• No results found

The impact of market timing on current capital financial structure

N/A
N/A
Protected

Academic year: 2021

Share "The impact of market timing on current capital financial structure"

Copied!
41
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

The impact of market timing on current

capital financial structure

Naam student: Tjipke van der Kooi Studentnummer: 11284404

Scriptie Executive Master of Finance and Control 5 oktober 2018

Begeleider: dr. S.W. (Sanjay) Bissessur Amsterdam Business School (UVA) Niet vertrouwelijk

(2)

Verklaring eigen werk

Hierbij verklaar ik, Tjipke van der Kooi, dat ik deze scriptie zelf geschreven heb en dat ik de volledige verantwoordelijkheid op me neem voor de inhoud ervan.

Ik bevestig dat de tekst en het werk dat in deze scriptie gepresenteerd wordt origineel is en dat ik geen gebruik heb gemaakt van andere bronnen dan die welke in de tekst en in de referenties worden genoemd.

De Faculteit Economie en Bedrijfskunde is alleen verantwoordelijk voor de begeleiding tot het inleveren van de scriptie, niet voor de inhoud

(3)

ABSTRACT

This paper investigates the impact of (historical) misvaluation on (current) capital financial structure. We research whether the current capital financial structure is the cumulative outcome of historical attempts of financial managers to time the equity markets in order to achieve relative cheap equity financing. If misvaluation has an impact on capital structure this will indicate an agency problem between existing and future shareholders. The concept of misvaluation is measured by a proxy of undervaluation based on the ratio of intrinsic fundamental value of equity to market value of equity. The seasonal equity issues in historical years are weighed by their importance to the cumulative total of equity issued in the previous years and the specific year under investigation. In contrast to several US studies on market timing theory and capital financial structure, our research doesn’t find evidence in support of the market timing theory. In fact, elaborating on the main study in five subsamples, we find that historical overvaluation shows a negative relation to financial leverage in the subsample with the strongest overvaluation. When firms are already overvalued and overvaluation rises further, European firms attract relative more debt than equity to finance their investments. This causes financial leverage to increase. Also the levels of financial leverage over the subsamples don’t differ significantly. The subsample results are the opposite of what is expected from the market timing theory and consistent with the trade-off theory. Due to the fact that our research doesn’t find evidence in support of the market timing theory, an agency problem between existing and future shareholders in an European context isn’t proven neither. Further research on the impact of the type of financial system, bankruptcy codes, corporate control mechanisms and patterns of ownership should be explored to fully understand under which circumstances market timing plays a role in the capital financial structure of firms.

(4)

1

TABLE OF CONTENT

1. INTRODUCTION ... 2

2. LITERATURE REVIEW ... 6

2.1 Agency problem theory ... 6

2.2 Capital financial structure ... 9

2.2.1 Trade off theory ... 9

2.2.2 Pecking order theory ... 9

2.2.3 Other determinants of Capital financial structure... 10

2.3 Equity misvaluation ... 10

2.4 Relation between misvaluation and capital financial structure ... 12

2.4.1 Research on the existence of market timing ... 12

2.4.2 Relation between market timing and capital financial structure ... 13

3. HYPOTHESIS ... 14

4. DATA METHODOLOGY ... 15

4.1 Historical equity finance weighted average V/P ... 16

4.2 Control variables ... 19

5. CROSS SECTIONAL TESTS ... 24

5.1 Full sample regression tests ... 24

5.2 Interpretation of the full sample test ... 24

5.3 Subsample data and tests ... 26

5.4 Interpretation of the subsample tests: hypothesis 1 ... 26

5.5 Interpretation of the subsample tests: hypothesis 2 ... 27

6. ROBUSTNESS TESTS ... 30 6.1 Fama-MacBeth ... 30 6.2 Interpretation of Fama-MacBeth ... 30 6.2 Macro economy ... 30 7. CONCLUSION ... 34 8. REFERENCES ... 36

(5)

2

1. INTRODUCTION

This paper examines if historical misvaluation impacts (current) capital financial structure due to financial managers who use misvaluation to time equity issues. More specifically, this paper examines if the company’s current capital financial structure is the outcome of an accumulation of historical attempts to time the equity market through strategic issuance of equity. By focusing on the relationship between historical overvaluation and current capital financial structure, this paper contributes to the existing literature in this field. Also it answers the question whether overvalued stock market valuation can create or exacerbate agency problems and costs.. If managers systematically time the equity market, an agency conflict between existing and future shareholders is created at the expense of corporate efficiency and value creation (Jensen, 2005).

Within the academic world, the topic of capital financial structure has been widely researched. Much empirical research that has been conducted, are based on traditional finance theories to explain capital financial structure of firms. The trade-off theory explains capital financial structure as a balance between benefits and costs of attracting debt, and is based on market frictions and transaction costs. The pecking order theory is based on information asymmetries and signaling effects. However, Baker and Wrungler (2002) have added a new dimension to the discussion on how capital financial structure is formed. They state that the current capital financial structure is the cumulative outcome of past attempts to time the market.

The study of Baker and Wrungler is built upon inefficient market theory which predicts that mismatches between market value and fundamental intrinsic value do exist (Black, 1986). Consequently, derived from the Agency Cost theory of Jensen and Meckling (1976), it can be expected that financial managers (agents) will adopt their financing channel strategy to exploit these mismatches, whether caused by true market misvaluation or perceived misvaluation. In case a company is overvalued, according to financial management, it will likely choose to issue new “cheap” equity. Due to the overvaluation, the effective cost of equity for existing shareholders is lowered. In case of undervaluation, managers fund investments from internal cash flows or debt issues, since issuing new equity can be regarded as “expensive” to existing shareholders.

There is a lot of empirical evidence on the existence of timing in issuing equity and debt using different methodologies like ex-post issue returns of stock or ex-ante stock-price run ups. In addition, other studies have elaborated on the research of Baker and Wrungler (2002) by researching the relation between overvaluation and the timing of stock or debt issues in relation to capital financial structure. Dong, Hirshleifer and Teoh (2012) find that equity issuance and total financing increase with equity overvaluation, but only among overvalued stocks, and that equity issuance is more sensitive to overvaluation than debt

(6)

3 issuance. Among others, Hovakimian, Hovakimiam and Tehranian (2004), Kayhan and Titman (2007) and De Bie and De Haan (2007) have studied the relation between overvaluation and capital financial structure and found evidence for market timing.

However, there is no consensus on the validity of the often-used construct of historical market-to-book is as a proxy for misvaluation. Mahajan and Tartaroglu (2008) argue historical market-to-book ratios can be seen as a proxy for growth opportunities. Furthermore, there remain inconsistencies in the interpretation of findings on the persistence of the effects of market timing on capital financial structure with regards to the (dynamic) trade-off theory or market timing theory. Leary and Roberts (2005), Kayhan and Titman (2007) and Hovakimian (2006) suggest that the impact of market timing on capital financial structure is balanced away in the short term and advocate the trade-off theory while Huang and Ritter (2006) find long lasting effects on capital financial structure.

