University of Amsterdam, Amsterdam Business School
MSc Business Economics, Finance track
Master’s Thesis Finance
CEO’s Optimism Impact on Firm Performance
Meng Zhao (11005769)
July, 2016
Thesis supervisor: Dr. Liang Zou
Statement of Originality
This document is written by Student [Meng Zhao] who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Acknowledgements
This thesis is written for the course Mater’s Thesis Finance in the second semester of academic year 2015-2016 and conclude my Finance-specialization in the MSc Business Economics, Finance track at University of Amsterdam, Amsterdam Business School. I highly indebted to my supervisor, Dr. Liang Zou, for giving useful comments and feedback. In addition, I would like to thank Dr. Florian Peters, which gave me guidance to complete this master thesis.
Abstract
Does managerial optimism help to explain the different firm performance? Panel fixed-effects econometric framework is used to investigate the impact of CEO optimism on firm performance in the United States. I regress managerial optimism against a set of accounting- and marketing-based measurements of firm performance. As such, I contribute a complete model with more control variables regarding CEO characteristics and accounting variables. To test these hypotheses, I use options-based measurement to define high- and low-optimism. However, the managerial optimism has a different impact on the firm performance, the results illustrate that optimistic CEOs can improve firm performance, but the low optimistic CEOs cannot.
JEL Classification: G02, G32, J33, L25
Keywords: Optimism, CEO Compensation, CEO Ownership, Firm Performance, Behavioral Finance
CONTENT STATEMENT OF ORIGINALITY ... 2 ACKNOWLEDGEMENTS ... 3 ABSTRACT ... 4 1. INTRODUCTION ... 6 2. LITERATURE REVIEW ... 8 2.1CEO’S OPTIMISM ... 8 2.3FIRMS’PERFORMANCE ... 11 3. METHODOLOGY ... 12 3.1HYPOTHESES ... 12
3.2MEASUREMENT OF CEOOPTIMISM ... 14
3.3MEASUREMENT OF HIGH- AND LOW-OPTIMISTIC CEOS ... 16
3.4MEASUREMENT OF FIRM PERFORMANCE ... 17
3.5METHODOLOGY ... 18
4. DATA ... 20
4.1DATA ... 20
4.2DESCRIPTIVE STATISTICS ... 22
5. RESULTS ... 25
5.1CEOOPTIMISM AND FIRM PERFORMANCE ... 26
5.2HIGH OPTIMISTIC CEOS AND FIRM PERFORMANCE ... 30
5.3LOW CEOOPTIMISM AND FIRM PERFORMANCE ... 30
6. ROBUSTNESS CHECKS ... 33
7. DISCUSSION ... 35
7.1LIMITATIONS ... 36
7.2CONCLUSION ... 37
1. Introduction
There has been much discussions revolving around the issue of what factors influence firm performance. To solve these kind of reality issues, numerous economists only focus on the traditional finance theories1. However, these hypotheses are barely realized in real world, because not only economic environment is imperfect, but also most decision makers are irrational. Therefore, abounding recent studies depart from a long-maintained rational expectations assumption. Alternatively, researchers start focus on expectation biases such as overconfidence (or underconfidence) and optimism2 (or pessimism), which are deviations from the notion of rationality behind expectation formation. Comforting to the definition of behavioral finance, that the behavior of investors and managers is not fully rational. The personal characteristics and psychology of managers in large corporations have an influence on corporate investment decisions. Therefore, in this thesis I try to do a research about CEOs behavioral effect, which mainly focus on the managerial optimism.
Malmendier and Tate (2005) show that normally the optimistic CEOs overestimate their ability of controlling the outcomes and underestimates the likelihood of failure. That is because CEOs may mistakenly take negative NPV3 projects because of managerial optimism. It is clear that personal characteristics and psychology of CEOs lead to distortions in corporate investment decisions in some content.
However, Campbell (2011) finds strong empirical support the view that there is an interior optimum level of managerial optimism that maximizes firm performance. In line with Compell’s further theoretical and empirical analysis in 2012, it suggests that Malmendier and Tate’s empirical findings of harmful effects of optimism are likely driven by the subset of CEOs with relatively certain optimism levels. The analysis
1
Traditional finance theories are based on rational behavior, the capital asset pricing model (CAPM), and the efficient market hypothesis. 2
Based on the self-serving attribution literature, Malmendier and Tate (2005) refer to optimism as overconfidence. In this paper, for simplified the model I don't distinguish overconfidence and optimism.
from Campell (2012) will be a good complements the work of Malmendier and Tate (2005, 2008).
In parallel to these behavioral finance theoretical developments, it is worth doing a research which combine behavioral finance and firm performance together. Analyzing firm performance is a good way to identify the performance gaps and improvement opportunities. So far, there is only few previous related research on the topic of CEO optimistic impact on investment policy in small ways. The key argument of this few literature is that the individual manager estimation bias matters for corporate decisions. For example, the seminal paper by Bertrand and Schoar (2003) shows that individual managers affect corporate investment, financing, and organizational practices. Even though unbiased beliefs preferable as I normally viewed, firms often employ optimistic managers and follow their beliefs in decision makings which mentioned by Malmendier and Tate (2005). Furthermore, Graham, Harvey, and Puri (2010) provide evidence of a matching of growth firms with more confident managers as well.
Moreover, this thesis is grounded on the research question “How Does CEO’s Impact on Firms Performance?”. Through answer this question, I can not only figure out the relation between optimistic CEO and firm performance, and also can get an insight about why firms hire certain optimistic CEOs. Actually, our contributions are aiming to extend emerging literature by showing that the managerial optimism style can explain different firm performance in a new viewpoint. To my best knowledge, my thesis is one out of few paper about how closely the firm performance can be explained by CEO optimism. In my thesis, the firm performance is measured by two methods, one is marketing-oriented of Tobin’s Q, and the other is accounting-oriented of return on asset (ROA). Meanwhile, it has to be kept in mind that there is a fine line between optimism and confidence, therefore the research on this topic could be a popular appeal as an explanation of behavioral finance. In the conclusion, this
research is intent on providing an instruction to improve the organization effectiveness by using much unique and recent information.
For this thesis, the remaining structure will be organized as follows. First, I will give literature reviews with respect to CEO optimism and firm performance. By taking the literatures as foundation, the hypotheses are pointed out and serves as the basis for the empirical analysis. Follow by that, the empirical models along with the data will be described. After a brief survey of limitations and discussion, I will present the conclusion.
2. Literature Review
After a short introduction about the topic, several papers concerning CEO optimism and firm performance are briefly discussed. The related literatures will be revealed as a foundation to formulate the hypotheses in the following section.
