Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis [1 ed.] 9783836645478, 9783836695473

Citation preview

Elmar Puntaier

Capital Structure and Profitability

Copyright © 2010. Diplomica Verlag. All rights reserved.

S&P 500 Enterprises in the Light of the 2008 Financial Crisis

Diplomica Verlag

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Elmar Puntaier Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis ISBN: 978-3-8366-4547-8 Herstellung: Diplomica® Verlag GmbH, Hamburg, 2010

Copyright © 2010. Diplomica Verlag. All rights reserved.

Dieses Werk ist urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funksendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen und der Speicherung in Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Verwertung, vorbehalten. Eine Vervielfältigung dieses Werkes oder von Teilen dieses Werkes ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland in der jeweils geltenden Fassung zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtes. Die Wiedergabe von Gebrauchsnamen, Handelsnamen, Warenbezeichnungen usw. in diesem Werk berechtigt auch ohne besondere Kennzeichnung nicht zu der Annahme, dass solche Namen im Sinne der Warenzeichen- und Markenschutz-Gesetzgebung als frei zu betrachten wären und daher von jedermann benutzt werden dürften. Die Informationen in diesem Werk wurden mit Sorgfalt erarbeitet. Dennoch können Fehler nicht vollständig ausgeschlossen werden und der Verlag, die Autoren oder Übersetzer übernehmen keine juristische Verantwortung oder irgendeine Haftung für evtl. verbliebene fehlerhafte Angaben und deren Folgen. © Diplomica Verlag GmbH http://www.diplomica-verlag.de, Hamburg 2010

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Contents List

Copyright © 2010. Diplomica Verlag. All rights reserved.

Abbreviations ...................................................................................................................3 1

Introduction .......................................................................................................5

1.1

General Definition and Justification of Issues and Objectives .............................5

1.2

Research Questions and Methodology ...............................................................7

1.3

Overview and Organisation of Chapters .............................................................8

2

Existing Theories and their Predictions........................................................11

2.1

Existing Theories ..............................................................................................11

2.1.1

The Trade-Off Theory .......................................................................................14

2.1.2

The Pecking Order Theory ................................................................................17

2.2

Predictions of Existing Theories ........................................................................20

2.2.1

Capital Structure and Industry...........................................................................20

2.2.2

Capital Structure and Profitability ......................................................................24

2.2.3

Capital Structure and Liquidity ..........................................................................26

2.2.4

Capital Structure, R&D and Tangible Assets ....................................................27

2.2.5

Capital Structure and Dividend Policy ...............................................................30

2.3

A Final Comment ..............................................................................................32

3

Methodology....................................................................................................33

3.1

Data Sampling ..................................................................................................33

3.2

Data Collection ..................................................................................................34

3.2.1

Missing Values and Adjustments ......................................................................35

3.3

Variables and their Definitions ..........................................................................36

3.3.1

Leverage and Gearing ......................................................................................36

3.3.2

Profits and Return .............................................................................................38

3.4

Hypotheses and Hypotheses Testing ...............................................................38

3.4.1

H1: Capital Structure and Industry .....................................................................39

3.4.2

H2: Capital Structure and Profitability ................................................................40

3.4.3

H3, H4 and H5: Capital Structure, Liquidity, R&D and Dividend Policy ...............41

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

4

Findings and Analysis ....................................................................................43

4.1

H1: The Impact of Industry on Capital Structure and Associated Variables .......43

4.2

H2: The Existence of a Correlation Between Leverage and ROCE ...................53

4.3

H3: The Existence of a Correlation Between Leverage and Liquidity ................58

4.4

H4: The Existence of a Correlation Between Leverage and R&D ......................61

4.5

H5: The Existence of a Correlation Between Leverage and Dividends .............65

5

Conclusions ....................................................................................................70

5.1

Capital Structure ...............................................................................................70

5.2

Profitability ........................................................................................................71

5.3

Liquidity .............................................................................................................72

5.4

Investments.......................................................................................................72

5.5

Dividends ..........................................................................................................73

6

Recommendations ..........................................................................................75

7

Reflections.......................................................................................................79

7.1

Objectives .........................................................................................................79

7.2

Strengths...........................................................................................................79

7.3

Weaknesses and Limitations ............................................................................80

Bibliography ...................................................................................................................82 Appendix A – S&P 500 industries, January 2004 and June 2009 ..................................89 Appendix B – Missing values of dead firms, based on balance-sheet records ..............90 Appendix C – Significant year-by-year correlations of core factors with Gearing 2 .......91 Appendix D – Independent t-test of delisted and non-delisted firms ..............................92 Appendix E – Independent t-test of the ten lowest and highest geared firms ................93

Copyright © 2010. Diplomica Verlag. All rights reserved.

Appendix F – Independent t-test of the ten less and most profitable firms ..................103 Appendix G – Mann-Whitney test of the ten lowest and highest geared firms .............113

2

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Copyright © 2010. Diplomica Verlag. All rights reserved.

Abbreviations

AV bn CCEG CD CE CEO CI CS DP EBIT EN FI H HC IN IT MA MM N NA NIATC NPV PAT PBT PCC R&D ROCE S&P Std. Dev. SG&A Sig. SPSS TE USD UT WACC

Average (e.g. profits from 2004 to 2008) billion Cash and Cash Equivalents Generic Consumer Discretionary (Sector/Industry) Capital Employed Chief Executive Officer Confidence Interval Consumer Staples (Sector/Industry) Dividends Payments Earnings before Interest and Tax Energy (Sector/Industry) Financials (Sector/Industry) Hypothesis Health Care (Sector/Industry) Industrials (Sector/Industry) Information Technology (Sector/Industry) Materials (Sector/Industry) Modigliani and Miller Number of Cases (Sample Size) Not Applicable Net Income Available to Common Net Present Value Profit After Tax Profit Before Tax Pearson Correlation Coefficient Research and Development Return on Capital Employed Standard and Poor‘s Standard Deviation Selling, General and Administration Expenses Significance (Level) Statistical Package for Social Sciences Telecommunication (Sector/Industry) US Dollar Utilities (Sector/Industry) Weighted Average Cost of Capital

3

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Copyright © 2010. Diplomica Verlag. All rights reserved.

4

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

1

Introduction

1.1

General Definition and Justification of Issues and Objectives

The publication of the Modigliani and Miller (MM) capital structure irrelevance theorem in 1958 and the subsequent preference of purely debt financing due to tax advantages in 1963, was in contradiction to traditional approaches which suggested an optimal capital structure. Meanwhile the theories of MM are academically accepted (Fama and Miller 1972; Kraus and Litzenberger 1973; Miller 1988; Frank and Goyal 2009) and out of competition with other approaches, since the underlying assumptions, especially the existence of perfect capital markets, is considered as unreal (Jackson 2009). However, in every economic boom, when access to capital becomes easier, financial markets seem to come close to the conditions of perfect markets, characterised by high competition and prosperity.

It is found that the western economic order is marked by asset bubbles that resulted in over one hundred crises over the last three decades (Stiglitz 2008) and which bring companies back to reality with a hard landing. Access to capital becomes extremely

Copyright © 2010. Diplomica Verlag. All rights reserved.

restricted and uncertainty dominates as the collapse of Lehman Brothers in September 20081 showed. Although signs were evident in 2007, the change from prosperity to depression can come overnight, where free market policy shows its true face, with unpredictable damages deeply wounding in the economy, and seeming to paralyse even the most experienced economists (Atkins and Guha 2009). 1

http://business.timesonline.co.uk/tol/business/industry_sectors/banking_and_finance/article4761892.ece (accessed 03.04.09)

5

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Since liquidity becomes a scarce resource and consumption declines, free cash flows that were previously available to finance an amply corporate structure, dividends and bonuses, are likely to fall. As debt, if any, must still be paid back – often to worse conditions than before (Johnson 2009) – corporations might run out of liquidity, as has happened to major US companies during the last twelve months (Tieman 2009; Sakoui 2009). Also, investments that ought to ensure future profits are likely to be reduced or to come to a still stand (Giles 2009), sending firms and the economy in a downward spiral. However, as experienced and predicted by Copeland (2005) and Greenspan (2008), systematic organisations which are considered as ‗too-big-to-fail‘ are offered bail-outs at the cost of society.

This study aims to investigate the impact of the capital structure on the profitability of large capitalised US companies. It does not, therefore, aim to test existing theories, nor does it try to find a model to predict one or another capital structure, since numerous attempts have previously been made that have so far struggled to capture the full complexity of the real world (Arnold 2008; Ross et al. 2008; Watson and Head 2007). Rather, it focuses on correlations between capital structure and profitability and major profitability-associated measures that can have an impact on a firm‘s survival, i.e. liquidity, dividends, investments and the impact of an industry-related target gearing

Copyright © 2010. Diplomica Verlag. All rights reserved.

ratio as a potential systematic risk. Thus, this work is supposed to contribute to the understanding of how resistant companies are to financial distress, and it provides evidence on the extent to which vulnerability can be reduced to prevent major systemic crises by means of their capital structure adjustments through the awareness of shareholders and corporate governors.

6

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

1.2

Research Questions and Methodology

The basis of this research project is a selection of secondary performance data over the period from 2004 to 2008 of firms listed in the Standard & Poor‘s 500 index (S&P 500) in January 2004. The index represents the 500 largest capitalised US companies among ten sectors that reflect the whole US market (Standard & Poor‘s 2008a). The combination of the US market and the S&P 500 companies, who have access to the widest range of financial sources, is expected to give a highly reliable result to find empirical evidence for the following research questions.

Question 1. According to the MM theorem and the pecking order theory that relies on information asymmetry between insiders and investors (Myers 1984), leverage should not depend on the industry a firm is in. However, evidence (Ross et al. 2008; Antoniou et al. 2008) suggests that firms in different industries operate with different capital structures. Thus, the first hypothesis (H1) is to verify whether industry-specific leverage exists.

Question 2. Since revenues are likely to decrease in an economic downturn, this reduces a firm‘s ability to meet debt payments, which is expected to have a negative

Copyright © 2010. Diplomica Verlag. All rights reserved.

impact on profitability. The second and central hypothesis (H2) is, therefore, that a negative correlation between gearing ratio and profitability exists, i.e. higher geared firms are less profitable. As this research question is the centre of attention, it merits a deeper investigation than all other hypotheses, especially for the years 2007 and 2008.

7

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Question 3. Most of the companies are affected by financial distress, not because they are not unprofitable, but because they have no liquidity (McLaney and Atrill 2008). If cash inflows decline, firms are likely to be unable to finance current expenses, including the interests on debt. Hence, the third hypothesis (H3) is the existence of a correlation between gearing ratio and liquidity. Higher geared firms are supposed to have lower cash positions, especially in 2008.

Question 4. Investments in R&D are crucial for survival and competitiveness of firms within some industries. Since higher geared firms must concentrate more to the avoidance of financial distress, they may tend to reduce expenses in long-term R&D projects that have no immediate effects. The aim of the fourth hypothesis (H 4) is to prove whether a correlation between gearing ratio and R&D expenditure exists, especially in 2008, that is expected to have a negative influence on future profits.

Question 5. Highly geared companies are encouraged to pay out higher dividends by transferring wealth from bondholders to shareholders (Ross et al. 2008), although managers should have an incentive to reduce them if liquidity becomes a scarce resource. The fifth hypothesis (H5) is, therefore, to find evidence of the existence of a

Copyright © 2010. Diplomica Verlag. All rights reserved.

correlation between gearing ratio and dividend policy, especially in 2008.

1.3

Overview and Organisation of Chapters

This book is organised into seven sections, each with a brief statement at the beginning and the end of the issues previously and subsequently discussed. Although this might

8

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

seem repetitious, it enables the reader to go through in multiple sessions. The following few paragraphs give an outline of the next six sections.

After the introduction and definition of the above stated research questions, section 2 attempts to review the existing literature on capital structure with a discussion of the main theories, after which a more detailed focus on the fields in respect of the research questions is provided.

Section 3 discusses and justifies the methodology used to answer the research questions, which refers to data sampling and data collection, the treatment of missing values, the variables defined and applied hypothesis testing methods.

Section 4 outlines the key findings in respect of the hypotheses initially stated, which are then analysed and discussed. Where appropriate, the results found are related to the most relevant findings discussed in the literature review.

Section 5 recalls the research objectives and findings previously obtained. After that, it concludes with the underlying assumptions required for the implications drawn from those findings, which are part of the next section.

Copyright © 2010. Diplomica Verlag. All rights reserved.

Section 6 attempts to identify management implications and recommendations to solve the issues. Based on the results revealed from the sample, it aims to identify measures in reducing vulnerability and systematic risk in order to achieve sustainable economic growth without adverse effects for society.

9

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Ultimately, section 7 points to the achievement of the objectives of this study, its

Copyright © 2010. Diplomica Verlag. All rights reserved.

strengths, weaknesses and limitations.

10

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

2

Existing Theories and their Predictions

In the previous section the objective of this paper and its importance, as well as the research questions are stated. This section reviews existing literature and discusses major capital structure theories that are subsequently related to the core aspects stated in the five research questions, where theories and findings are analysed in more detail and lead to the expected predictions.

2.1

Existing Theories

Following the publication of ―The Cost of Capital, Corporation Finance and the Theory of Investment‖ by Modigliani and Miller in 1958, scholars became busy for a while trying to provide evidence that capital structure, i.e. the mix of equity and debt, is not irrelevant for a firm‘s value. With the MM theorems, the traditional approach, according to which ―an optimal capital structure does exist for individual companies‖ (Watson and Head 2007:264) – determined by gearing level, volatility of profits, bankruptcy risk and the weighted average cost of capital (WACC) – was rejected (Ross et al. 2008). To explain

Copyright © 2010. Diplomica Verlag. All rights reserved.

that the capital structure is irrelevant, MM (1958:279) used the analogy of a dairy farmer, who ―cannot in general earn more for the milk he produces by skimming some of the butter fat and selling it separately, even though butter fat, per unit weight, sells for more than whole milk.‖ Thus, it does not matter whether a company is financed by 100 percent equity or 100 percent debt, or any mix of the two, because ―the increased cost

11

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

of borrowed funds as leverage increases will tend to be offset by a corresponding reduction in the yield of common stock‖ (MM 1958:274).

Since MM assumed no taxation in their first proposition, they revised it and considered also corporate taxation in a second proposition. MM (1963) came to the conclusion that debt financing should be favoured, since interests can be deducted from taxes payable that results in a decline of the WACC, if the proportion of debt increases. The optimum is, therefore, a capital structure consisting of 100 percent debt that guarantees the highest tax-shield. It should be noted that MM (1958; 1963) assumed perfect markets (where firms always have access to capital to meet their debt payments, making the capital structure independent from profit volatility), no transaction costs, no agency and bankruptcy costs, and that individuals have access to capital markets on the same conditions as firms.

Empirical evidence, however, has shown that firms rarely rely solely on debt financing and that they ―generally use less debt than equity‖ (Ross et al. 2008:419). It is clear that taxable profit must be available to offset interests from taxes, which defines the optimum amount of debt that a firm should take on. However, Graham (2000:1914) finds on a sample of ―87,643 [firm-year] observations from 1980 to 1994‖ that ―[t]he capitalized tax-reducing benefit of interest deductions is [9.7] percent of firm value‖

Copyright © 2010. Diplomica Verlag. All rights reserved.

(ibid.:1935) and could be increased by a further 15 percent, if personal taxation is not considered.

