios should have most of the explanatory power over the stock market returns. Statistically significant positive monthly abnormal returns would imply that the market is not fully implementing all information in the stock price.

However, in regressions made here, no significance of the coefficients was found, using both, the equally-weighted portfolio and the value-weighted portfolio, except of a weak positive abnormal returns for equally-weighted baseline model portfolio.

(31)

6. D

ISCUSSION

6.1. Reasoning of the results

Findings presented here about the Entrenchment Index, the company value and the company performance are not consistent with earlier finding of Gompers, Ishii, and Metrick (2003) and Bebchuk, Cohen and Ferrel (2004). However, it is important to take into account, that this paper is investigating a sample for a different time period. Also, there is no evidence that would support the hypothesis made earlier, that there is a correlation between shareholders’ rights and firm value.

Looking for a reason why there could be proven in a past correlation between the

Entrenchment Index and the Tobin’s Q, which is not found now, it is important to consider the endogeneity issue. Corporation decide about how their policy will be shaped and in the past there were not many limitations given by law. Any introduced provision on

shareholders’ rights was not at random, as it was a decision of a management. Companies who struggled with operating in the past, had worse position on the market and needed to protect themselves by setting up anti-takeover positions. Therefore, the policy was only a result of firm’s poor performance and not a cause of it. But nowadays most of anti-takeover positions are not considered effective anymore, as hostile takeovers became less popular. Companies with worse performance do not establish as often new provisions that are part of the index. Also, companies that had provisions in place in the past, usually keep them even if they are not financially constrained anymore. The same issue exists in the opposite way – companies that do not have any or few entrenchment provision in place will have difficulties with establishing ones, as shareholders will not let the management to set them up.

(32)

As we find no correlation between a corporate governance and a company value and a performance, we would not expect that there should be any effect of it on a stock price as well.

Regarding the Entrenchment Index and the stock return, there was found no evidence of abnormal returns of the long-short portfolio of stocks in marginal values of the index. Using similar methods, Gompers, Ishii and Metrick (2003) found that monthly abnormal returns of going long in the “Democracies” portfolio and short in the “Dictatorship” portfolio were 71 basis point with 1% significance for value weighted portfolio and 45% for equally-weighted portfolio with 5% significance. Bebchuk, Cohen, and Ferrell (2004) use the sample from 1990s, as Gompers, Ishii and Metrick, and the sample from 1990 to 2003 and still find a positive monthly abnormal returns, though not as high.

One of the possible explanation is the investors’ ability to observe and learn about the effects of the corporate governance provisions. Since publications of Gompers, Ishii and Metrick (2003) many articles were published and made available for market participants. Therefore, it is no surprise, that more than a decade later no abnormal returns calculated using their

method can be found.

Another explanation is lack of robustness of studies of Gompers, Ishii and Metrick (2003) and Bebchuk, Cohen, and Ferrell (2004). Returns of Gompers, Ishii and Metrick (2003) were not industry-adjusted, which was pointed out by Johnson, Moorman and Sorescu (2007). They reexamine earlier results and use industry clustering, to find statistically zero abnormal returns. In addition, they refer to publication of Lyon, Barber, and Tsai (1999), which uses industry clustering and that still yields big differences between their results, because the industry identification was using only 2-digit SIC code instead of 3-digit one. In this research a 4-digit SIC code was used, which leads to an extreme clustering method.

(33)

Another important issue is the way that the Entrenchment Index is calculated. Each provision in the index has equal value, while there is no evidence that all of the five provisions have the same impact on the company. There is a possibility, that also different combinations of their presence has variant effects. So far no research has been yet made in this direction, so that is yet to be determined.

