Are IPOs Really Attractive for Everyone?

Are IPOs Really Attractive for Everyone?

Finding the main factors that decide the short-term

and long-term performance of IPOs

Feng Li

Faculty of Economics

University of Groningen

Under the supervision of

Mark Helmantel

ABSTRACT

This paper analyzes the main factors that influence the initial return and aftermarket performance of initial public offerings (IPOs). Hypotheses on proportion, underwriter’s reputation, divergence of opinion, leverage and liquidity are documented in this study. The main finding is that top-tier underwriters are not always associated with lower levels of initial return, which is maybe motivated in some environments by the desire to avoid lawsuits or protect reputations. It also finds that the problem of information asymmetry is still an important consideration when firms are choosing an IPO. When issue costs are lowered to compensate for information asymmetry, equity financing is not always the last resort. The decision to go public in the period under review (1998-2003) has an impact on a firm’s aftermarket capital structure and long-term performance.

TABLE OF CONTENTS

INTRODUCTION 4

SECTION 1: THEORETICAL BACKGROUND AND LITERATURE REVIEW 6

TIMING OF IPOs AND THE INVESTOR’S OPINION 6

UNDERWRITER’S ROLE ON INITIAL RETURN 9

ISSUER’S CAPITAL STRUCTURE AND LIQUIDITY 10

SECTION 2: METHODOLOGY 11

IDENTIFICATION OF HOT AND COLD ISSUE MARKETS 11

TESTING THE SHORT-TERM RETURN OF IPOS 12

LONG-TERM UNDERPERFORMANCE MODEL 14

SECTION 3: DATA AND SAMPLE STATISTICS 15

SECTION 4: EMPIRICAL RESULTS 17

IDENTIFICATION OF RELATIVELY HOT AND COLD MARKETS 17

TESTING RELATIONSHIP BETWEEN THREE FACTORS AND SHORT-TERM RETURN OF IPOS 18

EFFECTS OF ISSUER’S CAPITAL STRUCTURE AND LIQUIDITY ON LONG-TERM U N D E R P E R F O R M A N C E 2 0 SECTION 5: CONCLUSIONS AND RECOMMENDATIONS 21

REFERENCES 24

APPENDIX 1: FIGURES 27

Introduction

Short-term and long-term performance of initial public offerings (IPOs) has attracted considerable attention in recent years. Previous studies show that IPOs worldwide often yield positive initial returns and abnormally negative long-term returns relative to matched non-issue firms. It is well documented (Ritter, 1987) that the offer price in IPOs is underpriced by about 10-15% on average compared to their dealings in the aftermarket. During 1990-1998, IPO companies in the U.S. left nearly $27 billion on the table, more than twice the amount ($13 billion) paid to investment bankers. These same companies earned profits of approximately $8 billion in the year before issue, that is, they gave up three years of the aggregate profits left on the table on the first day of trading (Loughran and Ritter, 2002). In contrast to the huge initial returns, the poor performance of IPOs in subsequent periods presents us with more of a surprise. Loughran and Ritter (1995) demonstrate that from 1970 through 1990 IPO companies underperformed 5.6% each year relative to matched non-IPO firms in the five years following issuance. Although some investors earned significant long-term returns, underperformance on average is widely documented in the literature (Aggarwal and Rivoli, 1990; Jog and Riding, 1987; Ritter, 1991; Lee, Taylor and Walter, 1996).

The simple explanation for this strong contrast between the two periods is that the initially optimistic assessment of issuers’ prospects is negated subsequently by disappointingly poor cash flows. I consider this due mainly to information asymmetry between investors and issuers, as well as asymmetry between investors’ opinions and behaviour. Even now, there is still a great deal of controversy over this phenomenon and no one comprehensive theory can overpower any others. In this study, I aim to capture common factors from the perspectives of investors, underwriters and issuing firms. I examine whether these factors play roles in generating the large initial return and subsequent aftermarket underperformance of IPOs.

market is the aggregate number of new issues in the year; I define criterion-industry distribution to be linked to the type of markets. This is due to the representative period (1998-2003) I have chosen, it experienced internet boom and burst, producing distinct IPO volume of various industries in each year. The second novel aspect is that I use both dummy and score variables of underwriter’s reputation based on Carter and Manaster’s reputation rankings to examine underwriter’s role on IPO performance.

Findings based on the short-term return model indicate a negative relationship between the proportion of issues and initial return, firms would like to issue a higher proportion of shares at a lower level of underpricing. The model also finds that a higher level of uncertainty or divergence in opinion results in higher initial return as compensation for bearing the risks.

