Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Dissertation PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 89

Determinants of Capital Structure

An Empirical Study from UK Firms

By

Lujie Chen

2007

A dissertation presented in part consideration for the degree of MA Finance and Investment

ACKNOWLEDGEMENTS

I would like to extend my appreciation and greatest gratitude to my supervisor Mr. Mark Billings for his support and guidance throughout my dissertation period. This dissertation cannot be completed without him.

I thank all lecturers and staffs of Nottingham University Business School for their help and effort.

I also would like to thank my parents for always being so supportive. Their enduring love inspires me throughout my studies in the United Kingdom.

Finally, I thank all friends I have met in Nottingham for giving me such a memorable year. I especially appreciate the support from Xiao Zang, who has always accompanied me throughout my hard time.

ABSTRACT
Capital structure has been one of the most controversial issues in the ground of finance during past years. There are a number of existing theories and empirical studies observing patterns involved in choosing a capital structure, however until now, there is no universal one. With the objective to provide an insight into determinants influence a firms level of debt, we apply ANOVA and multiple regression analysis of secondary data on 80 UK public companies operating in 10 industries. In addition, the study on the impacts of revised IAS 19 on pension deficit makes this dissertation differentiated from existing ones. We divide the time span into pre-pension scheme period 2001-2004 and post-pension scheme 2005-2006 to examine how the adjustment at balance sheet affects firms capital structure. The study demonstrates a disparity between empirical results and theoretical predictions. Overall, the model has a relatively low explanatory power and most of variables themselves show a conflicting sings and levels of significance in terms of long-term and short-term debt ratios. Growth rate, firm size, tax shields and asset compositions are significantly correlated to particular type of debt in different time period. In addition, firms do adjust their leverage ratios facing the disclosure of pension fund as a part of longterm liability. This research provides analysis tools for financial managers when looking to raise capital and assists managers with indications of what the market is anticipating. However, unavoidable research limitations suggest further studies.

## Key words: Capital Structure Leverage, Liability Pension Scheme

TABLE OF CONTENTS
CHAPTER 1: INTRODUCTION......................................................................................8 1.1 Capital Structure.......................................................................................................8 1.2 Research Objectives ................................................................................................9 1.3 The Structure of the Research .................................................................................10 CHAPTER 2: LITERTATURE REVIEW ........................................................................7 2.1 Theories and Empirical Studies Review...................................................................12 2.1.1 M&M Theory without Taxes ...........................................................................12 2.1.2 M&M Theory Correction................................................................................19 2.1.3 M&M Theory with Corporate and Personal Taxes ........................................22 2.1.4 Static Trade-off Theory..................................................................................25 2.1.4.1 Cost of Bankruptcy and Financial Distress..25 2.1.4.2 Financial Distress and Taxes Saving...27 2.1.5 Pecking Order Theory30 2.1.6 Pecking Order Theory VS. Static Trade-off Theory..32 2.2 Overview of Pension Scheme and Accounting Standards..34 2.3 Worldwide Capital Structure Patterns.....................................................................35 CHAPTER 3: METHODOLOGY.....................................................................37 3.1 Dependent Variables...............................................................................................37 . 3.2 Independent Variables and their Expected Signs....................................................39 3.3 The Impact of Pension Fund............................................47 3.4 Data Description....................................................48

.
3.5 Research Model.....49

3.5.1 One-way Analysis of Variance (ANOVA)...................................................50 3.5.2 Multiple Regression Analysis50 CHAPTER 4: RESULTS ANALYSIS...............................................53 4.1 Effects of Industry Classification..........................................................................54 4.2 Cross-sectional Research on Long-term Debt.....................................................57 4.3 Cross-sectional Research on Short-term Debt....................................................58 4.4 Effects of Pension Scheme..................................................................................60 4.4.1 Impacts on Long-term Debt....................................................................62 4.4.2 Impacts on Short-term Debt...................................................................63 4.5 The Summary of Statistical Results.....................................................................64 4.6 Research Limitations.65 CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS..................................68

REFERENCES..........................................................................................................71 APPENDIX ................................................................................................................79 Appendix One: Firms Financial Information Appendix Two: Tables of Coefficients Appendix Three: R2 Summary Appendix Four: Correlations between Independent Variables Appendix Five: Descriptive Figures from ANOVA Test

LIST OF FIGURES

Figure 1: Average Cost of Capital and Debt Ratio.16 Figure 2: Expected Yield on Common Stock and Debt Ratio.......................................17 Figure 3: Summary of M&M Propositions without Taxes..............................................18 Figure 4: Cost of Capital and Debt Ratio.................................................21

Figure 5: Summary of M&M Propositions with Taxes...................................................22 Figure 6: Bond Market Equilibrium...........23 Figure 7: Optimal Debt Ratio..28 Figure 8: Mean of Leverage Ratio by Industry..............................................................56

Figure 9: Standard Deviation of Leverage Ratio by Industry........................................56 Figure 10: Mean of Leverage Ratio by Industry for Post-pension Scheme Period.......61 Figure 11: Std. Deviation of Leverage by Industry for Post-pension Scheme Period.62

LIST OF TABLES
Table 1: Capital Structure of Americas Most Admired Companies................................9 Table 2: Cost of Financial Distress and Bankruptcy......................................................27

Table 3: Worldwide Capital Structure Patterns..............................................................36 Table 4: Industry Classification and Leverage...............................................................46 Table 5: Summary of Variable Indicators and Predicted Signs......................................47 Table 6: Sample of Targeted Industries and Companies ...49

Table 7: Summary of Collinearity Statistics ..52 Table 8: Signs and Significance of Independent Variables53. Table 9: Results from ANOVA54 Table 10: Regression Results of Long-term Debt....57 Table 11: Regression Results of Short-term Debt...59

Table 12: Comparison of Leverage between Pre-pension Scheme Period and Postpension Scheme Period..60 Table 13: Comparison of Long-term Regression Results between pre-pension Scheme Period and Post-pension Scheme Period..63 Table 14: Comparison of Short-term Regression Results between pre-pension Scheme Period and Post-pension Scheme Period..64

Chapter One: Introduction


1.1 Capital Structure
Securities issued by corporations may be classified roughly as equity and debt. The distinction between debt and equity is the basis for modern theory in finance area and practical issues of corporate capital structure. Capital structure refers to the way a corporation finance itself through some combination of equity, debt or hybrid securities. In simple terms, capital is the proportion of firm value financed with debt, the leverage ratio (Emery et al, 2004). In the real complex world, capital structure is not just debt versus equity. There are various forms of debt and equities, such as long-term debt, short-term debt, common shares and preferred shares. When firms are looking to raise capital, they attempt to find the particular combination that maximizes the overall market value of the firm. The aspect of long-term and shortterm debt is becoming especially important under the revised accounting policy- IAS 191. The new policy requires the company to make an estimate of pension obligation at balance sheet and to recognize a defined benefit liability. From 2005, firms started to disclose pension fund as an individual item of long-term liability. Under the emphasis of transparency, it is believed that companies should also adjust their target debt ratio, especially the long-term debt, since the total amount of liabilities appears to increase. Capital structure has been one of the most controversial issues in the theory of finance during past 40 years and now still there is no universal theory of the debt-equity choice, and no reason to expect one (Myers, 2001). The factor involved with choosing a capital structure are complex and the impact of each determinant on the value of firm are not always obvious. However, we are next trying to provide an overview of existing theories on capital structure. The modern theory of capital structure began with the celebrated paper of Modigliani and Miller (1958, hereafter M&M). They argue that in the existence of perfect financial market, capital structure is irrelevant to firms value. Since then, many economists have followed the path they mapped. Some years later, DeAngelo and Masulis (1980), Kim (1986) and Modigliani (1982) further reconcile Millers equilibrium
1

IAS 19 refers to International Accounting Standard 19, Employee Benefits

with the existence of optimal capital structure and they generate a result that the firms optimal capital structure will involve the static trade-off between the tax advantage of debt and various leverage-related costs. In contract, there is another leading theory called pecking-order theory, which is as a result of asymmetric information. The pecking order theory stems from Donaldson (1961) and the key idea is that mangers raise new finance in a particular sequence. The main proponent of this theory more recently has been Myers (1984) and the implication is that there is no optimal capital structure because capital structure is simply the accumulation of past shortages of internal cash flow. It is not appropriate to conclude that one theory is superior to another. Each type of model is particular good for certain explanations and has been argued by conflicting empirical studies.

1.2 Research Objectives


The empirical evidence shows consistent attributes of leverage ratios, as though there are definite reasons for following certain patterns. Some argue that firms are simply following the Behaviour Principal of Finance, which is to copy each other. However, we believe this is too simplistic.
Table One: Capital Structure of Americas Most Admired Companies

Companies Wal-Mart Berkshire Hathaway General Electric Dell Microsoft Johnson &Johnson Starbucks FedEx Southwest Airlines IBM

Debt-to-Equity Ratio, % (Book Value) 46 12 193 8 0 8 0 35 31 50

Debt-to-Equity Ratio, % (Market Value) 9 8 56 1 0 1 0 10 15 10

Source From: Americas Most Admired Companies Fortune (March 7, 2005), and authors calculation

The above table illustrates that these companies take different approaches when looking to raise capital. The study of capital structure attempts to explain the mix of securities

used to finance investment. Most of previous researches focus on testing a single theory or testing the explanatory power of capital structure models on (Buser and Hess 1983, Tizcinka and Kamma 1983, Bradley et al 1984, Titman and Wessels 1988, Ozkan 2001). As an extension of Bennett & Donnelly (1993), our research is aimed to help you to understand why some firms use a great deal of debt, while other firms use very little. By generating a cross-sectional data set; we will look at 80 UK listed non-financial firms operating in 10 industries from 2001-2006. One-way Analysis of Variance (ANOVA) and Multiple Regression will be employed to observe the proposed determinants. Different from previous studies, we will distinguish long-term and short-term debt in order to observe different borrowing behaviour. More importantly, the treatment of pension scheme is a significant concern in our studies. We will in addition test on how this adjustment influences companies borrowing behaviour by dividing the time length into pre-pension scheme period 2001-2004 and post-pension scheme period 2005-2006.

To summary, the research questions are as following: What are the determinants of firms leverage in terms of both short-term and longterm borrowing? How does these determinants influence firms capital structure? How does the treatment of pension fund affect firms borrowing decision?

1.3 The Structure of the Research

This dissertation is organized into six chapters. In the next chapter (chapter 2), we will provide a detailed literature review on capital structure. It contains both studies on existing theories of capital structure and empirical research on firms borrowing behaviour. In addition, the revised IAS 19 and the adjusted treatment of pension deficit at balance sheet will be introduced. Chapter three is the research methodology which discusses the proposed determinants proxies and gearing measurement. Also, data description and the research model will be presented in the later part of this chapter. In chapter four, we interpret our statistical results and some further issues regarding the validity of our research are discussed. Chapter five is the conclusion to a summary of our study. Finally, some implications and practical guidelines will be suggested.

10

Chapter Two: Literature Review


The study of capital structure attempts to explain the mixture of securities and capital sources used by companies to finance investment. Since the ground work by Modigliani and Miller (1958, 1963), numbers of theoretical and empirical studies have provided various predictions and explanation on corporations leverage behaviour.

The Modigliani-Miller results (1958) indicate that mangers cannot change the value of a firm by restructuring the firms securities. They argue that the firms overall cost of capital cannot be affected as debt is substituted for equity, even though debt appears to be cheaper than equity. The reason for this is that as the firm increases its debt level, the equity will become more risky and the cost of equity rises as a result. MM proves that the increase in the cost of equity exactly offsets the higher proportion of the firm financed by low-cost debt. Therefore, both the value of the firm and the firms overall cost of capital are invariant to leverage (Ross, 2005). However, MMs theory is Strictly under the assumption of perfect capital market and real-world mangers do not follow MM by treating debt and equity indifferently. In fact, almost any company has its own target debt-to-equity ratio to adhere. Because of this, many economists (including MM themselves) started to count other factors into their consideration, such as corporate taxes. In 1963, Modigliani and Miller relaxed those restraints, introduced corporate tax into the model, and obtained the revised conclusion. They argue that the increase of debt level can increase the value of the firm. Nevertheless, firm in real world are rarely 100% leveraged, because there is a cost of financial distress. Myers (1984) asserts that a firms optimal debt ratio is usually viewed as determined by a trade-off of the cost and benefits of borrowing, holding the firms assets and investment plans constant. Contrast to the static trade-off theory, there is another view of how financing decision are made. This is Pecking Order Theory, which stems from Donaldsons study (1961) and the key idea of pecking order theory is that mangers raise new finance in a particular sequence. Myers (2001) argued that until now, there is no universal theory of the debt-equity choice and no reason to expect one. Based on these theories, numerous empirical studies observed how theories influence firms financing. In this chapter, we will provide a theoretical literature review, including MM theory, trade-off theory and pecking order theory. Moreover, empirical studies of capital structure will be discussed as the guideline of proposed determinants. Finally, the summary of worldwide capital structure pattern will

11

be represented.

2.1 Theories and Empirical Studies Review


2.1.1 M&M Theory Without Taxes

Modigliani and Miller (1958) is a classic paper in the are of capital structure. They assume the existence of perfect capital market without taxes: (Bailey, 2004) Costless capital markets: no transaction costs or barriers to transactions, and financial assets are divisible. This is the standard assumption required to give the arbitrage principle its predictive force. The assumption is, of course, an idealization but one which is common in financial analysis. Neutral or no personal taxes. Taxes are neutral in the sense that the tax rate is the same across tax-payers, and for all income sources i.e. the tax system is non-discriminatory. Competitive markets: many perfect substitutes for all securities and all market participants are price takers. Investors and firms can borrow and lend on the same terms. This assumption can be counted within that of frictionless markets. It is listed separately here in order to highlight the possibility that the assumption might not hold. The claim can be accepted in the aspect that MM-theorems do not require that every investor can undertake the same financial transactions as firm but only that a sufficient number can do so. In particular, some investors (equity holders) may themselves be other firms (e.g. financial intermediaries). Under this case, it is obviously reasonable to say that terms on which investors can borrow or lend are the same with corporations do so. It is the fact that some investors do replicate the borrowing and lending strategies of firms, which is sometimes referred to as home made leverage. No information and bankruptcy costs.

12

In other word, the earnings of the firm exceed its debt obligations in every possible outcome. Existence of risk classes. This assumption plays a strategic role in the rest of the analysis. A risk class is defined as a set of firms, which has an identical pattern of earnings and payoffs. In these equivalent return classes, the return of equity in any firm is significantly correlated with ones in other firms. The assumption that firms belong to risk class allows the arbitrage principle by enabling the payoffs of one firms equity and bonds to be replicated with another. However, it is now generally accepted that the existence of risk classes is an inessential requirement and modern treatments tend to dispense with the assumption and rely on more abstract ways of applying the arbitrage principle. With the satisfaction of perfect market, M&M state that the value of the firm is unaffected by its choice of capital structure. Value Additivity Principle is the central idea supported their analysis, which says that if cash flow stream is split into a set of component streams then the present value of the original stream must equal the sum of the present values of the component streams. Using the assumption of perfect capital market and the Value Additivity Principle, M&M derived the following two basic proportions and the extensions with respect to the valuation of securities in companies with different capital structures. Propositions I-No Tax

Consider any company and let X stands as the expected return on the assets owned by the company. Market value of the debt and the market value of common share are denoted by D and S respectively. Assume a company which has no debt in its capital structure, with a constant perpetual net cash flow X. (Assume this must all be paid out as dividends). VU=PV(X)=X/kU (1) Where KU is the expected rate of return for shareholders of the unlevered firm

13

Now assume the company is partly financed by perpetual riskless debt, with nominal value D. If the value of the firm and the coupon rate are denoted as VL and I, then we have VL=PV(X-ID) + PV (ID) (2)

VL=PV(X)-PV(ID)+PV(ID)=PV(X)

(3)

Compare equation (3) and equation (1), we have VU=VL, (4)

This is to say; the market value of any firm is independent of its capital tructure and is given by capitalizing its expected rate of return at the k This proposition can be stated in an equivalent way in term of the firms average cost of capital K, which is the ratio of its expected return to the market value of all its securities. This process can be demonstrated as following: We know that VL=X/KL Where KL is the average cost of capital of the levered firm Therefore, KU=KL That is, the average cost of capital to any firm is completely independent of its capital structure and is equal to the capitalization rate of a pure equity stream of its class.

Proposition II- NO Tax


M&Ms second important insight is that even though debt is less costly to issue than equity, issuing debt causes the required return on the remaining equity to rise. Base on the core financial principle that investors expect compensation for risk, shareholders of levered firms demanding higher returns than shareholders in all-equity companies.

14

VL=SL+D=(X-ID)/kL+ID/I

(7)

Where SL is the value of equity in leveraged company and we multiply both sides by KL to get:

KLVL=X +KLD -ID

(8)

We know that X=KU(SL+D) (Remember that VL=VU) Then equation (8) becomes: KLVL= KU(SL+D) +KLD ID And this equation can be reformulated as: KL (VL-D)=KU (SL+D)-ID (10) (9)

According to equation (7), we can obtain that KLSL= KU (SL+D)-ID (11)

Divided by SL, we finally derived: KL=KU+(D/SL)*(KU-I) (12)

In proportion II, M&M derived the above proportion concerning the rate of return on common stock in companies whose capital structure include some debt: the expected rate of return or yield KL, on the stock of any company is the function of leverage. In words, the required rate of return on equity for a geared company is equal to that for an ungeared company, plus a premium for financial risk.