Moreover, Graham (ibid.:1902) finds that ―firms that use debt conservatively are large, profitable, liquid, in stable industries, and face low ex ante costs of distress‖, who also have ―growth options and relatively few tangible assets.‖ He identifies a trend towards a

12

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

more aggressive debt policy of US firms since the 1980s due to increased competition, measures taken by tax authorities to encourage debt financing and, according to Grinblatt and Titman (cited in Graham 2000), a reduction in transaction costs on global capital markets. Fama and French (2004:229) give evidence that ―[t]he number of new firms listed on major U.S. stock markets jump[ed] from 156 per year for 1973–1979 to 549 per year for 1980–2001‖, while the probability of surveillance declined due to a higher sensitivity to changes in supply and demand of equity. They believe that this increase resulted from lower costs of equity issuance, while such changes were not identified as industry-specific.

Thus, increased competition and lower transaction costs offer firms more flexibility in their capital structure management, but firms still do not exploit the full potential of the so-called ‗debt capacity‘ to lower the WACC and fail to optimise shareholder wealth. This indicates that MM ignored crucial factors that influence financing decisions (Fama and French 1998; Watson and Head 2007). As Ross et al. (2008:479) put it, ―[t]he theories of capital structure are among the most elegant and sophisticated in the field of finance‖, but ―the practical applications ... are less than fully satisfying‖; a view shared by Frank and Goyal (2009). This leads to different models of capital structure, other than the propositions of MM (1958; 1963), which seem inappropriate for real world conditions, characterised by imperfect capital markets that are never at an equilibrium

Copyright © 2010. Diplomica Verlag. All rights reserved.

(Klein 2007; Davis 2009; Soros 2009; Triana 2009). The most popular approaches are the ‗trade-off‘ theory and the ‗pecking order‘ model as proposed by Myers (1984), while ―[r]ecently, the idea that firms engage in ‗market timing‘ has become popular‖ (Frank and Goyal 2009:1). Both the trade-off and pecking order models are discussed in the next two sub-sections, since they are among the most cited models in existing literature.

13

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

2.1.1 The Trade-Off Theory

The trade-off theory dates back to Kraus and Litzenberg (1973:918), who argued that ―taxation of corporate profits and the existence of bankruptcy penalties are market imperfections that are central to a positive theory of the effect of leverage on the firm's market value.‖ By completing the equation with those two factors, firms can increase their debt level in order to maximise firm value as long as the marginal benefit of additional debt is not offset by costs of financial distress. However, Graham (2000:1934) points out that ―many firms overestimate the effect that using additional debt would have on the probability … of distress‖ which he estimates at 33 to 75 percent is far too high, if costs of financial distress of ―10 to 20 percent of firm value‖ are considered, as empirical evidence from Andrade and Kaplan (1998:1488) of thirty-one highly levered firms suggests.

Frank and Goyal (2009:5) point to the ‗agency perspective‘, defended by Jensen and Meckling (1976) and Jensen (1986:324), according to which ―debt reduces the agency costs of free cash flow by reducing the cash flow available for spending at the discretion of managers.‖ Stulz (1990) states that managers are likely to over-invest if free cash flows are high and under-invest if they are too low, resulting in agency costs. Because

Copyright © 2010. Diplomica Verlag. All rights reserved.

managers prefer equity and low debt levels that allow them to control the company‘s resources to improve their position, Stulz (1990) argues that shareholders could force managers to issue more debt that reduces over-investment, which thus results in an optimum combination of debt and equity to maximise shareholder wealth.

14

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

However, as Reich (2009) and Bakan (2005) emphasise, the larger a company and the more shareholders there are that own a small stake, the more they feel like investors rather than owners, unwilling and unable to control managers, who are well aware of this. This suggests a low debt ratio, even though a firm may be profitable with low volatile cash inflows, as Graham (2000) finds. This is consistent with Berger, Ofek and Yermack (1997), who find that a low CEO turnover and fewer monitoring activities result in a suboptimal capital structure. The agency theory is also supported by Jung, Kim and Stulz (1996), stating that firms still issue equity when alternative forms of financing are available, and thus reducing shareholder wealth. Such a phenomenon was also observed by Myers (1984:582), noting that ―[t]here are plenty of examples of firms issuing stock when they could issue investment-grade debt.‖ In contrast, Graham and Harvey (2001:226) find that under-investment problems appear more likely in ―more growth ... than non-growth firms‖, but the overall evidence is rather weak.

Graham and Harvey (2001) studied the capital structure behaviour of 392 chief financial officers of small and large firms and found that tax advantages play an inferior role for issuing debt, while personal taxation – in contrast to Miller (1977) – and transaction costs remain almost unconsidered. Also Myers (1984) and Titman and Wessels (1988) find no evidence that the latter is of any major importance for capital structure adjustments. As Frank and Goyal (2009) note, although corporation tax influences

Copyright © 2010. Diplomica Verlag. All rights reserved.

behaviour, it is difficult to anticipate and evaluate possible tax advantages, since the existence of transaction costs makes it more difficult to find evidence. However, ―tax advantage is most important for large, regulated, and dividend-paying firms‖ if the benefits achieve a certain satisfactory level (Graham and Harvey 2001:210). Large firms are also more worried about their credit ratings than financial distress (ibid.). Therefore, it is less surprisingly that only 19 percent of the firms have no target ratios – since credit

15

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

ratings also consider the probability of financial distress – while 37 percent have a flexible and 44 percent a more or less concrete target gearing ratio (ibid.). They point out that ―[t]argets are important if the CEO has short tenure or is young, and when the top three officers own less than 5% of the firm‖ (ibid.:211). Although, there is some support for the trade-off theory, it is not fully approved.

A dynamic model that considers financial restructuring, due to macroeconomic changes, is offered by Fischer, Heinkel and Zechner (1989), applying regression analysis on a sample of 999 firms on a quarterly basis from 1977 to 1985. They found that ―smaller, riskier, lower-tax, lower-bankruptcy-cost firms will exhibit wider swings in their debt ratios over time‖, as smaller firms are more sensitive to transaction costs – consistent with Titman and Wessels (1988) – and finally intervene only if debt considerably exceeds a desired level (Fischer, Heinkel and Zechner 1989:39).

Consistent with this approach is the research of Frank and Goyal (2009), carried out on over 200,000 publicly traded US firms from 1950 to 2003. The correlations between leverage and a series of factors (25 in total) give evidence of six ‗core factors‘ that have the highest impact on capital structure: ―1) Industry median leverage, 2) Tangibility, 3) Market-to-book assets ratio, 4) Profitability, 5) Log of assets, and 6) Expected inflation‖, of which only profitability is inconsistent with predictions of the bankruptcy-tax

Copyright © 2010. Diplomica Verlag. All rights reserved.

trade-off model (Frank and Goyal 2009:18). In consideration of all 25 factors, they constructed a regression model with market leverage as a dependent variable that gave evidence of a decline in the impact of those factors from 42 percent to 24 percent from 1950 to 2003 that explains debt levels (ibid.:19). Easier access to global capital markets and lower transaction costs might have contributed to such a development.

16

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Frank and Goyal (2009:4) share consistencies with Tsyplakov (2008), according to whom ―firms [tend to] stockpile retained earnings until the time is right to buy physical capacity‖, while tax-shields other than debt are also considered. The latter is supported by DeAngelo and Masulis (1980:20), who agree that ―each firm has a unique interior optimum capital structure in market equilibrium in a world characterized by … the equity-biased personal tax code … [and] corporate tax shield substitutes for debt and/or positive default costs.‖ To conclude, Frank and Goyal (2009:27) support the trade-off theory for ―factors such as industry leverage, firm size, tangibility, and market-to-book.‖ However, they point out that ―in dynamic trade-off models ... leverage and profits can be negatively related‖, while a weakness of the trade-off is ―that more profitable firms generally have lower leverage‖ (ibid.:27).

2.1.2 The Pecking Order Theory

The pecking order theory was recognised as managerial practice by Donaldson (1961) and later enhanced by Myers (1984), who implied information asymmetry between managers and investors due to changes in stock prices, when equity issues are announced that are considered negative news and therefore result in an asset revaluation by stockholders (Younghwan 2007; Frank and Goyal 2009). Therefore,

Copyright © 2010. Diplomica Verlag. All rights reserved.

Myers (1984) argues that firms should first use internal financing which also produces the lowest transaction costs. If this is not possible, priority should be given to the safest debt, while equity issuance – that also affects ownership – is the last means by which to finance positive NPV projects. Because there are costs related to the adjustment of a target debt-equity ratio, firms do not adjust their capital structure to changing circumstances as they would without costs that leads to ―no well-defined target debt-

17

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

equity mix‖, where ―[e]ach firm's observed debt ratio reflects its cumulative requirements for external finance‖ (Myers 1984:581).

In a replication of previous studies – such as Toy et al. (1974) for the US, Netherlands, Norway, Japan and France, and Kester (1986) for the US and Japan – Baskin (1989) identifies strong support for the pecking order theory on a sample of 378 Fortune 500 firms listed in 1960 over the period from 1960 to 1972. In a regression analysis, with book value debt-equity ratio as a dependent variable, and return on assets (before interest and tax) and growth as independent variables, Baskin (1989:33) observes that ―once [funding is] controlled, borrowing behaviour appears serially uncorrelated‖. Although debt is given the priority, since information asymmetry increases ―with those securities whose value is most dependent upon publically unknown future prospects‖, debt issuance is limited by bankruptcy costs that, however, seem to be ―more elastic than that of equity among [the] sample of large mature corporations‖ (ibid.:33). Baskin (1989:26) sees ―pecking order behaviour as the rational response not only to tax and transaction costs, but also as a signalling equilibrium‖ in imperfect equity markets due to information asymmetry that takes the form of a passive capital structure adjustment.

Support for the pecking order model is also provided by Graham and Harvey (2001) to some extent, who find debt over equity preference for small firms only. Despite the

Copyright © 2010. Diplomica Verlag. All rights reserved.

absence of information asymmetry, Graham and Harvey (2001:219) notice that flexibility is consistent with the pecking order model, since especially large firms are ―reluctant to issue common stock when they perceive that it is undervalued‖. However, they also state that firms who decide to remain flexible in their capital structure are likely to be firms who pay dividends and where information asymmetry is generally low, which is in contradiction with the pecking order theory, while ―the window of opportunity [to issue

18

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

equity] is most important for firms suffering from informational asymmetries‖ that refers to non-dividend-payers (ibid.:222). Their results indicate low association of leverage with signalling effects that is, however, more likely for speculative organisations who do not pay dividends and prefer equity issuance to communicate growth opportunities. The results indicate that small firms are ―more likely to suffer from informational asymmetries‖ (ibid.:222), ―but [there is] little evidence that executives are concerned about asset substitution, asymmetric information, transactions costs, free cash flows, or personal taxes‖ (ibid.:188).

Graham (2000) points out that neither the pecking order nor the trade-off model explains why firms tend to be debt-adverse. While Baskin (1989) finds notable supporting evidence for the pecking order theory that explains corporate behaviour under consideration of asymmetric information and disregards the trade-off theory, Frank and Goyal (2009:5) argue that ―[t]he pecking order theory is often used to explain financing decisions of firms.‖ However, both support the view that a static optimal capital structure, as proposed by DeAngelo and Masulis (1980), is not applicable. Frank and Goyal (2009:5) point to Shyam-Sunder and Myers (1999) and argue that ―[a] significant merit of the pecking order theory is that it predicts the effect of profits correctly‖, but they emphasise that ―[it] is not helpful in organizing many of the features we see in the way firms finance themselves‖ and refer to Fama and French (2002) and Frank and Goyal

Copyright © 2010. Diplomica Verlag. All rights reserved.

(2003). Although Frank and Goyal (2009) do not aim to test the capital structure theories, they find more supportive evidence for the trade-off model and emphasise the importance of the industry in which a business is in, a factor that is not considered by the pecking order theory.

19

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

The pecking order provides a rational explanation of management behaviour, but evidence is not fully satisfactory. Due to the absence of an optimal debt-equity ratio, a mature firm would be likely to have accumulated large amounts of debt that leads to a high leverage. In a recession, a firm could then face problems of paying debt-related interests, and if profits decrease – as is likely for the majority of industries (Guerrera et al. 2009) – that firm would also have minor growth opportunities, especially in the shortterm. Retained earnings erode and it might become extremely difficult to raise additional finance, due to insufficient collateral. Since share prices fall considerably, as happens periodically in any crisis, equity issuance is hardly a reasonable option. Thus, if no equity issue is possible, this may result in a failure, where Chapter 11 is the only valid option. Therefore, pecking order seems to work better in stable industries with less volatile profits.

Summarising these aspects, it seems that a firm‘s specific optimal debt level exists which, due to the complexity of influencing factors that are hard to identify and to predict, in most cases results in a flexible capital structure. This would equate to the view of Watson and Head (2007:272), which argues that ―the WACC curve will be flatter in practice than the U-shaped curve put forward by academic theories.‖

Copyright © 2010. Diplomica Verlag. All rights reserved.

2.2

Predictions of Existing Theories

2.2.1 Capital Structure and Industry

The pecking order theory suggests that leverage is industry-independent and thus in line with MM (1958; 1963). Also DeAngelo and Masulis (1980:23) point to a firm specific

20

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

leverage that has no direct association with the industry. However, advocates of the trade-off theory argue that industry does matter, as is supported by empirical evidence. Bradley et al. (1984), Ross et al. (2008) and Frank and Goyal (2009) emphasise the importance of industry, a factor that seems to be undervalued in most research papers. According to Ross et al. (2008:481), ―debt ratios tend to be quite low in high-growth industries with ample future investment opportunities‖, while they tend to be high in ―[i]ndustries with large investments in tangible assets‖. He also argues that ―almost any industry has a debt-equity ratio to which companies in that industry tend to adhere‖ (ibid.:439).

Industries with unique or homogeneous products ought to operate with lower debt levels due to high risk and costs of financial distress (Titman 1984; Graham 2000; Frank and Goyal 2009:9). This is confirmed by Titman and Wessels (1988), who find that, in a sample of 469 firms (mostly large) from 1974 to 1982, product uniqueness results in lower debt rates as for production and equipment manufacturers. Due to a low degree of product diversification, those firms are forced to have higher R&D and sales expenses (Titman and Wessels 1988) which is considered as weak collateral and increases profit volatility. Because in the sample period of Titman and Wessels (1988) the IT sector was less developed than it is today, low debt may also be used by firms within the IT sector. Graham (2000:1910) comes to the conclusion that firms within

Copyright © 2010. Diplomica Verlag. All rights reserved.

―sensitive‖ industries use debt moderately and, in contrast to Chevalier (1995b), (ibid.:1930) finds that ―firms in concentrated industries (high asset Herfindahl) use debt aggressively.‖ Chevalier (1995a) argues that in the supermarket industry an increase of leverage results in higher competition. Thus, if low profits indicate high competition, this would imply high debt levels for the respective industry.

21

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

The phenomenon of industry-specific leverage ratios may encourage firms to benchmark themselves with their equivalent considered as ‗best in class‘ (Frank and Goyal 2009:8). Also financial services organisations and consultancies tend to benchmark companies to capture the complexity of an organisation‘s influencing forces. This, however, can drive a whole industry in a misleading direction that may chronically over-leverage, as is identified by Wolf (2009) for the financial sector. He (2009:15) points out that ―[i]n a highly leveraged limited liability business, shareholders will rationally take excessive risks, since they enjoy all the upside but their downside is capped: they cannot lose more than their equity stake‖. Even though sophisticated mathematical tools are used to forecast market behaviour, they fail in the face of the unpredictable ―occurrence of the occasional extreme event‖, the so-called ―black swan‖ (Taleb and Spitznagel 2009:11). While in boom years all seems to work well, expectations of investors and managers‘ targets increase (Reich 2009), an industry may not be prepared to sustain a downturn that ultimately leads to a systematic failure as various financial crises have shown (Khor 2001; Porter 2005; Greenspan 2008).