2.1 CEO’s Optimism
The analysis of optimism relates several branches of the psychology and behavorial finance literature. Several researchers have investigated the effects of managerial optimism on firm performance. On the one side, Hirshleifer, Low, and Teoh (2012) show that optimistic CEOs can improve firm performance through effective innovation. On the other side, some researchers illustrate that firms with optimistic managers perform worsr than firms with rational managers as a result of overinvestment. (see, e.g., Goel and Thakor, 2008; Malmendier and Tate, 2008). The impact of managerial optimism on firm performance do spark a fierce debate.
Prior literature finds out that optimism is a strong and robust psychological trait across many samples of subjects, especially among top executives (Alicke and Govorun 2005; Graham et al. 2013; Moore and Healy 2008; Weinstein 1980). Therefore, in this paper our research mainly pitches into top executives such as CEOs.
An interesting survey named "Behaving Badly" did by James Montier in 2006 shows that over 74% of 300 fund managers thought their performance above average, most of the remaining 26% thought they were on the average, few or less people thought they were below average. Actually, based on the standard deviation in a large population, there only 50% of sample could be better than average. The result from this research is clearly showing that more than 24% people overestimate their performance. In fact, the estimation irrationally is not only a trait for funds managers, and it also could be a problem in the senior management, such as Chief Executive Officers (CEOs). Executives appear to be particularly prone to display optimism, both in terms of the better-than-average effect and in terms of “narrow confidence intervals” (Larwood and Whittaker (1977)). The better than average effect (BTA) is the first explanation about psychology of optimism (Svenson (1981)). BTA is defined as a common tendency that people overestimate one's ability to predict and control certain outcomes, such as in minor ways the personal capability, even in a broader perspective firm’s value, stock price and etc. Kruger and Camerer and Lovallo (1999) show that the effect is especially strong among highly skilled individuals, possibly due to insufficient weighting of the comparison group (‘base rate neglect’). Take CEOs as an example, CEOs personal think their capability is higher than the intermediate level, but actually most of them are overestimated themselves. Lots of research demonstrate that BTA is a ubiquitous phenomenon in various realms, with diverse populations in different cultures, on many dimensions, and with different measurements techniques. In line with our econometric knowledge, we know that is not possible. This finding is attributed to three main factors, each of which triggers optimism: the illusion of control, a high degree of commitment to good outcomes, and abstract reference points that make it hard to compare performance across individuals (Weinstein (1980); Alicke (1995)).
However, the personal attributes and behavioral biases, such as CEOs’ optimism, affect the quality of the information available to the board of directors and investors (Adams and Ferreira (2007)) as well as corporate investment decisions. Heaton (2002) analyses the effect of optimism on financing decisions without asymmetric information or moral hazard problems. However, CEOs may mistakenly take negative NPV4 projects because of managerial optimism. Moreover, Malmendier and Tate (2005) shows that normally the optimistic CEOs overestimate their ability of controlling the outcomes and underestimates the likelihood of failure. In consequence of manifestations of optimistic, the CEOs perhaps will continue the negative NPV projects for extended periods. Correspondingly, these bad projects will accumulate the poor firm performance and eventually materialize at their final maturity, leading to a market crash of the stock price.
There has been much discussions revolving around the issue of optimistic CEOs’ impact. As stated, managerial optimism may have a negative impact on firm performance because of taking negative NPV projects. However, we cannot early draw a conclusion that managerial optimism is harmful for firm performance. A coin has two sides. Therefore, previous empirical work on adverse consequences of CEO optimism raises the question of why firms hire optimistic managers. Theoretical research suggests a reason: optimism can benefit shareholders by increasing investment. Campbell (2011) shows theoretically that optimism can lead a CEO, who has a lower tolerance for risk, to choose a better investment level that maximizes shareholder value. Managerial optimism can beneficially offset the effect of CEOs risk aversion on the investment level. There is concave relation between firm value and optimistic CEOs. The result illustrates that there is an interior optimum level of optimism for optimistic CEOs, and leads to finer predictions about the effects of managerial optimism. Compell’s theoretical and empirical analysis in 2012 suggests
that Malmendier and Tate’s empirical findings of harmful effects of optimism are likely driven by the subset of CEOs with relatively certain optimism levels. The analysis from Campell (2012) will be a good complements the work of Malmendier and Tate (2005, 2008). With the help of a large samples, Campbell (2011) finds strong empirical support the view that there is an interior optimum level of managerial optimism that maximizes firm value.
Hirshleifer (2012) displays that firms with optimistic CEOs have greater return volatility. Through investing more in R&D innovation, firms obtain more patents and patent citations, and then attain greater innovative achievement for given development expenditures. The findings suggest that optimism helps CEOs exploit innovative growth opportunities, then enhance firm performance.
Goel and Thakor (2008) show theoretically that a CEO’s under estimation of project risk (optimism invariance) can offset their risk aversion, leading to improved investment levels, firm value. Moreover, Hirshleifer et al. (2010) find that there is a positive relationship between optimism and firm performance, by using the option-based optimism measurement as to measure CEO’s optimistic impact on returns on asset.
According to literature mentioned above, even though CEO optimism could accumulate poor firm performance, overall there will be a positive relation between CEO optimism and firm performance. After consulting the literature on CEO optimism, I expect to observe a positive relation between certain level of optimistic CEOs and firm performance.
2.3 Firms’ Performance
Firm performance is a subset of organizational effectiveness that covers operational and financial outcomes, which is the first standard to be evaluated by investors around
the world. Superior financial performance is a way to satisfy investors and can be represented by profitability, growth and market value (Cho & Pucik, 2005; Venkatraman & Ramanujam, 1986). The literature employs a number of different measures of firm performance to test the predictions of different agency cost hypotheses. These measures include (1) financial ratios from balance sheet and income statements, (2) stock market returns and their volatility, and (3) Tobin’s q, which mixes market values with accounting values. According to previous literature of firm performance measurement, this paper will be focus on the financial ratios and Tobin’s Q.
3. Methodology
This section elaborates on the empirical specification that is being employed to study the relationship between CEO optimism and firm performance in the United States. It starts with formalizing our expectations that follow from the literature review into testable hypotheses. It then proceeds by a discussion of the variables that serve as proxies for firm performance as well as factors that are controlled for. This discussion is summarized into a statistical relationship that will be tested using the data on CEO optimism. I conclude this section by briefly elaborating on predictions regarding the sign of variables.
3.1 Hypotheses
In the previous chapter the basic economic theories were described, why and how optimistic CEO effect firm performance. In this section, the hypotheses will be developed.
It is mentioned in previous literature review that there are certain motivations for hiring optimistic CEO. Optimism may improve firm performance along some dimension for example, optimism can benefit shareholders by increasing investment
in risky projects with greater return volatility, and with more invest in innovation, the greater innovative achievement will be obtained.