Some extensions of the Proportion I and II:


M&M have extended those two propositions in a number of useful directions: (1) allowing for a corporate profits tax under which interest payment are deductible; (2) recognizing

15

the extension of a multiplicity of bonds and interest rates; and (3) acknowledging the presence of market imperfections which might interfere with the process of arbitrage. ( Modigliani and Miller, 1958) By the same type of proof used for the original version of proportion I, it can be shown that the market value of firm in each risk class must be proportional in equilibrium to their expected return net of taxes. Although the form of propositions is unaffected, certain interpretation must be changed. In particular, the after-tax capitalization rate KL can be no longer identified with the average cost of capital which is KL=X/VL. It has been stated that the difference between KL and the true average cost of capital is a matter of some relevance in connection with investment planning within the firm.
* *

In their extension, proposition I remains unaffected as long as the yield curve is the same for all borrowers, the relation between common stock yields and leverage will no longer be the strictly linear one given by the original. (Figure 1)

Figure One: Average Cost of Capital and Debt Ratio

Source Adopted From: M&M (1958) The cost of Capital, corporation finance and the theory of investment.

16

According to proposition I, if the interest rate (I) increase with leverage, the expected return on common stock (KL) still tend to rise as D/S increase, yet at a decreasing rate. Beyond some high level of leverage, which is shown as LK in figure 2, the yield may even start to fall. The correlation between KL and D/S should go with the curve MD in figure 2. By contrast, with a constant rate of interest, the relation would be consistent with the line of MN.
Figure Two: Expected Yield on Common Stock and Debt Ratio

Source Adopted From: M&M (1958) The cost of Capital, corporation finance and the theory of investment

Also, M&M explain the downward sloping part of the curve MD using some examples as why investors, other than lotteries buyers, would purchase stocks in this range. After all, the yield curves of proposition II are a consequence of the more fundamental proposition I. Should the demand by the risk-lovers prove insufficient to keep the market to the peculiar yield curve MD. This demand would be reinforced by the action of arbitrage operators. The later would find it profitable to own a pro-rata share of the firm as a whole by holding its stock and bonds, the lower yield of the shares being thus offset by the higher return on bonds. (M&M, 1958)

A summary of the main M&M theory results without taxes can be presented as following:

17

Figure Three: Summary of M&M Propositions Without Taxes

Summary of Modigliani-Miller Propositions Without Taxes: Assumption: The existence of perfect market Results: Proposition I: VL=VU Proposition II: KL=KU+(D/SL)*(KU-I) Intuition: Proposition I: The market value of a company is independent of its capital structure. Proposition II: The required rate of return on equity for a geared company is equal to that for an ungeared company, plus a premium for financial risk.

Proposition III No Tax:


On the basis of their propositions with respect to cost of capital and financial structure, M&M derive a simple rule for optimal investment policy. If a firm in class risk k is acting in the best interest of stockholders for any financial decisions, it will create an investment opportunity if and only if the rate of return on the investment (KL##) is as large as or larger than KL. That is, the cut-off point for investment in the firm will in all cases be KL and will be completely unaffected by the type of securities used to finance the investment. To establish this result, M&M consider three major financing alternatives open to the firm. They are bonds, retained earnings, and common stock issues. In each case, an investment is worth undertaking if, and only if KL##KL. Proposition I implies M&Ms irrelevant theory and asserts that that is not worth concerning the issue of the optimal capital structure for a firm. However, this dose not mean that the owners or managers have no reasons for preferring one financial plan to another, or that there are no other policy or technical issues in finance at the level of the firm. This is especially true in the case of common stock financing. In summary, M&M propositions are important in modern finance because by understanding why capital structure has no value impact in perfect market, managers who operate business in imperfect market know how to choose one capital structure over another. They have shown, moreover, how the theory of firm value can lead to an

18

operational definition of the cost and how that concept can be used as a basis for rational investment decision-making within the firm. However, the result is restricted by a number of assumptions and is not suggested for practical guidelines. 2.1.2 M&M Theory Correction Modigliani and Miller (1963) Corporate Income Taxes and the Cost of Capital made a correction on their previous paper regarding the effects of the present method of taxing corporations on the valuations of firms. Their modified work illustrates that even though one firm may have an expected return after taxes twice that of another firm in the same class, it will not be the case that the actual expected return after taxes of the first firm will always be twice that of the second, if the two firms have different degree of leverage. They also assert that within any class, arbitrage will make values a function of expected after-tax returns, tax rate as well as leverage level. More importantly, they state that the tax advantages of debt are somewhat greater than what have been originally suggested. To this extent, the quantitative difference between the valuations implied by their propositions and by the traditional view is narrowed. The introduction of a corporation tax system makes a difference to the traditional proposal that capital structure is irrelevant. On an after tax basis: VU=PV {(1-T)X } VL=PV{(1-T)(X-ID) }+ PV (ID)

(14)

(15)

Note that interest is deducted before tax is calculated, therefore, the equation 11 becomes: VL=PV(1-T)X -PV(ID)+TPV(ID)+PV(ID)=PV(1-T)X+T.PV(ID)

(16)

Based on equation (14) VU=PV(1-T)X and PV(ID)=D, we finally get VL=VU+TD (17)

19

Obviously, the presence of debt increases the value of the firm and under the consideration of corporate taxes, every more debt the firm raises, the value of the firm is increased. In other words, firms physically should use as much debt finance as possible. However, firms are not financed by 100% debt in the real world due to the cost of financial distress.

Similarly, the cost of equity in the geared up firm consequently changes. Start from VL=SL+D, which can be rewritten as: VL=(X-ID)(1-T)/KL + D (18)

Take D from both sides and multiply KL, we get KLSL=(X-ID)(1-T)=X(1-T)-ID(1-T) It can be re-written as: KLSL =KU(VL-TD)-ID(1-T) (20) (19)

Because we know that VL=SL+D Equation (20) can be expressed as :

KLSL =KU(SL+D)-KUTD-ID(1-T)

(21)

Re-arange this equation, we obtain: KLSL=KUSL+KUD(1-T) -ID(1-T) Finally, we have KL=KU+D(1-T)(KU-I)/SL (23) (22)

The weighted average cost of capital under a tax system can also be derived as:

20

WACC=KLSL/VL+I (1-T)D/VL

(24)

The equation can be expressed by the following figure


Figure Four: Cost of Capital and Debt Ratio

# Note that Ki in the figure refers to our KL and the rest indicators are the same with our analysis Source Adapted From: M&M (1963) Corporate Income Taxes and the Cost of Capital: A correction

As we increase the amount of debt in the capital structure the WACC falls and tends to towards at i, which is the extreme gearing level. Moreover, when firms gear up, the cost of equity increases as a result of increasing financial distress. However, the rise is not at such a rate as to outweigh the tax shield on debt. A summary of the main M&M theory results with taxes can be presented as following:

21

Figure Five: Summary of M&M Propositions with Taxes

Summary of Modigliani-Miller Propositions With Taxes: Assumption: Corporations are taxed at the rate of T on earning after interest. No transaction cost. Individual and corporation borrow at the same rate. Results: Proposition I: VL=VU+TD Proposition II: KL=KU+D(1-T)(KU-I)/SL Intuition: Proposition I: Since corporation can deduct interest payments but not dividend payment, corporation leverage lower tax payments. Proposition II: The cost of equity rises with leverage, because the risk to equity rises with leverage

2.1.3 M&M Theory with Corporate and Personal Taxes M&M theories suggest that managers either should not worry about the capital structure decision or should borrow as much as possible to minimize taxes. Miller (1977) offers an explanation for the puzzle. He points out that in a world with differential personal taxes, tax shield can be devalued by the marginal personal tax disadvantage of debt and supply side adjustments. This will drive market price to an equilibrium implying leverage irrelevance to any given firm. Suppose that: T is corporate tax rate TS is marginal personal tax rate for shareholders Td is marginal personal tax rate for debt holders The value of unlevered firm now becomes: VU=PV X(1-T)(1-TS) (18)

And the value of the levered firm is:

VL=PV (X-ID) (1-T) (1-TS) + PVID (1-Td) =VU+D 1-(1-T) (1-TS)/ (1-Td) (19)

22

According to the above equation, only if TS=Td, the formula can be reduced to VU+TD. Leveraged corporation have the advantages of deducting their interest payments to bondholders in computing their corporate income tax. However this advantage of deductibility may not always be the case, as the interest rate on corporate debt is taxed as income for the holder of corporate debt. The interest paid on corporate debt must be high enough so that the after-tax income from holding corporate bonds is more than the income from equity. In this case, the before-tax cost of capital on debt must be higher than cost of equity if investors are to hold debt. This relationship can be presented in the following figure.
Figure Six: Bond Market Equilibrium

Source From: Miller(1977) Debt and Taxes

The horizontal line in this figure represents the supply curve of corporate debt. This line intersects the Y-axis at which the interest rate on corporate debt exactly offsets debts corporate tax advantage. The upward-sloping line represents the demand curve for debt, and indicates that bonds must offer higher rates to attract investors from higher tax brackets. Equilibrium occurs at B#. At this point, the only investors holding corporate bonds are tax-exempt investors and taxable investors facing a personal tax rate on

23

interest income less than or equal to the corporate tax rate. Investors with personal tax rates above the corporate tax rate would choose to hold municipal bonds rather than taxable corporate bonds. B# is the aggregate level of debt in the economy, but for any particular company, there is no net advantage to using debt or equity. (Megginson et al, 2007)

The equilibrium demonstrated in Miller (1977) illustrates that as the supply of debt from all corporations expand, investors with higher tax brackets have to be enticed to hold corporate debt. As a result, investors receive more of their income in the form of interest rather than capital gains. However, actual tax rates do no appear to support this equilibrium. Graham et al (2000) estimates that the tax rate paid by marginal investors in corporate debt is about 30 percent, which is well below the top bracket. He also observes the tax rate on equity income which is at 12 percent. In this case, the extra tax paid by investors cannot offset the corporate interest tax shield. Nevertheless, interest tax shields are still believed to be extremely valuable. Rubinstien (1973) merges the CAPM model with the original M&M framework on capital structure. This was reasonable because the M&M framework and the CAPM model share the same assumption facing the existence of perfect capital markets. Meanwhile, the most important assumption of M&Ms model that firms belonging to the same risk class were replaced by the assumption that firms have the same beta () risk of CAPM. Rubinstiens studies re-construct the basic proposition of the M&M theorems with a number of assumptions. Although Rubinstiens model represents an important progress and it provides the grounded work to prove the basic M&M irrelevance theorem using a mean-variance approach, his work has a number of limitations and is not extremely valuable. The study is constructed only from theoretical perspectives and not been empirically tested. Hsia (1981) generalizes the remaining propositions and this pregerss was also contingent on another independent discovery in the field of option pricing which was provided by Black and Scholes (1973). He applies the fact that the value of equity in a leveraged firm is identical to the value of a call option written on the firm, and that the

24

value of risky debt in a levered firm can be viewed as the value of risk-free debt plus a shock put option on the value of the firm. Both option are European type and have an exercise pricing equal to the future value of the risk-free debt. The maturity time of options equals to the maturity of the risk-free debt.

Rubinstiens and Hsias studies are considered as the most general work so far. Other proofs behind the M&M proportions have considerably complicated from a mathematical point of view. 2.1.4 Static Trade-off Theory Under the M&M theory, capital structure is irrelevant to firms value. Corporate income taxes, viewed in isolation, give firms a strong incentive to use leverage. Under M&Ms model, firms should theoretically borrow as much as they can to maximize tax advantages. However, in the real world we do not se firms financed by 100% debt. Sometimes must be missing from their model and this is the cost of financial distress.

2.1.4.1 Cost of Bankruptcy and Financial Distress:


Financial distress is defined as a condition where obligations are not met or are met with difficulty. A major disadvantage for a firm relaying heavily on debt is that it increases the risk of financial distress, and ultimately liquidation. This may become harmful for both equity and debt holders. 2 The risk of incurring the costs of financial distress has a negative effect on a firms value which offsets the value of tax advantages on borrowing. The most common example of a cost of financial distress is bankruptcy costs. Corporate bankruptcies occur when shareholders exercise their right to default. There are two forms of bankruptcy costs: direct and indirect (Megginson et al, 2007). Direct costs of bankruptcy are out-of-pocket cash expenses directly related to bankruptcy filling and administration. Document printing and filing expenses, as well as professional fees paid to lawyers, accountants, investment bankers, and court personnel are all direct bankruptcy costs. Indirect costs of bankruptcy are expenses that result from bankruptcy but are not cash expenses sent on the process itself. These costs include the diversion

http://cbdd.wsu.edu/kewlcontent/cdoutput/TR505r/page40.htm

25

of managements time, lost sales during and after bankruptcy, constrained capital investment and R&D spending, and the loss of key employees. Although indirect bankruptcy costs are difficult to measure, researchers have shown that they are significant. Many empirical studies indicates that relative to the pre-bankruptcy market value of large firms, direct costs are too small, comparing indirect costs, to provide an effective threat to the use of debt. Warner (1977) is the representative work for his argument. His study involved 11 railroads and is the first step in setting out a methodology for measuring an devaluating bankruptcy-related costs. He cautions that the costs are not small enough to be neglected completely in discussion of capital structure policy. But it would be reasonable to conclude that for firms of the size under consideration, the expected direct costs of bankruptcy are unambiguously lower than the tax saving on debt to be expected at present tax rates in standard valuation models (p.345). Warners work was criticized by Altman (1984) in the aspect that his results are based on a narrowly defined bankruptcy cost definition (lack of indirect bankruptcy costs) and the small sample size could not provide a whole picture. Altman investigated the empirical evidence with respect to both the direct and indirect cost of bankruptcy. Based on regression models, his results show very strong evidence that costs are not trivial. In many cases they exceed 20% of the value of the firm measured just prior to bankruptcy and even measure several years prior in some cases. In addition, the expected bankruptcy costs for many of the bankrupt firms are found to exceed the present value of tax benefits from leverage. This implies that firms were overleveraged and that a potentially important ingredient in the discussion of optimum capital structure is indeed the bankruptcy-cost factor. Finally, a study by Andrade and Kaplan (1998) of a sample of troubled highly leveraged firms estimates that costs of financial distress accounts as 10 to 20 percent of pre-distress market value. To summarize, the cost of financial distress and bankruptcy is:

26

Table Two: Costs of financial distress and bankruptcy

Indirect Examples
Uncertainties in customers minds about dealing with this firm- lost sales, lost profits, lost goodwill. Uncertainties in suppliers minds about dealing with this firm- lost inputs, more expensive trading terms. If assets have to be sold quickly the price may be low. Delays, legal impositions, and targets of financial reorganization may place restrictions on management action, interfering with the efficient running of the business, Management may give excessive emphasis to short-term liquidity, e.g cut R&D and training, reduce trade credit and stock levels. Temptation to sell healthy business as this will raise the most cash. Loss of staff morale, tendency to examine alternative employment. To conserve cash, lower credit terms are offered to customers, which impacts on the marketing effort,

Direct Examples
Lawyers fees

Accountants fees

Court fees Management time

Source adopted from: http://cbdd.wsu.edu/kewlcontent/cdoutput/TR505r/page40.htm

2.1.4.2 Financial Distress and Taxes Saving:


Clearly, the cost of financial distress and bankruptcy significantly influence capital structure decision in complex ways. This allows us to expand the basic equation 17, which was derived under M&M theory, to indicate financial distress factor. Megginson et al (2007) represented the new expression as: VL=VU+ PV (Tax Shield)-PV (Costs of Financial Distress) The argument leads to the optimal capital structure, which refers that mangers think of the firms debt-equity decision as a trade-off between interest tax shields and the costs of financial distress. Myer (1984) asserts that a firms optimal debt ratio is usually viewed as determined by a trade-off of the costs and benefits of borrowing, holding the firms assets and investment plans constant. The firm is supposed to substitute debt for equity,

27

or equity for debt, until the value of the firm is maximized. The following figure illustrates how debt to equity ratio affected firms market value.
Figure Seven: Optimal Debt Ratio

Source from: Stewart C. Myers (1984) The Capital Structure Puzzle

According to the figure, managers of all-equity companies can increases firm value by replacing equity with debt, thus generating more taxes saving. Without cost of financial distress, managers would maximize firm value by maximizing debt, a situation represented by the upper curve. We have to note that it is not in linear because of the cost of adjustment. The lower curve illustrates how the financial distress alters this conclusion. As a firm borrows more, it increases both the probability that it will go bankrupt and its expected bankruptcy costs. Beyond some point, the incremental expected bankruptcy costs offset tax advantages. Optimum is reached when managers find the mix of debt and equity that maximize firm value. If there were no costs of adjusting capital structure, then each firm should always be at its target debt ratio. There are a number of researches based on the existence of an optimal capital structure. Graham and Harvey (2001) surveyed 392 CFOs about the cost of capital, capital budgeting and capital structure. They find moderate support that most firms have target debt ratios and follow the trade-off theory. Marsh (1982)s study provides evidence that companies appear to make their choice of financing instrument as if they have target levels of debt in mind and more importantly, the results are consistent

