Capital structure in Central and Eastern European Countries: what is the role of EU membership?

(1)

Capital structure in Central and Eastern

European Countries:

what is the role of EU membership?

University of Groningen

Faculty of Economics and Business

Msc. International Business & Management

Supervisors: dr. I. Kalinic & drs. A. Visscher

Gerline de Pijper 1710516 g.j.de.pijper@student.rug.nl

(2)

1

Abstract

This study looked at the importance of EU membership on the capital structure of new member states. Due to EU membership macroeconomic circumstances change, so capital structure is likely to change as well. Agency theory, pecking-order theory and trade-off theory are being used to test which variables can explain the possible change in capital structure. The results are being controlled for size, age, industry and listed/non-listed companies. It emerges that capital structure does not change significantly due to EU membership; change in currency seems to have a bigger influence. Besides, among the considered literature, only agency theory partly explains the capital structure of the Eastern European companies. Finally tangibility has the largest influence on capital structure in this sample.

Key words: CEE countries, capital structure, transition economies, agency theory, pecking-order

(3)

2

List of acronyms

CEE Central and Eastern Europe

EU European Union

FDI Foreign Direct Investment

GDP Gross Domestic Product

NDTS Non-debt tax shield

OALD Oxford Advanced Learners Dictionary WW I World War I

WW II Word War II

List of tables

Table 1 – Paired samples t-test Czech Republic ... 44

Table 2 – Paired samples t-test Hungary ... 45

Table 3 – Paired samples t-test Latvia ... 45

Table 4 – Paired samples t-test Lithuania ... 45

Table 5 – Paired samples t-test Poland ... 46

Table 6 – Paired samples t-test Estonia ... 46

Table 7 – Paired samples t-test Slovakia... 47

Table 8 – Paired samples t-test Slovenia ... 47

Table 9 – Paired samples t-test Malta ... 47

Table 10 – Outcomes hypotheses testing ... 55

List of graphs

Graph 1 – Conceptual model ... 24

(7)

6

1. Introduction

In 1957, six countries started the European Economic Community. After that, in 1967 it became the European Communities, and in 1993 it was renamed to the European Union. From 1957 until 2004 there were six rounds in which more countries became member. Now, the EU has reached a number of twenty-seven members, and there are some candidate states that have to meet the requirements of the EU before they can obtain EU membership.

In 2004, in the sixth round, ten new member states were added to the European Union, which are: Cyprus, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia, and Slovenia. These countries became a member of the European Union, thereby opening up their economy for other European member states and applying to the rules and regulations of the EU. Furthermore, some of these countries even adopted the Euro in the years after their appliance to the EU.

What is especially interesting about the above mentioned countries is that most of them were considered ‘transition economies’ (Gabrisch & Werner, 1998), with relatively small private firms, a poorly developed ability to broach new markets, and legacies influenced by state socialism. These countries had to transition from a state regulated by socialism to a state driven by liberalism.

Nowadays, especially Eastern European countries are striving to reach the goal of becoming a member state of the EU. Among others, they expect to boost their economy due to an increased market and to reap other advantages from this membership as well, which are both economical and political (Kraus & Schwager, 2003). These advantages exist on both macroeconomic and microeconomic level. However, most of the studies focusing on the consequences of EU membership only focus on the macroeconomic consequences (Gabrisch, 1997; Kraus & Schwager, 2003; Turrión & Velázquez, 2004; Murphy, 2006; Butler, 2007). By becoming an EU member state, these Eastern European countries now have to follow the EU legacies, which encourages liberalism. It is expected that not only the macro-economy of these transition economies changed because of EU

membership, but also the micro-economy

(i.e. level characteristics). This study focuses on only one aspect of the consequences on firm-level, namely on the business structure of the companies originated in the new member states, with a specific focus on the capital structure.

(8)

7 and Modigliani (1958) were the founders of the capital structure with their trade-off theory. Lots of other researchers followed their example, to investigate the capital structure of companies, and they have developed other theories that can explain the capital structure in companies, like the pecking-order theory and the agency theory. However, most of these studies focused only on the developed countries worldwide. Only a few researchers have looked at whether the variables influencing capital structure are the same in developed and developing countries (Berk, 2006; Kayo & Kimura, 2011). Besides the fact that there is a research gap concerning the research of capital structure in these former transition economies, there is another interesting thing to look at. The variables influencing capital structure focus both on firm- and on country-level. These two types of characteristics are both likely to be changed by EU membership of the new members (Barrell et al, 2010), and therefore it is likely that the capital structure changes as well. However, it has not been tested yet whether becoming an EU member really changes the capital structure of companies originated in these new EU member states. Therefore, this study has two objectives. The main objective consists of examining whether capital structure changes due to EU membership. Secondly, I identify which variables can explain this possible changes in capital structure.

1.2 Problem statement

The research objective stated above cannot be reached in once. In order to be able to come to a conclusion, three steps are taken, which together lead to the conclusion. First of all the capital structure of companies located in the selected countries prior to EU membership is examined. Second the capital structure of these companies after EU membership is examined and a comparison is made. Finally variables which could explain the possible change in capital structure of these companies are examined.

Line of reasoning

(9)

8

1.3 Research relevance

The relevance of this research has two sides: first the problem is interesting because it fills a research gaps. Secondly, this research is relevant for management theory and managers themselves as well, for the following reasons. First of all it contributes to the existing knowledge of capital structure theories (i.e. trade-off theory, agency theory and pecking-order theory). These theories have been tested mainly in developed countries, so it is interesting to see whether they apply to former-transition countries as well (Nunkoo & Boateng, 2011; Delcoure, 2007). It tests whether companies in these countries follow one of these theories or not. Besides that, this research can give managers understanding of how EU membership can also bring about changes at firm-level, since most of the research on EU benefits and consequences only focuses on macro-economic level. However, it is logical to expect that EU membership also encourages changes on firm-level. Managers can use the outcomes of this research to see how the capital structure is being changed with influence of EU membership, and also whether and how the variables influencing capital structure can change following the EU membership. Furthermore, this study gives managers greater insights in the variables that influence capital structure in developing countries, and through which channels they can change this structure if they want or have to. Besides that, managers from companies originated in the candidate states can gain an understanding in what changes in their capital structure they can expect from becoming an EU member state. So this study both contributes to the capital structure theories, and it provides managers with more insights on how their capital structure is composed and influenced by EU membership and changing circumstances.

