INITIAL PUBLIC OFFERINGS
AND EARNINGS MANAGEMENT
An examination of the relation between IPO (under)pricing,
subsequent performance and earnings management
Judith Harmsma - 10266518
Bachelor Accountancy & Control
University of Amsterdam
June 2014 - final version
Under supervision of:
Mario Schabus, MSc
University of Amsterdam
Amsterdam Business School
Table of Contents
1 Abstract
3
2 Summary (Dutch)
3
3 Introduction
4
4 Initial Public Offerings
4.1 The IPO process
6
4.2 Theories on underpricing
4.2.1 Underpricing equilibrium
6
4.2.2 Theories involving underwriters
7
4.2.3 Signaling theory
9
4.3 Theories on overpricing
10
5 Earnings management
5.1 Definition
12
5.2 Measurement and detection
13
5.3 Opportunistic earnings management
14
5.4 Beneficial earnings management
15
6 Earnings management in IPOs
6.1 Incentives
17
6.2 Measurement
17
6.3 Evidence
6.3.1 Income maximization
19
6.3.2 Income smoothing
20
6.4 Effect of earnings management in IPOs
6.4.1 Short-term effects
20
6.4.2 Long-term effects
21
6.5 Factors influencing earnings management in IPOs
6.4.1 Third party involvement
22
6.4.2 Regulations
22
7 Discussion
24
8 Conclusion
26
1 Abstract
Earnings management around Initial Public Offerings (IPO) is topic of discussion. While it has been shown that management of IPO firms tends to manage earnings upwards, it seems that IPO shares are in general suffering underpricing. The offer price is lower than the first-day returns and IPOs tend to underperform in subsequent years. This thesis tries to address the discrepancy and brings together theories on underpricing and subsequent performance and earnings management. The main finding is that underpricing is related to information asymmetry and that earnings management can both reduce and increase information asymmetry.
2 Summary
Wanneer een bedrijf haar aandelen voor het eerst gaat verhandelen op de aandelenmarkt, vindt een zogenaamde IPO (Initial Public Offering) plaats. Het bedrijf schakelt normaal gesproken een bank in om de aandelen op de markt te brengen. Doordat er een groot gebrek is aan informatie en het management vaak over meer informatie beschikt dan de potentiële investeerders, treedt er een informatie asymmetrie probleem op. De investeerders zijn afhankelijk van de prospectus die de bank opstelt over de financiële situatie van het bedrijf. Dit leidt ertoe dat het management van het bedrijf een prikkel heeft om te participeren in 'earnings management'. Dit houdt in dat het inkomen en de winst van een bedrijf zodanig gemanipuleerd wordt dat het bedrijf beter voor de dag lijkt te komen. Management kan dit doen door kosten te vertragen en winst juist eerder op te nemen in de winst- en
verliesrekening. Het blijkt echter dat aandelen doorgaans lager geprijsd worden dan wat de markt er uiteindelijk voor geeft. 80% van de IPOs wordt op de eerste handelsdag voor een hoger bedrag verhandeld. Er zijn verschillende theorieën die proberen uit te leggen waarom IPOs zo vaak te laag geprijsd worden. In deze scriptie probeer ik een verband te leggen tussen 'earnings management' en het prijzen van de IPO. Informatie asymmetrie blijkt een grote factor te zijn in het onderprijzen van IPOs. 'Earnings management' kan de informatie asymmetrie zowel vergroten als verkleinen.
3 Introduction
Initial public offerings (IPOs) take place when a firm decides to take its shares publicly. Brau and Fawcett (2006) state that there are four main reasons for an IPO. First, the cost of capital of the issuing firm will decrease. Second, it is a way for insiders in the issuing firm to cash out. Third, public shares can be used as a currency in takeovers. If a firm is willing to be acquired by another firm, an IPO makes the acquisition easier. Fourth, the IPO may be part of a strategic move: It can increase publicity and reputation of the firm going public.
In order to market the stock the issuing firm hires an underwriter, this is usually an investment bank. The most common ways to market the shares is in a public auction or through the book-building process. In a public auction, all investors are allowed to bid on the shares. However, in the book-building process, only a few selected investors are able to buy the shares. These investors are invited to buy shares by the investment bank, which may suggest that the underwriters favor certain investors.
There has been a lot of research on earnings management around IPOs. The majority of these studies argue that management of an IPO issuing firm is likely to manage earnings. For example, Caramanis and Lennox (2008) find that abnormal accruals to increase income are more likely to occur if companies are facing an IPO. The issuing companies have an incentive to artificially boost reported earnings, because this leads to higher cash income for the firm. Management of the issuing firm usually has shares
themselves, which makes the incentive to boost earnings in order to achieve a higher price even more likely. These incentives suggest that shares will be overpriced. While the theory on earnings management indicates that overpricing will be likely in an IPO, the opposite seems to be true in practice. 80% of the IPOs are traded for a higher price after the first issuing day (the so-called aftermarket) and the surplus is on average 18.8%. There are several theories on underpricing and reducing this underpricing: The reputation of the underwriter may play a role, as well as opting for the book-building process instead of a public auction. A selection of these theories on underpricing will be further examined in this study, as well as evidence and theories about overpricing.
This thesis will focus on the relation between earnings management and the pricing of shares in an IPO. My research question is therefore: How does earnings management
around an initial public offering (IPO) relate to stock pricing in an IPO and subsequent performance? While it has been shown that earnings management happens around IPOs, the impact on the IPO prices is mixed. This problem is interesting, because the implication
that management has an incentive to overprice its shares contradicts the empirical evidence around IPO pricing. This study contributes to the literature, because it brings together
theories on the pricing of IPOs and earnings management. Most studies have been focusing on each subject separate or on empirical evidence around earnings management in IPOs. This paper will try to link the empirical evidence on earnings management to the theories on IPO pricing.
The remainder of this thesis is organized as follows. First, the IPO process will be explained as well as some theories about underpricing and overpricing. In the second section, earnings management in general will be further examined and subsequently, some studies about earnings management in IPOs will be outlined. Finally, I will try to relate the theories on stock pricing to earnings management in the discussion section.
4 Initial public offerings
4.1 The IPO process
Jenkinson and Jones (2009) provide insights in the process of pricing and allocations of IPOs, through a survey among institutional investors. It should be mentioned that their study only focuses on share allocation through the book-building process, which will be further explained in the following section.