The findings suggest that optimism helps CEOs exploit innovative growth opportunities and then improve firm performance. Therefore, in consistence with the previous literature, I expect there will be a positive relationship. This suggests the following hypothesis:
Hypothesis
I
:
CEO optimism has a positive impact on firm performance.The supplement to the first hypothesis, I aim to classified optimism into disparate levels. There are three levels of CEO confidence, such as excessive diffidence (low optimism), moderate optimism, and excessive optimism (high optimism), mentioned in the article from Goel and Thakor (2008).
Moreover, to identify the influence of different optimism levels on firm performance, I use a modified measurement from Malmendier and Tate (2005, 2008) and Malmendier, Tate, and Yan (2010) to classified who hold options very deep in the money and who exercises stock options at low levels of moneyness. For simplified our model, in this paper, I only focus on high- and low-optimism impacts. Moreover, just like Campbell mentioned in his paper, the low optimism group should include pessimistic CEOs, rational CEOs, and CEOs with positive optimism levels too low to offset their risk aversion. I use the term low optimism throughout to refer to this group. The detailed measurement will be pointed out in the section of data.
Therefore, I expect a different relationship between CEO’s low- and high-optimism on firm performance. Differ from positive relation with high-optimism, while the low- CEO optimism will generate a negative impact on firm performance. Consistent with
the theoretical literature the second hypothesis will be tested in this thesis is as followed:
Hypothesis
II
:
High optimistic CEO has a positive impact on firm performance.Hypothesis III:
Low optimistic CEO has a negative impact on firmperformance.
3.2
Measurement of CEO OptimismBased on the pervious literature, there are two alternative proxies for managerial optimism depend on options exercise behavior or press coverage. The options exercise measure (Malmendier and Tate 2005a) is contingent on the idea that a manager who chooses to be exposed to the firm’s idiosyncratic risk is likely to be confident about the firm’s prospects. Just like Ulrike (2001) mentions risk aversion and under diversification predict that CEOs should exercise options immediately after the vesting period if the amount in-the-money is beyond a rational benchmark. Under this approach, a CEO who voluntarily retains stock options after the vesting period in which exercise becomes permissible is viewed as optimistic. The second measure of optimism is conditional on the portrayal of the CEO in the news media, as developed by Malmendier and Tate (2005b, 2008).
The options-based CEO optimism measure has become widely used in recent empirical research (see, e.g., Malmendier and Tate, 2005, 2008; Campbell, Gallmeyer, Johnson, Rutherford, and Stanley, 2011; Malmendier, Tate, and Yan, 2011; Hirshleifer, Low, and Teoh, 2012). Malmendier and Tate (2008) and Hirshleifer, Low, and Teoh (2012) note several alternative reasons for such behavior, including positive inside information, signaling, board pressure, risk tolerance, and taxes, that fail to sufficiently explain the delay in exercise behavior among optimistic CEOs. Therefore, in this paper, I apply options-based measurement to define CEO optimism.
Since I do not have the detailed data of CEO optimism for our large sample of CEOs, I use a modified version of their optimism measure, as developed by Campbell (2011) to construct of CEO’s optimism. It is grounded on the option-based measurement of Holder67 (Campbell, 2011; Malmendier, 2011; Hirshleifer, 2012). Malmendier and Tate (2005) characterize CEOs as optimistic if they hold options at least twice during the sample period that are more than 67% in the money (Holder67). Following that, a dummy variable of confident CEO is generated, which takes a value 1 if a CEO postpones the exercise of vested options that are at least 67% in the money, and 0 otherwise. If a CEO is identified as optimistic by this measure, she remains so for the rest of the sample period. This treatment is consistent with the notion that optimism is a persistent trait.
Following Malmendier and Tate (2005), I apply the chosen cutoff across the full sample of CEOs and require that a CEO exhibits the option-holding behavior at least twice during the sample period. I take as a given their 67% moneyness cutoff for the full sample of CEOs to indicate optimistic CEOs. Use CEOs hold stock options that are more than 100% in the money to identify CEOs who are even more optimistic. I compute option moneyness as follows. The data of option-grant-specific exercise prices is not accessible, therefore, I using Core and Guay’s (2002) approximation method to estimate the average exercise price of the aggregated options. The realizable value per option is estimated as the total realizable value of the exercisable options (ExecuComp variable OPT_UNEX_EXER_EST_VAL) divided by the number of exercisable options (OPT_UNEX_EXER_NUM). By using the stock price at the fiscal year end (PRCC_F) minus the per-option realizable value, I obtain an estimate of the average exercise price of the options. The average percent moneyness of the options equals the realizable value of the exercisable options divided by the estimated average exercise price. The percent moneyness of the unexercised but exercisable option holdings is computed as follows:
Ø Realizable value of the exercisable options =!"#$% '($%)*$+%( ,$%-( ". #/( (0('1)2$+%( "3#)"42 5-6+(' ". (0('1)2$+%( "3#)"42
Ø Average exercise price of the options=Stock price at the fiscal year end - Per-option realizable value
Ø Average percent moneyness of the options = 7($%)*$+%( ,$%-( ". #/( (0('1)2$+%( "3#)"428,('$9( (0('1)2( 3')1( ". #/( "3#)"42
The optimistic CEO is defined as 1, if average percent moneyness of the options >0.67 at least twice during sample period, zero otherwise.
3.3 Measurement of High- and Low- Optimistic CEOs
To identify high optimistic CEOs, hold options deep in the money before exercise, I use a modified version of the stock option-based optimism measurement from Malmendier and Tate (2005, 2008) and Malmendier, Tate, and Yan (2010) to make an estimation which is similar with optimism measure. As an additional measure of high optimism, I draw upon Malmendier and Tate’s (2005) measure based on a CEO’s net stock purchases. I modify both of these measures by establishing classification cut-offs closer to the high-optimism end of the continuum. Based on the reverse logic of the options-based measure of high-optimism, I define a low optimistic CEO as one who exercises stock options at low levels of moneyness. In other words, the CEOs who exercises stock options that are less than 30% in the money and does not hold other exercisable options that are greater than 30% in the money are defined as low-optimistic CEOs. That hold and/or exercise options with moneyness between 30% and 100% is defined as the demarcation point.
The measurement of low optimism is similar with high optimism. Specifically, I estimate the average CEO stock option moneyness for each year as follows. First, I compute the realizable value per option as the total realizable value of the exercisable options (OPT_EXER_VAL) divided by the total number of exercisable
options(OPT_EXER_NUM). Second, I estimate the average exercise price of the options by subtracting the realizable value per option from the stock price at the fiscal year-end. Finally, the average percent moneyness of the options is calculated as the per-option realizable value divided by the estimated average exercise price.