(10)

9

2. Literature review

Before being able to explain exactly how the firm-level variables and capital structure are related it is important to know that there are different theories that attempt to explain these relationships. These theories can be divided into three different theories: the first is the agency theory, the second is the pecking-order theory and the third is the trade-off theory.

This section explains what is exactly meant by capital structure. Secondly the theories explaining capital structure are elaborated on.

2.1 Capital structure

Capital structure is a topic which is very important for every firm. Since there are two main sources of financing for a firm, namely debt and equity, the firm needs to make the choice on using debt or equity. This is called the capital structure decision, and it is one of the most important financial policy decisions and also one of the most researched topics in finance. The owners of a business have to contribute at least some equity, so the capital structure decision narrows down to what amount of debt to use (Frank & Goyal, 2009).

(11)

10 The first to study the capital structure of a company were Modigliani and Miller (1958). They say that in the absence of some assumptions (a world without taxes and transaction costs), capital structure does not affect firm value. So the decision of whether to use debt or equity does not have consequences for the value of a firm. In 1963 they conducted another study in which they included corporate taxes in their study, and concluded that firms should be totally debt financed. After their initial study, many studies followed in discussing the capital structure decision (Jensen & Meckling, 1976; Fama & Miller, 1972; Jensen, 1989; Harris & Raviv, 1990; Stulz, 1990). One of these follow-up studies was being done by Miller himself (1977), in which he included personal taxes to the capital structure decision and concluded again that companies should be completely debt-financed. In the end, lots of capital structure theories have been developed throughout these studies, which all tried to explain the financing behavior of corporations (Jensen & Meckling, 1976; Myers & Maljuf, 1984; Harris & Raviv, 1991; Frank & Goyal, 2003). All these studies highlighted the importance of two theories in making the capital structure decision: the trade-off theory and the pecking-order theory. Furthermore, the agency theory is strongly related to the pecking-order theory. Together, these three theories form the basis for the discussion on the capital structure decision and therefore these theories are elaborated on later in this section.

So, one can conclude from this section that the capital structure theory is a further refinement of corporate finance theory. It aims at finding an optimal capital structure which can be used to maximize corporate value, and for this aim, three theories have been developed: the pecking-order theory, the trade-off theory, and the agency theory (Jensen & Meckling, 1976; Myers & Maljuf, 1984; Harris & Raviv, 1991). These three theories all treat the financing decision a company has to make on the amount of debt and equity it will use to finance the corporation.

2.2 Agency theory

(12)

11 In order to understand the relationship between agency theory and capital structure, it is important to firstly explain what the agency theory holds and why agency problems occur. Agency theory is a theory that finds it origins in the 1960s, 1970s. At that time, the literature on risk-sharing among groups was booming and this literature was broadened by including agency problems. These problems occur when there is a separation of ownership and control in the company. When cooperating parties have different goals and there is division of labor agency problems can occur (Jensen & Meckling, 1976; Ross, 1973). Specifically, agency theory is directed at the ubiquitous agency relationship, in which one party (the principal) delegates work to another (the agent), who performs that work.

Agency problems

Agency theory is concerned with resolving two kinds of problems that can occur in this relationship: the first group is concerned with the problems between the management and the shareholders, and the second with the problems between the management and debt holders (Lundstrum, 2008). The first category can be divided into two groups: the first is that a problem arises when the desires and the goals of the principal and the agency conflict, or when it is difficult or expensive for the principal to verify what the agent is doing. Since the actions of the management are unobservable for the stockholders, there is uncertainty associated with these actions and with their efforts as well. This problem concerns the agency costs of aligning the desires and goals of the principal and the agent, and the costs of controlling the agents. The second problem can arise when the principal and the agent have different attitudes towards risk, so they may prefer different actions because of the different risk preferences (Eisenhardt, 1989), which can be costly as well. Shareholders want the management to act for their own good, but sometimes management wants to take other decisions that do not benefit the shareholders. This has to do with managers behaving opportunistically and/or in their own self-interest and with the managers being constraint by bounded rationality (Eisenhardt, 1989).

(13)

12 have incentives to reject positive NPV projects, and this rejection leads to underinvestment. However, a problem that is related to this second problem is that managers may want to increase firm size resulting in greater compensation. Consequently they may adopt non-profitable investments, even though the outcome is likely to be non-beneficial for the shareholders. This agency cost is likely to be exacerbated in the presence of free cash flow in the firm (Jensen, 1986). These two problems are acknowledged by rational debt holders, so they price their debt accordingly and thus ask a higher rate of return. The adverse consequences of the debt agency problems are entirely borne by the equity holders themselves, since the costs of debt financing increase.

The implications of the debt agency problems are widely studied in the literature (Jensen & Meckling, 1976; Myers, 1977; Williams, 1987; Harris & Raviv, 1990; Stulz, 1990; Mao, 2003; Berger & Bonaccorsi di Patti, 2006). These papers propose that a firm’s optimal capital structure is achieved by equating the marginal agency costs of debt and the marginal agency costs of equity. Furthermore, this theory assumes that the agency costs of debt increases monotonically with the amount of debt the firm employs.

Opportunistic behavior and bounded rationality

Opportunistic behavior and bounded rationality are two concepts that are at the core of the agency theory. Opportunistic behavior assumes that people will act in such a way to benefit themselves the most, or as is stated in the OALD: behaving opportunistic is ‘making use of an opportunity, especially to get an advantage for you’. So when behaving opportunistic, you put yourself and your own desires and goals at the first place, not the goals of the company. Bounded rationality on the other hand says that people have a limited ability to adapt optimally, or even satisfactorily, to complex environments (Simon, 1991). Bounded rationality can furthermore be described as: ‘the notion that in decision making, rationality of individuals is limited by the information they have, the cognitive limits of their minds, and the finite amount of time they have to make decisions (Simon, 1991; Gigerenzer & Selten, 2002). So there is a limit to human rationality and therefore people sometimes will not act in the way that is preferred by others.

Market imperfections

(14)

13 monitoring the agents are. So, although it is preferable that no market imperfections exist, this cannot be achieved in real life. Therefore, to keep agency costs as low as possible, actions should be taken to overcome the market imperfections. Many studies have tried to think of ways to diminish the agency costs (Florackin & Ozkan, 2009; Brockman et al, 2010; Davies & Prince, 2010), but that is outside the scope of this study, and also not important in assessing the influence of agency theory on capital structure.