This research combines methodologies and insights of market timing research that focuses on the relation between misvaluation and capital issuing activity and the firm capital financial structure. Although the dependence of capital financial structure on overvaluation via equity and debt issuing has already been researched before, this study uses a different valuable proxy of misvaluation. This variable eliminates the contaminating effect of future growth perspectives and firm risk away and therefore is a more consistent measure of misvaluation. Previous research on the relation between misvaluation and capital financial structure, like the research of Baker and Wrungler (2002), has been strongly criticized for using historical market-to-book as a proxy for overvaluation (Mahajan and Tartaroglu, 2008). Other research has shown that historical market-to-book can also be interpreted as a proxy for future growth perspectives of the firm. As a consequence, there is no consensus about the found negative coefficient between (cumulative) historical market-to-book and financial leverage as proof of market timing theory or trade-off theory. To my knowledge, there are no studies that use this benchmark as a proxy of misvaluation in researching the persistence of the impact of misvaluation on capital financial structure. In contrast to the majority of other studies on the relationship between misvaluation and capital financial structure, this study uses an European sample consisting out of the seven largest European countries by stock index market capitalization. The total full sample consists out of 20.865 firm-year observations.

The approach to measure the effect of historical misvaluation on current capital financial structure is to use the V/P proxy as applied by Dong et al (2012) in their test for misvaluation-effects on equity and debt issues. The intrinsic value (V) is obtained by applying the residual income model of Ohlsen (1995). The widely applied market-to-book ratio of equity relies on the backward looking value measure of book value. In contrast, the V/P-proxy applied in this study is a forward looking value measure, that excludes growth

(7)

4 opportunities and firm risk as factors. The V/P ratio is measured by the equity finance weighted average V/P formula (EFWAVP). The EFWAVP formula is taken from Baker and Wrungler (2002), who introduced the external finance weighted average market-to-book value as a proxy for overvaluation. The formula is used in this research to focus on the importance of historical, for equity issues weighted, overvaluation on current capital financial structure.

To test additionally for historical misvaluation effects on capital financial structure, quintile sorting by V/P is applied to test whether there are differences between subsamples in the relation between historical weighted overvaluation and current capital financial structure. Also, subsamples are used to test for the sensitivity of the relation between overvaluation and financial leverage in the different subsamples.

Our research indicates that cumulative historical equity weighted V/P values do not explain capital financial structure in the full sample. Our research therefore, doesn’t find evidence in support of the market timing theory. In fact, elaborating on the main study in five subsamples, we find that historical overvaluation shows a negative relation to financial leverage in the subsample with the strongest historical overvaluation. When firms are already overvalued and overvaluation rises further, European firms attract relative more debt than equity to finance their investments. This causes financial leverage to increase. Because persistent effects of equity timing efforts on capital financial structure is not found the results of the research are more consistent with the trade-off theory.

First, this study revisits the debate on the impact of misvaluation on capital financial structure by using a historical misvaluation variable that eliminates the effect of future growth perspectives. Also the use of an European sample contributes to the robustness of previous findings which are primarily focused on US firms. This is important, since most studies are focused on the US. The US has a highly market orientated financial system in which firms can tap the public capital markets easily, and therefore can relatively easily apply timing behavior in the financial markets. Europe has a more bank-orientated financial system and differs from the US in tax and bankruptcy codes, corporate control mechanisms and patterns of ownership (Rajan and Zingales, 1995). These differences can influence the preference of financial management and the opportunities to issue equity. Second, this study contributes to the design of performance based reward systems. Performance based reward systems at public firms depend partly on stock performance. If financial managers time equity issues to stock overvaluation core value will be destroyed in the long run (Jensen, 2005). A persistent impact of overvaluation on capital financial structure indicates an agency conflict between existing and future shareholders. The existence of such an agency conflict would stipulate the need for new designs of performance based reward systems. Reward systems would need to be corrected for short term behavior of financial management in equity issuing.

(8)

5 Instead they should be designed in such a way that they will rewarded on how financial management perform against strategic plans.

The remainder of this paper is as followed. In section 2 we first discuss Agency Cost theory and it’s predicted impact on capital financial structure, followed by a discussion of existing theory on the dependent variable of capital financial structure and the independent variable of overvaluation. It will be concluded by discussing the research focusing on the relation between overvaluation and capital financial structure. Section 3 is about the hypothesis tested and section 4 deals with the data and research methodology used in the research. Section 5 discusses the regression results. Section 6 focuses on the robustness of the regression results and the conclusion of the research is dealt with in the final section.

(9)

6

2. LITERATURE REVIEW

2.1 Agency problem theory

Historically there has been much focus on two kind of agency problems namely agency problems between shareholders and managers and agency problems between large and small shareholders. In his recent paper Jensen (2005) stresses the agency problem between existing and future shareholders caused by engagement of management trying to exploit overvaluation on capital markets for the benefit of their own and existing shareholders.

In their pioneering paper Jensen and Meckling (1976) have integrated elements of the theory of agency into the theory of the ownership structure of the firm. They define an agency relationship as a contract under which one or more persons, the principal(s), engage another person, the agent, to perform some service on their behalf which involves delegating some decision making authority to the agent. As agents are utility maximizers it can’t be expected that agents will always act in the best interest of the principals. In order to restrict divergences from their interests principals establish incentive structures and incur monitoring costs and bonding costs. Because it’s impossible that actual behavior of agents under all circumstances is in the best interest of the principal some residual agency costs will remain. This agency cost problem is inherent to the separation of ownership and control structures in which, regarding Jensen and Meckling (1976), the agent will only bear a fraction of the cost of any non-pecuniary benefits he takes out in an effort to maximize his own utility, depending on the share of equity or the design of incentive contracts. Also, there exists much research about the second agency problem between large and small shareholders. Small shareholders do not undertake any monitoring and free ride on the back of the larger shareholder who bear the costs of monitoring like voice or exit strategies (Erdmans, 2014). On the other hand, larger shareholder can harm smaller shareholders by collusion with management against smaller shareholders (tunneling). The large shareholder may tunnel corporate resources away by inducing business ties with other firms on unfavorable terms (Erdmans, 2014). Second, large shareholders’ voting decisions may be conflicted for example when a mutual fund ties with friendly directors to preserve business ties (Davis and Kim, 2007). Third, large stakes may cause the large shareholder to be concerned about idiosyncratic risk and induce the firm to forgo risky, value-creating investments (Dhillon & Rossetto 2014). Other costs consist out of reducing liquidity in the stock (Bolton & von Thadden, 1998) and over monitoring costs in the situation managers exert less effort because they fear their projects are not accepted by the shareholders (Erdmans, 2014).

