The influence of international cross-listings on the elements of CEO compensation

(1)

The influence of international

cross-listings on the elements of

CEO compensation

By Youri Zwart Bsc

Student number: 10286861

MSc Accountancy & Control, variant Accountancy Amsterdam Business School

Faculty of Economics and Business

Thesis supervisor: Dr. Alexandros Sikalidis Date: June 20, 2015

(2)

i

Statement of Originality

This document is written by Youri Zwart 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.

(3)

ii

‘The salary of the chief executive of a large corporation is not a market award for

achievement. It is frequently in the nature of a warm personal gesture by the

individual to himself.’

(4)

iii

Abstract

This thesis examines the relation between cross-listing and the elements of CEO compensation. Starting from a sample of UK firms I analyze the effect of cross-listing on the elements of CEO compensation (total, short-term, long-term, salary, bonus). I hypothesize that cross-listing would have a positive effect on the various elements of CEO compensation, as CEO’s that are able to run an international orientated firm are scarce and de demand is high. While the univariate analysis showed that cross-listed firms pay their CEO’s significantly higher compensation compared to non-cross-listed firms, the multivariate regression showed otherwise. In particular, the regressions analysis demonstrated that cross-listed does not have significant difference on total, short-term, long-term, and bonus CEO compensation. The regression with CEO salary as a dependent variable showed that cross-listing has negatively effect on CEO salary, the salary of a CEO of a cross-listed firm is $35.000 lower than a CEO of a non-cross-listed firm. These results are inconsistent with earlier research, Oxelheim and Randøy (2005) found a significant postive effect of cross-listing on total CEO compensation.

(5)

iv

Acknowledgements

Though only my name appears on the cover of this thesis, a great many people have contributed to its production. I owe my gratitude to all those people who have made this thesis possible. I would like to thank my thesis supervisor Dr. Alexandros Sikalidis for his support, knowledge and useful feedback throughout the duration of the master thesis.

Additionally, I would like to thank KPMG, my colleagues and fellow students, for the

opportunity as well as their support. I am looking forward to start as trainee at Corporate Clients in September.

Many friends have supported me over the years of my study. Their support and care helped me overcome setbacks and stay focused on my graduate study

Last but not least important, I owe more than thank to my parents and two brothers. They were always supporting me and encouraging me with their best wishes.

All faults are my own.

Youri Zwart June 2015

(6)

v

1. Introduction

The determinants of compensation of the chief executive officer (CEO) have been a topic of interest and the subject to criticism both in the academic literature and popular press. Just recently, when the ABN Amro Bank proposed salary increase of 100.000 euro for her managers, it caused great public outcry. Finance Minister Jeroen Dijsselbloem said the salary raise sends to wrong signal and he even decided to postpone the bank’s sale decision (Hensen & Heijden, 2015; Witt Wijnen, 2015). The decision about ABN Amro’ IPO would be made before April 1st (Jaarsveldt, 2015). The spotlight has tended to focus on CEO pay, because the CEO is usually the highest paid and most visible executive of a company (Nichols & Subramaniam, 2001). While the discussion in the popular press primarily concentrates on the fairness of CEO compensation, the academic literature focuses on several factors that can influence the height and structure of compensation. Some examples are board size, firm performance, board composition and firm size.

At the same time with markets’ globalization, more and more firms tend to cross borders to attract capital but also expand their activities; building factories abroad, hiring foreign personnel and increasing their sales in foreign markets. Through foreign board membership and the international product/service market firms are becoming more international, cross-listing is another step in this internalization process. Cross-listing means that firms list their shares on multiple exchanges in different countries. In recent years the number of firms that cross-listed their shares to the U.S. stock exchanges has increased significantly (Karolyi, 1998) So far, cross-listing is extensively studied, but these studies are focused on the impact of cross-cross-listing on the firm. But the topic of the impact of cross-listing on the top management and/or the CEO compensation remains unaddressed. The aim of this study is to focus on that gap in the literature and answer the following question: Does cross-listing influence CEO compensation? By answering this question, I will extend the existing literature related to CEO compensation aiming to provide evidence, which will be of help to compensation committees in their effort to develop a more effective compensation scheme. Using a multivariate analysis I examine whether CEO compensation is affected by the cross-listing policy of the firm and whether the cross-listings affects is short-term or long-term compensation.

(9)

2

The remainder of the paper is structured as follows. The next section contains the literature review and the hypotheses development. Then, the methodology of research is explained, including the sample that is used, the measurement of control variables, and the statistical method used. In the fourth section, the results of this study are presented. Finally, the fifth section consists of a conclusion, possible limitations of the research and in addition suggestions for further research will be mentioned.

(10)

3

2. Literature review and hypotheses development

In this section, the existing literature and theory related to the research problem will be reviewed. First, the theory behind the CEO and CEO compensation will be discussed, continued by the different ways to cross-list securities and the reasons for cross-listing. This chapter will be concluded by the hypotheses development.

2.1 Chief Executive Officer

In this chapter the background theory of the CEO will be discussed. This includes the separation of ownership and agency theory.

2.1.1 Theoretical background 2.1.1.1 Separation of ownership

The publicly listed companies have two main organizational characteristics: the separation of decision management from residual risk bearing and separation of ownership and control (Fama & Jensen, 1983). The companies are owned by the stockholders, but are controlled by the Board of Directors. There are two types of boards: one-tier and two tier boards. In the one!tier board model both executive (decision management) and non!executive (decision control) directors cooperate within one and the same board. In the two-tier board, the decision management and decision control are in separate boards. The decision management is the task of the Board of Directors and is accountable to the Supervisory Board (decision control).

2.1.1.2 Agency Theory

The separation of ownership and control can lead to conflicts and problems; this problem is discussed in the agency theory. This theory describes the principal agent relationship as a contract under which the principal engages agent to perform a service on their behalf for which

(11)

4

delegation of certain decision making to the agent is necessary (Jensen & Meckling, 1976). The agency is based on several assumptions (Eisenhardt, 1989). The agency theory assumes that individuals maximize own utility, which means that the agent is not always motivated to behave in the best interest of the principal or firm. In order to align the interests of the agent with those of the firm principal, the principal can create incentives for the agent to narrow this misalignment. The principal can do this through variable compensation and/or stock options contracts and by monitoring the behavior of the agent (Eisenhardt, 1989).

Furthermore, according to the agency theory there is an information gap between the agent and the principle. This means that the agent has more information about his real performance than the principle. To obtain this extra information the principle has to make monitoring costs. The monitoring costs are the cost of measuring or observing behavior, but also include costs incurred to ‘control’ the agent via operating rules, budget limits and compensation policies (Jensen & Meckling, 1976).

Moral hazard and adverse selection are also the result of the information problem (Eisenhardt, 1989). The fact that the principle cannot fully observe the actions and decisions of the agent can result in moral hazard. The agent becomes a ‘moral hazard’ when he does bear the fully consequences of his actions and therefore acts careless and ‘wild’. Adverse selection occurs upfront, before the agent signs his contract. The agent has more information about his skills and capabilities then the principle. The agent can take advantage of this ‘hidden’ information. This makes it difficult for the principle to select the right agent (Eisenhardt, 1989).