Asymmetric information

Another concept that plays a role in the agency theory is asymmetric information. One of the reasons why it is hard for the principal to monitor the agent is because of the different amount of information available for them. So, the larger the asymmetric information, the higher the costs of monitoring and also the higher the agency costs (Myers & Maljuf, 1984). If all parties would have an equal amount of information available it would be much easier to follow the thoughts of the management and to see rationally why they made certain decisions. However, since this is rarely the case, information asymmetry is one of the drivers of agency problems.

Agency costs and bankruptcy costs

This theory implies that an optimal capital structure exists, and so it should take into consideration both agency costs and bankruptcy costs (Mao, 2003). The agency costs are concerned with the costs of monitoring the managers and of aligning the desires and goals of the principals and the agents. Besides that, the agency costs are also concerned with preventing the managers from behaving opportunistically by for example implementing rules and restrictions for it. The bankruptcy costs, on the other hand, take into account the indirect costs as well (Altman, 1984). These indirect costs have to do with the disruptions that may occur in relationships between the firm and its customers and suppliers (Haugen & Senbet, 1988), so these are the costs related to long-run shifts in supply and demand facing the firm. The direct costs on the other hand are the costs associated with liquidating the firm (Haugen & Senbet, 1988), and also with transferring the ownership from stockholders to bondholders and with dismantling the assets of the firm upon liquidation (Altman, 1984). What is important to note here is that the bankruptcy costs are not likely to be higher if bondholders are in control than if stockholders are in control, because bondholders are no less rational than stockholders.

(15)

14 managers and reduces the manager’s bargaining power (Noe & Rebello, 1996) and the discretion over spending (Stulz, 1990).

2.2.1 Relevance agency theory

As explained above, the most important features of the agency theory relate to agency costs of debt, to agency problems that may arise and to the costs of issuing debt. When taking this into consideration, one can argue that these concepts can be related to growth, profitability and firm size. The features of agency theory can be used to argue why a certain relationship should exist between capital structure and for example growth and profitability. Therefore the decision has been made to use the agency theory to explain these relationships. Yet another reason for choosing the agency theory is that it is a theory that has been widely used to explain capital structure, and even though the explanation of capital structure with use of agency theory has been done mainly in developed economies, this does not say that it cannot be used to examine capital structure in developing theories. Therefore the agency theory is being used in this research.

2.3 Pecking-order theory

The second theory that is explained in this literature review is the pecking-order theory. It has been argued that the pecking-order theory is one of the most influential theories of capital structure (Ni & Yu, 2008).

Hierarchy of capital

(16)

15

Asymmetric information

The pecking-order theory is based on the assumption that there is asymmetric information within the company. Managers have more inside information than the investors and they act in favor of old shareholders (Myers & Maljuf, 1984; Ni & Yu, 2008). This theory further states that investors do not know the true value of newly issued stock for a new investment opportunity, nor can they accurately judge the value of newly issued stock for a new investment. Thus, when seeking external capital, financing through debt will always be preferable for firms whose stock price is undervalued. Since, when the stock price is undervalued, less capital can be generated by issuing new stock and therefore it is better to issue debt to generate capital (Sakai, 2010). On the other hand, when a firm’s stock is overvalued, this firm will prefer to raise funds through equity financing. However, because investors are aware of the adverse selection problem, issuing stock will not have an equilibrium price, causing the firm with overvalued stock to seek capital through debt in the end. The reason why the issuing stock will not have an equilibrium price is the following: if managers know more than the rest of the market about their firm’s value, and especially about the health and the prospects of their company, the market penalizes the issuance of securities whose expected payoffs are related to the assessment of such a value (Myers, 1984).

Because it could be difficult to generate capital due to asymmetric information, company managers may decide not to launch potentially profitable projects if they have to be financed with risky financial instruments (Myers & Maljuf, 1984). So, as a result, the pecking-order theory predicts a hierarchical order in a company’s financing policy. This policy is led by the financial sources which are at least subject to information costs and at the same time involve less risk. Furthermore, this policy will allow the firms the capacity to finance investments and avoid external financing (Sánchez-Vidal & Martín-Ugedo, 2005).

What was furthermore argued by Myers and Maljuf (1984) when they developed their theory is that if the company does not have enough funds to finance new investments, equity will only be issued if these new investments are either very profitable or they cannot be postponed nor financed through debt. It can also be that managers want to issue equity because they believe that the stock is overvalued enough that shareholders will be disposed to tolerate the market penalty.

Transparency

(17)

16 listing on a stock exchange (Tong & Green, 2005). There are more ways in which firms can become more transparent but it is outside the scope of this study to mention those. Thus transparency has an influence on asymmetric information costs and thereby on the financing decisions of companies.

Adverse selection

Since issuing equity is an information sensitive manner of generating capital, it involves a lot of costs to provide all the information the investors ask for. Since they will not be able to get all this information, because some of it is confidential, they will price the equity lower than how it might have been accurately priced. Therefore there is an adverse selection problem, and also adverse selection costs. Those costs are what ultimately affect the costs of issuing information-sensitive securities (Barath et al, 2008).

So, for the pecking-order theory, the most important cost which determines the choice of capital and thereby also the capital structure, next to the direct costs, is the cost of asymmetric information, and the cost of adverse selection (Lopéz-Garcia & Sogorb-Mira, 2008). As an explanation, these direct costs include the following: there are banking fees that have to be paid when issuing new equity; the firm may be able to reduce the taxable current dividends by limiting security issues; and the transaction costs, which are lower for issuing debt than for issuing equity (Baskin, 1989).

Firm control

There is something else except for the information costs involved with acquiring capital, namely the motivation to retain control of the firm (Hamilton & Fox, 1998; Holmes & Kent, 1991). When a firm issues new shares, these are offered to the general public. Thus, current shareholders’ relative ownership of the firm, as well as their control over the company, may be diminished with the new share offerings. Their ownership percentage goes down, and thereby also their control and influence and therefore the current shareholders, next to the company directors, are reluctant to issue equity (Sánchez-Vidal & Martín-Ugedo, 2005).

Development capital market

(18)

17 developed yet, it is more difficult for companies to get access to capital, since there is less supply and therefore it is harder to find cheap external capital. If it is not possible to get access to external debt, the only other way of getting capital is to issue shares. So according to the pecking-order theory, when a company has a high capital structure, the capital market is not very developed (Ni & Yu, 2008.).