28

With the notion that these target debt levels are themselves a function of company size, bankruptcy risk and asset composition. In contrast to previous studies that are based on time-series analysis of macro-data, Bradley, Jarrell and Kim (1984) used cross-sectional, firm-specific data to investigate the behaviour of 20-year average firm leverage ratios for 851 firms coving 25 two-digit SIC industries. The strong finding of intra-industry similarities in firm leverage ratios and of persistent inter-industry differences, together with the highly significant inverse relation between firm leverage and earning volatility, tends to support the modern balancing theory of optimal capital structure. In addition, Wald (1999) conducted a cross-country comparison and examined the factors correlated with capital structure. The result stated significant differences between firms. Specifically, differences appears in the correlation between long-term debt/ asset ratio and the firms riskness, profitability, size and growth. The findings of this study suggest different choices in capital structure across countries and legal institutions. this supports trade-off theory in that different corporations operating different target debt ratio. However, there are costs, and therefore delays, in adjusting to the optimum. Firms cannot immediately offset random events that bump them away from their capital structure targets, that is why always see random difference in actual debt ratios among firms which are having the same target debt ratio. Furthermore, due to the existence of market imperfections, firms can adjust only partially to their long term financial targets. According to Jalilvand and Harris (1984), large firms seem to adjust faster to the target level of long-term debt than do small firms. different business environment have

Target debt ratio varies from country to country, industry to industry, and firm to firm. As far as firm specific factors are concerned, the nature of the asset base, the stability of the cash flow, and the quality of management, will all be relevant. Static trade-off theory suggests that companies with safe, tangible assets and plenty of taxable income to shield ought to have high target ratios. Unprofitable companies with risky, intangible assets ought to rely primarily on equity financing (Brealy et al, 2006). However, empirical studies suggest that the trade-off model seems to have a relatively low R2 (Myers, 1984). Actual debt ratio vary widely across similar firms and an odd fact about real-life capital structure is that the most profitable companies commonly borrow the least (Titman and Wessel 1988, Rajan and Zingales 1995, Fama and French 2002 and Wald 1999), where

29

trade-off theory predicts exactly the reverse and fails to explain. Sarkar and Zapatero (2003) conduct an empirical study with a sample of firm in the S&P 500 Index. Their paper shows that mean reversion in the earnings process can reconcile the trade-off theory of capital structure with the empirical evidence. They re-formulate the trade-off theory with mean reverting earnings. Contrary to the traditional trade-off theory but consistent with empirical regularities, their model predicts a negative relationship between earnings and optimal leverage ratio when earnings are mean reverting. It also demonstrates that the speed of earnings reversion plays an important role in the determination of optimal capital structure as well as in the earning-leverage and volatility-leverage relationships. 2.1.5 Pecking Order Theory Contrast to the static trade-off theory, there is another view of how financing decisions are made. This is Pecking Order Theory, which stems from Donaldsons study (1961) of the financing practices of a sample of large corporation. He observes that management strongly favoured internal generation as a source of new funds even to the exclusion of external funds except for occasional unavoidable bulges in the need for funds (pp 67). The key idea of pecking order theory is that managers raise new finance in a particular sequence (Myers, 1984, pp581): Firms prefer internal finance. They adapt their target dividend payout ratios to their investment opportunities, although dividends are sticky and target payout ratios are only gradually adjusted to shifts in the extent of valuable investment opportunities. Sticky dividend policies, plus unpredictable fluctuations in profitability and investment opportunities, mean that internally generated cash flow may be more or less than investment outlays. If it is less, the firm first draws down its cash balance or marketable security portfolio. If external finance is required, firms issue the safest security first. That is, they start with debt, then possibly hybrid securities such as convertible bonds, then perhaps equity as a last resort. In this story, there is no well-defined target debt-equity mix, because there are two kinds of equity, internal and external, one at the top of the pecking order and one at the bottom. Each firms observed debt ratio reflects its cumulative requirement for external finance.

30

Pecking order theory starts with asymmetric information, which indicates that managers know more about their companies prospects, risks and values than do outside investors. Asymmetric information can in practice explain the dominance of debt financing over equity issues. The most of external financing comes from debt, even in the case of highly information efficient market. However, none of this says that firms ought to heavily relay on debt financing. In fact, a firm with ample internally generated funds does not have to sell any kind of security (Brealey et al, 2006) Pecking order theory explains why the most profitable firms generally borrow less- not because they have low target debt ratios but because they do not need outside money. Less profitable firms issue debt because they do not have sufficient internal funds for their capital investment and because debt financing is preferred to equity financing under the pecking order theory. More importantly, this theory demonstrates the inverse relationship between profitability and financial leverage within the industry. Suppose firms generally invest to keep up with the growth of their industries. Then rates of investment will be similar within an industry. Given sticky dividend policy, the least profitable firms will have less internal funds and will end up with borrowing more. The pecking order is considered as a descriptive reasonable empirical model of corporate leverage. When practical figures, the heavy reliance on internal finance and debt is clear. For all non-financial corporations over the decade 1979-1982, internally generated cash covered, on average, 62% of capital expenditures, including investment in inventory and other current assets. Brealy and Myers studies (1984) provide evidence that the bulk of required external financing came from borrowing. Net new stock issues were never more than 6% of external financing. This statistical figures make pecking order theory seem to be reasonable and at least provide a description of typical behaviour. Also, Shyam-Sunder and Myers (1999) find strong support for this model with a sample of 157 firms over the period 1971 to 1989. This is a persuasive and influential result. However, a few months after, their work was cited by Chirinko and Singha(2000), who argue that empirical evidence can evaluate neither the pecking order nor static trade off models. Alternative tests are needed that can identify the determinants of capital structure and can discriminate among competing hypotheses.

31

It has to be concerned that the pecking order theory assumes that the mangers act in the interest of existing shareholders, with the objective to maximize their value. However, this is not always the case, especially in the aspect of agency cost. Agency cost makes numerous scholars review their work and take new issues into consideration. Donaldson (1969) admits that the financing decisions of the firm he studies before were not directed towards maximizing shareholders wealth, and those scholars, such as Berle and Means (1932) and Berle (1954), attempting to explain those decisions have to restart by recognizing the managerial view of corporate finance. Myer and Majluf (1984) demonstrate the possible conflict between interests of managers and shareholders. Problems can particularly arise when a firms manager accumulates too much financial slack that they become immune to market discipline. Myers (1984) critically states that pecking order hypothesizes cannot explain everything. There are plenty of examples of firms issuing stock when they could issue debt. In 2003, Frank and Goyal constructed a study on financing behaviour of publicly traded American firms over the 1971 to 1988 period. According to their empirical results, firms internal financing, on average, is not sufficient to cover investment spending. As a result, most companies in their sample use significant external financing. In addition, it is widely found that debt financing does not dominate equity financing. Net equity issues track the financing deficit quite closely, while net debt does not do so. The current portion of long-term debt is not considered as part of the financing deficit. All these facts are in contrast to what is suggested by pecking order theory. This may be due the inclusion of small firms in their sample, which are hardly following the pecking order.

Furthermore, pecking order theory fails to explain the influence of taxes, financial distress, security issuance costs, agency costs, or the set of investment opportunities available to a firm upon that firms actual capital structure. It cannot explain why financing tactics are developed to avoid the consequences of managers superior information. 2.1.6 Pecking Order Theory VS. Static Trade-off Theory

The financial literature offers two competing models of financial decisions: static trade-off and pecking order theory. In the trade-off model, firms identify their optimal leverage by weighting the costs of financial distress and the tax benefits. At the optimal leverage

32

level, the benefit of the last unit of debt just offsets the cost. In contrast, pecking order theory arises due to the existence of asymmetric information and transaction costs. In this theory, firms raise funds in a particular sequence and follow two rules. Firstly, corporations prefer internal financing than external ones. Secondly, firms always issue the safest securities first.

It seems that one is competing the other one and they seem both reasonable to some extent. Scholars always try to run a race between them in order to find the circumstances in which one is superior to another (Myer and Majuf 1984, Fama and French 2002). They find that pecking order works best for large, mature companies that have access to public bond markets. These firms rarely raise equity and they prefer internal financing to external ones, yet turn to debt markets rather than equity when external financing is needed. This is not consistent with smaller, younger, growth firms, which are more likely to rely on equity instead of debt, Here, the pecking order theory stumbles (Shyam-Sunder and Myers 1999, Lemmon and Zender 2002, Frank and Goyal 2003). The trade-off theory still retains some explanatory power once pecking order motives are accounted for. The theory is particularly helpful in explaining inter-industry differences and works best for those companies with tangible assets. It suggests that the debt ratio is lower in volatile industries where value depends on intangible assets and growth opportunities. Nevertheless, it has to be noted that many scholars have proved that there is in fact no conflict between these two theories. Rajan and Zingales (1995) constructed a study of debt versus equity choice by large firms and they find that large firms tend to have higher debt ratios; firms with high propositions of fixed assets to total assets have higher debt ratios; more profitable firms have lower debt ratios and firms with higher ratios of market-to-book value have lower debt levels. These results interpret evidences for both the trade-off and pecking order theories. Fama and Frech (2002) state that though motivated by different forces, the two models share many predictions about dividends and leverage. Confirmed predictions shared by the trade-off and pecking order models are that more profitable firms and firms with fewer investments have higher dividend payouts. In a survey of US Chief Financial Officers, Graham and Harvey (2001) also find evidences consistent with both theories, although there was little evidence that asymmetric information was an important factor in financial decisions.

33

To sum up, it is always too simplistic to say one is superior to another. Either one is good at explaining certain issues and has obtained a number of empirical supports. More appropriately speaking, pecking order theory is offered as a complement to rather than a substitution for, the traditional trade-off model.

2.2 Overview of Pension Scheme and Accounting Standards


With the objective to examine how the adjusted accounting standards on pension scheme affect capital structure, we will next provide an introduction on how pension scheme works and what is the adjustment under new accounting standards. Pension is defined as post-retirement benefits and it is companys obligation to pay pensions to former employees. The purpose of a pension is to grant people some money when they are retired. Typically pension schemes are funded by the company putting money into a separate fund, and the fund pays out the pension. There are two major types of pension schemes, namely defined benefit and defined contributions. Under a defined contribution plan, the company pays fixed contribution to the fund and has no obligation to further payment if the play assets are not sufficient to pay the pension benefits. Defined benefit plans guarantee to provide a pension to the retired employee which is a proportion of their final salary or some average of their latter years working at the company.

Pension scheme was originally ruled based on SSAP 24 (Statements of Standard Accounting Practice), which has been criticized regarding a number of aspects. Firstly, SSAP 24 misleads the balance sheet position and does not require companies to report the assets and liabilities on the balance sheet accurately. Secondly, SSAP 24 requires the use of actuarial valuation basis for both assets and liabilities, which is difficult to justify and could give rise to unrealistic pension provision being made. After a relatively long discussion, in 2000, SSAP 24 was replaced by the new accounting standard IAS 19 (International Accounting Standard) which redefines how all employee benefits were to be accounted for.3 Different from SSAP 24, IAS 19, Employee Benefits, follow a balance sheet approach to accounting for the pension scheme contributed by the employer.
3

This is the same with FRS 17 (Financial Reporting Standard).

34

Defined contribution plans are relatively easy to account for: contributions payable to the plan are recognized as an expense as the employee provides services. For defined benefit plans, IAS 19 requires the company to make an estimate of the pension obligation at the balance sheet date ( the present value of its defined benefit obligations for current and past service of employees) and to recognize a defined benefit liabilities (a kind of provision), net of the fair value of the funded plan assets. The new accounting standard also defines how the balance sheet asset or liability should be built up. However, IAS 19 has been subject to some heavy criticism. It has been argued that a deficit should not be recognized as a business liability because the final amount that companies need to pay may be no actually estimated.

Based the revised IAS 19, companies adjusted their balance sheet format and disclosed pension as an individual item of liabilities from 2005. This adjustment has brought a significant impact on many companies. As companies begin to adopt the new standard, large pension deficits on the balance sheet appear to be an early sign of trouble. For example, IAS 19 will make company accounts appear more volatile, and pension deficits are likely to influence the rating of individual shares, in some cases positively, in other cases negatively. Some leading companies also have been affected by the new accounting standard. For example, British Telecom has warned that its pension fund will show a 5 billion pound deficit on their balance sheet. Actually, when figures are calculated they are likely to produce bigger deficits than it would under the old system. This will directly push down the companys share price, and could start to impact on companys profit.

2.3 Worldwide Capital Structure Patterns

Panno (2003) investigates the empirical determinants of capital structure choice by analyzing security issues made by companies in the UK and Italy between 1992 and 1996. He examines how companies actually choose between financing instruments at a given point in time and in different financial contexts. The results provide evidence of interesting differences between the two financial markets, generally supporting the idea that the UK market is more testable and in principle more consistent with capital structure theories. On the whole, the results support positive effects of size and negative

35

impact of liquidity conditions and bankruptcy risk on the financial leverage of companies. This, together with the negative correlation between leverage and available reserves, which are taken as an indicator of internally generated funds, tends to support the pecking order theory of capital structure. It is also suggested that firms in well-developed financial systems (UK) may have long-term target leverage ratios, while in less efficient markets (Italy) an optimal debt level does not seem to be a major concern. Finally, for both markets, the results are in line with the notion that the tax advantage of debt financing plays a relevant role in capital structure decisions. We can conclude that although with different constitutional environments, capital structure patterns seem not vary too much between countries. Megginson et al (2007) implemented a survey on companies capital structure around the world. Research and observation established a set of key facts that a capital structure theory should explain. In general, capital structure research documents the following patterns.
Table Three: Worldwide Capital Structure Patterns

I. Firms in the same industry often have similar capital structures regardless of their home
country. II. Capital structures vary across countries III. Leverage ratios vary inversely with financial distress costs IV. Corporate and personal taxes influence capital structures, but taxes alone cannot explain differences in leverage across firms,, industries, or countries V. Markets interpret leverage-increasing events as good news and leverage-creasing events as bad news VI. Corporations strive to maintain target capital structures VII. There is some evidence that, within industries, leverage varies inversely with profitability Source From: Megginson et al (2007) Corporate Finance

After looked at the literature review, we have provided an integrated picture on existing capital structure theories and empirical studies. With the objective to fill the gap of researchers that have not yet been conducted, the pattern of long-term and short-term debt ratios across industries for the period before and after the adjustment of accounting policy should be discussed in depth. In the following part, I will propose a study on the topic of capital structure in UK firms, answering the questions like What are the determinants of capital structure in UK firms? How these determinants affect firms long-term and short-term leverage? and How does the change in treatment of pension scheme affect firms capital structure?

36

Chapter Three: Methodology


Having reviewed a number of literatures in term of capital structure, we will now conduct an empirical study on UK listed non-financial companies, answering the following research questions:

What are the determinants of firms capital structure? How are these determinants influence companies borrowing decision? How dose the change of IAS 19 on pensions affect firms capital structure?

In this study, quantitative rather then qualitative method will be used. Analysis of variance (ANOVA) and multiple regression analysis will be applied to examine the significances and correlation between different variables. To run the research models, firstly and most importantly, dependent variables and independent variables need to be selected.

3.1 Dependent Variables:


With the aim to assess companies gearing level, how much a firm borrows is obviously measured as dependent variables. When companies raise fund, they will consider different forms of debt. Here we distinguish between short-term and long-term debt because they play different roles in financing decisions and believed to obtain different empirical results. Long-term corporate debt is a promise by the borrowing firm to repay the principal amount by a certain date, called the maturity date. Long-term debt always has a par value equal to the face value, and debt price is often expressed as a percentage of the per value. The borrower using long-term debt generally pays interest at a rate expressed as a fraction of par value (Ross, et al, 2005). According to World Bank Research, firms in industrial countries use far more long-term financing than short-term ones and the analysis concludes that long-term debt tends to be associated with productivity 4 . An active stock market and an ability to enter into long-term contracts also allow firms to
4

http://wbro.oxfordjournals.org/cgi/content/abstract/13/2/171

37

grow at faster rates than they could attain by relying on internal sources of funds or short-term credit alone.

However, for many companies, two primary sources of liquidity, which provide needed funds, are short-term debt and commercial paper programmes. Jalilvand and Harris (1984) argue that expectations of lower long-term interest rates in the future seem to postpone the issuance of long-term debt and increase the use of short-term debt. This is especially the case for companies in seasonal businesses in which large amounts of operating capital are needed for only a few months of the year. Short-term borrowing provides maximum flexibility at a minimum cost. Diamond and Rajan (2000) assert that companies do need short-term financing. Institutions like banks that want to enhance their ability to provide liquidity and credit to borrowers have to issue short-term debt. Similarly, countries that have poor disclosure rules and inadequate investor protections, will find borrowing becoming increasingly relying on short-term as they have limited longterm debt capacity.

Consequently, we compute both short-term and long-term debt as the measurement of leverage, which are demonstrated as dependent variables. Although these variables could have been combined to extract a common debt ratio attribute, there are reasons for not doing this. Some of the theories of capital structure have different implications for the different types of debt and hence the predicted coefficients in the structure model differ between long-term and short-term leverage. In terms of indicators, we apply the debt/total assets ratio rather than the actual amount of debt. This can protect from inaccuracy caused by different size of companies.