To minimize the agency costs, the principle can try to ensure that the agent will not harm the firm and align his interests with the agent’s. This will result is bonding cost, which reflect the costs incurred to guarantee to outside stock holders that the executive directors are limited in behavior costly to the firm.

Following the agency theory, the management and the stockholder of the company are the agent and respectively the principle. There are several options to solve the agency problem, one of the options to connect reward of the CEO to a performance measure that is aligned with the stockholder. Lilling (2006) states that in order to combat the principle-agent problem, the

directors of public companies use incentive-based contracts to align the interests of CEOs and shareholders. Firm performance can be such a measurement; this will be further discussed in paragraph 2.2.3.2.

(12)

5

2.2 CEO Compensation

This section will describe the elements of CEO compensation as well as the theories and factors influencing CEO compensation.

2.2.1 Elements of CEO compensation

CEO compensation is divided into three parts: cash compensation, equity compensation and other compensation. All three parts will be discussed in this paragraph.

2.2.1.1 Cash compensation

Cash compensation consists of the base salary and the cash bonus. The basic salary is the fixed salary and set in the beginning of the year. Salary can change each year due to inflation or based on good performance last year. The cash bonus is additional payment that a CEO receives if the company meets a specified target. The targets are short-term incentives and are accounting based (Lacker & Tayan, 2011).

2.2.1.2 Equity compensation

The equity bonus is stock based compensation. Stock options that are awarded to the CEO, give the CEO the right to purchase stock of the company at a pre-set price during a certain time period. These stock options motivate the manager to increase the stock price, because his bonus is the difference between the pre-set price and the price when he exercises the option. The stock option rewards has both benefits and drawbacks. On one hand, the stock options reduce the conflict of interest, but on the other hand the stock option rewards can also cause moral hazard problems (Chhabra, 2008). This is due to the asymmetric incentive, the CEO is awarded for good performance, but is not ‘punished’ for bad performance. This asymmetric information can cause the CEO to take more risk, which can cause bad decision-making (Rajgopal & Shevlin, 2002).

(13)

6

that is based on the long-term average stock price over the life of the option. Chharbra (2008) research shows that the implementation of this change addresses the moral hazard issue. Another option to address to moral hazard of the stock option is to simply reward the CEO with stock instead of stock options. The advantage of rewarding the CEO with stock is that the CEO will become more risk neutral because in contrary with the options the CEO gets ‘punished’ when the stock price drops.

2.2.1.3 Other Compensation

Other compensation of CEO refers to pension plans, health insurance, company car, company housing, private telephone use, and severance pay (Igalens & Roussel, 1999).

2.2.2 Theories on CEO compensation

In the academic literature there are several theories regarding CEO compensation. The market of CEO’s has a supply and a demand side. These can be explained using two theories, the Human Capital Theory and the Managerial Discretion Theory respectively. Both theories will be discussed below.

2.2.2.1 The Human Capital Theory

The human capital refers to the skills a CEO has and the fact that the CEO needs to be compensated for these skills. If a job has any requirements that limit the supply of the CEO candidates, then that would enhance the CEO pay. A CEO needs more skills to manage an international firm than to manage a domestically orientated firm, therefore a CEO who has these skills to manage an international firm are rare (Oxelheim & Randøy, 2005).

The paper of Harris and Helfat (1997) investigate the relationship between human capital and CEO compensation. They analyze the difference in compensation between internal and external successors and attribute this to differences in skill specificity. The results support the

(14)

7

proposition that differential skill specificity is associated with pay premiums to external successor

2.2.2.2 The Managerial Discretion Theory

Managerial discretion refers to the latitude of action or potential strategic options executives have. High-discretion contexts increase potential CEO impact on organizational outcomes, hence, their ability to directly influence firm performance (Finkelstein & Boyd, 1998). As a result, the absolute amount of CEO compensation is expected to be higher the greater the level of discretion. Hambrick and Finkelstein (1987) in Wangrow, Schepker and Barker (2015) identified three basic sources of managerial discretion: the task environment, the organization, and the CEO himself or herself. Finkelstein and Boyd (1998) investigated the relation between managerial discretion and compensation. They found a positive relationship between the CEO compensation and managerial discretion and also found strong performance will enhance this relationship.

2.2.3 Factors influencing CEO compensation

This paragraph describes the various factors that may influence CEO compensation. These factors are also implemented as control factors in the regression.

2.2.3.1 Size

The size of the company influences the level of CEO compensation. Tosi, Werner, Katz, & Gomez-Mejia (2000) find that organizational size is an imporatant determinants of CEO compensation. Their results show that firm size accounts for 40% of the variance in total CEO pay. Nourayi & Daroca (2008) the relationship between CEO total compensation and firm size, as well as the relationship CEO cash compensation and firm size. They used sales as the proxy for firm size and their regression shows that firm size is a significant variable. The paper of Conyon & Swalbach (2000), further discussed in paragraph 2.2.3.2, find that CEO pay is

(15)

8

positively influenced by company size. This relationship holds for the UK as well as Germany. The academic literature justifies the size premium, trough the human capital theory, which is discussed above.

2.2.3.2 Performance

Firm performance may also potentially influences CEO compensation. The paper of Lilling (2006) focuses on the relationship between market value and CEO compensation and finds that there is a positive relationship between market value of the firm and CEO compensation. Lilling (2006) concludes that the incentive based contracts are effective. Moreover, he concludes that CEO should be motivated by incentive-based contracts because much of his salary depends on the company’s performance. Jensen and Murphy (1990) also performed a study on firm performance and CEO compensation. They hypothesized that pay and performance were unrelated. They found that overall executive pay rose by $3.25 for every $1,000 increase in shareholder wealth. Although the results indicated that the relationship was positive and statistically significant, the authors still consider this amount immaterial. Hall and Liebman (1998) also investigated the correlation between firm performance and CEO compensation, they document a strong relationship between firm performance and CEO compensation. The paper of Conyon & Schwalbach (2000) investigated the determination of executive pay in the UK and Germany. They look at data over a large timeframe. The most important difference between the two countries lays in the structure of CEO pay, the executives from the UK received stock options, while the German executives did not. However their regression analysis shows a positive and significant relation between cash compensation and company performance. They also find that the executive has increased in both economies, but the UK has had a faster growth rate. Nourayi & Daroca (2008) also investigated the relationship between performance and CEO compensation (cash and total), they divided performance into market performance (total stock return) and accounting performance (ROA). They find that market performance is statistically significant for both cash compensation and total compensation. But accounting performance was only partially significant. The accounting performance was significant for cash compensation, when they partitioned their sample of regulated firms, but not for total compensation.

(16)

9 2.2.3.3 Leverage

Leverage is another factor; the leverage is an economic determinant for financial distress. Financing the company with more debt can have two effects on the compensation of the CEO. On one hand the debt holder may monitor the CEO more intense, resulting in a lower CEO compensation. On the other hand more debt can increase company risk.

