• No results found

The effect of Managers

N/A
N/A
Protected

Academic year: 2021

Share "The effect of Managers"

Copied!
47
0
0

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

Hele tekst

(1)

1

The effect of Managers’ signature from Internal Control Disclosure on Cost of Equity

Master Thesis Yan Shen

University of Groningen: S2272083 Uppsala University: 900131-P191

Email: 872949011@qq.com

Supervisor: Dr. C.L.M. (Niels) Hermes Assessor: Dr. R.B.H. (Reggy) Hooghiemstra

January 10th, 2014

Msc International Financial Management Faculty of Economics and Business

(2)

2

Abstract

This paper investigates the effect of manager’s signature from the internal control disclosure on firm’s cost of equity. The reports signed by manager show that the manager takes great responsibility for internal control disclosure, which is one important aspect of internal control disclosure. I assume the manager’s signature has impact on cost of equity and has different impact on cost of equity between big and small sized firms. Besides, how the internal control disclosure affect the cost of equity is tested. The sample is consisted of 2,039 specific firm years from 29 countries during the years from 2005 to 2007. I find evidence to support the negative relation between cost of equity and internal control disclosure after including six control variables. I do not find more evidence to support the assumption that manager’s signature can influence the firms’ cost of equity. The regressions do not show manager’s signature has different impact on cost of equity between large or small firms either.

Keywords: Manager’s signature; management responsibility; cost of equity; firm size

1. Introduction

An extensive empirical literature explores the relationship between the level of internal control disclosure and cost of equity. According to previous literatures, there is a negative relation between disclosure level and cost of equity (Botosan, 1997; Richardson & Welker, 2001; Ying & Zhengfei, 2006). After the publish of Section 404 of the Sarbanes-Oxley which requires management and the auditor to report on internal controls over financial reporting, the internal control disclosure, as an important part of disclosure is widely discussed. Ashbaugh-Skaife et al., (2009) and Beneish et al., (2008) find that higher internal control disclosure can lead to a lower cost of equity. They argue that in order to achieve a lower level of cost of equity, firms need greater internal control disclosure.

(3)

3

disclosure (IFAC, 2006; Deumes and Knechel, 2008; Ashbaugh et al., 2007), higher internal control disclosure can be achieved by disclosing more of these items, because these items can increase the awareness of shareholder and public and decrease the information asymmetry (e.g., Ashbaugh et al., 2009).

However, the effect of manager’s signature is debated by some scholars. Official statements and some studies (e.g. Hermanson, 2000; Treadway Commission, 1987) state that chief manager’s signature in the management report heighten his incentives to behave responsibly for the financial statements and internal control, which heighten the disclosure level. The higher the awareness is, the greater the attention is paid on the whole internal control procedure. However, some scholars think managers have incentives not to sign their names in internal control disclosure (Schuetze, 1993; McMullen et al., 1996; Raghunandan and Rama, 1994). Managers put their reputation at stake, when they make definitive statements about internal control that later turn out wrong, they may choose not to sign in their reports even if other aspects of internal control disclosure are perfect. Under this circumstance, not signing in the report cannot lead to a low quality of internal control disclosure. Therefore, manager’s signature does not impact the internal control disclosure.

(4)

4

internal control disclosure. In this study, I will focus on how manager’s signatures in internal control disclosures which shows his or her responsibility can affect cost of equity.

My regression analysis contains a sample of 2,039 specific firm years from 29 countries during the years from 2005 to 2007. Firstly, I will do a test to support the negative relation between the level of internal control and the cost of equity. Then, I will test how manager’s signatures influence the cost of equity in firms by using full sample, big firm subsample and small firm subsample. Fixed effect regression and sensitive tests will be used to test the hypothesis.

This paper is structured as follows. In the section 2, I will talk about the existing theory in related area and develop my hypothesis. In section 3, it contains data, sample and methodology. I will show and analysis the result in the section 4. Moreover, the overall conclusions and possibilities for future research will be covered in the last section.

2. Literature Review

2.1 Internal Control Disclosure and Cost of Equity

The cost of equity is a rate of return that a firm theoretically pays to its equity investors, so as to compensate for the risk they undertake by investing their capitals. Firms need to acquire capitals from others to operate and develop; individuals and organizations who are willing to provide their funds to other naturally desire to be rewarded. Lower cost of equity is more likely to be existed in a stable and predictable company than in a risky company. Managers always want to make higher profits, they may try to lower the cost of equity to achieve their goals.

Nevertheless, only American companies are required to make detailed financial disclosure about internal control. Firms in other countries are not mandated to make report on internal control; internal control disclosure is voluntary in these countries. As a result, the levels of internal control differ from each other.

(5)

5

the investors’ estimation costs (e.g. Barry and Brown, 1985; Handa and Linn, 1993). Firms can decrease the compensation for investors by either reducing the transaction costs or reducing the investors’ estimation costs.

The transaction cost theory and the investors’ estimation cost theory are both derived from information problem. Healy and Palepu (2001) think that the information problem is mainly influenced by the level of voluntary internal control disclosure. The information problem arises from information asymmetry between managers and investors. Managers always have more information than the outside investors. Healy and Palepu hold the idea that voluntary internal control disclosure can decrease the cost of equity by reliving the information problem. If the information problem exists in a firm, the investors will demand for information intermediaries, such as: financial analysts, rating agencies or auditing companies. As a result, the transaction

cost will be higher, therefore the cost of equity increased since managers need to pay more

compensation to the investors. On the other hand, firms do not need to pay additional compensation to investors when the risk of estimation is low. Barry and Brown (1985) think that investors require compensation for additional elements of risk when the risk of estimation is high. Firms with high level of disclosure can reduce this type of cost, which derived from information problem.

(6)

6

Asymmetric information causes the adverse selections, which produce the spread prices. They argue that the adverse selection can account for the existence of spread prices. Outsiders will face higher losses from insiders because of adverse selection. To decline the adverse selection, firms can enhance the internal control disclosure since the information asymmetries can induce the adverse selection. The bid-ask spread is declined when firms reduced the adverse selection. Therefore, firms could reduce the transaction cost by increasing the quality of internal control disclosure. Amihud and Mendelson (1986) test the effect of securities’ bid-ask spreads on their returns and find evidence that increasing liquidity of stock can reduce firms’ cost of equity, since the liquidity of stock was negatively related to bid-ask spread, and investors can reduce the transaction cost from the narrower bid-ask spread. Therefore, greater disclosure increases the liquidity of stock, which reduces the cost of equity. Diamond and Verrecchia (1991) state that revealing internal control information can reduce cost of capital by attracting demand from large investors. They also find a positive relation between the disclosure level and the stock liquidity. Higher stock liquidity can attract investors to take large positions in a firm’s stock. Transaction costs are reduced since the company trade is frequently with some particular large investors. Investors are less likely to demand for information intermediaries since trusts are built among these investors.

The investors’ estimation cost theory is analyzed and defined by Handa and Linn (1993). Estimation risk refers to investors’ uncertainty of the parameters of an asset’s return or future payoff. Investors cannot estimate the correct number of firm’s beta when the disclosure level is low, because beta is estimated by investors based on a firm’s financial history and other information. Investors require higher compensate from firms of high estimation risk, firms’ cost of equity increase in this way. Barry and Brown (1985) also study the estimation risk. Their results show better quality of information lower the estimation risk. As I talked above, low estimation risk decrease compensate for investors, which decrease the cost of equity at the same time.