When observing the debt level of firms, it is important to distinguish between market values and book values (Ross et al, 2005). It is a matter to choose between market value and book value of debt. In general, financial economists prefer the use of market values. MMs theory emphasized market value instead of book value and they stated that the level of gearing is independent to the market value of the company. Also the bankruptcy cost asserts that capital structure choice matters only to the extent that it affects the market value of the firm. Therefore, it is suggested by many scholars that market-based values of leverage should be applied where possible for empirical studies. (Beaver, Kettle and Scholes 1970, Rosenberg and McKibben 1973, Thompson 1976) However,

38

we use book value5 in our studies due to the restraint on data collection. This may lead to certain limitations, yet we cannot completely take away its validity. Bowman (1980) demonstrates that the cross-sectional correlation between the book value and market value of debt is very large, so the misspecification due to using book value measure is probably fairly small. Corporate treasurers also suggest that the use of book value is now becoming more popular because of the volatility of the stock market. The inherent volatility of the stock market makes market-based debt ratios change under a frequent basis. It is also the fact that restriction of debt in bond covenants are usually expressed in book values rather than market values. As a result, we use book value for both long-term and short-term debt ratio. Proxies for these two dependent variables are expressed as following: Long-term debt ratio=long-term debt/ total assets Short-term debt ratio=short-term debt/ total assets

3.2 Independent Variables and their Expected Signs:


In recent years, a number of theories have been proposed to explain the variation in debt ratios across firms. The theories suggest that firms select capital structures depending on attributes that determine the various costs and benefit associated with debt and equity financing. In this section, we present a discussion of the attributes that different theories and empirical studies suggest may affect the firms debt-equity choice. According to the worldwide capital structure patterns, leverage is affected by the industry classification, corporate and personal taxes and profitability. Marsh (1982) demonstrated that companies target debt levels are themselves a function of company size, bankruptcy risk and asset composition. Harris and Raviv (1990) summarized a number of empirical studies from US firms and they suggested that leverage increases with tangible assets, non-debt tax shields, growth rate firm size and decreases with profitability. Here, we extend Harris and Ravivs studies by adding some more
5

In accounting, book value is the value of an asset or liability according to its balance sheet. Book value is the value carried on the bookkeeping records of an economy entity, such as individual, corporate, government, or other organization.

39

determinants and apply into UK companies. Proposed determinants in our studies are denoted by tax shield on interests, tangible assets, profitability, growth rate, dividend payout rate, firm size and industry classification. Also, adjusted accounting policy will be discussed as an influential factor of companies capital structure. Although previous researchers have also concerned some other factors, such as cost of financial distress, it is hard to define an indicator for quantitative measurement. This issue will be discussed in more details in the next section as model limitation.

Interest tax shields:


The tax shield refers to the tax deduction for interests paid and investment tax credits. This is because interests is deducted before corporate tax, which means the more interests companies paid, the less tax amount will be deducted from profit. Consequently, tax shield can be indicated as the interests divided by profit before interests and taxes. In market value terms the balance sheet of a firm has the following form: Tax Benefit to Debt TCD Unlevered Assets Total Assets UA TA Equity Debt Total Assets E D TA

According to above table, every debt the company raises, the value is increased. Theoretically, MM suggested firms should borrow as much as they can to maximize tax benefit. Tax shield is a crucial issue in trade off theory. Under the trade off theory, financial managers of a firm often think of the debt-equity decision as a trade-off between interest tax shield and the cost of financial distress. DeAngelo and Masulis (1980) demonstrate a model of optimal capital structure that incorporates the impact of corporate taxes, personal taxes, and non-debt-related corporate tax shields. They argue that tax deductions for depreciation and investment tax credits are substitutes for the tax benefits of debt financing. As a result, firms with large tax shields relative to their expected cash flow include more debt in their capital structure. Therefore, we make the hypothesis that there is a positive correlation between leverage and tax shield. Tax Shield=Interest Paid / Profit before Interest

40

Profitability
Myers (1984) cites evidence from Donaldson (1961) that suggests that firms raise capital in a preferred sequence. He states that the past profitability of a firm, and hence the amount of earnings available to be retained, should be an important determinant of its current capital structure.

Although many theories and empirical studies have considered profitability as one of the most significant factors in capital structure, there is no consistent conclusion on the relationship between profitability and leverage. Tax based models suggest that profitable firms should borrow more in order to maximize their tax shield benefits. In contrast, the preference of internal fund to external ones suggest that firms which have been profitable in the past will have high-retained earnings and low borrowings. Jensen (1986) and Willamson (1988) define debt as a discipline device to ensure that managers pay out profits rather than build empires. For firms with sufficient free cash flow or high profitability, high debt can restrain management discretion. Moreover, worldwide capital structure patterns and empirical studies from Bennett & Donnelly (1993), Ozkan (2001), Kester (1986), Baskin (1989), Griner & Gordon (1995); Shyam-Sunder & Myers (1999) supporte the inverse relationship between profit and leverage. Thus, we predict a negative coefficient of profitability in our regression model. The proxy of profitability in our study is profit margin. Profit margin is an indicator of companys pricing policies and its ability to earn profit. Differences in competitive strategy and product mix cause profit margin to vary among different companies. Profitability= Net Income/Net Sales Revenue

Dividend rate:
According to M&M irrelevant theory, the dividend policy is irrelevant in a perfect market because the shareholders can effectively undo the firms dividend strategy. If a shareholder receives a greater dividend than desired, he or she can reinvest the excess. Conversely, if the shareholder receives a smaller dividend than desired, he or she can

41

sell off extra shares or stock. In this case, how much the firm pays out as dividend is irrelevant to its market value. This argument is similar to their irrelevant-leverage concept. However, corporations in the real world view the dividend decision as quite important, because it determines what funds flow to investors and what funds are retained by the firm for reinvestment.

According to pecking order theory, the highest preference for firms is to use internal financing before resorting to any form of external funds. It argues that firms have lower dividend rate and significant internal reinvestment turn to have lower leverage. This is because that a lower dividend payout rate indicated that firms spent a significant amount of profit in reinvestment and there are less needs to borrow from outside.

Therefore, we predicate a positive relationship between debt level and dividend rate. To indicate the proportion of profit that has been paid out as dividend, we use the following measurement: Dividend rate= Dividend Paid / Profit after

Firm size:
According to Storey, Keasey, Watson and Wynarczyk (1987) and Chittenden, Hall and Hutchinson (1996), the size of firm has a significant effect on capital structure. The relationship between firm size and leverage level has been fairly conflicting in both theoretical and empirical studies. In theory, larger companies with less asymmetric information problems and better access to capital market tend to have more equity than debt and thus have lower leverage. However, larger companies also tend to be more diversified, which means that they are less risky and have a lower bankruptcy cost. Due to the reverse relationship between risk and leverage6, lower probability of bankruptcy could result a higher debt level.

Early study Gupta (1969) asserts the debt ratio was negatively related to size of the corporation. He attributes this to the very high cost of outside equity funds for smaller
6

Bradley, Jarrell and Kim (1984), DeAngelo(1981) and Jaffe and Westerfield(1984) have supported the reverse relationship between volatility and borrowing.

42

corporations and the various psychological factors associated with their management which result in their being reluctant to take in new equity. Smith (1977) also provides evidence to support the negative correlation between firms size and leverage. He states that small firms pay much more than large firms to issue new equity. This suggests that small firms may be more leveraged than large firms and may prefer to borrow short-term rather than issue long-term debt because of the lower fixed costs associated with this alternative. In contracted, Warner (1977) and An, Chua, and McConnell (1982) argue that large firms should rely on high leveraged. They provide evidences that direct bankruptcy costs appear to contribute a larger proportion of a firms value as that value decreases. Marsh (1982) finds evidence that firms with greater bankruptcy risk are more likely to issue equity. Moreover, results from Jalilvand and Harris (1984)s studies indicate that larger firms tend to use more long-term debt in responding to their remaining financing need than do small firms. A number of researchers use natural logarithm of revenues to indicate the firm size7, however, we here apply the number of employees instead of revenues. This is because we have already included the profitability as one of our variables; the use of revenue here may lead to high correlation between two independent variables. As what has been discussed above, the coefficient is predicted as positive: Firm size= Log (Number of Employees)

Assets composition:
Trade off theory did explain how companies actually behave when making financial decisions, in that companies with most tangible and safe assets did borrow the less. Airlines can and do borrow heavily because their assets are tangible and relatively safe. However, Myers (1984) suggests that static trade-off theory seems to have an relatively low R2 which he considers as unacceptable and in the real world, there is little evidence that executives are concerned about assets substitution when deciding the target leverage ratio.

In our studies, we hypothesize a positive relation between tangible assets and debt level. To indicate this variable, we use:
7

For example: Titman and Wessels (1988)

43

Assets composition = Tangible assets/ Total assets

Growth rate
Growth opportunities are capital assets that add value to a firm but cannot be collateralized and do not generate current taxable income. It has been confirmed by many studies that growth rate is significantly correlated to the level of debt.

Both trade-off and agency cost theory suggest a negative correlation between growth and debt level. According to trade-off theory, fast growing companies borrow less because of increase expected cost of bankruptcy. Specifically, firms holding future growth opportunities, which are in the form of intangible assets, tend to borrow less than firms holding more tangible assets because growth opportunities cannot be collateralized. Take the agency cost into consideration; equity-controlled firms rarely invest optimally as they have tendency to expropriate wealth from the firms bondholders. The cost associated with this agency relationship is likely to be higher for firms in growing industries, which have more flexibility in their choice of future investment. Expected future growth should thus be negatively related to long-term debt levels. However, Myer (1977) notes this agency cost could be mitigated if the firms issue shortterm debt rather than long-term debt. This suggests that short term debt ratios might actually be positively related to growth rates if the short term debt can substitute the long term debt. This issue is also supported by the pecking order theory, which states that fast growing companies are likely to use more debt. This is because growth may be considered as an alternative quality signal, which would predict less need for leverage.

After all, the majority of empirical studies support the negative relationship between growth opportunities and leverage, applying various methodologies. Kim & Sorensen (1986), Smith & Watts (1992), Wald (1999), Rajan & Zingales (1995) and Booth et al. (2001) provide consistent evidence for the negative relationship. Wald (1999) uses a five-year average sales growth. Titman and Wessels (1988) measured capital investment scaled by total assets as well as research and development scaled by sales

44

to proxy growth opportunities. Rajan and Zingales (1995) use Tobins Q and Booth et al. (2001) use market-to-book ratio of equity to measure growth opportunities. Similar with Walds method, we will here apply the five years average turnover growth, which is already calculated by the firm and shown on their financial statement. As what have been argued by previous scholars, we foresee contradicted correlation of long-term and short-term debt variables, negative coefficients with long-term debt and positive one with short-term debt respectively. Growth Rate t=(Turnover t Turnover t-1) / Turnover t-1

Industry:
Both pecking order and trade off theory assume that firms leverage is affected by the industry they operate in. Bradley et al (1983) find that the industry factor has strong influence on firm leverage ratios. There are very significant inter-industry differences in debt ratios that persist over time. Trade off theory is particular helpful in explaining differences in capital structures across industries. It has been empirically supported by many studies that debt ratios tend to be very low in high growth industries with ample future investment opportunities such as the drugs and electronics industries. This is the case even when the need for external financing is great. Industries such as primary metals and papers, with relatively few investment opportunities and slow growth, tend to use the most debt. Titman (1984) suggests that firms that make products requiring the availability of specialized servicing and spare parts will find liquidation especially costly. This indicates that firms manufacturing machines and equipment should be financed with relatively less debt. Pecking order theory also supports the intra-industry similarities. DeAngelo-Masulis (1980) and Masulis (1983) argue that these similarities in leverage ratios were caused by tax code and tax rate.

To summary, the relationship between industry classification and firm leverage can be predicted as following:

45

Table Four: Industry Classification and Leverage

Level of Debt Higher Leverage

Industry Characteristics Safe and tangible assets firms (e.g airline) Slow growth industries (e.g metals and paper) Industries with low specialization

Lower Leverage

Risky and intangible assets firms High growth industries (e.g drugs and electronics industries) Industries requiring specialized servicing (e.g equipment manufacturing)

In our studies, firms operating in ten industries according to Financial Times London Share Service will be measured. Industry will be represented as dummy variables in the research model.

Overall, the indicator of both dependent and independent variables as well as predicted signs of their coefficients can be summarised as below:

46

Table Five: Summary of Variable Indicators and Predicted Signs

Variables
Dependent Variables Long-term Debt (LTD) Short-term Debt (STD) Independent Variables Interest Tax Shield (TaxS)

Indicators
Long term debt/Total Assets Long term debt/Total Assets

Predicted Signs

Interest Paid / Profit before Interest

Profitability (Prof)

Net Income/Net Sales Revenue

Dividend Rate(Div) Firm Size(Siz) Assets compositions (Ass) Growth Rate (Grow)

Dividend Paid / Profit after Log(Number of Employees) Tangible assets/ Total assets (Turnover
t

+ + + / LD: SD: +

Turnover

t-1)

Turnover t-1 Industry (Ind) 10 Dummy Variables

3.3 The Impact of Pension Fund:

Revised IAS 19 requires the company to make an estimate of the pension obligation and disclosed it on the balance sheet. Since 2005, most companies have started to treat pension fund as an individual item of long-term liability in balance sheet. With the aim to increase transparency by showing the pension fund surplus/deficit clearly on the balance sheet, and its cost in the P&L account, it is believed that companies should also adjust their targeted capital structure since the amount of liabilities appear to increase.

To capture how the new accounting standard has influenced companies capital structure, especially the level of debt, we divided the entire time period into pre-pension

47

scheme period 2001-2004 and post-pension scheme period 2005-2006. By comparing results obtained from both periods, we can detect the impact from different treatment of pension fund.

3.4 Data Description:


Our data come from Financial Analysis Made Easy (FAME)8, which provide a financial and accounting database covering the majority of UK public and private companies. Cross-sectional data set are collected from annual report of 80 non-financial UK listed companies in 10 industries. We excluded the financial firms because the balance sheet of the firms operate in the financial sector (banks, insurance companies and investment trusts) have an obviously different structure.

The time period under consideration is a time span of six years from2001 to 2006. In order to capture the impact of adjusted accounting standard, we divide the entire timeperiod into the pre-pension scheme period: - 2001-2004 and the post-pension scheme period: - 2005-2006. Similar with Bradley (1983), we use the average value of the testing period to avoid time bias.

While other studies employ the Standard Industrial Classification (SIC) codes for industry analysis, which provide an exogenous means for grouping firms into functionally defined industries. 9 We classify UK sample firms into 10 industries using sector information provided by the Financial Times London Share Service. Industries and firms observed are:

http://www.fame.bvdep.com/athens

For example, the test of the existence and determinants of inter-industry differences in leverage studied by Bowen et al. (1982)

48

Table Six: Sample of Targeted Industries and Companies Industries Automobiles Chemicals Number of firms 2 Sample of firms GKN, Torotrak Carclo, Croda, ICI JOHNSON MATTHEY, Porvair, Treatt, Victrex, Zotefoam, Elementis SMITH & NEPHEW Abacus, Chloride, Dialight Halma, Inensys, Laird, XP Power, Xaar, Volex Spectris, Renishaw Unilever, Tate&Lyl SABMiller, PGI, Kerry Group IAW Group , Glanbia, Diageo Devro, Dairy Crest Abbot Group, Aminex, BG Group, BP, Burren Energy XP Power, Hunting, JKX, Tullow, Dana Petroleum Acambis, Vernalis Astrazenca, Vectura Group BTG, Dechra Brixton, Liberty International Segro, CLS Holdings Daejan, Freeport Grainger, Savills, Minerva BT Group, KCOM Thus, Vanco Vodafone, Carphone Warehouse, Universe, Emap Arriva, British Airways Carnival, EasyJet, Ryanair Labrokes, Marston's, Luminar Alexon, Findel, Lookers, HMV, Beale, Greggs

10

Electronic Equipment 11 Food and beverage 10 Oil and Gas 10 Pharmaceuticals and Biotech 6 Real Estate 9 Telecommunication 8 Travel and leisure 8 Retailers

The sample of industries and firms are randomly chosen from Financial Times, and limited resources force us to skip those firms whose annual report from 2001-2006 are not available from FAME. Importantly, we have to make sure that all sample firms make clear estimates about the pension deficit of long-term liability from 2005 in their balance sheets. This will be discussed with more details in the limitation part.

3.5 Research Model:


Due to the different nature of independent variables, we apply different model for dummy variables and non-dummy variables. To regress dummy variables, which refer to the

49

industry classification in this case, we use one-way analysis of variance (ANOVA), examining how industry classification has affect companies leverage. To regress nondummy variables, which are the rest proposed determinants of capital structure, we apply multi-regression analysis.

3.5.1 One-way Analysis of Variance: (ANOVA)

We use analysis of variance because it compares the variance among the different groups10 (believed to be due to the independent variable) with the variability within each of the groups 11 (believed to be due to chance). To evaluate the impact of industrial classification on companies leverage, we applying the following model:

Yi=+ i Indi + e
Where, Yi refers to the two dependent variables (LTD and STD), which are introduced in the early part of the chapter. is the constant number which represents the interception for the regression line. Indi refers to the 10 dummy variables representing industry classification. i is the estimate of the coefficient number for explanatory variables and e is the error term. We have to be aware that basic assumptions under ANOVA have been made. The population of scores for the dependent variables should be normally distributed, and it is also a test for which groups are equally drawn. In addition, the categories of the independent variable are assumed to be fixed. Finally, error term has to be uncorrelated. (Malhotra, 2004)

3.5.2 Multiple Regression Analysis:

To examine how factors affect companies level of debt. We use the following two multiple regression models:

10 11

Refer to different level of debt in this study. Refer to different industries in this study.