Papa & Speciale (2011) studied the effects of financial leverage on managerial pay. Their results show that for firms with high levels of financial leverage it is less easy to create an incentive scheme, which aligns the interest of the manager with that of the shareholder. So, when a firm has a high level of debt, it becomes harder to measure the effort of the manager through total shareholders return. Papa & Speciale (2011) claim that this could be one of the reasons why during the crisis, while the banks had to be bailed out with tax payers money, the banks still paid high bonuses to their managers. They simply could not observe the manager; this can be seen as the information gap as referred to in paragraph 2.1.1.2.

2.2.3.4 Firm Risk

Firm risk is another factor affecting CEO compensation. Miller, Wiseman, & Gomez-Mejia (2002) examined the effects of unsystematic and systematic firm risk on CEO compensation. They used four proxies as measures for the firm risk: systematic and unsystematic market risk (beta and sigma taken from the CAPM), systematic and unsystematic income risk (calculated same way as market risk, only with the ROA instead of stock price). They find a relationship between the CEO compensation and the amount of risk a firm is facing. This relationship is stronger between firm risk and CEO compensation, compared to market risk and CEO compensation.

(17)

10 2.2.3.5 Industry sector

The industry sector in which a company operates can also influence the CEO compensation, some compensation elements can be more or less common in a specific industry. I can image that if you work in transportation a company car is provided, while when you are working in insurance sector this is not necessary.

2.2.3.6 CEO age

The specifics of the CEO do not only influence the level of compensation, but also the structure of the compensation. Adhikari, Bulmash, Krolikowski, & Sah (2015) responded to the ongoing debate regarding the hiring and compensation of younger versus older employees and investigated the CEO. They state that older CEO’s are seen as more reliable, efficient, thrustworthy and more risk adverse and find that the older CEO’s receive significantly higher compensation compared to the younger ones. Adhikari et al. (2015) also note that another reason could be that in order to motivate older CEO’s, who raised substantialy capital over time, to keep working and stay motivated, their compensation package must be highly competative.

2.3 How to cross-list securities in the U.S.

There is a number of different techniques to cross-list shares on to the American stock exchange as is explained in the paper of Karolyi (1998). Karolyi (1998) distinguishes two options to cross-list to the United States: The typical cross-listing and depositary receipts.

2.3.1 The typical listing

To list their shares using this method firms satisfy two requirements. First, they must qualify for listing according to standards set for overseas companies by the exchanges. The standards include certain income and shares figures. Second, the new listings in the U.S. must comply fully

(18)

11

with GAAP (Generally Accepted Accounting Principles) reporting and Securities and Exchange Commission (SEC) registration Requirements (Karolyi, 1998).

2.3.2 Depositary receipts

Depositary receipts (DRs) are negotiable certificates that indirectly represent ownership of shares in corporation for domestic investors. These certificates denote depositary shares that represent a specific number of underlying shares remaining on deposit in the issuer’s home market. DRs were developed by JP Morgan, in 1927, as a vehicle for investors to register and earn dividends on non-U.S. stock without direct access to the local market itself, all the dividend and payments are converted into US dollars (Karolyi, 1998). American Depositary Receipts are traded on a U.S exchange. There are four different types of ADRs, each with his own purpose, laws it has to apply to and markets on which they are traded. These ADR facilities consist of three levels of public offerings as well as private placement (Karolyi, 1998).

Level I ADR’s: These ADR securities are traded over the counter (OTC) with prices published in the Pink Sheets and are unlisted in the U.S. stock exchanges. The foreign firm isn’t subjected to full SEC disclosure and no U.S. GAAP reconciliation is required. The purpose of these ADR’s is to broaden shareholder base (Karolyi, 1998).

Level II ADR’s: These are listed a on the major U.S. stock exchanges (NYSE, AMEX, NASDAQ). The foreign firm is subject to partial U.S. GAAP reconciliation and SEC registration. These ADR’s are used to broaden shareholder base (Karolyi, 1998).

Level III ADR’s: These ADR’s are offered and listed in the major U.S. stock exchanges and the foreign firms are subject to full U.S. reconciliations and full SEC registration. The purpose of these ADR’s is to raise capital with new share issue (Karolyi, 1998).

Rule 144A ADR’s: These ADR’s are traded on U.S. Private Placement Market using PORTAL, which is a private electronic market on which only very large institutional investors can trade.

(19)

12

These ADR’s raise capital via placements (new shares) among Qualified Institutional Buyers (QIB) and do not have any U.S GAAP requirements or SEC registration (Karolyi, 1998).

2.4 Why do firms cross-list?

There is a great amount of theory and literature on the reasons for cross-listing. In this chapter, I will briefly outline the different hypotheses and theories behind cross-listing and the prior research will be discussed.

2.4.1 The market theory

Alexander, Eun and Janakiraman (1988) and Errunza and Losq (1985) examine the effect of market segmentation on the pricing of securities. The market segmentation hypothesis states that most investors have set preferences regarding the length of maturities that they will invest in. The securities with different maturity lengths cannot be easily substitute each other. If an investor chooses to invest outside their term of preference, they must be compensated for taking on that additional risk. Errunza and Losq (1985) show that ineligible securities command a “super-risk premium” when a subset of investors is restricted from investing in those securities. Securities that are eligible (with no restrictions on investors) are priced as if there were no segmentation. By cross-listing, these ineligible securities become eligible and lose their “super-risk premium” and therefore command a higher price.

Miller (1999) investigated the market reaction of international listings of ADRs around the announcement date. He did an event study on 181 stocks domiciled from 35 countries outside the U.S. in the period from 1985 to 1995. Miller (1999) measured the announcement effects on the stock price and finds a positive abnormal return. He also investigated institutional and geographical differences and finds that abnormal returns are largest for firms that list on major US exchanges such as NYSE or NASDAQ and smallest for firms that list on PORTAL. Miller suggests that indirect barriers such as liquidity risk and low investor recognition segment capital markets. Miller also investigated market reaction around the announcement of different levels of DR Programs. He finds that in the higher levels of DR programs, the increases in share value are

(20)

13

also higher. This is consistent with the hypothesis that more strict disclosure requirements, a greater shareholder base and liquidity results in a higher increase of share price. He concludes that the results provide empirical support for the hypothesis that dual listing can overcome capital barriers, resulting in a higher share price and a lower cost of capital.

2.4.2 The bonding hypothesis

Coffee (1999, 2002) and Stulz (1999) have a different hypothesis on the reason behind cross-listing, the bonding hypothesis. The bonding hypothesis is related to the agency theory. To solve the agency problem the management ‘bonds’ itself to a better governance regime, more rigorous standards then in the home country. Especially in the U.S. the management is subjected to much more regulation and monitoring. This limits the ability for private gains and provides opportunities for external financing and increases investor protection (King & Segal, 2003). This results in lower cost of capital, enhances the willingness to invest in a foreign firm and increases firm value. There are a number of studies the tested the bonding hypothesis. King and Segal (2004) examine the impact of cross-listing on the valuation of Canadian firms that list on both a Canadian and a U.S. stock exchange. They have two main findings; first, cross-listing has a positive impact on valuation over and above the positive impacts associated with firm size, profitability, cost of equity, and past sales growth. Second, the location of share trading is linked to valuation. Cross-listed firms that are traded actively in the U.S. market experience a significant increase of valuation over the long term. If firms that cross-listed aren’t actively traded in the U.S. market they are no differently from non-cross-listed Canadian firms.