The agency cost theory has direct and indirect impact on a wide array of major topics in corporate finance like capital financial structure. Capital financial structures can be influenced by issuing equity or debt to finance new investments or by recapitalization

(10)

7 transactions. Also dividend payout policy can be used to steer capital financial structure. In this respect the tendency of agents to maximize their own utility can also be expected to play a role in the issuing activity. Issuing activity on its turn influences capital financial structure.

In the traditional corporate finance theory based on market efficiency financial channels suffer from adverse selection (Akerlof, 1970). As financial markets know that managers wish to sell shares at times they perceive their shares to be overvalued (Stein, 2001) equity issues are seen to be bad news as shown by Asquith and Mullins (1986) and Mikkelson and Partch (1986). This makes managers of good firms reluctant to raise equity to finance themselves and their projects. The underlying assumption of efficient markets make that agency conflicts like short-termism or overconfidence are expected to result in relative scarce equity issues. However, the assumption of efficient markets has been widely questioned the last thirty years by empirical research focusing on the relationship between stock prices and investments. This relation has been studied by Fisher and Merton (1984), Morck Shleifer and Vishny (1990a) and others. Only semi-efficient or inefficient markets are able to explain the outcomes of the measured relationship of stock prices and investments. Stein (1996) for example hypothesizes that for firms which don’t have plenty of cash

available investment levels depend on the non-fundamental variations in stock prices. Given the fact that many firms don’t have the cash and internal cashflow available to finance their investments, the sentiment on financial markets can be seen as an important input for the ability of managers to create value. Also for management who acts in the 100% very best interest of existing shareholders it pays off to time the market (Stein, 1996).

The inefficient market approach influences capital financial via the mechanism of issuing activity. Keynes (1936) already stated that stock prices contain an important element of irrationality. As a consequence the effective cost of external equity diverges from other channels of financing. The inefficient approach predicts that a firm will raise more equity when it obtains a higher price relative to fundamental intrinsic value for the issued equity and repurchases equity when it pays a lower price than the fundamental intrinsic value of the equity it repurchases. In their research to the influence of stock prices on investment behavior of equity-dependent firms Baker, Stein and Wrungler (2003) show that firms will raise more capital in response to overvaluation. Also Stein (1996) proved that market

irrationality is a predicting force of stock returns and that firms raise more capital in response to overvaluation.

Because overvaluation has a very direct effect on the effective cost of issuing equity and less on the effect cost of debt, overvaluation gives management the opportunity to extract value from new shareholders (Dong et al, 2012). The extracted value is transferred to the existing shareholders. Even a manager with a long term value focus will issue more equity when it is overvalued because it generates a profit for the existing firm (Stein, 1996).

(11)

8 In case of undervaluation management won’t issue equity because it transfers value from the existing shareholders to the new shareholders. In contrast with equity, the influence of misvaluation on debt issuance is less evident. On the one hand, overvaluation encourages the issuance of risky debt. On the other hand, for a given amount of financing, greater overvaluation causes substitution from debt to equity issuance (Dong et al, 2012). In their study about the influence of overvaluation on financing activity Dong et al (2012) find an insignificant relation between debt issuance and overvaluation.

The agents’ own interests are also served by playing the earnings game and by issuing overvalued equity to finance ambitious investment programs. In a more recent paper Jensen (2005) expands the earlier determined range of agency costs by adding market and managerial optimism. Jensen builds on the examples of corporate scandals in the years around 2000 and creates a new version of agency costs. Jensen states that in the early stages of overvaluation financial management realizes, that after all the value creating alternatives have been taken, it cannot produce the performance required to justify the stock price. In this situation it’s nearly impossible for management to admit and communicate that the stock is overvalued. Doing so would harm the stock price immediately and this will severely affect the performance based incentive pay of management. Also it will imply negative reputational and career consequences for financial management. As an alternative financial management decides to cater with the current optimism. They start the earnings game and finance the acquisitions, greenfield investments and other major business decisions with new issued equity. By meeting or beating the inherent growth targets of the financial markets, in combination with the issue of new equity by which existing shareholders benefit, the stock price rises even further. Additionally, financial management benefits via agency cost mechanisms like empire building, a growing professional reputation and bigger financial rewards. Catering to the current sentiment is an agency cost because at the moment the market finds out that the high value and growth was an illusion it will backfire. As a consequence, all the overvaluation will disappear as well as the value of the core business. Moeller, Schlingemann and Stulz (2005) show that in the three-day period surrounding the announcement of acquisitions in the period 1998-2001 firms lost a total value of USD 240 billion. This loss has not been totally attributable to the value transfer to sellers. They explain the difference in values as “a reassessment of the standalone value of the acquirer”.

Building on the wide empirical evidence that financial markets are inefficient and with the insight that market overvaluation has compelling cross-over effects to financial management, who act partly in their own interest, it can be expected that market timing is a widespread phenomenon. In a study of Graham and Harvey (2001) CFO’s even report that stock market valuations are an important consideration in their firms’ decision to issue equity. Because issuing new equity is preferred above attracting new debt in a situation of

(12)

9 overvaluation it is predicted that agency cost theory links the concepts of overvaluation with capital financial structure.

2.2 Capital financial structure

2.2.1 Trade off theory

The trade-off theory is derived from the Modigliani and Miller theory that capital financial structure is irrelevant for firms (Modigliani & Miller, 1958). The trade-off theory builds further on Modigliani and Miller by adding market deficiencies like taxes and financial distress costs but retains market efficiency and symmetrical information as a basis. The market deficiencies like taxes and financial distress costs cause that firms try to optimize the total value for the firm by maximizing the net result of benefits of debt and the costs of debt. Therefore, they are not indifferent to capital financial structure.

The main benefits from debt come from the tax shield value of debt and the disciplining effect (Jensen 1986a). This causes firms with large and stable cash flows and assets that can be collateralized (Myers and Majluf, 1984) to attract more debt in order to profit from the tax shield created by the debt base. On the other hand, unique companies with high bankruptcy costs (Titman, 1984), high business risks and high growth opportunities (Bradley, Jarrel and Kim, 1984) use less debt because of relatively high distress costs.

The tradeoff theory states that every firm has its own optimal capital financial structure and adjusts its capital financial structure in response to temporary shocks that cause its leverage to deviate from the target. When stock prices rise to an overvalued level the financial leverage will be driven down which results in a leverage ratio lower than the target ratio. In order to convert back to the target under the trade off theory it is expected firms issue debt in order to increase the leverage. The tradeoff theory therefore expects that rising overvaluations have a positive effect on debt issues and therefore indirectly a positive effect on financial leverage measured by financial book leverage.