After a firm has announced an IPO and has found an underwriter, research reports about the issuing firm will be distributed among investors. The underwriter discusses the IPO with the investors and might set up meetings where the issuing firm has one-on-one sessions with the potential investors. Jenkinson and Jones (2009) list several reasons for these. First, they help investors make a proper valuation of the firm. Second, investors may collect some insight information about the offering, for example the popularity of the IPO among other investors. Third, these meetings may help the investors to receive the allocation (i.e. number of shares) they opt for.
After these meetings, the bidding on the shares begins. Investors can bid in three different ways. First, they can submit a strike bid in which they accept the price that is set for the share and the only decision is how many shares will be allocated to them. Second is the limit bid in which the investor sets a maximum price for the quantity they wish to receive. Third, there is a mix of the limit and the strike bid, meaning that neither the price nor the quantity is predetermined.
The actual allocation of the shares is not only based on the share price. The timing, size and type of bid also influence the number of shares allocated to the investor. Jenkinson and Jones (2009) find that investors perceive the size of the bid as the most important determinant for share allocation.
4.2 Theories on underpricing
4.2.1 Underpricing equilibrium
IPO shares are often underpriced. According to the perfect market theory, shares that are publicly traded should reflect the firm's value perfectly. However, Ritter and Welch (2002)
found in their study that on the first trading day, shares are traded 18.8 percent above the IPO price. They also found that 80 percent of the IPOs in their study was traded at a higher price than the issuance price. Both findings support the indication that IPO shares are underpriced.
Beatty and Ritter (1986) argue that Akerlof's lemons problem (1970) explains the consistent underpricing of the IPOs. They found that there is an equilibrium relation between underpricing and pre-IPO valuation uncertainty. Beatty and Ritter (1986) base their theory on the fact that not all shares increase in price. Some decline on the first trading day, which means an investor cannot costless distinguish between good and bad quality firms, leading to information asymmetry. This information asymmetry can be reduced if the investor incurs some costs to gather more information. However, the investor is only willing to incur these costs if he earns sufficient profits on the shares. Another way to reduce information
asymmetry is by signaling quality. Good quality firms will make an investment to signal their quality, which cannot be imitated by poor quality firms. There are several theories that try to explain the underpricing in the IPO market. First, I will examine theories involving
underwriters and then I will discuss the signaling theory.
4.2.2 Theories involving underwriters 4.2.2.1 Bookbuilding
The first theory that links underpricing to underwriters is based on the choice between issuing the shares through a public auction or with the book-building method. An issuing firm usually hires an investment bank to underwrite and market the shares. The investment bank has two ways in which it can market the shares. The first option is through a public auction. All investors are able to buy shares in these auctions. Underwriters try to gather information from investors about their willingness to buy shares. However, these investors do not have an incentive to provide the underwriter with correct information about their willingness to buy shares. In fact, they have an incentive to hide information from the underwriter. Without information about the willingness to pay from investors, the underwriter is likely to underprice the shares. The investor benefits from underpricing by buying the shares for a low price and selling them later (Benveniste & Spindt, 1989). The second option is the book-building process. The book-building process works as follows. First, the issuing firm selects an
underwriter to market the shares. Then, the underwriter evaluates and examines the financial prospects of the firm and sets a price range. The underwriter selects investors who might be interested in the shares and invites them to take a look at the firm. After the investors decide
that they want to buy the shares, the shares are allocated to them. Jenkinson and Jones (2009) find that investors who build a valuation-model, and thus are well informed, are less likely to share their view on the valuation of the IPO in the pre-bookbuilding stage. To make the investors reveal information about their willingness to buy the shares, the underwriter needs to reward the investors. Underpricing the shares and thus making investors benefit can provide this reward system. Evidence, on which method of IPO pricing is more accurate, is mixed. Sherman (2005) finds in her study that controlling access leads to a more accurate price, because the underwriter has control over the number of bidders participating in the IPO process. Jenkinson and Jones (2009) however, state that 71% of the responding investors set a price limit lower than the perceived true valuation of the firm to increase profits, which supports the theory on underpricing. Derrien and Womack (2003) find empirical evidence in favor of public auctions. They state that auctions lead to prices that better reflect current and past market conditions, in contrary to price established through the book-building process.
4.2.2.2 Firm commitment vs. best efforts
Another factor that may influence the level of underpricing is whether the underwriter markets the shares based on the firm commitment approach or on the best efforts approach. In a firm-commitment offer, the underwriter buys all the shares and sells them personally. In a best-effort arrangement, the underwriter serves as a broker between the issuing firm and the investors. Benveniste and Spindt (1989) argue that the firm commitment approach leads to higher levels of underpricing. The underwriter commits to buying all the shares that are not presold at the offer price. This gives the underwriter an incentive to sell as much shares as it can, especially when the price uncertainty of the shares is high. Because the underwriter allocates a lot of shares, initial prices will be too low in order to sell as much shares as possible, leading to underpricing. This can be overcome by selling the shares through the best-effort approach. In this case, the underwriter is not required to buy the shares left unsold. The disadvantage of the best-effort approach is the fact that it leads to high uncertainty about the proceeds for the underwriter.
4.2.2.3 Underwriter reputation
The third theory relating to underpricing is about the influence of the underwriter reputation. Carter and Manaster (1990) found empirical evidence that indicate that a more prestigious underwriter has an incentive to maintain its good reputation. This means that the underwriter is less likely to market shares of a risky firm. Good quality firms can thus reveal information about their quality by hiring a prestigious underwriter. Titman and Trueman (1986) support
this theory. They examined the reputation of the auditor in relation to the IPO, but state that this theory holds also for underwriting investment banks. Their main finding is that firms with favorable information about its value will opt for a higher-quality underwriter.
4.2.3 Signaling theory
The signaling theory is based on Akerlof's lemons problem (Akerlof, 1970). This theory assumes that the issuer has more information than the investor, which causes information asymmetry. The investor is not able to distinguish the good from the bad quality firms, so he is reluctant to buy shares. However, the good quality firms can signal their quality by pricing their shares lower than the market would expect them to be. This does mean that the issuing firm has lower initial income. The reason for doing this is because the firm expects to regain the lost income in the future: The firm makes a sacrifice, so when information becomes more widely available, the firm can benefit from new share issuance. Bad quality firms will not follow this strategy, because they know that they cannot recoup this loss. Allen and
Faulhaber (1989) assume in their model that the investment bank as an underwriter does not play an active role in the IPO. The signaling theory relies on future share issuance or so-called 'seasoned equity offerings' (SEO). Welch (1996) treats the timing of the SEO as endogenous. In his model, information becomes more widely available after the IPO and he argues that time is positively correlated to the amount of information. This means that higher quality firms are willing to wait longer before they issue SEO, because more information about their quality will be revealed, leading to a higher share price. Lower quality firms are reluctant to wait long, because information about their quality will not be favorable.
