University of Groningen
Faculty of Economics
The influence of Corporate Governance on the pricing
of the Initial Public Offering; The US story
Master Thesis
The influence of Corporate Governance on the pricing
of the Initial Public Offering; The US story
University of Groningen Faculty of Economics March, 2009 Berend Langelaar s1367218 Abstract:
This paper tests the influence of corporate governance on the IPO underpricing measured by the difference between the offering price and the first day closing price. I test the influence of three corporate governance indicators, inside ownership, board characteristics and executive compensation to find their support for the three main theories related to the influence of corporate governance on the underpricing; the information asymmetry hypothesis, the signaling hypothesis and the dispersed ownership hypothesis. This paper examines 138 US IPOs that took place in the years 2006-2008. In my dataset, I found evidence that the variables management ownership and equity compensation influence the amount of underpricing of an IPO. However, this paper can not link the outcome to one of the discussed theories.
Keywords: Initial public offering, Underpricing, Inside ownership, Board characteristics, Executive compensation
I.
Introduction
Corporate governance and underpricing of an initial public offering (IPO) are two topics that received much academic interest during the last decades. Researchers all over the world discuss the underpinnings of the two topics and some even try to link corporate governance and IPO underpricing by examining the influence of corporate governance on the underpricing of an IPO. However, the current literature does not give a clear overview of the different aspects of the influence and research is often based on an outdated dataset. This paper strives to explain the relation by testing the influence of the board characteristics, the inside ownership, and the executive compensation on the price setting of an IPO through analyzing 138 US initial public offerings in the period 2006-2008. I endeavor to enhance the current knowledge about the problems that arise due to the discrepancy between the expectations of the new owners and the behavior of the current management by examining different indicators of corporate governance. The research aims to help the continuing process of improving corporate governance structures, thereby increasing the prosperity of the owners of the company.
Corporate governance describes the interaction and relations between the different stakeholders that are involved in the company operations. Stakeholders have different interests in the company which, in some occasions, lead to situations of conflict. One point in time in which these conflicts play an important role is at an initial public offering (IPO). During this change in ownership structures, most of the time the inside ownership is strongly reduced because a new wide variety of investors retain a stake in the company. These new investors are able to acquire part of the shares against a discount, the so called underpricing1 of the shares. This paper discusses three theories, all based on the agency theory, that motivate why initial owners sell their stake at a discount, thereby directly transporting part of their wealth to the new owners. I discuss successively the asymmetric theory, the signaling theory and the dispersed ownership theory in order to help understand the underpinnings of the influence of corporate governance on the underpricing of an IPO. After explaining the different related subjects and discussing the different theories, I test for the influence of three indicators of corporate governance discussed in the academic literature, respectively the inside ownership, the board characteristics, and the executive compensation on the amount of underpricing.
1 Underpricing is the difference between the offering price and the first day closing price (Certo et al,
I aim to test whether these indicators of corporate governance can clarify why, during an IPO, some companies need to offer a bigger discount than others. Understanding the problems help to increase the transparency of the agent’s behavior and the efficiency of the overall business environment. This research helps to understand the factors of corporate governance that can be determined and adapted by the company and, as a result, helps to maximize the return of the selling shareholder by minimizing the discount offered to the public.
The outline of the paper is the following; after the introduction I discuss the theoretical framework and the related findings of previous empirical work in chapter II. This chapter ends with the hypotheses tested in my research. The section is followed on by the sample design and data used (part III). Part IV discusses the methodology and the related model. I finish my research presenting the test results (part V), the discussion (part VI) and the conclusion (part VII).
II.
Literature review
Theoretical framework
Before analyzing the influence of corporate governance on the amount of underpricing of an initial public offering, one has to understand what these two topics exactly mean, why these subjects are so important and how they generally are being measured.
In this paper, corporate governance is examined in the context of the price setting of an IPO. Schleifer and Vishny (1997) state that corporate governance “deals with the way in which the suppliers of finance to corporations assure themselves of getting a return on their investment.” What makes this theory especially important is that it explains why stakeholders can have conflicting interests. Insufficient efforts, extravagant investments, entrenchment strategies and self-dealing (Tyrole, 2006) of other stakeholders results in agency costs and harm the suppliers of finance. Important measures of corporate governance are both internal (board of supervisors, equity ownership structure, etc) and external (market for corporate control and the legal system). This paper aims to give insight in the internal mechanisms, because I want to find indicators that the companies can control in an attempt to reduce the required amount of underpricing needed to succeed the offering. These indicators will be discussed later on in this paper.
Like stated by Filatotchev and Bishop (2002), underpricing is “a direct wealth transportation from the founders and initial owners to the new shareholders.” Consequently, the selling shareholders have a strong incentive to reduce the amount of underpricing. The academic literature identifies
several reasons for offering the shares with a discount to the capital market. Table 1
contains an overview of the theories that lie behind the underpricing.
Table 1: Theoretical underpinnings of underpricing of an IPO
Theory Founder Explanation
Risk-averse underwriter hypothesis
Chalk and Peavy (1987) In order to avoid the costs and risks associated with an unsuccessful issue, underwriters set the offering price to low in order to succeed
Monopsony power hypothesis Rock (1985) The bargaining power of large reputable investment bankers over the initial public offering firm, makes it possible for bankers to lower the offering price available for preferred stock holders, thereby, offering an additional compensation for the buyers
Speculative bubble hypothesis McQueen and Thorley (1994)
Lowering the price in order to create oversubscription results in that fact that after the offering investors that do not obtain shares are prepared to pay higher prices Asymmetric information
hypothesis
Akerlof (1970) New investors want to be compensated for the differences between less informed outside shareholder and the more informed inside management
Implicit insurance hypothesis Tinic, 1980 Underpricing decreases the likelihood of legal fights due to underperformance of the stock
Market Feedback hypothesis Jegadeesh, Weinstein, and Welch (1993)
Selling the stocks to preferred stockholders reveals the appetite for the stock, to succeed the shares are offered with a discount
Bandwagon hypothesis Harvey Leibenstein (1950)
Underpricing is used to sell the stocks to the first buyers, which lowers the threshold for other potential buyers who are more frightened of being the only buyers
Signaling Hypothesis Spence (1974) Ownership by inside shareholders with a strong reputation signals the trust in the performance of the company therefore a discount is provided to them
Ownership dispersion hypothesis
Booth and Chua (1996) Lowering the offering price increases the demand for the shares, which enables the owners to sell the shares to a variety of shareholder reducing the risk of large block holders
Source Titnic (1988), Ritter (1998) and Certo et al (2001)
The three theories mentioned above are all rooted on the agency theory2. In the perspective of this paper the theory explains the problems arising between the management and the board, as well as with the current and new investors on the capital market at the time of an IPO. The current and future owners are the so called principals and the management the so called agent. Whereas the management should act in line with the expectations of the principals, they sometimes have conflicting interests. This means that if ownership and control are being separated the new principals want to be compensated for the value destroying behavior.