2.2.2 Pecking order theory

The pecking order theory (Myers and Maljuf, 1984) is based around the concepts of asymmetrical information and agency cost theory. The adverse selection problem exists because of asymmetric information between agents and principals. This adverse selection problem affects managers’ choice between internal and external financing and between issuing equity or debt.

To the extent that managers favor the interests of existing shareholders, they will wish to sell new shares at times of regarded overvaluation (Stein, 2001). As well informed

financial managers know that the attempt of issuing new equity will downplay the stock price to near or under the fundamental intrinsic value, because of the signal the equity issue sends

(13)

10 to the market, they will end up issuing debt so that they don’t harm existing shareholders who can sell their stock at current overvalued prices. The pecking order theory suggests a priority ranking in financing investments. Investment is financed first with internal funds, then by new issues of debt and finally with new issues of equity. Within the pecking order theory, the costs of deviating from the optimum point of capital financial structure are insignificant to the costs of raising external finance (Baker and Wrungler, 2002).

The dynamic version of the pecking order introduced by Lucas and McDonald (1990) predicts that firms issue equity after high market performance. From this it is expected that rising overvaluations have a positive effect on equity issues and therefore have an indirect a negative effect on financial leverage measured by book leverage.

2.2.3 Other determinants of Capital financial structure

In their investigation to find the main determinants of capital financial structure of public firms in the major industrialized countries (G7) the major objective of Rajan and Zingales (1995) is to establish whether capital financial structure in other countries is related to factors similar to those appearing to influence the capital financial structure of U.S firms. They base their testing on four specific variables that influence financial leverage on previous studies of Bradley, Jarrel and Kim (1984), Long and Malitz (1985) and Harris and Raviv (1991). They find that tangibility and size are positively related to leverage for all countries, except for Germany where size is negatively correlated with leverage. Due to the exception found for German firms Rajan and Zingales point at the possibility that the theoretical underpinnings are flawed or that different institutional contexts, like for example bankruptcy laws, do have severe impact on the real relationships. On the contrary to size and tangibility, they find negative coefficients for market-to-book and profitability to financial leverage.

Based on their finding that the negative correlation between market-to-book value with leverage is mainly driven by large equity investors puzzles them because it’s not totally consistent with the theoretical explanation that firms with higher book-to-market have higher costs of financial distress. In relation to this puzzle they point at market timing as a possible explanation.

2.3 Equity misvaluation

The classic theory of securities market equilibrium, based on fully rational investors introduced by Sharpe (1964) implies that market prices always accurately and fully reflect available information. Fama (1970) elaborated on this classical theory of market efficiency by differentiating between weak form, semi-strong form and strong form market efficiency. Within a world of market efficiency equity misvaluation can’t exist.

However, previous research has proven that markets don’t always fully reflect available information and can behave inefficient. According to Jensen (2005) a companies’

(14)

11 equity is overvalued when the market value is higher than the underlying value of the company. Except for pure luck the company won’t be able to deliver the performance to justify the too high market value.

There have been various explanations for misvaluation divided in a market approach and an individual investor approach. One theory as part of the market approach is that the market suffers from noise or positive feedback trades. Black (1986) states that on the one hand noise helps create financial markets but on the other hand also makes them imperfect. He describes noise trading as trading on noise as it were information. This noise trading actually puts noise intro the prices and creates misvaluation. De long, Shleiffer and Summers (1990b) state that noise traders follow positive feedback strategies in which they anticipate on positive or negative price cycles, consistent with overreaction or price bubbles, which causes the stock price differ even further from fundamental intrinsic value. As part of the individual investor approach Hirshleifer, Subrahmanyam and Titman (1994) developed a model that implies that investors will focus only on a subset of securities while neglecting other securities with identical exogenous characteristics. This reflects the behavioral finance concept of herding. De Long, Shleiffer and Summers (1990a) point at the unpredictability of noise traders' beliefs. Those beliefs create a price risk that deters rational arbitrageurs from aggressively betting against them. As a result, prices can diverge significantly from fundamental intrinsic values even in the absence of fundamental risk. Daniel, Hirshleiffer and Subrahmanyam (2001) show that misvaluation as much as risk, is a determinant of future stock returns.

Many researches which use overvaluation as the independent variable in explaining market timing or capital financial structure use Tobin’s Q or market-to-book value as a proxy for overvaluation. In their leading paper showing that capital financial structure can be explained as the consequence of cumulative timing attempts Baker and Wrungler (2002) use historical market-to-book value as a basis for their proxy on overvaluation. De Bie and De Haan (2007) also use historical market-to-book value as a proxy of overvaluation in their research about the effect of market timing on the capital financial structure of Dutch firms. Following Baker and Wrungler, current market-to-book ratios are used in the regression to control for growth opportunities. Mahajan and Tartaroglu (2008) however question the dual role of market-to-book values being a proxy for overvaluation as well as for growth opportunities. They state that like current market-to-book ratios historical market-to-book values can also be explained as characteristics that capture growth opportunities. Kayhan and Titman (2007) state that market-to-book values relate more to the pecking order theory and have nothing to do with market timing. They argue that it can be the case that firms with high market-to-book values are more willing to issue equity because they suffer less from asymmetric information problems or that they are more willing to be exposed to the increases

(15)

12 scrutiny of the public market when their shares are issued. Dong et al (2012) state that Tobin’s Q or market-to-book ratios do not only include misvaluation but also growth perspectives or firm risk and depend partly on accounting methods. Also they emphasize that market-to-book ratio’s and Tobin’s Q are backward looking value measures because they are based on book value. Therefore, those measures are and can be used for other firm characteristics like earnings growth perspectives, investment opportunities or managerial effectiveness but are less useful for misvaluation. To avoid the issue announcement effect on the mispricing itself (“mispricing affects market price”) they use the forward-looking measure of intrinsic value-to-price ratio. Daniel et al (2001) show that market-to-book value is relatively heavily weighted towards growth opportunities compared to misvaluation.

2.4 Relation between misvaluation and capital financial structure

The research on the relation between misvaluation and capital financial structure can be built on past research. This past research must be logically divided in two classes of research. First, there has been research focused on the existence of market timing. Second, there has been research on the relation between market timing and capital financial structure. In addition to this there has been research about the persistence of the effect of market timing on capital financial structure.

2.4.1 Research on the existence of market timing

The existence of market timing is explored either via researching the relation between ex-ante overvaluation and stock issuing activity or via post-issue stock returns.