Another application of the signaling theory is provided by Myers and Majluf (1984), They also assume that management has more information about the firm than shareholders. In their model, a firm has one asset and one investment opportunity, but does not have the means to make the investment. In order to gain enough cash, they can issue new equity. Myers and Majluf (1994) assume that management acts in the interest of old shareholders. That is, they want to maximize the intrinsic value of the existing shares, not the new issues. There are assumed to be two states of nature. If management decides to invest in the opportunity in state 1, the older shareholders are worse off. However, the worth of the firm is highest in state 1. If management decides to invest while in state 2, the old shareholders will benefit. Because the firm acts in the interest of the old shareholders, they will decide only to invest when state 2 occurs. But because the worth of the firm is higher in state 1, this will signal that management expects state 2 to occur. This will cause the share price to drop in order to reflect the value of the firm in state 2. The firm will thus decide not to issue new shares. In their theory, they rank equity issuance as a less preferred way to gain capital.
However, if it happens that in both states, the old shareholders are better off, the firm will decide to invest in the opportunity. Although Majluf and Myers (1994) do not specifically focus on IPOs, the theory can be extended to explain the underpricing of IPOs. If a firm decides go public and offer shares, they signal that this is beneficial to the 'old' shareholders (i.e. the owners and other private shareholders). Otherwise, they would not have done it, keeping in mind that they will act in the best interest of the old shareholders. The old
shareholders may be allocated shares prior to the IPO, which means they would benefit if the price would go up after the IPO. This would mean that old shareholders benefit from
underpricing and have an incentive to set the offer price too low. Thus, they do not act in the interest of the firm, but rather in their own interest.
There is mixed support for the signaling theory in the literature. Welch (1996) finds empirical evidence for his model: High quality firms wait longer to issue SEOs than lower quality firms. However, Kerins, Kutsuna and Smith (2007) examine Japanese IPOs and find no support for the signaling theory. The signaling hypothesis states that good quality issuers expect prices to rise, so that they can issue more shares in the future, for a higher price than the IPO shares. However, Kerins, Kutsuna and Smith (2007) find that only firms that
experience unexpected high aftermarket returns are issuing more shares. This contradicts the signaling theory, in which firms expect their shares to have a high aftermarket return.
4.3 Theories on overpricing
Although the majority of the studies show a consistent underpricing of IPOs, a few studies contradict this view. Overpricing of shares has particularly negative effects on the
underwriting investment bank (Paige Fields & Fraiser, 2003). By overpricing the shares, a bank might be willing to gain some favor from the issuing firm. However, it is possible that the underwriting bank gets stuck with these shares in their inventory, because no investor is willing to buy the shares. Nanda and Yun (1993) find that overpricing, contrary to
underpricing, will lead to a significant loss in the stock market of the lead underwriter. They also find that moderate underpricing has a positive effect on the wealth of the underwriting bank. This finding implicates that banks are reluctant to overprice shares and explains why underpricing is more common than underpricing.
Purnanandam and Swaminathan (2004) distinguish overvalued IPOs from undervalued IPOs. Overvalued IPOs tend to have low current profitability, but higher
profitability, but lower anticipated earnings growth. Purnanandam and Swaminathan (2004) matched IPOs to comparable non-IPO industry peer firm. They calculated the price-to-value ratio for each IPO firm and found that IPOs are systematically overvalued at the offer price. Their study contradicts the prediction of asymmetric information theories about IPO pricing, which say that IPOs that are most undervalued should have the highest first-day returns. Their main finding is that that overvalued IPOs underperform relatively to undervalued IPOs in the five years after the IPO. They also find that IPO firms underperform their matched non-IPO firms in the long run performance, which supports the view that non-IPOs are in general overpriced.
Zheng (2007) further examined Purnandandam and Swaminathan's study and found contradicting results. He suggests that the overvalued firms are not actually overvalued. As is said, Purnanandam and Swaminathan (2004) paired IPO firms to industry peers. Zheng (2007) argues that their model has three areas in which biases are possible. First, Purnanandam and Swaminathan (2004) use the price-to-value ratio to calculate the IPO valuation. They assume that pre-IPO accounting variables can be used to value the IPO. This indicates that there are no new primary shares. However, if there are new primary shares, it is possible that the issuing firm decides to issue more shares in their IPO. If the price of the shares stays the same, the shares get diluted, because the volume of the shares increases, while the firm's value stays the same. Zheng (2007) argues that Purnanandam and Swaminathan (2004) should leave the new primary shares out. Second, the paired firms may have different cash holdings, which affect the market values of the firms. If the firm does not use the cash, it does not affect sales or earnings. This means that firms with high cash holdings get higher price-to-sales ratios, which does not directly indicate overvaluation. Finally, Zheng (2007) argues that Purnanandam and Swaminathan (2004) do not take the leverage in account, the amount of debt may be very different among firms. They base their enterprise valuation only on the equity of a firm, while Zheng (2007) argues that they should also take into account the value of debt. Overall, Zheng (2007) argues that the pairs of IPO firms and non-IPO firms are not comparable. His study also shows that IPO firms do not underperform non-IPO firms, given the changes he made, which indicates that IPOs are not overpriced.
5 Earnings management
5.1 Definition
Scott (2009) defines earnings management as 'the choice by a manager of accounting policies, or actions affecting earnings, so as to achieve some specific reported earnings objective.' Healy and Wahlen (1999) have a more specific and elaborated description: 'Earnings management occurs when management uses judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes that depend on reported accounting numbers.'
Scott (2009) distinguishes between earnings management through accounting policy choice or earnings management through real actions. The accounting policy choice focuses mostly on discretionary accruals. Total accruals are the difference between net income and cash flow from operations. Earnings management based on accruals face an 'iron law': All accruals used to manage earnings upwards will result in decreasing earnings in the future. Scott recognizes four patterns in earnings management. The first is taking a bath. This form of earnings management usually happens around reorganizations. The firm is already reporting a loss and decides to enlarge this loss by, for example, making provision for future losses. The iron law suggests that this will enhance income in the future. The second one is income mineralization, which is not as extreme as taking a bath and usually happens when a firm is facing high profitability and tries to temper down these earnings. The third pattern is income maximization; this is common when managers try to earn bonuses. The fourth and final pattern is income smoothing. This leads to more constant earnings. Scott also mentions that IPO firms with high discretionary accruals have a lower long-run return compared to firms with low accruals. This could be explained by the iron law, which suggests that
increasing earnings by accruals will lead to lower earnings in the future. More of that will be explained in the sixth section.