50

Yi=+1Sizi+2Profi+3TaxSi+4Divi+5Groi+6Assi +ei Yi =+1Sizi+2Profi+3TaxSi+4Divi+5Groi+6Assi +ei


Endogenous variables Yi and Yi * represent long-term and short-term debt respectively. Exogenous variables are these proposed influential factors discussed in early part. They are referred to firm size (Siz), profitability (Prof), tax shield (TaxS), dividend rate (Div), growth rate (Grow) and assets compositions (Ass). The reason of choosing such model is because regression analysis is able to examine associative relationship between a metric dependent variables and one or more independent variables by identifying relative importance of independent variables (e.g what is the relative impact of firm size on long-term debt), predicting the values of the dependent variable (e.g how long-term debt would increase if the dividend rate has increased) and determining the structure or form of the relationship (e.g the mathematical equation relating the dependent and independent variables).

To apply the regression analysis, we assume that there are linear relationship between leverage and those six factors, the error term is constant, independent and under a normal distribution.

Diagnose the problems of Multicollinearity:

Classical linear regression model is that there is no multicollinearity among the regressors. The problem of multicollinearity occurs when there is a strong relationship between two or more explanatory variables. The sign is a low tolerance (less than 0.75) or high VIF, which could result inaccurate and artificially high R2.

Following tables represent the value of VIF/Tolerance coefficient of independent variables. We can see that none of the tolerance is lower than 0.75 and thus it appears that there is no problem of Multicollinearity

51

Table Seven: Summary of Collinearity Statistics

Long Term Debt


2001-2004 Tolerance TaxS Prof Ass Div Gro Siz .978 .956 .893 .902 .874 .849 2005-2006 VIF 1.023 1.047 1.120 1.108 1.144 1.178 2001-2004 Tolerance .915 .970 .863 .983 .850 .950 2005-2006 VIF 1.092 1.031 1.159 1.017 1.177 1.053 2001-2004 Tolerance .978 .956 .874 .893 .902 .849

Short-term Debt
2005-2006 VIF 1.023 1.047 1.144 1.120 1.108 1.178 2001-2004 Tolerance .970 .863 .850 .950 .915 .983 2005-2006 VIF 1.031 1.159 1.177 1.053 1.092 1.017

Data are summarized according to appendix two

To sum up, the statistical analysis of secondary data is applied in our study, observing the capital structure patterns of 80 UK listed firms operating in 10 industries. The research models used are one-way ANOVA and multiple regression analysis and results obtained from these models will be interpreted in the following part.

52

Chapter Four: Results Analysis


Before discussing our empirical results in details, we will first broadly evaluate the three most important features of regression model. They are signs of variables coefficients, significance of proposed determinants and finally the model fitness (R2). Statistical results can be summarized as below:
Table Eight: Signs and Significance Level of Independent Variable

Variables Tax Shield Profitability Asset Compositions Dividend

Expected Signs + + + LTD:

2001-2004 LD + + STD +# + +

2005-2006 LD +# STD # + +

Growth Size
# Significant at p0.05

STD: + +

+#

+ +#

Data are summarized according to appendix two

According to the above table, we can observe a huge diversity between the expected results and actual ones, which will be discussed in more details in the later part. Also, variables itself has different signs for pre-pension scheme and post-pension scheme period. In terms of different forms of debt, the sign of most variables stay the same. However, this is not the case for profitability observed from 2001 to 2004 and dividend rate examined from 2005 to 2006.

One of the most important features of regression analysis is the significance level of explanatory variables. According to our results, only a few proposed determinants have significant effects on firms leverage. When looking separately at long-term and shortterm debt ratios, those factors play different roles and the significance level of coefficients differs with each other.

53

Another aspect of regression models is goodness of fit, which is a summary measure of how well the sample line fits the data. In this study, it refers to the explanatory power of our proposed determinants of capital structure. According to Appendix Three, we can argue that R2 is relatively low in all regression models at around 0.3. However, R2 turns to be higher when including dummy variables (refer to industry classifications) in the independent variable. Low explanatory power indicates that relationships between variables are not necessarily linear and can be caused by improper model or measurement problems. In our case, it is possibly because data are measured as book value instead of market value, which can lead to misleading results. Also, R2 turns to be higher when we include dummy variables. In the following paragraphs, we will provide detailed interpretation on obtained statistical results, indicating how those proposed factors influence long-term and short-term debt and how results modify when taking the new IAS 19 into consideration.

4.1 Effects of Industry Classification

To examine the impact on industry classification, we included industries as dummy variables and apply the ANOVA analysis. Main results from ANOVA are as follow:
Table Nine: Results from ANOVA
ANOVA Sum of Squares 2.794 16.475 19.269 .028 .174 .203

df 9 70 79 9 70 79

Mean Square .310 .235 .003 .002

F 1.319

Sig. .243

LTD/TA

Between Groups Within Groups Total

STD/TA

Between Groups Within Groups Total

1.267

.270

Above table shows no evidence to conclude that industry classification affects both longterm and short-term debt (significance level>0.05). This result is inconsistent with static trade-off theory and pecking order theory, which emphasizes the intra-industry similarity and inter-industry differences on firms leverage. In addition, it provides no evidence to

54

support Bradley et al (1983)s investigation, which indicates that industry factor have strong influence on firm leverage ratios. However, trade-off theory admits the existence of adjusting cost and delays. Firms cannot immediately offset the random event that bump them away from their capital structure targets, therefore, it is not surprising to observe differences in actual debt ratios among firms having the same target debt ratio. What we have observed are actual debt ratios rather than the targeted debt ratio. Therefore, the insignificance may be because firms are not able to adjust their debt level to optimum without delay. Appendix four reports the correlation between industry and other independent variables. We can view that the only determinant, which is significantly correlated to industry, is tangible assets (significance level<0.05). This indicates that although firms are operating in the same industry, they are considerable different in terms of profitability, dividend rate, growth rate, tax shield and firm size. When looking at industries individually (figure 8 and 9), the lowest leverage ratio in our sample is 0.11 (oil and gas industry) and the highest ratio is 0.85 (pharmaceuticals and biotech industry) regarding to long-term debt. Turning to short-term borrowing, the lowest leveraged industry is 0.03 (automobile industry) and the highest ratio is 0.09 (food and beverage industry). We can conclude that the majority of industries in our samples have similar debt ratio, with the exception of pharmaceuticals and biotech industry.

55

Figure Eight: Mean of Leverage Ratio by Industry

Mean of Leverage Ratio by Industry


1 0.8 0.6 0.4 0.2 0 F ood and B ev erages P harm ac eutic a ls and B iotec h T elec om m unic ation O il and G as R eal E s tate T rav el and Leis ure E lec tronic E quip A utom obiles C hem ic als Retailers M ean V alu e

LTD/TA STD/TA

Industries

Source adopted from appendix five

Figure Nine: Standard Deviation of Leverage Ratio by Industry

St d. Devi at i on of Lever age Rat i o by Indust r y


Std.Deviation Value

Pharmaceutical s and Biotech

Telecommunicat ion Travel and Leisure

Automobiles

Electronic Equip Food and Beverages

1. 8 1. 6 1. 4 1. 2 1 0. 8 0. 6 0. 4 0. 2 0 Chemicals

Oil and Gas

Real Estate

Retailers

LTD/ TA STD/ TA

Indust r i es
Source adopted from appendix five

According to the value of mean and standard deviation, we can observe that firms in

56

pharmaceuticals and biotech industry borrowed far more long-term debt than other firms and firms within this industry differs a lot on long-term financing. This comes as surprising results and is not consistent with static trade-off theory, which argues that growth industries with intangible assets always have lower leverage. In addition, Myers (2006) particularly takes pharmaceutical industry as an example to explain the pattern of capital structure,

4.2 Cross-sectional Research on Long-term Debt

The regression result where long-term debt is the dependent variable are presented in the following table:
Table Ten: Regression Results of Long-term Debt
Model Unstandardized Coefficients B 1 (Constant) Tax Shield Profitability Tangible Assets Dividend Growth Rate Size .171 .030 -.127 .028 -2.33E-005 .004 Std. Error .251 .042 .137 .219 .001 .001 .061 Standardized Coefficients Beta .683 .077 -.101 .015 -.005 .394 -.002 .712 -.925 .129 -.041 3.439 -.019 .497 .479 .147 .898 .967 .001 .985 .978 .956 .893 .902 .874 .849 1.023 1.047 1.120 1.108 1.144 1.178 t Sig.

Collinearity Statistics Tolerance VIF

-.001 a Dependent Variable: LTD/TA

When dependent variables are measured in long-term debt ratios our proposed determinants have little explanatory power with R2 equal to 0.307. Only one of our six empirical proxies of determinants of firms leverage has significant influence on long-term debt ratio. It is growth rate (sig. level<0.05), which has been confirmed by many studies on its high correlation with debt ratio. However, the positive sign of its coefficient supports neither trade-off nor pecking order theory. The diversity may be caused by selection of different variable measurement. Wald (1999) uses sales growth and Titman and Wessels (1988) uses total assets as the indicator of growth rate while we use turnover to indicate the growth rate. The coefficients estimates between long-term debt and tax shield as well as tangible

57

assets are positive but insignificant. The positive correlation is consistent with our prediction and provides evidence for trade-off theory, which indicates that companies with safe, tangible assets and plenty of taxable income to shield ought to have high target ratios.

Moreover, variables of dividends and firm size hold negative relationship with firms longterm debt ratio. This result is contradictory with our hypothesis and in some ways rejected pecking order theories. Also, the insignificance of size differs with some scholars research (Storey, Keasey and Wynarczy 1987, Chittenden, Hall and Hutchinson 1996), which demonstrates the importance of firm size on capital structure. However, it provides support to Gupta (1969) and Smith (1977), asserting the debt ratio was negatively related to size of the corporation. The insignificance of profitability is surprising for us. Myers (1984), Donaldson (1961) and Brealey and Myers (1984)s studies state that the past profitability and the amount of earnings are important factors of its current capital structure. Looking at more details, the observed negative coefficient provides evidence for pecking order theory and is consistent with our hypothesis. Meanwhile, the inverse correlation between profitability and long-term debt provide evidence to support pecking order theory and those researches, who indicates that in real-life capital structure, most profitable companies commonly borrow the least. (Titman and Wessel 1988; Rajan and Zingales 1995; Fama and French 1999; Wald 1999).

4.3 Cross-sectional Research on Short-term Debt

The regression result where short-term debt is the dependent variable is presented in the following table:

58

Table Eleven: Regression Results of Short-term Debt


Model Unstandardized Coefficients B 1 (Constant) Tax Shield Profitability Growth Rate Tangible Assets Dividend Size .010 .008 .002 .000 -.040 4.92E-005 .017 Std. Error .025 .004 .014 .000 .022 .000 .006 .208 .019 .110 -.201 .095 .312 Standardized Coefficients Beta .381 1.956 .174 .980 -1.799 .861 2.727 .704 .048 .862 .330 .076 .392 .008 .978 .956 .874 .893 .902 .849 1.023 1.047 1.144 1.120 1.108 1.178 t Sig.

Collinearity Statistics Tolerance VIF

When dependent variables are measured in short-term rather than long-term debt ratios our proposed determinants seem to have more explanatory power. The R
2

has

increased to 0.329 from 0.307, which represents that when borrowing short-term debt, financial managers are more likely to take these aspects into consideration. Most of the variables coefficients go in line with our prediction. There are two significant variables for short-term debt, which are tax shield and firm size. Firms growth rate in this case is no longer significant. We can see that firms behave very diversely when borrow long-time and short term. The positive coefficient of tax shield is steady with results obtained from long-term debt. However, firm size shows a reverse correlation with long-term and shortterm debt. The positive relationship between firm size and short-term debt ratio is consistent with our hypothesis and it is in line with McConnell (1982) and Marsh (1982)s study, asserting that large firms should be more highly leverage. Profitability turns to be the least important variable with significance level 0.862. Different from long-term debt, the sign of coefficient shows a positive correlation with debt ratio. This finding supports static trade-off theory against pecking order theory in that most profitable corporations with tangible assets borrow the most. Growth rate also turns to be insignificant when we regressing short-term debt. However, the positive coefficient is consistent with pecking order theory and the hypothesis. The result supports Myers (1977)s finding which suggest a positive correlation between growth rate and short-term debt because agency cost could be mitigated if firms issue short-term debt rather than long-term debt. The beta value of tangible assets and dividend on short-term debt are different from

59

ones on long-term debt. It is surprising to see that tangible assets have a negative relationship with short-term debt, which observes the opposite results from trade-off theory. This may be caused by the selection of sample firms and their desired debt ratio cannot be actually observed. However, the test on dividend falls on our prediction with pecking order theory, which suggests a positive correlation with leverage.

4.4 Effects of Pension Scheme

To examine how the treatment has influenced firms financing behaviour, we divided our period into pre-pension scheme and post-pension scheme period. To compare results obtained from 2005-2006 to ones from 2001-2004, we can observe the impacts of revised IAS 19 on firms capital structure. According to table 12, we can see that most firms adjusted their debt ratio after 2005. The mean of both long-term and short-term debt ratio has increased from 2005. This comes as surprising results because in order to keep the book value of liability stable, corporations are assumed to borrow less face to the inclusion of pension deficit in balance sheets. The increase of both short-term and long-term debt will definitely make firms look suffering more deficits. Also, the rise of standard deviation indicates that different firms have different pension positions.

Table Twelve: Comparison of Leverage between Pre-pension Scheme Period and Postpension Scheme Period

2001-2004 Without Pension Fund


N LTD/TA STD/TA 80 80 N LTD/TA STD/TA 80 80 Minimum 0 0 Minimum 0 0 Maximum 4.2689 0.2599 Maximum 12.186 4.471 Mean 0.255011 0.056951 Mean 0.438621 0.120403 Std. Deviation 0.4938795 0.050671 Std. Deviation 1.5965297 0.5050481

2005-2006 With Pension Fund

Source was summarized according to Appendix Five

Results on industry analysis have little different with pre-pension scheme period 2001-

60

2004. Pharmaceuticals and biotech still have the highest leverage debt (2.316917 for LTD/TA and 0.772417 for STD/TA) and interestingly, firms in telecommunication industry are just a little behind on long-term borrowing, with the mean of 1.145838. Compare to figure 8; it is reasonable to assume that pension deficit is larger in firms operating telecommunication industry.

Figure Ten: Mean of Leverage Ratio by Industry for Post-pension Scheme Period

M ean of Lever age Rat i o by Indust r y 2. 5 2 1. 5 1 0. 5 0 Automobiles Chemicals Electronic Equip Mean Value

LTD/ TA STD/ TA

Pharmaceutical s and Biotech

Telecommunicat ion

Food and Beverages

Oil and Gas

Real Estate

Travel and Leisure

I ndust r i es
Data are summarized from appendix six

Retailers
61

Figure Eleven: Std. Deviation of Leverage by Industry for Post-pension Scheme Period

St d. Devi at i on of Lever age Rat i o by Indust r y Std.Deviation Value 6 5 4 3 2 1 0 Automobiles Chemicals Electronic Equip

LTD. TA STD/ TA

Pharmaceutical s and Biotech

Telecommunicat ion

Food and Beverages

Oil and Gas

Real Estate

Travel and Leisure

I ndust r i es
Data are summarized from appendix six

4.4.1 Impacts on Long-term Debt

The model fitness has dropped from 0.307 of pre-pension scheme period to 0.250 of post-pension scheme period (appendix three), which means the explanatory power of proposed determinants has weakened when testing on data collected from 2005-2006. This finding indicates that have disclosed pension fund in their balance sheet, firms started to take more factors into considerations rather than just those proposed determinants. Looking at the following comparative table, we can see that new accounting policy has changed the signs and significance of some variables. Tax shield and growth rate have showed a contradictory correlation with long-term debt. In terms of significance, profitability and firm size have turned to become significant apart from growth rate.

Retailers
62

Table Thirteen: Comparison of Long-term Debt Regression Results between Pre-pension Scheme Period and Post-pension Scheme Period

2001-2004 Beta (Constant) Tax Shield Profitability Tangible Assets Dividend Growth Rate Size .077 -.101 .015 -.005 .394 -.002 Sig. .497 .479 .147 .898 .967 .001 .985

2005-2006 Beta Sig. 0.034 -0.08 0.491 -0.178 0.048 0.094 -0.02 -0.234 -0.178 0.43 0.86 0.05 0.039

Data are summarized according to appendix two

4.4.2 Impacts on Short-term Debt When regressing the short-term debt from 2005-2006, the model fitness is even lower, with the value of 0.189. The poor explanatory power indicated that financial managers seldom consider these proposed factors when facing short-term finance decision. In theory, the inclusion of pension liability should have less effect on short-term debt rather than long-term ones, because pension deficit is recognized as one item of longterm liability. However, just like long-term debt, the sign of variable coefficients and the significance of determinants have changed under the revised accounting policy. Tax shield, tangible assets and growth rate showed a reverse correlation with short-term debt in post-pension scheme period. The most significant determinants become the most insignificant from 2005 (p value changes from 0.008 to 0.817). Interestingly, tax shield is still significant yet has negative relationship with short-term leverage ratio, which is also the case with long-term debt ratio. This is hard to explain because firms with more debt ought to have more tax shield no matter what they have included in their long-term liability.