(7)

7

related to the internal control disclosure. However, after controlling the firm characteristics and analyst forecast bias, the relationship becomes insignificant. Ashbaugh-Skaife et al., (2009) utilize unaudited pre-SOX 404 disclosures and SOX 404 audit opinions to assess the effect of SOX internal control deficiencies on cost of equity. They find that firms initially disclose internal control deficiencies can experience an increase in the cost of equity. But a decrease in the cost of equity is shown after a subsequently disclosures. Beneish et al., (2008) examine the relation between internal control weakness and information uncertainty. They use a sample of 336 firms and find that the low quality of internal control disclosure increases the information uncertainty, which also increase the cost of equity. The effect of disclosure on cost of equity is also discussed (Botosan, 1997; Richardson & Welker, 2001). Botosan (1997) studies the association between disclosure and cost of equity capital by using a sample of 122 manufacturing firms. After including the control variables firm size and beta, she finds a negative relationship between cost of equity and disclosure level. Besides these, Richardson and Welker (2001) utilize a Canadian sample during the years from 1990 to 1992, and get same result of the negative relationship between financial disclosure and cost of equity.

Most theories support that there is a negative relation between cost of equity and disclosure. But Berton (2004) claims that the enhanced disclosure can give positive signals to markets. The stock would be traded more frequently because of investors’ confidence, which can add share price volatility. Investors are difficult to make prediction for future payoff, thus enhanced disclosure increase the estimation risk and cost of equity.

From the above, we can see that most studies suggest a negative relation between cost of equity and internal control disclosure level. To do the following research, it is useful to test the relation between cost of equity and internal control by using my sample. So I predict:

Hypothesis 1: The level of firms’ internal control disclosure is negatively related to firms’ cost of

equity.

(8)

8

Managers of American companies are required to take manager’s responsibility for internal control by including their names in the report under the rules of SOX section 404. Managers outside America are not required to put their names in the report, they can decide whether to sign in reports or not.

There is a signal effect of manager’s signature. The credibility of voluntary disclosure is doubt by investors and public, CEOs and managers always have incentives to make self-serving disclosure. They may over-value or under-value their firms’ stock price in their annual report. Besides, managers can also under-value the risks of firm (Frost, 1997). Manager’s signature can increase the credibility and quality of internal control disclosure. Treadway Commission (1987) state that chief manager’s signature in the management report will heighten his or her responsibility for the financial statement and internal control disclosure. COSO (1992) and POB (1993) also state that manager’s signature send a signal to the society and the investors, showing that they are confidence to be monitored by both outsiders and insiders. Additionally, the manager’s signature also sends a signal that the information asymmetry level is low in the internal control. In line with the transaction cost theory, investors will trust the signed disclosure and are less likely to demand for information intermediaries, such as: financial analysts, rating agencies or auditing companies. Besides, insiders have less ability to build a wide bid-ask spread when the information asymmetry is small, which means manager’s signature can decrease the transaction cost and cost of equity. On the other hand, even if disclosure contains detailed information of other internal control items, investors will still look for information intermediaries when they do not find the signatures in the report.

(9)

9

equity will be lower when the litigation cost decrease. As a result, the manager’s signature in the report can heighten the disclosure level, which can reduce the cost of equity at the same time.

However, the usefulness of management responsibility is doubted (e.g. Schuetze, 1993), including manager’s signature, which may not have impact on the level of internal control disclosure and cost of equity. Raghunandan and Rama (1994) and McMullen et al., (1996) both claim that if manager sign his or her name in the internal control disclosure, he or she put their reputation at stake when they make definitive statements about internal control that later turn out to be wrong. Managers do not sign in reports do not mean the quality of reports is low. They may disclose other aspects of internal control perfectly but do not sign in reports, since managers do not want to take the responsible that they cannot control. If managers sign their names in report do not represent a high quality of the internal control disclosure. It is possible that manager do not disclose some important information which related to them when they sign in reports.

Francis et al., (1994) propose the negative litigation cost theory. They find litigation can prevent managers from providing more detail information. Under the high pressure, managers who sign their names are less likely to disclose bad information and future risks. Instead, they are more likely to make misleading or vague disclosure, even underestimating risks, which can decrease the internal control disclosure quality and thus increase the cost of equity. Francis et al., (1994) utilize a sample of firms in the biotechnology, computing, electronics and retailing industries during the years from 1988-1992, who find voluntary internal reports might not be an effective deterrent to litigation.

(10)

10

investments need to acquire equity from other investors, while these investments have high risks. Managers need to pay more compensation for the investors because of additional risks. As a result, manager’s signature can increases the cost of equity.

According to the theories above, we can see that manager’s signature have an impact on the cost of equity. Manager’s signature gives confidence to outside investors, which can decrease the transaction cost and cost of equity. While the management entrench theory hold the idea that managers can make manager-specific investment, which can increase the investor’s estimation risk and cost of equity. In my opinion, only a few managers will do manager-specific investment. If shareholders acknowledge their managers do manager-specific investments that harm their interest, shareholders will punish managers, which can decrease the chances of manager-specific investments. I think manager’s signature in internal control disclosure can reduce the cost of equity.

So I want to test:

Hypothesis 2: Manager’s signature in internal control disclosure has a negative effect on cost of

equity.

Some scholars think firms’ characteristics can influence the decision of manager’s signature. Carcello and Palmros (1994) analyze the relation of litigation and disclosure by using 655 cases of bankruptcy, who find that litigation is more likely to be involved in large firms. The litigation cost is higher in big companies. They also find that managers in big firms are more likely to be monitored in order to avoid litigation costs. So managers in big companies are more likely to sign their names in the reports. Furthermore, firm’s ownership structure is more complex in big firms. Managers in large firms have less chances not to disclose important information which is related to them when they sign their names in disclosure. Sometimes managers do not have the chance to disclose misleading information, since they are monitored by other managers and shareholders (Jensen and Meckling, 1976). So the negative effect of litigation cost is less likely to happen in big firms.

(11)

11

analyze the voluntary management reports on internal control before Section 404 of the Sarbanes-Oxley Act. In their 397 medium-sized firms’ sample, voluntary managers’ signatures are more likely to be included in larger firms.

I think managers of big company face higher litigation costs. Managers are more likely to be affected by positive litigation cost theory rather than negative litigation cost theory. Signatures show managers are confidence in the quality of internal control disclosure. If managers think the quality of internal control disclosure is low, they will choose not to sign in the reports because of high litigation cost. Managers of big firms have less chance to make misleading internal control disclosure or disclose less information, since they are monitored by other managers and shareholders. Managers in big firms sign their names in the report, which can also show positive signal to the public that they are willing to be monitored. Therefore, I want to test:

Hypothesis 3a: Manager’s signature in internal control disclosure has a negative effect on cost of

equity in big companies.

While a low proportion of smaller companies include signatures (McMullen et al., 1996). It is not clear how the manager’s signature affect the cost of equity in small size firms. Managers of small company sign their names in the report do not mean they have confidence that the quality of internal control disclosure is high. It is possible that the litigation cost is smaller in small firms. Managers do not take litigation cost into consideration when they sign their names. As a result, manager’s signature of small firms’ internal control disclosure cannot guarantee the quality of internal control disclosure, which does not influence the cost of equity.

On the other hand, managers in small companies do not sign in reports do not always means a higher cost of equity and bad quality of disclosure. It is possible that managers notice that other small companies do not include signatures of managers too, so they think it is not necessary to sign in internal control disclosure.