Next to the principals and the management, the board is the third stakeholder that plays an important role in the agency theory. Companies are required by law to appoint a board; they help to monitor the behavior of the management. Most members of the board are not involved in the daily operations of the company and should therefore be able to review the performance of the management independently and, if needed, help to overcome problems of mismanagement.
Adopting the right corporate governance mechanisms in order to overcome problems between the principal and the agent, helps to minimize the need to compensate for the separation of ownership and control. Effective corporate governance aligns the behavior of the agent and the expectations of the principals. The three derived theories help to understand the different kind of relations of corporate governance with the amount of underpricing of an initial public offering and are discussed hereafter.
The first theory put forward in the academic literature, the asymmetric information theory, concludes that one stakeholder posses more information than the other(s). In the context of this paper, I focus on the information asymmetry between the issuer and the investors on the capital market. Myers and Majluf (1984) and Dierkens (2007) state that management normally has advantage over the other stakeholders because they have the benefit of predicting firms specific events. The authors state that the importance varies over the lifecycle of the company however should increase in times of raising capital. Raising external capital leads to the occurrence that outside shareholders have incomplete information about the functioning of the management and they are not able to react effectively by undertaking proper actions.
Benviste and Spindt (1989) state that the shareholders want to be compensated for the discrepancies between ownership and control and, therefore, the shares are offered to the public with a discount, the so called underpricing of the shares. The seller has to be aware of the fact that the shares are less valuable for the outside investor because the outside investor has to incorporate the cost to incur due to monitoring the management of the company. Conversely, the current shareholders obviously want to minimize the amount of underpricing because it is costly to them.
Second theory that explains the influence of corporate governance on the underpricing of an IPO is the so called signaling theory. One of the early studies on the influence of signaling comes from Leland and Pyle (1977). They state that issuers know the “collateral, industriousness and moral rectitude” better than the investors do. They refer to this as “inside” information. They state that the lenders would benefit from knowing the true characteristics of the company. However, due to moral hazard, these investors are not able to retain the information. The authors state that providers of finance, like shareholders, are not capable to distinguish the two types of companies, called “good” and “bad” companies. Good companies will behave in line with the expectation of the company and strive to do what is best for the shareholders. Bad companies try to mimic as if they do their best, but actually they put in as less effort as possible and try to seek for private benefits. In order to show the difference, the good management tries to send a signal to the shareholders, one that can not by copied by the bad management. Leland and Pyle call this the signaling theory. The theory concludes that the management wants to signal that their behavior is in best interest of the (new) shareholders. They want to prove that they show value maximizing behavior. Leland and Pyle state that proving the quality by undertaking proper actions “speaks louder than words.” Tyrole (2006) also discuss the influence of dissipative signals. He states that for signals to be informative they have to be costly. If signals are costly they become less appealing for the bad companies to copy. Leland and Pyle mention one of such action, the willingness of inside person, like the management, to invest in the company. Other signals put forward by Tyrole are the payout policy, short-term maturities, diversification and incomplete insurance, and underpricing. This means for the management that they can be stimulated through the compensation structure, the possibility to get fired easily in the short run, to increase the risk by incomplete insurance against economic downfalls and the possibility to underprice the shares in an attempt to signal the quality of the firm. Concluding, we can state that signaling can influence the underpricing by adopting costly signals in order to show that the management is a “good” management (Ritter, 1998).
share ownership after the offering, because this limits the possibilities of large block holders3 to influence the company’s management. Furthermore, they state that it protects the management against hostile takeovers. Stoughton and Zechner (1998) however come up with a conflicting theory concerning the ownership dispersion. On the one hand they say that it is optimum that, in terms of risk sharing, the shares are widely dispersed. In some situations the company is better of if hostile bidders are not able to acquire the company (think of the consequences for the strategy of the company and the related turmoil for employees). On the other hand they say that large shareholder have a bigger incentive to monitor the management, as a result this boost the shareholder’s value and, in the long run, will favor all the stakeholders.
So, in the context of an IPO, the influence of large block holders and the related consequences for the corporate governance structures can inside owners decide to sell their stake with a discount to the public. One condition for this reason of underpricing is that the selling shareholder should have an advantage of dispersed ownership, for example, when the selling shareholder has a position in the management of the company.
To sum up, the three theories rooted on the agency theory have strong implications for the price setting of an initial public offering. Where the asymmetric information increases the underpricing of an offering, the signaling and dispersed ownership hypothesis are less stringent about the direction of the influence. In case of the signaling hypothesis, lowering the offering price can send out a signal to potential investors, but adapting the right costly non-underpricing signals would be an effective signal as well. The dispersed ownership hypothesis can, as stated above, lower the offering price to stimulate the diffusion of the shares which in turn can be positive for some stakeholders like employees, however, reducing the underpricing stimulates large blockholders to acquire significant number of shares and make it possible for these blockholders to be actively involved in the monitoring process.
Empirical Evidence
This section discusses previous findings of academics about the one-way causality of the influence of corporate governance on the amount of underpricing of an initial public offering. Filatotchev and Bishop (2002), Hartzell et al (2004) and Nikbahkt (2007) discuss in their papers different indicators of corporate governance. They say that the influence of corporate governance can be examined by testing for inside ownership, adopting the right board characteristics and constructing effective executive compensation structures. Arugaslan et al (2004) acknowledge this, but noticed a shift in the focus of the literature discussed. The authors conclude that prior
literature merely focuses on the influence of uncertainty and asymmetric information on the IPO, while the current academic interest is mainly interested in the influence of the ownership on the IPO process. However, no other authors support their view. Although more indicators of corporate governance exist, I choose to focus on respectively inside ownership, board characteristics and executive compensation because this information is publicly available in the offering document that comes with the shares issue.
Table 2 gives an overview of the related theories, the way corporate governance is being measured and the conclusion about the consequence of the indicators for the three different theories. Because all three theories try to explain the same existence, the underpricing, there is not a one on one relation between each indicator of corporate governance and one of the theories. The theories explain the logic behind the underpricing of the shares.
