• No results found

Financial Inflexibility and the Value Premium in the US Consumer Sector

N/A
N/A
Protected

Academic year: 2021

Share "Financial Inflexibility and the Value Premium in the US Consumer Sector"

Copied!
22
0
0

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

Hele tekst

(1)

Financial Inflexibility and the Value

Premium in the US Consumer Sector

Oane de Graaf

Koene Parmentierloane 13, 9269 VP, Feanwalden, oanedegraaf@hotmail.com

Supervisor: Dr. Halit Gonenc

Abstract

This paper tests whether the financial inflexibility explains the value premium in the US Consumer Sector. In this paper, financial inflexibility is defined as the inability of a firm to smooth dividends over time, caused by an inability to adjust investments on exogenous shocks. This financial inflexibility index is constructed using three variables, namely the costly reversibility, total leverage and financial constraints. Based on the literature I expect a positive relation between the value premium and the financial inflexibility index. However, when adding the financial inflexibility index in the Capital Asset Pricing Model and three-factor model as an extra risk three-factor, no relation is detected.

Keywords: Asset Pricing; Value Premium; CAPM; Leverage; Fixed Asset

JEL codes: D92, G12, G31, G32

University of Groningen Faculty of Economics and Business

Msc Finance

(2)

2

I. Introduction

When a firm has a relatively low ratios of book-to-market equity, earnings to price and cash flow to price it is considered a growth firm. When these ratios are relatively high than it is considered a value firm. Recent research shows that portfolios of stocks of value firms perform on average better than portfolios of growth firms. Fama and French (1992, 1996) for example, find that in the period between 1962 and 1989 there exists a so called value premium in the US. Thereafter, Fama and French (1998) and Bauman et al. (1998) find international evidence that value portfolios perform better than growth portfolios. Next to that, quite some country specific empirical studies concerning the existence of the value premium have been done (example given: Bauman et al., 2001, Drew and Veeraraghavan, 2002 and Athanassakos, 2011). Most of them attain to the same conclusion, namely that the value premium does exist.

The next question is how the value premium can be explained. Researchers are not consentient in their explanation. Behavioural theory explains the value premium as an overreaction of the market. For example, De Bondt and Thaler (1985) find in their research that portfolios that underperform other portfolios in the first period have a tendency to outperform those portfolios in the next period. In later research, Lakonsihok et al. (1994) and Haugen (1995) argue that the value premium in average returns arises because the market undervalues distressed stocks and overvalues growth stocks (Fama and French, 1998). When these price errors are corrected, distressed (value) stocks have high returns and growth stocks have low returns (Fama and French, 1998).

On the other hand, there is the neoclassical theory, which argues that fluctuations in stock prices should be explained by a certain risk factor. This idea is the basis for the Capital Asset Pricing Model, which is the basic model used to explain pricing movements in finance. In this model the only independent variable is the beta, representing the risk of the asset against the market. However, research shows that when using only this variable the value premium is not explained. Therefore, according to the neoclassical theory there has to be another risk factor which explains the value premium.

(3)

3

risk premium taking into consideration this cyclical dividend pay-out of firms. The index Poulsen et al. (2013) construct is based on three sources that can cause financial inflexibility, namely costly reversibility, total leverage and financial constraints.

Zhang (2005) defines costly reversibility as the reasoning that it is more costly for firms to reduce capital in a bad economic environment than to expand capital in a good economic environment, in other words firms have more adjustment costs in reduction than in expanding. Zhang (2005) argues that, firms having a higher fixed assets ratio assess more unproductive capital, making it more difficult to reduce their capital in bad times. Cooper (2006) contributes that firms with a high number of fixed assets are also inflexible in a positive economic environment. High fixed assets firms have the unproductive capital, which changes to productive capital in a growing economy. Therefore, these firms do not invest that much when increasing. Because this lack of investing, the dividend pay-out to shareholders in positive economic environments will expend. Hence, with more costly reversibility the cyclical dividend pay-out increases and the financial inflexibility increases as well.