6.2. Importance of the study

Understanding the correlation between shareholders’ rights provision and the company value is crucial. Since publication of Gompers, Ishii and Metrick (2003), there exists a common belief that indeed governance provisions are linked with the company value, which was not entirely proved or claimed in a mentioned paper. However, on the example of the further fate the IRRC Takeover Defence database, it can be noticed that the demand for data on corporate governance is growing. Firstly, acquired by the MSCI together with the acquisition of the RiskMetricks in June 2010 and then purchased in April 2014 by the Vestal Capital Partners. Nowadays, it operates until name of the Institutional Shareholder Services (ISS) and offers much more than just data on shareholders’ rights provisions. From proxy voting services, to governance research and advisory, even providing its own analytics platform of a non-financial data, the ISS Inc. provides their services to over 1700 institutional clients on 117 markets.2

In fact, information about governance rating are as important for investors as for corporations, when the score of the firm’s index is a decisive factor in planning where to place an investment. The more academic independent research will be done in that matter, the better quality of information will be available for a wider range of market participants,

2_{Institutional Shareholder Service Inc, available:} https://www.issgovernance.com/about/about-iss/

(34)

decreasing the importance of indexes created by corporations, that have monopoly over corporate governance data. As mentioned earlier by Bebchuk, Cohen, and Ferrell (2004), shareholder advisory firms tend to create indexes with massive amounts of provisions, increasing the noise of the variables, that do not matter for companies. As a result, large indexes give misleading results, so it is important to work on developing improved governance measures.

(35)

7. C

ONCLUSION

This research provides information on the correlation of the corporate governance and the company value and the operating performance. Also, it ought to find whether choosing an investment strategy based on the values of the Entrenchment Index can generate abnormal returns.

Looking at the results of regressions on the company value (Tobin’s Q) and the operating performance (net profit margin) it is not possible to support the hypothesis that limitations to shareholders’ rights have any influence over company value or performance. As the

correlation between the corporate governance and values dissolves, there is a weaker

indicator of correlation between corporate governance and stock returns. Indeed in this study are found no abnormal returns, that investors can make if they invest in better governed firms. There is no evidence found that investing in well-governed stocks increases an investment strategy’s portfolio alpha.

However, there is also no evidence to claim that the corporate governance is not correlated with firm values. A corporate governance issue usually suffers from the problem of

endogeneity. Because corporations can decide themselves about what policies to establish, it is not a random event and it is not possible to say if the policy changed a company

characteristic or if an appearance of some characteristics drives management to establish the policy.

Moreover, there is still a question – is the industry-adjusted Tobin’s Q an accurate measure of the company value? Is the net profit margin an accurate measure of the firm operating

performance?

And, is the Entrenchment Index still a good measure of shareholders’ right provisions? Looking at the changes in the sample, it can be noticed a shift from the marginal levels of the

(36)

Entrenchment Index towards the middle of the distribution. Firms’ policies over the years became more unified, so it is obviously more difficult to see the influence of a difference between the levels of the index distribution.

A further research on the topic should focus on improving the measurement of shareholders’ rights. The Entrenchment Index is straightforward and it does not account for the possibility that different combinations of the provisions can have diverse effects on a company.

(37)

A

PPENDIX

A.1. 13

TABLEA.3: THE ENTRENCHMENT INDEX AND TOBIN’S Q

This table reports fixed-effects regressions of industry-adjusted Tobin’s q on various controls and two specifications of the entrenchment index. Tobin’s q is the ratio of the market value of assets to the book value of

assets, where the market value of assets is computed as book value of assets plus the market value of common stock less the sum of book value of common stock and balance sheet deferred taxes. Industry-adjusted Tobin’s q is

equal to Tobin’s q minus the median Tobin’s q in the industry, where industry is defined by three-digit SIC Code. Entrenchment index i (i=1, 2, 3, 4, and 5-6) is equal to 1 if the firm has an entrenchment level i and 0 otherwise. Insider Ownership is equal to the fraction of shares held by officers and director. ROA is the ratio of net income to

assets. CAPEX/assets is the ratio of capital expenditures to assets. RnD per Sales is the ratio of research and development expenditures to total sales. Leverage is the ratio of long-term debt plus debt due in one year to assets. Columns 1 and 2 provide OLS robust estimates of regression with year fixed effects, and Columns 3 and 4

provide the results of regressions with fixed firm and year effects. Robust standards errors appear in the parentheses. Significance levels are indicated by *, **, and *** for 10, 5, and 1 percent respectively.