2.3.1 Relevance pecking-order theory

Since the pecking-order theory cannot be used only to see whether firms prefer internal over external financing or not, but also to draw conclusions on the development of the capital markets it is of big relevance for this study. The countries that are being studied in this research were transition economies in the ‘90s and so it is interesting to be able to draw some conclusions on the development of the capital markets in these economies by using the pecking-order theory.

Furthermore, the pecking-order and the trade-off theory are not mutually exclusive (Sakai, 2010). They can exist together at the same time, and therefore few researches have strictly compared the relevance of the two theories (Ghosh & Cai, 2007). Research more often focuses on proving the validity of one theory in explaining some phenomena, while other phenomena are validated by the other theory (Sakai, 2010). That is the reason why three theories are being used in this study, one of the theories cannot explain exactly why a company has a certain capital structure. Variables relating to the capital structure can only be explained by combining these theories.

Finally, the pecking-order theory is important in explaining the capital structure of a company, because it relates profitability, free cash flow, dividend payouts and growth to the capital structure of a company, as was explained above in taking into consideration asymmetric information, adverse selection and the hierarchy of capital (Tong & Green, 2005; López-Garcia & Sogorb-Mira, 2008; Sánchez-Vidal & Martín-Ugedo, 2005).

2.4 Trade-off theory

Static trade-off theory

(19)

18 give the total bankruptcy costs of the firm. It is argued that the higher the probability of bankruptcy, the lower the level of debt should be (Harris & Raviv, 1991; Frank & Goyal 2003).

The trade-off theory is a theory that builds upon the work of Miller and Modigliani (1958) who were the first to come up with the idea of an optimal capital structure. As explained above, they assumed that taxes and bankruptcy costs do not play a role in determining the capital structure. The trade-off theory on the other hand relaxed these ‘no tax’ and ‘no bankruptcy costs’ assumptions and said that taxes and bankruptcy costs do play a big role in determining the capital structure. It is argued that debt not only provides a tax shield, but also increases the risk of bankruptcy (Howe & Jain, 2010). So, both the tax shield and the risk of bankruptcy increase with a larger capital structure.

The trade-off theory says that the deadweight bankruptcy costs are traded off against the tax saving of debt (Frank & Goyal, 2009). The bankruptcy costs are deadweight because these costs are already paid and there is no way in which the company can get them back. These costs are paid despite of any investments made. The tax saving of debt says that when one borrows money one has to pay rent for this debt next to paying back the sum of money. The rent that is paid over the debt can be distracted from the amount of money over which you have to pay tax. So you lower the costs spent on taxes and therefore there is a cost saving when using debt.

Although this theory states that there is an optimum capital structure for the firm, this notion is not shared among all researchers. For example Graham (2000) argues that the tax savings (which are the benefits of debt) are much higher than the bankruptcy costs (which are the costs of debt) so there will not be a trade-off between the two. Besides Fama and French (2002) find that being more leveraged has more to do with the profitability of the company than with the marginal benefits and costs of debt.

Trade-offs

Although the benefits and costs of using debt are generally referred to as the tax savings of using debt and the costs of financial distress, there are multiple ways in which these costs and benefits can be obtained.

(20)

19 Furthermore, there is a third perspective on the trade-off of costs and benefits of using debt. This third perspective is called the ‘product and factor market based trade-off theory’. Theory suggests that capital structure could either enhance or impede productive interactions among the stockholders, and when firms make unique products they will lose customers if they are likely to fail (Titman, 1984).

So, capital structure can influence the choice of firms to offer a high quality product or not (Maksimovic & Titman, 1991). The two latter perspectives differ from the traditional trade-off perspective in that the costs of debt are from disruption to normal business operations and they do not depend on the relatively small direct costs of bankruptcy (Frank & Goyal, 2009). These two perspectives trade off the advantages of debt with the liquidation costs, rather than the bankruptcy costs.

Target ratio

When assuming that the trade-off theory holds, one acknowledges that companies have a target debt ratio to which they have to apply (Brounen et al, 2005). This target debt ratio is, among others, based on the industry capital structure. Furthermore, the target debt ratio is calculated by balancing the beneficial tax shields with the costs of financial distress. By balancing this, the appropriate amount of corporate debt can be calculated. Research revealed that indeed many firms have a target debt ratio that they strive to reach (Brounen et al, 2005; Howe & Jain, 2010). However, even though there is a lot of theory on this aspect of capital structure, empirical evidence lags behind (Brounen et al, 2005).

Dynamic trade-off theory

(21)

20 non-profitable investments. In the dynamic trade-off theory on the other hand, firms with higher profits have a lower capital structure because this shifts over time (Frank & Goyal, 2009).

Evidence for the dynamic trade-off theory is given in empirical research looking at the capital structure of European companies (Brounen et al, 2005). They show that over half of the European companies have a target ratio, but this ratio is not a strict one. They allow their capital structure to fluctuate somewhat around this target ratio, so they have a dynamic target ratio.

The main importance of this theory is that it explains why capital structures fluctuate, which is one of the hypothesized consequences of EU membership in this research.

2.4.1 Relevance trade-off theory

The trade-off theory is important in explaining the capital structure of the companies the sample, since it is predicted that the tax benefits and the bankruptcy costs of debt are among the main determinants of capital structure (Frank & Goyal, 2009). This research would be incomplete if the factors that are associated with the trade-off theory are not taken into consideration, especially because these two concepts are the cornerstones of the theories on capital structure.

The static trade-off theory has to be taken into consideration since it explains largely the basis of the dynamic trade-off theory. The variables in both theories are the same, but the assumptions differ.

Relevance dynamic trade-off theory

(22)

21

3. Hypotheses and model development

So now that explanations of the three main theories in the discussion on the capital structure have been given, hypotheses are being formed, based on these theories. In that way it can be seen whether previous literature is in consistence with this research. First of all the hypothesis relating to the main objective of this research is determined. Secondly the hypotheses relating to the agency theory, the pecking-order theory and the state-off theory are given.