A second explanation provided by Frank and Goyal (2009:8) is that ―industry effects reflect a set of correlated, but otherwise omitted, factors‖. All companies within the same industry have to deal with the same micro- and macroeconomic circumstances that ―could reflect product market interactions or the nature of competition‖, but also

Copyright © 2010. Diplomica Verlag. All rights reserved.

―industry heterogeneity in the types of assets, business risk, technology, or regulation‖ (ibid.:8). Thus, firms with unique products who operate with a more specialised workforce are more vulnerable to macroeconomic changes that results in cash flow volatility, which increases the probability and cost of financial distress and implies debt conservatism, also because tax-shields remain unused if profits are omitted (ibid.). Therefore, different industries are supposed to have a different debt level range,

22

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

because ―higher risk should result in less debt under the trade-off theory‖ (ibid.:10) with higher debt levels for more regulated industries, since their cash flows are more stable. However, as Frank and Goyal (2009) correctly point out, under the pecking order theory, risky businesses imply more volatile stock prices and cash flows that would result in higher debt levels, since debt is accumulated and equity would be the last means to raise capital required.

As Frank and Goyal (2009:8) note, ―[g]rowth increases costs of financial distress, reduces free cash flow problems, and exacerbates debt-related agency problems‖ that, under trade-off, would result in lower debt levels, since those firms use significantly more equity than firms without such growth opportunities. If growth pushes down leverage, this implies that industries with high leverage are mature with limited potential for growth. However, the opposite is implied under the pecking order model, because ―firms with more investments – holding profitability fixed – should accumulate more debt over time‖ (Frank and Goyal 2009:8). In their study, they find that leverage and growth are negatively correlated over the whole period from 1950 to 2003, if considered as terms of market-to-book value in correlation to long-term debt (or total debt) to market assets, while long-term debt to book assets results in a positive correlation. Frank and Goyal (2009:15) also find that a high median implies higher leverage, bringing them to the conclusion that ―the most important single empirical factor is industry leverage‖

Copyright © 2010. Diplomica Verlag. All rights reserved.

(Frank and Goyal 2009:27).

23

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

2.2.2 Capital Structure and Profitability

While MM (1958) and Myers (1984) see capital structure as being independent from profits, the bankruptcy-tax trade-off theory and the alternative trade-offs are very explicit about the optimal leverage.

However, several studies (Graham 2000; Jung, Kim and Stulz 1996) indicate that managers – whether intentionally or unintentionally – do not always contribute to sustainable shareholder wealth. Reich (2009) points out that hyper-competition forces CEOs to generate profits in the short-term – which contributes to an increase in share prices, which in turn indicates favourable future profits – at the expense of long-term profitability. While some firms will always be more profitable than others, and although higher profits reduce the risk of financial distress and increase the effectiveness of taxshields through higher debt levels, ―all published statistical studies conducted over 50 years in five countries show a prominent negative relationship‖ between profitability and leverage (Baskin 1989:28).

Such a study conducted by Kester (1986), who compared US and Japanese manufacturing companies, concludes that there is no significant difference in their

Copyright © 2010. Diplomica Verlag. All rights reserved.

leverage if measured in market value, but in terms of book value, Japanese companies are less profitable since they accumulate more debt due to a faster growth, because ―[d]uring expansion, stock prices go up, expected bankruptcy costs go down, taxable income goes up, and cash increases‖ which encourages borrowing (Frank and Goyal 2009:11). A similar ―relationship between debt, growth, and profits is … found‖ by Baskin (1989:28), based on a sample of ―378 firms from the 1960 Fortune 500‖ from

24

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

1960 to 1972 and also by Antoniou et al. (2008), based on a sample of 4,854 firms from major world economies (US, UK, France, Germany and Japan) from 1987 to 2000. Baskin (1989:33) also finds that ―debt leverage varies positively with past growth and inversely with past profits.‖ Thus, the pecking order predicts that ―[i]f investments and dividends are fixed, then more profitable firms will become less levered over time‖ (Frank and Goyal 2009:7).

The use of more debt is not only promoted by the tax-bankruptcy trade-off, but also by the agency cost model, according to which ―the discipline provided by debt is more valuable for profitable firms as these firms are likely to have severe free cash flow problems‖ (Jensen cited in Frank and Goyal 2009:7). Thus, higher profitability implies higher debt levels, but research is still ongoing, since complexity in dynamic trade-off models seems to be vast (Strebulaev 2007). However, according to Frank and Goyal (2009:11) ―agency problems are likely to be more severe during downturns as manager‘s [sic] wealth is reduced relative to that of shareholders‖. In their latest research, they find that profitability and a high market-to-book ratio are significantly negatively correlated to leverage, while industry median, tangible assets, size and inflation are positively correlated. The negative correlation of profits to total debt-to-book assets of -0.334, p < 0.01 was found over the whole sample period, whereas profitability is defined as operating income before depreciation to assets (Frank and Goyal 2009).

Copyright © 2010. Diplomica Verlag. All rights reserved.

However, Frank and Goyal (2009:19) point out that ―[t]he impact of profits declines sharply‖ from -0.54 in the 1950s to -0.05 at the end of the sample period, while ―the effects of firm size and dividend paying status have both increased in economic importance.‖

25

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Although Frank and Goyal (2009:26) fail to give evidence of whether firms consider market or book value in their leverage decisions and why dividend payers operate with lower debt than non-dividend payers, they find that with ―book leverage, the effects of market-to-book, firm size, and expected inflation factors all lose the reliable impact that they have when studying market-based leverage.‖ However, ―industry median leverage, tangibility, and profitability remain reliable and statistically significant‖, for which an explanation is given by Barclay et al. (cited in Frank and Goyal 2009:26), who note that ―book-leverage is backward looking while market leverage is forward looking‖ and thus market-leverage reflects future prospects that should also be related to variables that are able to absorb information about future expectations. Nevertheless, ―industry median leverage, tangibility, and profitability appear as robust factors in various definitions of leverage.‖ (Frank and Goyal 2009:18)

2.2.3 Capital Structure and Liquidity

As long as an economy is growing, insufficient liquidity is only a problem if the future prospects of a company let assume an inability to repay a loan and if satisfactory collateral is unavailable. In a recession the laws change and access to capital becomes more difficult, leading to an increase in the cost of capital (Johnson 2009). Since profits

Copyright © 2010. Diplomica Verlag. All rights reserved.

decline in most industries (Guerrera et al. 2009) and share prices are generally undervalued, it is inconvenient to issue equity that makes takeovers easy for those who have cash. A higher geared firm must certainly pay higher interests that decrease free cash flows, but it must also finance dividend payments and profitable investments that might motivate it to increase borrowing more than less leveraged firms would (Baskin

26

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

1989). Although the pecking order predicts that mature firms operate with high debt levels, those firms are expected to have less liquidity.

One possibility to increase liquidity is the reduction of inventories, as has been observed over the last few decades (Frank and Goyal 2009), but these are only a small part of a firm‘s assets. Another way is to increase short-term borrowings, which can be cheaper than long-term borrowing, and an option preferred by smaller firms (Titman and Wessels 1988). Since unique assets can only be sold below the real market value or not sold at all, the Swiss National Bank (cited in Ackermann 2009:1) argues that ―large banks should hold capital and liquidity buffers that account for the systemic risk they pose‖, while reducing leverage, rather than restricting systematic relevant organisations by their size.

Insufficient liquidity does not only increase the probability of financial distress, it is also argued that ―illiquid firms face high ex ante borrowing costs‖ (Graham 2000:1909). Hence, highly geared companies are supposed to have a lower interest cover and thus a weaker cash position due to higher interest payments, while firms with better cash positions are expected to be more profitable.

Copyright © 2010. Diplomica Verlag. All rights reserved.

2.2.4 Capital Structure, R&D and Tangible Assets

While financial institutions invest intellectual property to invent new products, partly to undermine restricting laws (The Economist 2009), firms in most other sectors are required to invest in new technology to ensure future growth and competitive advantage that, for non-financials, can be a considerable proportion of their earnings. Since a cut in

27

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

those expenses has no immediate effect, such long-term investments may be reduced if profits and liquidity decline. This is more likely if leverage is high, since interest payments must be met. In fact, Andrade and Kaplan (1998:1489) observe that high leveraged firms have less R&D expenditure, which results from more ―mature industries that did not require large amount[s] of R&D.‖

R&D, along with SG&A expenditure, is an intangible asset that suggests lower debt levels for those firms with large investments in R&D (Graham 2000; Antoniou et al. 2008; Frank and Goyal 2009). Baskin (1989:29) refers to studies from Long and Malitz (1985) and Bradley et al. (1984) who demonstrate that ―intangible assets (often lost in bankruptcy) ... ought to discourage debt finance.‖ Baskin (1989:32) argues, that ―[u]nless the investment is completely inelastic … it also must rationally adjust to reflect the availability of funds‖, whose costs, under the pecking order, become lower as the ability to reduce the probability of financial distress increases. This leads to higher profits, which in turn increase R&D expenditure. For Fama and French (1999:1954), ―debt plays a key role in accommodating year-by-year variation in investment‖ but, in a more competitive environment, where the cost of capital decreases, investments do not increase (Fama and French 2004). Furthermore, because the cost of equity contains information about risk and, therefore, has an impact on investors‘ expected return,

Copyright © 2010. Diplomica Verlag. All rights reserved.

seasoned firms can reduce their equity if their costs are low (Fama and French 2004).

Tangible assets that are good collateral (Graham 2000) enable firms to negotiate beneficial conditions that can lower the costs of capital. In addition, if collateral is used to issue debt, then this should increase with economic growth, since the market value of collateral is higher during those times, as concluded by Frank and Goyal (2009:11). Also, ―lower expected costs of distress and fewer debt-related agency problems predict

28

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

a positive relation between tangibility and leverage‖ (Frank and Goyal 2009:9). Myers (1984:581) states that firms with ―[s]pecialized, intangible assets or growth opportunities are more likely to lose value in financial distress‖ and thus will have higher debt levels that, however, are lower if information asymmetry is low (Frank and Goyal 2009). Frank and Goyal (2009:9) predict that in order ―[t]o protect unique assets that result from large expenditures on SG&A and R&D, these firms will have less debt.‖ This seems to be partly consistent with Baskin (1989:33), whose ―findings constitute evidence of the sensitivity of investment to internally generated funds, and they are consistent with asymmetric information acting to partially isolate firms from financial markets‖.

R&D expenses might also be used as debt-substitute to benefit from the tax-shield, as demonstrated by DeAngelo and Masulis (1980), since higher profits lead to progressively higher taxes that decrease marginal expenses. Thus, if ―non-debt tax shields are a substitute for the tax benefits of debt financing‖ (Frank and Goyal 2009:9), it can be assumed that firms with lower debt have higher R&D expenditure. Empirically, Frank and Goyal (2009:15) find that expenditure for R&D had a significant negative correlation to total debt-to-market value (-0.127, p < 0.01), while almost uncorrelated for long-term debt to book assets (-0.057, p < 0.01) and similar results were found for product uniqueness.

Copyright © 2010. Diplomica Verlag. All rights reserved.

Thus, higher leverage is expected to have a negative impact on R&D expenditure, while generally higher investments are predicted for sectors with unique products, since they are more dependent on such factors.

29

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

2.2.5 Capital Structure and Dividend Policy

Shareholders expect a return on their investments, either in the form of an increase in share prices or through dividends for placing equity at the disposal of a business. Since some organisations, such as pension funds and charities, periodically require revenues, dividends might be preferred that also reduces transaction costs (Watson and Head 2007). This reality might be overseen in the simplified MM argument of 1961, according to which, dividend policy does not matter in perfect markets. Dividend announcements, due to information that they contain, influence stock prices and thus the market value of a firm which has an impact on equity issues, according to the pecking order. This concept was approved by Asquith and Mullins (1986) on a sample of 168 publicly listed US companies from 1964 to 1980, who began to pay dividends.

Also Baskin (1989:27) argues that ―[d]ividends provide signals both to current and future earnings‖ and found that higher dividend payments in 1965 resulted in considerably higher debt levels in 1972. He (ibid.:31) observed that the level of dividend payouts is rather stable, since ―[s]erial correlation after twelve years is 0.714 and amounts to about 50% explained variance‖. In contrast, Antoniou et al. (2008) find a negative correlation of gearing and dividend payments for US firms only, while Lintner (1956) argues that

Copyright © 2010. Diplomica Verlag. All rights reserved.

dividends depend on profits and successful investment strategies that allow a stable compensation for shareholders. Also Rozeff (1982) argues that more investment opportunities lead to lower dividend payout ratios that, however, do not reduce agency costs, which are offset by higher transaction costs for debt issuance.

30

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Thus, according to the pecking order theory, dividends contain information about the future that makes dividend policy a ‗sticky‘ subject (Myers 1984; Baskin 1989), ―while capital spending varies over the business cycle‖ (Baskin 1989:32) that increases debt levels. According to Baskin (1989:31), ―the need to adhere to stable dividend policy appears to be much stronger than those motivating adherence to some statically defined optimal capital structure.‖ In a severe downturn, however, where profits and financial resources shrink, it is worth questioning whether a firm should stick with dividend policy or would be better selling undervalued assets to generate liquidity. An observation made by Graham (2000:1933) is that ―dividend-paying firms issue debt more conservatively than do non-dividend-paying firms, even though they presumably have less severe informational problems.‖ This implies higher leverage for nondividend-payers as is confirmed by Frank and Goyal (2009). Thus, firms who pay dividends do not only have less interests to pay, but would also be able to reduce dividend payments in bad times and thus are able to avoid financial distress, while for them it is easier to continue investments.

Another explanation is that non-dividend-payers have more growth opportunities, requiring new investments that are leveraged with external debt, while mature firms do not always have the possibility to invest in profitable projects. Baskin (1989:27) states that ―an increase in equity issues necessarily results in greater dividends, and greater

Copyright © 2010. Diplomica Verlag. All rights reserved.

dividends in turn give rise to a larger burden of personal taxation.‖ As a result, investors would prefer lower dividends in favour of faster growing share prices. While young firms with large growth potential do not have these problems with free cash flows (Stulz 1990), rational investors of mature firms expect dividends to avoid over-investment. Also large firms, who generally pay higher dividends, tend to expand more slowly than small ones, as observed by Baskin (1989:32), which implies higher dividend payouts.

31

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

2.3 .

A Final Comment

As the literature review shows, there are different ways of looking at things. Although the various approaches seem well grounded in rationality, they often come to contradicting conclusions and until empirical evidence is absent, none is approved. The complexity of the real world makes it necessary for academics to make simplifications that are rarely able to capture the full range of influencing factors. It might also be the case that human beings have a tendency to see what they are searching for and are likely to be biased by a selective perception (Gadamer 1980) that unintentionally results in biased findings. However, different outcomes and explanations also tell us that reality is as complex as suspected and that it is not static. This might be one reason why static capital structure models seem to be outdated. Also the pecking order does not give full explanations of managers‘ behaviour, which may simply be limited by the financing options available, as stated by Baskin (1989). However, there is unambiguous indication that financing matters, especially in turbulent times, where management failure is likely to affect society. This leads to the next section, where the methods applied in this study

Copyright © 2010. Diplomica Verlag. All rights reserved.

are explained.