A striking finding in my study is that top-tier underwriters are not always associated with lower levels of underpricing, which is inconsistent with other literature. In some environments, such as during the internet boom, underwriters are unwilling to price offerings at high levels because they want to avoid lawsuits and protect their prestigious reputation if and when the bubble eventually bursts. The paper proposes an implication that the problem of information asymmetry is still an important consideration for firms choosing an IPO. When issue costs are lowered to compensate for information asymmetry, firms prefer to go public rather than employ other financing methods. It suggests that an equity issue is not always the last resort. In hot markets, more equity financing drives the issuer’s aftermarket capital structure and makes it persistent in subsequent periods.

SECTION1: Theoretical Background and Literature Review

The large initial return and subsequent underperformance of IPOs has spawned various theoretical and empirical literatures. The most popular explanation is based on asymmetric information between insiders and investors: initial returns are regarded as compensation to investors. It considers subsequent asymmetric information to be not as serious as in early IPO periods, since investors’ required returns also decrease with low asymmetric information costs. This explanation is consistent with the fact that the huge initial return of an IPO is often accompanied by a lower return afterwards. I also consider the information asymmetry problem to have an important impact on a firm’s performance. However, the efficiency of the capital market, underwriter’s role and a firm’s characteristics are also strong explanations for this phenomenon. In this section I will give my perspective and conjecture from the following three aspects.

Timing of IPOs and the Investor’s Opinion

With respect to the average 10-15% initial return in recent decades, giving rise to the term ‘underpricing’ mentioned in the literature, many studies often draw the conclusion that an IPO is undervalued relative to its first-day closing price. In this article, however, I posit that IPOs are not undervalued, but overvalued relative to their true value. I choose ‘true value’ here as the benchmark because the market price on the first day is not a ‘fair value’ considering the securities market in the short term is not efficient enough to reflect the stock’s intrinsic value in its price. Therefore, when stocks are overvalued in the short term, firms will note this and be happier to go public.

hot-market effect on the amount of issued equity capital in IPOs. He argues that IPO firms in a cold market raise capital by an average of about 60% of their pre-IPO asset value, while the same ratio for the hot market is 95%. That is, an IPO firm with past underperformance has more opportunity to successfully issue equity and raise more capital in a hot market than in the normal or cold market. CFOs also admit that timing plays an important role in their financing decisions. After comparing the market price with the intrinsic or appropriate value of firms (investment bankers estimate the true value with their superior information and observations of the market situation), managers conduct a flexible financing strategy, for example, issuing equity when stocks are overvalued by the market and repurchasing shares at a lower price to reduce future dividends.

Managers take evident advantage of this window of opportunity to raise money on favourable terms. However, on the other hand, why do investors cater to these managers’ intentional issues and purchase stocks at such high prices? In my opinion, this is due to market inefficiency in the short term and information asymmetry between insiders and outsiders. Draho (2001) reveals that public information acts as a coordinating device in the IPO process. This public signal enables investors to make a judgment on a firm’s value, but more importantly, it also affects the investors’ beliefs based on other investors’ beliefs, especially frenetic investors (in the stock market, investors are inclined to be influenced by other investors’ opinions ). Thus optimistic assessments by a few influential investors could trigger heavy buying behaviour in other ‘normal’ investors. Then again, underwriters often intentionally underprice the hot offerings relative to a maximum price, which stimulates investors to participate in IPOs and drives up the stock prices. The high stock price and initial returns make investors’ enthusiasm persist for a while.

purchasing at the high offer price. The situation in the cold market is quite the contrary. Because investors are reluctant to put more capital in stock market, issuers have to give up more money to attract investors through underpricing (even set the offer price lower than the true value), so they have to keep their equity issues to a necessary minimum to avoid high issue costs. Therefore, the lower (higher) issue proportion should be matched with the higher (lower) level of underpricing. This results in my first hypothesis:

Hypothesis 1: The proportion of issuing shares relates negatively to initial returns1

Loughran and Ritter (2002) document in their article that although on average the amount left on the table is $9.1 million, the median is only $2.3 million. This suggests that only a minority of firms left large amounts of money on the table. Philip, Stephen and Terry (1996) also observe a highly skewed distribution on underpricing which is consistent with Rock’s (1986) ‘winner’s curse’ model in the IPO market. They state that the degree of underpricing is directly correlated to uncertainty of the issuer’s value in the Australian securities market. In fact, they are not the first to propose this theory. Back in 1977, Miller originally proposed that divergence of opinion among investors is responsible for asset overvaluation and subsequent poor performance in markets with restricted short-selling2, such as in the IPO markets. With regard to the presence of short-sale constraints in IPO markets, the short-term price reflects only the optimistic investors’ opinion because the pessimists will wait until the short-sale

1

Issue cost relates to the true value and offer price of the IPO firms. in the hot market, offer price is more than true value which induces firms issue more proportion, but in the cold market, the firm’s true value is lower than offer and market price, the issue cost= (offer price –true value)*proportion. So firms have to issue a lower level of equity to save the issue costs.