Ø Per-option value realized from exercising = !"#$% ,$%-( '($%)*(: .'"6 (0('1)2)49 2#"1; "3#)"42 5-6+(' ". "3#)"42 (0('1)2(:
Ø Average exercise price of the exercised options=Stock price at the fiscal year end - Per-option realized from exercising
Ø Average percent moneyness of the exercised options = <('="3#)"4 ,$%-( '($%)*(: .'"6 (0('1)2)49
8,('$9( (0('1)2( 3')1( ". #/( (0('1)2(: "3#)"42
Low-optimism is defined as 1, if average percent moneyness of the exercised options < 0.3 and average percent moneyness of the options< 0.3 at least twice during sample period, zero otherwise.
3.4 Measurement of Firm Performance
The first indicator of firm performance is Tobin’s Q. Tobin's q is the “ratio of market value of the outstanding financial claims on the firm to the current replacement cost of the firm's assets” (Lewellen and Badrinath, 1997). It is considered a good indicator capturing future growth opportunities and long-term financial performance as expected by the stock market (Aivazian et al., 2005). Following the literature, I measured Tobin's q as the sum of firm's long-term liabilities, market value of firm at the end of financial year, and liquidation value of preferred stock divided by total assets.
I measure Q as the ratio of market value of assets to book value of assets. Market value of assets is defined as total assets (item 6) plus market equity minus book equity.
Market equity is defined as common shares outstanding (item 25) times fiscal year closing price (item 199). Book equity is calculated as total assets (item 6) minus total liabilities (item 181) minus preferred stock liquidating value (item 10) plus balance sheet deferred taxes and investment tax credit (item 35) when available minus post retirement assets (item 330) when available.
Tobin’s Q = Market value of asset / Book value of asset = !"#$% 822(#>?$';(# @A-)#B=C""; @A-)#B!"#$% 822(# where:
Book value of asset = total asset
Market Equity = common shares outstanding * fiscal year closing price
Book Equity = total assets - total liabilities -preferred stock liquidating value + balance sheet deferred taxes and investment tax credit
The second widely used firm performance indicator is Return on asset (ROA), and ROA is defined as operating income before depreciation minus depreciation and amortization normalized by total assets ((item No. 13 − item No. 14)/item No. 6). In this paper, Tobin’s Q is our main indicator of firm performance, and ROA is a complement to our robustness check.
3.5 Methodology
Before I start to describe our methodology, there is an assumption I have to point out first. In our paper, I take all the recorded CEOs into account, which means considering the entire management team instead of only elaborating on the CEO. In this case, there will be more than one CEO in firms at the same time.
With the help of penal dataset, the firm performance regression is built as follows, which include firm fixed effect αi, as well as year fixed effect δt designed to capture specific impact. Himmelberg (1999) argue that fixed effects estimators should be used in examination of the relationship between managerial ownership and firm
performance. Even though our main variable is CEO optimism instead of ownership, using firm fixed effects to eliminate the impact of time-invariant firm characteristics. Moreover, in a widely cited study, Himmelberg, Hubbard, and Palia (1999) show that when firm fixed effects are controlled for, total insider share ownership of all directors and officers has no identifiable impact on Q. The fixed-effects regression model to empirically measure the relation between CEO optimism and firm performance will be given as follows:
Y
i,t= α
i+ δ
t+ β
0CO
i,t+ β
1Ownership + β
2Compensation + β
3Firm
Size + β
4M/B + β
5ROE + ε
i,tThe Yi,t indicates firm performance which is measure by ROA and Tobin’s Q
respectively. By measuring the firm performance in two aspects, it provides a completed explanation of firm’s current operating performance and reflects the market’s expectations about future performance for our research. The base model employs Tobin’s Q as a marketing-based measure of firm performance to explain the depend variable. Alternative accounting measures such as ROA is used in alternative specifications as robustness checks. The full model, however, adds a wide variety of control variables in order to account for the firm size, firm value, and profitability of the firm. Although my predictions focus on CEO optimism, I would also expect boards to terminate low-ability CEOs, regardless of the CEO’s optimism level. Thus, it is important that I control for CEO ability. I include controls for CEO compensation, and ownership. Hermalin and Weisbach (1998) argue that CEO past performance capture a CEO’s perceived ability, so my tests should be able to separate out the effects of low ability from the effects of optimism on firm performance. The other control variables deal with CEO characteristics of the empirical specification contain CEO ownership, CEO compensation. Oswald and Jahera (1991) find that ownership of directors and officers is positively related to performance. Mehran (1995) shows
that firm performance is positively related to the percentage of equity held by managers. Therefore, I expect there will be a positive relation between ownership and firm performance. The coefficient β0 will give rise an answer to our research question.
4. Data
The previous section is outlined how the hypotheses are developed about CEO optimism relating to firm performance. In this section, the dataset and the measurement of data will be described at the first place. Second, the measurement of CEO optimism and firm performance will be outlined. Third, the related tests about data will be outlined.
4.1 Data
This paper employs panel data from 1992 to 2015, constructed by merging the executive data in ExecuComp with accounting data in Compustat and stock information data in CRSP. Most of the variables are obtained from the Wharton Research Data Services (WRDS) database. Observations are not all available for the whole period. In this paper, the list of firms is restricted by eliminating firms in the finance, insurance, real estate (SIC codes between 6000 and 6799). The firms are required to have available data appear at least three consecutive years in the sample. The variables are winsorized at the 5% level in both tails.
The main variables in the model built in this thesis are dependent variable firms performance and independent variable CEO optimism. The firm performance is measured in two ways, from an accounting perspective, I consider returns on asset (ROA), and in the market is measured with Tobin’s Q. The most of important control variables are CEO compensation and CEO ownership. The ExecuComp database provides total compensation (data item TDC1) for 189,427 firm-year observations.
Normally, CEOs will own substantial amounts of company stock. Therefore, there is a close relationship between shareholder wealth and executive wealth, which is based on numbers of shares held by the CEO. Actually, there are two perspectives to measure the CEO ownership—the dollar value of the CEO’s holdings or the value of his shares as a percentage of his annual cash compensation. In terms of the dollar value measurement of CEO ownership, CEO delta5 and CEO option holdings vega have to be taken into account. There is a particular description in Hirshleifer (2011). Hirshleifer calculated delta and vega values using the 1-year approximation method of Core and Guay (2002). His results are robust to controlling for CEO incentives using percentage stock ownership and option holdings instead of delta and vega. According to the incentive consequences of stock ownership, I find out that what really matters is the percentage of the company’s outstanding shares owned by the CEO. Moreover, according to literature, CEO ownership is defined as the aggregate number of shares held by the CEO, including restricted shares but excluding stock options (whether vested or unvested), expressed as the percentage of the firm’s total shares outstanding. Therefore, in this paper the CEO ownership is defined as the percentage of aggregate number of shares held by the CEO divided by the firm’s total shares outstanding. Considering that CEOs primarily choose the level of investments, innovations, merges, and etc., I focus on the effect of CEO stock ownership, rather than all officers’ and directors’ stock ownership, on discretionary investments. However, most of shares held information is missing and many of the observations that come from ExecuComp only available from 2007 to 2015. Accordingly, the observation of CEO ownership is quite limited, because shares held information is missing.