In another paper Doidge, Karolyi and Stulz (2004) find that at the end of 1997, foreign companies with shares cross-listed in the U.S. had Tobin's q ratios that were 16.5% higher than the q ratios of non-cross-listed firms from the same country. The Tobin’s q ratio is even higher for firms that cross-list to the major U.S. exchanges up till 37%.

2.4.3 The signaling hypothesis

In the signaling hypothesis, the reason behind the cross-listing decision is to credibly convey their inside information to the uninformed outside world (Witmer, 2005). The signaling theory is

(21)

14

closely related to the bonding hypothesis when the inside information is about management benefits. However, under the bonding hypothesis, management is trying to change the incentives of the management’s benefits, whereas under the signaling hypothesis, the management is communicating the amount of management benefits.

2.4.4 The shareholder base hypothesis

Under this hypothesis the benefit of cross-listing doesn’t affect the existing shareholder base, the benefits comes from the fact that the company is able to expose itself to new investors in a new country, extending its shareholder base and decreasing its discount (Witmer, 2005).

Lee and Valero (2010) investigated whether international cross listing facilitates improvements in the information environment of foreign firms. They analyze whether analyst coverage intensity and recommendation dispersion changes when foreign firms cross-list their stocks as ADRs in the United States and find that the analyst coverage increases significantly, especially on organized exchanges and ADR’s that raise capital. The recommendation dispersion decreases, this is mainly observed by firms that non-capital raising ADR programs. Another paper from Baker, Nofsinger and Weaver (2002) investigated the relationship between cross-listing and visibility. This study shows that international firms listing their shares on the New York Stock Exchange (NYSE) or the London Stock Exchange (LSE) experience a significant increase in analyst coverage and print media attention, these two factors were used as proxies for visibility. The increased visibility is evidence for the shareholder base hypothesis, it proofs that through cross-listing the company is able to expose itself to new investors.

2.5 Research on international cross-listings

Hail and Leuz (2009) preformed a study on whether cross-listing in the U.S. reduces cost of capital. They use cost of capital estimates based on current market prices and analyst forecasts to measure the growth expectations around listings. They find strong evidence that cross-listings on U.S. exchanges are associated with a significant decrease in firms’ costs of equity capital. They do not find similar cost of capital effects for listings on the London Stock Exchange. Hail and Leuz (2009) also investigate cross-sectional differences in the cost of capital

(22)

15

effects and find that firms from countries with weak disclosure regulation and weak protection of outside investors against self-dealing by corporate insiders benefit the most from cross-listing on U.S. exchanges.

Perotti, Cordfunk and Lynch (1997) investigated whether foreign cross-listing increased firm value with a sample of Dutch firms. By measuring the stock market reaction on the announcement date of the foreign listing they find that the stock price increases, they say that stock price increase seems to be permanent, given the insignificant abnormal returns following the announcement and conclude that shareholders look forward to an increase in value.

So building on the previous literature, theory and all the reasons listed above, I can summarize that, through cross-listing on U.S. markets firms can: lower liquidity risk, raise stock price, reduces cost of capital, expands shareholder base, improves the information environment, improve visibility and enhance bonding and signaling. Although there are a bunch of different reasons for firms to cross-list their stocks, the common goal of all of them is to increase firm value and performance.

2.6 Hypotheses

In the literature and theory review above, I presented several theories explaining the motivation for cross-listing and the factors of CEO compensation. Apart from the control factors in paragraph 2.3.2, paragraph 2.2 above explains two theories of CEO compensation. From these theories I can conclude that the efficiency of the labor market determines executive compensation.

First, the human capital theory, which states that there is a smaller pool of CEO’s who are capable of running an internationally orientated rather than just a domestically orientated firm. This smaller pool results in an increase in CEO compensation. Second, the managerial theory, which states that higher influence of a CEO results in a higher compensation. Both the theories suggest that the CEO compensation is higher for the ‘international’ CEO’s. Third, is the literature discussed, the literature on cross listing shows a positive relationship between cross-listing and firm performance and the literature on CEO compensation shows better firm performance results in higher CEO compensation.

(23)

16

By connecting the literature of CEO compensation to the literature of cross-listing I expect that cross-listing can be positively linked to CEO compensation. There are some studies that examine whether CEO compensation and firm performance are linked. Lilling (2006) investigated this link between CEO compensation and firm performance. The paper uses the log of the market value, which is an intuitive way from an incentive point of view, the higher the market value for a company, the higher the CEO salary should be. Lilling (2006) finds empirical evidence that there is a positive link between the CEO compensation and the performance of the company.

Sanders and Carpenter (1998) did research on match between internalization and CEO pay. Their definition of internalization was: A firm's dependence on foreign markets for customers, factors of production and the geographical dispersion of this dependence. They found that the degrees of internationalization to be positively associated with CEO pay. Oxelheim and Randøy (2004) come up with another influence. They speak of a risk effect for the CEO, due to the harsher monitoring and risk of dismissal in an international firm. The risk effect also has a positive effect on CEO compensation.

Building upon the prior literature, the theory and the connection I make between the literature of CEO compensation and cross-listing, the following hypothesis is formulated.

Hypothesis 1: Cross-listing has a positive effect on CEO compensation

To further investigate the effect of cross-listing on CEO compensation further research is performed on the composition of the CEO compensation. This research will then investigate the horizon of executive performance and compensation. Managers should ideally seek for opportunities that enhance the long-term firm value, but short term value is needed for the firm to survive (Porter, 1992; Marginson & McAulay, 2008). Marginson & McAulay (2008) state that balancing the needs of both the long term and the short term is important for the firm and come up with two possibilities for managers: short-termism, which Mullins in Jacobs (1991) defines actions to secure short-term results that preclude term achievement and long-termism which are actions that focuses on the long term to the detriment of the short term (Marginson & McAulay, 2008). An important factor that influences the decision between

(24)

short-17

termism and long-termism is the capital market. Jacobs (1991) states that CEO’s of listed firms are pressured into long-termism in order to meet stock market expectations. So when a firms cross-lists into a more developed market, the pressure on the CEO to focus on long-termism increases. By focusing on long-term compensation firms motivate their managers to increase long-term firm value and encourage manager to look for projects, which increase the long-term firm value (Lacker & Tayan, 2011). These two factors lead to the following hypothesis.

Hypothesis 2: The compensation of cross-listed firms is more long-term orientated compared to non-cross-listed firms

(25)

18

3. Research Methodology

In this section the research design and method for testing the hypotheses are discussed. First, the sample will be discussed, followed by the measures of the dependent variable. Then, control variables are discussed and the last part covers the research design with the regression model.

