• No results found

Capital structure and the influence on stock returns : an empirical research on the performance of portfolios based on debt-related fundamental ratios

N/A
N/A
Protected

Academic year: 2021

Share "Capital structure and the influence on stock returns : an empirical research on the performance of portfolios based on debt-related fundamental ratios"

Copied!
27
0
0

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

Hele tekst

(1)

1

Capital structure and the influence on

stock returns

An empirical research on the performance of portfolios based on debt

-related fundamental ratios

Author:

J. M. Wiegman

Student number:

10000631

Thesis supervisor:

J. J. G. Lemmen

(2)

2

I. Preface and Acknowledgements

Stock market participants are often chasing quick returns. This results in volatility in the financial markets which are often driven by emotions. Fear and greed are natural habits of human beings and they influence the participants’ investing behavior. From time to time, irrational speculation takes over true investing in the underlying company.

Since I have started myself with an investment strategy and portfolio, questions arise by others about the return of the portfolio. These short-term returns, driven by price fluctuations are only distracting the real purpose of investing. In my opinion, strong and profitable companies will eventually show better returns in the long-term. Even when the stock might underperform the market in a short-term (1-5 years) period.

In the classic 1949 Intelligent Investor, Benjamin Graham described that an investor should not necessarily be intelligent as the name of the book suggests. An investor should be disciplined and not led by other market participants’ behavior. An investor should value a company on basis of their annual reports, which consists of the financial statement, among others. These data should be used to determine the true firm value instead of the market prices as explained by the efficient market hypothesis. As a consequence, I want to back-test stock performance on the basis of fundamental ratios.

(3)

3

II. Abstract

This article describes a value investing strategy based on fundamental characteristics. The main focus is on fundamental ratios that are influenced by the firm’s capital structure. The article consists of background information on theories about the capital structure and the effects of monetary policy in the stock market. Finally, portfolios are constructed based on the outcome of the fundamental ratios. The performances are compared among 10 different portfolios. We can conclude that the ratios alone are not sufficient to divide winners against losers in the stock market.

Keywords:

Capital structure, Monetary policy, Stock returns, Portfolio performance, Value investing JEL Classification: G11

(4)

4

III. Table of Contents

I. Preface and Acknowledgements ... 2

II. Abstract ... 3

III. Table of Contents ... 4

1. Introduction ... 5

2. Literature review ... 7

2.1 Capital structure theories ... 7

2.2 Monetary policy and stock returns ... 11

2.3 Ratios ... 13 3. Methodology ... 15 3.1 Model ... 15 3.2 Data ... 15 4. Results ... 16 4.1 Regression analysis ... 16

4.2 Portfolio performance analysis ... 19

5. Conclusion ... 24

Reference list ... 25

(5)

5

1. Introduction

Recently many positive macro-economic figures about the US economy were published. The Federal Reserve has started debates about the timing of an interest rate hike. The Federal Reserve sets the Fed Funds rate, which is a very short interest rate affecting market rates and longer-term interest rates. General expectations indicate a first interest rate hike later this year or in the beginning of 2016 (CME Group, 2015). Since many companies use leverage in their capital structure it is interesting to investigate the consequence of normalization of interest rates on companies’ performance. In general, higher interest rates will increase the fixed charges of a firm so that it needs higher earnings to cover these interest expenses to leave the earnings-interest expense ratio unaffected.

The capital structure generally consists of debt and equity. There are in fact determinants of the capital structure, which differ across industries and companies. For decades, researchers have been examining theories about the capital structure which resulted in both generalities and contradictions. Some of these theories will be discussed in this article to obtain a clearer view of the effect of the capital structure on the value of a company. Besides, I will explore the effects of the expansionary monetary policies on market valuations. Extremely low interest rates, as shown in figure 1.1 and 1.2, have led to increasing debt issuance and thus changes in the valuation of companies.

The empirical research will focus on the fundamental ratios of 258 non-financial S&P500 companies. Fundamental ratios are calculated on data retrieved from the financial statements of companies. Thus, the ratios indicate the fundamental strength and efficiency of the company. It is solely a quantitative approach towards the valuation of a company. The ratios chosen are partly influenced by leverage and for this reason they are linked to the firm’s capital structure, among other fundamental characteristics. The chosen ratios for this article’s model are described in section 2.3.

The outcome of the ratios will divide the companies into different portfolios. The performance of the different portfolios will be backwardly tested for the period of 2003-2014. As can be seen with a closer look at the Fed Funds rate in figure 1.2, this period was subject to changing interest rates. Gradually increasing interest rates prior to the financial crisis of 2008 were followed by fast decreasing rates towards close to zero rates we have seen in the past 6 years.

(6)

6

Figure 1.1, Data: Federal Reserve

Figure 1.2, Data: Federal Reserve

In section 2 the academic content about the capital structure of companies will be described. This section captures the leverage ratio and the issuance of new debt since the historical low interest ratios in many advanced economies following the economic crisis of 2008. In this section also some attention is paid to previous studies with explanatory variables on the return of stocks. Fama & French (1992) for example have used firm size and book-to-market value as explanatory variables, besides the beta of Markowitz’ mean-variance efficient returns, to predict stock returns. I conclude this section with a brief introduction to the fundamental ratios. Section 3 covers the methodology and will disclose more information about the data and the two-step approach of this article’s research. In section 4 I will describe the results based upon the model of section 3, so that portfolios can be constructed on basis of the fundamental ratios. Furthermore, I will construct the portfolios and discuss their performance over the 2003-2014 period. Finally, in section 5 I will draw conclusions based on the theories and empirical findings.

0 5 10 15 20 25 1954 -07 1956 -07 1958 -07 1960 -07 1962 -07 1964 -07 1966 -07 1968 -07 1970 -07 1972 -07 1974 -07 1976 -07 1978 -07 1980 -07 1982 -07 1984 -07 1986 -07 1988 -07 1990 -07 1992 -07 1994 -07 199 6-07 1998 -07 2000 -07 2002 -07 2004 -07 2006 -07 2008 -07 201 0-07 2012 -07 2014 -07

Federal funds rate

Federal funds rate

0 1 2 3 4 5 6 7 2000 -11 2001 -05 200 1-11 2002 -05 2002 -11 2003 -05 2003 -11 2004 -05 2004 -11 2005 -05 2005 -11 2006 -05 2006 -11 200 7-05 2007 -11 2008 -05 2008 -11 2009 -05 2009 -11 2010 -05 2010 -11 2011 -05 2011 -11 2012 -05 201 2-11 2013 -05 2013 -11 2014 -05 2014 -11 2015 -05

Effective Fed funds rate

(7)

7

2. Literature review

2.1 Capital structure theories

The capital structure is important in determining the financial health of a company. In this article the emphasis will be on debt related features of a company, so a closer examination of the capital structure of a company is needed. First of all, we will look at the theories about the possible existence of an optimal capital structure. In their famous pioneering article Modigliani & Miller (1958) have examined a theoretical framework for an optimal capital structure. The optimal debt-to-equity ratio will maximize the value of the firm. However, in their first proposition they derive the conclusion that the total value of an unlevered firm is equal to a levered firm for any degree of debt, without taxation taken into account. Nevertheless, there are tax benefits on debt resulting in an increasing value from higher debt levels.