3.1 Hypothesis relating to the change in capital structure

The first hypothesis that is tested in this research is based on the expectation that the capital structure of Eastern European companies changed due to the CEE countries becoming a member of the EU. This hypothesis is based on the dynamic trade-off theory (Fischer et al, 1989), which was described in the previous section. Since the EU is said to bring many advantages to the member states, it is expected that the circumstances for and the environment of companies in the new member states change as well. They are expected to have a larger market area, more money available, more customers etc. (Chen et al, 1997). So therefore it is likely that the capital structure in these companies changes as well, due to changing circumstances and a changing environment. Thus the first hypothesis is the following:

H1: Becoming an EU member state influences the capital structure of companies in these states.

3.2 Hypotheses relating to variables influencing capital structure

The second set of hypotheses attempts to explain the possible change in capital structure detected by testing the first hypothesis. By using the agency theory, the pecking-order theory and the trade-off theory the variables that have the largest influence on capital structure of companies in developed countries are determined. A set of seven variables with the largest influence has been chosen and the influence of these variables is restated into hypotheses that are tested for the companies in CEE countries. These hypotheses are tested for the two periods separately to see whether the influence of these variables on the capital structure changed over time.

Agency theory

(23)

22 As explained above, managers behave opportunistically when they have the chance, and they are restricted by bounded rationality. This is something that especially occurs when the company has a period of high growth, since during this period there are more free cash flows. And as explained above, when a firm has more free cash flows available, managers have more incentives to use this money for non-profitable investments, or for investments that do not utilize the assets in the most effective way. So, with the availability of more free cash flows, more debt should be used, since this mitigates the agency problems that can occur with the availability of more money. So, agency theory states that growth is positively related to the use of debt. However, no consensus has been reached on whether this theory rightly predicts the relationship between capital structure and growth and profitability (Kayo & Kimura, 2011; Nunkoo & Boateng, 2010). Chen et al (1997) on the other hand argue that based on agency theory and the acknowledgement that underinvestment problems exist, firms with higher growth opportunities should use less debt, so that there is a negative relationship between growth and the capital structure. The decision has been made to follow the first line of reasoning, since this one is more recent, and supported by more researchers.

H2a: Growth has a positive influence on the capital structure

Firm size is another important indicator of the capital structure, and this can be related to the agency theory as well, especially to the bankruptcy costs mentioned in the agency theory. Titman and Wessels (1988) argue that larger firms are more diversified, which makes them less prone to bankruptcy risk. Besides that, larger firms are generally more transparent as well, which makes it easier for them to issue debt and spread the issuing costs (Byoun, 2008). Besides, being more transparent also reduces the uncertainty and unobservability associated with the managers’ actions. Therefore, even though reduced uncertainty does not ask for more debt, larger firms are likely to have more debt, which proposes a positive relationship between firm size and capital structure. This can be restated into the following hypothesis:

H2b: Firm size has a positive influence on the capital structure

(24)

23

H2c: Tangibility has a positive influence on the capital structure

Pecking-order theory

The pecking-order theory states that firms prefer to finance their investments with internally generated funds over externally generated funds. One of the variables which is obviously related to the issue of the availability of internal capital is the profitability of a firm. When the firm is able to become more profitable, there are more retained earnings and therefore the firm has the possibility to raise more funds internally (Rajan & Zingales, 1995; Barath et al, 2008; Nunkoo & Boateng, 2010). In this way more funds is available for new investments and less externally generated funds has to be attracted. So when a firms’ profits become larger, the amount of debt they use decrease. This is restated in the hypothesis below:

H2d: Profitability has a negative influence on the capital structure

Trade-off theory

With the introduction of the costs of financial distress on the optimal capital structure by Jensen & Meckling (1976) the trade-off theory was complete, as was indicated above as well. They concluded that firms should use debt to the extent that the marginal benefits of additional debt and the costs of financial distress are equal, as this indicates that the firms are using an optimal debt ratio.

The trade-off theory says that the bankruptcy costs are formed by the probability of bankruptcy and the bankruptcy-related costs. If the probability of bankruptcy becomes higher, the bankruptcy costs become higher and then banks are less prone to borrow money to these firms (Harris & Raviv, 1991; Frank & Goyal, 2003) since their risk has gone up (Hackbarth et al, 2007). So if firms are more likely to go bankrupt, indicated by higher bankruptcy costs, they have less access to debt. This is restated in the hypothesis below:

H2e: Bankruptcy costs have a negative influence on the capital structure

Furthermore, the costs of financial distress are displayed as the healthiness of the company, which is indicated by the distance from bankruptcy (Bradley et al, 1984; Chen et al, 1997; Nivorozhkin, 2005; Sabiwalsky, 2010; Kayo and Kimura, 2011). The healthier the company, the less likely it is to go bankrupt. When the company is less likely to go bankrupt, the lower the costs of financial distress and so the more likely it is that the firm will use more debt. Otherwise stated:

(25)

24 Second of all, there is the other side of the trade-off theory, which relates to the benefits of additional debt, which is described as the non-debt tax shields (Bradley et al, 1984; Berk, 2006; De Jong & Van Dijk, 2007; Akhtar & Oliver, 2009). The reason why additional debt can lead to tax benefits has been explained above already, but I repeat it in short again. When adding additional debt, the rent that has to be paid over this sum of money can be distracted from the tax that needs to be paid, so the rent on the debt is tax-free while tax needs to be paid over equity. Therefore debt is preferred over equity when only taking into account the taxation. It is however also argued that additional debt has not only positive effects. Certain expenditures, like depreciation, are tax deductible, and this diminishes the positive effects of having more debt. Therefore, it is proposed that the level of non-debt tax shields is positively related to the level of debt, and therefore to the capital structure. It must be acknowledged however that this positive relationship is not confirmed by all empirical tests (Kayo & Kimura, 2011), it could be that the relationship is negative in the end since the positive effects of tax-free debt are being offset by the tax deductibility of certain expenditures as explained above. However, for this study a positive effect is hypothesized since this is supported by the trade-off theory. So, the following hypothesis is formulated concerning the non-debt tax shield:

H2g: The non-debt tax shield has a positive influence on the capital structure

Graph 1 shows the expected influence of the selected variables on the capital structure. These expected influences are tested twice, once for the period before becoming an EU member state and once again for after becoming an EU member state. Statistical tests have to show whether the hypotheses stated above are accepted or rejected and whether the conceptual model depicted below is correct or not.