3.1 Sample

This research focuses on the compensation of CEOs of publicly held cross-listed and non-cross-listed companies from the United Kingdom over the year 2010. The accounting data will be available on World Scope and the CEO compensation is available in the People Intelligence database of Compustat. To select the companies, all companies from the United Kingdom available in Word Scope are selected, from this sample the companies that are not listed on the London Stock Exchange (LSE) are subtracted. The data from World Scope and Compustat were then integrated to get to the dataset used in this research.

3.2 Measures

The dependent variable denotes the natural logarithm of CEO compensation. This includes all the major components of CEO remuneration: annual salary, cash bonuses, equity bonus and other compensation. Annual salary is the fixed salary that is paid to CEO and is set at the beginning of the year. The natural logarithm is used to mitigate the effect of outliers. The cash bonus is an additional payment a CEO receives when the company meets a specified accounting target, this is also defined as short-term compensation. (Lacker & Tayan, 2011). The natural logarithm short-term compensation is the dependent variable is the second regression. The natural logarithm is used to mitigate the effect of outliers. The equity bonus is restricted stock and stock options that are awarded to the CEO. These are measured at market value when the CEO receives them (Lacker & Tayan, 2011). The equity bonus is long-term compensation and depends on long-term market performance. The natural logarithm of the long-term compensation

(26)

19

is used for the third regression, the natural logarithm is used to mitigate the effect of outliers. Other compensation refers to pension plans, severance pay, health insurance and company car.

The dependent variable is lagged one year with respect to the independent variables. This is due the fact that CEO compensation is not constantly calculated. The one-year lag will result in a more accurate picture of the CEO pay. This is according earlier studies (Oxelheim and Randøy, 2004).

The independent variable selected is cross listing. To measure cross listing a dummy variable is used. A dummy variable can take on two values, 0 or 1. A ‘0’ indicates the absence and the ‘1’ indicates the presence of the variable. So this dummy variable equals 1 if a firm cross-lists and 0 otherwise.

3.3 Control variables

The first control variable is firms’ size, the proxy to measure Size is the natural logarithm of total market capitalization. The natural logarithm is used to mitigate the effect of outliers. This proxy is also used in the study of Oxelheim & Randøy (2005). Performance is another control variable in this study (Tosi et al., 1999; Core, Holthausen, & Larcker, 1999). Performance is divided in two measures; firms stock performance and accounting performance (Oxelheim & Randøy, 2005). The firms’ stock performance is measured by total stockholder return, which is computed as the sum of the increase in stock price and the dividends per share divided by beginning price per share (Abowd, 1990). The accounting performance is measured as the Return on Equity. The reason for this twofold is that is it common for compensation packages have elements of both accounting and firm performance, as discussed in paragraph 2.5.1. The third control variable is leverage. Leverage is defined as the ratio of total debt to total assets. Another potentially important determinant of CEO compensation is firm risk. Price volatility is used to measure as a proxy of Risk.

CEO age is a CEO specific variable for which need to be controlled in this study. Age is equal to the age in years of the CEO at the end of the fiscal year.

(27)

20

To control for the industry, a dummy variable is used. A dummy variable can take on two values, 0 or 1. A ‘0’ indicates the absence and the ‘1’ indicates the presence of the variable. The companies in the sample are divided over six different industries using the general industry classification is World Scope. The classifications consist of: Industrial (1), Utility (2), Transportation (3), Banks/Savings & Loans (4), Insurance (5) and Other Financial (6). These six classifications will lead to six dummies to control for the industry.

3.4 Research Design

Drawing on previous research on cross-listing and CEO compensation, the model for testing the hypotheses was developed with a variety of independent variables to minimize specification bias (Oxelheim & Randøy, 2005). I use a cross-sectional ordinary least-square (OLS) regression to test the hypothesis presented in the preceding section. This results in the following models. (1): CEO compensation (year) = α + β1*Cross-listing + β2*Size + β3*Accounting

Performance + β4*Stock Performance+ β5*Leverage + β6*Risk + β7*Age + β8*Industry

(2): CEO long-term compensation (year) = α + β1*Cross-listing + β2*Size + β3*Accounting

Performance + β4*Stock Performance+ β5*Leverage + β6*Risk + β7*Age + β8*Industry

(3): CEO short-term compensation (year) = α + β1*Cross-listing + β2*Size + β3*Accounting

(28)

21

4 Results

In this section I will discussed the descriptive statistics as well as the results of the ordinary least squares (OLS) regression.

4.1 Descriptive statistics

Table 1 panel A reports the descriptive statistics of the total sample, the panel shows that mean of the sample of CEOs total compensation and cash-compensation are just over $0,5 million and equity-compensation around $50,000. The mean of the market capitalization is $57 million and the mean of the market yield is slightly negative (-0.70), which means that on average the share price has dropt a little bit. The dividend payout mean is $0.07, so not a lot of companies have paid out any dividend in the year of the sample. The ROE amounts to 11.2% and the leverage is 18,2%, which is quite low. The volatility is around $35.14 and the average CEO age is 53.

Panel B and panel C describe the descriptive statistics of the cross-listed and non-cross-listed firms. Looking at the means of the variables one can see that the mean of total compensation, short-term (cash) compensation and long-term compensation are larger for cross-listed firms than for non-cross-cross-listed firms. Another observation I make is that cross-cross-listed firms are significantly larger than non-cross-listed firms. Furthermore, the panels show that the leverage is higher for cross-listed firms compared to non-cross-listed firms. This suggest that cross-listed firms rely more heavily on debt financing and contradicts the theory that cross-listed firms have greater access to the equity market. Finally, the results show that the volatility of cross-listed firms is higher compared to non-cross-listed firms. Table 2 shows that all these differences apart from dividend are significant.

Table 3 shows the composition of the sample among the general industry classification. The panel shows that in this sample the cross-listing is very successful. In almost all the classification categories there are more cross-listed firms than non-cross-listed firms, classification category 6 is the only category, which is the exception.

(29)

22

Panel 4 shows Pearson-Watson correlation matrix, most of the coefficients in the matrix are below 0,50, so in general no multi-collinearity exists between the variables. Multi-collinearity means that there is a high correlation between several independent variables. There are some variables that have a high correlation factor, the different forms of compensation are highly correlated, which makes sense because total compensation is the sum of short-term and long-term compensation. Market capitalization and total and short-long-term compensation are also correlated.