63

Table Fourteen: Comparison of Short-term Debt Regression Results between Pre-pension Scheme Period and Post-pension Scheme Period

2001-2004 Beta (Constant) Tax Shield Profitability Tangible Assets Dividend Growth Rate Size 0.208 0.019 0.11 -0.201 0.095 0.312 Sig. 0.704 0.048 0.862 0.33 0.076 0.392 0.008

2005-2006 Beta -0.219 0.125 -0.081 -0.114 -0.047 0.026 Sig. 0.078 0.029 0.055 0.512 0.127 0.689 0.817

Data are summarized according to appendix two

4.5 The Summary of Statistical Results

Take a general look at our results, we find it is not consistent with Agrawal and Nagarajan (1990)s studies on US firms. Agrawal and Nagarajan (1990) state that firms are averse to leverage of any kind, with little short-term as well. In addition, they have levels of cash and marketable securities well above their leverage counterparts. Whereas in our samples, only little firms have no debt and cash and marketable securities are only a small fraction of liabilities. Surprisingly, long-term debt in certain firm even exceeds the amount of total assets such as Vernalis in pharmaceuticals and biotech industry (LTD/TA=4.2689), which makes the industry have the highest debt ratio and standard deviation.

To conclude, same as many previous studies, statistical results obtained from our empirical studies have a relatively low explanatory power and are conflicting with leading theories. Although some of our proposed determinants have significant influence on firms leverage, the most of variables are insignificant. For example, the insignificance of industry classification provides no evidence to support both static trade-off theory and pecking order theory. In terms of coefficients signs, most of variables go in line with our hypothesis; however, observation on some variables rejects our hypotheses for both long-term and short-term debt ratio, such as dividend rate. The reason for this will be

64

discussed in the following research limitation part. The change in the treatment of pension deficit makes firms adjust their leverage ratio for both long-term and short-term aspects. For some determinants, the sign of coefficients even differ between pre-scheme period and post-scheme period.

After all, the conflicting results are in some ways due to the research shortcomings, which will be discussed in details in the following part.

4.6 Research Limitations

Limitations on Data and Sample Collection

Firstly of all, due to the time and data constraint for constructing this study, the findings are based on a relatively small sample of UK public companies (80 companies in 10 industries). Companies and industries are randomly chosen without any criteria. Apparently, the small sample size and avoidance of private companies cannot provide an integrated picture of UK firms and may lead to a research bias. In addition, data examined in this study are cross-sectional data, which are data on one or more variables collected at the same point in time. Just as other types of data, crosssectional data has their own problems, specifically the problem of heterogeneity. We have diagnosed the problem of multicollinearity yet ignored the possibility of heterogeneity and autocorrelation, which may lead to misspecification. (Gujarati, 2003).
Book Value vs. Market Value

When observing the capital structures of firms, it is important to distinguish between market values and book values (Ross et al, 2005). For example, suppose a firm buys back shares of its own stock and finances the purchase with new debt. This would suggest that the firms debt level should go up and its reliance on equity should go down. After all, the firm has fewer outstanding shares and more debt regard to the mix of equity

65

and debt. The analysis is more complicated than it seems because the market value of the firms remaining share of stock may go up and offset the effect of the increased debt. The matter of applying book value or market value is believed to be the central issue of why there is a disparity between the theoretical results and the empirical tests on capital structure. The theory is based on market-value measure of leverage. With few exceptions12, most empirical tests use book-value measures due to the data restriction. Differences between market value and book value are important sources of spurious correlation, which refers that the debt ratio can still show a significant correlation to proposed determinants even if debt levels are set randomly. If managers set debt levels in terms of book value rather than market value, then differences in market value across firms will not necessarily affect the amount of debt they issue. Therefore, suppose if some firms use book value targets while others use market value targets, the relationship between dependent variables and independent variables have been examined will then be spurious.
Issues related to independent variables

In this study, we failed to consider the signs of some items in P/L statement, such as dividend rate and tax shield. Dividend payout ratio was defined as current ordinary dividends divided by profit after taxes. Since companies may still pay dividends even when there is a loss, the statistical ratio may then not reflect the real situation and can be very misleading. Therefore, dividend ratio is suggested to set as zero when no dividend was paid and 100% when dividends exceeded earnings (Marsh, 1982). However, we failed to consider this issue and ignored the sign of profit, which may lead to the inaccurate sign of variable coefficient. For example, the dividend of XP Power was calculated as 625.88, which certainly contained the problem and may have led to the disparity between empirical results and theoretical predictions. We can then explain why obtained dividend results are not consistent with our prediction for both long-term and short-term debt. Similarly, the ignorance of a loss when calculating tax shield may cause misleading interpretation.

For example, Beaver, Kettle and Scholes (1970), Rosenberg and McKibben (1973) and Thompson (1976) use market value for bonds and preferred stock in their studies.

12

66

As what has been discussed before, the relatively low explanatory power R2 may be caused by exclusion of some important variables, which are difficult to measure statistically. The most obvious one is the cost of financial distress. Financial distress is especially important under the argument of static trade-off theory and it has been proved by many scholars that the cost of financial distress and bankruptcy significantly influence capital structure decision in complex ways. Although some scholars argue that firm size could be a proxy of financial distress, there are still no empirical evidences to support that argument. The most common measurement of financial so far is the Z-Score, which we fail to consider in our study (Altman, 1968). Moreover, according to Marsh (1982), market condition and timing variables claimed to be an important determinant of the debt and equity choice. In addition, we know from Marsh (1977) that there is a substantial difference between the pre-issue share price performance of debt and equity issuers, with equity issues tending to follow unusually good share price performance. After all, taking these variables into consideration would have increased the explanatory power and obtained a better descriptive model of firms leverage. In addition, given more time, I would have included the total debt ratio as another dependent variables. A study on total debt level could help us gain an overview of firms capital structure. In fact, most companies in real world only count the amount of total debt when make decision on capital structure.

67

Chapter Five: Conclusions and Recommendations


It is argued that in well-functioning capital markets with no taxes, debt policy does not matter. However, few financial managers could accept this issue as practical guidelines. With the aim to gain an insight into how companies actually choose between equity and debt, we conducted statistical models to examine what are the determinants of a companys leverage. In our studies, data were collected from financial reports of 80 UK public companies operating in 10 industries13. After choosing proxies for each variable, ANOVA and multiple regression analysis were developed to observe their coefficiences and significances. In general, our model came up with a relatively low explanatory power (R2 is around 0.3), although it is higher when we include the dummy variable in our model. As discussed in the research limitation, the poor explanatory power maybe mainly due to the measurement of book value. In addition, results varied between long-term and shortterm debt in the term of variable coefficients, significances and signs. The obtained signs of proposed determinants were not always consistent with our predictions. Results on certain variables are conflicting with our hypothesises for both long-term and short-term leverage, such as dividend rate. The reason for this has been explained in the limitation part and is believed due to the ignorance of profit signs. Our empirical results rejected the significance of industry classification regarding to both long-term and short-term debt ratio. The insignificance is not in line with both existing theories and numerous empirical studies such as Bradley (1983) and Titman (1984). When looking at industries individually, the mean of leverage stays constant in shortterm borrowing, yet they differ significantly for long-term debt. This can be explained by the study of Bennett and Donnelly (1993), who assert that long-term debt is more representative of a firms policy regarding to its capital structure. Companies in pharmaceuticals and biotech industry have borrowed far more long-term debt than others. This result is contradictory with Brealey et al s argument (2006) it is rare to find a pharmaceutical company that is not predominantly equity-financed. Glamorous growth companies rarely use much debt despite rapid expansion and often heavy requirements

13

The data set is selected from FAME, which provides financial reports for most of UK companies.

68

for capital (pp 469) To explain this disparity, we noted that certain company in pharmaceutical and biotech industry rely extremely on debt with the debt ratio 4.2689, which may have caused high mean value. Growth rate is the only significant factor influences firms long-term debt, yet it is not helpful when explaining short-term leverage. Results showed that there is a positive correlation between growth rate and leverage. When regressing short-term debt, tax shield and firm size turned to become significant with p value less than 0.05. Same with our prediction, tax shield showed a positive relation with both long-term and short-term debt. Results on firm size seem vary between different forms of debt with unpredicted signs. The insignificance of profitability is surprising for us, which supports neither static trade-off theory nor pecking order theory. A great number of researches find that profitability plays an important role in firms financing decision and it could significantly influence how much firms borrow. Again, the choice of book value is possibly the reason for this conflicting result. In addition, profitability has different signs in aspects of longterm and short-term debt. In terms of long-term, the negative sign supports pecking order theory and a number of empirical studies such as Kester (1986). Yet its positive correlation with short-term debt ratio can be explained by tax-based model. Furthermore, tangible assets and dividend could explain neither long-term nor short-term leverage. Stick to what has been argued in static trade-off theory; tangible assets came with positive coefficients. The negative coefficients of dividend rate showed no evidence to support pecking order theory.

Furthermore, transparency of pension fund in balance sheet is believed and has been proved by our empirical studies, to have impacts on firms leverage. Companies have adjusted their target ratios from 2005, yet in unexpected ways. The amount of leverage has increased after 2005, which is contrary to the hypothesis. Apart from tangible asset, all other proposed determinants indicated a different correlation with either long-term or short-term debt. Industry classification still has no effects on companies capital structure. Importantly, the explanatory power has decreased under the disclosure of pension deficit. However, both growth rate and firm size became significant to long-term debt apart from growth rate. In terms of short-term leverage, firm size no longer has impact.

To explain why our studies provide some conflicting evidence with theoretical

69

expectations, the research limitation has been discussed. To take an example, the number of sample size may not truly reflect the whole picture of capital structure in UK companies. Despite these limitations, our studies have some valuable implications and provide practical guidelines to financial managers. The predictive model from regression analysis could be used by investment analysts to forecast the financial policy of particular companies. Furthermore, financial managers looking to raise capitals could use the model to gain an insight into the decision other managers would make under the same circumstances. Used in this way, the model provides managers with some indication of what the market was anticipating. However, the limitations of this study cannot be avoided and therefore some further researches are suggested. For example, researches with larger sample size, covering more non-financial companies and more industries are advocated and given more time and sufficient resources, total debt should have been included as another dependent variables.

70

REFERENCE
Agrawal A. and Nagarajan N. (1990) Corporate Capital Structure, Agency Costs, and Ownership Control: The Case of All-Equity Firms, Journal of Finance, 45(4), p1325 Altman E. I. (1984) A Further Empirical Investigation of the Bankcrupcy Cost Question, Journal of Finance, 39(4), p1067-1089 Altman E. I. (1968) Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy, Journal of Finance, September Altman E., Haldeman R. and Narayanan P. (1977) Zeta Analysis: A New Model to Identify Bankruptcy Risk of Corporation, Journal of Banking and Finance, 1

Altman E. I. (2000) Predicting Financial Distress of Companies: Revising the Z-Score and Models, Ang J., Chus J and McConnell J (1982) the Administrative Costs of Corporate Bankruptcy: A Note, Journal of Finance, 37, p219-226 Andreade G. and Kaplan S. N. (1998) How Costly Is Financial (not Economic) Distress? Evidence From Highly Leveraged Transactions that Became Distressed, Journal of Finance, 53, pp 1443-1493 Bailey R. E (2004) Corporate Finance: The Modigliani-Miller Theorems, Economics of Financial Markets

Bancel F. and Mittoo U. R. (2004) Cross-country Determinants of Capital Structure Choice: A Survey of European Firms, Financial Management, 33(4), p103-132 Baskin J. (1989) an Empirical Investigation of the Pecking Order Theory, Finance Management, 18, p26-35 Beaver W., Kettler P. and Scholes M. (1970) The Association between Market Determined and Accounting Determined Risk Measures, the Accounting Review, 45(4), p654-682 Bennett M. and Donnelly R. (1993) the Determinants of Capital Structure: Some UK Evidence, British Accounting Review, 25, p43-59.
71

Berle A. and Means G. (1932) the Modern Corporate and Private Property, MacMillan, New York Berle A. (1954) the 20th Century Capitalist Revolution, Harcourt, Brace and World, Inc Booth L., Varouj A., et al. (2001) Capital Structure in Developing Countries, Journal of Finance, 56, p87-130. Bowen R. M., Daley L. A., and Huber Jr., C.C. (1982) Evidence on the Existence and Determinants of Inter-Industry Differences in Leverage, Financial Management, 11(4), p10-20 Bowman. R (1980) The Important of a Market-Value Measurement of Debt in Assessing Leverage, Journal of accounting research, 18(1), p242-254

Bradley M., Jarrell G. A. and Kim E. H. (1984) On the Existence of an Optimal Capital Structure: Theory and Evidence, Journal of Finance, 39(3), p857-878 Brealey R. A., Myers S. C. and Allen F. (2006) Corporate Finance, Eighth Edition, McGraw-Hill, New York Black, F and Scholes, M (1973) The pricing of options and corporate liabilities, Journal of political economy Buser S. and Hess P. (1983) The marginal cost of leverage, the tax rate on equity and the relation between taxable and tax-exempt yields. Ohio State University, Working Paper. Chirinko R. S. and Singha A. R. (2000) Testing the Static Tradeoff Against Pecking Order Model of Capital Structure: a Critical Comment, Journal of Financial Economics, 58(3), p417-425

Chittenden F., Hall G et al. (1996). Small Firm Growth, Access to Capital Markets and Financial Structure: Review of Issues and an Empirical Investigation. Small Business Economics 8(1): p59-67.

Copeland T. E., Weston J. F. and Shastri K. (2005) Financial Theory and Corporate Policy, Pearson Addison Wesley, UAS

72

Diamond. D and Rajan R. (2000) Banks, Short Term Debt and Financial Crises: Theory, Policy Implications and Applications, National Bureau of Economic Research DeAngelo H. (1981) Business Risk and Optimal Capital Structure, Unpublished working paper, University of Washington

DeAngelo H. and Masulis R. (1980) Optimal Capital Structure under Corporate and Personal Taxation, Journal of Financial Economics, 8, p3-19 Diamond D, Rajan R (2000), A Theory of Bank Capital, The Journal of Finance 55 (6), P 24312465. Donaldson G. (1961) Corporate Debt Capacity: A study of Corporate Debt Policy and the Determination of Corporate Debt Capacity, Boston, Division of Research, Harvard Graduate School of Business Administration Donaldson G. (1969) Strategy for Financial Mobility, Boston, Division of Research, Harvard Graduate School of Business Administration

Du X. H. (2006) Determinants of Capital Structure: Some UK Evidence, MA Dissertation, University of Nottingham Elliott B. and Elliott J. (2006) Financial Accounting and Reporting, 10th Edition, Prentice Hall, England Emery D. R., Finnerty J. D. and Stowe J. D. (2004) Corporate Financial Management, Second Edition, Pearson Prentice Hall, USA Fama E. F. and French K. R. (2002) Testing Trade-Off and Pecking Order Predictions About Dividends and Debt, the Review of Financial Studies, 15(1), P1-33

Frank M. Z. and Goyal V. K (2003) Testing the Pecking Order Theory of Capital Structure, Journal of Financial Economics, 67(2), p217-248 Graham. R., King. R., and Bailes, J. (2000), The Value Relevance of Accounting Information during a Financial Crisis: Thailand and the 1997 Decline in the Value of the Baht, Journal of International Financial Management and Accounting, Vo 11(2), p 84107.

73

Graham J. R. and Harvey C. R (2001) the Theory and Practice of Corporate Finance: Evidence from the Field, Journal of Financial Economics, 60, p187-243

Griner E. and Gordon L. (1995) Internal Cash Flow, Insider Ownership and Capital Expenditures: a Test of the Pecking Order and Managerial Hypothesis, Journal of Business Finance Accounting, 22, p179-97

Gupta M. C. (1969) the Effects of Size, Growth and Industry on the Financial Structure of Manufacturing Companies, Journal of Finance, 24, p517-529 Gujarati D. N. (2003) Basic Econometrics, Fourth Edition, McGraw-Hill, New York

Harris M. and Raviv A. (1990) Capital Structure and the Informational Role of Debt, Journal of Finance, 45, p321-49 Harris M. and Raviv A. (1991) the Theory of Capital Structure, Journal of Finance, 46(1), p297-355 Hsia C. C. (1981) Coherence of the Modern Theories of Finance, Financial Review, 16(1), p27-42

Jalilvand A. and Harris R. S. (1984) Corporate Behavior in Adjusting Capital Structure and Dividend Targets: An Econometric Study, Journal of Finance 39(1), p127 Jensen M. C. (1986) Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers, American Economic Review, 2, p323-29. Jaffe j. and Westerfield R. (1984) Risk and Optimal Debt Level, Working Paper, University of Pennsylvania Kim, E. (1986). Evidence on the Impact of the Agency Costs of Debt on Corporate Debt Policy. Journal of Finance and Qualitative Analysis 21, p131-44. Kester C. W. (1986) Capital and Ownership Structure: a Comparison of United States and Japanese Manufacturing Corporations, Financial Management, 15, p97-113. Lemmon M. L. and Zender J. F. (2002) Debt Capacity and Tests of Capital Structure Theories, Working Paper, University of Utah Li Y. (2006) Determinants of Capital Structure: The Case of Chinese Listed Companies,

74

A Dissertation Presented in Part Consideration for the Degree of MA Finance and Investment, University of Nottingham Lumby S. and Jones C. (2005) Corporate Financial Theory and Practice, 7th Edition, Thomson, London Masulis R. W. (1983) the Impact of Capital Structure Change on Firm Value: Some Estimates, Journal of Finance, 38, p107-126 Malhotra N. K (2004) Marketing Research: an Applied Orientation, 4th edition, Pearson education

Martin J. D. and Scott D. F. (1974) A Discriminant Analysis of the Corporate Debt-Equity Decision, Financial Management, 4, pp71-79 Marsh P. R. (1977) an Analysis of Equity Rights Issues on the London Stock Exchange, Unpublished Ph.D. dissertation, London Graduate School of Business Studies Marsh P. (1982) the Choice between Equity and Debt: An Empirical Study, Journal of Finance, 37(1), p121 Megginson W. L., Smart S. B. and Gitman L. J. (2007) Corporate Finance, Second Edition, Thomson South-Western, Canada Miller M. H. (1977) Debt and Taxes, Journal of Finance, 32(2), p261-275 Modigliani F. and Mille M. H. (1958) The Cost of Capital, Corporate Finance and The Theory of Investment, the American Economic Review, 48(3), p261 Modigliani F. and Mille M. H. (1963) Corporate Income Taxes and the Cost of Capital: A Correction. American Economic Review, 53(3), p443-453.