5.2 Measurement and detection
Almost all studies estimating earnings management base their models on the accruals approach. Accruals are defined as the difference between reported accounting earnings and cash flow from operations (Healy, 1985). These accruals are then divided between
discretionary and non-discretionary accruals. Management has the ability to influence discretionary accruals. Healy (1985) assumes that discretionary accruals will add up to zero over the period of employment of a manager.
DeAngelo (1986) examined management buyouts of public stockholders. In a buyout with high leverage, management has an incentive to decrease the stock price. This makes a high leveraged firm cheaper and more attractive to take-over firms. Public buyouts result in private firms, which no longer have to disclose any information about their performance. This makes the risk of discovery and punishment of earnings management prior to a buyout unlikely. Public stockholders suffer from the lower stock price. An investment bank is often hired to independently set a price for the buyout. In determining the offer price, investment banks usually focus on earnings, which give management an incentive to understate these earnings. DeAngelo (1986) estimates the level of earnings management based on accruals.
One of the most cited studies in the field of earnings management is the study of Teoh, Welch and Wong (1998). They examined the effect of earnings management on the long-run market performance of initial public offerings. Net income is the total of cash flow from operations and accruals. Teoh, Welch and Wong (1998) divided these accruals in long-term and current components. Short-long-term accruals are, for example, adjustments regarding current assets or liabilities that are used for operational purposes. Speeding up revenue recognition or delaying expenses lead to an increase in current accruals. Long-term accruals involve non-current assets and can be adjusted by changing depreciation policies.
Discretionary accruals are subject to management's choices and can be manipulated, while non-discretionary accruals are not manipulated. Teoh, Welch and Wong (1998) focus on accruals that are different from industry peers' accruals. Thus, they suppose that industry peers only have discretionary accruals. That way they can differentiate between non-discretionary and non-discretionary (manipulated) earnings.
As is mentioned before, earnings management is often hard to detect. An interesting approach in detecting and measuring earnings management is provided by Lo (2008). Since earnings management is often regarded as fraudulent and undesirable for shareholders, management will try to hide earnings management as well as they can. Lo (2008) therefore suggests that investigation to earnings management should be similar to an investigation of a
crime. He suggests that investigators should focus on the seven elements of a 'crime': whether earnings management happened, who are the suspects, which 'weapons' are used (e.g. accruals), who are the victims, what are the motives, was there an opportunity to
commit earnings management and if there are any other explanations which would clarify the use of accruals.
Lo (2008) treats earnings management as a crime and due to accounting scandals in the past, earnings management has a bad name. However, earnings management is not necessarily bad for a firm and its stockholders. It can actually be beneficial. First,
opportunistic earnings management will be examined followed by beneficial earnings management.
5.3 Opportunistic earnings management
If a firm's manager acts in his own interest by manipulating earnings, it is called opportunistic earnings management. Opportunistic earnings management leads to a greater information asymmetry between management and stockholders and might possibly lead to a decrease in firm value. This kind of earnings management arises when management believes that
stockholders do not have enough information to detect earnings management (Healy & Wahlen, 1999). Scott (2009) mentions that opportunistic earnings management is likely to arise when management issues new shares, for example by earlier revenue recognition. Current investors benefit from this earnings management, while new investors are paying too much for the shares. According to Scott (2009), improving disclosure quality can reduce bad earnings management.
Beneish and Vargus (2002) argue that earnings management through accruals is hard to detect. Auditors and regulators do not have full insight in the valuation of accruals, because the relation between accruals and future earnings is often too complex for outsiders to understand. They focus particularly on the relation between income-increasing accruals and insider trader information. Their study reveals that the market often overprices income-increasing accruals. This can be attributed to either changes in economic conditions or management manipulating earnings. Because accruals are always made in advance, it is hard to distinguish between these two. However, Beneish and Vargus (2002) show that income-increasing accruals are at least partially attributable to earnings management. They find that accruals are often unfairly rated as high quality by investors. Beneish and Vargus (2002) show that these income-enhancing accruals are positively related to abnormally high
insider selling, which gives at least the appearance of opportunistic earnings management through accruals.
Discretionary accruals are, in general, likely to happen more often prior to issuance of new equity (Marquardt & Wiedman, 2004). However, firms that provide an earnings forecast voluntarily prior to an issuance of new equity do not show any evidence of earnings
management in the months prior to this equity offering. Marquardt and Wiedman (2004) also examined the impact of opportunistic earnings management on the value relevance of financial reports. They measure value relevance by calculating the explanatory power of accounting information for prices and returns. They define value relevance as the relation between accounting information and the market value of a firm. In their study, Marquardt and Wiedman (2004) find evidence that value relevance declines when discretionary accruals are high. This is consistent with the view that managements' incentives to manage earnings have a negative impact on the informativeness of earnings announcement to investors. Thus, indicating opportunistic earnings management.
5.4 Beneficial earnings management
As is mentioned before, information is often asymmetric and hard to obtain for investors. It is also difficult for firms to signal their quality. Earnings management can help reducing the information asymmetry. Agents are well informed about a firm, but it can be costly to provide this information to the principal. Scott (2009) explains the reducing of information asymmetry by earnings management as follows. Managers with positive inside information about a firm's future performance are willing to signal this to the investors. However, it is not credible if management only announces the good prospects. If a firm, in one year, has higher than expected profit, it can decide to create a provision and reduce net income to the expected level. The market knows that management will not increase earnings in the short run, because according to the iron law, future income will decrease. That is known as income smoothing. Thus, by creating a provision, management can signal information about future performance to the market.
Earnings management can also be beneficial, because information about the performance of management will reach the owner with a delay. This delay will lead to decisions about management performance ex post and allows for managers to work things out (Arya, Glover & Sunder, 1998). The owner will fire the manager when it is apparent that he underperforms. If the owner will do this based on ex ante information, there is a chance
that management is fired too often and this could be costly and might result in a decrease of welfare for the owners. Earnings management will lead to a shift of costs and revenues in the short-run and will make performance measurement in the short run difficult. However, it can help owners to detect poor performance in the long run, and prevents them from firing management too soon.