Dong et al (2012) find that equity issuance and total financing increase with equity overvaluation, but only among overvalued stocks, and that equity issuance is more sensitive than debt issuance to misvaluation. Eckbo and Masulis (1995) show that stock issues occur after stock price run ups. However, Korajczyk, Lucas, and McDonald (1991) and Bayless and Chaplinsky (1996) point at information asymmetry effects however as an explanation for the found relation. Furthermore, under the rational q theory of investment (Brainard and Tobin 1968; Tobin 1969), a stock price run up or a high stock price relative to book value indicates an improvement in growth opportunities, which encourages the firm to invest more, and perhaps to raise more capital as financing.

Empirical studies by Loughran and Ritter (1995), Ikenberry, Lakonishok and Vermaelen (1995) and Ritter (2003) show that managers time the issuance of equity and the repurchase of shares. Loughran and Ritter show that companies issuing stock during 1970 to 1990, whether an IPO or a SEO, significantly underperform relative to non-issuing firms for five years after the offering date. They explain the difference in returns between issuing and non-issuing firms by the announcement effect of equity issuing. When managers issue new equity the market interprets the company as overvalued. The announcement effect indicates

(16)

13 that the market believes managers have a timing ability. There is no unanimous agreement about timing effects, however. Brav, Gecy and Gompers (2000) show that returns after an IPO are similar to the returns of non-issuing firms. They also find similar results for returns following a SEO and returns of non-issuing firms. They conclude that model specification is an important consideration in testing anomalous returns which strongly influences returns.

2.4.2 Relation between market timing and capital financial structure

Baker and Wrungler (2002) have extended the existing work on equity market timing by providing an alternative hypothesis for explaining observed capital financial structure. They find that, controlling for firms’ growth opportunities by using current market-to-book ratio, leverage is inversely related to historical market-to-book ratio, which they interpret as evidence supporting an equity market timing theory. They further argue that the significance of historical market-to-book ratio in explaining leverage is inconsistent with the trade-off theory. The effect of past equity issues on capital financial structure is persistent even during a period of 10 years. In trade-off theory, as firms adjust to their optimal capital financial structure, temporary shocks such as market timing attempts should not have a long-lasting effect on firms’ leverage. The study of Huang and Ritter (2006) supports the conclusion that past security issues have a long-lasting effect on capital financial structure.

Other studies also find evidence that misvaluation, exploited via market timing efforts, affects capital financial structure. Kayhan and Titman (2007) find a strong relation between historical equity finance weighted average market-to-book and debt ratios. On the other hand, there are other studies which question this long lasting effect of equity market timing on capital financial structure. Other studies of Leary and Roberts (2005) Flannery and Rangan (2006), Kayhan and Titman (2007), Alti (2006) and Hovakimian (2006) also find this relationship but suggest that the impact of equity market timing on leverage is short lived. Leary and Roberts (2005) show that a typical US firm rebalances its capital financial structure in three to five years following the equity issuance.

In addition to the questioning if equity market timing plays a persistent role in capital financial structure the dynamic trade-off models which support the trade-off theories, as showed by a study of Fischer, Heinkel and Zechner (1989) suggest long adjustment periods and large deviations from target capital financial structure in the presence of even small adjustment costs. Hence, slow adjustment imposes a relation between historical ratios and leverage. Hennessy and Whited (2005), Liu (2005) and Hovakimian (2006) conclude that a negative coefficient between market-to-book and financial leverage is more consistent with trade off theory than with equity market timing theory of capital financial structure. Core behind this conclusion is the disagreement if historical market-to-book values are a proxy for market misvaluation or for future growth opportunities.

(17)

14

3. HYPOTHESIS

The inefficient market approach combined with agency cost theory predict that CFO’s will adapt financing channel strategies to exploit market overvaluation. Many previous research has proven that markets don’t always fully reflect available information and can behave inefficient creating misvaluation. This misvaluation can come in the form of overvaluation which lowers the effective cost of equity and makes issuing equity relatively more attractive or undervaluation which has the opposite effect. Financial catering in a situation with overvaluation serves existing shareholders because issuing equity is a relative cheap way of financing new value creating projects. This increases their existing equity value stake. With this, effectively wealth is transferred from new to existing shareholders. From agency cost theory it can derived that issuing equity also serves management. Issuing equity caters to high expectations of the market and can maintain overly optimistic perceptions about investment opportunities (Polk and Sapienza 2009; Jensen 2005). This serves management own corporate careers. This is an agency cost because when the overvaluation becomes apparent to the market not only the overvaluation but also core value will be destructed (Jensen, 2005). The hypotheses are used to test whether historical overvaluation, through equity market timing, affects (current) capital financial structure

Hypothesis 1. There exists a negative relation between the level of financial leverage and

the degree historical overvaluation

If firms systematically time their equity issues to the market, historical overvaluation will have a significant negative effect on financial leverage. If firms do not systematically time their equity issues the relation between historical overvaluation and financial leverage will be insignificant. If historical overvaluation is statistically significant associated with financial leverage that means that current capital financial structure is influenced by the cumulative attempts of financial management to time the equity market

Hypothesis 2. The sensitivity of weighted historical overvaluation on financial leverage is

non-lineair to the degree of overvaluation

The stronger the overvaluation is the cheaper issuing new equity becomes and the stronger the catering incentives for management will be to live up to high expectations. Therefore, the relation between historical overvaluation and financial leverage is expected to be stronger when overvaluation is high. According to Jensen (2005) the expected negative relation between overvaluation and financial leverage should be especially expected in cases of substantial overvaluation.

(18)

15

4. DATA METHODOLOGY

The sample includes companies listed on selected European stock indexes that are covered by Compustat in the period 1997-2017. The European sample is restricted to companies listed on the stock indexes of the UK, Germany, France, Italy, Spain, Switzerland and the Netherlands. These stock exchanges do have the largest market capitalization in Europe and therefore give a fair representation whether equity timing efforts influence capital structure on European public markets. The sample selection is solely based on the country where the stock exchange is based because not the country of incorporation but the liquidity and development of the public stock exchange determines the degree of opportunities to act on misvaluation via equity issuing activity. Because the selection is made on index instead of companies’ countries of incorporation the sample also includes for a small part non-European countries who have an non-European listing. Measurement will take place on a yearly basis due to the fact that firms publish financial reports annually. The sample is restricted to firms for which the necessary accounting items are available in Compustat to calculate the V/P ratio, yearly equity issues and capital financial structure. Firms in the regulated (two digit SIC of 49) and financial industry (firms with a SIC code between 6000 and 6999) are eliminated. They are eliminated because it can be expected that financial managers of regulated firms have less discretionary space to adopt financial catering strategies because of the financial policy restrictions those regulations bring with them. Firms in the financial sector are also heavily bounded by both market scrutiny and strict government regulations regarding capital financial structure which in effect prohibits them to adopt their capital financial structure. Firms for which book value of assets is less than EUR 10mio and firms for which the financial book or market leverage is higher than 1 are also eliminated. We further restrict for outliers at the 1% and 99% interval in the all firm-years dataset for capital financial structure (book debt ratio and market debt ratio). Because earnings analyst forecast data on EPS for companies on European stock indexes are not available on I/B/E/S or other available databases realized leading EPS and realized leading book value of equity available on Compustat are used instead. Missing control variables for a certain firm year will not limit the sample size.