Table 2: Theories and the corporate governance indicators
Topic Explanation
Information asymmetry Lowering the offering price is a compensation for uninformed investors (vs informed investors)
Signalling theory Lowering the price induces a threshold (one that can not be copied by “bad companies”)
Dispersed ownership theory Lowering the offering price increases the demand for the stock which stimulates ownership dispersion
Inside ownership Should enhance the motivation of the management/board to act in best interest of the company
Board characteristics Should influence the drive of the board to perform in line with the expectation of the shareholders
Executive compensation Should be designed to stimulate the management to show value maximizing behaviour
Information asymmetry Optimizing the influence of Corporate Governance in order to minimize the need for IPO underpricing should reduce the amount of information asymmetry
Signalling theory Optimizing the influence of Corporate Governance in order to minimize the need for IPO underpricing should reduce the cost of the signal needed to show the type of company
Dispersed ownership theory Optimizing the influence of Corporate Governance in order to minimize the need for IPO underpricing should reduce the motivation of large blockholders to monitor the management and, thus, the threat of hostile takeover T h e o r y C o rp o ra te G o v e rn a n ce in d ic a to r C o n c lu si o n
Table 3 Overview previous academic research
Author Topic Data Set Conclusions
Inside ownership
Hartzell et al (2004) Inside ownership 107 real estate investment trusts (REITs) 1991 - 1998
Higher inside ownership results in a higher initial value and reduces the discount
Petersen (2007) Inside ownership German initial public offerings from 1997 to 2002
Finds a negative impact of the divestment of the inside shareholders on the IPO valuation
Brau and Fawcett (2006) Management ownership Survey amongst CFO of 519 US IPOs (Jan. 2000- Dec. 2002)
Attribute most underpricing to market uncertainty and the need to reward investors for taking the risk of the IPO.
Brennan and Franks (1997) Management ownership Sample of 69 UK IPOs 1986-1989
Favoring smaller shareholders makes it more easy for managers to focus on their own strategy instead of being afraid for hostile activities
Aggerwal et al (2001) Management ownership 264 US manufacturing firms (1987)
Higher ownership is positively correlated with the first-day underpricing
Booth and Chua, (1996), Brennan and Franks, (1997), Brau and Fawcett (2006) and Chemmanur (1993).
Management ownership Various data sets Management that sells bigger stakes, first sells a part of the shares with a bigger discount
Howton et al (2001) Board ownership Sample consists of 412 US firms 1986-1994
Higher board ownership increases the amount of underpricing instead of reducing it
Booth and Chua (1996) Board ownership Sample of 2151 US companies (1977-1988)
Board of directors tend to stimulate underpricing in order to increase the demand for the stocks Filatotchev and Bishop
(2002)
Board ownership Sample of 251 UK IPOs in the United Kingdom 1999-2000
Do not find any significant support for the hypothesis that non executive director’s ownership is associated with the underpricing of an IPO Banu Durukan (2006) Board ownership 182 Turkish IPOs (1994-2003) Weak relation between board ownership and IPO
underpricing Board characteristics
Howton et al (2001) Outside board members Sample consists of 412 US firms 1986-1994
Finds that there is a positive relation between the number of outside board members and the initial return of the IPO
Filatotchev and Bishop (2002)
Outside board members Sample of 251 IPOs in the United Kingdom 1999-2000
Increased number of outside directorships reduces the amount of underpricing
Li and Naughton (2007) Size of the board IPOs
issued in China between 1999 and 2001
The short term return, related to the pricing of the IPO, is positively related with the board size Carter, Dark, and Singh
(1998)
Size of the board 512 US firms 1990-1996 Board size and IPO underpricing are inversely related
Compensation
Lowry and Murphy (2006) Stock options packages 874 US IPOs (1996 - 2000) Companies with IPO options show lower IPO underpricing
Nikbakht et al (2007) Executive compensation 99 US stocks July 1, 2000 - June 30, 2004
Paper shows an influence of the level of CEO salary; however, the influence is relatively small
Hopp and Dreher (2007) Executive compensation Panel data for 29 countries 1988-2005
Inside ownership
Hartzell et al (2004) examined US REITs on the presence of inside ownership in general and in specific the influence on the pricing of an initial public offering. They find that higher inside ownership results in a higher initial value and reduces the discount. New shareholders know that the inside owners have a bigger incentive to act in best interest of the company and therefore the new shareholders are more willing to pay a higher price. This finding is supported by Petersen (2007). Petersen examined the influence of the divestment of inside owners on the valuation of the company at the time of the IPO. He found a negative impact of the divestment of the inside shareholders on the IPO underpricing. The authors state that the divestment results in the necessity to increase the discount thus reducing the offering price.
While Hartzell and Petersen test the influence of inside shareholders in general, many academics choose to focus on one inside stakeholder. Broadly speaking, the literature defines three types of inside ownerships; the management, the board and the employees. This paper focuses on both the influence of the ownership by the management and ownership by the board for the reason that these two are displayed in the offering document (S1).
Zingales (1995) and Brau and Fawcett (2006) examined the influence of the management
ownership in the pricing of an IPO. They conclude that managers find it extremely important to maintain in control. Therefore it is an interesting option to sell the shares with a bigger discount in order to stimulate a wider variety of shareholder to buy the share. Merely, Brennan and Franks (1997) state that favoring smaller shareholders makes it easier for managers to focus on their own strategy instead of being afraid for hostile takeovers. Aggarwal et al (2001) acknowledged this and stated that setting the offering price strategically low maximizes the personal value of the management owners. They find that higher ownership is positively correlated with the first-day underpricing. They argue that management that wants to sell a bigger stake first sells a part of the shares with a bigger discount. This attracts the attention of potential investors and consequently it “shifts the demand curve for the stocks outwards.” After the lock up period4 the management is able to sell the rest of the shares for a higher price. This vision is supported by Booth and Chua, (1996), Brennan and Franks, (1997), Brau and Fawcett (2006) and Chemmanur (1993).
One of the most influential researches on the influence of the board ownership is conducted by Howton et al (2001). The authors conducted a research on the role of the ownership on the IPO pricing irregularities. The academics state that the board can potentially reduce both
4 Lock up: period that the pre IPO owners are not aloud to sell the rest of his stock. This period is
asymmetric information and agency problems. They show that the IPO underpricing is directly related to ownership by insiders. The authors can not find support for the hypothesis that high ownership by the board lowers the asymmetric information between the company and potential investors on the capital market. They find the remarkable fact that higher board ownership increased the amount of underpricing instead of reducing it. Booth and Chua (1996) examined the relation for of the board ownership and the offering price too and, again, they concluded that the board of directors tends to stimulate underpricing in order to increase the demand for the stocks, which makes them able to sell the rest of their stocks more easily after the lock-up period.
In respective to Howton et al, Filatotchev and Bishop (2002) examined the influence of ownership by non executive directors, however, they do not find any significant support for the hypothesis that non executive director’s ownership is associated with the underpricing at the time of an offering. Also Banu Durukan (2006), who tested the influence for the Istanbul Stock Exchange found just a very weak positive relation.
Concluding, although in the academic world there is no consensus about the influence of inside ownership on the pricing of an IPO, quite some researchers do find that companies should focus on enhancing the management and board ownership.
Board characteristics
Second indicator of corporate governance focuses on the board. While more characteristics are being discussed in the academic context, I focus on the relative number of outside board members and the board size, because information about these topics is publicly available and, evidently, possible to measure and compare.