Modigliani, F. (1982). "Debt, dividend policy, Taxes, Inflation and Market Valuation." The Journal of Finance, 37(2), p255-273 Myers S. C. (1984) The Capital Structure Puzzle, Journal of Finance, 39(3), p575 Myers S. C. and Majluf N. S (1984) Corporate financing and investment decisions when

75

firms have information that investors do not have. Journal of Financial Economics, 13:2: 187-221.

Myers, S. C. (1977) Determinants of Corporate Borrowing. Journal of Financial Economics, 5(2): p147-75. Myers S. C. (2001) Capital Structure, Journal of Economic Perspective, 15(2), p81-102, Ozkan A. (2001) Determinants of Capital Structure and Adjustment to Long Run Target: Evidence from UK Company Panel Data, Journal of Business Finance and Accounting, 28(1&2), p175-198

Ogilvie J. (1999) The Capital Asset Pricing Model and Modigliani and Millers Capital Structure Theory, Management Accounting: Magazine for Charted Management Accountants, 77(2), p70 Pallant J. (2005) SPSS Survival Manual, 2nd Edition, Open University Press, United Kingdom Panno A. (2003) An Empirical Investigation on the Determinants of Capital Structure: the UK and Italy Experience, Applied Financial Economics, 13(2), p97-112 Rajan R. G. and Zingales L. (1995) What Do We Know About Capital Structure? Some Evidence from International Data, Journal of Finance, 50(5), p1421-1460 Rosenberg B. and Mckibben W. (1973) The Prediction of Systematic and Specific Risk in Common Stocks, Joumal of Financial and Quantitative Analysis, 8(2), p317-333 Ross S. A., Westerfield R. W. and Jaffe J. (2005) Corporate Finance, Seventh Edition, McGraw-Hill, New York

Rubinstein M. E (1973) The Fundamental Theorem of Parameter-Preference Security Valuation, The Journal of Financial and Quantitative Analysis, Vol. 8 (1), p. 61-69

Sarkar S. Zapatero F. (2003) The Trade-Off Model with Mean Reverting Earnings: Theory and Empirical Tests, the Economic Journal, 113, p834-860 Storey D., Keasey K., et al. (1987) The Performance of Small Firms: Profits, Jobs and Failures, Croom Helm, London.

76

Shyam S. L. and Myers S. C. (1999) Testing Static Tradeoff Against Pecking Order Models of Capital Structure. Journal of Financial Economics, 51, p219244.

Smith C. (1977) Alternative Methods for Raising Capital: Rights Versus Underwritten Offerings, Journal of Financial economics, 5, p273-307 Smith C. W. and Watts R. L. (1992) the Investment Opportunity Set and Corporate Financing, Dividend, and Compensation Policies, Journal of Financial Economics, 32, p263-92

Thompson D. (1976) Sources of Systematic Risk in Common Stocks, Journal of Business, 49(2), p 173-188 Titman S. (1984) the Effect of Capital Structure on a Firm's Liquidation Decision, Journal of Financial Economics, 13, p137-51. Titman S. and Wessels R. (1988) the Determinants of Capital Structure Choice, Journal of Finance, 43(1), p1-19 Trzcinka C. and Kamma S. (1983) Marginal Taxes, municipal bond risk and the Miller equilibrium: new evidence and some predictive tests. Working paper, UCLA, State University of New York at Buffalo. Tunvirachaisakul K. (2005) Capital Structure of Thai Firms-Target & Crisis, A Dissertation Presented in Part Consideration for the Degree of Master of Business Adiministration (Finance), University of Nottingham Wald J. K. (1999) How Firm Characteristics Affect Capital Structure: an International Comparison, Journal of Financial Research, p161-187

Walton, P. and Aerts W. (2006) Global Financial Accounting and Reporting: Principal and Analysis, Thomson, London

Warner J. B. (1977) Bankruptcy Costs: Some Evidence, Journal of Finance, 26, pp337348

77

Williamson O. (1988) Corporate Finance and Corporate Governance, Journal of Finance, 43, p567-591. http://cbdd.wsu.edu/kewlcontent/cdoutput/TR505r/page40.htm Financial Analysis Revised (Chapter 8: Module 1.4: Financial Statement Analysis) Session 4: Indicators and Sources of Financial Distress http://www.fame.bvdep.com/athens

78

Appendix
Appendix One: Firms Financial Information 2005-2006 Company
GKN Torotrak Carclo Croda ICI JOHNSON MATTHEY Porvair Treatt Victrex Zotefoam Elementis SMITH & NEPHEW Abacus Chloride Dialight Halma Inensys Laird XP Power Xaar Volex Spectris Renishaw Unilever Tate&Lyl SABMiller PGI Kerry Group IAW Group

LTD/TA
0.2100 0.1500 0.2700 0.2300 0.0400 0.2200 0.1800 0.0729 0.0000 0.0235 0.2421 0.0375 0.0921 0.0243 0.0000 0.0000 0.5396 0.2293 0.1040 0.0208 0.1136 0.1760 0.0000 0.1409 0.0042 0.1881 0.0437 0.3309 0.2360

STD/TA
0.0100 0.0500 0.0500 0.6600 0.0100 0.0100 0.0100 0.0481 0.0000 0.0058 0.0056 0.0570 0.2333 0.0952 0.0491 0.1032 0.0081 0.0060 0.2111 0.0120 0.2096 0.0475 0.0000 0.1351 0.1479 0.0635 0.1827 0.0213 0.0171

Ind
Automobiles Automobiles Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Food and Beverages Food and Beverages Food and Beverages Food and Beverages Food and Beverages Food and Beverages

TaxS
0.1900 0.8100 0.4500 0.1600 0.0700 0.2400 0.2900 0.0963 0.0030 0.3142 0.0959 0.0287 0.6656 0.2591 0.2044 0.0733 2.1460 0.2023 0.1170 0.0114 0.8131 0.2218 0.0478 0.1458 0.2745 0.1148 22.8194 0.2267 0.1182

Prof
0.0400 0.0200 1.7500 0.1000 0.3100 0.0400 0.0400 0.0996 0.3609 0.0854 0.0064 0.1843 0.0091 0.0541 0.0925 0.1662 0.0909 0.0896 0.1063 0.1990 0.0328 0.1013 0.2098 0.1208 0.0364 0.1652 0.0503 0.0575 0.0787

TanAss
0.3900 0.3400 0.3100 0.3600 0.2700 0.3100 0.1400 0.3170 0.5142 0.7269 0.2902 0.1960 0.1471 0.1139 0.1287 0.1552 0.1640 0.1221 0.4667 0.2109 0.1149 0.1346 0.2751 0.1671 0.3088 0.2551 0.3004 0.2591 0.2730

Div
0.9800 0.6900 0.6800 1.4500 0.3600 0.5400 0.8000 0.4009 0.3756 0.9954 0.1593 0.1982 2100 1.5424 0.2620 0.7083 0.0910 0.5601 0.5040 0.1577 0.0000 0.4216 0.5319 0.5319 1.2288 0.4121 0.3353 0.1440 0.1624

Gro
2.20% 31.39% 14.22% 36.80% 9.29% 15.94% 1.76% 5.57% 18.97% 9.27% 0.87% 6.81% 19.57% 11.34% 27.23% 9.00% -5.83% 14.41% 8.64% 9.62% 0.86% 5.58% 8.49% -2.43% 13.74% 18.39% 12.64% 3.47% 12.30%

Siz
4.5600 3.6700 3.2100 3.3000 4.8200 3.8900 2.6800 2.2467 2.5545 2.3865 3.3122 3.9281 2.9530 3.1250 3.1250 3.4950 4.5266 4.0067 2.5359 2.4472 3.9893 3.7706 3.2709 5.3021 2.9564 4.6752 4.1499 4.3668 3.5740

79

Glanbia Diageo Devro Dairy Crest Abbot Group Aminex BG Group BP Burren Energy XP Power Hunting JKX Tullow Dana Petroleum Acambis Vernalis Astrazenca Vectura Group BTG Dechra Brixton Liberty International Segro CLS Holdings Daejan Freeport Grainger Savills Minerva BT Group KCOM

0.3580 0.2823 0.1672 0.3231 0.2719 0.0032 0.1155 0.0502 0.0017 0.1040 0.1232 0.0000 0.1455 0.0133 0.0011 12.1860 1.5205 0.0003 0.0000 0.1936 0.3384 0.3837 0.2862 0.4960 0.1574 0.0857 0.7230 0.0149 0.3001 0.3120 0.3045

0.0158 0.0600 0.0212 0.0251 0.1426 0.0017 0.0135 0.0753 0.0007 0.1013 0.1525 0.0000 0.0019 0.0000 0.0635 4.4710 0.0405 0.0018 0.0293 0.0284 0.0000 0.1444 0.0054 0.0197 0.0094 0.1082 0.0178 0.0103 0.0881 0.1227 0.0002

Food and Beverages Food and Beverages Food and Beverages Food and Beverages Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Pharmaceuticals and Biotech Pharmaceuticals and Biotech Pharmaceuticals and Biotech Pharmaceuticals and Biotech Pharmaceuticals and Biotech Pharmaceuticals and Biotech Real Estate Real Estate Real Estate Real Estate Real Estate Real Estate Real Estate Real Estate Real Estate Telecommunication Telecommunication

0.2488 0.1697 0.1017 0.2422 0.2576 0.0636 0.0041 0.0196 0.0104 0.1170 0.1993 0.0074 0.0916 0.0304 0.0418 0.1383 0.0657 0.0093 0.0012 0.1511 0.2686 0.2098 0.2303 0.2558 0.1541 1.1167 0.4986 0.0103 0.7881 0.4406 0.2019

0.0337 0.2486 0.1375 0.0417 0.0587 0.2851 0.4537 0.1261 0.6656 0.1063 0.0358 0.7240 0.4280 0.5509 0.5959 0.4852 0.3006 0.8422 0.4327 0.0450 0.4750 0.4250 0.4500 0.8775 0.3360 0.1181 0.2671 0.1599 0.5774 0.1152 0.0795

0.3173 0.1490 0.6047 0.4148 0.3821 0.2887 0.4805 0.4168 0.5554 0.0467 0.2697 0.4935 0.5820 0.4155 0.1629 0.4392 0.2651 0.1320 0.1244 0.0638 0.4250 0.0001 0.0327 0.0039 0.4183 0.4168 0.0939 0.0364 0.0057 0.6088 0.3255

0.2509 0.5449 0.4874 0.1040 0.8898 0.0000 0.1318 0.2213 0.1147 0.5040 0.1940 0.0579 0.1675 0.0000 0.0000 0.0000 0.3648 0.0000 0.0000 0.3956 0.1194 0.1489 0.1346 0.0000 0.2718 1.0188 0.1928 0.4259 0.8864 0.5354 0.1314

0.28% -9.01% 1.50% 2.01% 30.18% 3.22% 32.33% -4.17% 77.26% 8.64% 20.11% 66.45% 63.83% 40.65% 38.30% 4.27% 9.29% 77.32% 11.04% 11.55% -4.11% -4.28% 13.90% -1.55% 0.97% -0.96% -2.49% 26.22% 34.09% 4.21% 15.14%

3.5890 4.3578 3.3407 3.8535 3.7457 1.4314 3.7081 4.0894 3.0894 2.5359 3.3905 2.5514 2.2820 1.7559 2.4472 2.2476 4.8192 2.0453 2.0810 2.8357 1.9934 2.8215 2.6144 2.0191 2.1538 2.2014 2.3118 4.1813 1.5051 5.0056 3.4041

80

Thus Vanco Vodafone Carphone Warehouse Universe Emap Arriva British Airways Carnival EasyJet Ryanair Labrokes Marston's Luminar Alexon Findel Lookers HMV Beale Greggs

0.1058 0.1729 0.0683 0.1278 7.8477 0.2277 0.1938 0.3264 0.2212 0.1676 0.3343 0.7052 0.4760 0.2598 0.0000 0.4208 0.1304 0.0000 0.1220 0.0000

0.0081 0.0359 0.0019 0.0445 0.0228 0.0316 0.0918 0.0443 0.1136 0.0128 0.0335 0.6628 0.0243 0.0006 0.0212 0.0508 0.0322 0.1629 0.0051 0.0000

Telecommunication Telecommunication Telecommunication Telecommunication Telecommunication Telecommunication Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Retailers Retailers Retailers Retailers Retailers Retailers

1.1495 0.1916 0.1750 0.1403 0.0351 0.1864 0.1409 0.3298 0.1676 0.1325 0.1709 0.2259 0.4797 34.9132 0.0142 0.3003 0.3464 0.0746 0.3815 0.0020

0.0501 0.0827 0.3220 0.0279 0.1206 0.1292 0.0649 0.0630 0.1493 0.0652 0.2108 0.1357 0.1253 0.0204 0.0585 0.0789 0.0142 0.0623 0.0045 0.0835

0.5863 0.1365 0.1226 0.1633 0.0608 0.0242 0.5951 0.6833 0.7739 0.2960 0.5479 0.1608 0.8197 0.5840 0.1274 0.1388 0.3045 0.3390 0.5842 0.7380

0.0000 0.0000 0.2404 0.2848 0.0126 0.8636 0.4118 0.0282 0.3201 0.0000 0.0000 3.6457 0.6591 2.7015 0.2912 0.4851 0.3255 0.3401 14.3426 0.4044

23.26% 32.95% -4.02% 30.19% 47.74% -8.27% -1.68% 4.36% 6.36% 21.85% 28.44% 47.41% 7.99% 26.08% 11.88% 13.20% 14.22% 0.30% -4.25% 4.53%

3.2652 2.7320 4.7846 4.1386 2.3345 3.7709 4.5108 4.9765 4.4138 3.6145 3.4153 4.5124 4.0840 3.8638 3.9255 3.5092 3.6709 4.1177 3.1796 4.2796

2001-2004

Company
GKN Torotrak Carclo Croda ICI JOHNSON MATTHEY Porvair Treatt Victrex Zotefoam Elementis SMITH & NEPHEW

LTD/TA
0.2400 0.2400 0.2900 0.1400 0.0300 0.1600 0.1300 0.1089 0.0253 0.0287 0.1517 0.1157

STD/TA
0.0300 0.0200 0.0600 0.0500 0.0100 0.0300 0.0400 0.0419 0.0000 0.0489 0.0123 0.0837

Ind
Automobiles Automobiles Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals

TaxS
0.2600 1.0200 0.3900 0.1800 0.0100 0.1300 0.1100 0.0738 0.0199 0.0902 0.8397 0.0714

Prof
0.0800 0.0300 -3.0900 0.1000 0.2300 0.0400 -0.1300 0.0761 0.3192 0.1524 -0.0265 0.1693

TanAss
0.4000 0.3300 0.3500 0.4200 0.3100 0.3000 0.1800 0.3577 0.5051 0.7774 0.3294 0.2092

Div
1.2300 1.2500 0.3400 27.0400 0.4200 0.4600 0.8100 0.5361 0.3775 1.2760 0.0053 0.3568

Gro
1% -4.81% -9.56% -1.95% -1.16% -5.38% -7.55% 7.30% 11.70% 4.93% -8.04% 2.54%

Siz
4.5700 3.7900 3.3900 3.2500 4.7700 3.8500 2.8700 2.2149 2.3579 2.3743 3.3749 3.8858

81

Abacus Chloride Dialight Halma Invensys Laird XP Power Xaar volex Spectris Renishaw Unilever Tate&Lyl SABMiller PGI Kerry Group IAW Group Glanbia Diageo Devro Dairy Crest Abbot Group Aminex BG Group BP Burren Energy XP Power Hunting JKX Tullow Dana Petroleum Acambis Vernalis

0.0825 0.0843 0.0003 0.0000 0.3688 0.2449 0.1418 0.0431 0.3960 0.3027 0.0000 0.2363 0.0052 0.1827 0.3028 0.1405 0.2367 0.2560 0.2124 0.2554 0.3775 0.1305 0.0028 0.0844 0.0756 0.0533 0.1418 0.2301 0.0000 0.2464 0.1090 0.1000 4.2689

0.1772 0.0807 0.0591 0.0768 0.1173 0.0263 0.0788 0.0184 0.0379 0.0366 0.0000 0.1939 0.0496 0.0822 0.1379 0.0258 0.0263 0.1386 0.1920 0.0288 0.0123 0.0539 0.0047 0.0746 0.0609 0.0472 0.0788 0.0390 0.0567 0.0668 0.0369 0.0468 0.0746

Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Electronic Equip Food and Beverages Food and Beverages Food and Beverages Food and Beverages Food and Beverages Food and Beverages Food and Beverages Food and Beverages Food and Beverages Food and Beverages Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Oil and Gas Pharmaceuticals and Biotech Pharmaceuticals and Biotech

0.1080 0.1232 0.0602 0.0165 3.9878 0.2082 0.2199 0.1339 0.3695 0.2919 0.0947 0.3225 0.0871 0.1488 0.9037 0.1941 0.1520 10.677 1 0.2765 0.1014 0.3275 0.1441 0.0005 0.0681 0.0843 0.2774 0.2798 0.2228 0.0183 0.2648 0.1160 0.0382 0.0237

0.0564 0.0262 0.0910 0.1594 0.1190 -0.0001 0.0430 0.0590 -0.0204 0.0662 0.1795 0.0824 0.0566 0.1391 0.0089 0.0540 0.0643 0.0148 0.1581 0.0427 0.0310 0.0661 -0.2139 0.3449 0.0735 0.2921 0.0430 0.0282 0.3158 0.2167 0.3150 0.0899 -0.2685

0.1536 0.0929 0.1948 0.1904 0.229 0.1597 0.0596 0.2053 0.2357 0.1581 0.2466 0.1754 0.4614 0.3141 0.7688 0.3286 0.323 0.4128 0.1450 0.6032 0.4309 0.3913 0.3239 0.5684 0.508 0.7794 0.096 0.315 0.5118 0.6038 0.3707 0.1339 0.2266

0.7182 0.7785 0.6918 0.7232 0.8328 1.5292 625.8800 0.0353 0.0672 0.6624 0.7170 0.8147 0.8599 0.6566 0.0214 0.1683 0.1714 0.1392 3.0069 0.3470 0.6964 0.5568 0.0000 0.1889 0.5917 0.0265 625.8838 0.7186 0.0893 0.1397 0.0000 0.0000 0.0000

7.89% -2.84% -8.59% 2.83% -21.68% -16.52% -2.39% 12.55%


-11.12%

2.8834 3.1785 3.1785 3.4637 4.8515 3.7101 2.3968 2.3878 3.9986 3.7392 3.2009 5.3997 3.0605 4.5639 4.1701 4.3666 3.5614 3.5755 4.6909 3.3692 3.8794 3.6266 1.6604 3.6588 3.7779 3.0002 2.3975 3.4001 2.5502 2.0746 1.7709 2.394 2.1414

7.83% 6.34% -1.66% -6.60% 33.01% -9.78% 13.07% 9.05% -7.02% -6.53% -8.79% 0.71% 26.41% 12.81% 1.24% 12.37% 30.02% -2.39% 1.92% 16.53% 250.69% 45.96% 224.82% 185.59%

82

Astrazemc a Vectura Group BTG Dechra Brixton Liberty Internation al Segro CLS Holdings Daejan Freeport Grainger Savills Minerva BT Group Kcom Thus Vanco Vodafone Carphone Warehouse Universe Emap Arriva British Airways Carnival EasyJet Ryanair Labrokes Marton's Luminar Alexon Findel Lookers HMV

0.5720 0.0467 0.0000 0.1330 0.4114

0.0259 0.0430 0.0142 0.1038 0.0127

Pharmaceuticals and Biotech Pharmaceuticals and Biotech Pharmaceuticals and Biotech Pharmaceuticals and Biotech Real Estate

0.0131 0.0161 0.0034 0.2169 0.4988

0.2350 -0.5893 -0.4723 0.0361 0.4532

0.3105 0.1874 0.0165 0.0962 0.9345

0.4227 0.0000 0.0000 0.5104 0.7485

-1.16% 16.96% 4.59% 6.47% 4.03%

4.7685 1.9138 2.277 2.7427 1.6628

0.3924 0.3122 0.5255 0.1909 0.0814 0.6229 0.0745 0.5061 0.4847 0.1803 0.0710 0.1645 0.0759 0.0440 1.1829 0.2089 0.1987 0.4741 0.3113 0.0842 0.1527 0.0538 0.3757 0.2695 0.0021 0.3436 0.1176 0.6769

0.0019 0.0077 0.0212 0.0736 0.0185 0.0638 0.0227 0.0054 0.0775 0.0001 0.0093 0.0804 0.0130 0.1590 0.0732 0.1350 0.0853 0.0481 0.0594 0.0119 0.0019 0.2599 0.0583 0.0294 0.0430 0.0202 0.0829 0.1176

Real Estate Real Estate Real Estate Real Estate Real Estate Real Estate Real Estate Real Estate
Telecommunication Telecommunication Telecommunication Telecommunication Telecommunication Telecommunication Telecommunication Telecommunication

0.5365 0.4226 0.6605 0.2737 0.0626 0.3754 0.0617 1.1975 0.4344 0.1808 0.2064 0.3687 0.1824 0.0203 0.4040 0.2002 0.1845 1.0290 0.2300 0.0769 0.1168 0.3184 0.4705 1.3524 0.0552 0.2484 0.2382 0.5688

0.3227 0.4172 0.2161 0.3544 0.4074 0.3032 0.1084 -0.0494 0.1153 -0.1597 -0.1952 0.0401 -0.3717 0.0164 -0.0066 -0.0606 0.0448 0.0101 0.1118 0.0886 0.2640 0.0305 0.0950 0.1045 0.0560 0.0727 0.0170 0.0297

0.8979 0.7244 0.9073 0.788 0.9129 0.0874 0.1031 0.8788 0.5797 0.7256 0.7877 0.2247 0.1045 0.1700 0.1754 0.0310 0.5764 0.7506 0.8771 0.2785 0.5172 0.5106 0.8131 0.6534 0.1390 0.1874 0.3880 0.3175

0.8307 0.8275 0.0000 0.4066 0.2404 0.1904 0.5107 0.1143 0.3200 0.0000 0.0000 0.0000 0.1062 0.2015 7.3341 0.1879 0.5404 0.4232 0.3224 0.0000 0.0000 1.0685 0.9040 0.0944 0.2630 0.5527 0.4220 2.2273

1.74% 4.70% 15.53% 2.18% 16.15% 37.01% 21.44% 11.61% -2.42% 13.01% 11.62% 39.49% 24.35% 22.83% 1.68% -1.62% 1.44% -4.06% 6.15% 43.98% 34.28% 33.28% -3.61% 13.58% 5.70% 14.05% 17.19% 5.16%

2.9186 2.7439 2.0616 2.0482 2.5115 2.0107 4.0735 1.4843 5.0440 3.3465 3.2973 2.4249 4.7911 3.8858 2.3909 3.7665 4.5012 4.9302 4.3428 3.4202 3.4124 4.7163 4.0942 3.8638 3.8948 3.4379 3.4851 4.1152

Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Travel and Leisure Retailers Retailers Retailers Retailers

83

Beale Greggs

0.0865 0.0000

0.1121 0.0000

Retailers Retailers

0.2021 0.0019

0.0232 0.0887

0.4847 0.7085

8.0533 0.3490

-1.33% 10.44%

3.1726 4.2358

Appendix Two: Tables of Coefficients Regression on Long-term Debt

2001-2004
Model Unstandardized Coefficients B .171 .030 -.127 .028 -2.33E-005 .004 -.001 Std. Error .251 .042 .137 .219 .001 .001 .061 Standardized Coefficients Beta .683 .077 -.101 .015 -.005 .394 -.002 .712 -.925 .129 -.041 3.439 -.019 .497 .479 .147 .898 .967 .001 .985 .978 .956 .893 .902 .874 .849 1.023 1.047 1.120 1.108 1.144 1.178 t Sig. Collinearity Statistics Toleranc VIF e

(Constant) Tax Shield Profitability Tangible Assets Dividend Growth Rate Size

2005-2006
Model 1 (Constant) Tax Shield Profitability Tangible Assets Dividend Growth Rate Size Unstandardized Coefficients B 1.524 -.028 -.881 .742 .000 -.017 -.310 Std. Error .704 .040 .556 .936 .001 .009 .197 Standardized Coefficients Beta 2.163 -.080 -.178 .094 -.020 -.234 -.178 -.692 -1.584 .793 -.177 -1.952 -1.572 .034 .491 .048 .430 .860 .050 .039 .915 .970 .863 .983 .850 .950 1.092 1.031 1.159 1.017 1.177 1.053 t Sig. Collinearity Statistics Tolerance VIF

Regression on Short-term Debt

2001-2004
Model 1 (Constant) Tax Shield Profitability Growth Rate Unstandardized Coefficients B .010 .008 .002 .000 Std. Error .025 .004 .014 .000 Standardized Coefficients Beta .381 .208 .019 .110 1.956 .174 .980 .704 .048 .862 .330 .978 .956 .874 1.023 1.047 1.144 t Sig. Collinearity Statistics Tolerance VIF

84

Tangible Assets Dividend Size

-.040 4.92E-005 .017

.022 .000 .006

-.201 .095 .312

-1.799 .861 2.727

.076 .392 .008

.893 .902 .849

1.120 1.108 1.178

2005-2006
Model 1 (Constant) Profitability Tangible Assets Growth Rate Size Tax Shield Dividend Unstandardized Coefficients B .295 -.343 .312 -.002 -.063 -.005 5.68E-005 Std. Error .228 .180 .302 .003 .064 .013 .000 Standardized Coefficients Beta 1.295 -.219 .125 -.081 -.114 -.047 .026 -1.912 1.032 -.659 -.987 -.402 .232 .078 .029 .055 .512 .127 .689 .817 .970 .863 .850 .950 .915 .983 1.031 1.159 1.177 1.053 1.092 1.017 t Sig. Collinearity Statistics Tolerance VIF

Appendix Three: R2 Summary

Regression on Long-term Debt

2001-2004
Model Summary Adjusted R Square Std. Error of the Estimate

Model 1

R Square

.401(a) .307 .205 .4706778 a Predictors: (Constant), Size, Profitability, Tax Shield, Dividend, Tangible Assets, Growth Rate

2005-2006
Model Summary Adjusted R Square Std. Error of the Estimate

.331(a) .250 .110 1.5671772 a Predictors: (Constant), Size, Profitability, Dividend, Tax Shield, Tangible Assets, Growth Rate

Model 1

R Square

Inclusion of dummy variables


Model Summary Adjusted R Square .271 Std. Error of the Estimate .4350471

Model 1

R .408(a)

R Square .389

85

a Predictors: (Constant), Industry, Size, Profitability, Tax Shield, Dividend, Growth Rate, Tangible Assets

Regression on Short-term Debt

2001-2004
Model Summary Adjusted R Square .207 Std. Error of the Estimate .0474305

Model 1

R .436(a)

R Square .329

a Predictors: (Constant), Size, Profitability, Tax Shield, Dividend, Tangible Assets, Growth Rate

2005-2006
Model Summary Adjusted R Square Std. Error of the Estimate

.268(a) .189 -.165 .5062158 a Predictors: (Constant), Dividend, Tax Shield, Profitability, Size, Tangible Assets, Growth Rate

Model 1

R Square

Inclusion of dummy variables


Model Summary Adjusted R Square .124 Std. Error of the Estimate .0477150

Model 1

R .450(a)

R Square .397

a Predictors: (Constant), Industry, Size, Profitability, Tax Shield, Dividend, Growth Rate, Tangible Assets

86

Appendix Four: Correlations between Independent Variables

Correlations Tax Shield -.065 .567 80 -.065 .567 80 .113 .320 80 .278(*) .013 80 -.097 .397 79 .134 .235 80 .063 .581 80 -.004 .972 80 .061 .589 80 -.025 .825 79 -.116 .305 80 .099 .381 80 .193 .085 80 .005 .962 79 .053 .638 80 -.018 .875 80 80 -.204 .072 79 -.035 .758 80 -.086 .448 80 79 -.068 .552 79 -.166 .143 79 80 -.304(**) .006 80 80 80 1 Tangible Assets .278(*) .013 80 .061 .589 80 .193 .085 80 1 Growth Rate .134 .235 80 -.116 .305 80 .053 .638 80 -.035 .758 80 -.068 .552 79 1

Industry Industry Pearson Correlation Sig. (2-tailed) N Tax Shield Pearson Correlation Sig. (2-tailed) N Profitability Pearson Correlation Sig. (2-tailed) N Tangible Assets Pearson Correlation Sig. (2-tailed) N Dividend Pearson Correlation Sig. (2-tailed) N Growth Rate Pearson Correlation Sig. (2-tailed) N Size Pearson Correlation Sig. (2-tailed) N 1

Profitability .113 .320 80 -.004 .972 80 1

Dividend -.097 .397 79 -.025 .825 79 .005 .962 79 -.204 .072 79 1

Size .063 .581 80 .099 .381 80 -.018 .875 80 -.086 .448 80 -.166 .143 79 -.304(**) .006 80 1

80 80 * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

87

Appendix Five: Descriptive Figures from ANOVA Test 2001-2004


N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Lower Upper Bound Bound .240000 .240000 .060845 .049763 .149548 .047521 -.915759 .195084 -.018042 .118667 -.071921 .145104 -.038531 .019511 .030627 .037879 .036478 .016864 .005499 .019362 .001003 .011247 .045675 .175215 .252855 .291552 .167259 2.62262 6 .497650 .621092 .361333 .480821 .364919 .088531 .055849 .098300 .139601 .067422 .085903 .045057 .117513 .137547 .114019 .068228 Minimum Maximum

LTD/TA

Automobiles Chemicals Electronic Equip Food Beverages Oil and Gas and

2 10 11 10 10 6 9 8 8 6 80 2 10 11 and 10 10 6 9 8 8 6 80

.240000 .118030 .151309 .220550 .107390 .853433 .346367 .301525 .240000 .204450 .255011 .025000 .037680 .064464 .088740 .051950 .051383 .025278 .068438 .069275 .062633 .056951

.0000000 .0799386 .1511528 .0992534 .0836918 1.6858511 .1968117 .3822477 .1451316 .2633523 .4938795 .0070711 .0253990 .0503658 .0710983 .0216289 .0328932 .0257319 .0587010 .0816634 .0489653 .0506710

.0000000 .0252788 .0455743 .0313867 .0264657 .6882458 .0656039 .1351450 .0513118 .1075131 .0552174 .0050000 .0080319 .0151859 .0224833 .0068396 .0134286 .0085773 .0207539 .0288724 .0199900 .0056652

.2400 .0253 .0000 .0052 .0000 .0000 .0745 .0440 .0538 .0000 .0000 .0200 .0000 .0000 .0123 .0047 .0142 .0019 .0001 .0019 .0000 .0000

.2400 .2900 .3960 .3775 .2464 4.2689 .6229 1.1829 .4741 .6769 4.2689 .0300 .0837 .1772 .1939 .0788 .1038 .0736 .1590 .2599 .1176 .2599

Pharmaceuticals and Biotech Real Estate Telecommunication Travel and Leisure Retailers Total STD/TA Automobiles Chemicals Electronic Equip Food Beverages Oil and Gas

Pharmaceuticals and Biotech Real Estate Telecommunication Travel and Leisure Retailers Total

88

2005-2006
N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Lower Upper Bound Bound -.201186 .561186 .055856 .011005 .120725 .020440 -2.795017 .143568 -1.119298 .186130 -.059233 .083331 -.224124 -.059526 .028141 .023951 .003864 -1.129201 .003438 .000391 -.062238 -.018088 .008010 .207344 .225304 .294155 .145260 7.42885 1 .475410 3.41097 3 .484945 .283633 .793912 .284124 .230826 .149150 .113989 .094036 2.67403 4 .086184 .066534 .308163 .108822 .232795 Minimum Maximum

LTD/TA

Automobiles Chemicals Electronic Equip Food and Beverages Oil and Gas Pharmaceutical s and Biotech Real Estate Telecommunica tion Travel and Leisure Retailers Total

2 10 11 10 10 6 9 8 8 6 80 2 10 11 10 10 6 9 8 8 6 80

.180000 .131600 .118155 .207440 .082850 2.316917 .309489 1.145838 .335538 .112200 .438621 .030000 .085650 .088645 .068970 .048950 .772417 .044811 .033463 .122963 .045367 .120403

.0424264 .1058824 .1594941 .1212187 .0872437 4.8711268 .2158549 2.7094239 .1787124 .1633571 1.5965297 .0282843 .2029418 .0900622 .0629325 .0630265 1.8120385 .0538243 .0395585 .2215263 .0604657 .5050481

.0300000 .0334830 .0480893 .0383327 .0275889 1.9886292 .0719516 .9579260 .0631844 .0666903 .1784974 .0200000 .0641758 .0271548 .0199010 .0199307 .7397616 .0179414 .0139860 .0783214 .0246850 .0564661

.1500 .0000 .0000 .0042 .0000 .0000 .0149 .0683 .1676 .0000 .0000 .0100 .0000 .0000 .0158 .0000 .0018 .0000 .0002 .0006 .0000 .0000

.2100 .2700 .5396 .3580 .2719 12.1860 .7230 7.8477 .7052 .4208 12.1860 .0500 .6600 .2333 .1827 .1525 4.4710 .1444 .1227 .6628 .1629 4.4710

STD/TA

Automobiles Chemicals Electronic Equip Food and Beverages Oil and Gas Pharmaceutical s and Biotech Real Estate Telecommunica tion Travel and Leisure Retailers Total

89

You might also like