In the real world taxes do exist and so are the tax benefits of having debt on the balance sheet. Interest expenses are deducted from operating income before tax charges. For this reason lower taxable income follows from higher interest expenses. So there is a value increasing benefit of taxes. However, Kim (1978) states that the tax benefits of debt are only sufficient for a small amount of debt, because there are costs involved with debt as well, which are discussed in the article of Modigliani & Miller (1958). The costs of interest-bearing debt will be discussed in the next paragraph.

In their second proposition Modigliani & Miller derive a formula from the weighted average cost of capital theorem, which indicates that a higher debt-to-equity ratio requires a higher expected rate of return on equity. This is caused by the financial risk involved for equity holders of positive net-debt firms. This risk, where investors require a premium for, follows from bankruptcy and distress costs, among other less important causes. The risk of default lowers the total value of a company. The bankruptcy costs involved with unsustainable levels of debt lead to the conclusion that the firm’s total value is a concave function for the level of debt outstanding. This means that the value is increasing with debt until an optimal point, where after it slightly decreases, because the benefits will be offset by the costs.

For this reason Kim (1978) suggests that there is a unique global maximum in the function, which contains the optimal level of debt. A company should try to achieve this optimal point on behalf of their debt- and equity-holders. What is important to understand are the reasons for a firm to issue debt, when there are risks involved. I will continue with the underlying reasons for bond issuance next.

Leland (1994) has concluded that a higher optimal debt level results from an increase in the risk-free rate. A rise in the risk-free interest rate leads to increasing costs of debt financing.

(8)

8

When the spread between the risk-free rate and corporate bond yields remains equal, the higher risk-free rate leads to higher corporate interest expenses. Because of the tax benefits resulting from interest charges, the optimal debt level will be higher. The assumption of equal yield spreads could be violated when uncertainty leads to volatile markets and a flee to safety. The increase in spreads could lead to higher corporate interest expenses, even with a decreasing risk-free interest rate (Attinasi et al., 2009). Attinasi et al. (2009) describe the increasing spreads in the Euro Area following the turmoil and uncertainty during the recent crisis.

Though, intuitively it seems more likely to issue debt when interest rates are low. Indeed, the historically low interest rate levels seen in recent years have resulted in more issuance of long-term debt by companies as is shown in figure 2.1.

Figure 2.1: Total long-term debt outstanding, 264 non-financial US companies from the data set of this article. Source: Bloomberg

As short-term debt is part of the capital structure as well, it is important to look at the change in the term debt outstanding during the same period. Figure 2.2 shows the total short-term debt level for the companies in this article data set during the period of 2002-2013. From the two graphs we can see that the short-term debt actually decreased. This decrease, in contrast to the increase in long-term debt, suggests that the low interest rates of the last 5 years were an incentive for managers to issue longer-term debt over short-term debt. The reason behind this difference might be the historically low interest-rates, which are expected to normalize eventually. In order to benefit of the access to the relatively cheap money now, managers can choose for the issuance of debt with long maturities.

0 500000 1000000 1500000 2000000 2500000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Long-term debt

Long-term debt In mln $

(9)

9

Figure 2.2: Total short-term debt outstanding, 258 non-financial US companies from the data set of this article. Source: Bloomberg

Various researchers have done theoretical research about the variation in capital structures between firms, but little empirical research has been done so far. Titman & Wessels (1988) have done empirical research on theories of optimal capital structure. They describe a couple of determinants of the capital structure in closer detail, which include at least the following factors: firm size, volatility of earnings and the collateral value of assets.

Titman & Wessels found a negative relation between the firm size and short-term debt financing, likely caused by transaction costs. Short-term debt, which has to be paid within one year, has relatively high roll over costs for smaller firms. This makes the issuance of long-term debt over short-long-term debt more attractive.

The costs for debt and equity financing could differ greatly between firms. Overall, there is an order in which management prefers to use capital for investments. In the first place they use the retained earnings for positive net present value investment opportunities. Second, they will use debt-financing and if the first two are not accessible, equity will be issued (Myers, 1984).

Bradley et al. (1984) have found an inverse relationship of leverage ratios to the volatility of earnings. More stable flows of cash and earnings result in the ability of a company to take on more debt without risking default. They find explanatory power in the volatility of earnings for inter- and intra-industry differences in the leverage ratio. The results from these firm-specific factors imply that the optimal leverage ratio differs per firm. This difference enhances the theory of firm specific optimal capital structures. Besides, they show an inverse relation of the debt ratio to financial distress costs. This supports the existence of the concave function described earlier and thus a unique optimal debt-to-equity ratio for a specific firm as described in the first part of this section.

0 500 1000 1500 2000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Short-term debt

Short-term debt In mln $

(10)

10

Moreover, Barclay et al. (2003) point in their article at the different characteristics of the capital structure. The characteristics differ among companies in terms of leverage ratio, covenants and maturity. In this article the maturity of outstanding debt is neglected, as long-term debt is taken as a whole. Barclay et al. could, in accordance with other theories, not fulfill one general leverage equation to determine the optimal debt-to-equity ratio. So far it seems impossible to generally determine the perfect capital structure, because this differs across companies and industries. This makes the addition of other variables important in predicting the true value of a company. More about the collateral value will be discussed in a later part of this article, through the assets coverage ratio used in the model.

So far I have described some of the determinants of the capital structure and the theories about the possibility of an optimal debt-to-equity ratio. The empirical research will investigate the performance of portfolios on the basis of ratios which are influenced by the capital structure. For the empirical research to be interesting we need relations between the capital structure and the stock returns.

Empirical research done by Bhandari (1988) shows that the expected returns of common stocks is greater with a higher debt-to-equity ratio, controlling for beta and firm size. Thus, there is a relation between leverage and average return, which is positive in Bhandari’s research. This might be caused by the higher risk premium required by investors to hold equity stock of an heavily indebted firm. The higher risk premium is in harmony with the previous described proposition of Modigliani & Miller (1958). However, an investment in a company holding less debt than the optimal leverage ratio should ignore this required premium, as an incremental increase in the debt-to-equity ratio will actually positively influence the firm’s value. For small amounts of debt the relation suggests that it is preferred to issue some debt, because this will have the tax benefits without a big increase in default risk. This reasoning is valid for any debt level to the left of the optimal point, according to the theories.