(26)

25

4. Method

Within this section, more details are given on the exact sample for this research. Furthermore, the variables which are included in the hypotheses above are elaborated on by operationalizing them. Besides, the data sources and the data collection are explained so that an understanding in the exact way of doing the actual analysis can be understood more easily. Finally, the research method itself is explained and an explanation is given on the choice of this exact method.

4.1 Sample

Countries

As was already stated in the introduction, ten new member states joined the European Union in the sixth round in 2004. This sixth round of adding new member states forms the basis of this research. However, not all ten member states are the focus of this study. The decision was being made not to focus on all ten countries that joined the EU in 2004. The countries in Central and Eastern Europe are being chosen as the core group of interest for this study, since these countries were considered transition economies (Gabrisch & Werner, 1996) after the Soviet Union was dissolved in 1991. Since these countries were communistic at the time the Soviet Union existed, they had a hard time in becoming liberal market economies. Therefore it is interesting to see whether joining the EU had an influence on their capital structures, since this is a way of looking at the availability of capital in the market and at the development of the capital markets in these countries.

What is furthermore of importance when deciding the core group of focus for this study is to look at their currency. Some of these countries adopted the Euro after they became an EU member state, which is likely to influence their capital structure as well. Since this could change the outcomes of this research, the decision has been made to include these countries as a control group. The countries that have adopted the Euro are Estonia, Slovakia and Slovenia. So, instead of excluding these countries from the research, they form the first control group to test whether the results from the core group are consistent or not.

(27)

26 availability of data. Most of the companies in Cyprus have data available for only 2003 and 2004, which is a too small amount for this research.

So, now that the two control groups have been identified, it is also possible to name the countries which do belong to the core group of interest for this study. These are the former-transition economies of Central and Eastern Europe that did not adopt the Euro. This is a group of five countries, namely: the Czech Republic, Hungary, Latvia, Lithuania and Poland.

What needs to be taken into consideration as well are the country differences. Although the countries are similar in some economic or political conditions, lots of differences exist as well. Therefore the decision has been made to perform the statistical tests for each country individually, so to control for the country differences.

Time span

As was mentioned above, the countries that are included in this study became an EU member in 2004. To see whether there are changes in the capital structure before and after joining the EU, two groups of periods are made. The first group contains data from 2001 until 2003, so that it can be seen what the capital structure was before the EU membership. The second group runs from 2004 till 2006, so that the time span of both groups is equal. The capital structure of companies after EU membership is computed so that a comparison can be made between the two periods.

So for this research, the two periods that are being studied are 2001-2003 and 2004-2006. Companies

It is furthermore important to make a decision on which companies to include in the sample and how many companies to include. First of all, the decision has been made to use 100 companies per country, since the sample needs to be substantial in order to be able to draw conclusions that are generalizable (Keller, 2005). Furthermore, the companies needed to have enough data available, so the decision was made to include only companies that had data available for the entire time span. Furthermore, one of the requirements for companies to be included was that they had to be located in the country of interest.

So, the decision of the companies that are included in the sample was being made based on the availability of data and on the location of the company.

Industry

(28)

27 research is to draw conclusions on the entire country, not just on one industry. However, a dummy variable should be included to control for industry differences, to make the results more consistent and reliable (Bradley et al, 1984; Nivorozhkin, 2005).

4.2 Variables and operationalization

In the previous section the hypotheses and the conceptual model were being explained, so it is important to relate the concepts included into these things to the empirical reality. The way this can be done is to operationalize the concepts into variables that are measurable. This is being done below, starting with the dependent variable: capital structure; followed by the independent variables: growth, profitability, firm size, tangibility, non-debt tax shield, distance from bankruptcy, dividends, agency costs and bankruptcy costs; ending with the control variables: industry, and listing. Dependent variable

 Capital structure: to operationalize the capital structure both equity and debt should be included

in the definition, since these are the two sources of capital available. Following the example of other researchers (Chen et al, 1997; Ahktar & Oliver, 2009; Nunkoo & Boateng, 2010), the capital structure is measured as follows: __________book value long term debt____________

book value of long term debt + market value of equity Independent variables

 Growth: the growth opportunities of the company is measured following the method of Kayo

and Kimura (2011), namely the ratio of the firm total market value (debt plus equity market value) to total assets.

 Profitability: to measure the profitability of the company, the following method suggested by

Kayo and Kimura (2011) and by Titman and Wessels (1988) is being used: it is the ratio of profit before interest, tax and depreciation to total assets

 Firm size: to measure the firm size, the method suggested by Kayo and Kimura (2011) is being

used, namely to use the natural logarithm of sales.

 Tangibility: in order to measure tangibility, the ratio of fixed to total assets is being used, as also

being done by Kayo and Kimura (2011).

 Non-debt tax shield: to measure this, the method suggested by Fatouh et al (2008) is being used,

namely to measure this as the ratio of total depreciation to total assets.

 Distance from bankruptcy: this is being measured by Altman’s, modified by MacKie and Mason

(29)

28

 Bankruptcy costs: finally, bankruptcy costs are measured as the earnings volatility (Gaver &

Gaver, 1993; Skinner, 1993), which is the change of cash flows to the average asset over the sample period.

Control variables

 Industry: in order to see in which industry the company is operating, the US SIC code is used,

since this is a widely known code.

 Listing: if the company is listed on a stock exchange it is labeled as one, while not being listed is

labeled as 0.

To include the control variables in the regression analysis, the following things are done. Industry SIC code is used to make 10 groups (0-1000; 1001-2000 etc.) which are transformed into dummies. Listing is split into two groups (listed or non-listed), which are also transformed into dummies. Furthermore size is split into three groups (size<3, 3<size<6 and size>6); from this two dummies are being made. And finally age is split into three groups as well (age < 15; 15< age < 25 and age>25), again two dummies are used.

4.3 Data sources and data collection

Data sources

In order to be able to obtain the data that is needed for this research an existing databases is used. Orbis, which is a database of Bureau van Dijk, is one of the databases that is available through the university website, and it is very well suited for this research because Orbis contains comprehensive information on companies worldwide ( http://www.bvdinfo.com/Products/Company-Information/International/ORBIS.aspx). It covers over around 80 million companies around the world, of which half of them are European companies. It contains company financials, financial strength indicators, directors and contacts, stock data for listed companies, detailed corporate structures, industry research, and M&A deals. So it contains all the information that is needed for this research.