This research focuses on the impact of historical overvaluation combined with equity issuing timing efforts on capital financial structure. Timing efforts are captured by a new variable formed by multiplying two variables. To form this variable equity misvaluation as well as issuing activity data is needed. Equity misvaluation is measured at the beginning of the year, for which the intrinsic valuation (V) and market valuation (P) in the last month of the previous fiscal year are used. Equity issue activity is measured during the next fiscal year. To align the different firms in the panel data research fiscal years are set equal to calendar

(19)

16 years. Although Dong et al (2012) set calendar years equal to fiscal years ending between March t-1 through February t based on June as a cutoff adapted with an allowed four-month publication period, this doesn’t help our research because realized leading EPS is taken as an input for ROE calculations instead of analysts’ forecasts. The V/P value in our research is therefore not affected by analysts forecast around publication dates.

In portfolio sorting tests firms are sorted into equity misvaluation quintiles to test the second hypothesis.

4.1 Historical equity finance weighted average V/P

As part of the equity finance weighted average V/P ratio, equity issues are derived in an indirect way by using accounting data from Compustat. In order to calculate equity issues, book value of equity, retained earnings, deferred taxes and total assets are needed. For this the method of Baker and Wrungler (2002) is applied. Because book equity equals balance sheet paid-in capital plus retained earnings, net equity issues (EI) are measured as the change in book equity minus the change in retained earnings plus the change in deferred taxes [delta book equity (CEQ) + delta deferred taxes (TXDB) – delta retained earnings (RE)] divided by total assets (AT) at the end of the previous year. Equity issuance calculation outcomes are trimmed at the 1st and 99th percentile to mitigate the influence of outliers.

As a proxy for misvaluation this research uses V/P ratios instead of book-to-market ratios. The V/P method is borrowed from Dong et al (2012) and is used as a measure of undervaluation. The V/P ratio filters out effects like growth opportunities, risk profile or information asymmetry (Dong et al, 2012) that are inherently present in book-to-market ratios. In addition, the intrinsic value (V) based on the residual income method as showed below, is in contrary to book value of equity, invariant to accounting methods which can influence the book value of equity. Also intrinsic value focuses on the future while book value of equity is a consequence from the past. Because the market value of equity is a result of market expectations towards future earnings potential V/P is a more adequate proxy of equity misvaluation than book-to-market values. V/P ratio measures the ratio of intrinsic value of equity to market value of equity instead of total asset values because timing efforts of financial management depend on the relative “cost” of equity.

Intrinsic value (V) at year t is based on the residual income model of Ohlson (1995) and will be calculated as the book value of equity (balance sheet paid in capital plus retained earnings) at year t plus the discounted value of an infinite sum of expected residual incomes. In line with Dong et al (2012) this research changes the infinite residual income model of Ohlson for a finite residual income model with an explicit forecast period of three years. The explicit forecast period of 3 years is adopted to capture the major value creating part of growth opportunities into fundamental intrinsic value. As stated by Dong et al (2012) Lee,

(20)

17 Myers and Swaminathan (1999) report that the quality of the intrinsic values was not sensitive to the choice of an explicit forecast horizon beyond 3 years. Year 3 discounts the residual income in year 3 as a perpetuity.

The residual income consists out of ROE (t +i) as the return on equity for period t+I of subtracted by Re (t), as the required return on equity and multiplied by the book equity at the beginning of the year t. Negative B (t) observations are deleted. The residual income is discounted with the required return on equity. The last term discounts the residual income of year three as a perpetuity. The forecasted ROE (t+i) are computed as

where EPS is calculated through multiplying realized leading EPS with the amount of shares outstanding. The book value of equity at the beginning of the year is determined as the average of the book value of equity at the beginning of the previous year and the book value of equity at the beginning of the current year. It is required that each of the fROE will be less than 1.

For future book values of equity in the intrinsic value (V) calculation the realized leading book equity values are used.

To determine the required return on equity the CAPM model is used. The annualized cost of equity is determined on an annual basis for the full sample. Because it is assumed that the research sample represents a sample of the whole market the beta for the whole sample period is set at 1. The risk free rate is set equal to the effective yield of German 10 year bonds and the market risk premium is set every year as the average historical annual market risk premium. Historical market risk premium is used as a proxy for required market risk premium. The required market risk premium can fluctuate between different investors. However, it can reasonably be assumed that historical market premium reflects the required market risk premium of the total investor market as a whole. As a proxy for the European average annual market risk premium the implied equity risk premium for US stocks is taken with the S&P 500 as the benchmark as generated by Damodaran.

(21)

18 As stated by Dong et al (2012) residual income valuations have been found too low in previous researches so the benchmark for fair valuation is not equal to 1 for V/P. However, this is acceptable because the research focuses on relative comparison of valuation across firms and time. Lower VP values indicate overvaluation compared to higher V/P values. Further, both V and P values are eliminated at the 1-99th percentile range to mitigate the influence of outliers. Resulting V/P ratios are again eliminated at the 1-99th percentile range.

As a final step an equity finance weighted average V/P variable (EFWAVP) is formed. For this the formula of Baker and Wrungler is used. Instead of the Market-to-book (M/B) values used by Baker and Wrungler we use V/P values. Also, we just use the formula to calculate the weight of equity issues instead of total issuing (equity and debt) as applied by Baker and Wrungler. Lastly, we include the V/P value and the issues in year t in the formula because in year t financial management can also act upon overvaluation which influences financial leverage at the end of the fiscal year t. The applied formula is:







,

=







. 









The formula is based on the hypothesis introduced by Baker and Wrungler (2002) that managers do not rebalance to a pre-defined target level financial leverage ratio and that capital financial structure is the outcome of historical timing efforts of management in raising external finance (whether equity or debt). The EFWAVP variable has higher values for firms that raised external finance when the V/P ratio was high and low values for firms who raised external finance when V/P values were low or bought back stocks when the V/P ratio was high. The formula gives more weight to V/P values in years were much equity was raised and therefore according to Baker and Wrungler “precisely indicate which lags are likely to be the most relevant” for current capital financial structure. As Baker and Wrungler we set the minimum weight in the denominator to 0. In contrast to the method of Baker and Wrungler issues in years where the denominator is set at zero are excluded from the EFWAVP variable. This method allows negative weights for negative issue years (with the minimum restriction that the total cumulative equity issue amount is above 0) to be part of the EFWAVP variable. As a result, the EFWAVP variable can only turn negative in case of a negative V which causes the V/P to turn negative as well as the EFWAVP variable as a consequence. EFWAVP outcomes higher than 10 and lower than -10 are eliminated to mitigate for too extreme outcomes of historical misvaluation.