32

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

3

Methodology

In section 2 a set of the most relevant literature with focus on the core aspects related to the capital structure i.e. profitability, liquidity, R&D and dividends, was presented, which now leads to the methodology applied to test the stated hypotheses. This section consists of ‗Data Sampling‘ that discusses to what extent the data can be assumed to be representative and, thus, refers to the reliability, followed by ‗Data Collection‘ which refers to the sources of the data. Section 3.3 is devoted to giving explanations to the variables chosen and how they are defined that ensures replicability. Finally, in ‗Hypotheses and Hypotheses Testing‘ the five research questions, and the applied statistical evaluation methods that allow a valid conclusion, are explained and justified.

3.1

Data Sampling

To conduct a reliable study on the impact of the capital structure on profitability, it was necessary to look for a large sample that is representative of the wider economy. The US market as a precursor of the free market policy (Yergin and Stanislaw 2002), where

Copyright © 2010. Diplomica Verlag. All rights reserved.

the latest financial crisis originated and quickly swapped from the financial sector to other industries, seems to be a good representation. To gain the best evidence of funding chosen by firms and their effects on profitability, the focus lies on large capitalised companies who have the most options in terms of their choices of funds. A sample that fulfils this requirement in order to apply a positivist approach, was found in the S&P 500 index, which, according to Standard & Poor‘s (2008a:1) ―with 33

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

approximately 75% coverage of U.S. equities is also an ideal proxy for the total market.‖ The index consists of 500 publicly traded companies, classified in 10 sectors, which enables to analyse industry-specific characteristics. This approach is similar to Baskin‘s (1989:28) sample of the Fortune 500 companies, which ―comprise a material component of the economy‖. Appendix A gives an overview of the industries and their respective proportion that is not equal for all sectors.

As with most indices, the S&P 500 is updated periodically. Because companies listed in the second quarter of 2009 were less affected by the credit crunch, the sample refers to those listings of January 2004; a time when the economy had recovered from the downturn after the dot-com bubble burst in 20012. Thus, the sample covers a five-year period, from 2004 to 2008.

3.2

Data Collection

This study is based on secondary data that was extracted from Datastream, similar to Antoniou et al. (2008), using the constituent list of January 2004 that was cross-checked with the official list of Standard & Poor‘s. Data of a random sample of firms was also compared with data published by The Wall Street Journal (2009a), where in some

Copyright © 2010. Diplomica Verlag. All rights reserved.

cases minor differences were observed, which probably arose from different accounting standards and rounding. However, the available data from Datastream can be considered as reliable.

2

http://www.marketwatch.com/investing/fund/SPY/charts?countryCode=US&submitted=true&intflavor=advanced&origurl=%2Ftools %2Fquotes%2Fintchart.asp&startdate=01%2F01%2F1999&enddate=01%2F01%2F2009&hiddenTrue=true&comp=Enter%20Sym bol(s)%3A&compidx=aaaaa~0&compind=aaaaa~0&uf=7168&ma=1&maval=50&lf=1&lf2=4&lf3=0&type=2&size=1&optstyle=1013 (accessed 08.05.09)

34

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

To ensure the correct meaning of ratios, preference was given to their calculation on the basis of the respective absolute values extracted from the balance sheets, income statements and cash flows.

3.2.1 Missing Values and Adjustments

Unfortunately, 17 of the 500 firms had 100 percent missing values that made it necessary to exclude them from the initial sample, so that the effective sample comprised 483 firms. Of the remaining firms, for 88 cases data was only partly available from 2005 to 2008; this, however, represents less than 8 percent of the overall data. A full missing value analysis can be found in Appendix B that is based on the balance sheets available for the respective firms. In contrast with Frank and Goyal (2009:23), missing values were not substituted through ―multiple imputation‖, since the present sample is very small, compared to Frank and Goyal, who observed 272,537 cases over 54 years. Furthermore, because mean-substitution for missing values, as applied by Baskin (1989), can result in distortions if one of two variables required for a ratio is mean-substituted, while the other is not, it was preferable to work with less data of high quality rather than a larger, more uncertain dataset, as suggested by Fama and French (1998). However, partly consistent with Kester (1986), the five-year mean of each

Copyright © 2010. Diplomica Verlag. All rights reserved.

variable was calculated, which built the basis from which ratios were derived.

Since not all firms have their fiscal year end on 31st December, it was necessary to adjust those cases that had not already been adjusted by Datastream. It was defined that firms who have their fiscal year end between January and 31 st May, were backward-adjusted, since the majority of the activities occur in the former year (e.g.

35

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

31.05.08 is equal to 31.12.07), while firms with their fiscal year end from 1st June onwards remained unchanged and thus were considered as having their fiscal year end in the same calendar year (e.g. 01.06.08 is equal to 31.12.08). Out of the 483 firms, 146 firms‘ fiscal year end was not in December, of which 40 were already adjusted by Datastream, 34 were backward-adjusted and 72 remained unchanged. Also cash dividend payments revealed from the cash flow statements needed to be adjusted to the respective accounting period to enable the comparability with variables from balance sheets and income statements. In total, 95 adjustments were made, of which the dividends paid by 36 firms were forward-adjusted for one quarter, and 15 firms‘ cash flows adjusted for two quarters, while another 29 were backward-adjusted for one quarter and 15 for two quarters. All of these firms were made up of those whose fiscal year end was not in December. Finally, it should be noted that the data was not inflation-adjusted due to a relatively short sample period and stable economy.

3.3

Variables and their Definitions

Different definitions and meanings of terms, especially financial ratios, can lead to a misunderstanding of findings by the reader. Thus, to ensure replicability and an

Copyright © 2010. Diplomica Verlag. All rights reserved.

unambiguous terminology, it is worth noting the variables and their definitions.

3.3.1 Leverage and Gearing

It should be anticipated that in this paper, the terms ‗leverage‘ or ‗gearing‘ are used synonymously, although it might not be technically correct. However, there is still a wide

36

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

range of definitions given to them in the literature. While Ross et al. (2008) defines them as

, they can also be expressed as

. Furthermore, questions arise about

which debt (long-term only or short-term as well) and which equity (market or book values) to apply. Baskin (1989), for instance, prefers total debt, because the total amount of borrowings matter and not the maturity, while Frank and Goyal (2009) investigate the effect of long-term, total debt, book and market value. In addition, market values are highly volatile (Ross et al. 2008; Gapper 2009; Taleb and Spitznagel 2009; Saigol 2009) – especially in the sample period considered here – giving preference to book values – ―despite their imperfections – because they attempt to measure the amount of capital actually obtained from various sources‖ (Baskin 1989:28). Thus, the gearing ratios here are defined as follows:

To overcome the effects of discrepancies of long-term and total debt, both values are considered for the majority of the analysis, however, priority was given to the total-debt value. As is evident from the above formula, ‗Total Liabilities + Shareholders‘ Equity‘ is

Copyright © 2010. Diplomica Verlag. All rights reserved.

defined as ‗Capital Employed‘.

37

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

3.3.2 Profits and Return

To investigate whether higher geared companies have higher returns, interest and debt payments as the main advantage for high leverage, according to the trade-off theory, must be considered. Therefore, in favour of ‗Net Income Available to Common‘ (NIATC), the ‗Earnings Before Interest and Tax‘ (EBIT) are disregarded, which are certainly higher for those with high gearing levels (DeAngelo and Masulis 1980). This, unconventionally, enables to see evidence of the real effects of higher gearing ratios on profitability and to measure shareholder wealth, the ‗Return on Capital Employed‘ (ROCE) is defined as:

Having ensured the correct use of the central variables, the next step is to further develop and define the hypotheses initially stated, with consideration of the proxies associated and the respective hypotheses tests applied.

Copyright © 2010. Diplomica Verlag. All rights reserved.

3.4

Hypotheses and Hypotheses Testing

One objective of this study is to provide results that are not only reliable, but also the subject of a high validity, as it is the aim of any researcher to provide certain conclusions whether the null hypothesis (no correlation) can be rejected or not.

38

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

3.4.1 H1: Capital Structure and Industry

Despite some theoretical models, the majority of existing literature indicates the existence of industry-specific gearing ratios. Thus, it was seen as sufficient to rely on descriptive statistics, such as mean and standard deviation, while taking into consideration the effects of different gearing ratio definitions. The proxies are, therefore, Gearing 1 and Gearing 2, as defined in section 3.3.1, whereas ‗AV‘ indicates the fiveyear average.

To provide an overall first impression of the S&P 500 landscape, the same descriptive measures are also applied to other relevant ratios that are used in other hypotheses. This is done in order to understand the impact of the industry also on profitability, liquidity, R&D and dividends. To simplify the comparison of the various variables, the same denominator is used for all proxies, i.e. capital employed. ROCE, NIATC to CE, R&D to CE and Dividend Payments to CE are the defined variables.

Since the five-year average of the respective variables may give a statistically misleading result and assumes that no changes occurred within this five-year period, line charts for all variables are applied to highlight their volatility, so that the effect of the

Copyright © 2010. Diplomica Verlag. All rights reserved.

industry becomes evident.

39

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

3.4.2 H2: Capital Structure and Profitability

While all the correlations between capital structure and liquidity, R&D and dividends affect profitability indirectly, this hypothesis is the core of the present book and merits a wider analysis.

The first step is to calculate the Pearson Correlation Coefficient (PCC) for Gearing 1, Gearing 2 and ROCE on the overall sample and then for the single industries, using the five-year average of those variables. Although the PCC is rarely criticised, it seems to be the best suitable approach to deal with a sample of that size (Kinnear and Gray 2009; Bryman and Bell 2007), because, unlike the covariance, it results in a value between -1 and 1 independent of the variables' unit, while Spearman's rank correlation and Kendall's tau statistics are subject to ordinal data (Kinnear and Gray 2009). However, Anscombe (cited in Kinnear and Gray 2009:399) emphasises that in some cases the Pearson correlation can give misleading values, which makes it necessary to verify ―whether there really is a robust linear relationship between two variables‖, according to the shape of the scatterplot. Since scatterplots indicate little if correlations are weak, they are not displayed in this analysis. The number of cases of each industry varies between 23 and 87, while only one sector (Telecommunication) consists of

Copyright © 2010. Diplomica Verlag. All rights reserved.

12 firms only. This, however, is sufficient for the application of the PCC (Price 2000). To increase the level of robustness and to identify a trend of a possible shift over the years, the PCC of Gearing 2 and ROCE is also calculated for the individual years.

According to the hypothesis stated, it is expected that a higher gearing level leads to lower profitability, implying that the ten lowest geared firms generate considerably

40

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

higher profits than the ten highest geared firms. To find evidence on whether to accept or reject this hypothesis, once again, the PCC is calculated, while the means of both groups are compared and statistically tested through an independent t-test. To verify the significance of different means and to improve the robustness of the results, the MannWhitney Test is also applied. Since a correlation analysis is inappropriate to identify whether the capital structure implies profitability or vice versa, which would require a regression analysis and which is also not the objective of the present research study, the same procedure is made for the ten most profitable and the ten least profitable firms of each industry too. If a strong correlation between capital structure and profitability can be observed, then the most profitable firms of the second sample are consistent with the ten lowest geared firms of the second sample.

To eliminate ambiguity, it is worth briefly noting the format of how the correlations are reported. Pearson‘s r in the following sections is reported as: rX(dƒ) = a, p < s, whereas, ‗X‘ indicates the respective industry (see List of Abbreviations), if the correlation does not refer to the whole sample size; dƒ = N – 2 degrees of freedom, according to Price (2000), with N = number of cases; and ‗p < s‘ the significance level ‗s‘. All correlations, including those regarding H2, are based on two-tailed significance levels, while unrealistic significance levels of 0.000 being substituted with 0.001, as suggested by

Copyright © 2010. Diplomica Verlag. All rights reserved.

Kinnear and Gray (2009).

3.4.3 H3, H4 and H5: Capital Structure, Liquidity, R&D and Dividend Policy

Similar to H2, the means to prove these hypotheses, the PCC is applied for the samples, subdivided in their industries. A full replication of the second hypothesis was

41

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

not considered to be particular useful, since this would go beyond the scope of this study. Nevertheless, it should be noted that these hypotheses are of equal importance for the survival of any company. Thus, the analysis is mainly limited to the evaluation of the correlation of Gearing 2 and the proxy of the respective hypothesis. To achieve robust results, both the five-year average and year-by-year correlations are calculated with reference to the ten lowest and ten highest geared firms, and the ten least and most profitable firms, where appropriate. Finally the data is compared with the results gained from the other hypotheses, which are analysed and discussed in the next

Copyright © 2010. Diplomica Verlag. All rights reserved.

section by means of the methodology explained.

42

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

4

Findings and Analysis

After defining the appropriate methods to evaluate the data collected, this part is devoted to the critical analysis and discussion of the findings in respect to existing literature. The attempt to focus on the most ‗burning‘ points is made in order to avoid an overly-detailed description of the data that might say little about the implications. As Miles and Shevlin (2001) point out, all variables are correlated to one another somehow, even if in the real world they have little in common. Hence, statistical results should be treated carefully, however sophisticated the statistical tools to prevent such bias are.

4.1

H1: The Impact of Industry on Capital Structure and Associated Variables

The discussion in section 2.2.1 indicates strong support for the view that the industry a business is in predefines the amount of debt and equity. In fact, the result revealed from descriptive statistics confirms this view to some extent. Figure 1 shows the five-year average (AV) of Gearing 1 (long-term debt to capital employed) and Gearing 2 (total debt to capital employed) of all ten industries represented in the S&P 500 index with the

Copyright © 2010. Diplomica Verlag. All rights reserved.

respective standard deviations at a confidence interval (CI) of 95 percent that ignores outliners.

43

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Figure 1: Gearing levels and standard deviations of the ten S&P 500 industries (five-year average)

Table 1: Gearing levels and standard deviations of the ten S&P 500 industries

Copyright © 2010. Diplomica Verlag. All rights reserved.

Although some industries operate with considerably more debt than others, the mean debt levels are well below 50 percent for all industries and thus consistent with Ross et al. (2008). It is evident that the highest gearing levels belong to Telecommunication and Utilities, while the lowest belong to Information Technology. However, the argument that faster growth results in more debt (Kester 1986) does not hold true when comparing the

44

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

industry means. It seems that one explanation is the degree of tangible assets, which is low for the IT sector (Graham 2000; Titman and Wessels 1988).

In addition, unique product policy, as is the case for most IT firms and energy suppliers, implies a lower debt ratio (Titman 1984; Titman and Wessels 1988; Graham 2000), but does not explain the high debt level of the Telecommunication sector, which has an unexceptionally large standard deviation. It might result from the small sample size of ten firms within this sector and indicates high market concentration that is consistent with Chevalier (1995b) and Graham (2000).

While the difference between Gearing 1 and Gearing 2 is generally small, it can be observed that Financials rely much more on short-term debt than their peers, as it is their business to borrow on short-term and lend on long-term (Wolf 2009) where possible, through easy access to overnight loans that lowers the costs, since the risk decreases as the maturity of a loan shortens. Consumer Staples and Industrials also use more short-term debt than other industries, but to less of an extent than Financials. One reason might be that the working capital is significantly higher than, for instance, in the IT or energy sector, where the time-span between production and consumption is smaller.

Copyright © 2010. Diplomica Verlag. All rights reserved.