(30)

23

Table 1 Descriptive statistics

Panel A: Descriptive statistics of the full sample

Variable Observations Mean Std. Dev. Min Max

Dummy cross-listing 1490 0.557 0.497 0 1 Ln(Total compensation) 444 13.133 1.154 7.601 16.187 Ln(Short-term compensation) 437 14.144 1.078 7.601 15.576 Ln(Long-term compensation) 409 10.797 1.548 6.658 15.665 Ln(Market-capitalization) 1490 17.862 2.354 11.608 25.466 Market Yield 461 -0.707 1.392 -8.5 1.84 Dividend 1490 0.0688 0.147 0 0.78 ROE 427 0.113 0.195 -0.766 0.9753 Leverage 467 0.182 0.179 0 0.966 Volatility 1235 35.117 14.222 8.39 74.25 CEO Age 430 52.705 6.717 37 73 General Ind. 2 1490 0.232 0.154 0 1 3 1490 0.012 0.106 0 1 4 1490 0.006 0.078 0 1 5 1490 0.013 0.115 0 1 6 1490 0.262 0.440 0 1

(31)

24

Panel B: Descriptive statistics of cross-listed sample

Variable Observations Mean Std. Dev. Min Max

Ln(Total compensation) 288 13.495 1.034 7.601 16.188 Ln(Short-term compensation) 285 13.486 0.973 7.601 15.576 Ln(Long-term compensation) 275 11.028 1.622 6.658 15.665 Ln(Market-capitalization) 830 0.073 0.145 0 0.78 Market Yield 304 -0.809 1.506 -8.16 1.84 Dividend 830 0.073 0.145 0 0.78 ROE 281 0.140 0.199 -0.766 0.975 Leverage 309 0.195 0.181 0 0.966 Volatility 665 37.775 13.511 13.86 72.63 CEO Age 292 52.253 6.182 37 70 General Ind. 2 830 0.033 0.178 0 1 3 830 0.016 0.124 0 1 4 830 0.011 0.104 0 1 5 830 0.022 0.150 0 1 6 830 0.148 0.356 0 1

(32)

25

Panel C: Descriptive statistics of non-cross-listed sample

Variable Observations Mean Std. Dev. Min Max

Ln(Total compensation) 156 12.465 1.062 8.517 15.084 Ln(Short-term compensation) 152 12.503 0.968 8.854 15.084 Ln(Long-term compensation) 134 10.332 1.262 6.908 13.522 Ln(Market-capitalization) 660 16.885 1.916 11.608 23.326 Market Yield 157 -0.507 1.114 -8.5 1.167 Dividend 660 0.063 0.148 0 0.78 ROE 146 0.060 0.178 -0.719 0.714 Leverage 158 0.158 0.172 0 0.8 Volatility 570 32.017 14.413 8.39 74.25 CEO Age 138 53.659 7.666 40 73 General Ind. 2 660 0.013 0.116 0 1 3 660 0.006 0.077 0 1 5 660 0.001 0.039 0 1 6 660 0.406 0.491 0 1

(33)

26

Table 2: T-test of difference in sub-samples

Difference Mean Non CL Mean CL P-Value

Ln(Total compensation) -1.030 12.465 13.495 0.000*** Ln(Short-term compensation) -.983 12.503 13.486 0.000*** Ln(Long-term compensation) -.705 10.322 11.028 0.000*** Ln(Market-capitalization) -1.754 16.884 18.639 0.000*** Market Yield 0.303 -0.507 -0.810 0.026** Dividend -0.010 0.063 0.073 0.188 ROE -0.079 0.060 0.140 0.000*** Leverage -.037 0.158 0.195 0.035** Volatility -5.758 32.017 37.775 0.000*** CEO Age 1.406 53.659 52.253 0.043**

Significance levels t-statistics * p<0.1 **p<0.05 ***p<0.01 Variable definitions are shown in the appendix

(34)

27

Panel D: Pearson's correlation test

Variables L(TC) Ln(STC) Ln(LTC) CL Ln(MC) Mark.

Yield Div. ROE Lev Vol Age GIC

Ln(Total Compensation) 1.0000 Ln(Short term Compensation) 0.9905 1.0000 Ln(Long term compensation) 0.6225 0.5852 1.000 Dummy cross-listing 0.4268 0.4348 0.2142 1.0000 Ln(Market Capitalization) 0.7842 0.7983 0.4656 0.3702 1.0000 Market yield -0.2520 -0.2403 -0.1497 -0.1034 -0.3143 1.0000 Dividend 0.3165 0.3423 0.1551 0.0341 0.3459 -0.4142 1.0000 ROE 0.2615 0.2777 0.1374 0.1929 0.3822 -0.2327 0.3471 1.0000 Leverage 0.1233 0.1339 0.1490 0.0972 0.1444 0.0601 0.0843 0.1212 1.0000 Volatility -0.2599 -0.3066 -0.1256 0.2019 -0.3883 0.1308 -0.3464 -0.3420 -0.0634 1.0000 CEO age -0.0641 -0.0701 -0.0225 -0.0987 -0.0213 -0.1031 0.0511 -0.0602 0.0415 -0.1179 1.0000 General industry classification -0.0327 0.0134 0.0143 -0.2556 0.1070 0.0085 0.1056 0.0387 0.1632 -0.4586 0.0726 1.0000

(35)

28

Table 4: Composition of general industry classification General

Industry Classification

Non-cross-listed Cross-listed Total

1 Frequency 378 639 1,017 Row percentage 37.17% 62.38% 100% Total percentage 25.37%' 42.89%' 68.26% 2 Frequency 9 27 36 Row percentage 25.00% 75.00% 100% Total percentage 0.61%' 1.81%' 2.42% 3 Frequency 4 13 17 Row percentage 23.53% 76.47% 100% Total percentage 0.27%' 0.87%' 1.14% 4 Frequency 0 9 9 Row percentage 0% 100% 100% Total percentage 0% 0.60% 0.60% 5 Frequency 1 19 20 Row percentage 5.00% 95.00% 100% Total percentage 0.07%' 1.27%' 1.34% 6 Frequency 268 123 391 Row percentage 68.54% 31.46% 100% Total percentage 17.97%' 8.25%' 26.24% Total 660 830 1,490

This item represents the company's general industry classification.

It is defined as follows:

1 Industrial

2 Utility

3 Transportation

4 Bank/Savings & Loan

5 Insurance

(36)

29

4.2 Multivariate regression

4.2.1 Total compensation

Table 5 describes the regression estimates for the empirical model described by equation 1, it test the hypothesis that cross-listing has a positive effect on CEO compensation. The dummy variable of cross-listing is not significant, which reflects that the total compensation for cross-listed firms is equal to the total compensation of non-cross-listed firms. Therefore, the first hypothesis is rejected. When I take the control variables in consideration, one can see that market capitalization (size) is significantly positive related to CEO compensation (p<0,01). So with increase in size, the compensation of the CEO also increases. This is consistent with the previous literature of Tosi, Werner, Katz, & Gomez-Mejia (2000) which is discussed in paragraph 2.2.3.1. Other control factors, market yield and dividend have negative significant influence on CEO compensation (p<0.10 and p<0.05 respectively). The negative influence of the market performance is inconsistent with the existing literature discussed in paragraph 2.2.3.2 (Lilling, 2006; Jensen & Murphy, 1990). Another remarkable observations comes from the coefficient of the ROE, which proves to be insignificant, this is also inconsistent with the literature. So the CEO’s of the cross-listed firms are not paid for performance, instead the regression shows an increase in their compensation when the firm becomes larger. This could result in riskier behavior by the CEO’s trying to make the firm bigger. The other control factors leverage, volatility and CEO age proof to be insignificant.