So I predict:

Hypothesis 3b: Manager’s signature in internal control disclosure has no effect on cost of equity

(12)

12

3. Data and Methodology

3.1 Data

I use the data sample which is collected by Hooghiemstra et al., (2013). 29 countries are selected differing in GDP and levels of total market capitalization are quite different. So the country sample includes both developed economies and developing economies. Furthermore, the specific firm samples included big-sized firms, medium-sized firms and small-sized firms. Finally, a sample of 4,370 firm-year observations for 1,559 distinct firms during the years from 2005 to 2007 is collected.

To estimate the firms’ cost of equity, four models are used: model of Gebhardt, Lee and Swaminathan (GLS, 2001), model of Claus and Thomas (CT, 2001), model of Ohlson and Juettner-Nauroth (OJ, 2005) and model of Easton (modified price-earnings growth or MPEG, 2004). To make my analysis more comparable, I ensure all observations can run these four models. The data are collected from the Compustat and I/E/B/S. For example, if I can use GLS model to estimate a firm specific year’s cost of equity, but I cannot use OJ model because of data missing, I drop this firm specific year from the sample. At the beginning, GLS model has 2064 observations, CT model has 2148 observations, OJ model has 2489 observations and MPEG model has 2547observations. To ensure all the observations can run these four models, 2039 observations are kept finally. The number of observations in each country is shown in the table

1.

3.2 Independent Variables Internal control items

I follow Hooghiemstra et al. (2013) and separate the internal control disclosure into 7 equally

(13)

13

Manager’s signature (MS)

Manager’s signature (MS) is a dummy variable, which has been discussed in section 2; equal to 1, if managers acknowledge explicitly its responsibility for internal control in the annual report, equal to 0, if managers do not acknowledge its responsibility or do not sign in annual report for internal control. Manager’s signature is included in COSO (1992, 2004), corporate governance codes and the literature (e.g., Deumes and Knechel, 2008).

Strategic and Operational Risk (SOR)

The strategic and operational risk (SOR) contains strategic risk and operational risk. Collins and Ruefli (1992) define the strategic risk as the probability of a firm to lose competition advantage against the other firms. The strategic risks include lower industry profits, lack of new technology, brand erosion, competitors’ pressure, new project failure, and bargain power from customers. Slywotzky and Drzik (2005) regard strategic risk as the biggest risk for a company. The operational risk comes from human error or external events. It focuses on the risks arising from the people, systems, processes and change of outside environment, through which a company operates (Ong, 2006). Most studies combined these two risks. Disclosure of strategic and operation risk is a qualitative estimation. In this paper, SOR is a dummy variable, which equals 1, if a firm disclose its strategic and operational risk; otherwise it equals 0.

Financial Risk (FR)

Disclosure of the financial risk (FR) is a quantitative estimation of forward-looking financial risk (Robert and Gordon, 2005). This item is also included in COSO (1992, 2004), IFAC (2006). It is a dummy variable, it equals 1, if a firm disclose its financial risk; otherwise it equals 0.

Financial Reporting Risk (FRR)

The possibility that the internal control disclosure of a company contains false information is financial reporting risk (FRR). This item is included in COSO (1992, 2004), IFAC (2006). It is a dummy variable, it equals 1, if firm disclose its financial reporting risk; otherwise it equals 0.

(14)

14

The internal control measures (ICM) are the process of identification, assessment and management of company’s risk. Reports include this item will enhance the internal control disclosure by stimulating managers to identify and assess the risk of company from time to time. Internal control measures (ICM) is included in COSO (1992, 2004), and the literature (e.g., Deumes and Knechel, 2008). It is a dummy variable, it equals 1 if firm disclose its internal control measures or activities to control risks; otherwise it equals 0.

Framework (FR)

Framework (FR) helps to identify, assess and manage company’s risk. Providing framework also helps shareholders to assess company’s performance. Section 404 of Sarbanes-Oxley Act of 2002 requires companies to state the frameworks, which are used to evaluate internal control. In the internal control disclosure, the criteria ―Internal control-Integrated Framework‖ should be included to evaluate internal control. This item is included in COSO (1992, 2004), IFAC (2006) and the literature (e.g., Deumes and Knechel, 2008). Framework is a dummy variable, it equals 1 if firm disclose its framework; otherwise it equals 0.

Effectiveness (EF)

Managers’ opinion about the effectiveness of internal control is discussed by a lot of scholars (Ashbaugh-Skaife et al., 2007; Deumes and Knechel, 2008; Bronson et al., 2006). The annual reports or other disclosures should include a statement that controls were effective. Bronson et al., (2006) claim that only a small proportion of companies disclose effectiveness. But this item is useful to decrease the financial statement users’ uncertainty of quality of disclosure. This item is included in COSO (1992, 2004). Effectiveness (EF) is a dummy variable, it equals 1 if the annual report presents an opinion on the effectiveness of internal control; otherwise it equals 0.

3.3 Control variables Firm Size

(15)

15

and the firm size (e.g. Ohlson, 1995; Feltham and Ohlson, 1995). So I predict firm size is negatively related to cost of equity.

Financial Leverage

Financial leverage refers to the amount of debt in the capital structure of the business firm. According to Modigliani and Miller (1958), the use of financial leverage can improve the firm's return on equity and earnings per share. Too much financial leverage, however, can lead to the risk of default and bankruptcy. Fama and French (1992) find a positive relationship between leverage and cost of equity, the higher risk resulting from large financial leverage can lead to

higher cost of equity capital. Richardson and Welker (2001) also find a positive relationship

between cost of equity and financial leverage, though the effect is not significant. I predict financial leverage is positively related with cost of equity.

Market value

Market value is the price, at which an asset would trade in a competitive auction setting. Greenwald et al. (1984) propose that firms with high market value are trusted by investors, and managers’ compensation of risk depends on the current market value. The transaction cost will be reduced by the trust since investors are less likely to find auditor companies and rating agencies, then firms will pay less compensation to the investors, which can reduce the cost of equity.

Prior researches show a significant negative association between market value and cost of equity (e.g. Botosan, 1997; Botosan and Plumlee, 2002). As a result, I predict market value is negatively related to cost of equity.

Margin

(16)

16

Income growth

Bronson et al. (2006) argue that a rapid income growth firm will have less internal control deficiencies, which can lead to low information asymmetries. High quality of internal control disclosure can decrease the information asymmetries, which can narrow the bid-ask spread. In line with the transaction theory, investors are less likely to demand for information intermediaries when the bid-ask spread is small. Therefore, rapid income growth is predicted to decrease the cost of equity.

Sales growth

Bronson et al., (2006) distinguish sales growth from income growth. Sales growth is different from income growth since sales growth has a chance to be unprofitable. Bronson et al., also argue that firms with rapid sales growth are more likely to have internal control deficiencies, which can increase the investors’ transaction cost. Besides, Fazzari et al., (1988) identify sales growth as a significant determinant of capital expenditures and cost of equity capital. Therefore, sales growth is positively related to the cost of equity.

3.4 Dependent Variables: cost of equity

Cost of equity is the dependent variable. Nowadays, scholars seldom use CAPM model to measure cost of equity, instead they use the models of Gebhardt, Lee and Swaminathan (GLS, 2001), Claus and Thomas (CT, 2001), Ohlson and Juettner-Nauroth (OJ, 2005), Easton (modified price-earnings growth or MPEG, 2004), or the average of cost of equity by using these models (e.g. Ogneva et al., 2007; Hou et al., 2012).