In an environment where organizations are decentralized and information is not perfectly known by everyone, managed earnings can contain more information than
unmanaged earnings (Arya, Glover & Sunder, 2003). However, accruals can only be value-enhancing if they are meet certain criteria. Arya et al. (2003) discuss three accounting properties that help increase the value of accruals. First, accruals have to be verifiable. Second, accounting policy choices and the use of accruals commits management to certain obligations. Finally, the iron law ensures that accounting earnings are correct over the life of a firm. For example, increasing inventory in one year will lead to an increase in cost of inventory in a subsequent year.
6 Earnings management in IPOs
6.1 Incentives
The lack of information about a firm that decides to take its shares publicly results in
information asymmetry between investors and management. Firms are required to provide a prospectus including three years of financial statement. Since this is often the only
information about a firm, investors rely heavily on this prospectus (Chaney & Lewis, 1994). The high dependence on the prospectus provides an incentive for management to change the accounting policies. Additionally, IPO firms in the United States are allowed to change their accounting choices and policies prior to the offering, which makes the information asymmetry even bigger (Teoh, Welch & Wong, 1994). Chaney and Lewis (1994) also note that underwriters often base their estimation of the offer price on the price/earnings ratio.
The lack of transparency and the ease with which accounting policies can be
changed, create incentives for a firm to manage earnings, both prior to an IPO and after the issuance. Management usually owns shares in their own company and is often required to keep their shares for a predetermined period of time, the lock-up period. This creates an incentive to keep the share price high after the IPO and to manage earnings upwards. Underwriters may also push management to use an aggressive reporting policy (Chaney & Lewis, 1998). In order to maintain their good reputation, they could want IPO firms to at least meet their earnings forecast. During the road show, firms often hint projections of earnings to potential investors. In order to avoid lawsuits from misled shareholders and to maintain goodwill of these shareholders, firms may feel a lot of pressure to keep the share price up in the months after the IPO (Teoh, Wong & Rao, 1998).
6.2 Measurement
Sample selection to measure earnings management in general is difficult. Firms are reluctant to share sensitive information about their ways to manipulate earnings. Chaney and Lewis (1998) focused on earnings management post-IPO. All the firms they examined had information about their cash flows and earnings available on COMPUSTAT for five years subsequent to the IPO date. Measuring earnings management prior to an IPO is harder.
Firms are not required to issue an annual report containing financial statements. Therefore, the only information available is in the prospectus. Teoh, Welch and Wong (1998) examined earnings management pre-IPO. The firms they included in their sample had to had to have information available in COMPUSTAT for the year of the IPO and the year prior to an IPO. 1,649 of the initial sample of 1,974 had data available in COMPUSTAT.
As is mentioned before, the most common method in measuring earnings
management is through the accruals approach. Total accruals are the difference between cash flow from operations and reported earnings. IPOs, however, are in many ways not comparable to 'normal' public firms. For example, they experience a much higher growth rate than average other firms. It is hard to estimate a benchmark in determining the level of earnings management. IPO firms are often experiencing high growth, which makes them hard to compare to other firms. Teoh, Wong and Rao (1998) matched each IPO firm to a non-IPO firm in the same industry. They obtained close matches. The mean sales growth of IPO firms is 116 percent, while the matched samples had a mean sales growth rate of 115 percent. They used a model of Jones (1991) in estimating total accruals and abnormal accruals. They assumed that accruals are dependent of long-term assets (property, plant and equipment) and the changes in revenues. These factors determined the amount of accruals in response to a firm's economic condition. To estimate the expected level of accruals, they filtered out the increases in trades receivables, because this balance post is prone to earnings management. They then subtracted the expected level of accruals from total accruals in order to calculate the level of abnormal accruals. They did this for both current and long-term accruals. The abnormal current and long-term accruals are treated as a proxy for earnings management, while expected accruals are a proxy for the normal accruals given the changes in economic conditions. Ducharme, Malatesta and Sefcik (2001) also determine managed earnings by subtracting unmanaged earnings from total earnings. They also note that accruals depend on the economic conditions of a firm and are not always opportunistic. To filter out these non-opportunistic accruals, they match each IPO firm to at least 10 non-IPO industry-matched firms and perform regressions. These regressions are assumed to be the benchmark and all deviating accruals are assumed to be discretionary.
Friedlan (1994) states that DeAngelo's model for estimating discretionary accruals is not applicable for IPO firms. As is mentioned before, DeAngelo (1986) assumes that
changes in non-discretionary accruals are random and add up to zero. Thus, the absolute change in total accruals is assumed to be discretionary. This is not valid for IPO firms, because they experience a high growth rate. In his sample of 155 firms, 143 firms were experiencing growth. If a firm doubles its sales, total accruals are likely to double as well. Because DeAngelo (1986) uses absolute numbers, her model does not account for this
growth and will overestimate the level of discretionary accruals. Friedlan (1994) adjusts DeAngelo's model by dividing total accruals by sales for both the benchmark period and the test period. This makes accruals proportional, which allows the accrual approach to be applicable for IPO firms. Friedlan (1994) then tests the hypothesis that discretionary accruals are zero. That is, he examines if the total accruals have grown proportionally relative to the firm growth. If total accruals have grown significantly, he assumes that this is due to
discretionary accruals.
6.3 Evidence
6.3.1 Income maximization
Friedlan (1994) applied his adjusted accruals model in his study to accounting choices of issuers of IPOs. He took a sample of 155 firms that went public from 12 different industries. He took the year previous to the IPO as a benchmark. Friedlan (1994) found that accruals in the year of the IPO were significantly higher than accruals in the benchmark period, thus supporting the hypothesis that firms do manage earnings upward prior to an IPO. Some firms also provided interim statements that were more current than the annual statements.
Friedlan (1994) found that increases in earnings were not accompanied by increases in cash flows. He also finds that firms that have positive cash flow do no show any presence of managed earnings. Contrary, firms that experienced negative cash flows do have
significantly high accruals. This result suggests that better-performing firms do not manage earnings upward, while firms that perform poorly do. Teoh, Welch and Wong (1994) also use discretionary accruals as their main proxy for the presence of earnings management.
However, they found that discretionary current accruals in IPO firms were unusually high in comparison to industry peers.