The analysis of the mean-difference of Gearing 2 of firms that are still in the S&P 500 constituency list of June 2009 (338 firms) and those that were delisted after January 2004 (144 firms) shows a higher mean for the former (Appendix D). Since the results are not significant, mainly indicating other reasons for delists, this issue is not considered in the following analyses. Nevertheless, a statistically significant, but weak mean difference is found for profitability in terms of ROCE.

45

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

The examination of the development of the capital structures showed relatively stable gearing ratios for both Gearing 1 and Gearing 2. The increase of the debt level within the telecommunication sector resulted from a heavy debt issuance by AT&T in 2006 – a 100 percent increase of its long-term debt – and a 62 percent increase of the long-term debt of Verizon Communications in 2008. AT&T‘s continuously growing share price from 25 USD in January 2006 to some 42 USD at the end of 20073 indicates that managers still favour debt over equity, since it seems that (1) they have more information about future performance, according to the pecking order theory, and (2) want to keep their power (Stulz 1990).

Copyright © 2010. Diplomica Verlag. All rights reserved.

Figure 2: Changes in Gearing 1 from 2004 to 2008

Figure 3: Changes in Gearing 2 from 2004 to 2008 3

http://markets.ft.com/tearsheets/performance.asp?s=T%3ANYQ&vsc_appId=ts&ftsite=FTCOM&searchtype=equity&searchOption =equity (accessed 06.08.09)

46

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

While there is a slight increase in the gearing ratios from 2004 to 2008 of the overall S&P 500 sample, the only industry that increased the gearing level of some five percent, is the IT sector. This seems to be favoured by easier access to capital markets and costs of capital that become lower over time (Graham 2000; Fama and French 2004; Frank and Goyal 2009). However, a considerable difference between Gearing 1 and Gearing 2 in the change of debt levels could not be observed.

As Figures 4 and 5 show, Financials had a considerable increase of debt in absolute terms. Since this effect is not visible in the gearing levels, a similar increase in equity issues must have occurred that indicated prosperity and enabled expansion.

Copyright © 2010. Diplomica Verlag. All rights reserved.

Figure 4: Changes in long-term debt in billion USD from 2004 to 2008

Figure 5: Changes in total debt in billion USD from 2004 to 2008

47

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

The examination of the profitability, in terms of net income available to common (NIATC) over the five-year period, resulted in an overall slow decline in 2007 that became greater in 2008, as indicated in Figure 6.

Figure 6: Net income available to common (NIATC) in billion USD from 2004 to 2008

Figure 7: Operating income in billion USD from 2004 to 2008

The sharpest decline in profits was noted for Financials in 2008, while the

Copyright © 2010. Diplomica Verlag. All rights reserved.

telecommunications industry and some other industries seem not to have been affected by the crisis. Only Consumer Discretionary, Consumer Staples and Materials experienced a decline in profits, which is sharper for energy firms. Thus, despite some outliners, the profits in absolute values are relatively stable.

48

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

An analysis of the industry-specific ROCE, defined as net income available to common (NIATC) to capital employed (CE), showed that profits are highly industry-related (Figure 8). The relatively small profit margin of Financials and Utilities indicate high competition and less diversified products (Grant 2008; Chevalier 1995a). This might be one explanation for the low profit margins and relatively high debt levels that encouraged excessive loan offers by banks prior to 2008. The high standard deviation of 5.34 percent on a mean ROCE of 2.06 percent of telecommunication services (Table 2) should not be over-estimated, since the sample size is low. However, a

Copyright © 2010. Diplomica Verlag. All rights reserved.

mature and oligopolistic market explains the low profitability.

Figure 8: ROCE in terms of net income available to common (NIATC) to capital employed (five-year average)

Hypothetically, high investment opportunities are expected for firms with non-unique, non-standardised products that operate in markets with high growth potential, if investments are a means to ensuring future profits. As Figure 9 makes clear, IT and

49

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

healthcare firms have considerably higher R&D expenditure, if measured by proportion to capital employed.

Figure 9: Research and development expenses to capital employed (five-year average)

The five-year average is also confirmed by the respective absolute values in US dollars (Figure 10), where a general increase in R&D expenditure is evident, especially in Health Care and Information Technology. However, since the NIATC of those sectors is almost at the same level if considered in absolute values, Materials and Industrials

Copyright © 2010. Diplomica Verlag. All rights reserved.

(Figure 6), which invest surprisingly conservatively in R&D, show that the healthcare and IT sectors are much more dependent on innovation, as was found by Titman (1984), Titman and Wessels (1988) and Graham (2000).

50

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Figure 10: Research and development expenditure in billion USD from 2004 to 2008

Although dividend payments to capital employed is not appropriate to measure and analyse which companies pay the highest dividends – since not all sectors require the same amount of financial resources to generate an equivalent operating profit – it is evident from Figure 11 that dividend payouts are related to the industry to some extent. While dividend policy of firms within Consumer Staples is kept relatively flexible,

Copyright © 2010. Diplomica Verlag. All rights reserved.

dividends payouts of Utilities, Industrials and Consumer Discretionary seem to be fixed.

Figure 11: Dividend payments to capital employed (five-year average)

51

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

The lowest dividend payments in relation to the capital employed belong to the IT sector, where free cash flows are needed to finance future investments, while this does not apply for Consumer Staples, which includes food, tobacco and household products. Thus, higher dividend payments indicate saturated markets with less growth opportunities (Baskin 1989). The argument that a more flexible capital structure results in higher dividends, according to Graham and Harvey (2001) seems to apply for Consumer Staples, Telecommunication and Information Technology, while it is inconsistent with Frank and Goyal (2009), and Graham (2000), at least for industry-toindustry comparison.

Summarising the key facts, it can be argued that debt in all industries is lower than equity, while primarily Financials use higher amounts of short-term debt than other sectors. Despite an increase in debt of the financial sector, both long-term debt and total debt have been relatively stable during the sample period. Sectors with less volatile cash flows, such as Consumer Staples, Telecommunication and Utilities imply higher debt levels, as predicted by Bradley et al. (1984). In addition, industries within mature markets with more standardised products seem to be more likely to pay higher dividends, especially Consumer Staples, while they keep investments in R&D at a relatively low level. In contrast, industries with high growth opportunities, as in, for instance, IT and Health Care, require above-average R&D expenditure to ensure future

Copyright © 2010. Diplomica Verlag. All rights reserved.

profits. Thus, industry has a strong influence on the capital structure, which is consistent with the findings of Bradley et al. (1984), Ross et al. (2008) and Frank and Goyal (2009) that induces the rejection of the null hypothesis.

52

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

4.2

H2: The Existence of a Correlation Between Leverage and ROCE

As previously found, the industry strongly affects capital structure, which favours an industry-specific analysis for this and subsequent hypotheses. Nevertheless, a correlation of the overall sample size was applied, using the five-year average of Gearing 1 and ROCE, and Gearing 2 and ROCE as proxies. While for the former a significant correlation of r(483) = -0.177, p < 0.01 was found, the latter resulted in r(482) = -0.210, p < 0.01. This confirms the relevance of the short-term debt that seems to play a considerable role as a source of financing, which is in line with Baskin (1989) and Frank and Goyal (2009). Although these correlations are significant, they seem relatively small and were expected to be considerably stronger for some industries as shown in Table 2.

Copyright © 2010. Diplomica Verlag. All rights reserved.

Table 2: Correlation of Gearing 2 and ROCE (five-year average), ranked in order of Gearing 2 (lowest to highest)

The results show a relatively strong negative correlation of gearing and profitability for the IT sector, although these firms are the lowest geared ones. In contrast, profits of firms within the telecommunication sector, which operate with apparently high debt levels, have little to do with capital structure. Despite the fact that this market is 53

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

oligopolistic and that the sample size is rather small, the analysis shows that capital structure decisions are less relevant for firms that operate within an environment with high market concentration.

The strongest negative correlations are found for Energy, Utilities and smaller ones for Industrials and Consumer Discretionary. While Industrials and Consumer Discretionary use similar debt-equity ratios, they are very different to Energy and Utilities (Table 1) and support the industry argument of H1. Similar correlations were found for Gearing 2 and earnings before interest and tax (EBIT) to capital employed (Table 3) for the IT, energy, industrials and consumer discretionary sectors; and partly also for the materials sector. Since higher geared firms were expected to have a higher EBIT/CE ratio than lower geared firms, this confirms the effectiveness of the tax-shield to some extent, while negative effects of higher gearing levels result from other factors.

Copyright © 2010. Diplomica Verlag. All rights reserved.

Table 3: Correlation of Gearing 2 and EBIT to CE (five-year average), ranked in order of Gearing 2 (lowest to highest)

However, for Utilities it appears that the strong negative correlation, with ROCE as proxy (-0.459), remains untouched as no correlation could be found with EBIT/CE. This implies that firms do not issue debt, according to the debt-capacity as the trade-off theory and empirical evidence from Graham (2000) predict. An explanation might be the 54

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

few growth options that lead to their second-highest debt level, while profits are the second-lowest of the sample. Such a profile was identified by Graham (2000), Barclay et al. (2006) and Antoniou et al. (2008) for companies who accumulate more debt that, in this industry, does not allow offsetting interest payments from taxes to a large extent.

It seems that firms target an industry specific debt-equity ratio that has negative implications on profitability if the debt-levels are too high. However, a correlation does not express that an increase of equity of a company automatically implies an increase of its profits, because a number of factors such as corporate and business strategy or macroeconomic circumstances also influence the profitability of a company (Howell 2007). But it can be argued that a higher gearing level does lower profits.

To further develop this argument, the ten highest and lowest geared firms of each industry were analysed and demonstrated even stronger significant negative correlations for Energy, Consumer Discretionary, Materials and Utilities, as shown in

Copyright © 2010. Diplomica Verlag. All rights reserved.

Table 4.

Table 4: Correlation of Gearing 2 and ROCE (five-year average), of the ten lowest and highest geared firms of each industry, ranked in order of Gearing 2 (lowest to highest)

55

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Energy, Consumer Discretionary and Materials have slightly weaker correlations if calculated with Gearing 2 and EBIT/CE (rEN(18) = -0.455, p < 0.05; rCD(18) = -0.449, p < 0.05; rMA(18) = -0.480, p < 0.05). The mean-differences of ROCE are found to be significant with the independent sample t-test at a confidence interval equal to 95 percent and Mann-Whitney test for all these industries, except Utilities, due to low profit margins (Appendix G). Thus, lower geared firms generate higher profits or, in other words, more profitable firms use less debt, which may result from more free cash flows that enable higher stock repurchases.

Consistent results are observed in samples of the ten least and most profitable firms for Energy and Utilities (Table 5), both with a significant mean-difference in gearing ratio, according to the independent sample t-test at a confidence interval equal to 95 percent (Appendix F).

Copyright © 2010. Diplomica Verlag. All rights reserved.

Table 5: Correlation of Gearing 2 and ROCE (five-year average), of the ten less and most profitable firms of each industry, ranked in order of Gearing 2 (lowest to highest)

While for Energy and Materials the correlations of Gearing 2 and EBIT/CE are rEN(18) = -0.634, p < 0.01 and rMA(18) = -0.491, p < 0.05 respectively, the moderate negative correlations throughout are consistent with Kester (1986), Baskin (1989), Antoniou et al. (2008), and Frank and Goyal (2009). 56

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

The analysis of Gearing 2 and ROCE of the whole sample for every single year (Table 6), the aim of which was to find evidence of a possible trend, showed no significant correlation for financial and telecommunication services. For firms within the healthcare and consumer staples sectors, significant correlations of almost the same strength appear for 2007 only, but in the opposite direction. Information Technology – which increased its gearing level in 2007 – and Materials have strong negative correlations in the first three years of the sample which decline and become insignificant in 2007 and 2008, while it seems that for Industrials and Consumer Discretionary the negative correlation increases over time. Although the results are consistent with the

Copyright © 2010. Diplomica Verlag. All rights reserved.

average-proxy, a clear trend cannot be identified.

Table 6: Correlation of Gearing 2 and ROCE from 2004 to 2008, ranked in order of Gearing 2 (lowest to highest) – continued next page

57

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Table 6: Correlation of Gearing 2 and ROCE from 2004 to 2008, ranked in order of Gearing 2 (lowest to highest) – continued

As observed in previous studies, gearing ratio and profitability are negatively correlated, if significant correlation coefficients are found and it seems that recession proof industries are able to take on higher debt levels as Graham (2000) finds. On the other hand, firms from the industrial and consumer discretionary sectors appear to be less flexible and more sensitive to a recession, while for Financials, other factors are relevant that generate profits. Thus, the hypothesis that higher gearing lowers profitability

can

be

accepted

for

all

industries,

despite

Financials

and

Telecommunication.

4.3

H3: The Existence of a Correlation Between Leverage and Liquidity

Copyright © 2010. Diplomica Verlag. All rights reserved.

Since liquidity is essential to keep a firm‘s operations running, it is hypothesised that highly levered firms have less liquidity, which is likely to decline in a downturn. Figure 12 shows stable liquidity levels for the majority of sectors in terms of cash and cash equivalents generic (CCEG) in absolute values.

58

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Figure 12: Cash and cash equivalents generic (CCEG) in billion USD from 2004 to 2008

The only exceptions are Financials, which has above-average liquidity that increased in its absolute value after 2006, but which is constant if calculated in proportion to the capital employed, as Figure 13 represents.

Figure 13: Cash and cash equivalents generic (CCEG) to capital employed (CE) from 2004 to 2008

Copyright © 2010. Diplomica Verlag. All rights reserved.

The expected decline in liquidity in 2008 could not be observed, which is probably the result of a ‗time delay‘ due to sufficient reserves being available until the end of 2008. However, the relatively stable liquidity level has a positive effect on the five-year average considered to calculate the correlation with Gearing 2. The correlation coefficient applied on the full sample is r(480) = -0.383, p < 0.01. A single industry

59

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

analysis (Table 7) identifies only five sectors where higher gearing has a negative effect on liquidity.

Table 7: Correlation of Gearing 2 and CCEG to CE (five-year average), ranked in order of Gearing 2 (lowest to highest)

The highest negative impact of capital structure on liquidity was found for Materials and Industrials, while Health Care, Consumer Discretionary and Information Technology are less affected but remain at a significant level. These correlations are even stronger if applied to the sample of the ten lowest and ten highest geared firms (rCD(18) = -0.495, p < 0.05;

rHC(18) = -0.453,

p < 0.05;

rIN(18) = -0.652,

p < 0.01;

rMA(18) = -0.567,

p < 0.01) and to the sample of the ten least and ten most profitable firms (rCD(18) = -0.458, p < 0.05; rIN(18) = -0.520, p < 0.05; rMA(18) = -0.578, p < 0.01).

All these sectors have a similar NIATC/CE ratio and are likely to have pro-cyclical cash

Copyright © 2010. Diplomica Verlag. All rights reserved.

flows that make those firms more vulnerable in periods of recession. The analysis also shows that higher interest payments result in less liquidity with significant correlations for the IT sector of rIT(78) = -0.438, p < 0.01 and Industrials of rIN(56) = -0.438, p < 0.01. This is consistent with Graham (2000), since all of them have lower gearing levels than the sectors without any significant correlation, except Energy.