CT and GLS model are based on the residual income valuation model. Residual income is the income generated by a firm after accounting for the true cost of capitals. OJ and Easton are derived from abnormal earnings growth-based model.

The formula of the GLS model is:

(17)

17

TV= ∑ ( ) ( ) ( )

( )

Where equal share price at the end of fiscal year t. I collect the data at the end of fiscal year

from Compustat, and find share price from Compustat item 199 by using the data of 7 months after the end of fiscal year.

equal book value of equity of fiscal year t divided by numbers of shares outstanding, and

( ), where dividend payout ratio can be

calculated using the data from Compustat.

, I collect FEPS data from I/B/E/S. It is common that the company

has not release its annual report when I/E/B/S updates its forecast. To ensure that my estimates are based on the public available information, I use 7 months lag to estimate the future earnings

per share. Specifically, = ( )/ . LTG means long term growth

rate, it is reported by I/E/B/S. When LTG is missing, I estimate the long term growth as:

*(

) +.

TV means terminal value with T=12. The formula of the CT model is:

∑ ( ) ( ) ( )( ) ( )( )

Where equal share price at the end of fiscal year t. I collect the data at the end of fiscal year

from Compustat, and find share price from Compustat item 199 by using the data of 7 months after the end of fiscal year.

equal book value of equity of fiscal year t separated by numbers of shares outstanding. In

the CT model, T equals 6. The method to calculate and is same as the method in

GLS model.

g is the future growth rate of residual earnings in perpetuity. To calculate g, I follow Hail & Leuz

(18)

18

use next year’s country-specific median of the realized monthly percentage changes in the consumer price index as a proxy for future inflation. As deflation cannot persist forever, I replace negative values by the country’s historical inflation rate, computed as the median of the monthly inflation rate over the entire sample period. Similarly, I replace values exceeding 10% by the country’s historical inflation rate.

The formula of the OJ model is:

( ) ( )

( )

( )

The method to calculate g, is as same as the method in CT model. Where

is the share price at the end of fiscal year t, I collect the fiscal year end date from Compustat, then plus 7 months and find data from Compustat item 199. k equal common dividends divided by income before extraordinary items, by using Compustat item 21 divided by Compustat item 18.

The formula of MPEG is:

is the share price at the end of fiscal year t, I collect the data at the end of the fiscal year

from Compustat, then plus 7 months and find data from Compustat item 199. FEPS is the future

forecast of earnings per share which was defined previously. is dividend payout in the

time t+1, it equals dividend payout ratio multiply by .

(19)

19

The first hypothesis I want to test is the relationship between cost of equity and internal control disclosure. First, I test the bivariate relationship of internal control disclosure and cost of equity. Second, to check the sensitive of the variables, I add control variables by two steps. First of all, I add ln-assets, financial leverage and market value. Secondly, I add all the control variables at the same time. I use the panel fixed effect regression to test my hypothesis. The full empirical model can be denoted as follows:

(1)

Where:

is the cost of equity estimated by the four models which have mentioned before -

, and I add as the average of the four model. i means years, j means

specific company, and m represents models of cost of equity.

is the disclosure level of firm j in the year i. The detail explanations are shown in section

3.2.

is the ln asset of firm j in year i, I measure firm size as the natural logarithm of the

firm’s total assets (Compustat item 6) in that year. It is a continuous variable.

is the firm j’s financial leverage in the year i. I measure firm’s financial

leverage by using the natural logarithm of total long-term debt (Compustat item 9) divided by the natural logarithm of stockholder equity (Compustat item 216).

is the market value of firm j in the year i. I use the nature logarithm of the

market value of outstanding common equity (Compustat item 216) in the year i-1 divided by pricing data in December before year i. I find the pricing data from I/E/B/S, SAS format.

is firm j’s profit margin in the year i, it is calculated by using the natural logarithm of

(20)

20

is compound five-year income growth rate of firm j, the five years’ income

before extraordinary data can be found in Compustat item 18.

, IC means the income before extraordinary item.

is compound five-year sales growth rate of firm j, the five years’ sales data can

be found in Compustat item 21. =

-1.

(21)

21

is a dummy variable, if manager sign his or her name in internal control disclosure of firm

j in the year i, =1; 0 otherwise.

is an interaction term. It tests the effect of firm size on the relationship

between manager’s signature and cost of equity.

, , , , and are defined before.

I use the natural logarithm of the firm’s total assets (Compustat item 6) to separate big and small companies, I calculate the median of the total assets for the entire firm of the specific years, 1,020 firm specific years’ assets are larger than median number; 1,019 firm specific years’ assets are smaller than median number.

4. Results

The table 2 presents the characteristics of dependent and independent variables. The full

sample includes 2,039 observations.

The first five lines provide the descriptive statistics for the implied cost of equity estimates. We can see that cost of equity under the GLS model has the lowest number which is 11.3%. This result is consistent with Gode & Mohanram (2003), Hail & Leuz (2006) and Boubakri et al.

(2010), their estimation of Rgls show the GLS model produce the lowest cost of equity. The

average of four models is 11.9%, which is consistent with other papers (12.44 % in Boubakri et

al. (2010); 12.49% in Hail & Leuz (2006)).

(22)

22

growth, sales growth, margin and market value. So the extreme values of bottom 1% and top 1% are discarded in these variables.

The table 3 shows the correlations between the dependent and independent variables in the

full sample and two subsamples.

We can see that the correlations between five models are high, according to Hail & Leuz (2006) and Boubakri et al., (2010), this phenomenon is normal. For example, in Boubakri et al.,

(2010)’s work, the correlation between and reaches 0.948. These models’ aim is to

estimate cost of equity, and all of them use the data of , so they should have high

correlations.

Additionally, correlations among seven internal control disclosure items are not high, none of them reach 0.5. But these items have high correlation with the total internal control disclosure, since these items are the aspects of internal control. Besides, the correlations between manager’s signature and estimated cost of equity in the full sample and the big firm samples are highest compared with other internal control items, which show the relative importance of manager’s signature.

Finally, other independent variables do not show high correlations, the correlations among firm size, firm leverage, income growth, sales growth, margin and market value do not exceed 0.3. Therefore, the correlation matrix indicates that multicollinearity should not be a problem.

The table 4 provides test of the hypothesis 1. In the panel A, I present the univariate analysis

(23)

23

disclosure to some extent. We can see that the P –values decrease by including all control variables. As a result, I find evidence to support the hypothesis 1 after including the control variables. The results suggest that there is a negative relationship between internal control disclosure and cost of equity. Firms can decrease the cost of equity by including more internal control disclosure items in reports. It is consistent with the transaction cost and the estimation risk theories. When the level of information asymmetry decrease, the bid-ask spread is lower. Firms do not need to pay much compensation to investors, since investors do not demand for information intermediaries (e.g. Copeland and Galai, 1983). Additionally, firms do not need to pay much compensation to investors when the estimation risk is low. The higher internal control disclosure can decrease the estimation risk (e.g. Handa and Linn, 1993). The negative association between the cost of equity and internal control disclosure is also consistent with previous empirical findings (e.g. Ashbaugh-Skaife et al., 2009; Beneish et al., 2008; Botosan, 1997). As a result, to achieve a low level of cost of equity, firms need greater voluntary internal control disclosure.