Bhandari (1988) claims after all that the positive relation of leverage and average return is not only due to a higher risk premium required by investors. For this reason empirical research has to be done on the effect of the leverage ratios on the common stock return. In the paper of Welch (2004), he explained the effect of changes in stock prices on the capital structure. The changing market capitalization leads to differing debt-to-equity ratios over time. The higher market value of equity results in a lower debt ratio, which indeed results in a lower risk premium required by investors. Higher stock valuation will always lower the earnings yield, ceteris paribus. The lower earnings yield follows from a lower required risk premium. The earnings yield is defined as the net income over market capitalization (Net income / Market

(11)

11

Capitalization). Thus, it tells something about the percentage return that investors obtain at the current market price. The earnings yield is useful for comparison between the stock yield and corporate bond yields. These dynamics support the previous described proposition II of Modigliani & Miller (1958). Unfortunately Welch is not able to describe the underlying reasons for companies to issue debt.

2.2 Monetary policy and stock returns

Previously, I already concluded that interest rates do affect the capital structure (Leland, 1994). Besides, the link between the capital structure and stock returns has been examined. There are many different market interest rates that are influenced by the Federal Funds rate. In order to predict the stock performance by debt related fundamental ratios, there must be a relation between monetary policy and stock returns. For this reason I will discuss the possible influence of monetary policy on the stock market.

Monetary policy is set by the Federal Open Market Committee (FOMC). The FOMC targets the Federal Funds rate by operating in the market (Fed, 2015). This very short-term interest rate influences market rates (e.g. long-term rates, mortgage rates). Through several transmission channels the monetary policy conducted by the Federal Reserve could have real effects on the economy.

Bernanke & Blinder (1992) indicated that the Federal Funds rate is a good indicator of monetary policy and moreover, the interest rate is predicting changes in real economic variables. In principle the interest rate is a conventional tool, which could influence the real economy through a couple of transmission channels. One main conclusion of their article shows how monetary policy affects the composition of bank assets. Over time, these changes lead to different lending patterns affecting the money tightness in the economy. The money tightness affects the “price” of money (interest rates) and so they have an effect on the capital structure and investment decisions.

To continue with an explanation of the monetary policies as a reaction to the financial crisis, I will first introduce some empirical research about the potential effect of monetary policy on the stock market. From the findings of Bernanke & Kuttner in their 2005 paper we can conclude that unexpected changes in the monetary policy do lead to a change in the stock market. Bernanke & Kuttner have made a distinction between expected and unexpected policy changes to capture this effect. According to the researchers, the effect of monetary policy on the expected future excess returns is a main driver in this relationship. Logical reasoning gives some insight in the economic reasons behind the stock market change. Tight money in the economy will increase the riskiness of stocks, through higher interest costs and weakening of the balance sheet (Bernanke & Kuttner, 2005). In his article, Thorbecke (1997)

(12)

12

concludes in accordance with Bernanke & Kuttner, that expansionary monetary policy leads to a rise in stock prices. Since stock prices reflect the discounted expected future cash flows, Thorbecke suggests that monetary policy must increase these cash flows or lower the discount factor.

The recent crisis of 2008 has forced central banks to use unconventional monetary policy tools to recover the economy. Fawley & Neely (2013) have explained the differences in the Quantitative Easing (QE) policy of the major central banks. The QE policy was initiated to reduce uncertainty in the financial markets. However, it soon became a tool to hit inflation targets and to stimulate the recovery of the economy.

What we have seen in practice is the relatively low interest rates in advanced economies since 2010 (See figure 1.1 and 1.2). Next I will investigate the results on bond issuance with the lower interest rates. Leland’s (1994) conclusion that higher interest rates will lead to a higher optimal debt-to-equity ratio is violated by the fact that more corporate bonds were issued since the extremely low interest rates in recent years (Lo Duca et al. (2014). Lo Duca et al. show with their analysis that the QE program in the US has led to an increase in the issuance of corporate bonds with a lower average rating and shorter maturity. In the period from 2010 bond issuance in advanced economies was high compared to historical standards (Lo Duca et al., 2014). This trend could be graphically concluded from figure 2.1, in which the chart shows an increase in the absolute amount of long-term debt issued by the 264 non-financial US companies of this article’s dataset.

Fama and French (1992) have made a model to explain the cross-section of expected stock returns. They used firm size and book-to-market value as additional explanatory variables. These relatively easy to measure ratios have shown great results. Still, they are not linked directly to the debt structure of a company. In fact, Fama & French concluded that their size and book-to-market equity factors capture the leverage effect. For this reason I will make a cross section of expected stock returns by focusing on fundamental debt-linked ratios. These ratios will be described in closer detail in the next part.

After the careful examination in this section, we can conclude that the theories suggest an effect of the capital structure on the stock returns. In the empirical research I will test the performance of portfolios, which are composed on the basis of the companies’ value for a couple of debt ratios. These ratios are based on the company’s fundamentals and provide a clear view of the company’s financial strength, profitability and efficiency.

(13)

13

2.3 Ratios

In this part the ratios used in the empirical model will be discussed in more detail. The Asset coverage ratio is defined as the Tangible Assets scaled by the Total Assets. (Tangible Assets / Total Assets). It is expected that firms with valuable assets, which can be used as collateral, issue more debt. According to a model of Myers & Majluf (1984) there are costs involved in issuing debt by the fact that managers have more information than investors. This adverse selection problem is resolved by issuing secured debt. However, Leland (1994) found that the optimal leverage ratio is lower with protected debt and the possible gains will be less, which suggests that it will be unattractive to issue collateralized debt. Besides, Leland (1994) came up with an important distinction between short- and long-term debt. Short-term debt may only be rolled over if the value of the firm’s assets is sufficiently great. Short-term debt in fact behaves as protected debt. In contrast, long-term debt is often issued without any covenants. Earlier we already concluded from Titman & Wessels (1988) that smaller firms issue less short-term debt. The fact that smaller firms have less assets to use as collateral, strengthens the plausibility of Leland’s conclusion.

The ability of a company to issue secured debt is greater if assets on the balance sheet have a known value. For this reason the asset coverage ratio is linked to the capital structure and more goodwill and intangible assets could result in poorer performance when firms are in financial distress. Besides the ability to issue debt, a company in financial distress is able to sell tangible assets, which would indicate less fear of bankruptcy by investors. It is plausible to include the inverse asset coverage ratio (IAA) in the analysis, as measured by the Total Intangible Assets divided by the Total Assets.