With regards to the strength of the model, one can see that the model is significant (p<0.01) and the adjusted R-squared which is the percentage of variance explained by the model is 63.4%. The conclusion is for this hypothesis is that there is no significant difference in the total compensation of CEO’s between cross-listed and non-cross-listed firms.

(37)

30

Table 5: Influence of cross-listing on Total CEO Compensation

Ln(Total Comp) Coefficient Standard Error T P > | t |

Dummy cross-listing -0.067 0.094 -0.71 0.480 Ln(Market-capitalization) 0.406 0.023 17.39 0.000*** Market Yield -0.052 0.030 -1.73 0.084* Dividend -0.739 0.307 -2.41 0.017** ROE -0.127 0.225 -0.56 0.573 Leverage 0.199 0.246 0.81 0.420 Volatility 0.006 0.005 1.25 0.211 CEO Age -0.006 0.006 -1.01 0.312 General Ind. 2 -0.455 0.279 -1.63 0.104 3 -0.285 0.244 -1.17 0.233 4 -0.425 0.411 -1.03 0.303 5 -0.261 0.287 -0.91 0.365 6 -0.329 0.111 -2.97 0.003** _cons 5.676 0.566 10.03 0.000 Number of obs. 359 F 48.76 Prob. > F 0.0000 R-squared 0.6467 Adjusted R-squared 0.6343

(38)

31

4.2.2 Short-term or long-term compensation

Table 6, panel A and B, describes the regression estimates for the empirical model described by equation 2, it test the hypothesis that cross-listed firms focus more on long-term compensation then short-term compensation. Panel A shows the effect of cross-listing on the short-term (cash) compensation and panel B shows the effect of cross-listing on the long-term (equity) compensation. From panel A one can see that cross-listing has negative effect on short-term compensation, but this effect is not significant. The control variables show the same results as the regression of the total compensation both market capitalization (size) and dividend are significant. The coefficient of market capitalization has a positive effect on CEO compensation while dividend has a negative effect on CEO compensation. This is consistent with the previous literature of Tosi, Werner, Katz, & Gomez-Mejia (2000). The model is strong (p<0.01) and has an adjusted R-squared of 63.4%.

Panel B shows that cross-listing has no significant effect on the long-term compensation, so this show that cross-listed and non-cross-listed pay equal amounts of long-term incentives to their CEO’s. The control variables show that market capitalization as well as leverage has a positive significant effect on the long-term compensation. Leverage has not been significant in the other regressions. This result is consistent with the paper of Papa and Speciale (2011), who state that shareholders have to increase the variable part of CEO pay to provide the CEO with enough incentives. The effect of market capitalization is consistent with the previous literature of Tosi, Werner, Katz, & Gomez-Mejia (2000). Looking at the performance of the model, one can see that the model is significant (p<0.01), but the adjusted r-squared is quite low. Only 21.6% of the model is explained.

The overall conclusion of my second hypothesis is that cross-listed companies do not pay significantly more or less, short term or long term compensation. Therefore the second hypothesis will be rejected.

(39)

32

Table 6: Influence of cross-listing on short-term and long-term compensation Panel A: Regression coefficients between cross-listing and short-term compensation

Ln(Short Term)

Coefficient Standard Error t P > | t |

Dummy cross-listing -0.012 0.086 -0.14 0.888 Ln(Market-capitalization) 0.367 0.021 17.28 0.000*** Market Yield -0.038 0.027 -1.40 0.162 Dividend -0.604 0.278 -2.17 0.030** ROE -0.087 0.203 -0.43 0.667 Leverage 0.334 0.225 1.49 0.138 Volatility 0.002 0.004 0.38 0.703 CEO Age -0.007 0.005 -1.36 0.175 General Ind. 2 -0.498 0.252 -1.98 0.049** 3 -0.326 0.220 -1.48 -0.759 4 -0.228 0.371 -0.61 0.540 5 -0.162 0.260 -0.62 0.533 6 -0.210 0.102 -2.05 0.041** _cons 6.523 0.520 12.54 0.000 Number of obs. 354 F 51.56 Prob. > F 0.0000 R-squared 0.6635 Adjusted R-squared 0.6506

(40)

33

Panel B: Regression coefficients between cross-listing and long-term compensation

Ln(Long Term) Coefficient Standard Error T P > | t |

Dummy cross-listing -0.210 0.206 -1.02 0.310 Ln(Market-capitalization) 0.396 0.051 7.74 0.000*** Market Yield -0.022 0.065 -0.33 0.738 Dividend -1.039 0.693 -1.50 0.135 ROE -0.379 0.480 -0.79 0.430 Leverage 0.970 0.541 1.79 -.074* Volatility 0.006 0.103 0.62 0.535 CEO Age -0.006 0.012 -0.44 0.658 General Ind. 2 -0.966 0.593 -1.63 0.104 3 0.040 0.548 0.07 0.941 4 -1.77 0.869 -2.04 0.042** 5 -0.673 0.606 -1.11 0.268 6 -0.153 0.248 -0.62 0.534 _cons 3.452 1.252 2.76 0.006 Number of obs. 335 F 8.11 Prob. > F 0.0000 R-squared 0.2472 Adjusted R-squared 0.2167

(41)

34 4.2.2.2 Further investigation

To further investigate the influence of cross-listing on short-term/cash compensation, I divided the cash compensation into its original building stones; salary and bonus. These two variables were then taken as the dependent variable and two separate OLS regressions were performed. The descriptive statistics of these variables are shown in Table 7 and the outcomes of those regressions in table 8 panel A and B respectively. Table 7 shows that the means of the CEO salary and CEO Bonus are around $400.000, but the spread of the CEO Bonus is significantly wider.

Table 8 Panel A describes the regression outcomes for the influence of cross-listing on CEO salary. One can see that CEO’s of cross-listed firms earn less then CEO’s of non-cross-listed firms. The coefficient is $-35.086, which means that a CEO of a cross-non-cross-listed firm on average earns $35.086 less, then a CEO of a non-cross-listed firm. This result is significant (p<0.10). The control variables show that market capitalization and CEO age strongly positively related to CEO salary (p<0.01). The fact that CEO age is strongly positively related to salary is consistent with the existing literature (Adhikari, Bulmash, Krolikowski, & Sah, 2015). Their paper title says it all, Old is Gold. The older CEO’s earn more salary than younger colleagues. Both market performance variables (market yield and dividend) and leverage also positively related to CEO salary (p<0.05). ROE on the other hand has a significant negative effect on CEO salary (p<0.05). The performance of the model indicates that it is significant (p<0.01) and the adjusted r-squared is 70.5%.

Panel B reports the regression outcomes for the influence of cross listing on the yearly cash bonus of the CEO. The variable for cross listing is not significantly different from zero, this means that the yearly cash bonus of CEO’s of cross-listed firms is equal to the yearly cash bonus of CEO’s of non-cross-listed firms. Looking at the control variables, one can see that market capitalization and dividend are again highly positively related to the CEO bonus (p<0.01). But leverage has a significant negative effect on the CEO bonus (p<0.10). I discussed this in paragraph 2.2.3.3, the more debt causes the debt holders to monitor the CEO more intensely and that lowers the compensation. Looking at the performance of the model, the model is significant (p<0.01) and the adjusted R-squared is 43.8%.