The second constraint Poulsen et al. (2013) use in their research is the total leverage. Poulsen et al. (2013) defines operating leverage as the sensitivity in operating income with changes in revenue, it therefore increases when fixed costs increase. Financial leverage is defined as the sensitivity of net income with changes in operating income, and therefore increases when interest expenditure grows. Hence, a firm with a high total leverage has increasing fixed costs and interest expenditure. This means that when the revenue fluctuates, the net income will be more sensitive than firms with lower total leverage. As a consequence, Gulen et al. (2011) argue that a high leverage makes firms inflexible, with high leverage firms are less able to adapt investments to meet the relatively high volatility of their cash flows. Empirical studies (García-Feijóo and Jorgenson, 2010, Manelker and Rhee, 1984 and Ho et al., 2004) find that operating leverage has an effect on the systematic risk of firms. However, for the financial leverage the empirical research is less clear. Manelker and Rhee (1984) find a significant relation between financial leverage and stock returns. On the other hand, Darrat and Mukherjee (1995) and Lord (1996) find no relation between these variables. The third variable in the index are the financial constraints. This is defined by Poulsen et al. (2013) as the frictions that prevent firms from funding all desired investment. When an investment opportunity appears a firm which has more inabilities to get funding is less able to profit from this opportunity. This can affect the profitability of the firm, enlarging the risk of cyclical dividend pay-out (Livdan et al., 2009).

(4)

4

taking one industry instead of all industries that gives other results, which are more closely related to one another. The reasoning behind this is, that the firms in the dataset are less different from each other, when only using one industry. Moreover, I will use a different time period than Poulsen et al. (2013). Poulsen et al. (2013) use in their analyses the period 1970 until 2009, where I will use data from 1998 till 2012. This means that the financial crisis (from approximately 2008 onwards) will probably have an influence in the results of this research. When using a larger time period, like Poulsen et al. (2013), the economic shocks will be less visible than when using a shorter time period.

(5)

5

II. Hypothesis

Poulsen et al. (2013) argue that the financial inflexibility explains the value premium. Value firms are, as explained before, firms with a relative high book-to-market value. Financial inflexible firms are firms having trouble reversing there capital investments. Poulsen et al. (2013) quote Cooper (2006, p. 140) which states that if capital investments are largely irreversible, it is probable that their profits and therefore their dividends fluctuate more over time. Investors will hedge for this risk factor, making the market value decrease, while the book value of the assets remains relatively equal. Hence, firms with a high book-to-market ratio are more likely to be financial inflexible.

Poulsen et al. (2013) also find empirical evidence that the financial inflexibility explains the value premium. However, this variable does not explain the whole value premium. When putting the inflexibility factor in the Capital Asset Pricing Model and the Fama and French three-factor model it has extra explanatory power, therefore implying that the variable is a risk factor that influences the differences in return. However, not the complete mispricing errors are explained by adding the inflexibility index to the model, that is why they conclude that financial inflexibility is a factor which explains part of the value premium.

In my research I also include an inflexibility index using operating leverage instead of total leverage. This is done, because in the literature it is not clear if the financial leverage has an impact on the market premium. Darrat and Mukherjee (1995) and Lord (1996) find no relation between these two variables, where for example Manelker and Rhee (1985) do find a relation. This is why I will compare the explanatory power of the inflexibility index including total leverage and the inflexibility index when only using the operational leverage.

(6)

6

III. Methodology

In this model I will use the same model as used by Poulsen et al. (2013). Poulsen et al. (2013) create an index for financial inflexibility. The index is based on three compartments: costly reversibility, total leverage and financial constraints.

Costly reversibility is measured by the fixed assets ratio, which is the gross property, plant and equipment divided by the total assets of a firm. Rajan and Zingales (1995) argue that the fixed assets ratio is related to asset specifity and tangibility.

The second compartment is the total leverage. It is not possible to measure leverage directly, different techniques have been used to measure leverage of companies. García-Feijóo and Jorgenson (2010) make a series of assumptions in the classical leverage model, so they can to conclude that operating and financial leverage should be estimated by elasticity measures. Then there are two approaches which have been used in the recent literature, the time-series regression or the point-to-point approach. The point-to-point approach estimates leverage as a ratio of changes in earnings to changes in sales or fixed assets to total assets (García-Feijóo and Jorgenson, 2010). The time-series regression uses a regression of earnings before interest and depreciation, net sales (or net revenue) and net income to estimate the operating and the financial leverage. In this research I will use the time-series regression, because it is more theoretically adequate (Dugan and Shriver, 1989). Nevertheless, both measures seem to suffer from the same biases (Lord, 1998). I will estimate the following three regressions on 5-year overlapping intervals. So, for the time period 1998-2002; the time period 2003-2007 and the time period 2008-2012 I make different regressions for each individual firm.