2

restrictions are removed. This inflates the share price, especially when there is a huge demand in early IPO markets. Then, with the removal of only short-sale restrictions, both optimists and pessimists participate in trading and the share price can freely reflect its fundamental value producing poor performance in the long term.

Miller narrows the uncertainty to divergence3. Paraphrasing roughly, this means that as the uncertainty (divergence) increases, investors will expect more money to be left on the table via underpricing, which leads me to propose my second hypothesis:

Hypothesis 2: The level of underpricing has a positive correlation to the degree of uncertainty (divergence) about an issuer’s value

Underwriter’s Role on Initial Return

In the process of pricing equity offerings, the underwriter’s role should not be underestimated. The underwriter strategically designs a simultaneous underpricing and rationing situation in the IPO market. The underwriter has observed or estimated the demand at a maximum price in pre-market activities, and then intentionally makes the offer price at a discount given the observed demand. On the other hand, however, why do the issuers accept the underwriter’s proposal and give up the amount of money left on the table? Ritter (1998) argues that underpriced IPOs ‘leave a good taste’ with investors, aiming to sell out future shares at a more favourable price in Seasoned Equity Offerings (SEOs, follow-on offerings). Brennan and Franks (1997) suggest that when there is excess demand, issuers can ration the ownership strategically and ensure issuer’s control after going public.

As we know, underwriters can charge a great deal in fees for IPOs: the greater

3

their reputation, the more money they can charge. Issuers are willing to pay for the prestige of an underwriter because it is a guarantee of efficient work in pre-market activities, and it may also colour investors’ opinions about the quality of the issuing firm. Choosing a prestigious underwriter is an effective way to alleviate the information asymmetry problem. Moreover, empirical studies of different measurements of the underwriter’s reputation provide evidence that underwriter prestige is associated with a lower level of underpricing (Block and Stanley, 1980; Carter and Manaster, 1990; Franzke, 2003; McDonald and Miller, 1988). Issuers expect an underwriter with a good reputation to help them go public at a lower cost by means of effective underpricing. I test this relationship in my third hypothesis:

Hypothesis 3: In the normal market, the better the underwriter’s reputation, the less the initial return

Issuer’s Capital Structure and Liquidity

In contrast with short-term returns, managers may not take advantage of investors’ speculative sentiment and market inefficiency to increase long-term performance. Bayless and Chaplinsky (1996) suggest that stock price reaction to equity issues is dependent on how the issuers signal their value and growth opportunity in the long-term. Firm-specific characteristics such as, for example, high levels of capital spending send out a signal of strong investment opportunity, while high operating risk can influence investors’ beliefs about a firm’s value. Management influence is mitigated to the aftermarket performance compared to the short-term return because of stock market’s long-term efficiency.

advantage of timing when investments are regarded as growth opportunities by irrational investors, while high-leverage firms issue equity in an undervalued market, result in fact that subsequent underperformance of the overvalued stocks in the hot market is due more to the financial market’s long-term efficiency rather than investors’ clear achievements or disappointment with the firm’s net cash flow due to either (1) bad luck (2) irrationally overoptimistic forecasts or (3) fads (Ritter, 1991). In my thesis, I will test the capital-structure hypothesis:

Hypothesis 4: The lower the leverage of an issuer, the poorer the long-term performance relative to market returns

Another reason for poor aftermarket performance, I think, is the liquidity of stocks. Studies on stock returns by Brennan and Subrahmanyam (1996) and Datar et al. (1998) suggest that they are cross-sectionally related to their liquidity: a higher premium is needed by investors to compensate for stock illiquidity and a lower return for liquid stocks. These researchers state that high trading volume contributes to lower required stock returns. This effect persists after controlling for determinants of stock returns (issue size, book-to-market ratio and firm’s beta). This argument leads to my last hypothesis:

Hypothesis 5: The liquidity of stocks has a negative relationship to their future required returns

SECTION 2: Methodology

Identification of Hot and Cold Markets

reduction of adverse selection stimulates firms to issue new equity at lower costs. Industry distribution is an additional and new criterion for examining hot and cold markets. The time span I have chosen (1998-2003) is a representative period, which experienced both the boom and burst of internet and related industries. It also witnessed a great deal of high risk industries going public successfully. These firms raised large amounts of funds to assist early R&D expenditure through IPOs and venture capital. A period with more new equity is defined as a relatively hot market with a high proportion of internet and high-risk industries.