60

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

For Financials, however, the most significant factor that accounts for variability of liquidity among the variables observed, is the ROCE with rFI(74) = 0.700, p < 0.01. Other sectors, where liquidity depends largely on the ROCE, are Industrials with rIN(56) = 0.494, p < 0.01, Consumer Staples with rCS(34) = 0.449, p < 0.01, Utilities with rUT(35) = -0.370, p < 0.05, and Health Care with rHC(44) = 0.299, p < 0.05. While for the stated sectors similar correlations to liquidity were also found for the proxy EBIT/CE, the latter was found to be relevant for telecommunication services with rTE(8) = -0.733, p < 0.05 and Materials with rMA(31) = 0.425, p < 0.05. For firms within the remaining sectors, profits seem to be of low importance for their cash positions. This results are widely consistent with those of the sample of the ten lowest and ten highest geared and ten least and ten most profitable firms. Although in seven out of the ten sectors, higher profitability results in considerably better cash positions, only for Financials and Utilities is a significant mean-difference in Gearing 2 of the ten most and ten least profitable firms found, where the most profitable firms use less debt.

Thus, less debt implies better cash positions due to higher profitability, whereas the pecking order theory does not support this notion, since it predicts higher debt for more profitable mature firms. Consequently, these findings are consistent with Graham (2000)

Copyright © 2010. Diplomica Verlag. All rights reserved.

and Frank and Goyal (2009) and inconsistent with Baskin (1989).

4.4

H4: The Existence of a Correlation Between Leverage and R&D

Since higher gearing levels reduce free cash flows, it is hypothesised that a decline in revenues forces managers to reduce investments, especially in R&D. This would not have any immediate effect on profits, but may avoid financial distress or a reduction of

61

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

managers‘ wealth, whose average tenure became reduced to a few years (Reich 2009). Higher levered firms are also expected to have less R&D in general which declines further if financial distress arises. Figure 13, however, shows an almost stable R&D expenditure in relation to capital employed over the period 2004 to 2008 among all sectors, even in 2008.

Figure 13: R&D expenditure to capital employed (CE) from 2004 to 2008

This chart is largely consistent with the R&D expenditure in absolute values drawn in Figure 10 and suggests that investments are planned in advance, while cost cuts, if any, occur on a ‗time-delay‘. An increase in R&D expenditure as a debt-substitute to optimise tax-shield benefits in periods of higher profits, as argued by DeAngelo and Masulis (1980), could not be observed. However, it should be noted that profits of firms with high

Copyright © 2010. Diplomica Verlag. All rights reserved.

R&D expenditure were relatively stable.

Statistically, the correlation of the five-year average of Gearing 2 and R&D/CE, applied on the full sample size, resulted in r(381) = -0.329, p < 0.01 and thus represents a significant negative effect on investment activities. The single industry analysis of Table 8 shows generally negative correlations, but only for the IT sector was a significant negative correlation found, while it is positive for Consumer Staples. 62

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Table 8: Correlation of Gearing 2 and R&D expenditure to CE (five-year average), ranked in order of Gearing 2 (lowest to highest)

The argument that low levered firms have higher R&D expenditure, as stated by Andrade and Kaplan (1998), Graham (2000), and Frank and Goyal (2009), becomes weakly supported. Yet this applies only for the IT sector. For Consumer Discretionary and Consumer Staples, R&D seems to depend largely on ROCE, since relatively strong correlations were found that, however, point in the opposite direction (rCD(64) = -0.450, p < 0.01; rCS(24) = 0.664, p < 0.01). The results were even stronger in the sample of the ten least and the ten most profitable firms (rCD(14) = -0.569, p < 0.05; rCS(9) = 0.863, p < 0.01). In the sample of the ten lowest and highest geared firms, however, R&D expenditure in the consumer discretionary and healthcare sectors were found to be highly correlated to liquidity. Strong correlations of R&D and cash and cash equivalents generic (rCD(12) = 0.602, p < 0.05; rHC(15) = 0.779, p < 0.01) suggest that firms with more free cash flows invest more in R&D which, according to Stulz (1990), implies overCopyright © 2010. Diplomica Verlag. All rights reserved.

investment. Such evidence, however, was not found for other industries.

If higher geared firms make fewer investments in R&D, it can be expected that they may have more tangible assets. Although there are some negative correlations between R&D/CE and non-current assets (NCA) to CE (rEN(10) = -0.735, p < 0.01; rIN(43) = -0.339, p < 0.05; rIT(73) = -0.256, p < 0.05; rMA(31) = -0.387, p < 0.05), a positive 63

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

correlation between Gearing 2 and NCA/CE exists for most industries. As indicated in Table 9, for five out of the ten industries, significant correlations at the 0.05 level were found.

Table 9: Correlation of Gearing 2 and non-current assets (NCA) to CE (five-year average), ranked in order of Gearing 2 (lowest to highest)

Surprisingly, NCA/CE is strongly negative correlated to CCEG/CE for the majority of industries (rDS(82) = -0.577, p < 0.01; rCS(34) = -0.680, p < 0.01; rEN(21) = -0.774, p < 0.01; rHC(41) = -0.531, p < 0.01; rIN(56) = -0.682, p < 0.01; rIT(79) = -0.779, p < 0.01; rMA(31) = -0.498, p < 0.01; rTE(8) = -0.931, p < 0.01; rUT(33) = -0.615, p < 0.01). This implies an association of more liquidity with more intangible assets that, in turn, implies lower gearing levels due to insufficient collateral (Titman and Wessels 1988; Graham 2000; Antoniou et al. 2008; Frank and Goyal 2009). Significant correlations are also found for NCA/CE and ROCE, however, not with a homogeneous algebraic sign

Copyright © 2010. Diplomica Verlag. All rights reserved.

(rCS(34) = -0.455,

p < 0.01;

rIN(56) = -0.338,

p < 0.01;

rMA(31) = -0.545,

p < 0.01;

rTE(8) = 0.666, p < 0.05). Similar results are derived for the ten least and ten most profitable firms and for the ten lowest and ten highest geared firms.

Thus, if less profitable firms have more tangible assets, according to the trade-off theory, they can have less debt. Although valuable collateral is available that favours 64

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

debt, this only appears to be the case for telecommunication services. The pecking order theory, which predicts higher debt for more profitable firms, assumes more NCA if firms do not invest largely in intangible assets. Hence, Consumer Staples and Telecommunication, for instance, were expected to have a positive correlation of ROCE and NCA that, again, is only true for the telecommunication industry.

4.5

H5: The Existence of a Correlation Between Leverage and Dividends

As the probability of financial distress increases, managers are expected to take measures to ensure the survival of an organisation. Thus, it is hypothesised that managers and shareholders accept cuts in dividends in troubled times in order to ensure long-term sustainability. Figure 14 indicates that this is not the case, since a slight increase in dividend payouts can be observed, while Financials and the healthcare sector hardly reduced their dividends at all. Telecommunication firms‘ increase in dividends is caused by AT&T after a large issuance of debt in 2006 that led

Copyright © 2010. Diplomica Verlag. All rights reserved.

to higher profits.

Figure 14: Dividend payments in USD from 2004 to 2008

65

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Taking into consideration the full sample size, a very weak correlation of Gearing 2 and dividend payments (DP) to NIATC can be observed that equals r(480) = -0.090, p < 0.05, if based on the five-year average. In the single industry analysis, a significant negative correlation is found for Energy only, while none of the other sectors have correlations below the 0.05 significance level (Table 10).

Table 10: Correlation of Gearing 2 and Dividend Payments (DP) to net income available to common (NIATC) (five-year average), ranked in order of Gearing 2 (lowest to highest)

Slightly better results are seen for the proxies Gearing 2 and DP/CE. While a general tendency towards negative effects can be observed, as Table 11 shows, relatively strong negative correlations result for Energy and Materials that is positive for

Copyright © 2010. Diplomica Verlag. All rights reserved.

Consumer Staples.

Table 11: Correlation of Gearing 2 and Dividend Payments (DP) to CE (five-year average), ranked in order of Gearing 2 (lowest to highest)

66

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Since the five-year average gearing ratio of firms within the Consumer Staples sector is the third highest, an explanation might be that this industry generates the secondhighest profits in terms of NIATC/CE of all sample industries and limited growth opportunities. This view is consistent with the agency perspective, according to Jensen and Meckling, (1976), Rozeff (1982), Jensen (1986) and Stulz (1990). Despite Energy, the lower geared sectors, who also invest more in R&D, apply a more conservative dividend policy, since they do not have the problem of free cash flows, as stated by Frank and Goyal (2009) and Graham (2000).

The most impressive result is, however, the negative correlation of gearing level and dividend payments within the energy and materials sectors that, however, is consistent with Antoniou et al. (2008). While for both sectors, who operate with conservative debt levels and generate moderate profits, the dividend policy seem largely dependent on the gearing level, for all other sectors, other determinants seem to be more relevant. Table 12 shows the correlations of dividend payments and ROCE on the five-year average basis, which seems, in part, to be the answer to the question of which factor

Copyright © 2010. Diplomica Verlag. All rights reserved.

most determines dividend payments.

Table 12: Correlation of ROCE and Dividend Payments (DP) to CE (five-year average), ranked in order of Gearing 2 (lowest to highest)

67

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Despite Telecommunication, where the validity of the results can be questioned due to a small sample size, and Materials, the results of which have previously been discussed, the numbers in Table 12 indicate considerable positive correlations, especially for Consumer Staples and Financials. This indicates that dividends are less ‗sticky‘, as argued by Myers (1984) and Baskin (1989), especially for sectors with a higher meangearing ratio. Yet the results are even stronger when applied to the ten lowest and ten highest

geared

rCD(18) = 0.557,

firms p < 0.05;

(rIT(18) = 0.618, rMA(18) = 0.626,

p < 0.01; p < 0.01;

rEN(18) = 0.530,

p < 0.05;

rFI(18) = 0.677,

p < 0.01;

rCS(18) = 0.905, p < 0.01; rUT(18) = 0.586, p < 0.01), which are widely consistent with the ten least and ten most profitable firms and also with the equivalent EBIT/CE ratio as proxy. Although R&D expenditure of Consumer Staples is relatively low, a relatively strong correlation to dividend payments can be noted (rCS(24) = 0.664, p < 0.01). Thus, these results are consistent with Lintner (1956).

In summary, it can be argued that gearing has an effect on dividend payments for the energy sector and to some extent for Materials. For all other sectors, except Telecommunication, profits are more important in determining dividend payments, especially for Consumer Staples, Financials and Utilities – which have an aboveaverage gearing level – where significant correlations above 0.600 were found. Thus, higher geared firms seem to apply more flexible dividend policies due to meeting higher

Copyright © 2010. Diplomica Verlag. All rights reserved.

interest payments, since the interests cannot be offset from insufficient taxable profits. Furthermore it seems that these firms have less potential for growth, as argued by Baskin (1989), since the potential of an increase in share prices is limited (Stulz 1990). As the evaluation of all four proxies over the single years in Appendix C shows, higher gearing has negative effects on all defined variables in almost all industries. Only for

68

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Consumer Staples could a positive effect of higher gearing levels be observed, while Health Care, Financials and Telecommunication remain widely unaffected.

These findings suggest that the industry has a strong impact on various factors examined, not only on leverage. The key findings are outlined in the next section, which

Copyright © 2010. Diplomica Verlag. All rights reserved.

builds the basics for the management implications.

69

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

5

Conclusions

After analysis and extensive discussion of the results revealed from the data collected, the crucial facts are now summarised. Subsequently, a comment regarding the extent to which the initial research questions are answered is provided, which will restrict the recommendations in the next section.

To recap, the aim of this research is to find evidence of the extent to which capital structure influences profitability and thus the vulnerability of firms. The analysis of almost 500 firms of the S&P 500 constituency list of January 2004 showed that industry matters.

5.1

Capital Structure

While short-term debt plays a considerable role, total debt is generally less than equity for all ten sectors and, despite the financial sector, it seems that short-term debt is higher for those sectors that are forced to use more working capital. Gearing levels

Copyright © 2010. Diplomica Verlag. All rights reserved.

appear relatively stable over the sample period, but there are signs that access to favourable capital becomes easier the longer the boom lasts. In times of recession, capital structure reorganisation is restricted by the access to additional debt at reasonable interest rates and undervalued share prices, which make capital structure adjustments rather sluggish. A serious decline in profits for 2008 is noted for the

70

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

financial sector only, while, in general, all key factors observed (profitability, liquidity, investments, dividend payments) are relatively stable over the sample period.

5.2

Profitability

Low profit margins indicate high competition and mature industries that encourage high debt levels, which are also seen in industries with high market concentration. Aboveaverage gearing levels apply to industries with non-unique products and less volatile profits, while unique products increase the risk of financial distress. Firms with unique products within high concentrated markets, however, have a high standard deviation in gearing levels which underlines the importance of other factors like strategy and market power in determining leverage and profitability.

High leverage has an overall negative effect on profitability, especially for sectors with a lower industry-mean in gearing. This suggests that sectors with more growth potential use less debt, while higher debt levels belong to sectors with less growth opportunities and more standardised products. For instance, the utilities sector adopts debt levels that are far too high to offset interest payments from tax advantages. Thus, the trade-off model fails to explain such a behaviour that reduces shareholder wealth, but increases

Copyright © 2010. Diplomica Verlag. All rights reserved.

management control. Nevertheless, the overall consistency of the results for Gearing 2 with ROCE and EBIT indicate the effectiveness of the tax-shield for most industries. A clear trend of gearing and profitability over the sample period cannot be identified, while the most valuable correlations are found for Industrials, where the negative impact on profitability increases over time.

71

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

5.3

Liquidity

Cash and cash equivalents generic as proxy for liquidity show a stable picture for the sample period. Only Financials increased liquidity due to an increase in total liabilities. The expected decline in liquidity is not evident, suggesting sufficient reserves and a time-delayed liquidity problem. Cash positions appear negatively correlated to leverage for the overall sample size and especially for sectors with pro-cyclical cash flows. Thus, higher profit volatility has negative effects on liquidity, which generally implies less debt for the respective industries. As expected, the most significant factor that determines the variance of liquidity for most industries is profitability in terms of ROCE, and EBIT in rare cases, while results are mostly consistent. In seven of the ten sectors, higher profitability results in better cash positions. However, a significant mean-difference in leverage is observed only for Financials and Utilities, where the most profitable firms use less debt. Lower gearing results, therefore, in higher profitability and more liquidity.

5.4

Investments

A negative correlation for leverage and R&D on the full sample size is found that in the

Copyright © 2010. Diplomica Verlag. All rights reserved.

single industry analysis results negatively for the IT sector only, while it is positive for Consumer Staples. The findings suggest that R&D expenditure depends largely on ROCE and less on the gearing ratio. Nevertheless, the argument that firms adjust their non-debt tax-shields according to their profits, as argued by DeAngelo and Masulis (1980) cannot be confirmed. For Consumer Discretionary, Health Care and IT, strong positive correlations for R&D and liquidity indicate that more free cash flow promotes

72

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

investment and carries with it the danger of over-investment. High investments in R&D are found for firms with unique products, on which they depend, but they also have more growth opportunities. Intangible assets are positively correlated with higher liquidity and higher profits, while firms within less profitable sectors have more tangible assets. This is in contradiction with the trade-off theory, which predicts less debt, but in reality firms within those sectors have higher debt levels. Also the pecking order, which predicts higher leverage for firms with more intangible assets due to higher information asymmetry, fails on this point.

5.5

Dividends

Firms that have the potential to expand are found to pay lower dividends, while firms generally tend to pay out less the higher the debt level. Especially for the energy and materials sectors, dividend policy is mainly related to the degree of leverage, although they apply a conservative debt policy and generate moderate profits. For all other industries the results indicate a moderate correlation with ROCE that becomes stronger as the industry mean increases, except in the case of Telecommunication. This can be explained by the agency perspective, where in mature markets, a reduction of free cash flows through higher gearing and higher dividend payouts increases the efficient use of

Copyright © 2010. Diplomica Verlag. All rights reserved.

capital employed. Results also show that dividends appear less sticky for most industries than argued by Myers (1984), especially for higher geared sectors. Thus, industries with fewer growth opportunities are more highly geared, invest less in R&D and make dividend payments more dependent on their profits, which might transfer wealth from creditors to shareholders.