(1) (2) (3) (4) Entrenchment Index E -0.008 0.012 (0.011) (0.016) Entrenchment Index 1 0.039 -0.021 (0.096) (0.105) Entrenchment Index 2 0.238** -0.035 (0.0634) (0.113) Entrenchment Index 3 0.183* 0.0446 (0.0540) 0.112 Entrenchment Index 4 0.165* 0.054 (0.057) 0.116 Entrenchment Index 5-6 0.165* 0.044 (0.065) 0.119 Log(Assets) -0.086 -0.009 -0.149** -0.146** (0.009) (0.009) (0.053) 0.0526 Log (Company Age) -0.039* -0.037* -0.213** -0.211**

(0.020) (0.019) (0.049) 0.107 Delaware Incorporation -0.080* -0.083* -0.087* -0.090*

(0.039) (0.021) (0.049) 0.049 Insider Ownership 0.0011 0.001 0.072 0.007 (0.002) (0.002) (0.005) 0.004 Insider Ownership Squared -0.00004 -0.00001 -0.00003 -0.0003

(0.00002) (0.00002) (0.00004) 0.00004 ROA 2.100*** 2.108*** 0.744*** 0.749*** (0.274) (0.276) (0.151) 0.151 CAPEX/Assets 1.109*** 1.145*** 0.595 0.546 (0.157) (0.146) (0.493) 0.489 Leverage 0.008 0.004 -0.308** -0.304** (0.136) (0.138) (0.144) 0.146 R&D per Sales 0.178*** 0.180*** 0.039* 0.038**

(0.039) (0.039) (0.021) 0.021 Year fixed effects Yes Yes Yes Yes Firm fixed effects No No Yes Yes Number of observations 2722 2722 2722 2722 R_squared 0.1480 0.1502 0.0160 0.0160

(38)

TABLE A.4: THE ENTRENCHMENT INDEX AND NET PROFIT

MARGIN

This table reports fixed-effects regressions net profit margin on two specifications of the Entrenchment Index and control variable, book-to-market Ratio. Net profit margin is the ratio of the Net Income to the Sales. The Entrenchment Index i (i=1, 2, 3, 4, and 5-6) is equal

to 1 if the firm has an entrenchment level i and 0 otherwise. Columns 1 and 2 provide OLS robust estimates of regression with year fixed effects, and Columns 3 and 4 provide the results of regressions with fixed firm and year effects. Coefficient of the Book-to-Market Ratio and the constant were omitted. Robust standards errors appear in the parentheses.

Significance levels are indicated by *, **, and *** for 10, 5, and 1 percent respectively. (1) (2) (3) (4) Entrenchment Index E 0.011 0.015 (0.009) 0.014 Entrenchment Index 1 -0.078* -0.040 (0.369) (0.300) Entrenchment Index 2 -0.125*** -0.041 (0.0218) (0.029) Entrenchment Index 3 -0.237** -0.057*** (0.062) (0.012) Entrenchment Index 4 -0.153** -0.009 (0.036) (0.045) Entrenchment Index 5-6 -0.140** -0.019 (0.033) (0.043) Year fixed effects Yes Yes Yes Yes Firm fixed effects No No Yes Yes Number of observations 4822 4822 4822 4822 R_squared 0.0001 0.0004 0.0003 0.0007

(39)