Another debt-linked fundamental ratio which indicates the long-term health of a company, is the Long-term Debt as part of the Total Assets (LTD). Too much debt will eventually lead to default if the earnings are not sufficient to cover the interest expenses. In accordance with the concave function of Kim (1978) we might expect a worse performance for companies, which issued more debt than the optimal level. Because the determination of the firm’s optimal debt level is beyond the scope of this article, I will assume that more debt lead to a worse performance. For this reason I will not assume the concave function, but rather a linear function, where companies with less debt are regarded stronger.

Besides the long-term stability of a firm it must fulfill its short-term obligations as well. The other two explanatory variables will focus on the operating performance and cash flows of the company. To find the efficiency and profitability of a firm in the short run I will use the Free Cash Flow over Debt (FCF) and the Operating Income over Interest Expenses (OPI). A low OPI ratio makes investment in either debt or equity of the firm riskier, because the ability to

(14)

14

pay off the obligations becomes smaller. A negative ratio is even worse, as it means that the firms made an operating loss because interest expenses are non-negative by definition. In case a firm incurs no interest expenses at all (i.e. debt-free company), I will assume a ratio equal to the operating income for calculation purposes.

Finally, I will take the Return on Equity into account (ROE). This ratio shows the efficiency in which the company uses the investor’s capital. When stockholders’ equity is negative, the ratio could lead to biased results in the stock performance. Since I will use diversified portfolios in the performance comparison I ignore this fact in this article.

I will conclude with the ratios in formula form. The ratios are, together with the abbreviations, shown below:

𝐼𝑛𝑡𝑎𝑛𝑔𝑖𝑏𝑙𝑒 𝐴𝑠𝑠𝑒𝑡𝑠 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 (IAA) 𝐿𝑜𝑛𝑔−𝑡𝑒𝑟𝑚 𝑑𝑒𝑏𝑡 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 (LTD) 𝐹𝑟𝑒𝑒 𝐶𝑎𝑠ℎ 𝐹𝑙𝑜𝑤 𝐷𝑒𝑏𝑡 (FCF) 𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑛𝑔 𝑖𝑛𝑐𝑜𝑚𝑒 𝐼𝑛𝑡𝑒𝑟𝑒𝑠𝑡 𝑒𝑥𝑝𝑒𝑛𝑠𝑒𝑠 (OPI) 𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒 𝐴𝑣𝑔 𝐸𝑞𝑢𝑖𝑡𝑦 (ROE)

The use of the above mentioned ratios solely differs companies on the basis of their relative financial health and profitability. A very important determinant in investment decisions is the current market valuation of the stock. After all the stock return is a function of the price at which the stock was purchased. Even the strongest companies could underperform in a one year period because the stock price is already highly inflated due to high expectations by investors. Fama & French (1992) already used the price-to-earnings ratio (PE) in a cross-section of stock returns. In their article the use of this ratio, among others, showed great results.

In my empirical research, I mainly focus on the debt-linked ratios, but to take into account the current market valuations for each period, I use the PE ratio as an additional ratio in section 4.

(15)

15

3. Methodology

3.1 Model

The model is designed in a two-step approach. First of all, I will regress the firm’s scores for the different ratios on the stock return. The stock price return is calculated as the percentage change over a year. Dividends are included, because they are important for the total return of investors. The stock prices I used, are the adjusted closing prices of the first trading day of April following the fiscal year end. This conservative approach assures us that the financial statements are published, so that the returns are calculated after the portfolio could be constructed. Regulations by the SEC oblige companies to file their 10-K statements within 90 days of the fiscal year end (Security and Exchange Commission, 2015).

Second, I will construct portfolios based on the scores of the explanatory variables. For every ratio the firms will be listed on their score and given a rank between 1 and 258 according to their position. The scores of the individual ratios will be summed to get a relatively clear distinction between strong companies and weak companies on the basis of debt-related ratios. With 258 companies in the analysis, I will drop the eight worst performing companies for each year. As a consequence, each portfolio consists of 25 companies to get a total of ten portfolios. Financial theories suggest that a portfolio of 25 companies is well diversified and diversifying benefits diminish significantly after 25-30 companies (Markowitz, 1991). The investment strategy is rather easy to implement. A portfolio will be constructed and kept for one year. After a year new financial statement data are available and the process is redone. Since portfolios are rebalanced every year, the ‘best’ portfolio could consist of many different companies during the period of interest. However, it will be named portfolio 1, portfolio 2 etc.

The performance of the portfolios will be evaluated in section 4. In the evaluation the return of the individual portfolios will be calculated for each year. Furthermore, I will calculate the standard deviations per year per portfolio. With this information I can calculate the Sharpe ratio for each portfolio to give a mutual comparison. The Sharpe ratio will fit as best, since the portfolios can be constructed as the only investment portfolio of an investor.

3.2 Data

The data used in this article are retrieved from Bloomberg. The data consist of 258 non-financial companies currently part of the S&P500 index. The data include yearly figures from these companies’ financial statements at the end of the fiscal year from 2002 until 2013. The data are used to calculate the earlier described fundamental ratios.

(16)

16

4. Results

4.1 Regression analysis

In the regressions I use all of the 258 non-financial companies in the data set. The regressions are run over the total period, which makes the total sample size in each regression 2838. Equation 4.1 is used to calculate the coefficient of each ratio:

𝑅𝑖,𝑡+1= 𝛼 + 𝛽1∗ 𝐼𝐴𝐴𝑖,𝑡+ 𝛽2∗ 𝐿𝑇𝐷𝑖,𝑡+ 𝛽3∗ 𝐹𝐶𝐹𝑖,𝑡+ 𝛽4∗ 𝑂𝑃𝐼𝑖,𝑡+ 𝛽5∗ 𝑅𝑂𝐸𝑖,𝑡+ 𝜀𝑖,𝑡 (4.1)

Where 𝑅𝑖,𝑡+1 is the stock return of the 𝑖𝑡ℎ company in year t+1 calculated on basis of the adjusted closing price of the first trading day of April in year t+1 and the adjusted closing price of the first trading day in April one year further.

Returns

Coef.