Contrary to Teoh, Welch and Wong's study, Ball and Shivakumar (2008) found evidence that earnings and reporting quality is actually higher prior to an IPO. They argue that this is because the market demands more transparency from public firms than they do from private firms. Public firms face higher regulatory standards and might face regulatory actions in case of insufficient reporting quality. Aharony, Lin and Loeb (1993) studied
earnings management in industrial companies. Their approach was to estimate discretionary accruals by using total accruals. As is mentioned before, discretionary accruals are accruals that can be manipulated by management. Non-discretionary accruals were assumed to be stationary. They defined total accruals as net income minus operating cash flows. They only
found weak evidence for earnings management. The accruals were not significantly higher in the years prior to an IPO.
Lo (2008) refutes Ball and Shivakumar's finding that IPO firms do not manage earnings, by stating that their studies are not comparable. They only investigated UK firms. These firms are required to provide two sets of financial statements. The UK requires not only public firms, but also private firms to file statements. Ball and Shivakumar (2008)
compared these statements to the statements IPO firms are including in their prospectus and found no evidence of differences in accruals. Firms that participate in an IPO in the United States are not required to file statements prior to an IPO, which makes the study of Ball and Shivakumar and the study of Teoh, Welch and Wong incomparable (Lo, 2008).
6.3.2 Income smoothing
As is mentioned before, income smoothing can be used as an instrument to communicate firms' permanent earnings. Chaney and Lewis (1998) define income smoothing as the case where, in the long run, variability in earnings is less severe than variability in cash flow. They find that firms that have a high market performance also experience a high variability of cash flows relatively to earnings. That is, firms that seem to have constant earnings, but variable cash flows from operations, seem to perform better. Chaney and Lewis (1998) argue that firms with higher future earnings forecasts are able to smooth income better than aggressive firms. Firms that participate in income increasing, instead of income smoothing, face the repercussion of their high amount of discretionary accruals in future years. Chaney and Lewis (1998) find evidence that well-performing IPO firms use income smoothing as a signal in order to distinguish themselves from poor-performing firms. This indicates that earnings management is not necessarily bad for the firm.
6.4 Effects of earnings management in IPOs
6.4.1 Short-term effects
The general belief among managers is that external users are not able to fully understand the fundamental value of a firm and thus have an incentive to participate in earnings management. Ducharme, Malatesta and Sefcik (2001) found that earnings management prior to an IPO leads to an increase in the initial firm value. They examined what the relation is between earnings management prior to an IPO and the firm value at offering. Ducharme et al. (2001) found that the initial firm value is higher if a firm participated in earnings
management prior to an IPO. That is, they found that an increase in income due to earnings management by accruals leads to an increase of the firm valuation. In a perfect market, investors should take discount all accruals when estimating the fundamental of a firm. Thus this indicates that investors do not account properly for accruals.
As is mentioned before, management has an incentive to overstate earnings in order to receive more cash. However, IPOs are in general heavily underpriced. The offer price is much lower than the trading price at the end of the day. The effect of earnings management in IPOs on the offer price may not be reflected, more about this will follow in the discussion section.
6.4.2 Long-term effects
The underpricing of IPOs suggests that investors are overoptimistic about earnings potential of IPOs, while IPO underperform in the long run. Ritter (1991) provides three different explanations for long-run underperformance: risk mismeasurement, bad luck or
overoptimism. Overoptimism finds most support in the literature regarding the long-run performance of IPOs.
The greater the use of accruals prior to an IPO, the greater the price correction will be in future years (Teoh, Welch & Wong, 2002). As is mentioned before, management has an incentive to keep up the share price in the lock-up period right after the IPO. Teoh, Welch and Wong (2002) took in account this period and examined the level of earnings
management and the performance of IPO firms in subsequent years. They found a strong negative relation between earnings management and stock returns. Teoh, Welch and Wong (2002) differentiated between aggressive and conservative firms. Aggressive firms are in the highest quartile based on IPO-year accruals, while conservative firms are in the lowest quartile. They found that aggressive firms earn 20 to 30 percent less return than conservative firms. Teoh, Wong and Rao (1998) extend this view. They find that opportunistic accruals around an IPO mislead investors and leads to over optimism about future performance. After the IPO, investors realize that the accruals do not reflect the fundamental value of the firm, which causes the share price to drop.
The long-term underperformance of IPO shares can also be explained from an analyst point of view. Financial analysts use financial statements to form an opinion about the future performance of a firm. However, as is stated before, these statements may contain accruals. Although analysts should be able to see through these overoptimistic statements and to account for the accruals, the opposite seems to be true. Teoh and Wong (2002) show that firms with high accruals are valuated overoptimistically. Investors rely on analysts' reports and will drop their shares when the firm does not live up to their expectations, leading
to a decreasing share price. Ducharme, Malatesta and Sefcik (2001) call the
underperformance-phenomenon the 'disappointment hypothesis': Firms that have high pre-IPO accruals will have disappointing returns in subsequent years.
6.5 Factors influencing earnings management in IPOs
6.5.1 Third party involvement
The reputation of third-party certifiers seems to play an important role in the level of earnings management in an IPO. Third-party certifiers are auditors, underwriters, attorneys and venture capitalists. Prestigious underwriters have an incentive to back only high-quality issuing firms. Underwriters do not only face the risk of legal penalties, in addition it has been shown that the share price will also drop due to reputation loss (Karpoff, Lee & Martin, 2008). Furthermore, high quality auditors can also mitigate earnings management. Auditors are highly depended on their quality in order to remain credible. It has been shown that attorneys also decrease asymmetric information and lead to a lower level of underpricing (Beatty & Welch, 1996). Finally, venture capital financing can serve as a certification of quality (Brau & Johnson, 2009).
These parties can help reduce earnings management in two different ways. First, the third party can act as a watchdog. The certifier will go through the books of the issuing firm in order to prepare the prospectus. If this firm turns out to be involved in aggressive accrual management, the third party will steer them to a more conservative accrual approach, in order to maintain their good reputation. Second, firms signal their quality by selecting
prestigious certifiers. If a firm does not engage in earnings management, they will show their good quality by hiring certifiers with a good reputation (Brau & Jonhson, 2009). Chang, Chung and Lin (2010) find a significant negative relation between the underwriters' reputation and abnormal discretionary accruals, indicating that the reputation of the underwriter reduces the amount of earnings management. They suggest that firms that are willing to participate in earnings management are more likely to hire a lower quality underwriter, because these underwriters are less likely to monitor the issuing firm thorough.