A.1. 15

TABLEA.5: MONTHLY ABNORMAL RETURNS

This table documents the monthly abnormal returns, and their associated robust standard errors in parenthesis, associated with different trading strategies for the period of January 2007 - December 2015. The monthly abnormal returns where calculated using two different methods. In the baseline model, abnormal returns were calculated by regressing the return associated with a particular trading strategy on the three Fama-French (Fama and French 1993) – the HML factor which captures book-to-market effects, the SMB factor which captures firm size effects and

the value-weighted market return in excess of the risk-free rate for further explanation) – and a momentum factor which was calculated using the procedures described in Carhart (1997). The trading strategies analyzed consist of going long a portfolio of stocks with a certain entrenchment index score and, simultaneously,shorting another portfolio of stocks with a higher entrenchment score. The long and short portfolios of stocks were constructed using equal weightings of each stock (equal-weight) and by weighting the holding of a stock in the portfolio by its

common stock market capitalization(value-weight). With industry-adjusted returns, the monthly abnormal returns were calculated by first subtracting from each firm’s monthly stock return the median industry return for

the industry in which the firm operates. The 3-digit SIC code was used for industry classification.Robust standards errors appear in the parentheses. Levels of significance are indicated by *, **, and *** for 10, 5, and 1

percent respectively.

Baseline model Industry -adjusted Long-Short Portfolios Equal-Weight Value-Weight Equal-Weight Value-Weight Index 0 - Index 5-6 0.0093101* 0.0257619* -0.028625 -0.0016754 (0.0055496) (0.0135484) (0.0033966) (0.0032585) Index 0 - Index 4-5-6 0.0089929* 0.0241692* -0.0093756 -0.0083175 (0.005424) (0.0127664) (0.0069749) (0.0068125) Index 0-1 - Index 4-5-6 0.0090102* 0.0241299* -0.0012223 0.0000724 (0.0054035) (0.0127519) (0.0019189) (0.0001419) Index 0-1 - Index 3-4-5-6 0.00907* 0.0221949* -0.0021476 0.000611 (0.005393) (0.011899) (0.00193) (0.0001408) Index 0-1-2- - Index 3-4-5-6 0.00904* 0.0229285* 0.0003744 -0.0000462* (0.005436) (0.0122245) (0.000669) (0.0000261)

(40)

TABLEA.6: THE ENTRENCHMENT INDEX AND TOBIN’S Q FOR

YEARS 2010-2015

This table reports fixed-effects regressions of industry-adjusted Tobin’s q on various controls and a specification of the entrenchment index. Tobin’s q is the ratio of the market value of assets to the book value of assets, where the market value of assets is computed as book value of assets plus the market value of common stock less the sum of

book value of common stock and balance sheet deferred taxes. Industry-adjusted Tobin’s q is equal to Tobin’s q minus the median Tobin’s q in the industry, where industry is defined by three-digit SIC Code. Entrenchment

index i (i=1, 2, 3, 4, and 5-6) is equal to 1 if the firm has an entrenchment level i and 0 otherwise. Insider Ownership is equal to the fraction of shares held by officers and director. ROA is the ratio of net income to assets. CAPEX/assets is the ratio of capital expenditures to assets. RnD per Sales is the ratio of research and development

expenditures to total sales. Leverage is the ratio of long-term debt plus debt due in one year to assets. Column 1 provides OLS robust estimates of regression with year fixed effects, and Column 2 provides the results of regressions with fixed firm and year effects. Robust standards errors appear in the parentheses. Significance levels

are indicated by *, **, and *** for 10, 5, and 1 percent respectively.

(1) (2)

Entrenchment Index E 0.027 -0.013

0.029 -0.055

Log(Assets) -238** -0.235**

0.089 0.090

Log (Company Age) -0.152 -0.161

0.153 0.154

Delaware Incorporation -329 -0.318

0.201 0.202

Insider Ownership 0.003 0.004

0.010 0.011

Insider Ownership Squared 0.00008 -0.00007 0.0002 0.00225 ROA 0.792 0.791*** 0.254 0.256 CAPEX/Assets 0.707 0.693 0.716 0.714 Leverage -0.456 -0.459** 0.228 0.229

R&D per Sales -0.244 -243

0.4121 0.413

Year fixed effects Yes Yes

Firm fixed effects No Yes

Number of observations 1466 1466

(41)

(42)