Std. Err.

t

P>|t|

[95% Conf. Interval]

FCF 0.00 0.00 -0.68 0.50 0.000 0.000 ROE 0.00 0.00 -0.47 0.64 -0.002 0.001 OPI 0.00 0.00 0.28 0.78 -0.001 0.001 LTD 0.12 0.05 2.35 0.02 0.020 0.068 IAA -0.14 0.04 -3.87 0.00 -0.209 -0.068 Cons 19.75 1.53 12.95 0.00 16.764 22.745

Table 4.1.1, Regression results based on a 11-year period (2003-2004) of 258 non-financial US companies. FCF stands for Free Cash Flow over Total Debt, ROE for Return on Equity, OPI for Operating Income over Interest Expenses, LTD for Long-term Debt over Total Assets and IAA stands for Intangible Assets over Total Assets. Data: Bloomberg

The results for the regression ran over the total test period (2003-2014) are shown in table 4.1.1. For the 258 non-financial US companies used in this sample, the only significant effect on a 5% level of the ratios on the stock returns are the Intangible Assets over the Total Assets (IAA) and the Long-term Debt over the Total Assets (LTD), whereas the coefficient of the IAA ratio is significantly negative at the 1% level. The IAA ratio shows the expected result, where increasing intangible assets on the balance sheet lead to a decrease in the rate of return of the company’s stock. An increase of one percent point in the IAA ratio shows a decrease of the stock return by 0.138 percent point. However, long-term debt increases the stock return in this sample. This effect could support the theories described in section 2 about an optimal capital structure and the concave function of debt on value. According to these theories, some debt would indeed increase the firm’s value until the optimal debt-to-equity ratio. In the regression in table 4.1.1 I have only used a linear relationship, thus the optimal point cannot be described. What follows from the regression is that an increase in long-term debt will increase the stock return. In this case the bankruptcy and distress costs incurred by a highly

(17)

17

levered firm are ignored, which is not plausible according to some theories and practical experience where firms are declared bankrupt after default on their debt.

Returns

Coef.

Std. Err.

t

P>|t|

[95% Conf. Interval]

LTD 0.20 0.11 1.82 0.07 -0.015 0.417

LTD² 0.00 0.00 -1.17 0.24 -0.005 0.001

IAA -0.14 0.04 -4.03 0.00 -0.214 -0.074

PE -0.10 0.02 -5.53 0.00 -0.140 -0.067

Cons 23.18 1.84 12.01 0.00 18.507 25.708

Table 4.1.2, Regression results based on a 11-year period (2003-2004) of 258 non-financial US companies. LTD stands for Long-term debt over Total assets, LTD² is the squared value of LTD and IAA stands for Intangible assets over Total assets and PE is the price-to-earnings ratio. Data: Bloomberg

In table 4.1.2 a little side step is made in an attempt to capture the concave function as described in the theories. The LTD ratio is squared and regressed, together with the LTD, IAA and PE ratio on the stock returns. The LTD² is not even significantly different from zero at the 10% level, so in this data set I cannot demonstrate the existence of a concave function for the value of a firm to the level of debt.

Returning to table 4.1.1 shows us that the short-term ratios used in this article (FCF, ROE, OPI) are not significantly affecting the stock returns. This could be caused by the relative heavy fluctuations in the stock markets in short-term periods. Many studies about investor overreaction in the stock markets have been done. For example, Bondt & Thaler (1987) have examined the winner-loser effect and they conclude that this effect cannot be attributed to risk changes. The winner-loser effect describes the convergent movement in stock markets, where winners in one year will be the losers in the following year.

Moreover, Baker & Wurgler (2006) examined cross-section returns and conclude that investor sentiment has a significant effect on stock returns. For this reason, short-term fundamental ratios as used in this article could be insignificant due to sentiment in particular years during the test period. Besides, the difference between the relative usefulness of the long-term debt ratios compared to the short-term debt ratios in my model, could result from the conservative approach towards the calculation of the stock returns.

The volatility of the S&P500 during the period 2002-2014 increased dramatically during the recent crisis as can be shown in figure 4.1. In my opinion these heavy fluctuations are non-fundamentally based and for this reason the stock performance over a one-year period ignores the short-term strength and profitability of a company. This could have led to the insignificant short-term fundamental ratios used in this article.

(18)

18

Figure 4.1, VIX index to measure the volatity of the S&P500 index from 2002-2004. Data: CBOE

In the previous analysis, market valuations are neglected. Valuations at the beginning of the year could influence the stock performance greatly. As described before, investors overreact and irrationally valuate companies in the short-term. Thus, strong and reliable companies’ stock can underperform in a particular year, because of the high valuation at the beginning of the year. For this reason I will include the price to earnings ratio (PE) as measured by the market capitalization divided by net income. For companies with negative outstanding equity shares, I assume a market capitalization of zero. This may not be conservative in a way, but prevents good scores for loss-making companies.

The PE ratio is added to equation 4.1 to obtain equation 4.2 as shown below:

𝑅𝑖,𝑡+1= 𝛼 + 𝛽1∗ 𝐼𝐴𝐴𝑖,𝑡+ 𝛽2∗ 𝐿𝑇𝐷𝑖,𝑡+ 𝛽3∗ 𝐹𝐶𝐹𝑖,𝑡+ 𝛽4∗ 𝑂𝑃𝐼𝑖,𝑡+ 𝛽5∗ 𝑅𝑂𝐸𝑖,𝑡+ 𝛽6∗ 𝑃𝐸𝑖,𝑡+ 𝜀𝑖,𝑡 (4.2)

Returns

Coef.

Std. Err.

t

P>|t|

[95% Conf. Interval]

FCF 0.00 0.00 -0.68 0.49 0.000 0.000 ROE 0.00 0.00 -0.52 0.60 -0.002 0.001 OPI 0.00 0.00 0.30 0.76 -0.001 0.001 LTD 0.08 0.05 1.64 0.10 -0.017 0.186 IAA -0.14 0.04 -3.93 0.00 -0.210 -0.070 PE -0.10 0.02 -5.48 0.00 -0.139 -0.066 Cons 23.18 1.64 14.13 0.00 19.967 26.403

Table 4.1.3, Regression results based on a 11-year period (2003-2014) of 258 non-financial US companies. FCF stands for Free cash flow over total debt, ROE for Return on Equity, OPI for Operating income over Interest expenses, LTD for Long-term debt over Total assets and IAA stands for Intangible assets over Total assets. Data: Bloomberg

In table 4.1.3 the PE ratio is included in the regression analysis. The results show that a one point increase in the price-to-earnings ratio at the end of the company’s fiscal year leads to a decrease of 0.103% in the stock return during the next year (April-April). The coefficient is

0 20 40 60 80 100

VIX index

VIX index % per year

(19)

19

significantly negative at the 1% level. For this reason I will use the PE ratio as an additional variable in the construction of portfolios in the next part.

By adding the PE ratio, the LTD ratio is not significantly different from zero at the 10% level compared to the regression in table 4.1.1. However, I will still use this variable in the portfolio construction, because of the close link to the firm’s capital structure as described in section 2.

4.2 Portfolio performance analysis

In the first analysis I have constructed portfolios consisting of 25 companies by using all five of the described ratios. The results are rather counterintuitive. In table 4.2.1 the results are shown. The first portfolio returned an average of 15.43% per year. Though, portfolio 10 returned an average of 22.12% a year. The fact that financials are excluded in the sample and the relatively smallness of the sample data probably influence these results. Financial companies were hit hard by the economic crisis of 2008 and so did their stock performance. The S&P financials index dropped from 485.42 points to 171.28 in a one year period from April 1, 2008 till April 1, 2009, which is a decrease of 64.72% (Standard & Poor, 2015). The average of the 10 portfolios in the same period turned out to be -31.01%, which is less than half of the financial companies returns in the same year. For a better comparison I will compare the volatility of the 10 portfolios over the 11-year test period. Furthermore, I use the Sharpe ratio to compare the relative performance.