=∝ + + , (1)

=∝ + + , (2)

=∝ + + , (3)

In this regression the α1 is set to one. So there is only a constant in this model. Furthermore, the EBIT is the Earnings before interest and taxes, Sales is the sales revenue and NI is the Net income of a firm. For logs of negative earnings I use the transformation used by Ljungfist and Wilhelm (2005), which is common in accounting research: ln (1+EBIT/Total Assets) if EBIT≥0 and ln (1-EBIT/Total Assets) if EBIT<0. I make this transformation to adjust for the fact that it is not possible to calculate negative logarithms, using this transformation it is.

(7)

7

, = ! , + ,"# (4)

, = $ , + ,%# (5)

Where OL (FL) is the estimate of the degree of operating (financial) leverage, measures the average sensitivity of the percentage deviation of EBIT (NI) from its trend, relative to the percentage deviation of sales (EBIT) from its trend (Poulsen et al., 2013). The total leverage is calculated as the sum of the operating leverage and the financial leverage.

Recent literature tries to find a measure for the financial constraints of a firm. Fazzari et al. (1988) for instance, find a negative relation between the dividend ratio of firms and their financial constraints. The intuition is that firms which have more trouble getting finance for their projects hold on to more of their previous earnings, so they can use that earnings for financing their future projects, therefore using less of their earnings to pay dividend. The size and age of a firm are also assumed to have an impact on the constraints in financing projects. Several papers explain that when the firm is relatively large and mature the financing is easier than for relatively smaller and young firms (Devereux and Schiantarelli (1990), Gilchrist and Himmelberg (1995) and Whited (2006)), since it should be easier to collect information from these firms. More variables have been used to explain the financial constraints of a firm, like group membership (Whited, 2006), profitability (Musso and Schiavo, 2008) and repaying ability (Musso and Schiavo, 2008).

Whited and Wu (2006) constructed a measure for financial constraints of firms in the US, which Poulsen et al. (2013) also uses. It contains the independent variables: cash flow, dividend position, debt-to-assets ratio, natural logarithm of total assets, industry sales growth and firm-specific sales growth.

&&' = −.091-$' − .0620 12! ' + .021 0' − .044 4' + .102 5' − .035 5' (6) Where i stands for the firm and t stands for the year, CF is the cash flow, DIVPOS is a dummy variable which is one when the firm pays out dividend and zero when the firm does not. The TLTD is the long-term debt divided by the assets, LNTA is the log of total assets (inflation adjusted in 2004) and ISG is the average inflation adjusted sales growth in the industry of firm i. Finally, SG measures the firm specific inflation-adjusted sales growth.

The ‘composite inflexibility index’ is then calculated measuring a normalized value, for all the three measures, giving the following metric:

(8)

8

Where Xi,t is the firm-year measure of the given proxy, X is the time series average of

the proxy of all observations and σx is the time series standard deviation of the given proxy. Summerizing these three proxies gives the composite inflexibility index constructed by Poulsen et al. (2013):

$ 8', = ∑A 8',

BC (8)

I wil add this index as an independent variable in the Capital Asset Pricing Model. Also in the Fama and French (1993) three-factor model I will add the inflexibility index.

Without the inflexibility index variable the Capital Asset Pricing Model and the Fama and French three-factor model are as follows.

D', − DB, = +∝ (DE, − DB, ) + (9)

D', − DB, = +∝ FDE, − DB, G +∝H I +∝AJI + (10)

The first model (equation (9)) is the Capital Asset Pricing Model, where the Ri,t is the

companies return in a given month for a given firm, the Rf,t is the risk-free rate given a

specific month. And the Rm.t is the return of the market in a given month. The α0 is an

intercept to capture the mispricing in accordance with Poulsen et al. (2013).

The second model (equation (10)) is the Fama and French three-factor model. The added factor SMB is the small minus big factor, which is the monthly size factor, adjusting for the difference in size. The HML is the high minus low factor, which is the monthly value factor, adjusting for the difference in value of the firms.