6.5.2 Regulations
Kao, Wu and Yang (2009) also examined the effect of regulations on earnings management in Chinese IPOs. Chinese regulations on IPO pricing are pricing regulations and penalty regulations. Pricing regulations state that the IPO price should be based on the earnings per
share (EPS). The key determinant for EPS is the accounting performance, which means that firms may have an incentive to overstate earnings. Kao, Wu and Yang (2009) found evidence in favor of this theory. Penalty regulations punish firms that fall more than 10% short on their realized earnings compared to the management earnings forecasts, which may reduce earnings management in IPOs. Kao, Wu and Yang (2009) found that the punishment deters firms from engaging in earnings management.
Disclosure of earnings forecasts may also affect the level of earnings management in an IPO. In general, issuing firms only provide underwriters with information in order to market the shares. Jog and McConomy (2003) examined Canadian firms that voluntarily disclose information about their forecasted earnings. Their study shows that firms that do provide this information suffer less from underpricing and declining future earnings. This result may suggest that earnings management can be reduced by disclosure of earnings forecasts. Friedlan (1994) distinguished between firms that provide interim statements and firms that only provided annual statements. He found that firms with only annual statements experience a higher level of income increasing than firms that do provide interim statements. This result also suggests that disclosure of information, other than the annual statements, has a
7 Discussion
This section will try to answer the research question: How does the pricing and subsequent performance of Initial Public Offerings relate to earnings management? As is mentioned before, IPOs take place in an environment with great information asymmetry. Investors do not know exactly how to value an IPO, because there is barely information available. They rely heavily on prospectus, which is usually written by underwriters, attorneys or auditors and contains financial statements of the past three years. Management therefore has an
incentive to manage earnings in order to make their firm attractive to invest in. This happens most often through accrual management: Costs will be delayed and income is recognized earlier. Earnings management can be beneficial as well as opportunistic. Chaney and Lewis (1998) found that firms that smooth income through accruals perform better than firms that increase income aggressively through accruals. Although it has been shown by several studies that firms manage earnings upward (Ducharme et al; Friedlan; Teoh et al.), the offer price seems to be too low and the returns in years after the offering are disappointing. This section will examine the relation between earnings management and the pricing of IPOs and subsequent performance.
The underpricing of IPOs contradicts the fact that management engages in earnings management in order to increase share prices. Theories on underpricing can be divided in two types: Underwriter-based theories and signaling theories. Both have different
implications for the existence of earnings management in IPOs. Theories on underpricing based on underwriters are twofold. On one hand, they suggest that underpricing exists because of reputation loss in case of overpricing. It has been shown that the market will punish the underwriter with a severity of twelve times the cost of the legal issue (Karpoff, Lee & Martin, 2008). On the other hand, they say that underpricing exists because underwriters may favor certain investors and offer them a lower price. The theory is in line with the finding that there is a negative correlation between earnings management and the reputation of the underwriter. It has also been shown that a higher reputation leads to less sever underpricing in IPOs. By hiring a reputable underwriter, the firm signals its quality to the market.
Underpricing is shown to be more severe in case of information asymmetry (Boulton, Smart & Zutter, 2011). The choice of hiring a good quality underwriter reduces some of this
asymmetry about the quality of the firm, thus reducing underpricing. These findings are consistent with the theory on earnings management. A high reputation is negatively correlated to earnings management and thus underpricing is less severe.
As is mentioned before, the book-building process might be factor in explaining the underpricing of IPOs. However, I did not find any literature that examined the relation between book-building and earnings management. Book-building could both increase and reduce information asymmetry between investors and firms. Underwriters might push firms to show their best results, but are also reluctant to overprice the shares in order to maintain their good reputation. However, they do want the IPO firms to meet their earnings forecast to keep investors happy. The implications for earnings management in the book-building
process are mixed and ask for further research.
Another theory discussed about underpricing was the signaling theory. This theory states that good quality firms underprice their shares to signal information about their quality to potential investors. Because they are able to recoup this loss, they distinguish themselves from bad-quality firms. The good quality firms will issue new shares in the subsequent period when information about their quality is further revealed. However, Chaney and Lewis (1998) found that higher quality firms tend to smooth their income rather than increasing it in an IPO. However, their study primarily discusses earnings management after the offer date. They show that firms that engage in income smoothing are performing better in the long run. To examine the relation between income smoothing and the pricing of IPOs, further research is required. The signaling theory of Myers and Majluf (1984) does not focus on IPOs, but can be extended to fit the IPO market. They assume that firms will try to benefit old shareholders. If these shareholders are sold shares at the offering date, they are better off when the prices go up on the first trading day. This might indicate that old investors benefit from lower offer prices and suggests overpricing of shares.
Earnings management in the long run might explain the underperformance of IPOs. As is mentioned before, the iron law states that increase of income results in decreasing income in the long run. Firms that decrease the information asymmetry are usually of good quality and do not have anything to hide. In fact, they want to signal their quality to potential investors. It has been shown that firms that provide interim documents and voluntarily disclosure do not suffer from long run underperformance.
The pricing of IPOs is heavily dependent on information asymmetry prior to an IPO. Earning management can both reduce and increase information asymmetry. Income
maximization leads to a higher level of underpricing, while firms that try to signal their quality through income smoothing experience a lower level of underpricing. This also holds for subsequent performance: Firms that maximize income through earnings management underperform in subsequent years, while firms that smooth income seem to perform better.
8 Conclusion
Underpricing of IPOs is highly depended on the level of information asymmetry in firms, which can be both reduced and decreased by earnings management. The higher the level of abnormal accruals, the lower the quality of the accounting information. This leads to high uncertainty among investors and thus to underpricing. However, firms that smooth income try to signal their permanent income to potential investors and therefore reduce information asymmetry. This might result in less severe underpricing among good quality firms. This thesis examined several underpricing theories, divided in theories based on underwriters and theories based on signaling. Theories regarding underwriters found that there are several factors contributing to the underpricing of IPO shares. First of all, the choice between a public auction and allocation through the book-building process makes a difference. Second, the level of underpricing is dependent on whether the underwriter markets the shares through firm-commitment or best effort. Third, several studies show that the reputation of the
underwriter influences the pricing of IPOs. More specifically, Nanda and Yun (1996) find that the market will punish the underwriter if shares are overpriced and will reward the underwriter if shares are slightly underpriced. Good quality underwriters therefore have an incentive to underprice IPO shares. Underpricing can also be explained through the signaling theory: Good quality firms will underprice their shares, because they know that they will recoup this gain in a subsequent offer when information becomes more widely available. Bad quality firms will not follow this strategy, because they cannot regain the loss.