(22)

19

4.2 Control variables

In the multivariate tests the control factors of size, profitability, tangibility as found on influence of leverage in developed countries by Rajan an Zingales (1995) and current market-to-book, as researched by several studies in the field, are added to the regression. The definitions of the first three factors are derived from the research of Baker and Wrungler (2002). Size is defined as the log of net sales (SALE). Size increases leverage because large firms have more stable cash flows due to, among others, diversification effects and are therefore less likely to enter in financial distress. According to Rajan and Zingales (1995) size can be seen as an inverse proxy for bankruptcy risk. This has a positive effect on the amount of leverage the company uses. Profitability is measured as earnings before interest, taxes and depreciation (EBITDA) divided by lagged assets (AT) expressed in percentage terms. With higher profitability, in line with pecking order theory explanation, the company has more internal funds to fund their operations and investments and as a result has a lower financial leverage (Myers and Majluf, 1984). However, a contrary view based on the free cashflow theory of Jensen (1986) is that profitable firms have more cashflow available which can cause overspending. The potential of overspending will be, in case of an effective market for corporate control needs to be combatted with a pressure to pay out cash to shareholders and therefore levering up. However, managers of profitable firms prefer to avoid the disciplinary role of debt (Jensen, 1986) which will ultimately lead to a negative correlation between profitability and financial leverage. Tangibility of assets is defined as net plant, property, and equipment (PPENT) divided by total lagged assets (AT). Tangibility has a positive effect on financial leverage because they can be used as collateral for external providers of capital (Baker and Wrungler, 2002) therefore diminishing the risk of the lender suffering the agency costs of debt like risk shifting (Rajan and Zingales, 1995). The last control variable applied is current market-to-book value as a control for investment/growth opportunities (Rajan and Zingales, 1995). According to Myers (1977) highly levered companies are more likely to pass up profitable investment opportunities. Therefore, the more growth opportunities a company has the lower the leverage will be to profit from future growth opportunities. Based on the research of Dong et al (2012) current market-to-book is a ratio of equity instead of a ratio of total asset values as sometimes applied. Market-to-book is defined as market value of equity to book value of equity. The market value of equity is calculated as common shares outstanding (CSHOI) times daily close share price (PRCCD) at the fiscal year end (FYRC). The book value of equity is defined according to the definition of Baker and Wrungler (2002) as total assets (AT) minus total liabilities (LT) minus preferred stock (PSTKL) plus deferred taxes (TXDB). In contrast to the definition of Baker and Wrungler convertible debt is omitted because of a lack of availability in Compustat. All control variables are trimmed at the 1st and 99th percentile to mitigate the influence of outliers.

(23)

20

4.3

Capital financial structure variable

For the capital financial structure the book leverage and market leverage are taken as variables. Book leverage is measured as book debt divided by total assets (AT). Book debt is defined as total assets (AT) minus book equity (CEQ) minus deferred taxes (TXDB). Market leverage is defined as book debt divided by total assets (AT) minus book equity (CEQ) plus market equity. Market equity is defined as common shares outstanding (CSHOI) times daily close share price (PRCCD). Both book and market leverage are measured as a percentage. Capital financial structures, both book as market leverage, with a financial leverage higher than 100% are eliminated. Also outliers at the 1st and 99th percentile are eliminated to mitigate the influence of outliers.

4.4

Data descriptive statistics

The original raw accounting data sample consisted out of 51.228 firm year observations in the period 1998-2017. Due to the eliminations on the described minimum/maximum values and the outliers and due to the calculations made in order to derive equity issues (one period lag needed for the Asset Total) and Intrinsic values (one years lag and three period-years leading variables of realized ROE needed to calculate the Terminal value) the sample shrinked to the sample period 1998-2014 with 20.865 firm year observations. Table 1 reports the summarized descriptive statistics for the sample.

From table 1 it can be seen that during the sample period firms on average issued every year 1,87% of their asset total in new equity. It can be seen that the intrinsic values (V) are relatively low compared to market values (P). The main cause is the applied average 7,8% required return on equity. Firms in the sample who don’t have a ROE above this required return on equity in the third leading year, on which the important terminal value is calculated, score most of the time negative intrinsic values. On average the V/P ratio values 0,13x which indicates that the intrinsic value is on average just 13% of market value. The V/P is heavily influenced however by the fact that negative intrinsic values (V) are allowed for in the sample. If negative intrinsic values (V) are not allowed for the sample drops with 5.533 firm year observations and the average V/P value rises to 0,34. These V/P values can either indicate that stocks in this period were structurally overvalued or can just indicate that V values based on the residual income formula of Ohlsen can be found too low on average as claimed by Dong et al (2012). It’s out of the scope of this research to answer this question because the topic of research is to find the relative relation between EFWAVP and financial leverage.

The EFWAVP factor in the p25-p75 percentile range values near 0. This is caused partly because in the applied model of Baker and Wrungler years with negative cumulative equity weights are put on 0%. Another cause is that the V/P ratio values in the whole sample,

(24)

21 and therefore also on these percentiles, have a low score with an average of 0,13x.The market leverage is on average lower than the book leverage. This indicates that sample firms have on average a higher market valuation of equity than their book equity.

Because a panel data research with different years and different firms for various variables is executed a time pattern descriptive table can deliver additional insights into the data used for the regressions. Table 2 shows that on average firms were during all years, except in 2001, net issuers of capital. That firms on average bought back shares in 2001 can possibly be explained by the dotcom bubble which busted in 2000 and resulted especially in 2001 and 2002 in strong declining share prices for stock listed companies as broad category and made repurchasing own shares possibly an attractive investment. The EURO STOXX 50 index declined 20.25% in 2001 and 37.30% in 2002. That on average the sample firms were net issuers in 2002 while the stock indexes were severely hit is puzzling however. It can be seen that the V/P values in 2005 and 2006 were negative indicating that firms in the sample had a negative intrinsic value on average when discounted to the relevant required return on equity. The EFWAVP rises steady from 1998 up to 2003, then show a steady decline from 2003 till 2007 and thereafter a steady rise again from 2007 till 2014. The rise in the period 2007-2014 indicates that the degree of overvaluation diminished. It’s very clear that companies became less and less dependent on tangible assets in the period 1998-2014. This is not very surprising due to the growing importance of technology and information as a source of competitive power. During the sample period book financial leverage has always been higher than market financial leverage which is consistent with an average market-to-book value of equity of 1.84x and all years above 1x. Only during the financial crisis of 2008 the book and market financial leverage where almost equal due to the plummeting of the stock markets (-44.28% of the EURO STOXX 50). Market-to-book values also felt to the lowest level in the sample (1.34x)