Li and Naughton (2007) and Agrawal and Knoeber (1996) tested the influence of these indicator of corporate governance, however, no significant evidence about the influence in the short run has been found by the authors. Same goes for the research of Sanders and Boivie (2004).
underpricing. They conclude that the board structure can help to overcome problems with and compensate for the lack of executive’s experience and improve the corporate network. Directors from outside the company can therefore raise the value of the firm during an IPO.
The size of the board has much academic interest as well (e.g. Boone et al, 2007 and Bakers and Gompers, 1998). However research about the influence on the amount of underpricing is rather scarce. Li and Naughton (2007) do test the influence of the board size and find a significant influence of the board size on the pricing of the IPO. The authors find that the short term return related to the pricing of the IPO, is positively related with the board size, meaning that increased board size, increases the amount of underpricing. Hartzell et al (2004) and Howton et al (2001) also try to relate the board size to the amount of underpricing; unfortunately these authors did not find a significant influence of the board size on the IPO valuations for respectively US firms and REITs.
Executive compensation
Although many academics have endeavored the influence of executive remuneration on the performance of company, just a few relate the remuneration to the IPO pricing and its related underpricing mechanism. Lowry and Murphy (2006) tested the one-way causality of the influence of executive compensation on the IPO underpricing. The authors tested to what extend IPO pricing was influenced by the adopted stock options packages. They found that one third of the companies that went public have adopted IPO options packages. Their research further concludes that companies with IPO options for the executives on average show lower amounts of underpricing during an IPO than companies without option structures. This means that options packages can effectively reduce the amount of underpricing. The result can not be supported by Nikbakht et al (2007). They examined the influence of executive compensation on the IPO pricing. The academics measured to what extend the cash pay, bonuses, stock and stock options grants contribute to the pricing of an IPO. Their research does show an influence of the level of CEO salary; however, the influence is relatively small.
Finally, a somewhat different research comes from Hopp and Dreher (2007). Their research compares the influence of remuneration on underpricing between different countries. Their results show that countries characterized by more excessive pay packages also show a higher degree of underpricing. Hopp and Dreher examined the influence for 29 countries over the period 1988-2005.
remarkable, because every year hundreds of companies go public worldwide while diluting value for the current shareholder due the necessity of a discount. Therefore, it is interesting that some researches do find a significant influence of some parts of the corporate governance indicators on the amount of underpricing needed to succeed an offering to the public. This means that changing the indicators effectively reduces the need to compensate for information asymmetry, to signal the quality of the firm and to stimulate dispersed ownership in order to reduce block ownership. This means that understanding the relation and adapting the different corporate governance related elements reduces the stakeholder’s conflicts and increases the initial owner’s wealth.
To further increase the transparency of the influence on the three indicators, this paper conducted different tests in order to give an answered on one main question; what is the influence of corporate governance on the price setting of an IPO?
III.
Sample design and data
Sample and data collection
I collected data concerning the US initial public offerings over the period 2006 – September 2008. The list of IPOs was collected from the website of the US Securities and Exchange Commission (SEC) and Bloomberg. The list also contained seasonal offers; however these were excluded from the data set because these companies went public at another point in time. From the US IPOs (2006-2008) I selected a sample of 180 IPOs that took place in the examined period. After deleting 42 companies for missing essential information like compensation and ownership structure, my data set finally contained 138 companies.
The actual filings of the offering (S.1) were extracted from the EDGAR database. The EDGAR database contains the filings of all companies that went public on one of the US stock exchanges. The US government obliges all companies that want to have an US listing, both foreign and domestic, to report there filings in the same way. The companies are required to file documents like registration statements and periodic reports electronically via the EDGAR database. This makes the offerings transparent and comparable. However, foreign based companies, most often Chinese companies, do not file the compensation of their executives in a proper way. In case not all information is included in the filings, the company is excluded from the sample tested.
shareholders” of the filings provide the information used in this report. The management section provides information about the executives and board of advisors and discusses the remuneration of the executives. This information is used to measure the corporate governance structure of the company. Finally, the principal and selling shareholder section gives an overview of the change in ownership following on the offering.
Besides the information extracted from the S.1 filing, I also control for some firm specific elements. This information is extracted using the Factset data source. I choose to account for the market cap and the employees. These control variables are being discussed in a subsequent section. 5
Table 4: Explanations of the variables used
Variable Description
Dependent
RETURN Underpricing measured by the relative difference between the offering price and the 1st
day closing price Independent
Inside ownership
MANOWN Change in management ownership as a result of the offering measured by the absolute difference between before and after the offering (%)
BOARDOWN Change in board ownership as a result of the offering measured by the absolute difference between before and after the offering (%)
INSIDEOWN Change in total inside ownership as a result of the offering measured by the absolute difference before and after the offering (%)
Board characteristics
BOARDSIZE The number of board members at the time of the offering
% INDEP The percentage of independent members in the board at the time of the offering
Compensation structure*
FIX The natural logarithm of the fixed compensation earned by the top 5 executives of the offering company
INCENT The natural logarithm of the incentives earned by the top 5 executives of the offering company
EQUITY The natural logarithm of equity compensation earned by the top 5 executives of the offering company
OPTIONS The natural logarithm of the options received by the top 5 executives of the offering company
OTHER The natural logarithm of the other compensation earned by the top 5 executives of the offering company
Control variables
EMPLOYEES Number of employees of the company at the time of the offering MARKETCAP The natural logarithm of the market capital at the time of the offering
* The US law on compensation requires to publish the compensation of the top 5 executives
Variables
Dependent Variable
errors were found. The return is measured by the relative difference between the closing price on the first trading day and the initial offering price. This information is used as the dependent variable in the research. The return is the dependent variable because the underpricing is the result of inefficiencies and suboptimal elements that already exists within the company. Accordingly, the discount provided to the public tries to compensate for this appearance.
Independent variables
There is a widespread debate about the different ways to measures the corporate governance mechanisms. Like stated in the sections before, I choose to examine the influence of inside ownership, board characteristics and the executive compensation.
Inside ownership
Several types of inside ownership are being discussed in the academic literature. I chose to examine the influence of management ownership, board ownership and the total inside ownership by testing the influence of the relative change in ownership on the IPO underpricing. This means that I collected data on their ownership in percentage before and after the offering and calculated the absolute change. The information about the ownership is extracted from the principal and selling shareholder section in the prospectus. The prospectus gives on overview of the stakeholders that, at the time of an IPO, have more than 1% of the shares. Less than 1% is not defined and therefore not examined (in case of the management and board ownership).
Management ownership is the ownership in terms of shares of the company. Management is directly responsible for the day-to-day operations of the company; therefore, their behavior influences the operating performance of the company. Ljunqvist (2006) acknowledges this vision and states that management ownership is important because of the possible implication it can have on the management’s incentives “to make operating and investment decisions.” Decreased ownership would mean that the management is less stimulated to behave and act in best favor of the company and so a decrease in ownership should increase in the underpricing.