73

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

The consistency of all results to a large extent with the samples of the ten lowest and ten highest geared firms and the ten least and ten most profitable firms of each industry, strengthens the results of the full industry samples. Nevertheless, the overall effects of higher gearing levels among all industries, except Consumer Staples, are negative. Weak support is found for mean-differences of firms that are listed in the S&P 500 index in June 2009 and for those that are not included in the index. The trade-off and pecking order theories help to find reasons but are not satisfactory in explaining financing behaviour in full. It seems that organisations act in accordance with their competitors, but that the majority still attempt to act conscientiously, which belittles the impression that ―the corporation is a pathological institution, a dangerous possessor of the great power it wields over people and societies‖ (Bakan 2005:2). Corporations affect and are affected by the economy, but have to accept the rules of the game.

Due to largely missing data for R&D, the results are less reliable; this also applies to the small

sample

size

of

Telecommunication

and

limits

the

extent

to

which

recommendations can be drawn from these findings, which are discussed in the next

Copyright © 2010. Diplomica Verlag. All rights reserved.

section.

74

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

6

Recommendations

Until this point, the discussion and analysis has focused on the correlations between capital structure and profitability, including some major associated factors, and the likely consequences of highly geared firms in relation to an economic downturn. Based on these findings, the next step is to seek implications for management and corporate governors in favour of sustainable growth and the prevention of a possible contribution to a systematic risk.

Although firms within most sectors use conservative debt levels, empirical evidence clearly shows that deleveraging improves performance for most sectors. The strong industry-effect which affects all factors analysed, does not allow the generalisation of a single capital structure model to predict corporate financing decisions. Nevertheless, the most support is found for the trade-off theory, while dynamic rather than static models are more likely to capture real world dynamics, which future research is presumed to focus on. Similar to the study undertaken by Frank and Goyal (2009), a vast sample period is required. This is because when recessions shed light on how high leverage affects firms‘ performance, the results occur on a ‗time-delay‘. In addition, more

Copyright © 2010. Diplomica Verlag. All rights reserved.

consideration should be focused on industry-specific characteristics rather than following the idealistic approach of a universal model, given the strong impact of the industry on management behaviour.

The central question is, however, whether measures can be taken to reduce the negative impact on the economy and society, caused by inappropriate financing 75

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

behaviour. Results indicate an extremely complex field that makes it almost impossible to define large-scale and comprehensive regulations that would restrict free market policy. At the extreme, this would lead to a public regulated economy, a philosophy that dates back to the early years of the 19th century (Chambers 2009), where individuals and organisations have no incentive to innovate new products and services, since they are unable to improve their wealth. Economic growth would, thus, be impossible, because ―[i]n the absence of continuous technical change, there can only be short-term and transient growth periods‖ (Griffiths and Wall 2007:617).

On the other hand, protective measures must be taken, since it has already been noted by Adam Smith (1999:184-185) that ―[w]orkmen ... when they are liberally paid by the piece, are very apt to over-work themselves, and to ruin their health and constitution in a few years‖, which can lower economic growth if limits are not defined. This is also true for organisations, since the western economic order does not merit ‗good behaviour‘ due to increased competition (Bakan 2005; Wagner-Tsukamoto 2007; Reich 2009), which makes it necessary to encourage managers to adopt policies that make companies less vulnerable to financial distress in turbulent times. Tax authorities can contribute to this through promoting equity financing, but as results suggest, firms require a certain degree of benefits to take action.

Copyright © 2010. Diplomica Verlag. All rights reserved.

A second way to promote such behaviour is to give more power to corporate governors and non-governmental organisations in charge of monitoring management decisions, as it is proposed by Porter (2005). Arguments of advocates for free market policy, such as Friedman (2006) and Greenspan (2008), who proclaim the failure of any attempt to regulate markets, cannot be blindly accepted. The findings demonstrate that firms target industry-specific debt levels that may hold a systematic risk. Corporate governors and

76

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

monitoring agencies are therefore required to be independent and given the authority to intervene in a direct or indirect manner to prevent a potential systematic risk and major negative effects on economy and society.

It remains questionable, however, whether the Federal Reserve is the right organisation to prevent such a risk, as it is proposed by the US treasury secretary, Tim Geithner, for Financials (O‘Connor and Guha 2009). However, ―unless responsibility for systematic risk is concentrated in one agency, no one will be accountable for the failure to address it‖ (O‘Connor and Guha 2009:7). This may also support managers in resisting the pressure of share holders interested solely in short-term returns (Reich 2009), which lowers liquidity and increases the probability of financial distress.

The most vulnerable firms are those that have unique products, pro-cyclical cash flows and high debt levels. To reduce the risk of financial distress it is recommended to increase product portfolio diversification and to target conservative debt levels, while buffering liquidity to prevent bankruptcy. This, however, requires aligning management goals with those of shareholders in order to avoid over-investment. But to ensure that the interests of executives are in line with those of different shareholder groups, who all have their own agendas (Stern 2009), seems to be a difficult task. A valuable approach ―[t]o maximise long-term free cash flow‖, recommended by Mauboussin (2009:6), is to

Copyright © 2010. Diplomica Verlag. All rights reserved.

―properly manage its relationships with key stakeholders‖, which seems to be the key to ensuring long-term sustainability.

Financial crises are ‗part and parcel‘ of the western economic system; their occurrence was recognised long ago by Keynes and Marx (Eaton 1951). But any crisis is also a window of opportunity for implementing measures that improve the economic system.

77

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Unfortunately only past experience can be used as such, while crises are characterised by their uniqueness (Greenspan 2008) and markets by uncertainty, since certainty would erase trade. The ―central activity [of the financial industry] is creating and trading assets of uncertain value‖ (Wolf 2009:15) which requires non-financial firms as well as financials to prevent them from running out of liquidity. Based on the conclusions made, it is evident which of the analysed groups are the most vulnerable, since in a recession ―cash is king‖ (Greenspan 2008:529). Responsible behaviour does not only protect shareholders, it also contributes to economic stability, where the role of non-

Copyright © 2010. Diplomica Verlag. All rights reserved.

governmental organisations should be taken seriously.

78

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

7

Reflections

While the previous section attempted to discuss management implications from the preceding conclusions, it now comes to the point to comment on the fulfilment of the objectives of this study, and its strengths and weaknesses.

7.1

Objectives

An extensive analysis that provided reliable and valuable answers to all research questions to a large extent was not expected, since this depends on the availability and quality of the data and methodology chosen. In this sense, the objectives were widely fulfilled due to well defined and straightforward research questions at an early stage that could be developed step by step. The large amount of data that was evaluated required prioritisation in order to enable a focus on the central question and to reduce complexity. This is, however, at the expense of certain factors that may have a minor influence on core aspects, but it is unavoidable to overcome the complexity of the real

Copyright © 2010. Diplomica Verlag. All rights reserved.

world (Brown and Jackson 1990; Wagner-Tsukamoto 2008).

7.2

Strengths

An effective sample size of almost five hundred firms, combined with different approaches to analyse the effects of capital structure, provides a relatively robust basis. 79

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

With the exception of the telecommunication sector, all industries consisted of a sufficient sample size to provide reliable results in most correlations analysed. Since it appears that a single capital structure model is unable to capture the full range of financing behaviours, it was the right choice to omit this idealistic approach and to focus on empirical evidence of associations among variables over a partly turbulent period. This does not imply a devaluation of capital structure modelling, but it is a complex process that requires sufficient resources and excellent expert knowledge to provide acceptable results.

7.3

Weaknesses and Limitations

Although there are a number of strengths, it is important to outline weaknesses and limitations of this study, also to promote future research. Probably the most notable restriction is the focus on large capitalised US firms, the findings of which cannot fully be assigned to smaller firms, since size affects capital structure (Baskin 1989; Graham 2000; Graham and Harvey 2001; Akhtar 2005; Antoniou et al. 2008; Frank and Goyal 2009). The sample of S&P 500 firms, which was assumed to increase homogeneity and reliability of industry-comparison, has still an uncertain component that merits attention. The predefined sectors of Standard & Poor‘s (2008b:1) are broad categories ―of 10

Copyright © 2010. Diplomica Verlag. All rights reserved.

economic sectors aggregated from 24 industry groups, 68 industries, and 154 subindustries‖ that may explain some weak evidence of, for instance, Financials, which consists of commercial and investment banks, assurances and so on. Furthermore, the effects of the economic downturn, which reached most firms in a time-delayed manner, show only a few effects. An analysis of a ten-year period – five years pre and five years post crisis – would certainly provide more evidence of the long-term impact of high

80

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

leverage. It is also necessary to use better data of R&D expenditure, which was of poor quality and without any rational pattern of largely missing values, all of which limited the findings obtained.

Restrictions are also the underlying assumptions of unappreciated market values, inflation, firm age and the degree of a firm‘s diversification, all of which matter, according to Frank and Goyal (2009). It is assumed that, by focusing exclusively on S&P 500 firms, the significance of these factors was reduced. A weakness might also be the impact of mergers and acquisitions during the sample period (Appendix B) in respect to missing data, which is assumed to appear elsewhere if the firm was acquired or merged with a firm of the S&P 500 sample. In the case of a firm not merging or not being acquired from a firm within the sample, it was assumed that the performance for the sample period with missing values is constant. Thus, effects of distortion, if any, can be assumed to be insignificant. However, since the real impact is unknown, it represents a limitation. Finally it is crucial to point out that the Pearson correlation assumes a linearity that, at the extreme, may not be very effective. The objective, after all, was to identify correlations of a broad range that justify the use of the PCC, rather than examine abnormal cases.

A comment should be given to the literature reported here. The vast amount of directly

Copyright © 2010. Diplomica Verlag. All rights reserved.

and indirectly related previous research required the selection of a few core studies of the last decades. It represents a small proportion, but is required to maintain cohesiveness while attempting to give a comprehensive picture of existing literature. This refers also to the importance of dividend policy that unfortunately could not be appreciated in full. Also the application of a cross-sectional data analysis should be considered in related future projects.

81

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Bibliography Ackermann, J. (2009) ‗Smaller banks will not make us safer‘ Financial Times 29th July Available at: http://www.ft.com/cms/s/0/9aef3d00-7c6d-11de-a7bf-00144feabdc0.html (accessed 30.07.09) Akhtar, S. (2005) 'The Determinants of Capital Structure for Australian Multinational and Domestic Corporations' Australian Journal of Management 30(2):321-341 Andrade, G. and Kaplan, S. N. (1998) 'How Costly is Financial (Not Economic) Distress? Evidence from Highly Leveraged Transactions that Became Distressed' The Journal of Finance 53(5):1443-1493 Antoniou, A., Guney, Y. and Paudyal, K. (2008) 'The Determinants of Capital Structure: Capital Market-Oriented versus Bank-Oriented Institutions' Journal of Financial and Quantitative Analysis 43(1):59-92 Arnold, G. (2008) Corporate Financial Management 4th ed. Harlow: Financial Times Prentice Hall Asquith, P. and Mullins, Jr. D. W. (1986) 'Signalling with Dividends, Stock Repurchases, and Equity Issues' Financial Management 15(3):27-44 Atkins, R. and Guha, K. (2009) ‗In search of the exit‘ Financial Times 29th June p. 9 Bakan, J. (2005) The Corporation: The Pathological Pursuit of Profit and Power London: Constable Barclay, M. J., Smith Jr., C. W. and Morellec, E. (2006) 'On the Debt Capacity of Growth Options' Journal of Business 79(1):37-59 Baskin, J. (1989) ‗An Empirical Investigation of the Pecking Order Hypothesis‘ FM: The Journal of the Financial Management Association 18(1):26-35 Berens, J. L. and Cuny, C.J. (1995) 'The capital structure puzzle revisited' Review of Financial Studies 8(4):1185-1208

Copyright © 2010. Diplomica Verlag. All rights reserved.

Berger, P. G., Ofek, E. and Yermack, D. L. (1997) 'Managerial Entrenchment and Capital Structure Decisions' The Journal of Finance 52(4):1411-1438 Bradley, M., Jarrell, G. A. and Kim, E. H. (1984) 'On the existence of an optimal capital structure: Theory and Evidence' The Journal of Finance 39(3):857-878 Bray, A. et al. (2005) 'Payout Policy in the 21st Century' Journal of Financial Economics 77(3):483-527 Brown, C. V. and Jackson, P. M. (1990) Public Sector Economics 4th ed. Oxford: Basil Blackwell

82

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Bryman, A. and Bell, E. (2007) Business Research Methods 2nd ed. Oxford: Oxford University Press Bullock, N. (2009) ‗Lear funding affords hope to others in bankruptcy‘ Financial Times 9th July p. 35 Chambers (2009) ‗socialism‘ Chambers 21st Century Dictionary Available online at: http://www.chambersharrap.co.uk/chambers/features/chref/chref.py/main?query=sociali sm&title=21st (accessed 20.09.09) Chevalier, J. A. (1995a) 'Capital Structure and Product-Market Competition: Empirical Evidence from the Supermarket Industry' American Economic Review 85(3):415-435 Chevalier, J. A. (1995b) 'Do LBO Supermarkets Charge More? An Empirical Analysis of the Effects of LBOs on Supermarket Pricing' The Journal of Finance 50(4):1095-1112 Clemons, S. (2009) 'George Soros on How to Reregulate the Financial Markets' The Washington Note Available online at: http://www.thewashingtonnote.com/vlog/ (accessed 06.04.09) Copeland, L. S. (2005) Exchange Rates and International Finance 4th ed. Harlow: Financial Times Prentice Hall Davis, J. (2009) 'What can replace efficient market theory? Financial Times Fund Management 13th July p. 24 DeAngelo, H. and Masulis, R. (1980) 'Optimal capital structure under corporate and personal taxation' Journal of Financial Economics 8(1):3-29 Donaldson, G. (1961) Corporate Debt Capacity: A Study of Corporate Debt Policy and the Determination of Corporate Debt Capacity Boston: Harvard University Easterby-Smith, M., Thorpe, R. and Jackson, P. (2008) Management Research 3rd ed. London: Sage Publications Eaton, J. (1951) Marx against Keynes: A Reply to Mr. Morrison’s “Socialism” London: Lawrence and Wishart

Copyright © 2010. Diplomica Verlag. All rights reserved.