(30)

29 numerical databases, and it covers financial instruments, as well as equity and fixed-income

securities and indicators for over 175 countries

http://thomsonreuters.com/products_services/financial/financial_products/a-z/datastream/). However, this database has the same limitation as Orbis, namely that it is also a secondary data source. So, for collecting the data necessary, Orbis is the main database that is used, and DataStream is used as a complementary database.

Data collection

For collecting the data, it was already mentioned that Orbis is be used as the main source and DataStream as a complementary database. Furthermore, the data is gathered and filed into different categories per country. When collecting the data, the companies that have the largest amount of financial data available are selected first, and after that the list of companies is filled with companies that have less, but still enough data available that is needed for this research.

When collecting the data, all the information necessary is obtained first, and after that this data is converted into ratios that are needed to be able to measure the effect of the independent variables on the capital structure.

4.4 Data analysis

Now it is important to explain the different analyses that are being used to reach a conclusion on the research objective. In this section the statistical tests used are explained, advantages and disadvantages are given plus an explanation of choosing for these specific tests.

Paired samples t-test

First of all, to test whether there is a significant difference between the capital structure before and after attending the EU, paired samples t-tests are performed, which can test whether the capital structures of both periods are significantly different or not. The analyses are performed on country-level, so the conclusion of whether the capital structure has changed over time is drawn multiple times, for each country individually. After that, the control groups are compared to the main research group, to see whether the EU membership really accounted for the change in capital structure.

(31)

30 For this test it is assumed that the two samples that are being compared have different variances. Besides, the data has to be distributed normally, for which the Kolmogorov-Smirnov is performed (Keller, 2005). An α of 0,10 is being used to test for the significant change.

Regression analysis

After the t-tests have been performed, regression analyses are being done to test which variables can explain the possible difference in capital structure. By doing these analyses, it can be seen whether the proposed causal relations do exist in reality as well. Furthermore, tests are performed to secure the validity and the reliability of the results. When doing a regression analysis, one of the requirements for the data is a normal distribution. Besides, the correlation between the variables has to be low. Finally, outliers have to be removed. When all these requirements have been met, the regression analysis can be performed (Keller, 2005).

The method that is used to examine the relationships between the independent and the dependent variables, as stated above, is a regression analysis. The regression analysis is used to measure the linear, causal relationship between the independent variables and the dependent variable. Although regression analysis is often being criticized for its simplicity (Dos Santos and Porta Nova, 2007), this is also its major strength. The regression analysis is very often used because of this simplicity (Ingene and Lusch, 1980; Lord and Lynds, 1981). The unstandardized beta indicates whether the relationship is positive or negative. The effect is supported by the statistical tests if the coefficient of the independent variable is significant when using the two-tail test (p ≤ 0.10). The regression analysis can also be used to control for industry effects by including dummy variables.

Research quality

(32)

31

5. Country descriptions

Since it adds to the understanding of the reader to have some information on the sample countries, this information is given here. First of all a short country description is given for each country, together with some information on the capital markets in these countries (Ni & Yu, 2008; Delcoure, 2007; Nivorozhkin, 2005). After this has been done, the actual analyses start and the information given in this section can help in understanding the outcomes of these tests better.

5.1 Czech Republic

The Czech Republic is a country that is situated in Central Europe, and it officially exists since 1993, the capital is Prague. It was formerly known as Czechoslovakia and after the WWII it fell within the Soviet sphere of influence. Although the country tried to create socialism, these efforts were ended by Warsaw Pact troops. After the Soviet Union fell in 1989, Czechoslovakia regained its freedom and on January 1st 1993 the country underwent a ‘velvet divorce’ into two national components, namely the Czech Republic and Slovakia. In 2004 the Czech Republic joined the EU.

The Czech Republic has a population of around 10 million people (July 2011 est., CIA World Fact Book). GDP increased steadily over time, until 2008, after which it decreased (graph 2). It is a stable and prosperous market economy, with a relatively healthy conservative inward looking financial system, and a small open, export-driven economy which remains sensitive to changes in the economic performance of its main export markets. The car industry is the largest single industry and accounts for almost 20 percent of Czech manufacturing.

After it was considered a transition economy during the 1990s it is a parliamentary democracy now. The Czech privatization occurred in two waves in 1992 and 1995. The second wave of the privatization program boosted the share of the domestic product generated by the private sector to 70 percent (Nivorozhkin, 2005). Although the Czech Republic is considered a relatively successful transition economy, it faced problems with the enterprise and banking sector reform. There were insufficient bankruptcy law and poor enforcement, which resulted in insolvent companies, and overdue debts (EBRD, 2000).

A small number of foreign-owned banks came to dominate the sector after privatization and consolidation of Czech banks. These banks owned more than 90 percent of the total banking assets in 2001.

(33)

32 was established in 1992 and the market did not serve a source of capital for listed companies. In 2001 the market was illiquid with only six stocks actively traded.

5.2 Estonia

Estonia is one of the Baltic States, it is situated in Central Europe, next to the Baltic Sea, and the capital is Tallinn. After it was ruled by Denmark, Sweden, Germany and Russia, it became independent in 1918. However, it was forced into the USSR in 1940, but it regained its freedom with the collapse of the Soviet Union in 1991. After that, Estonia was free to promote economic and political ties with the West and it did so heavily and finally they joined the EU in 2004. Besides, Estonia adopted the Euro in 2011 (World Fact Book, CIA).

Estonia has a population of almost 1,3 million people (July 2011, est. World Fact Book, CIA). GDP increased steadily until 2008 and in 2009 it decreased (graph 2). It has a modern market-based economy and one of the higher per capita income levels in Central Europe and the Baltic Region. Estonia has a successive government, which pursued a free market, a pro-business economic agenda and which have wavered little in their commitment to pro-market reforms. The budgets are balanced due to sound fiscal policies and the public debt is very low. The electronics and telecommunications sectors are the largest industries in Estonia (World Fact Book, CIA).

Immediately after Estonia became independent from Russia, it started a small-scale privatization. In 1992 a large-scale privatization began, and liberalization of most consumer prices and a currency board arrangement were introduced. The private sector’s share in GDP increased gradually to 70 percent in 1996. The privatization of industrial companies was largely completed in 1999 (Nivorozhkin, 2005).

The privatization of Estonian banks proceeded rapidly after the first state-owned bank was privatized in 1995. During the 1990s the banking sector was characterized by active consolidation through mergers and by entry of foreign banks. The share of non-performing bank loans was very low and stable, just like the domestic credit to the private sector.