As can be seen in table 4.2.2, there is a trend in the yearly volatility of the constructed portfolios. Portfolio 1 has a yearly volatility of 19.76% compared to the 27.63% volatility of portfolio 10. It turns out that the riskiness, as measured by the volatility, increases with the lower scoring companies in my analysis. To conclude this analysis, we I compare the Sharpe ratios of the different portfolios. The Sharpe ratio measures the average return for one percentage point of volatility. The Sharpe ratio of portfolio 1 and 2 are the only ones below 0.80, where portfolio 4 has a Sharpe ratio above 1. The fact that the portfolios in the middle (3-7) perform relatively well might be due to the fact that some debt on the balance sheet increases the firm’s value and the ability for managers to invest in growth opportunities.

(20)

20

2003-2004 2004-2005 2005-2006 2006-2007 2007-2008 2008-2009 2009-2010 2010-2011 2011-2012 2012-2013 2013-2014 Portfolio 1 35.08 5.64 27.95 14.19 3.21 -28.00 46.53 24.36 10.41 6.32 24.05 Portfolio 2 34.27 13.44 25.60 18.29 1.22 -38.05 38.62 28.56 4.70 18.78 27.37 Portfolio 3 32.22 22.53 24.22 11.28 9.75 -26.22 39.25 28.81 -0.04 22.76 30.33 Portfolio 4 41.55 24.04 32.27 23.61 15.57 -33.68 41.56 22.28 7.30 23.61 32.20 Portfolio 5 40.83 18.14 31.86 23.49 9.31 -27.88 57.94 23.72 0.16 14.20 30.05 Portfolio 6 40.30 17.19 24.56 22.03 17.26 -34.14 41.13 26.75 -6.01 22.03 25.88 Portfolio 7 48.51 27.01 28.30 20.75 -1.15 -31.64 42.22 35.27 2.93 20.44 20.94 Portfolio 8 33.26 23.36 20.78 21.36 1.78 -33.88 70.57 28.54 5.61 28.59 22.16 Portfolio 9 42.01 8.71 23.89 24.46 4.01 -29.93 59.69 23.15 7.03 28.67 22.61 Portfolio 10 71.81 31.07 30.88 28.25 -8.46 -26.66 52.35 14.88 -3.35 24.39 28.16 Table 4.2.1, yearly portfolio returns including the P/E ratio as a variable in the portfolio construction. Portfolio 1 consists of the 25 best scoring companies according to the portfolio construction strategy, portfolio 2 consists of stocks 25-50 in the list, etc. Data: Bloomberg

Table 4.2.2, Average returns, volatility per year. Sharpe ratio as measured by the average return per year divided by yearly volatility. Portfolio construction based on FCF, ROE, OPI, LTD and IAA ratios. Portfolio 1 consists of the 25 best scoring companies according to the portfolio construction strategy, portfolio 2 consists of stocks 25-50 in the list, etc. Data: Bloomberg

The analysis is redone with the PE ratio as an additional explanatory variable. From the regression in tables 4.1.2 and 4.1.3 we might suggest that adding the PE ratio will improve the portfolio results. The PE ratio is ranked from the smallest positive number down to the largest number. Any zeros and negative numbers are eventually given the lowest ranking, because they cannot objectively determine the current market valuations when the data is either inadequate (PE is zero) or the net income in the given period was negative (PE ratio is

Average Volatility Sharpe ratio Portfolio 1 15.43 19.76 0.78 Portfolio 2 15.71 21.24 0.74 Portfolio 3 17.72 18.43 0.96 Portfolio 4 20.94 20.81 1.01 Portfolio 5 20.17 22.31 0.90 Portfolio 6 17.91 21.32 0.84 Portfolio 7 19.42 22.50 0.86 Portfolio 8 20.19 25.18 0.80 Portfolio 9 19.48 22.86 0.85 Portfolio 10 22.12 27.63 0.80

(21)

21

negative, because he market price is strictly positive). Also the scores are doubled to put an extra weight in the value of the PE ratio. This is done because the ratio shows the strongest significance in the regression analysis and it is the only variable that takes into account the current market valuations.

In table 4.2.4 we can see the improvement in the Sharpe ratio of portfolios 1 and 2. However, portfolios 3 and 4 show a decrease in their Sharpe ratio value. The average Sharpe ratio of all the portfolios improves from 0.85 to 0.87. This could be caused by dropping out some of the companies, which were included in the analysis of tables 4.2.1 and 4.2.2.

From the volatility levels in table 4.2.4 we can see a convergence among the 10 portfolios after using the PE ratio in the portfolio construction. The more equal volatility levels across the ten portfolios compared to the results in table 4.2.2, suggest the existence of the winner-loser effect as described in the study of Bondt & Thaler (1987). They have found an effect where bad performers in one year show a greater performance in the sequent year and vice versa. The addition of the PE ratio in the construction process of the portfolios obviously caused the smaller differences between the volatility levels of the portfolios.

Only portfolio 8 shows an outperformance in their average return, which is offset by the high volatility of 30.38% per year. With the current data and ratios an explanation for this outlying result remains open.

Table 4.2.3, yearly portfolio returns including the FCF, ROE, OPI, LTD, IAA and the PE as additional variable in the portfolio construction. Portfolio 1 consists of the 25 best scoring companies according to the portfolio construction strategy, portfolio 2 consists of stocks 25-50 in the list, etc. Data: Bloomberg

2003-2004 2004-2005 2005-2006 2006-2007 2007-2008 2008-2009 2009-2010 2010-2011 2011-2012 2012-2013 2013-2014 Portfolio 1 36.36 9.10 27.23 16.25 8.31 -31.18 42.42 36.65 10.14 11.52 26.76 Portfolio 2 37.74 23.58 20.60 11.62 4.28 -26.03 48.17 15.72 8.15 20.58 30.43 Portfolio 3 31.23 17.68 40.51 25.63 16.83 -38.14 34.18 25.37 1.66 16.27 23.93 Portfolio 4 46.41 12.70 19.52 14.53 5.02 -29.40 40.95 27.82 -5.81 21.16 28.61 Portfolio 5 33.96 18.03 26.10 20.22 -2.29 -27.98 49.47 27.97 2.29 21.81 29.63 Portfolio 6 40.87 22.36 29.58 13.56 2.28 -31.15 54.02 26.35 5.34 20.01 21.88 Portfolio 7 40.26 26.40 28.29 25.95 12.12 -32.43 46.36 23.49 4.91 24.05 17.53 Portfolio 8 77.20 23.33 38.37 31.81 -2.35 -31.84 63.14 26.76 -2.93 23.13 32.57 Portfolio 9 41.10 17.03 25.22 22.94 12.38 -27.12 51.72 23.46 1.34 25.04 24.80 Portfolio 10 29.84 22.81 21.05 24.05 -2.35 -30.95 57.33 20.81 2.25 27.87 23.06