I supplement the accounting data with various items from the Compustat database. Firm size is defined as the natural logarithm of assets (item 6) at the beginning of the
5 Delta is defined as the dollar change in a CEO’s stock and option portfolio for a 1% change in stock price, and measures the CEO’s incentives to increase stock price. Vega is the dollar change in a CEO’s option holdings for a 1% change in stock return volatility, and measures the risk-taking incentives generated by the CEO’s option holdings.
year. Market value is defined as Common shares outstanding (item 24) times closed price (item 25). Book value is defined as total asset (item 6) minus total common equity (item 60) minus deferred taxes (item 74). Market-to-book ratio is defined as the market value of assets normalized by book value. I use the return on equity (ROE) as a proxy for the firm profitability. ROE is defined as net income (item 172) normalized by total stockholder’s equity (item 216).
4.2 Descriptive Statistics
In the last part of this section, the descriptive statistics is displayed. The sample used in this master thesis is extracted out of the WRDS database. The main purpose of this paper is to estimate the relationship between CEO optimism and firm performance, and the selection is based on companies listed in the S&P 1500 between 1992 to 2015. The chosen time period is based on the availability of CEO optimism.
The dataset does not include financials and real estate investment firms because of distinctive balance sheet structures which prohibits a meaningful interpretation of the estimation results. All variables are specified as annual data. The descriptive statistics of the dataset, and the correlation matrix of explanatory variables are shown in Table 1 and Table 2 respectively.
Table 1: Descriptive Statistics
Obs. Mean Std. Dev. Min Max
Tobin’s Q 194186 2.05 2.49 0.20 203.96 CEO Optimism 214423 0.09 0.29 0 1 CEO Compensation 189427 2203.57 5214.24 -5172.15 655448 CEO Ownership 3862 0.01 0.10 0 3.73 ROA 43509 0.13 0.21 -32 1.25 ROE 212893 0.07 9.77 -790.61 1156 Market-to-Book Ratio 194667 5.70 19.55 0 1564.53 Firm Size 170783 5994.83 22549.7 0.001 797769
High optimistic CEO 214438 0.06 0.24 0 1
Low optimistic CEO 214438 0.02 0.12 0 1
The table above presents the summary statistics for all variables. As noted earlier, the confidence level indicators are zero-one dummy variables, therefore, the mean of variable represents the percentage of CEOs that is defined as optimism, high optimism and low optimism. It is clear can be seen from table that 29% CEOs are defined as optimistic. The proportion of high- and low- optimistic CEOs are 24% and 12% respectively. The other variable has to be noted is market-to-book ratio. M/B measures how much a company worth at present. The minimum of M/B is close to zero. A low market/book ratio could suggest a company’s assets are undervalued, or that the company’s prospects are good and earnings/value should grow. Moreover, technology companies and other businesses that don’t have a lot of physical assets tend to have low book-to-market ratios. Technology companies are part our dataset, therefore, M/B close to zero could be a case in this paper. The mean of CEO compensation $ 2203,57. It shows that the standard deviation of CEO compensation is very large. A possible reason for these differences is that there are used both small and large firms in this study. Moreover, it appears that the compensation is subject to a skewed distribution whereby the mean is considerably larger than the median. Large companies may therefore steer average outcomes in an upward direction. Some firms
reward large stock and stock option packages to their CEOs, while this type of compensation is non-existent in other firms. The observation that large companies may influence the outcomes is further bolstered by the proxy variables for firm size.
Table 2: Correlation Matrix of Explanatory Variables
CEO Overconfi -dence CEO Compensa -tion CEO Ownershi p ROE Market-to -book Ratio Firm Size CEO Optimism 1 CEO Compensation -0.023 1 CEO Ownership -0.006 -0.019 1 ROE 0.047 0.049 -0.011 1 Market-to-book Ratio 0.196 0.021 -0.013 0.133 1 Firm Size -0.067 0.243 -0.028 0.094 -0.007 1
The issue of multicolinearity may arise if two or more variables are highly correlated. It may affect the estimation of the regression parameters (Hair et al., 2010). It can be detected either by examining the correlation matrix or by the variance inflated factor (VIF). As Table 2 of the correlation matrix shown, there is no remarkable high correlations, which means multicollinearity is not an issue in this paper. I also provide the most common Variance Inflation Factor (VIF) multicolinearity detection test results for each independent variable (Naseret al., 2002). If the (VIF) is more than 10 for any independent variable, it indicates that this variable is highly explained by other variables and might be considered for exclusion from the model (Silver, 1997). For this study, the VIF for all the independent variables was reported as reported in Table 3. However, the VIF for all the independent variables was far lower than the cut
off value of 10 as suggested by (Hair et al., 2010) and thus confirming the absence of the multicolinearity issue.
Table 3: Multicollinearity
Variable VIF 1/VIF
CEO Optimism 2.85 0.350429
High Optimistic CEO 2.85 0.350854
Low Optimistic CEO 1.01 0.992042
Firm Size 1.26 0.793567 CEO Compensation 1.25 0.800876 Market-to-Book Ratio 1.04 0.963046 ROE 1.02 0.978385 CEO Ownership 1.01 0.990915 Mean VIF 1.54
5. Results
In this section, the the regression coefficients used to estimate the impact of optimism CEO on firm performance are shown. It starts with a discussion of the results in the regression tables, and then is a comparison with the predictions mentioned in the stated hypotheses. In particular, I are interested in weather the dataset confirms our prediction that there is a different relation between firm performance and optimism, such as a positive relation with (high-) optimism and a negative relation with low optimism.
The heterogeneity test is conducted in the first place. The null hypothesis of heterogeneity test is that the model is homogeneity (or constant variance). The result is Prob>chi2 = 0.0000, which means the null hypothesis can be rejected at 1 percent level, and the model suffers from heterogeneity.
A Hausman test then is carried out to select between fixed or random effects. The null hypothesis is a preferred model with random effects. The alternative is a preferred model with fixed-effects (Hausman, 1978). Hausman basically tests whether the unique errors (ui) are correlated with the regressors: the null hypothesis is that they are not correlated. The result of test is Prob>chi2 = 0.006, which is smaller than 0.05, thus the null hypothesis can be rejected. For achieving a robust result, the option ‘robust’ is used to control for heterogeneity in fixed-effects. The Hausmant test result is consistent with our theoretical literature, therefore, fixed-effects regression is used in this paper.