The Tallinn Stock Exchange was established in 1996 and by 2000 the stock market capitalization reached 35 percent of GDP (EBRD, 2001).

5.3 Hungary

(34)

33 announced withdrawal from the Warsaw Pact were met with a military intervention by Moscow. Already in 1968 Hungary began liberalizing the economy, introducing a so-called ‘Goulash Communism’. In 1990 the free market economy initiated and finally Hungary joined the EU in 2004. Hungary has a population of almost 10 million people (July 2011, est.). GDP increased until 2008, fell in 2009 and increased to $190 billion in 2010 (graph 2). Hungary was a centrally planned economy and transitioned to a market economy, but the per capita income was nearly two-third of the EU-25 average. The private sector accounts for more than 80 percent of GDP. There is a lot of foreign ownership and FDI in Hungary, which are worth more than $70 billion. The budget deficit had been reduced from 9 percent in 2006 to 3,2 percent in 2010.

By the late 1980s and 1990s fundamental laws on the banking system, foreign investments, the foundation of companies, trade, competition, labor, intellectual property and bankruptcy were laid down, while at the same time imports, prices and wages were liberalized. Hungary was the first country in CEE to launch market-based privatization, which resulted in a rapid increase in the FDI. By the end of the 1990s the privatization process was essentially complete with less than 20 percent of state assets remaining in government control.

Hungary has an impending inability to service its short-term debt, which led to obtain a financial assistance package worth over $25 billion.

The Budapest Stock Exchange was reopened in 1990 as the first post-communist stock-exchange in the Central and East European region. Currently, the BSE has 50 members of which 40 companies are floating their securities and the securities of more than 80 funds are available for investors (www.itdh.com).

5.4 Latvia

Latvia is the second of the Baltic States, and the capital is Riga. Latvia used to be under the control of Germany, Poland, Sweden and finally under the control of Russia. During WW I the Latvian republic emerged, but it was annexed by the Russians again in 1940. Following the break-up of the USSR in 1991, Latvia reestablished its independence, but the last Russian troops did not leave before 1994. In the spring of 2004 Latvia joined the EU and its major goal for now is to be able to adopt the Euro in 2014.

(35)

34 significantly to its GDP. The transit services are highly developed, along with timber and wood-processing, agriculture and food products, and manufacturing of machinery and electronic devices. Besides, the country is very active in the services sector. The currency of Latvia is pegged to the euro, and there was a large fiscal deficit of almost 10 percent of GDP in 2010. Latvia has targeted its major financial policy goal to enter the euro zone in 2014. The foreign investment in Latvia is however still modest compared to other north-central European countries.

In 1991 the privatization in Latvia commenced by adopting the major principles of state ownership transformation. Mass privatization began with the introduction of privatization certificates in 1992 and by the end of 1993 almost 70 percent of the small-scale privatization was completed. The large-scale privatization proceeded at a slower pace from 1991 to 1993. Therefore the privatization agency was created in 1994 which became responsible for implementing privatization. In 2000 the privatization in Latvia was almost complete, leaving only a small number of politically sensitive large state companies. The private sector accounted for nearly 68 percent of GDP in 2000 (www.fdi.net). In 1993 the Riga Stock Exchange (RSE) was founded by 4 Latvian commercial banks. The RSE completely owns the Latvian Central Depository, which is the depository for equity and debt securities. Right now, the Nordic and Baltic states are trying to develop an integrated Nordic Baltic securities market (www.danskebank.com).

5.5 Lithuania

Lithuania is the third and most southern country of the Baltic States and Vilnius is the capital. In the 14th century Lithuania was the largest state in Europe, when also Belarus and Ukraine were part of their territory. In the 16th century Lithuania and Poland formed a single dual state, namely the Polish-Lithuanian Commonwealth. In 1795 the remnants were partitioned by surrounding countries, but Lithuania regained its independence after World War I, but was annexed by the USSR in 1940. On March 11th 1990 Lithuania was the first of the Soviet republics to declare its independence but this was not recognized by Moscow until September 1991. The last Russian troops left in 1993 and Lithuania restructured its economy for integration into the Western world. In the spring of 2004 Lithuania joined the EU (World Fact Book, CIA).

(36)

35 current account surplus of 3,4 percent while it had a deficit of roughly 15 percent in 2007. Imports are being reduced sharply to maintain this number (World Fact Book, CIA).

Privatization in Lithuania started in September 1991 through mass privatization. During the first stage of privatization more than 5000 enterprises were sold, which was nearly the entire amount of small businesses and some 60 percent of the industrial enterprises. A significant part of this privatization process took place in the agricultural sector, due to the Law on Privatization of Agricultural Enterprises. The second phase of privatization started in July 1995, when the Law on the Privatization of State-Owned and Municipal Property was being accepted. These properties were being sold for cash to foreign investors as well. The privatization of the banking sector was being completed in 2002 and right now the privatization is almost totally complete.

The stock exchange of Lithuania, the NASDAQ OMX Vilnius, was established in 1993 and it is owned by the OMX. Lithuania, Latvia and Estonia want to establish an integrated Nordic Baltic securities market with the Nordic states (www.nasdaqomxbaltic.com).

5.6 Malta

Malta is a country that is situated in southern Europe, an island in the Mediterranean Sea. The capital is Valletta. In 1814 Great Britain formally acquired possession of Malta and Malta kept supporting the UK through both World Wars and it remained in the Commonwealth when it became independent in 1964. In 1974 Malta became a republic and since about the mid-1980s, Malta transformed itself into a freight transshipment point, a financial center and a tourist destination. In 2004 Malta became a member of the EU.

Malta has a population of 408.000 people (July 2011, est.) and it had a GDP of $11, 7 billion in 2010, after it slightly decreased in 2009 (graph 2). Malta is a country which is highly dependent on imports, since it produces only about 20 percent of its food needs, had limited fresh water supplies, and few domestic energy sources. The most important industry in Malta is the tourism sector, which accounts for one third of total GDP.

In recent years, Malta’s financial services industry has grown and this sector is sector is centered on the indigenous real estate market and is not highly leveraged. Furthermore Malta has a very stable banking system and strong prudent risk-management practices. In 2010 Malta had an excessive deficit according to the EU targets, but this target is about to be reached again in 2011.

Capital structure in Central and Eastern European Countries: what is the role of EU membership?