(22)

22

Average Volatility Sharpe ratio Portfolio 1 17.60 20.35 0.86 Portfolio 2 17.71 19.41 0.91 Portfolio 3 17.74 21.25 0.83 Portfolio 4 16.50 21.34 0.77 Portfolio 5 18.11 20.83 0.87 Portfolio 6 18.64 22.14 0.84 Portfolio 7 19.72 20.79 0.95 Portfolio 8 25.38 30.38 0.84 Portfolio 9 19.81 20.45 0.97 Portfolio 10 17.80 22.24 0.80

Table 4.2.4, Average returns, volatility per year. Sharpe ratio as measured by the average return per year divided by yearly volatility. Portfolio construction based on FCF, ROE, OPI, LTD, IAA and PE ratios. Portfolio 1 consists of the 25 best scoring companies according to the portfolio construction strategy, portfolio 2 consists of stocks 25-50 in the list, etc. Data: Bloomberg

The last analysis will be done with only the most significant ratios from the regression in table 4.1.2. I will include the long-term debt ratios (LTD & IAA) together with the PE ratio to construct the portfolios.

Tables 4.2.5 and 4.2.6 show the results of the portfolios. Portfolio 1 is in the last four years of the test period the best performer. The volatility is great as well, however the Sharpe ratio has increased dramatically compared to the previous performances. Portfolios 1 and 4 now have the highest Sharpe ratio. This result greatly supports the importance of long-term fundamentals in determining the strength of enterprises. Additionally, the effect of the price-earnings ratio in the increase in this performance suggests that in the short-term under- and over-valuations occur in stock valuations. Low PE ratio stocks seem to outperform the market in the next year, however, analysis on individual stock returns is beyond the scope of this article as I have compared the total portfolio returns.

(23)

23

2003-2004 2004-2005 2005-2006 2006-2007 2007-2008 2008-2009 2009-2010 2010-2011 2011-2012 2012-2013 2013-2014 Portfolio 1 39.03 21.37 39.90 14.57 14.12 -35.55 48.79 32.68 11.63 26.92 36.90 Portfolio 2 32.75 21.99 37.23 19.56 6.84 -35.67 41.24 30.30 -0.44 8.79 25.77 Portfolio 3 73.31 12.91 16.71 21.89 4.33 -23.73 43.05 26.67 -4.19 21.34 19.36 Portfolio 4 41.26 12.86 27.64 26.33 -0.47 -27.69 54.80 30.74 1.98 22.84 23.58 Portfolio 5 47.97 29.47 41.18 19.36 16.40 -31.21 64.00 26.62 5.84 21.59 27.31 Portfolio 6 46.42 28.32 27.18 14.83 9.81 -28.97 49.23 23.76 0.89 21.27 20.69 Portfolio 7 34.33 21.51 29.48 20.76 -12.10 -33.17 40.47 20.92 3.02 16.89 37.07 Portfolio 8 32.82 17.81 20.70 16.12 -0.14 -25.68 51.02 19.92 -1.02 26.13 21.77 Portfolio 9 39.73 15.40 21.35 27.33 4.03 -29.50 50.27 23.39 8.48 20.29 26.25 Portfolio 10 36.17 8.86 19.34 28.30 12.87 -35.49 49.93 17.83 2.89 26.14 31.57 Table 4.2.5, yearly portfolio returns including the LTD, IAA and PE ratio as variables in the portfolio construction. Portfolio 1 consists of the 25 best scoring companies according to the portfolio construction strategy, portfolio 2 consists of stocks 25-50 in the list, etc. Data: Bloomberg

Average Volatility Sharpe ratio Portfolio 1 22.76 22.85 1.00 Portfolio 2 17.12 21.86 0.78 Portfolio 3 19.24 24.94 0.77 Portfolio 4 19.44 22.28 0.87 Portfolio 5 24.41 24.45 1.00 Portfolio 6 19.40 21.40 0.91 Portfolio 7 16.29 22.39 0.73 Portfolio 8 16.31 19.93 0.82 Portfolio 9 18.82 20.63 0.91 Portfolio 10 18.03 22.15 0.81

Table 4.2.6, Average returns, volatility per year. Sharpe ratio as measured by the average return per year divided by yearly volatility. Portfolio construction based on LTD, IAA and PE ratios. Portfolio 1 consists of the 25 best scoring companies according to the portfolio construction strategy, portfolio 2 consists of stocks 25-50 in the list, etc. Data: Bloomberg

(24)

24

5. Conclusion

From the theories and empirical analysis in this article we can conclude that questions remain about stock market predictions. The capital structure is definitely an important driver for the stock returns, but evidence for an optimal capital structure is yet to be found. It could hardly be the fact that one optimal debt level generally exists, because these debt levels differ among companies and certainly industries.

Besides, I have tried to implement a rather easy investment strategy based on the fundamental characteristics of companies. These characteristics are measured with ratios from the data of the financial statements. The focus in this article was particularly on debt linked ratios, such as the long-term debt level and debt expenses incurred by firms, among others. From the analysis I can conclude that the ratios used in this article are not sufficient to differentiate between strong and weak firms in the selection process. In the sample data I have used, the only significance is found in long-term ratios, which turn out to be less flexible over the years in the test period. I found gradual changes in the long-term debt on the balance sheet compared to more heavy fluctuations in the short-term debt on companies balance sheet. Fluctuations in earnings and cash flows are less stable in the short-term and thus seem to be less influential in predicting the stock performance over a one-year period.

More research has to be done on the performance of portfolios during longer time periods. For this reason the systematic approach in constructing portfolios could be extended with more fundamental ratios. The addition of more ratios increase the distinction between strong and weak companies which should result in greater differences between the portfolio returns.

(25)

25

Reference list

Attinasi, M. G., Checherita-Westphal, C. D., & Nickel, C. (2009). What explains the surge in euro area sovereign spreads during the financial crisis of 2007-09? European Central Bank, Working paper series, 1131 (December 2009).

Baker, M., & Wurgler, J. (2006). Investor sentiment and the cross‐section of stock returns. The Journal of Finance, 61(4), 1645-1680.

Barclay, M. J., Marx, L. M., & Smith, C. W. (2003). The joint determination of leverage and maturity. Journal of Corporate Finance, 9(2), 149-167.