The last element of the inside ownership is measured by the total inside ownership. Because the ownership less than 1% is not defined in the prospectus, the inside shareholders with a minor stake can not be included in the analysis. The prospectus, however, does mention the percentage of total ownership (all inside stakeholders, like employees, management and the board). Therefore, this variable is also included in the analysis. Same as for the management and board ownership, it is also expected that a decrease in total inside ownership means that the insiders are less stimulated to perform in best interest of the company. Consequently, decreased inside ownership should lead to a higher amount of underpricing.
Board characteristics
Second indicator, the board characteristics, is measured through testing the influence of two variables on the amount of underpricing; the number of outside board members and the size of the board.
Due to the Sarbanes Oxley act (2002), at least half of the board must consist of members that come from outside the company. After reading all the prospectuses I concluded that quite some companies not yet comply with this act, but they stated in their offering document that they are planning to do so during their next shareholders meeting. In order to measure the board
independence, I collected information on the total number of board members and the number of members that come from outside the company, called independent board members. I calculated this variable as the relative number, so in percentage, of independent board members on the board. Based on previous literature (e.g. Baysinger and Butler, 1985 and Agrawal and Knoeber, 1996), I expect that a higher percentage of independent board members leads to a lower amount of underpricing. The outside directors have fewer incentives to misbehave due to the risk of harming their own reputation and the absence of private benefits. Therefore, they have less conflicting interests. Important aspect is that the board does not profit from these outside board members and, most of the time, do not profit from take-over activities. As a result, they have a more objective position at the time of an IPO.
2001). The increased monitoring decreases the need for underpricing, because investor are aware of the fact that it is more difficult and, consequently, less likely to show value diluting behavior by the management.
Information about the board characteristics is extracted from the prospectuses as well. The management section of the filings gives on overview of the size of the board of directors and the number of independent directors.
Executive compensation
The final indicator examined in this paper is the influence of executive compensation on the IPO underpricing. The compensation should contain incentives to stimulate to behave in line with the expectations of the shareholders. If the compensation is too low or does not stimulate behavioral actions at all, the management will seek for private benefits and, consequently, destroy shareholder value. Stimulating, in the context of the compensation structure, means that the management receives a reward for acting in best interest of the company. This means that the variable remuneration parts should only be paid out when the executive performed in accordance with the expectations of the shareholders. Therefore the higher the compensation, the better the performance of the management and the less underpricing is needed.
In general the different parts of the compensation are fixed, bonus, stocks, options and
“other” compensation. For the fixed compensation I do not expect any relation, for all the variable elements I expects a relation in line with mentioned above.
The information about the compensation is extracted from the compensation summary table, part of the management section depicted in the prospectus of the offering company. I take the natural logarithm of the compensation parts as mentioned in the compensation section in order to account for far outliers and use these as the different variables in my regressions.
Control variables
Like stated in the previous section I use the market cap and number of employees as the control variables in my research. The variables used account for firm specific characteristics (see Bethel et al, 1993, Gibbs, 1993). Like stated in the in the methodology, I expect that larger companies, both in terms of market capital as in number of employees receive more media attention and more attention of analysts. Consequently, these companies have a decreased necessity to underprice in order to receive interest of different potential investors on the capital market.6
Descriptive statistics
Table 5 shows the descriptive statistics for the dataset used in this paper. The total dataset contains 138 observations, however in some offering documents information about one of the variables is missing. The dependent variable does not have far outliers. The influence of far outliers in the independent variables is strongly reduced by calculating the natural logarithm return in case of the compensation variables.
Table 5 Descriptive statistics7
Obs. Mean Median Min. Max Std. Dev. Prob.
RETURN 138 0.161 0.097 -0.269 0.972 0.258 0.000 MANOWN 137 -0.059 -0.024 -0.986 0.036 0.114 0.000 BOARDOWN 137 -0.135 -0.100 -0.725 0.194 0.141 0.000 INSIDEOWN 137 -0.142 -0.116 -0.598 0.036 0.130 0.000 BOARDSIZE 137 6.949 7.000 1.000 13.000 1.848 0.001 % INDEP 138 0.594 0.625 0.000 1.000 0.233 0.005 FIX 138 10.480 12.213 0.000 13.921 4.484 0.000 INCENT 138 7.798 11.044 0.000 14.496 5.613 0.000 EQUITY 138 2.551 0.000 0.000 15.237 5.029 0.000 OPTIONS 138 5.916 8.825 0.000 14.780 5.502 0.000 OTHER 138 6.766 9.383 0.000 14.914 5.059 0.000 EMPLOYEES 133 6.331 6.114 1.792 10.621 1.552 0.528 MARKETCAP 137 20.177 20.009 16.939 24.137 1.146 0.000
Note: Return is the dependent variable and is the underpricing calculated by the difference between the offering price and the first day closing price. The independent variables are measured by; Management ownership (MANOWN), Board ownership (BOARDOWN) and inside ownership (INSIDEOWN ) as the absolute difference in percentage between before and after the offering, size of the board (BOARDSIZE) is provided as the number of members at the time of the offering, the relative number of independent board members (% INDEP) is determined at the time of the offering, and different type of compensations (FIX, INCENT, EQUITY, OPTIONS and OTHER) is calculated as the natural logarithm of the values reported in the offering document. EMPLOYEES and MARKETCAP are the control variables for the firm size during the offering resp nominal and natural logarithm value.
The summary of descriptive statistics shows that the average amount of underpricing of a company that went public is 16.1%8. This result clearly shows that for this dataset the selling shareholders left money on the table by actually selling the stocks below the shareholders value. Other remarkable fact is that in one occasion the first day closing price was 97.2%9 above the offering price, while the closing price of the worst performing offering is just 26.9% below the offering price. This huge difference clarifies why the median (9.7%) is below the average first day return.
7
All descriptive statistics see appendix: Table A1 Descriptive statistics all info
8 This is consistent with the result of previous results; E.G Loughan and Ritter (18%), 2002 and Filatotchev and Ritter
2002 (29%)
9
Further notable facts are the change in management ownership (-5.9% of the total company shares) and change in board ownership (-13.5% of the total company shares) is quite a big difference. This means that on average, the board sells the biggest stake in absolute terms. The compensation is tested through its log return to account for far outliers. The results show that for componsation there is strong difference between the minimum and maximum rewarding per element.
The correlations and covariance between different variables used in the models are displayed in table A2 and table A3 in the appendix. The dataset contains a strong correlation between the variables within the corporate governance indicator inside ownership. All three variables (management ownership, board ownership and inside ownership) show a correlation of more than 0.404. Besides, there is a clear correlation between the independent variable “incentive” and the variables “fix” and “other compensation”.