Including the composite inflexibility index in these models gives:

D', − DB, = +∝ FDE, − DB, G +∝H $ 8', + (11)

D', − DB, = +∝ FDE, − DB, G +∝H I +∝AJI +∝K $ 8', + (12) As explained in before, in this paper I also include a model where the operational leverage is used to calculate the inflexibility index. Instead of the total leverage, which is used in the previous models. This model be as follows:

D', − DB, = +∝ FDE, − DB, G +∝H I +∝AJI +∝K $ 8 8$ ', + (13)

(9)

9

IV. Data

In this research I use data of firms in the Standard and Poor’s 1500 index of the US consumer products sector. The financial data, which I use to calculate the inflexibility index, and the stock returns are both from Thomson and Reuters’ Datastream. The inflexibility index is calculated yearly and therefore the financial data is also collected yearly. The stock return data is collected monthly. The SMB, HML, Rm-Rf and Rf variables in the Fama and French three-factor model are published in the database on the site of Kenneth R. French. I used here the monthly US database. From the original of 246 companies 150 firms remained in the final sample, after excluding observations which were not available, had negative total assets, fixed assets or sales or a negative book-to-market ratio.

Ten portfolios are composed of the 150 firms, sorted by their book-to-market ratio. The first portfolio contains the 15 firms having the lowest book-to-market ratio, or the growth stocks. While the tenth portfolio contains the 15 firms having the highest book to market ratio, otherwise considered as the valued stocks. I want to secure that the portfolios will keep their characterised stocks, therefore the portfolios are reallocated every five years. They are reallocated using the book-to-market ratio of the first year of the five year period. Than the portfolio will consist of the same companies for five years, after five years the reallocation will select different portfolios.

In table 1 the main characteristics of the sample of firms are given. As can be seen, the total assets of the including firms fluctuates from 28 thousand dollars to more than 303 million dollars. The standard deviation of the sample is more than 9 million dollars, so there is quite some difference in the capitalization of the firms.

Table I

The Company Descriptive Statistics for 150 US Firms for the Period 1998-2012

The Characteristics are the main discriminants of the US firms in the US consumer products sector, which are used for the calculation of the constraints of the inflexibility index. For this sample of firms,

the Maximum, Minimum, Average, Median and Standard Deviation is presented.

Characteristics Maximum Minimum Average Median Stnd Dev

Plant, Prop. & Equipm. 123,223,000 1,638 3,518,236 631,359 9,743,901 Total Assets 303,828,000 28,293 7,075,800 1,424,396 24,569,001 EBIT 34,299,000 -41,907,000 558,982 154,093 1,946,790 Net Sales 177,089,000 25,007 6,356,213 1,877,966 15,557,509 Net Income 22,071,000 -44,461,000 254,796 75,350 1,487,192 Depreciation 16,519,000 604 301,745 57,814 1,225,356 Total Dividends 5,348,000 0 77,509 4,450 255,255

(10)

10

V. Results

A. Portfolio Descriptions

In table 2 the averages of the book-to-market ratio, returns, inflexibility index and the determinants of the inflexibility index are given. Portfolio 1 contains the companies with the ten percent lowest book-to-market ratio, in other words this are the companies with the highest market value compared to their book value, and therefore considered growth companies. On the other hand, portfolio 10 contains the ten percent companies with the highest book-to-market ratio, also called the value companies. In the return column the average monthly returns on the stock market for the companies is given. The average return fluctuates from 0.41% to 0.89%, there does not seem to be a clear connection between the book-to-market ratio and the returns.

The normalized total leverage, the normalized fixed assets ratio and the normalized Whited and Wu (2006) ratio form the financial inflexibility index. The last column excludes the financial leverage in calculating the financial inflexibility index, it only includes the operational leverage. I hypothesized in the theoretical part of this paper that the inflexibility index would be larger when the book-to-market ratio increases, furthermore the total leverage, fixed assets ratio and Whited and Wu (2006) factor should also increase when the book-to-market ratio increases. However, in this table it does not seem like inflexibility index increases when the to-market ratio increases, there seems no connection between the growth in book-to-market ratio and the fluctuation in inflexibility index (for neither the inflexibility index for total leverage, nor the index excluding financial leverage). Furthermore, the operating leverage, total leverage, fixed assets ratio and the WW factor do not seem to show a clear pattern as the book-to-market ratio increases.

The correlation table for the inflexibility index and the variables calculated for this index are presented in table 3. The inflexibility index and the variables are expected to positively correlate with one another, this is because they are expected to react in the same direction when the book-to-market ratio is changing. Moreover, they are expected to positively correlate with the book-to-market ratio, since the hypothesis is that the inflexibility index is an explanation for the value premium. It therefore should increase (decrease) as the book-to-market ratio increases (decreases). Because of this, the book-to-market ratio and the inflexibility index are expected to correlate positively.