Bernanke, B. S., & Blinder, A. S. (1992). The federal funds rate and the channels of monetary transmission. The American Economic Review, 901-921.

Bernanke, B. S., & Kuttner, K. N. (2005). What explains the stock market's reaction to Federal Reserve policy?. The Journal of Finance, 60(3), 1221-1257.

Bhandari, L. C. (1988). Debt/equity ratio and expected common stock returns: Empirical evidence. The Journal of Finance, 43(2), 507-528.

Bondt, W. F., & Thaler, R. H. (1987). Further evidence on investor overreaction and stock market seasonality. The Journal of Finance, 42(3), 557-581.

Bradley, M., Jarrell, G. A., & Kim, E. (1984). On the existence of an optimal capital structure: Theory and evidence. The Journal of Finance, 39(3), 857-878.

Fama, E. F., & French, K. R. (1992). The cross‐section of expected stock returns. The Journal of Finance, 47(2), 427-465.

Fama, E. F., & French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33(1), 3-56.

Fawley, B. W., & Neely, C. J. (2013). Four stories of quantitative easing. Federal Reserve Bank of St. Louis Review, 95(January/February 2013).

Grossman, S. J., & Shiller, R. J. (1980). The determinants of the variability of stock market prices.

Guo, H. (2004). Stock prices, firm size, and changes in the federal funds rate target. The Quarterly Review of Economics and Finance, 44(4), 487-507.

Jensen, G. R., & Mercer, J. M. (2002). Monetary Policy and the Cross‐Section of Expected Stock Returns. Journal of Financial Research, 25(1), 125-139.

Kim, E. H. (1978). A mean‐variance theory of optimal capital structure and corporate debt capacity. The Journal of Finance, 33(1), 45-63.

Leland, H. E. (1994). Corporate debt value, bond covenants, and optimal capital structure. The Journal

of Finance, 49(4), 1213-1252.

Lo Duca, M., Nicoletti, G., & Martinez, A. V. (2014). Global corporate bond issuance: what role for US quantitative easing?.

Markowitz, H. M. (1991). Foundations of portfolio theory. The Journal of Finance, 46(2) 469-477.

Modigliani, F., & Miller, M. H. (1958). The cost of capital, corporation finance and the theory of investment. The American Economic Review, 48(3) 261-297.

Myers, S. C. (1977). Determinants of corporate borrowing. Journal of Financial Economics, 5(2), 147-175.

(26)

26

Myers, S. C. (1984). The capital structure puzzle. The Journal of Finance, 39(3), 574-592.

Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics, 13(2), 187-221.

Piotroski, J. D. (2000). Value investing: The use of historical financial statement information to separate winners from losers. Journal of Accounting Research, 38 1-41.

Smirlock, M., & Yawitz, J. (1985). Asset returns, discount rate changes, and market efficiency. The Journal of Finance, 40(4), 1141-1158.

Stoll, H. R., & Whaley, R. E. (1983). Transaction costs and the small firm effect. Journal of Financial Economics, 12(1), 57-79.

Thorbecke, W. (1997). On stock market returns and monetary policy. The Journal of Finance, 52(2), 635-654.

Titman, S., & Wessels, R. (1988). The determinants of capital structure choice. The Journal of

Finance, 43(1), 1-19.

Welch, I. (2004). Capital structure and stock returns. Journal of Political Economy, 112(1), 106-132.

Internet sources

CME Group, Interest rate contracts. (2015, April 14) Retrieved from:

http://www.cmegroup.com/trading/interest-rates/fed-funds.html

Federal Reserve Bank of New York, Federal Funds data. (2015, June 7) Retrieved from:

http://www.newyorkfed.org/markets/omo/dmm/fedfundsdata.cfm

Federal Reserve, Board of Governors of the Federal Reserve System. (2015, June 7) Retrieved from:

http://www.federalreserve.gov/releases/h15/data.htm

Federal Reserve, Board of Governors of the Federal Reserve System. (2015, June 9) Retrieved from:

http://www.federalreserve.gov/monetarypolicy/

S&P 500 Financials, S&P Dow Jones Indices (2015, June 22) Retrieved from:

https://us.spindices.com/indices/equity/sp-500-financials-sector

US Securities and Exchange Commission, Form 10-K (2015, June 15) Retrieved from:

http://www.sec.gov/answers/form10k.htm

VIX historical price data, CBOE (2015, June 25) Retrieved from: http://www.cboe.com/micro/vix/historical.aspx

(27)

27

Appendix A

For a clearer view of the investment strategy, I will discuss the portfolio construction in greater detail. The investment strategy is similar in use for all kind of ratios that could be included. In this appendix I will only discuss the portfolio construction where I have used the LTD, IAA and PE ratios.

First of all, we need to obtain the data of all the companies. This will be done separately for each year in the period of interest. The companies will be sorted by the outcome of the calculated ratio. Then I will rank the companies from 1 until the total number of companies in the dataset. This is done for each ratio. This way, it is possible to obtain a “total score” based on the relative strength of the company. Obviously, we need to determine what actually divides a strong company from a weak one.

In this analysis I have used the following differentiation:

- LTD ratio: Higher long-term debt-to-total assets is worse (in contradiction to the regression results)

- IAA ratio: Higher intangible assets-to-total assets is worse (in accordance with the regression results)

- PE ratio: Higher PE ratio is worse (in accordance with the regression results). In extension, I have given negative or zero PE ratios the lowest rank (highest score). By sorting the companies on the basis of their total score in this analysis I could construct portfolios and calculate the portfolio returns in a given year. This analysis could be easily extended with other ratios to obtain a better differentiation between companies to achieve better results.

Referenties

GERELATEERDE DOCUMENTEN

As the weather variables are no longer significantly related to AScX returns while using all the observations, it is not expected to observe a significant relationship

The momentum returns of the low and high Ivol tercile portfolios have a negative exposure to the SMB size factor, while the momentum return of the medium Ivol portfolio has a positive

However, these data points are still present and might influence the alphas in such a way that it does not report significant positive returns for the most active funds during

Investors still prefer banks that have high Tier 1 capital ratios, both weighted and unweighted and a low RWA density, so a density which is at least half a standard

The influence of international soccer results on market indices for southern European Countries This table reports the estimated coefficients and the corresponding p-values of

(e.g., a portfolio with maximum included rank 70 with size 40 includes the worst (loser) and best (winner) ranked stocks at the end of the ranking period between rank 30 and 70)...

For the influence of OPEC announcements on oil firms’ stock prices in general, so without distinguishing for ‘small’ or ‘large’ firms, I find positive significant returns for

The general mechanical design of the Twente humanoid head is presented in [5] and it had to be a trade-off between having few DOFs enabling fast motions and several DOFs