Due to the lack of information, management has an incentive to manage earnings. Earnings management can be beneficial or opportunistic. It has been shown that
underpricing is more common when information asymmetry is bigger (Boulton, Smart & Zutter, 2011). Earnings management can reduce this information asymmetry if the firm uses it to smoothing income. That way, the firm signals its quality to potential investors. It can also be used to maximize income and make the firm appear of higher quality. In that case,
earnings management will increase the information asymmetry. It has been shown that firms that participate in opportunistic earnings management around IPOs are more likely to
9 References
Aharony, J., Lin, C., & Loeb, M. P. (1993). Initial public offerings, accounting choices, and earnings management. Contemporary Accounting Research, 10(1), 61-81.
Akerlof, G. A. (1970). Market for "lemons": Quality uncertainty and the market mechanism.
The Quarterly Journal of Economics, 84(3), 488-500.
Allen, F., & Faulhaber, G. R. (1989). Signaling by underpricing in the IPO market. Journal of
Financial Economics, 23, 303-323.
Arya, A., Glover, J., & Sunder, S. (1998). Earnings management and the revelation principle.
Review of Accounting Studies, 3(1-2)
Ball, R., & Shivakumar, L. (2008). Earnings quality at initial public offerings. Journal of
Accounting and Economics, 45(2-3), 324-349.
Beatty, R. P., & Ritter, J. R. (1986). Investment banking, reputation and the underpricing of initial public offerings. Journal of Financial Economics, 15(1-2), 213-232.
Beneish, M. D., & Vargus, M. E. (2002). Insider trading, earnings quality, and accrual mispricing. The Accounting Review, 77(4), 755-791.
Benveniste, L. M., & Spindt, P. A. (1989). How investment bankers determine the offer price and allocation of new issues. Journal of Financial Economics, 24, 343-361.
Boulton, T. J., Smart, S. B., & Zutter, C. J. (2011). Earnings quality and international IPO underpricing. The Accounting Review, 86(2), 483-505.
Brau, J. C., & Johnson, P. M. (2009). Earnings management in IPOs: Post-engagement third-party mitigation or issuer signaling? Advances in Accounting, 25(2), 125-135. Brau, J. C., & Fawcett, S. E. (2006). Initial public offerings: An analysis of theory and
practice. The Journal of Finance, 61(1), 399-436.
Caramanis, C., & Lennox, C. (2008). Audit effort and earnings management. Journal of
Accounting and Economics, 45(1), 116-138.
Carter, R., & Manaster, S. (1990). Initial public offerings and underwriter reputation. The
Journal of Finance, 45(4), 1045-1067.
Chaney, P. K., & Lewis Craig M. (1998). Income smoothing and underperformance in intial public offerings. Journal of Corporate Finance, 4(1), 1-29.
DeAngelo, L. (1986). Accounting numbers as market valuation substitutes: A study of management buyouts of public stockholders. The Accounting Review, 61(July), 400-420.
Derrien, F., & Womack, K. (2003). Auctions vs. bookbuilding and the control of underpricing in hot IPO markets. Review of Financial Studies, 16(1), 31-61.
DuCharme, L. L., Malatesta, P. H., & Sefcik, S. E. (2001). Earnings management: IPO valuation and subsequent performacne. Journal of Accounting, Auditing & Finance,
16(4), 369-396.
Friedlan, J. M. (1994). Accounting choices of issuers of initial public offerings. Contemporary
Accounting Research, 11(1), 1-31.
Healy, P. M., & Wahlen, J. M. (1999). A review of the earnings management literature and its implications for standard setting. Accounting Horizons, 13(4), 365-383.
Jenkinson, T., & Jones, H. (2009). IPO pricing and allocation: A survey of the views of institutional investors. Review of Financial Studies, 22(4), 1477-1504.
Jog, V., & McConomy, B. J. (2003). Voluntary disclosure of management earnings forecasts in IPO prospectuses. Journal of Business Finance & Accounting, 30(1-2), 125-168. Jones, J. (1991). Earnings Management during Import Relief Investigation. Journal of
Accounting Research, 29(2), 193–228.
Kao, J. L., Wu, D., & Yang, Z. (2009). Regulations, earnings management and post-IPO performance: The chinese evidence. Journal of Banking & Finance, 33(1), 63-76. Kerins, F., Kutsuna, K., & Smith, R. (2007). Why are IPOs underpriced? Evidence from
Japan's hybrid auction-method offerings. Journal of Financial Economics, 85, 637-666. Lo, K. (2008). Earnings management and earnings quality. Journal of Accounting and
Economics, 45(2-3), 350-357.
Myers, S. C., & Majluf, N., S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics,
13(2), 187-221.
Nanda, V., & Yun, Y. (1997). Reputation and financial intermediation: An empirical
investigation of the impact of IPO mispricing on underwriter market value. Journal of
Financial Intermediation, 6, 39-63.
Paige Fields, L., & Fraser, D. R. (2004). Effects of IPO mispricing on the risk and reputational capital of commercial banks. Review of Financial Economics, 13, 65-77.
Purnanandam, A. K., & Swaminathan, B. (2004). Are IPOs really underpriced? Review of
Financial Studies, 17(3), 811-848.
Ritter, J. R., & Welch, I. (2002). A review of IPO activity, pricing and allocations. The Journal
of Finance, 57(4), 1795-1828.
Scott, W. R. (2009). Earnings management. In S. Scully (Ed.), Financial accounting theory (5th ed., pp. 402-427). Toronto: Pearson Prentice Hall.
Sherman, A. (2005). Global trends in IPO methods: Book building versus auctions with endogenous entry. Journal of Financial Economics, 78(3), 615-649.
Teoh, S. H., Welch, I., & Wong, T. J. (1998). Earnings management and the long-run market performance of initial public offerings. The Journal of Finance, 53(6), 1935-1974.
Teoh, S. H., Wong, T. J., & Rao, G. R. (1998). Are accruals during initial public offerings opportunistic? Review of Accounting Studies, 3(1), 175-208.
Titman, S., & Trueman, B. (1986). Information quality and the valuation of new issues.
Journal of Accounting and Economics, 8(2), 159-172.
Welch, I. (1996). Equity offerings following the IPO theory and evidence. Journal of
Corporate Finance, 2(3), 227-259.
Zheng, S. X. (2007). Are IPOs really overpriced? Journal of Empirical Finance, 14(3), 287-309.