IV.
Methodology and Hypotheses development
To examine the influence of the corporate governance on the IPO underpricing, I use information on the amount of underpricing (dependent variables), information on variables that potentially have relationship with dependent variable (independent variables) and information on the control variables that take into account firm specific extraneous factors that might influence the model.
After collecting all the cross sectional data, I test the one-way causality of the influence of corporate governance on the IPO underpricing. I use a multiple regression model to test the influence of governance indicators (see for related studies Hill, 2006, Filatotchev and Bishop 2002 and Nikbath et al, 2007 for same type of regression models). The basic model of the simple linear regression model10 is:
t t
t x u
y =
α
+β
+Where t denotes the observation number and u the disturbance term.
The models used in this paper are an extension of the basic model and look the following:
Model (1) ε β β β β β β α+ + + + + + +
= MANOWN BOARDSIZE INDEP OTHER MARKETCAP EMPLOYEES
RETURN 1 2 3% 4 5 6
10
Model (2) ε β β β β β α+ + + + + +
= BOARDOWN INCENT EQUITY MARKETCAP EMPLOYEES
RETURN 1 2 3 4 5 Model (3) ε β β β β β α + + + + + +
= INSIDEOWN INCENT OPTIONS MARKETCAP EMPLOYEES
RETURN 1 2 3 4 5 Model (4) ε β β β β β β α+ + + + + + +
= INSIDEOWN BOARDSIZE INDEP FIX MARKETCAP EMPLOYEES
RETURN 1 2 3% 4 5 6 Model (5) ε β β β β β α+ + + + + +
= BOARDOWN BOARDSIZE INDEP MARKETCAP EMPLOYEES
RETURN 1 2 3% 4 5 Model (6) ε β β β β β β β α+ + + + + + + +
= MANOWN BOARDSIZE INDENP FIX EQUITY MARKETCAP EMPLOYEES
RETURN 1 2 3% 4 5 6 7 Model (7) ε β β β β α+ + + + +
= MANOWN INCENT EQUITY MARKETCAP
RETURN 1 2 3 4
The construction of the models is based on the correlation between the different independent variables. All independent variables are tested in two different models at least. The independent variables of each model are selected based on their mutual correlation which has to be low in order to be both accepted in one model. The correlation between each variable is depicted in table A2 (appendix). Besides the models, this paper also tests the influence one by one to find if there is any correlation. The models allow to test if there is any significant relation between the independent and the dependent variable. One additional model (number 7) is constructed based on the outcome of each individual link with the return of the IPO displayed in table A4 in the appendix. This model is based on the outcome of the adjusted R squared.
I use market cap and number of employees as control variables11 in my research to account for firm specific characteristics. I expect that larger companies, both in terms of market capital as in number of employees receive more media attention as well as from analysts. Consequently, these companies have a decreased necessity to underprice in order to receive interest of different potential investors on the capital market. This view is supported by several previous conducted researches (see Bethel et al, 1993, Gibbs, 1993).
Based on the previous literature and empirical findings, I test the influence of indicators of good governance on the IPO pricing. The general hypothesis discussed in the literature review12 is answered by testing the sub hypotheses related to each of the three corporate governance indicators.
V.
Findings
In this section of the paper I discuss the outcome of the 7 models analyzed. I test if there is any relation between the independent variables and the underpricing of an IPO.
The first test conducted is the basic ordinary least square method (OLS). This method assumes that the estimators are normally distributed in the residuals and helps to find the unknown parameters. In order to be valid, the OLS has a number of assumptions that must be checked in accordance, the so called best linear unbiased estimator (BLUE) assumptions13.
Table 6 displays the results of the normality test for the residuals. The Jarque-bera shows that the models are not normally distributed. Brooks (2002) states in his book that if “sample sizes are sufficient large, violation of normality is virtually inconsequential.” Brooks does not elaborate on when a dataset is sufficient large. Based on the finding of Brooks, the non-normality is accepted and the influence of the different variables is tested.
11
I checked the data on the year effect, however a significant effect was found. My dataset contains to small and to many different industries to include this effect in my analysis
12
H1 = Ceterus paribus, adopting corporate governance indicators reduces the underpricing of an IPO. 13
Table 6 Normality test of the residuals per model
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
Mean 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Median -0.060 -0.043 -0.039 -0.073 -0.053 -0.052 -0.062 Minimum -0.421 -0.453 -0.424 -0.431 -0.452 -0.480 -0.444 Maximum 0.770 0.760 0.754 0.796 0.770 0.765 0.759 Std. Dev. 0.259 0.255 0.254 0.259 0.260 0.254 0.251 Skewness 0.971 0.954 0.934 1.011 0.972 0.902 0.993 Kurtosis 3.463 3.501 3.485 3.598 3.514 3.408 3.554 Jarque-Bera 21.584 21.236 20.317 24.103 21.890 18.520 24.077 Prob. 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Obs. 130 131 131 130 130 130 136
Note: OLS accounts for the existence of heteroscedasticity. Return is the dependent variable and is the underpricing calculated by the difference between the offering price and the first day closing price. The independent variables are measured by; Management ownership (MANOWN), Board ownership
(BOARDOWN) and inside ownership (INSIDEOWN ) as the absolute difference in percentage between
before and after the offering, size of the board (BOARDSIZE) is provided as the number of members at the time of the offering, the relative number of independent board members (% INDEP) is determined at the time of the offering, and different type of compensations (FIX, INCENT, EQUITY, OPTIONS and
OTHER) is calculated as the natural logarithm of the values reported in the offering document. EMPLOYEES and MARKETCAP are the control variables for the firm size during the offering resp
nominal and natural logarithm value.
Table 7 on the page 25 shows the outcomes of the ordinary least squares tests conducted for each model. The F-statistic tests the null-hypothesis that all of the slope coefficients in the regression are zero (Brooks, 2007). Based on the outcome of the OLS, only model 7 shows significant viable results. This means that only this model can be used for further analysis of the relations. The other six models do not provide any interesting information; accept that no relation can be proved due to their insignificance.
Model 7 shows significant results for the one way influence of the two independent variables and one control variable on the return. The first variable that has a significant influence on the underpricing of an IPO is the management ownership. With a coefficient of 0.225 the management ownership has a rather strong influence on the underpricing. The influence, however is, is quite remarkable. The coefficient implicates that a stronger decrease of the ownership by the board results in lower underpricing of the offering. Theory expects it to be the other way around; if de management decides to sell a bigger stake, the amount of underpricing needed to succeed should be higher.
underpricing. The higher the equity compensation part, so the amount of stocks retained, the lower the amount of underpricing. This is in line with the theory discussed previously.