A.1. 17

TABLEA.7: THE ENTRENCHMENT INDEX AND TOBIN’S Q - BIG

AND SMALL CAPS

This table reports fixed-effects regressions of industry-adjusted Tobin’s q on various controls and a specification of the entrenchment index. Tobin’s q is the ratio of the market value of assets to the book value of assets, where the market value of assets is computed as book value of assets plus the market value of common stock less the sum of

book value of common stock and balance sheet deferred taxes. Industry-adjusted Tobin’s q is equal to Tobin’s q minus the median Tobin’s q in the industry, where industry is defined by three-digit SIC Code. Entrenchment

index i (i=1, 2, 3, 4, and 5-6) is equal to 1 if the firm has an entrenchment level i and 0 otherwise. Insider Ownership is equal to the fraction of shares held by officers and director. ROA is the ratio of net income to assets. CAPEX/assets is the ratio of capital expenditures to assets. RnD per Sales is the ratio of research and development

expenditures to total sales. Leverage is the ratio of long-term debt plus debt due in one year to assets. Columns 1 and 3 provide OLS robust estimates of regression with year fixed effects, and Columns 2 and 4 provides the results

of regressions with fixed firm and year effects. Columns 1 and 2 presents results for companies with market capitalization above 10 billion USD and columns 3 and 4 present results for companies with market capitalization

below 2 billion USD. Robust standards errors appear in the parentheses. Significance levels are indicated by *, **, and *** for 10, 5, and 1 percent respectively.

(1) (2) (3) (4) Entrenchment Index E -0.051* 0.014 -0.004 -0.019

0.017 0.027 0.0147 0.023 Log(Assets) -0.131*** -122 -0.124*** -0.130

0.0185 0.110 0.022 0.083 Log (Company Age) 0.0183 -0.244 -0.076** -0.163

0.025 0.291 0.022 0.163 Delaware Incorporation -0.119*** -0.148* -0.074** -0.016*

0.024 0.070 0.017 0.069 Insider Ownership -0.027** -0.19 -0.0005 0.011 0.008 0.016 0.002 0.007 Insider Ownership Squared 0.0015*** 0.0009 0.00002 -0.00006

0.0003 0.0005 0.00003 0.0005 ROA 2.728*** 0.791 1.485*** 0.653*** 0.415 0.644 0.259 0.171 CAPEX/Assets 2.101* 0.284 0.806* 0.848 0.870 1.485 0.271 0.680 Leverage -0.288* -0.672 0.084 -0.214 0.0961 0.326 0.198 0.204 R&D per Sales -0.088 0.085 0.122* 0.039 0.241 0.429 0.038 0.028 Year fixed effects Yes Yes Yes Yes Firm fixed effects No Yes No Yes Number of observations 746 746 1386 1386 R_squared 0.2879 0.1324 0.1481 0.1024

(43)

TABLE A.8: THE ENTRENCHMENT INDEX AND TOBIN’S Q FOR

YEARS 2010-2015 - BIG CAPS

This table reports fixed-effects regressions of industry-adjusted Tobin’s q on various controls and a specification of the entrenchment index. Tobin’s q is the ratio of the market value of assets to the book value of assets, where the market value of assets is computed as

book value of assets plus the market value of common stock less the sum of book value of common stock and balance sheet deferred taxes. Industry-adjusted Tobin’s q is equal to

Tobin’s q minus the median Tobin’s q in the industry, where industry is defined by three-digit SIC Code. Entrenchment index i (i=1, 2, 3, 4, and 5-6) is equal to 1 if the firm has

an entrenchment level i and 0 otherwise. Insider Ownership is equal to the fraction of shares held by officers and director. ROA is the ratio of net income to assets. CAPEX/assets

is the ratio of capital expenditures to assets. RnD per Sales is the ratio of research and development expenditures to total sales. Leverage is the ratio of long-term debt plus debt due in one year to assets. Column 1 provides OLS robust estimates of regression with year

fixed effects, and Column 2 provides the results of regressions with fixed firm and year effects. Table reports results only for companies with market capitalization above 10 billion

USD. Robust standards errors appear in the parentheses. Significance levels are indicated by *, **, and *** for 10, 5, and 1 percent respectively.