As far as the control variables are concerned, firm size, financial leverage, market value and

income growth provide significant results in the panel C. Firm size is positively related to the

cost of equity. The results suggest that big size firms have less cost of equity. The coefficient is positive and significant at 10% level under GLS model, the result is not consistent with previous studies (e.g. Feltham and Ohlson, 1995). The significant and positive association between

financial leverage and cost of equity is shown under the OJ and the MPEG model. The positive

effect is consistent with the previous theories as well as my prediction (Modigliani and Miller, 1958). Market value shows strong negative relation with cost of equity, the results are significant at 1% level under all the models. So the results suggest that market value is negatively related to cost of equity, it is in line with the previous studies as well as my prediction (e.g. Botosan, 1997). Furthermore, there is a nearly persistent negative relation between cost of equity and income

growth. The coefficients are significant at 5% level under the OJ and the MEPG models, which

means the rapider the income growth, the lower the cost of equity.

(24)

24

the panel C, manager’s signature has no effect on cost of equity in small size firms. In the table

5(2), I add ln asset, financial leverage and market value. The coefficients of manager’s signature

remain insignificant, and p-values of manager’s signature do not show a show a downward trend. The results cannot support the hypothesis 2 and 3a. Then I run regression by including all control variables with an interaction term. The P-values of manager’s signature do not show a show a downward trend either. As a result, I do not find evidence to support the hypothesis 2 and 3a. The argument manager’s signature has a negative effect on cost of equity cannot be supported. The insignificant results confirm the hypothesis 3b, which suggests that manager’s signature does not affect the cost of equity in small size firms. It is consistent with previous literatures (e.g. McMullen et al., 1996; Schuetze, 1993).

Including manager’s signature in small firms cannot lead to a high level of internal control disclosure. Managers in small companies do not face high litigation cost, they are less likely to take litigation cost into consideration when they sign their names. As a result, signatures of small firms’ internal control disclosure cannot guarantee the quality of internal control disclosure, which do not influence the cost of equity.

While the results of hypothesis 2 and 3a do not consistent with positive litigation theory and

signal effect theory (e.g. Skinner, 1994; Khurana and Raman, 2004; COSO, 1992). It seems that

(25)

25

In the table 5(3), ln asset, financial leverage, market value and income growth show significant results. The control variable ln asset shows a positive association with cost of equity in small size firms and full samples; and significant at 5% level under the GLS model in the full sample and at 10% level under GLS and CT models in small firms’ sample, which means the smaller firms are more likely to have lower cost of equity, which is not consistent with previous findings (e.g. Ohlson, 1995). The coefficients of firm size are not significant in the large firm sample. The results suggest that firm size only impact the cost of equity in the small size firms. The compensation to investors is influenced by the number of investors. I think the small companies have fewer outside investors than big ones. Therefore the small companies have less cost of equity.

In the full sample of table 5(3), two positive and significant coefficients of financial leverage are showed. Which suggest that larger leverage increase the risk as well as the cost of equity (e.g. Fama & French, 1992). While in the sample of big size firms and small size firms, the positive and significant relationship between financial leverage and cost of equity are also shown. Firms with large leverage decrease the trust from investor. Therefore investors are more likely to demand for the information intermediaries which increase the compensation. Furthermore, big financial leverage increases the risk of investors. Investors require higher compensation for the additional risk they take. As a result, financial leverage is positively related to cost of equity.

Market value shows significant and negative relation with cost of equity in all the samples;

especially in the full sample, all the coefficients are significant at 1% level. The results are consistent with my negative prediction and previous literature (Greenwald et al., 1984). Therefore, a firm with high market value is more likely to be trusted by outsiders, which can decrease the transaction cost and cost of equity.

(26)

26

In the table 5 (3), the signs of interaction term show confusing results, three coefficients are positive while other two are negative. But no significant results are provided, which means the relationship between manager’s signature and cost of equity is not affected by firm size. Consistent with the results in the table 4, profit margin and sales growth do not provide significant results too. As a result, the interaction term, profit margin and sales growth are not important variables in my sample.

5. Conclusions, Limitations and Future Research

This paper has examined how internal control disclosure and manager’s signature affect cost of equity of 2,039 specific firm years from 29 countries during the years 2005 to 2007. In the first hypothesis, I argue that internal control disclosure is negatively related to cost of equity of firm. Since greater internal control disclosures can reduce the transaction costs and investors’ estimation costs, which decrease the compensation to investors. (e.g. Copeland and Galai, 1983; Barry and Brown, 1985). In the second hypothesis, I argue that manager’s signature has a negative effect on the cost of equity. In the third hypothesis, I propose that manager’s signature has a negative effect on the cost of equity in big companies, while manager’s signature has no effect on the cost of equity in small firms. Big firms face higher litigation cost, and managers have incentive to disclose more information to reduce the litigation cost when they take responsibility to the internal control (Carcello and Palmros, 1994). While for managers in small firms, they sign or not sign in report do not impact the internal control disclosure.

I find evidence to support the hypothesis 1 after six control variables are included. The amount of internal control information firms present is negatively related to cost of equity. This result is consistent with transaction cost theory, investors’ estimation risk theory and empirical studies (e.g. Copeland and Galai, 1983; Glosten and Milgrom, 1985; Barry and Brown, 1985; Handa and Linn, 1993; Ashbaugh-Skaife et al., 2009; Beneish et al., 2008).

(27)

27

(e.g. Schuetze, 1993). Besides, the impact of manager’s signature shows no differences between the big size firms and small size firms. It suggests that firm size does not influence the relationship between cost of equity and manager’s signature.

Financial leverage, market value and income growth are important control variables in my

regressions. Financial leverage has a positive effect on the cost of equity. Large financial leverage can lead to the risk of default and bankruptcy, and the higher risk resulting from large financial leverage which can increase the cost of equity (Modigliani and Miller, 1958; Fama and French, 1992). Besides, market value shows a strong negative effect on cost of equity. Firms with high market value are trusted by investors, which could decrease the compensation for investors, thus decrease the cost of equity (Greenwald et al., 1984). Income growth is also an important factor, which is negatively related with cost of equity. Since rapid income growth firm have less internal control deficiencies, which can decrease investors’ transaction cost (Bronson et al., 2006). Besides these three factors, firm sizes affect the cost of equity in smaller size firms. The small firms tend to have lower cost of equity, which is contrast to my prediction.

This thesis provides some evidences to support the negative relationship between the cost of equity and the level of internal control disclosure level. Besides, it enriches the empirical studies of internal control disclosure and cost of equity. Furthermore, a broader sample is used to test this relationship outside America in this thesis. It suggests that besides American firms, firms in other countries can also achieve low cost of equity by increasing the quality of internal control disclosure. Furthermore, it enriches the theories of the relation between the cost of equity and manager’s signature, though no results are found in this field. The results suggest that manager’s signature does not have a direct effect on cost of equity. Manager’s signature cannot stop investors from demanding for information intermediate. On the other hand, manager’s signature does not have a big impact on the quality of internal control disclosure. Managers do not take litigation cost into consideration when they sign their names in reports. Their signatures do not guarantee the quality of internal control disclosure.

(28)

28

do not show significant results in the regressions. But it does not mean that they are not important factors. It is possible that these two variables have significant effect in America or some countries in the sample which I use.