(11)

11

show a positive correlation. However, the fixed assets ratio shows a negative correlation, this does not correspond with the literature.

Table II

The Average Determinants of the Inflexibility Index for the Different Portfolios

In the first column the ten Portfolios are presented, the Portfolios are determined using the book-to-market ratio. The Average Book-to-Market ratio for the Portfolios is presented in the second column. The Expected Return represents the average monthly expected return for the different Portfolios. The Total Leverage, Fixed Assets Ratio and WW Factor are the determinants for the Inflexibility Index.

This variables are normalized. The Fixed Assets Ratio is the determinant for the costly reversibility and the Whited and Wu (2006) Factor is the determinant for the financial costraints. The Inflexibility

Index (-Financial Leverage) represents the Inflexibility Index where, instead of Total Leverage, the

Normalized Operating Leverage is used in the calculation.

Table III

The Correlation Table for the Determinants of The Inflexibility Index

In this table the correlation between the Book-to-Market ratio, the Operating Leverage, the Total

Leverage, Fixed Assets Ratio, the Inflexibility Index and the Inflexibility Index(-Financial Leverage) is

presented. B/M Portfolio Book-to-Market Expect. Return Oper. Lever. Total Lever. Fixed Assets WW Factor Inflex. Index Inflex. Index (-Finan. Lev) 1 (Growth) 0.085 0.0060 0.012 0.012 0.263 -0.028 0.247 0.248 2 0.212 0.0069 0.012 0.010 -0.029 0.282 0.263 0.266 3 0.275 0.0070 0.022 0.022 -0.068 -0.047 -0.094 -0.093 4 0.353 0.0064 0.033 0.033 0.047 -0.467 -0.388 -0.387 5 0.417 0.0089 0.015 0.011 0.026 -0.073 -0.036 -0.032 6 0.529 0.0059 0.037 0.036 -0.007 -0.061 -0.032 -0.031 7 0.631 0.0041 0.000 -0.001 0.122 0.126 0.247 0.249 8 0.741 0.0045 -0.173 -0.174 0.030 0.112 -0.032 -0.031 9 0.962 0.0065 -0.058 -0.059 -0.111 0.079 -0.091 -0.090 10 (Value) 1.789 0.0068 0.098 0.111 -0.205 0.077 -0.017 -0.030 Book-to-Market Oper. Lever. Total Lever. Fixed Assets WW Factor Inflex. Index Inflex. Index (-Finan. Lev.) Book-to-Market 1.000 0.028 0.031 -0.015 0.023 0.023 0.021 Oper. Lever. 0.028 1.000 0.997 -0.034 -0.005 0.554 0.556 Total Lever. 0.031 0.997 1.000 -0.034 -0.006 0.555 0.553 Fixed Assets -0.015 -0.034 -0.034 1.000 0.037 0.579 0.578 WW Factor 0.023 -0.005 -0.006 0.037 1.000 0.596 0.597 Inflex. Index 0.023 0.554 0.555 0.579 0.596 1.000 0.999

(12)

12

B. The Capital Asset Pricing Model

The results of the Capital Asset Pricing Model, equation (9), using ordinary least squares for the different portfolios are presented in table 4. Where, again, the first portfolio is the one with the lowest book-to-market ratios and the tenth portfolio is the one with the highest book-to-market ratios. The first column represents the Alpha zero, or the intercept, the Rm-Rf is the independent variable in the model, and the Adjusted R-squared is the fit of this model in comparison to the individual results, adjusted for the number of variables. The first number is the coefficient of the variable(s). The number between brackets is the probability that the coefficient of the independent variable is zero, using a t-test. So if the probability is low, the variable is significant, and the hypothesis that the independent variable does not influence dependent variable is rejected.

The goal of the recent literature is to search for a proxy for the value premium, because the CAPM does not explain the value premium as recent literature shows. So in a model without a compensation proxy for the value premium, the expectation is that there is some mispricing in the model. This is why the Alpha zero is added to the model, to adjust for mispricing errors.