(25)

22

Table 1

Summary statistics of Issuance, Overvaluation and Capital financial structure

The table reports the mean values of the equity issuance, intrinsic value, market value, the current degree of overvaluation, the historical weighted degree of overvaluation, the control variables and the book and market financial leverage. Equity Issuance is the determined as net equity issue dividend by lagged total assets. VP is the ratio of Intrinsic value at the start of year t to market value at the end of t-1. The intrinsic value is determined by the residual income model of Ohlsen (1995) where the third leading year is taken to determine the terminal value. Market value is the result of common shares outstanding times daily close share price at the fiscal year end. The EFWAVP factor is determined by using the formula of Baker and Wrungler whereby B/M is replaced with V/P and only equity issues are measured. The EFWAVP ratio is the sum of all the VP ratios in the sample up till the specific year multiplied by the weights of the equity issue/buyback in the periods (t). Size is scaled as the log of net sales, Profitability is measured as earnings before interest, taxes and depreciation (EBITDA) divided by total lagged assets and Tangibility of assets is defined as net plant, property, and equipment divided by total assets. The Current M/B ratio indicates the ratio of market value of equity to book value of equity. The book value of equity is defined according to the definition of Baker and Wrungler (2002) as total assets minus total liabilities minus preferred stock plus deferred taxes.

Descriptive stats N Mean SD p25 p50 p75 Min Max

Equity Issuance 20.865 1,87% 12,09% -1,02% 0,40% 3,57% -46,77% 82,56% Intrinsic Value 20.865 166,54 596,14 -0,84 14,78 94,72 -1.190,54 7.085,32 Market value 20.865 1026,03 2508,36 40,10 147,92 657,32 4,47 21294,16 VP ratio 20.865 0,13 0,87 -0,01 0,15 0,31 -42,46 27,74 EFWAVP ratio 20.865 0,22 1,12 -0,02 0,17 0,53 -9,94 9,54 Size 20.865 5,60 1,87 4,22 5,45 6,91 0,74 10,88 Profitability 20.865 10,63% 7,97% 6,66% 10,61% 15,02% -28,01% 35,15% Tangibility 20.865 23,84% 20,00% 7,51% 19,15% 34,59% 0,49% 88,86% M/B current 20.865 1,84 1,49 0,87 1,41 2,31 0,20 10,74 Book leverage 20.865 52,74% 18,14% 40,15% 54,26% 66,18% 4,49% 92,41% Market leverage 20.865 45,45% 21,12% 28,88% 45,16% 61,55% 3,43% 93,68%

(26)

23

Table 2: Time pattern Issuance, Independent variables, Control variables and Capital financial structure

The table reports the mean values of the equity issuance, overvaluation and the resulting independent variable of EFWAVP. Equity Issuance is the determined as net equity issue dividend by lagged total assets. VP is the ratio of Intrinsic value at the start of year t to market value at the end of t-1. The intrinsic value is determined by the residual income model of Ohlsen (1995) where the third leading year is taken to determine the terminal value. Market value is the result of common shares outstanding times daily close share price at the fiscal year end. The EFWAVP factor is determined by using the formula of Baker and Wrungler whereby B/M is replaced with V/P and only equity issues are measured. The EFWAVP ratio is the sum of all the VP ratios in the sample up till the specific year multiplied by the weights of the equity issue/buyback in the periods (t). Size is scaled as the log of net sales, Profitability is measured as earnings before interest, taxes and depreciation (EBITDA) divided by total lagged assets and Tangibility of assets is defined as net plant, property, and equipment divided by total assets. The Current M/B ratio indicates the ratio of market value of equity to book value of equity. The book value of equity is defined according to the definition of Baker and Wrungler (2002) as total assets minus total liabilities minus preferred stock plus deferred taxes.

(27)

24

5. CROSS SECTIONAL TESTS

5.1 Full sample regression tests

To control for other characteristics of firms who can explain capital financial structure the relevant control variables as described in section 4.2 are added to the regression. Because each firm can exist up to 17 times in the sample time-series autocorrelation on the firm level can exist due to the fact that the independent variable financial leverage is related to the financial leverage of the previous year. Also cross-sectional autocorrelation can exist due to, for example, general economic conditions in Europe which have relative more influence on the financial leverage of some firms than for others. To solve for cross-observation correlation as well as for heteroscedasticity the standard errors are clustered by firm and year. Table 3 reports the results of the panel regressions book financial leverage and market financial leverage.

5.2 Interpretation of the full sample test

The first hypothesis predicts that financial leverage will decrease with overvaluation due to systematic timing efforts of financial management in issuing equity. With the EFWAVP variable, which is rather a measure of undervaluation than of overvaluation, it can be therefore expected that the higher the value of EFWAVP the higher financial leverage will become. The more a firm historically has been undervalued the higher the financial leverage will be in the current capital financial structure. For overvaluation the opposite is true. The lower the EFWAVP becomes, the more equity is issued in historical years of overvaluation which eventually will result in a lower leverage.

In table 3 it can be seen that the coefficient of EFWAVP for book financial leverage is lightly negative with a coefficient of -0.00008 (t=-0.03) and for market financial leverage lightly positive with a coefficient of 0.00003 (t=0.01) These coefficients are close to zero and insignificant. The regression indicates that cumulative historical equity weighted V/P values do not explain capital financial structure. This is in contrast with the findings of Baker and Wrungler (2002) who find a strong effect of their M/Befwa variable on capital financial structure and even found this relation to be persistent for more than 10 years.

The cause of the difference in findings can possible be explained by the fact that the External Finance Weighted Average calculation of Baker and Wrungler includes both equity and debt issues while we include only equity issues in our Equity Finance Weighted Average calculation. However, Baker and Wrungler find that the effect of M/B on changes in leverage mainly comes though net equity issues. Also we use V/P values as a purer measure of

Referenties

GERELATEERDE DOCUMENTEN

Chapter 2 Suppression of chikungunya virus replication and differential innate responses of human peripheral blood mononuclear cells during co- infection with dengue

In particular, major demographic declining changes in rural villages tended to be observed in the extreme scenarios, namely in conditions of desolate areas of low population density

[r]

[r]

Omdat betrokken partijen van tevoren niet altijd zullen weten of de koper een gelieerde partij is, is het aan te raden dat de beoogd curator zo snel mogelijk na zijn aanwijzing een

Although the pension notes mention the weighted average duration of the defined benefit plans or pension liabilities only, this information was not used, because the reported

The correlation tables show a significant positive relation between firm performance and cash compensation for the periods pre- and during the crisis, for the period

As a result of hydrological limitations on Nigerian hydropower dams, effect of gas price on cost of energy produced and diversifying gas technology, harnessing electric energy