Some limitations still exist in my paper. First of all, the previous literatures focus on the firms in a single country such as the US or Canada (e.g. Francis et al., 2005; Botosan, 1997). My analysis contains 29 countries outside the US and Canada, some countries have large samples, while others have relative small samples. The country biases may exist in my sample, which may influence the estimation of cost of equity and other variables. Secondly, the data of dependent variables internal control disclosures are collected from annual reports and other material reports. Probably, managers make misleading or false statements in the reports but I can not find. Therefore, the level of internal control disclosure can be overestimated.

(29)

29

Reference

Amihud, Y., & Lev, B. (1981). Risk reduction as a managerial motive for conglomerate mergers. The bell journal of economics, 605-617.

Amihud, Y., & Mendelson, H. (1986). Asset pricing and the bid-ask spread. Journal of financial

Economics, 17(2), 223-249.

Ashbaugh, S., Collins, D. W., & Lafond, R. (2009). The effect of SOX internal control deficiencies on firm risk and cost of equity. Journal of Accounting Research, 47(1), 1-43. Ashbaugh-Skaife, H., Collins, D. W., & Kinney Jr, W. R. (2007). The discovery and reporting of

internal control deficiencies prior to SOX-mandated audits. Journal of Accounting and

Economics, 44(1), 166-192.

Barry, C., & Brown, S. J. (1985). Differential information and security market equilibrium. Journal of Financial and Quantitative Analysis, 20(4), 407-422.

Beneish, M. D., Billings, M. B., & Hodder, L. D. (2008). Internal control weaknesses and information uncertainty. The Accounting Review, 83(3), 665-703.

Berton, L. 1994. Companies pressure accounting panel to modify demands for more data. Wall

Street Journal, 72(3), 323-349.

Botosan, C. A. (1997). Disclosure level and the cost of equity capital. Accounting review, 323-349.

Botosan, C. A., & Plumlee, M. A. (2002). A re-examination of disclosure level and the expected cost of equity capital. Journal of accounting research, 40(1), 21-40.

Botosan, C. A., & Stanford, M. (2005). Managers' motives to withhold segment disclosures and the effect of SFAS No. 131 on analysts' information environment. The Accounting

Review, 80(3), 751-772.

(30)

30

Bronson, S. N., Carcello, J. V., & Raghunandan, K. (2006). Firm characteristics and voluntary management reports on internal control. Auditing: A Journal of Practice & Theory, 25(2), 25-39.

Carcello, J. V., & Palmrose, Z. V. (1994). Auditor litigation and modified reporting on bankrupt clients. Journal of Accounting Research, 32, 1-30.

Claus, J., & Thomas, J. (2001). Equity premia as low as three percent? Evidence from analysts' earnings forecasts for domestic and international stock markets. The Journal of

Finance, 56(5), 1629-1666.

Collins, J. M., & Ruefli, T. W. (1992). Strategic risk: An ordinal approach. Management

Science, 38(12), 1707-1731.

Committee of Sponsoring Organizations of the Treadway Commission (COSO). (1992). Internal

Control—Integrated Framework. New York: AICPA

Committee of Sponsoring Organizations of the Treadway Commission (COSO). (2004).

Enterprise risk management—Integrated framework. New York: AICPA.

Copeland, T. E., & Galai, D. (1983). Information effects on the bid-ask spread. The Journal of

Finance, 38(5), 1457-1469.

Deumes, R., & Knechel, W. R. (2008). Economic incentives for voluntary reporting on internal risk management and control systems. Auditing: A Journal of Practice & Theory, 27(1), 35-66.

Diamond, D. W., & Verrecchia, R. E. (1991). Disclosure, liquidity, and the cost of capital. The

journal of Finance, 46(4), 1325-1359.

Easton, P. D. (2004). PE ratios, PEG ratios, and estimating the implied expected rate of return on equity capital. The accounting review, 79(1), 73-95.

Fama, E. F., & French, K. R. (1992). The cross-section of expected stock returns. the Journal of

Finance, 47(2), 427-465.

Fazzari, S., Hubbard, R. G., & Petersen, B. C. (1988). Financing constraints and corporate investment. Brookings Papers on Economic Activity, Vol. 1,141-195

(31)

31

Francis, J., Philbrick, D., Schipper, K. (1994). Shareholder litigation and corporate disclosures.

Journal of Accounting Research.32, 137-165.

Frost, C.A. (1997). Disclosure policy choices of UK firms receiving modified audit reports.

Journal of Accounting and Economics 23, 163-188.

Gebhardt, W. R., Lee, C., & Swaminathan, B. (2001). Toward an implied cost of capital. Journal

of accounting research, 39(1), 135-176.

Glosten, L. R., & Milgrom, P. R. (1985). Bid, ask and transaction prices in a specialist market with heterogeneously informed traders. Journal of financial economics, 14(1), 71-100. Gode, D., & Mohanram, P. (2003). Inferring the cost of capital using the Ohlson–Juettner

model. Review of accounting studies, 8(4), 399-431.

Greenwald, B., Stiglitz, J. E., & Weiss, A. (1984). Informational imperfections in the capital market and macroeconomic fluctuations. The American Economic Review, 74(2), 194-199. Hail, L., & Leuz, C. (2006). International differences in the cost of equity capital: Do legal

institutions and securities regulation matter?. Journal of accounting research, 44(3), 485-531. Handa, P., & Linn, S. C. (1993). Arbitrage pricing with estimation risk. Journal of Financial and

Quantitative Analysis, 28(01), 81-100.

Healy, P. M., & Palepu, K. G. (2001). Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature. Journal of accounting and

economics, 31(1), 405-440.

Hermanson, H. M. (2000). An analysis of the demand for reporting on internal control.

Accounting Horizons, 14(3), 325-341.

Hooghiemstra, R.B., N. Hermes, N., and J. Emanuels (2013), A Cross-Country Study of the Direct and Indirect Effects on National Culture on Internal Control Disclosures: An Agency Perspective, Unpublished working paper, Groningen: University of Groningen.

Hou, K., Van Dijk, M. A., & Zhang, Y. (2012). The implied cost of capital: A new approach. Journal of Accounting and Economics, 53(3), 504-526.

IFAC. (2006). About IFAC. International Federation of Accountants, available at: http://www.ifac.org/About/#Activities

(32)

32

Khurana, I. K., & Raman, K. K. (2004). Litigation risk and the financial reporting credibility of Big 4 versus non-Big 4 audits: Evidence from Anglo-American countries. The Accounting

Review, 79(2), 473-495.

McMullen, D. A., Raghunandan, K., & Rama, D. V. (1996). Internal control reports and financial reporting problems. Accounting Horizons, 10, 67-75.

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.

Modigliani, F., & Miller, M. H. (1963). Corporate income taxes and the cost of capital: a correction. The American Economic Review, 53(3), 433-443.

National Commission on Fraudulent Financial Reporting (the Treadway Commission). (1987). Report of the National Commission on Fraudulent Financial Reporting. Washington,

D.C.:Government Printing Office

Nissim, D., & Penman, S. H. (2001). Ratio analysis and equity valuation: From research to practice. Review of accounting studies, 6(1), 109-154.

Ogneva, M., Subramanyam, K. R., & Raghunandan, K. (2007). Internal control weakness and cost of equity: evidence from SOX Section 404 disclosures. The Accounting Review, 82(5), 1255-1297.

Ohlson, J. A. (1995). Earnings, book values, and dividends in equity valuation. Contemporary

accounting research, 11(2), 661-687.