However, in table 4 this alpha zero is not bigger than 0.003 in the portfolios, and the probability is far from significant. This means that the model does not need much adjusting for mispricing errors, so according to this model no proxy for the value premium has to be added. The Rm-Rf is positive and significant, this is in comparison with the theory of the CAPM. As the risk-factor increases, so does the expected return. The coefficient of the risk factor is larger for the tenth portfolio comparing to the other portfolios, suggesting that the value premium is visible in this model. The adjusted R-squared seems to be slightly higher for the portfolios in the middle.

(13)

13

no positive result detected, and therefore the inflexibility index is not an extra risk factor in this model.

Table IV

The Capital Asset Pricing Model

The Capital Asset Pricing Model is explained by equation (9). For the ten Book-to-Market Portfolios this model is regressed. The Alpha represents the Alpha zero and is the constant in the CAPM. The

Rm-Rf represents the coefficient of the Return on the Market minus the Risk Free Return and is the

independent variable of the model. The significance of the variables is calculated using a t-test, the probability of the t-test is given between the brackets. When a variable has a significant influence on

the Expected Return, the probability will be low. And the Adjusted R-squared represents the percentage of the expected return that can be explained by the model.

B/M Portfolio Alpha Rm-Rf Adj. R-squared

(14)

14

Table V

The Capital Asset Pricing Model including the Inflexibility Index

The Capital Asset Pricing Model including the Financial Inflexibility Index is explained in equation (11). For the ten Book-to-Market Portfolios this model is regressed. The Alpha represents the Alpha zero and is the constant in the model. The Rm-Rf represents the coefficient of the Return on the Market minus the Risk Free Return and is an independent variable of the model. The Inflexibility Index is the second independent variable, and is the additional variable compared to Table IV. The significance of

the variables is calculated using a t-test, the probability of the t-test is given between the brackets. When a variable has a significant influence on the Expected Return, the probability will be low. And the Adjusted R-squared represents the percentage of the expected return that can be explained by

the model.

B/M Portfolio Alpha Rm-Rf INFLX Adj. R-squared

(15)

15

C. Fama and French Three-factor Model

The Fama and French three-factor model (equation (10)) includes two additional variables compared to the Capital Asset Pricing Model. In table 6 the results of this model are presented. The additional risk factors SMB and HML have a significant influence on the expected return when using a 1 percent significance in all the portfolios, except the SMB factor in the first portfolio. The coefficient of the SMB and HML are positive, meaning that, as the SMB and HML factor increase this has an increasing effect on the expected return. As in the CAPM, the Rm-Rf variable shows a positive significant result. The HML(high-minus-low) factor, is the value factor, and is an attempt by Fama and French (1993) to correct for the value premium. That is why it is according to the theory that the coefficient increases as the portfolios consist of a higher book-to-market ratio.

When comparing the three-factor model to the Capital Asset Pricing Model, in table 4, you can conclude that the three-factor model has a higher adjusted R-squared than the CAPM. Therefore, it seems like this model is a better fit for the data. Nevertheless, where the CAPM does not show any significant alphas the three-factor model does. The five portfolios which contain the highest value firms have a significant alpha on a 10 percent significance. This means that there are mispricing errors in these portfolios.

(16)

16

Table VI

The Fama and French (1993) Three-Factor Model

The Fama and French (1993) Three-Factor Model is explained by equation (10). It includes three independent variables, namely the Rm-Rf (Return of the Market minus the Risk Free Return), the

SMB Factor (Small Minus Big) and the HML Factor (High Minus Low). The Alpha, again, represents the

Alpha zero and in the constant in this model. As in the previous tables, the significance of the variables is calculated using a t-test, the probability of the t-test is given between the brackets. When

a variable has a significant influence on the Expected Return, the probability will be low. And the

Adjusted R-squared represents the percentage of the expected return that can be explained by the

model.

B/M Portfolio Alpha Rm-Rf SMB HML Adj. R-squared

(17)

17

Table VII

The Fama and French (1993) Three-Factor Model including the Inflexibility Index

The Fama and French (1993) Three-Factor Model including the Inflexibility Index is explained by equation (12). It includes the three independent variables included in the Three-Factor Model, the

Rm-Rf (Return of the Market minus the Risk Free Return), the SMB Factor (Small Minus Big) and the HML Factor (High Minus Low). Besides, this model includes the Inflexibility Index as an additional

independent variable. As in the previous models, The Alpha represents the Alpha zero and in the constant in this model. Additionally, the significance of the variables is, again, calculated using a t-test, the probability of the t-test is given between the brackets. When a variable has a significant

influence on the Expected Return, the probability will be low. And the Adjusted R-squared represents the percentage of the expected return that can be explained by the model.