Table 7 Results multiple regression models
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
Variable Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient (t-statistic) (t-statistic) (t-statistic) (t-statistic) (t-statistic) (t-statistic) (t-statistic)
C -0,064 -0,469 -0,429 -0,181 -0,083 -0,310 -0,524 (-0.149) (-1.164) -1,071 (-0.413) (-0.198) (-0.688) -1,405 MANOWN 0,222 0.327** 0.225* (-1.479) (-2.022) (1,844) BOARDOWN 0,122 0,093 (0,975) (0,723) INSIDEOWN 0,235 0,205 (1,590) (1,363) BOARDSIZE -0,010 -0,009 -0,008 -0,016 (-0.821) (-0.789) (-0.671) (-1.355) %INDEP 0,024 0,003 0,036 -0,017 (0,240) (0,035) (0,368) (-0.183) FIX 0,004 0,003 (0,716) (0,624) INCENT 0,006 0,006 0,006 (1,387) (1,473) (1,650) EQUITY -0.009** -0.010*** -0.096** -0.010*** (-2.493) (-2.624) (-2.400) -2,797 OPTIONS 0,003 (0,704) OTHER -0,004 -0,004 (-0.902) (-0.004) MARKETCAP 0,012 0,030 0,028 0,016 0,012 0,025 0.033* (0,595) (1,408) (1,322) (0,751) (0,572) (1,149) (1,832) EMPLOYEES 0,012 0,004 0,006493 0,012 0,013 0,013 (0,904) (0,298) (0,468) (0,839) (0,967) (0,943) Obs 130 131 131 130 130 130 136 Adj R-squared -0,018 0,016 0,024 -0,022 -0,024 0,004 0,037 Prob(F-statistic) 0,709 0,224 0,156 0,786 0,809 0,397 0.063*
Note: OLS accounts for the existence of heteroscedasticity. Return is the dependent variable and is the underpricing calculated by the difference between the offering price and the first day closing price. The independent variables are measured by; Management ownership (MANOWN), Board ownership (BOARDOWN) and inside ownership (INSIDEOWN ) as the absolute difference in percentage between before and after the offering, size of the board (BOARDSIZE) is provided as the number of members at the time of the offering, the relative number of independent board members (% INDEP) is determined at the time of the offering, and different type of compensations (FIX, INCENT, EQUITY, OPTIONS and
OTHER) is calculated as the natural logarithm of the values reported in the offering document. EMPLOYEES and MARKETCAP are the control variables for the firm size during the offering resp nominal and natural logarithm value. ***
indicates significance at the 1% level, ** at 5% level and * at 10% level
VI.
Discussion
The main objective of this paper is to give insight in the influence corporate governance on the underpricing of an initial public offering. Like mentioned in the literature, there is no consensus in the academic literature about to what extend corporate governance influences the direct wealth transportation from original owners to the new public owners, some papers even doubt if there is a clear-cut relation.
This paper examines the influence of corporate governance by testing three different indicators; the inside ownership, the board characteristics and the executive compensation. Furthermore, this paper aims to link the influence to the three main reasons for underpricing widely discussed in the academic world, namely, the information asymmetric theory, the signaling hypothesis, and dispersed ownership theory. The three theories are developed over de last decades en help to understand why on average the shares are sold with a discount. I found that on average selling shareholders underprice the share by 16.1%, partly caused by the corporate governance mechanisms, thereby, leaving money on the table.
Although many academics do find interesting relations of three corporate governance indicators14, I can just partly support their findings. The first indicator, inside ownership, was tested by examining the influence of management ownership, board ownership and inside ownership and the percentage change in absolute terms between the pre- and post-IPO share ownership. Although this paper does not find any support for the influence of board and inside ownership, I do find a relation between management ownership and IPO underpricing. I find a significant coefficient of 0.225 in the only significant model (7). This outcome is not in line with the expectations. An increase in the stake sold by the management, results in decreased underpricing. The findings of this paper can to a certain extent support previous findings of e.g. Hartzell et al (2004), Peterson (2007) and Harjato and Garen (2005) that the ownership is directly related to underpricing, however the influence is different than expected. Main reasons for the difference in outcome is due to dissimilarity in the examined period (e.g. my dataset does not include the dot.com bubble) and focuses just on the US (market is relatively transparent).
The second indicator is the board characteristics; I chose to examine the influence of the board size and the number of outside (independent) board members. Already during the literature review it became clear that the two variables received limited academic attention and the papers that examined the indicator all came up with there own conclusions, so there is no real consensus on the impact amongst the academics. Also this paper did not find any relationship between the
14
two variables and IPO underpricing. None of the models tested show significant results for the variables. This is in line with the researches conducted by e.g. Hartzell et al (2004) and Sanders and Boivie (2004).
Final indicator examined is the compensation15 of the management. This paper examines different kinds of compensations based on the parts that the companies have to report in there offering document (S.1.). Lowrey and Murphy (2006) and Nikbaht et al (2007) find that companies with stock options for their executives show lower amounts of underpricing and that the executives’ salary has a significant, but relatively small influence on the underpricing. This paper only finds a relation between the equity component of the compensation and the underpricing of an initial public offering. After calculating the value of the natural logarithms for the different compensation elements and accounting for heteroscedasticity, the equity variable shows a significant result. The coefficient is -0.010, meaning that the influence is relatively small. The finding is in line with the expectations. If the amount of equity compensation increased, the need for underpricing decreases. This means that higher equity compensation will result in the fact that the management will act in best favor of the shareholders and so, compensation in the form of underpricing is not required.
Overall I can state that this paper finds some influence of corporate governance on the IPO underpricing. Although an exact linkage between the results and the theory is not possible to examine, I find that the influence of the management ownership and the equity compensation do influence the amount of underpricing. When going public, the management should take this conclusion into consideration to minimize their wealth transportation to the new owners.
Most important question remains; why do not most of the models show significant viable results, while key papers within the area of interest do find quite some significant models? The differences in outcome of the relation analyzed merely come from a couple of differences. Most of the papers published on this topic, have a relatively outdated dataset. One of the oldest datasets on the linkage of corporate governance and the IPO underpricing dates from 1977 (Booth and Chua, 1996), while the most recent data comes from the year 2005 (Hopp and Dreher, 2007). All other papers contain information within this time range, meaning that my dataset is way more up-to-date. Besides, many papers examine countries other than the US (e.g. Petersen, 2007, Brennan and Franks, 1997, Filatotchev and Bishop, 2002 and Lu and Naughton, 2007). Other important differences are the focus on one specific industry (e.g. Aggarwal et al, 2001 and Hartzell et al, 2004), the bigger size of the dataset (e.g. Booth and Chua, 1996 and Howton et al, 2001), the
15
slightly different way of measuring the independent variable (e.g. Petersen, 2007) and some even do not specify if their model is significant (Hartzell et al, 2004).