Ohlson, J. A., & Juettner-Nauroth, B. E. (2005). Expected EPS and EPS growth as determinants of value. Review of accounting studies, 10(2-3), 349-365.

Ong, M. K. (2006). Risk management: a modern perspective. Academic Press/Elsevier, 1-740 Public Oversight Board (POB). 1993. In the Public Interest. Stamford: POB.

(33)

33

Robert, F. & Gordon, T (2005). Financial statement disclosure: the needs and practices related to financial risk. Society of Actuaries, 1-115.

Richardson, A. J., & Welker, M. (2001). Social disclosure, financial disclosure and the cost of equity capital. Accounting, Organizations and Society, 26(7), 597-616.

Schuetze, W. (1993). Reporting by independent auditors on internal controls. CPA Journal (October): 40

Shleifer, A., & Vishny, R. W. (1989). Management entrenchment: The case of manager-specific investments. Journal of financial economics, 25(1), 123-139.

Skinner, D., (1994). Why firms voluntarily disclose bad news. Journal of Accounting Research 32, 38-61

Slywotzky, A. J., & Drzik, J. (2005). Countering the biggest risk of all. Harvard Business

Review, 83(4), 78.

(34)

34

Table 1: Observations in each country Country

AUS 87 ESP 54 JPN 229

AUT 11 FIN 90 KOR 62

BRA 94 FRA 160 MEX 42

CHE 93 GBR 213 MYS 70

CZ 11 GRC 74 NZL 15

DEU 138 HUN 18 POL 31

DNK 39 IDN 61 RUS 9

ITA 86 IND 61 SGP 26

SWE 85 THA 74 TUR 54

TWN 13 ZAF 39 ALL 2039

(35)

35

Table2:Characteristic of data set

Mean Median Maximum Minimum Std.Dev Skewness Kurtosis obervations

(36)

36

Table 3: Correlation between variables

Panel A: Full Sample

RMEA N

RMPE G

(37)

37

Panel B: Small firms

(38)

38

Panel C: Big firms

Panel C: big firms RMEA N RMPE G

ROJ RCT RGLS MS SOR FR FRR ICM FW EF ICDIS C Ln asset leverage Income growth Sales growth Margi n Market value RMEAN 1.000 RMPEG 0.881 1.000 ROJ 0.863 0.900 1.000 RCT 0.712 0.398 0.364 1.000 RGLS 0.848 0.650 0.623 0.530 1.000 MS 0.113 0.093 0.116 0.060 0.107 1.000 SOR -0.041 -0.024 -0.017 -0.035 -0.058 0.074 1.000 FR -0.002 0.000 -0.010 -0.014 0.020 0.161 0.339 1.000 FRR -0.025 0.027 0.009 -0.083 -0.026 0.135 0.257 0.259 1.000 ICM 0.029 0.050 0.049 -0.012 0.015 0.256 0.287 0.467 0.205 1.000 FW 0.022 0.018 0.060 0.006 -0.011 0.208 0.101 0.177 0.160 0.259 1.000 EF 0.016 0.028 0.043 -0.012 -0.003 0.349 0.128 0.176 0.150 0.234 0.237 1.000 ICDISC 0.026 0.046 0.061 -0.025 0.009 0.577 0.524 0.614 0.549 0.662 0.533 0.542 1.000 Ln asset -0.070 -0.042 -0.059 -0.030 -0.100 -0.082 0.074 -0.201 0.090 -0.196 -0.124 -0.106 -0.132 1.000 Leverage 0.085 0.110 0.115 0.027 0.033 0.011 -0.029 -0.011 -0.002 0.071 0.049 -0.045 0.011 0.032 1.000 Income growth -0.020 -0.027 -0.022 -0.011 -0.006 -0.007 0.003 0.019 -0.047 0.022 -0.011 -0.004 -0.008 -0.036 -0.047 1.000 Sales growth -0.013 -0.015 -0.030 0.001 0.001 -0.078 -0.056 0.035 0.020 -0.051 -0.085 -0.055 -0.068 0.009 -0.084 -0.001 1.000 Margin -0.030 -0.023 -0.023 -0.028 -0.025 0.038 0.018 0.015 0.033 0.019 0.057 0.042 0.059 0.015 -0.019 0.000 -0.053 1.000 Market value -0.206 -0.154 -0.205 -0.128 -0.192 -0.158 0.059 -0.004 0.115 -0.023 0.066 -0.132 -0.010 -0.081 -0.050 -0.011 0.038 0.019 1.000

(39)

39

Table 4: panel fixed effect regression of the relation between internal control disclosure and cost of equity

Panel A:simple regression including internal control disclosure

Prediction Rmean Rgls Rct Roj Rmpeg

Intercept 0.123 (0.00) 0.117 (0.00) 0.127 (0.00) 0.126 (0.00) 0.122 (0.00) icdisc - -0.001 (0.31) -0.001 (0.28) -0.0002 (0.88) -0.001 (0.37) -0.001 (0.27) R-squared 0.75 0.77 0.58 0.72 0.69 Adjusted R-squared 0.53 0.57 0.22 0.47 0.43 Observations 2039 2039 2039 2039 2039 P-value in brackets p<0.01***, p<0.05** and p<0.1*

(40)

40

Panel B: regression including internal control disclosure, ln asset, financial leverage and market value

Prediction Rmean Rgls Rct Roj Rmpeg

Intercept -0.018 (0.87) -0.09 (0.50) -0.095 (0.63) 0.064 (0.68) 0.047 (0.74) icdisc - -0.001 (0.16) -0.001 (0.12) -0.0006 (0.74) -0.001 (0.27) -0.001 (0.15) ln asset - 0.010* (0.09) 0.016** (0.02) 0.0149 (0.17) 0.005 (0.51) 0.006 (0.43) financial leverage + 0.005 (0.27) -0.009 (0.11) -0.002 (0.77) 0.019*** (0.00) 0.013** (0.02) market value - -0.027*** (0.00) -0.042*** (0.00) -0.022*** (0.00) -0.021*** (0.00) -0.021*** (0.00) R-squared 0.77 0.79 0.60 0.74 0.72 Adjusted R-squared 0.56 0.60 0.41 0.50 0.47 Observations 1774 1774 1774 1774 1774 P-value in brackets p<0.01***, p<0.05** and p<0.1*

(41)

41

Panel C: fixed effect regression of the relation between cost of equity and internal control disclosure including all control variables

Prediction Mean GLS OJ CT MPEG

C ? 0.043 (0.72) -0.034 (0.80) 0.19 (0.22) -0.13 (0.52) 0.14 (0.36) Icdisc - -0.001 (0.14) -0.002 (0.13) -0.001 (0.26) -0.0005 (0.76) -0.002* (0.09) Ln asset - 0.008 (0.24) 0.014* (0.07) -0.001 (0.85) 0.017 (0.12) 0.001 (0.84) Financial leverage + 0.006 (0.19) -0.008 (0.17) 0.023*** (0.00) -0.003 (0.65) 0.016** (0.01) Market value - -0.03*** (0.00) -0.005*** (0.00) -0.023*** (0.00) -0.026*** (0.00) -0.025*** (0.00) Margin - -0.0002 (0.11) -0.000001 (0.37) 0.000002 (0.15) -0.00002 (0.30) -0.00002 (0.14) Income growth ? -0.001 (0.29) 0.0001 (0.96) -0.005** (0.02) -0.002 (0.31) -0.004** (0.02) Sales growth + 0.0007 (0.88) -0.004 (0.48) -0.0001 (0.98) 0.0076 (0.37) -0.0002 (0.96) R-squared 0.77 0.79 0.74 0.69 0.73 Adjusted R-squared 0.55 0.59 0.51 0.40 0.47 Observation 1624 1624 1624 1624 1624 P-value in brackets p<0.01***, p<0.05** and p<0.1*