5.1 CEO Optimism and Firm Performance
Table 4 on the next page reports the estimation results for the dependent variable marketing-based Tobin’s Q and accounting-based ROA respectively. I can see from Table 4 that there is a positive and significant coefficient on variable CEO optimism for both Tobin’s Q and ROA, which indicates that CEO optimism positively influences firm performance. The coefficient of CEO optimism indicates the average increase in firm performance 0.184 for Tobin’s Q and 1.18% for ROA. However, comparing with ROA regressions, CEO optimism is highly significant across all Tobin’s Q regressions and also yields high parameter values. It is quite clear that CEO optimism have a significantly larger impact on Tobin’s Q rather than ROA. In other words, there is no significant effect of optimistic CEOs on current operating performance (ROA) instead of market expectations about further performance (Tobin’s Q).
These results provide evidence in favor of accepting Hypothesis I. The first regression only contains dummy variable of CEO optimism has a low explanatory power as shown by a low R-square 0.029 and 0.011. For example, the R-square for the model 1 which was used to test for hypothesis 1 is 0,029. This means that, barely 2.9% of the
regression model or variation in the dependable variable (Tobin’s Q) is explained by the independent variables. This value seems to be normal suggesting that the regression equation or the variables used seem reasonable to account for variations in Tobin’s Q. It would be better if the value for R square was high. The explanatory power is greatly improved after the inclusion of various control variables. In the regression (3), the R-squared value is close to 0.506, indicating that the model explains a rather large part of the variation in the model.
Table 4: CEO Optimism and Firm Performance, 1992-2015
Tobin's Q ROA （1) （2) （3） (4) （5) （6） CEO Optimism 0.495 ** * 0.299 *** 0.184*** 0.0135 ** * 0.0153 ** 0.0118* -40.31 -5.92 -4.35 -11.79 -2.6 -2.46 CEO Compensation 0.0000157* * 0.00000948 0.00000 1 0.00000156 ** -2.64 -1.79 -1.58 -2.59 CEO Ownership -0.377 -0.0217 -0.031 -0.0277 (-1.59) (-0.11) (-0.98) (-1.12) ROE 0.266*** 0.0133** -8.22 -3.27 Market-to-Boo k Ratio 0.0918 *** 0.00230*** -30.38 -7.56 Firm Size -0.00000804 ** * -0.00000013 5 (-5.56) (-0.72) _cons 2.333 ** * 2.187 *** 1.607*** 0.126*** 0.124*** 0.113*** -134.79 -52.16 -36.03 -77.91 -15.61 -15.03 R-Square 0.029 0.054 0.506 0.011 0.019 0.082 N 194174 3390 2838 43503 2200 1755 t statistics in parentheses * p < 0.05, ** p < 0.01, *** p < 0.001Table 5: High Optimistic CEO and Firm Performance, 1992-2015
Tobin's Q ROA （1) （2) （3) （4) （5) （6） High Optimistic CEO 0.611** * 0.364*** 0.176*** 0.0181** * 0.0187** 0.011 -43.64 -6.17 -3.63 -13.55 -2.66 -1.93 CEO Compensation 0.0000163* * 0.00000996 0.0000010 4 0.00000162** -2.74 -1.88 -1.64 -2.67 CEO Ownership -0.42 -0.0283 -0.0317 -0.0279 (-1.76) (-0.14) (-1.01) (-1.12) ROE 0.266*** 0.0133** -8.22 -3.28 Market-to-Boo k Ratio 0.0919*** 0.00228*** -30.29 -7.41 Firm Size -0.00000818** * -0.00000014 3 (-5.65) (-0.76) _cons 2.332** * 2.178*** 1.612*** 0.127*** 0.125*** 0.113*** -134.9 -51.65 -36.04 -78.2 -15.7 -15.14 R-Square 0.031 0.055 0.505 0.010 0.013 0.075 N 194174 3392 2838 194174 3392 2838 t statistics in parentheses * p < 0.05, ** p < 0.01, *** p < 0.001In terms of control variables, the CEO compensation turns out to be positive in sign, but is statistically not different from zero in all cases. The CEO ownership yields insignificant parameters across all regressions. The other accounting control variables show a positive sign and significant in most of specifications.
5.2 High Optimistic CEOs and Firm Performance
According to related literature I mentioned in the pervious part, I specified the optimistic CEO into two groups, which are high- and low-optimistic CEO. The regression results of high optimistic CEO effect will be pointed out by referring to Table 5 in the next page. In this section, the main variable CEO optimism is replaced by high optimistic CEO, and the other variables are the same as before. The dependent variables will be measured by marketing-based Tobin’s Q and accounting-based ROA as well.
I can see from Table 5 that there is a positive and significant coefficient on variable CEO optimism for both Tobin’s Q and ROA, which indicates that CEO optimism positively influences firm performance. The result of main explanatory variable named high optimistic CEO is consistent with the result of CEO optimism. The results of coefficient of CEO optimism have changed slightly, and the conclusions concerning significance and sign have not changed. The coefficient of high-level CEO optimism indicates the average increase in firm performance 0.176 for Tobin’s Q and 1.11% for ROA. For the other control variables, the results almost same.
5.3 Low CEO Optimism and Firm Performance
In this section, our results are focus on the low-level CEO optimism impact on firm performance. Comparing with pervious regressions, the results for low-optimistic CEO are complicated.
I can see from Table 6 that there is a negative and significant coefficient on variable CEO optimism for Tobin’s Q, which indicates that CEO optimism positively influences firm performance. In column (1), the coefficient of CEO optimism indicates the average decrease in firm performance -0.163 for Tobin’s Q. In column (2), after adding control variables deal with CEO characteristics, the coefficient changes a little bit but the sign keeps negative. The measures in column (3) loses its significant after inclusion of accounting variables. However, comparing with ROA regressions, CEO optimism is highly significant across all Tobin’s Q regressions and also yields high parameter values. The results are not robust in ROA regressions. It is quite clear that CEO optimism have a significantly larger impact on Tobin’s Q rather than ROA. In other words, there is no significant effect of optimistic CEOs on current operating performance (ROA) instead of market expectations about further performance (Tobin’s Q). After adding the other control variables deal with CEO characteristics and other accounting variables, the coefficients turn to be insignificant. As the dependent variable changed into ROA, unfortunately, the coefficients of ROA on firm performance are not significant at all cases.
These results provide evidence in favor of accepting hypothesis III. The first regression only contains dummy variable of CEO optimism has a low explanatory power as shown by a low R-square 0.029 and 0.011.