B/M Portfolio Alpha Rm-Rf SMB HML INFLX Adj. R-squared

(18)

18

D. Operational leverage vs. Total leverage

In the theoretical background I explained that for the connection between the value premium and the operating leverage the empirical evidence is more visible, compared to the empirical evidence for the financial leverage. In table 3, the correlation between the book-to-market ratio and the total leverage and operational leverage is given. The hypothesis would be that the operating leverage would correlate more to the book-to-market ratio, than the total leverage. However, I see the opposite, the total leverage has a correlation of 0.031 and the operating leverage has a correlation of 0.028. Therefore, the total leverage seems to have a stronger connection with the market premium than the operating leverage. The inflexibility excluding the financial leverage is also given in table 3. And it is not surprising that the index without financial leverage is less correlated than the index including financial leverage, because the correlation of the total leverage is higher than that of the operating leverage.

(19)

19

Table VII

The Fama and French (1993) Three-Factor Model including the Inflexibility Index (-Financial Leverage)

The Fama and French (1993) Three-Factor Model including the Inflexibility Index(-Financial Leverage) is explained by equation (13). It includes the three independent variables included in the Three-Factor Model, the Rm-Rf (Return of the Market minus the Risk Free Return), the SMB Three-Factor (Small Minus Big) and the HML Factor (High Minus Low). Besides, this model includes the Inflexibility

Index(-Financial Leverage) as an additional independent variable. The difference with the Inflexibility Index

is that this index is calculated using only the operational leverage, and not the total leverage. As in the previous tables, The Alpha represents the Alpha zero and in the constant in this model. Also, the significance of the variables is calculated using a t-test, the probability of the t-test is given between the brackets. When a variable has a significant influence on the Expected Return, the probability will be low. And the Adjusted R-squared represents the percentage of the expected return that can be

explained by the model.

B/M Portfolio Alpha Rm-Rf SMB HML INFLXEXFL Adj. R-squared

(20)

20

VI. Conclusion

The research of Poulsen et al. (2013) explains the connection of the value premium and the financial inflexibility. The value premium should compensate for the high financial inflexibility of value firms. The literature explains that financial inflexibility contains of three variables, namely costly reversibility, leverage and financial constraints. In these three variables there should be a positive relation with the book-to-market ratio. Furthermore, when the financial inflexibility is compensating for the financial inflexibility it would be an additional risk factor. That is why in this research I added the financial inflexibility index as an additional risk factor.

(21)

21

VII. References

Athanassakos, G., 2009, Value versus Growth Stock Returns and the Value Premium: The Canadian Experience 1985-2005, Canadian Journal of Administrative Sciences, Vol. 26, pp. 109-121

Bauman, W. S., Conover, C. M. and Miller, R. E., 1998, Growth versus Value and Large-Cap versus Small-Cap Stocks in International Markets, Financial Analysts Journal, Vol. 54, pp. 75-89

Bauman, W. S., Conover, C. M. and Miller, R. E., 2001, The Performance of Growth Stocks and Value Stocks in the Pacific Basin, Review of Pacific Basin Financial Markets and Policies, Vol. 4, pp. 95-108

Cooper, I., 2006, Asset Pricing Implications of Nonconvex Adjustment Costs and Irreversibility of Investment, Journal of Finance, Vol. 61, pp. 139-170

Darrat, A. F. and Mukherjee, T. K., 1995, Inter-Industry Differences and the Impact of Operating and Financial Leverages on Equity Risk, Review of Financial Economics Vol. 4, pp. 141-155

De Bondt, W. F. M., Thaler, R., 1985, Does the stock market overreact? Journal of Finance, Vol. 40, pp. 793-805

Devereux, M. and Schiantarelli, F., 1990, Investment, Financial Factors and Cash Flow: Evidence from U.K. Panel Data, Asymmetric Information, Corporate Finance, and Investment, pp. 279-306

Drew, M. E. and Veeraraghavan, M., 2002, A Closer Look at the Size and Value Premium in Emerging Markets: Evidence from the Kuala Lumpur Stock Exchange, Asian Economic Journal, Vol. 16, pp. 337-351