(42)

35

Because the dataset is limited regarding the long-term compensation unfortunately no further research on this subject can be done.

Table 7 Further research: descriptive statistics of CEO salary and CEO bonus

Variable Observations Mean Std. Dev. Min. Max

CEO Salary 426 398,084.9 269,786.1 200 2,038,885

CEO Bonus 368 400,273.4 593,065.8 0 4,422,000

Table 8: Influence of cross-listing on CEO salary and CEO bonus Panel B: Regression coefficients between cross-listing and CEO salary

Salary Coefficient Standard Error T P > | t |

Dummy cross-listing -35,085.97 21,177.34 -1.66 0.099* Ln(Market-capitalization) 97,272.39 5,296.14 18.37 0.000*** Market Yield 14,417.98 6,855.54 2.10 0.036** Dividend 189,312.4 72,729.32 2.60 0.010** ROE -108,174.3 49,839.77 -2.17 0.031** Leverage 133,795 55,668.97 2.40 0.0017** Volatility 656.58 1,030.95 0.64 0.525 CEO Age 3903.00 1,290.30 3.02 0.003** General Ind. 2 -101,395.8 62,200.54 -1.63 0.104 3 -75,436.15 53,886.61 -1.40 0.162 4 153,271.1 91,032.72 1.68 0.093* 5 -50,562.37 63,587.55 -0.80 0.427 6 -110,748.2 25,360.30 -4.37 0.000*** _cons -1,667,183 128,280.6 -13.00 0.000 Number of obs. 347 F 64.70 Prob. > F 0.0000 R-squared 0.7164 Adjusted R-squared 0.7053

(43)

36

Panel B: Regression coefficients between cross-listing and CEO bonus

Salary Coefficient Standard Error T P > | t |

Dummy cross-listing -36,332.11 71,115.08 -0,51 0.610 Ln(Market-capitalization) 150.685,60 18,267.2 8.25 0.000*** Market Yield -3,232.58 21,697. -0.15 0.882 Dividend 811,685.5 229,334.7 3.54 0.000*** ROE -46,080.56 163,080.3 -0.28 0.778 Leverage -313,415.5 188,945.6 -1.66 0.098* Volatility 3,511.62 3,604.93 0.97 0.331 CEO Age -28.91 4,387.20 -0.01 0.995 General Ind. 2 -247,028.8 207,817.8 -1.19 0.236 3 -373,832.2 168,950.8 -2.21 0.028** 4 1,297,041 286,868.1 4.52 0.000*** 5 -124,364.5 199,897.9 -0.62 0.534 6 55,246.74 87,556.61 0.63 0.529 _cons -2,655,508 435,669.9 -6.10 0.000 Number of obs. 308 F 19.42 Prob. > F 0.0000 R-squared 0.4620 Adjusted R-squared 0.4382

(44)

37

5 Conclusion

This study emphasizes the effect of cross-listing on CEO compensation. I argue that CEO’s of cross-listed firms are receiving higher compensation compared to CEO’s of non-cross-listed firms. The reason behind my hypothesis is that CEO’s of cross-listed firms are harder to find, this is based on the human capital theory and the managerial theory as well as the positive relationship between cross-listing and CEO compensation. The univariate analysis proofs that my estimations were right. The firms that are cross-listed are larger and the CEO compensation is higher. My multivariate regression, which controls for factors specific to the firm, the CEO and the industry, showed no significant results for the total compensation, short-term compensation and long-term compensation. The control variables for size (market capitalization) and market performance (market yield and dividend) proved to have a significant effect on the CEO compensation.

When I preformed additional research on the short-term compensation I did find a significant negative effect of cross-listing on CEO’s salary. This result indicates that firms who are cross-listed pay their CEO’s $35.000 less than the firms who are not cross-listed. The reason of this could be that the CEO’s of cross-listed firms are more bonus paid, but evidence to support this claim is not found in this study.

This study has its limitations, one of which comes from the sample size, which is limited. One reason for this is that the amount of firms in the UK that do cross-list their securities is per definition limited and this population was further reduced when I had to merge the two databases of Compustat and World Scope. Another limitations arises from the limited availability on the numbers of the CEO compensation, especially the long-term (equity) compensation. This limitation fazed me of conducting further research on this form of compensation.

Further research can investigate the effect of cross-listing on the different elements of CEO compensation. Another possibility is to focus on the supply-and-demand of the CEO’s and their respective compensation. More thorough knowledge on the CEO compensation can help board or compensation committees to develop a more suited compensation scheme.

(45)

38

References!

Abowd, J. M. (1990). Does performance-based managerial compensation affect corporate performance? Industrial & Labor Relations Review , 43 (3), 52S-73S.

Adhikari, H., Bulmash, S., Krolikowski, M., & Sah, N. (2015). Dynamics of CEO compensation: Old is gold. The Quarterly Review of Economics and Finance , article in press.

Alexander, G. J., Eun, C. S., & Janakiramanan, S. (1988). International Listings and Stock Returns: Some Empirical Evidence. Journal of Financial and Quantitative Analysis , 23 (2), 135-151.

Baker, H. K., Nofsinger, J. R., & Weaver, D. (2002). International cross-listing and visibility. Journal of Financial and Quantitative Analysis , 37 (3), 495-521.

Chhabra, A. B. (2008). Executive stock options: Moral hazard or just compensation. The Journal of Wealth Management , 11 (1), 20-35.

Coffee, J. (2002). Racing towards the top?: The impact of cross-listings and stock market competition on international corporate governance. Columbia Law Review , 102 (7), 1757-1831. Coffee, J. (1999). the future as history: the prospects for global convergence in corporate

governance and its implications. Northwestern University Law , 93, 641-708.

Conyon, M. J., & Schwalbach, J. (2000). Executive compensation: evidence from the UK and Germany. Long Range Planning , 33 (4), 504-526.

Core, J., Holthausen, R., & Larcker, D. (1999). Corporate governance, chief executive officer compensation and firm performance. Journal of Financial Economics , 51 (3), 371-406.

Doidge, C., Karolyi, A., & Stulz, R. (2004). Why are foreign firms listed in the US worth more? Journal of Financial Economics , 71 (2), 205-238.

Eisdorfer, A., Giaccotto, C., & White, R. (2013). Capital structure, executive compensation, and investment efficiency. Journal of Banking & Finance , 37, 549-562.

Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of management review , 14 (1), 57-74.

Errunza, V. R. (2000). Market Segmentation and the Cost of Capital in International Equity Markets. The Journal of Financial and Quantitative Analysis , 35 (4), 577-600.

Errunza, V., & Losq, E. (1985). International Asset Pricing under Mild Segmentation: Theory and Test . The Journal of Finance , 40 (1), 105-124 .

Fama, E., & Jensen, M. (1983). Seperation of ownership and control. The Journal of Law and Economics , 26 (2), 301-326.

The influence of international cross-listings on the elements of CEO compensation