Table 6: Low CEO Optimism and Firm Performance, 1992-2015
Tobin's Q ROA （1) （2) （3) （4) （5) （6) Low -CEO Optimism -0.163** * -0.192 * -0.0334 -0.0008 3 0.00319 0.00151 (-6.99) (-2.07) (-0.47) (-0.43) -0.32 -0.18 CEO Compensation 0.0000170* * 0.00000962 0.0000010 4 0.00000162* * -2.65 -1.8 -1.63 -2.68 CEO Ownership -0.194 0.0108 -0.03 -0.0269 (-0.76) -0.05 (-0.95) (-1.08) ROE 0.269*** 0.0133** -8.29 -3.28 Market-to-Boo k Ratio 0.0932 *** 0.00240*** -30.84 -7.93 Firm Size -0.00000828 ** * -0.00000014 (-5.70) (-0.75) _cons 2.366*** 2.248*** 1.643*** 0.128*** 0.126*** 0.114*** -136.16 -54.87 -37.29 -78.87 -15.86 -15.15 R-Square 0.019 0.039 0.507 0.001 0.006 0.071 N 194174 3392 2838 43503 2201 1756 t statistics in parentheses * p < 0.05, ** p < 0.01, *** p < 0.0016. Robustness Checks
In this section robustness checks have been conducted in order to improve the accuracy of the results. By adding a new lagged variable of Tobin’s Q and ROA into our first hypothesis regression, the original static model replaced by dynamic model.
Estimating a dynamic panel regression of changes in performance on lagged changes in Tobin’s Q and ROA, the R-squares are improved to 67.9% and 57.4% which indicate that the explanatory power of the model is fairly high. In our robustness checks, the explanatory power is greatly improved after inclusion of new lagged variable, and the statistic model also turn to be a dynamic model. The result can be found in next page in Table 7.
The model as follows:
Y
i,t= α
i+ δ
t+ β
0CO
i,t+ β
1Ownership + β
2Compensation + β
3Firm
Size + β
4M/B + β
5ROE + β
6ROE Y
t-1+ ε
i,tAfter lagged variables added into regressions, the coefficient of CEO optimism on firm performance decreases around 14 percent, which is from 0.184 to 0.161 of Tobin’s Q and from 0.0118 to 0.0101. The sign and the significant level are the same as before. Comparing with the static model, the relation between CEO ownership and firm performance is negative and insignificant. It is noted that the coefficient of CEO ownership on firm performance is turn to be positive and highly significant, which is consistent with existing literature. The coefficient and significant of the other variables keep the same before and after adding the lagged variable.
In the robustness check, not only the accuracy of the model has been improved, but also the coefficient and significant level of CEO ownership are corrected
Table 7: CEO Optimism and Firm Performance, 1992-2015
Tobin's Q ROA Tobin's Qt-1 0.242*** -12.33 ROAt-1 0.296*** -9.53 CEO Optimism 0.161*** 0.0101* -3.95 -2 CEO Compensation 0.00000441 0.00000198** -0.87 -3.26 CEO Ownership 0.259*** 0.0119** -8.46 -2.77 ROE 0.0457 0.00701 -0.24 -0.28 Market-to-Book Ratio 0.0854*** 0.00175*** -28.09 -5.66 Firm Size -0.00000591*** -0.000000383 (-4.28) (-1.85) _cons 1.129*** 0.0775*** -19.8 -10.25 R-Square 0.679 0.574 N 2791 1391 t statistics in parentheses * p < 0.05, ** p < 0.01, *** p < 0.0017. Discussion
In this section an analysis of the results presented in section 6 of this master thesis will be pointed out. The results of the estimation regressions will provide a source of information to check about our predictions about developed hypotheses. Furthermore, this section will end with a conclusion about the relationship between CEO optimism and firm performance. Before that, the limitations of this research and suggestions for future research will be identified.
First the results indicate that optimistic CEO advance firm performance in some extent. According to pervious literature, I use the timing when the CEO exercise their granted options to measure the optimism level of CEOs. The results from Table 5 indicate that the optimism helps CEOs offset risk averse, invest more in innovation, and then boost firm performance. From these observations it can be stated that there is a relation between CEO optimism and firm performance, our result from first regression support our hypothesis. For complement to our hypothesis, the first rough model will be refined into more models with optimism tessellation. The results of the second regression model as outlined in Table 6, which help to identify if CEOs with relatively high or low optimism levels show different impacts on firm performance. The results of second hypothesis are statistically significant evidence indicate that CEO’s with relative high optimism levels have an effect on escalation of firm performance. Second, the results indicate that there is a negative relation between CEO optimism and accrual firm performance.
This indicates that the more confident a CEO, the more risk investments are made. Therefore, it can be stated that hypotheses I, there is a positive relation between CEO optimism and firm performance, and hypotheses II, there is a positive relation between CEO optimism and accrual earnings management, can both be supported. Additionally, the results indicate that optimistic CEO have statistically significant
influence on firm performance. The results of the second regression model presented in Table 7, show that extremely low confident CEO’s influence have a negative important on firm performance. From all regressions, I find out that the results are more significant for marketing-based indicator Tobin’s Q instead of accounting-based ROA. In the last model of low-optimism CEO, the coefficient of ROA on firm performance even lost significant.
7.1 Limitations
One clear limitation of this thesis lies with the availability of data. Considering CEO optimism is a new concept in filed of behavioral finance, therefore, there is no completed package of CEO optimism and CEO ownership datasets. Take CEO ownership as an example, I measure it by using the definition of the percentage of the company’s outstanding shares owned by the CEO, but the data about how many shares held by CEO is quite limited. The data is only available from 2007 to 2015, and after screening there are only 3,862 observations available.
The other limitation comes from the measurement of variables I used in our regressions, such as CEO optimism. Considering the data of CEO optimism is calculated, therefore, the accuracy of CEO optimism variable is limited in this thesis. Actually, in line with previous literature, there are two ways to measure optimism, one is options-based and the other is press-based. The options-based optimism measure is based on the premise that it is typically optimal for risk-averse, undiversified executives to exercise their own-firm stock options early if the option is sufficiently in the money (Hall and Murphy (2002)). In terms of proxies of optimism, since options-based is widely used for years, therefore, in this thesis I only focus on optioned-based measure. Furthermore, Ulrike (2001) constructs three indicator variables to partite CEOs into “late” and “timely” option exercises, such as longholder, identified CEOs who, at least once during their tenure, hold an option until the year of
expiration, even though the option is at least 40% in-the-money entering its final year; Pre-/Post-Longholder, Post-Longholder is equal to 1 only after the CEO for the first time holds an option until expiration (provided it exceeds the 40% threshold). Pre-Longholder is equal to 1 for the rest of the CEO years in which Longholder is equal to 1; Holder 67, CEOs hold their options all the way until expiration. In this paper, I do not include them all, and I only choose Holder 67 as a proxy of optimism, but other measurements or indicators may be better suited for CEO optimism.