Fama, E. F. and French, K. R. (1998) ‗Taxes, Financing Decisions, and Firm Value‘ The Journal of Finance 53(3):819-843 Fama, E. F. and French, K. R. (1999) ‗The Corporate Cost of Capital and the Return on Corporate Investment‘ The Journal of Finance 54(6):1939-1967 Fama, E. F. and French, K. R. (2002) ‗Testing Trade-Off and Pecking Order Predictions About Dividends and Debt‘ Review of Financial Studies 15(1):1-33 Fama, E. F. and French, K. R. (2004) ‗New lists: Fundamentals and survival rates‘ Journal of Financial Economics 73(2):229-269 Fama, E. F. and Miller M. H. (1972) The Theory of Finance Chicago: Holt, Rinehart and Winston 83

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Financial Times (2009) ‗AT&T‘ Markets Available online at: http://markets.ft.com/tear sheets/performance.asp?s=T%3ANYQ&vsc_appId=ts&ftsite=FTCOM&searchtype= equity&searchOption=equity (accessed 06.08.09) Fischer, E. O., Heinkel, R. and Zechner, J. (1989) 'Dynamic Capital Structure Choice: Theory and Tests' The Journal of Finance 44(1):19-40 Frank, M. Z. and Goyal, V. K. (2003) 'Testing the pecking order theory of capital structure' Journal of Financial Economics 67(2):217-248 Frank, M. Z. and Goyal, V. K. (2009) ‗Capital Structure Decisions: Which Factors Are Reliably Important?‘ Financial Management 38(1):1-37 Friedman, M. (2006) 'Free Markets and the End of History' NPQ: New Perspectives Quarterly 23(1):37-43 Funnell, B. (2009) ‗Debt is capitalism‘s dirty little secret‘ Financial Times 1st July p. 13 Gadamer, H. G. (1980) ‗The Eminent Text and Its Truth‘ The Bulletin of the Midwest Modern Language Association 13(1):3-10 Gapper, J. (2009) 'Apple's network helps prevent a fall' Financial Times 25th June p. 13 Giles, C. (2009) 'Leaders should debate new growth model' Financial Times 8th July p. 25 Graham, J. R. (2000) 'How Big Are the Tax Benefits of Debt?' The Journal of Finance 55(5):1901-1941 Graham, J. R. and Harvey, C. R. (2001) 'The theory and practice of corporate finance: evidence from the field' Journal of Financial Economics 60(2/3):187-243 Grant, R. M. (2008) Contemporary Strategy Analysis 6th ed. Oxford: Blackwell Greenspan, A. (2008) The Age of Turbulence London: Penguin Books Griffiths, A. and Wall, S. (2007) Applied Economics 11th ed. Harlow: Financial Times Prentice Hall

Copyright © 2010. Diplomica Verlag. All rights reserved.

Guerrera, F., Mackenzie, M. and Stacey, K. (2009) ‗US Groups under pressure to show signs of recovery‘ Financial Times 7th July p. 15 Harris, M. and Raviv, A. (1990) 'Capital Structure and the Informational Role of Debt' The Journal of Finance 45(2):321-349 Howell, D. C. (2007) Statistical Methods for Psychology 6th ed. Belmont: Thomson Wadsworth Hughes, J. (2009) ‗In death do we part‘ Financial Times 9th July p. 11

84

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Jackson, T. (2009) 'Buffett and other blows to efficient market theory' Financial Times 29th June p. 18 Jensen, M. C. and Meckling, W. H. (1976) 'Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure' Journal of Financial Economics 3(4):305-360 Jensen, M. C. (1986) 'Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers' American Economic Review 76(2):323-329 Johnson, S. (2009) ‗How to breathe life back into bonds‘ Financial Times 20th July p. 9 Kester, W. C. (1986) 'Capital and Ownership Structure: A Comparison of United States and Japanese Manufacturing Corporations' Financial Management 15(1):5-16 Khor, M. (2001) Rethinking Globalization London: Zed Books Kinnear, P. R. and Gray, C. D. (2009) SPSS 16 Made Simple Hove: Psychology Press Klein, N. (2007) The Shock Doctrine: The Rise of Disaster Capitalism New York: Metropolitan Books Kraus, A. and Litzenberger, R. H. (1973) 'A State-Preference Model of Optimal Financial Leverage' The Journal of Finance 28(4):911-922 Lintner, J. (1956) 'Distribution of Incomes of Corporations Among Dividends, Retained Earnings, and Taxes' American Economic Review 46(2):97-113 Long, M. S. and Malitz, I. B. (1985) ‗Investment Patterns and Financial Leverage‘ in B. Friedman (ed.) Corporate Capital Structure in the United States Chicago: University of Chicago Press pp. 325-348 Maubussin, M. (2009) ‗Managing for shareholder value is still key‘ Financial Times Fund Management 3rd August p. 6 McLaney, E. and Atrill, P. (2008) Accounting: An Introduction 4th ed. Harlow: Financial Times Prentice Hall Miles, J. and Shevlin, M. (2001) Applying Regression & Correlation: A Guide for Students and Researchers London: Sage

Copyright © 2010. Diplomica Verlag. All rights reserved.

Miller, M. H. (1977) 'Debt and Taxes' The Journal of Finance 32(2):261-275 Miller, M. H. (1988) 'The Modigliani-Miller Propositions After Thirty Years' The Journal of Economic Perspectives 2(4):99-120 Miller, M. H. and Modigliani, F. (1961) ‗Dividend Policy, Growth, and the Valuation of Shares‘ Journal of Business 34(4):411-433 Modigliani, F. and Miller, M. H. (1958) 'The Cost of Capital, Corporation Finance and the Theory of Investment' American Economic Review 48(3):261-297

85

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Modigliani, F. and Miller, M. H. (1963) ‗Corporate Income Taxes and the Cost of Capital: A Correction‘ American Economic Review 53(3):433-443 Moore, J. S. and Reichert, A. K. (1983) 'An Analysis of the Financial Management Techniques Currently Employed by Large U.S. Corporations‘ Journal of Business Finance and Accounting 10(4):623-645 Myers, S. C. (1984) 'The Capital Structure Puzzle' The Journal of Finance 39(3):575592 Myers, S. C. (2001) 'Capital Structure' Journal of Economic Perspectives 15(2):81-102 O‘Connor, S. and Guha, K. (2009) ‗Cuncil has no teeth yet to resist the Fed‘ Financial Times 19th June p. 7 Porter, T. (2005) Globalization and Finance Cambridge: Polity Press Price, I. (2000) Common Statistical Tests: Critical Values Armidale: University of New England Available online at: http://www.une.edu.au/WebStat/unit_materials/c6_comm on_statistical_tests/critical_values.html (accessed 27.06.2009) Regan, M. P. and Stanton, E. (2009) ‗AIG, Fannie Mae, Freddie Mac Surge in U.S. Trading (Update 4)‘ Bloomberg 5th August Available online at: http://www.bloomberg. com/apps/news?pid=20601103&sid=aQW7FVe_hdDQ (accessed 20.08.09) Reich, R. (2009) Supercapitalism: The Battle for Democracy in an Age of Big Business London: Icon Books Ross, S. A. et al. (2008) Modern Financial Management 8th ed. New York: McGrawHill/Irwin Rozeff, M. S. (1982) 'Growth, Beta and Agency Costs as Determinants of Dividend Payout Ratios' Journal of Financial Research 5(3):249-259 Saigol, L. (2009) 'Currency volatility makes international acquisitions a minefield' Financial Times 22nd June p. 19 Sakoui, A. (2009) 'Help is at hand for companies in distress as crisis takes its toll' Financial Times Business Turnround 22nd June p. 5

Copyright © 2010. Diplomica Verlag. All rights reserved.

Shyam-Sunder, L. and Myers, S. C. (1999) 'Testing static trade-off against pecking order models of capital structure' Journal of Financial Economics 51(2):219-244 Simon, B. (2009) ‗Fresh doubts over Delphi‘s future cast long shadow‘ Financial Times 10th July p. 22 Smith, A. (1999) The Wealth of Nations Books I-III London: Penguin Books Soros, G. (2009) 'My three steps to financial reform' Financial Times 17th June p. 11

86

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Standard & Poor‘s (2008a) S&P 500 Factsheet New York: McGraw-Hill Available online at: http://www2.standardandpoors.com/portal/site/sp/en/us/page.category/indices/2,3,1, 0,0,0,0,0,0,0,0,0,0,0,0,0.html (accessed 21.05.09) Standard & Poor‘s (2008b) GICS (Global Industry Classification Standard) Available at: http://www2.standardandpoors.com/portal/site/sp/en/us/page.topic/indices_gics/2,3,1,7, 0,0,0,0,0,0,0,0,0,0,0,0.html (accessed 30.07.2009) Standard & Poor‘s (2009a) S&P 500 Constituent List 02.01.2004 Available online at: http://www2.standardandpoors.com/portal/site/sp/en/us/page.topic/indices_500/2,3,2,2, 00,0,2,2004,0,2,3,0,0,0,0,0.html (accessed 03.06.2009) Standard & Poor‘s (2009b) S&P 500 Constituent List 02.06.2009 Available online at: http://www2.standardandpoors.com/portal/site/sp/en/us/page.topic/indices_500/2,3,2,2, 00,5,2,2009,0,2,3,0,0,0,0,0.html (accessed 03.06.2009) Stern, S. (2009) 'How to encourage managers to act more like owners' Financial Times 7th July p.12 Stiglitz, J. E. (2008) ‗The Way Out‘ Time South Pacific 1(42):28-30 Strebulaev, I. A. (2007) 'Do tests of capital structure theory mean what they say?' The Journal of Finance 62(4):1747-1787 Stulz, R. M. (1990) 'Managerial discretion and optimal financing policies' Journal of Financial Economics 26(1):3-27 Taleb, N. N. and Spitznagel, M. (2009) ‗Time to tackle the real evil: too much debt‘ Financial Times 14th July p. 11 The Economist (2009) 'Greed—and fear' The Economist 22nd January Available online at: http://www.economist.com/displaystory.cfm?story_id=12957709&fsrc=nwlgafree (accessed 29.01.09) The Times (2008) 'Lehman Brothers collapse sends shockwave round world' The Times 16th September Available online at: http://business.timesonline.co.uk/tol/business/ industry_sectors/banking_and_finance/article4761892.ece (accessed 03.04.09)

Copyright © 2010. Diplomica Verlag. All rights reserved.

The Wall Street Journal (2009a) MarketWatch Available online at: http://www.marketwatch.com/quotes/ (accessed 15.06.09) The Wall Street Journal (2009b) ‗S&P 500 Index from 01/01/1999 to 01/01/2009‘ MarketWatch Available online at: http://www.marketwatch.com/investing/fund/SPY/ charts?countryCode=US&submitted=true&intflavor=advanced&origurl=%2Ftools%2Fqu otes%2Fintchart.asp&startdate=01%2F01%2F1999&enddate=01%2F01%2F2009&hidd enTrue=true&comp=Enter%20Symbol(s)%3A&compidx=aaaaa~0&compind=aaaaa~0& uf=7168&ma=1&maval=50&lf=1&lf2=4&lf3=0&type=2&size=1&optstyle=1013 (accessed 08.05.09) Tieman, R. (2009) 'Rescue remedy achieves global acceptance' Financial Times Business Turnround 22nd June p. 5

87

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Titman, S. (1984) 'The Effect of Capital Structure on a Firm's Liquidation Decision' Journal of Financial Economics 13(1):137-151 Titman, S. and Wessels, R. (1988) 'The Determinants of Capital Structure Choice' The Journal of Finance 43(1):1-19 Triana, P. (2009) 'Efficient market theory is not to blame' Financial Times Fund Management 22nd June p. 11 Wagner-Tsukamoto, S. (2007) ‗Moral Agency, Profits and the Firm: Economic Revisions to the Friedman Theorem‘ Journal of Business Ethics 70(2):209-220 Wagner-Tsukamoto, S. (2008) ‗Human Nature and Organization Theory: On the Economic Approach to Institutional Organization‘ Academy of Management Review 33(3):787-790 Watson, D. and Head, A. (2007) Corporate Finance: Principles & Practice 4th ed. Harlow: Financial Times Prentice Hall Witzel, M. (2009) 'How maths theories can destroy markets' Financial Times 23rd July p. 12 Wolf, M. (2009) ‗Reform of regulation has to start by altering incentives‘ Financial Times 24th June p. 15 Yergin, D. A. and Stanislaw, J. (2002) The Commanding Heights – The Battle for the World Economy New York: Free Press

Copyright © 2010. Diplomica Verlag. All rights reserved.

Younghwan, L. (2007) ‗The signalling aspect of pre-announcement insider trading and the stock price response to seasoned equity offerings‘ Journal of Academy of Business and Economics 1st March Available online at: http://www.thefreelibrary.com/The%20 signaling%20aspect%20of%20pre-announcement%20insider%20trading%20and%20 the...-a0172010691 (accessed 06.08.09)

88

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Appendix A – S&P 500 industries, January 2004 and June 2009

Sector Consumer Discretionary Consumer Staples Energy Financials Health Care Industrials Information Technology Materials Telecommunication Utilities Total

Firms listed in January 2004 N 87 37 23 83 47 59 83 33 12 36 500

% 17% 7% 5% 17% 9% 12% 17% 7% 2% 7% 100%

Firms listed in January 2004 and June 2009

Firms listed in June 2009 N 80 41 39 80 54 58 75 28 9 35 499

% 16% 8% 8% 16% 11% 12% 15% 6% 2% 7% 100%

N 57 29 16 46 37 46 53 24 5 27 340

% 66% 78% 70% 55% 79% 78% 64% 73% 42% 75% 68%

Copyright © 2010. Diplomica Verlag. All rights reserved.

Sources: Standard & Poor‘s (2009a) S&P 500 Constituent List 02.01.2004 Standard & Poor‘s (2009b) S&P 500 Constituent List 02.06.2009

89

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Appendix B – Missing values of dead firms, based on balance-sheet records

Copyright © 2010. Diplomica Verlag. All rights reserved.

Industry Consumer Discretionary (87 firms) Delist Merger Unknown Consumer Staples (37 firms) Delist Reorganized Unknown Energy (23 firms) Delist Merger Financials (83 firms) Delist Merger Aquired Unknown Health Care (47 firms) Delist Merger Unknown Industrials (59 firms) Delist Merger Unknown Information Technology (83 firms) Delist Merger Unknown Suspended Materials (33 firms) Delist Merger Unknown Telecommunication (12 firms) Delist Merger Aquired Utilities (36 firms) Delist Merger Total no. of firms

Firms with Missing Values from ... to ... 2004-2008 2005-2008 2006-2008 2007-2008 2

2 1

2

1

1

2008

4 1

2 1 1 3

1 4

1 2 3 2

2

1 1 3 1

1 1

5

2 2 1

1

1 2 1

1

1 1 3

1

1 1

1 1 1

3

2 4

2 1 3

1 1 1

1

1

1 1

2

1

1 1

1 1

1

17 (3.4%) of 500 Missing Values (no. of firms x no. of years) % in relation to 483 firms x 5 years = 2,415 values

14 (2.9%) of 483 56 (2.3%)

2 2 19 (3.9%) 23 (4.8%) of 483 of 483 57 (2.4%) 46 (1.9%) 191 (7.9%)

32 (6.6%) of 483 32 (1.3%)

90

Capital Structure and Profitability: S&P 500 Enterprises in the Light of the 2008 Financial Crisis : S&P 500 Enterprises in the Light of the 2008 Financial Crisis, Diplomica Verlag, 2010.

Appendix C – Significant year-by-year correlations of core factors with Gearing 2 Industry

Copyright © 2010. Diplomica Verlag. All rights reserved.

Information Technology

Year

2004 2005 2006 2007 2008 Energy 2004 2005 2006 2007 2008 Health Care 2004 2005 2006 2007 2008 Industrials 2004 2005 2006 2007 2008 Consumer 2004 Discretionary 2005 2006 2007 2008 Materials 2004 2005 2006 2007 2008 Financials 2004 2005 2006 2007 2008 Consumer 2004 2005 Staples 2006 2007 2008 Utilities 2004 2005 2006 2007 2008 Telecommuni- 2004 2005 cation 2006 2007 2008

Gearing 2 to ROCE

Gearing 2 to CCEG/CE

Gearing 2 to R&D/CE

Gearing 2 to DP/CE

r(79) = -0.458, p