To conclude, main limitation of this paper is the examination of just three years, focusing on just one country and the number of companies examined does not make it possible for analyzing industry differences. Further research is needed in order to reveal more causal relations of corporate governance and the discount provided to new owners if a company goes public for the first time. Next papers might be able to find influence by examining and comparing different countries, including more variables like the age of the executives and the ties with other companies or extending the examined period to look for changes in the influences over the past.
VII.
Conclusion
This paper tries to give insight in the one-way causality of corporate governance on the price setting of an initial public offering. I collected data concerning the US initial public offerings over de period 2006 – September 2008. I selected at random 180 IPOs that took place in he examined period. After deleting 42 companies for missing essential information like compensation and ownership structure, I ended up with a data set containing 138 companies.
I examine three indicators (inside ownership, board characteristics and executive compensation) in an attempt to prove the relation with the theories about the influence of corporate governance on the IPO underpricing put forward by many academics, respectively the information asymmetry, signaling theory and the dispersed ownership theory.
VIII.
Appendix
A1 Descriptive statistics all info
Obs. Mean Median Min. Max Std. Dev. Prob. Skewness Kurtosis Jarque-Bera
RETURN 138 0,161 0,097 -0,269 0,972 0,258 0,000 1,144 3,826 34,023 MANOWN 137 -0,059 -0,024 -0,986 0,036 0,114 0,000 -4,852 35,170 6445,040 BOARDOWN 137 -0,135 -0,100 -0,725 0,194 0,141 0,000 -1,258 5,270 65,520 INSIDEOWN 137 -0,142 -0,116 -0,598 0,036 0,130 0,000 -1,051 3,939 30,260 BOARDSIZE 137 6,949 7,000 1,000 13,000 1,848 0,001 0,349 4,449 14,756 % INDEP 138 0,594 0,625 0,000 1,000 0,233 0,005 -0,679 2,960 10,615 FIX 138 10,480 12,213 0,000 13,921 4,484 0,000 -1,885 4,647 97,292 INCENT 138 7,798 11,044 0,000 14,496 5,613 0,000 -0,619 1,502 21,716 EQUITY 138 2,551 0,000 0,000 15,237 5,029 0,000 1,514 3,425 53,777 OPTIONS 138 5,916 8,825 0,000 14,780 5,502 0,000 -0,058 1,183 19,055 OTHER 138 6,766 9,383 0,000 14,914 5,059 0,000 -0,462 1,476 18,275 EMPLOYEES 133 6,331 6,114 1,792 10,621 1,552 0,528 0,208 3,240 1,276 MARKETCAP 137 20,177 20,009 16,939 24,137 1,146 0,000 0,551 4,397 18,082
Table A2 Correlation
RETURN MANOWN BOARDOWN INSIDEOWN BOARDSIZE % INDEP FIX INCENT EQUITY OPTIONS OTHER EMPLOYEES MARKETCAP RETURN 1 MANOWN 0.098 1 BOARDOWN 0.075 0.404 1 INSIDEOWN 0.094 0.414 0.858 1 BOARDSIZE -0.038 0.125 -0.120 -0.103 1 % INDEP -0.011 0.118 -0.023 0.052 0.077 1 FIX 0.027 -0.125 -0.107 -0.194 0.122 0.113 1 INCENT 0.081 0.031 -0.103 -0.176 -0.013 0.288 0.559 1 EQUITY -0.144 0.098 -0.004 0.079 -0.073 -0.025 -0.009 0.091 1 OPTIONS 0.043 -0.196 -0.080 -0.099 0.126 0.176 0.306 0.224 -0.170 1 OTHER -0.088 -0.018 -0.160 -0.225 0.082 0.215 0.516 0.631 0.119 0.283 1 EMPLOYEES 0.077 0.084 -0.001 -0.096 0.276 -0.131 0.040 0.129 0.141 -0.142 0.082 1 MARKETCAP 0.094 0.053 0.119 0.124 0.055 -0.180 -0.217 -0.192 0.168 -0.171 -0.121 0.440 1
Note: Return is the dependent variable and is the underpricing calculated by the difference between the offering price and the first day closing price. The independent variables are measured by; Management ownership (MANOWN), Board ownership (BOARDOWN) and inside ownership (INSIDEOWN ) as the absolute difference in percentage between before and after the offering, size of the board (BOARDSIZE) is provided as the number of members at the time of the offering, the relative number of independent board members (% INDEP) is determined at the time of the offering, and different type of compensations (FIX,
INCENT, EQUITY, OPTIONS and OTHER) is calculated as the natural logarithm of the values reported in the offering document. EMPLOYEES and MARKETCAP are the control variables for the firm size during
the offering resp nominal and natural logarithm value.
Table A3 Covariance
RETURN MANOWN BOARDOWN INSIDEOWN BOARDSIZE % INDEP FIX INCENT EQUITY OPTIONS OTHER EMPLOYEES MARKETCAP RETURN 0.068 0.003 0.003 0.003 -0.018 -0.001 0.031 0.117 -0.186 0.061 -0.114 0.031 0.028 MANOWN 0.003 0.012 0.006 0.006 0.024 0.003 -0.060 0.019 0.053 -0.117 -0.010 0.014 0.007 BOARDOWN 0.003 0.006 0.020 0.015 -0.030 -0.001 -0.066 -0.080 -0.003 -0.061 -0.112 0.000 0.019 INSIDEOWN 0.003 0.006 0.015 0.016 -0.023 0.002 -0.107 -0.122 0.049 -0.068 -0.140 -0.019 0.018 BOARDSIZE -0.018 0.024 -0.030 -0.023 3.223 0.032 0.960 -0.127 -0.648 1.231 0.731 0.769 0.112 % INDEP -0.001 0.003 -0.001 0.002 0.032 0.055 0.115 0.370 -0.029 0.223 0.248 -0.047 -0.048 FIX 0.031 -0.060 -0.066 -0.107 0.960 0.115 19.225 13.490 -0.184 7.289 11.183 0.270 -1.081 INCENT 0.117 0.019 -0.080 -0.122 -0.127 0.370 13.490 30.315 2.481 6.692 17.168 1.100 -1.202 EQUITY -0.186 0.053 -0.003 0.049 -0.648 -0.029 -0.184 2.481 24.484 -4.566 2.920 1.084 0.946 OPTIONS 0.061 -0.117 -0.061 -0.068 1.231 0.223 7.289 6.692 -4.566 29.540 7.603 -1.198 -1.056 OTHER -0.114 -0.010 -0.112 -0.140 0.731 0.248 11.183 17.168 2.920 7.603 24.426 0.627 -0.678 EMPLOYEES 0.031 0.014 0.000 -0.019 0.769 -0.047 0.270 1.100 1.084 -1.198 0.627 2.403 0.775 MARKETCAP 0.028 0.007 0.019 0.018 0.112 -0.048 -1.081 -1.202 0.946 -1.056 -0.678 0.775 1.293