7.2 Conclusion
This paper explores the relation between CEO optimism and firm performance by using the data from 1992 to 2015. I assert that the managerial optimism problem is typically associated with financial decisions makers just as CEOs, because senior managers’ irrational decisions will have important impacts on corporate decisions. I focus on two measurements of firm performance, which related to marketing-measured of Tobin’s Q and accounting-measured of ROA. In our model, the relation between CEO optimism and firm performance depends on whether optimism ameliorates or exacerbates the over- or under-investment problem related to each type of CEO. The results show that there is positive relation between optimism CEO and firm performance. The high optimistic CEO aligns the interests between shareholders and managers and, thereby, mitigates the under-investment problem.
On the contrary, low optimistic CEO would worsen firm performance. Why there are opposite impacts of CEO optimism on firm performance, that is because increasing optimism at low optimism cannot improve the underinvestment problem. Managers are not willing to bear larger costs from investment in certain projects, therefore, the achievements from overinvestment cannot offset the problems of underinvestment. In summary, our results provide a plausible reason for the different relation between firm performance and managerial optimism. Increasing managerial optimism enough has a positive effect on firm performance at high optimism because high optimism
improves under investment problems and thus increases investments and innovations achievement. However, at lower levels, increasing optimism distorts investment decisions because managerial risk-aversion exacerbates the underinvestment problem. Thus, firm performance decline because discretionary investments are not in the interest of shareholders, rather they are in the interest of the management.
Additional robustness checks indicate that our results hold to add new variables, such as lagged variable of Tobin’s Q and ROA, and using alternative autoregressive specifications. The relation between CEO optimism and firm performance does not change by adding new control variables. However, the accuracy of model is improved, which can be seen from the increasing R-square.
In this master thesis the relationship between CEO optimism and firm performance is investigated. To my best knowledge, it is the first research to investigate this relationship using the optimism measure introduced by Malmendier & Tate (2005, 2008). Moreover, it is the first time that accounting variables such as firm size, firm value and firm profitability are added into regressions. This thesis contributes to the existing literature is that it provides a new point of view to explain the different firm performance by referring the behavioral finance.
For further research it is recommended to additional data on the other European countries and include these in the regressions, in order to obtain more reliable results and to be able to generalize the influence of optimism CEO over firm performance. It is also worthwhile to extend the model slightly. As I can see from our robustness checks, the R-square is improved by changing the origin static model into autoregressive model. However, additional variables should be carefully chosen based on previous findings, as always, there could be some relations between independent variables which causes endogeneity problem.
References:
Alicke, M. D., & Govorun, O. (2005). The better-than-average effect. In M. Alicke, D.Dunning, & J. Krueger (Eds.), The self in social judgment(pp. 85-106). New York,NY: Psychology Press.
Aivazian, V.A., Ge, Y, and J. Qiu, 2005, "The Impact of Leverage on Firm Investment: Canadian Evidence," Journal of Corporate Finance 11, 277-191. Adams, R., Ferreira, D., 2007. A theory of friendly boards. Journal of Finance 62,
217–250.
Bertrand, M. and A. Schoar, 2003, “Managing with Style: The Effect of Managers on Firm Policies,” Quarterly Journal of Economics 118, pp. 1169-1208.
Cho, H., & Pucik, V. (2005). Relationship between innovativeness, quality, growth, profitability, and market value.Strategic Management Journal, 26(6), 555-575. doi: 10.1002/smj.461
Campbell, T. C., Gallmeyer, M., Johnson, S. A., Rutherford, J., & Stanley, B. W. (2011). CEO optimism and forced turnover. Journal of Financial Economics,101(3), 695-712.
Core, J., & Guay, W. (2002). Estimating the value of employee stock option portfolios and their sensitivities to price and volatility. Journal of Accounting Research, 40(3), 613-630.
Chakravarthy, B. (1986), Measuring Strategic Performance, Strategic Management Journal, 7, pp.437-458.
Graham, J.R., Harvey, C.R., 2001. The theory and practice of corporate finance: evidence from the field. Journal of Financial Economics 60, 187–243.
Graham, J.R., Harvey, C.R., Puri, M., 2012. Capital allocation and delegation of decision-making authority within firms. Unpublished working paper, Duke University.
Glick, W. H., Washburn, N. T., & Miller, C. C. (2005). The myth of firm performance. Proceedings of the Annual Meeting of American Academy of Management. Honolulu, Hawaii.
Larwood, L., and W. Whittaker, (1977). Managerial Myopia: Self-Serving Biases in Organizational Planning. Journal of Applied Psychology 62: 194-198.
Lewellen, W.G. and S.G. Badrinath, 1997, "On the Measurement of Tobin's Q," Journal of Financial Economics (April), 77-122.
Malmendier, U., & Tate, G. (2005). CEO optimism and corporate investment. The journal of finance, 60(6), 2661-2700.
Malmendier, U., & Tate, G. (2008). Who makes acquisitions? CEO optimism and the market’s reaction, Journal of Financial Economics 89, 20–43.
Malmendier, U., Tate, G., & Yan, J. (2011). Optimism and Early- Life Experiences: The Effect of Managerial Traits on Corporate Financial Policies.The Journal of Finance, 66(5), 1687-1733.
Mehran, Hamid, 1995, Executive Compensation Structure, Ownership, and Firm Performance, Journal of Financial Economics 38, 163-184.
Naser, K., Al-Khatib, K. Karbhari, Y. 2002. Empirical evidence on the depth of corporate information disclosure in developing countries: the case of Jordan. International Journal of Commerce and Management 12:3, 122-155.
Oswald, S. L., & John S. Jahera, J. (1991, May). The Influence of Ownership on Performance: An Empirical Study. Strategic Management Journal, 12(4), 321-326.
Heaton, J. B. (2002). Managerial optimism and corporate finance. Financial management, 33-45.
Hausman, J., 1978. Speci"cation tests in econometrics. Econometrica 46, 1251-1271. Himmelberg, C.P., Hubbard, R.G., Palia, D., 1999. Understanding the determinants of managerial ownership and the link between ownership and performance. Journal of Financial Economics 53, 353–384.
Healy, P. J., & Moore, D. A. (2007). Bayesian optimism (Tepper Working Paper No. 2007-E16). Pittsburgh, PA: Carnegie Mellon University, Tepper School of Business.
Hirshleifer, D. A., Teoh, S. H., & Low, A. (2010). Are optimistic CEOs better innovators? Journal of Finance 67 (4), 1457–1498.
Hawawini G, Subramanian V, and Verdin P (2003) Is performance driven industry- or firm- specific factors? A new look at the evidence. Strategic Management Journal vol. 24: 1-16.
Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (2010). Multivariate data analysis, Prentice-Hall, Upper Saddle River, NJ.
Kruger, J. (1999). Lake Wobegon be gone! The "below-average effect" and the egocentric nature of comparative ability judgments. Journal of Personality and Social Psychology, 77(2), 221-232.
Svenson, O., (1981), Are I all less risky and more skillful than our fellow drivers? Acta Psychologica 47: 143-48.
Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of Personality and Social Psychology, 39, 806–820.