Dugan, M. T. and Shriver, K. A., 1989, The Effects of Estimation Period, Industry, and Proxy on the Calculation of the Degree of Operating Leverage, Financial Review, Vol. 24, pp. 109-122

Fama, E. F. and French, K. R., 1992, The cross-section of expected returns, Journal of Finance, Vol. 47, pp. 427-465

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

Fama, E. F. and French, K. R., 1996, Multifactor Explanations of Asset Pricing Anomalies, Journal of Finance, Vol. 51, pp. 55-84

Fama, E. F. and French, K. R., 1998, Value versus Growth: The international Evidence, Journal of Finance, Vol. 53, pp. 1975-1999

(22)

22

García-Fejióo, L. and Jorgensen, R. D., 2010, Can Operating Leverage Be the Cause of the Value Premium?, Financial Management, pp. 1127-1153

Gilchrist, S. and Himmelberg, C. P., 1995, Evidence on the role of cash flow for investment, Journal of Monetary Economics, Vol. 36, pp. 541-572

Gulen, H., Xing, Y. and Zhang, L. , 2011, Value versus Growth: Time-Varying Expected Stock Returns, Financial Management, Vol. 40, pp. 381-407

Haugen, R., 1995, The New Finance: the case against efficient markets. Prentice-Hall: Englewood Cliffs, NJ

Ho, Y. K., Xu, Z. and Yap, C. M., 2004, R&D investment and systematic risk, Accounting and Finance, Vol. 44, pp. 393-418

Lakonishok, J., Shleifer, A. and Vishny, R. W., 1994, Contrarian Investment, Extrapolation, and Risk, Journal of Finance, Vol. 49, pp. 1541-1578

Livdan, D., Sapriza, H. and Zhang, L., 2009, Financially Constrained Stock Returns, Journal of Finance, Vol. 44, pp. 1827-1862

Ljungfist, A. and Wilhelm, W. J., 2005, Does Prospect Theory Explain IPO Market Behavior?, Journal of Finance, Vol. 40, pp. 1759-1790

Lord, R. A., 1996, The Impact of Operating and Financial Risk on Equity Risk, Journal of Economics and Finance, Vol. 20, pp. 27-38

Lord, R. A., 1998, Properties of time-series estimates of degree of leverage measures, The Financial Review, Vol. 33, pp. 69-84

Manelker, G. N. and Rhee, S. G., 1984, The Impact of the Degrees of Operating and Financial Leverage on Systematic Risk of Common Stock, Journal of Financial and Quantitative Analysis, Vol. 19, pp. 45-57

Poulsen, M., Faff, R. and Gray, S., 2013, Financial Inflexibility and the Value Premium, International Review of Finance, Vol. 13, pp. 327-344

Powell, T. C., 1996, How Much Does Industry Matter? An Alternative Empirical Test, Strategic Management Journal, Vol. 17, pp. 323-334

Rajan, R. G. and Zingdales, L., 1995, What do we know about Capital Structure? Some Evidence from International Data, Journal of Finance, Vol. 50, pp. 1421-1460

Whited, T. M. and Wu, G., 2006, Financial Constraints Risk, The Review of Financial Studies, Vol. 19, pp. 531-559

Whited, T. M., 2006, External finance constraints and the intertemporal pattern of intermittent investment, Journal of financial Economics, Vol. 81, pp. 467-502

Referenties

GERELATEERDE DOCUMENTEN

The most useful contributions which polytechnics can make to regional and rural development – and in particular to smart specialisation processes – are those contributions

The list suggests an expansion of the conference to three kinds of tracks, each with their own evaluation criteria: technical solutions to be evaluated on novelty and

From all these overviews, we conclude that there are still challenges to realize the epitaxial piezoMEMS devices, such as, control of the thin film quality, enhancement of the

Because due to the listing, the international community get informed about the violent armed group, and moreover, states listed armed groups as terrorists with the aim

o Your highest educational level Post graduate degree Degree or diploma Post- matric certificate Grade 12 (Matric). Other (If other, please

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication:.. • A submitted manuscript is

U hebt, zo blijkt uit uw conceptbeslissing, het voornemen om alsnog OB-alg te indiceren voor extra begeleiding tijdens het vervoer van en naar de instelling waar verzekerde zijn

Hypothesis 1b that value stocks do not earn, on average, higher size adjusted returns than growth stocks in the Dutch Stock Market between June 1 st , 1981 and May 31 st , 2007