(42)

42

Table 5 (1) simple panel fixed effect regression of the relation between MS and cost of equity Panel A: Full sample

Prediction Rmean Rgls Rct Roj Rmpeg

c 0.119 (0.00) 0.114 (0.00) 0.125 (0.00) 0.121 (0.00) 0.117 (0.00) MS - -0.0005 (0.87) -0.003 (0.38) 0.002 (0.71) 0.0002 (0.96) -0.0009 (0.83) R-squared 0.75 0.77 0.58 0.72 0.69 Adjusted R-squared 0.53 0.57 0.22 0.47 0.42 Observations 2039 2039 2039 2039 2039

Panel B: Big firms

Prediction Rmean Rgls Rct Roj Rmpeg

c 0.109 (0.00) 0.104 (0.00) 0.116 (0.00) 0.110 (0.00) 0.107 (0.00) MS - 0.002 (0.62) -0.001 (0.74) 0.008 (0.26) 0.001 (0.77) 0.000 (0.93) R-squared 0.74 0.77 0.59 0.72 0.68 Adjusted R-squared 0.49 0.54 0.19 0.46 0.36 Observations 1019 1019 1019 1019 1019

Panel C:Small firms

Prediction Rmean Rgls Rct Roj Rmpeg

(43)

43

p<0.01***, p<0.05** and p<0.1*

This table presents results obtained by regression of the different models’ cost of equity on manager’s signature.

Table 5 (2) simple panel fixed effect regression of the relation between MS and cost of equity Panel A: Full sample

Prediction Rmean Rgls Rct Roj Rmpeg

c -0.018 (0.87) -0.095 (0.48) -0.093 (0.63) 0.065 (0.67) 0.046 (0.75) MS - -0.001 (0.73) -0.004 (0.29) 0.0009 (0.88) 0.0001 (0.97) -0.001 (0.71) ln asset - 0.010 (0.10) 0.016** (0.02) 0.014 (0.18) 0.005 (0.54) 0.006 (0.46) Financial leverage + 0.005 (0.27) -0.009 (0.11) -0.002 (0.77) 0.019*** (0.00) 0.013** (0.02) market value - -0.026*** (0.00) -0.04*** (0.00) -0.022*** (0.00) -0.021*** (0.00) -0.020*** (0.00) R-squared 0.77 0.79 0.59 0.74 0.72 Adjusted R-squared 0.55 0.60 0.31 0.50 0.47 Observations 1774 1774 1774 1774 1774

Panel B: Big firms

Prediction Rmean Rgls Rct Roj Rmpeg

(44)

44

Table 5 (2) simple panel fixed effect regression of the relation between MS and cost of equity (continued)

Panel B: Big firms

Prediction Rmean Rgls Rct Roj Rmpeg

market value - -0.023*** (0.00) -0.043*** (0.00) -0.022** (0.03) -0.023*** (0.00) -0.013* (0.06) R-squared 0.77 0.80 0.60 0.77 0.72 Adjusted R-squared 0.53 0.60 0.30 0.53 0.44 Observations 957 957 957 957 957

Panel C: Small firms

Prediction Rmean Rgls Rct Roj Rmpeg

c 0.0003 (0.99) -0.177 (0.30) -0.238 (0.33) 0.235 (0.24) 0.177 (0.34) MS ? -0.003 (0.55) -0.003 (0.61) -0.009 (0.35) -0.0007 (0.92) -0.0001 (0.98) ln asset - 0.012 (0.17) 0.025** (0.02) 0.027* (0.09) -0.002 (0.86) 0.0008 (0.94) Financial leverage + -0.0006 (0.51) -0.02* (0.06) -0.004 (0.77) -0.0001 (0.99) 0.0006 (0.95) market value - -0.031*** (0.00) -0.044*** (0.00) -0.024** (0.02) -0.029*** (0.00) -0.028*** (0.00) R-squared 0.80 0.80 0.60 0.76 0.75 Adjusted R-squared 0.61 0.62 0.31 0.54 0.52 Observations 817 817 817 817 817 P-value in brackets p<0.01***, p<0.05** and p<0.1*

(45)

45

Table 5 (3) panel fixed effect regression of the relation between MS and cost of equity including all control variables Panel A: Full sample

Prediction Rmean Rgls Rct Roj Rmpeg

c 0.044 (0.72) -0.043 (0.76) -0.117 (0.57) 0.190 (0.22) 0.137 (0.38) MS - -0.011 (0.65) -0.006 (0.81) -0.047 (0.24) -0.001 (0.96) -0.002 (0.93) MS* ln asset 0.0004 (0.71) -0.0006 (0.68) 0.002 (0.22) 2.93E-05 (0.98) -0.0001 (0.94) Ln asset - 0.007 (0.27) 0.014* (0.07) 0.016 (0.15) -0.001 (0.83) 0.001 (0.85) Financial leverage + 0.006 (0.19) -0.008 (0.18) -0.004 (0.65) 0.023*** (0.00) 0.016** (0.01) Market value - -0.030*** (0.00) -0.046*** (0.00) -0.026*** (0.00) -0.023*** (0.00) -0.024*** (0.00) Margin - -2.21E-06 (0.12) -1.37E-06 (0.40) -2.38E-06 (0.31) -2.57E-06 (0.16) -2.52E-06 (0.15) Income growth + -0.001 (0.30) 9.49E-05 (0.96) 0.003 (0.30) -0.005** (0.02) --0.004** (0.02) Sales growth + 0.0009 (0.85) -0.004 (0.48) 0.008 (0.35) -4.79E-05 (0.99) -0.0001 (0.98) R-squared 0.77 0.79 0.60 0.75 0.73 Adjusted R-squared 0.55 0.58 0.40 0.50 0.47 Observations 1624 1624 1624 1624 1624

Panel B:Big firms

Prediction Rmean Rgls Rct Roj Rmpeg

Referenties

GERELATEERDE DOCUMENTEN

Either all the organizational costs are allocated to projects, or the present value of a project has to be definitely positive in order to keep the organization going.. Some

The volume effect of changes in the index is studied by Harris and Gurel (1986), Beneish and Whaley (1996) and Chen et al. In this thesis, both effects are analyzed for

The British passenger market consists of a number of different operators, the domestic operators are all private, whereas the foreign operators are dominated by state

In werklikheid was die kanoniseringsproses veel meer kompleks, ’n lang proses waarin sekere boeke deur Christelike groepe byvoorbeeld in die erediens gelees is, wat daartoe gelei

Active efflux pump systems extrude noxious compounds and antibiotics from the cell reducing their intracellular concentration (Rodrigues et al. 2011), therefore

Key Words: Economic Sanction, Import Restrictions, Export performance, Iran, Directly Imposed Sanctions, Globalization, Panel Data, Case Study, TIES

In line with human capital theory, it was argued that managers with a higher level of education are more corrupt, since education increases a person's stock

The lower limit of the band is determined by the lowest (unweighted) average asset beta (either on the basis of weekly data, or on the basis of daily data), while the upper limit