Testing the Short-term Return of IPOs

Firstly, I consider the level of initial return a fairly appropriate compensation for bearing risk or as the cost of providing information because in the price-setting process investment banks have incorporated most public information into the offer price, that is, the pricing process is comparatively efficient. Before the IPO, the company observes the market conditions and provides a range of preliminary offer prices for SEC filings. They then have the opportunity to amend the offer price if they have new market information in the filing period. Although investment banks can adjust the offer price based on its market price (upward in a hot market and downward in a cold market), they retain some necessary room for investors. Thereby the initial return is always regarded as a dependent variable to test other determinants of short-term returns.

because it has been justified by theoretical and empirical studies. Diether, Molloy and Scherbina (2002) document a positive relationship between price volatility and divergence of opinion. Yan Gao (2006) also uses return volatility to test Miller’s theory in the IPO market. In my model, I use return volatility of IPOs over market as a factor to reflect the degree of divergence. As for underwriter reputation, I follow Carter and Manaster’s reputation rankings for IPO underwriters. They rank underwriters from 0 to 9 by several factors: the higher the score, the higher the reputation.

To identify a clean relationship between these three factors and IPO performance, I need to control for all variables of the model.

Initial return =α0+α1* ln(Offer size) +α2* Proportion of issue +α3*σIPO /σ

Nasdaq +α4* Underwriter reputation +α5* Hot market dummy +α6* Internet

dummy +α7* Underwriter reputation* Hot market dummy + ε4

Where

Initial return = (First-day closing price－offer price) / offer price Offer size = Offer price* issue shares

Proportion of issue = Shares offered / shares outstanding

σIPO / σNasdaq = SD of IPO returns / SD of market returns for 30 days and 60 days

starting from the second day of IPO. It represents the level of divergence. IPO daily returns = (Pt+1－Pt) / Pt.

Underwriter reputation = According to Carter and Manaster (1990), reputation is a

score ranging from 0 to 9. I use both the dummy variable

4

Note to short-term model: Expected Sign Ln ( offer size ) = nature logarithm of equity issue size in $millions (＋) Proportion of issue = Shares offered / shares outstanding (－)

σIPO /σNasdaq （＋） Underwriter reputation (both score and dummy variables) (－) Hot market dummy （＋） Internet dummy （＋）

and score. When using the dummy variable, it equals 1 if the score is 8 or higher, and 0 otherwise.

Hot market dummy = 1 in hot year and 0 otherwise.

Internet dummy = 1 in dotcom and dotcom-related industries and 0 otherwise.

Long-term Underperformance Model

An ideal measure of the long-term performance of IPO firms relative to markets would be the total share return. It equals the one-year (and three-year) stock price plus dividends in this period, divided by the first-day closing price. I use the total share return of IPO firms over NASDAQ composites as the dependent variable. The market-to-book ratio is defined as the assets minus the book value of the equity plus the market value of the equity after issuance, all divided by the assets. Market-to-book value is an indicator of investment opportunities, risks5, or a determinant of the optimal leverage ratio. Although issuing firms usually have low leverage after the IPO, they possibly rebalance it through debt issue. Firms with more investment opportunities would rather remain low leverage to avoid more subsequent equity issue. Baker and Wurgler (2002) have documented that high market-to-book ratio (especially the ratios of those firms issued in hot markets) has considerable and persistent effects on a firm’s leverage. They also find that stock returns are lower in subsequent periods if equity accounts for a large proportion of the financing. Therefore I expect a higher market-to-book ratio as a proxy of a firm’s leverage after its IPO is associated with poor long-term performance. Trading volume as an indicator of stock liquidity will contribute to lower required stock returns. In the model, this equals a daily trading volume of 30 days and 60 days after issuance, divided by shares issued, after which I calculate the average trading volume index as an independent variable.

5

The following model is designed to test long-term underperformance of IPOs.

TSR IPO/ TSR NASDAQ = β0 + β1*ln(Market-to-book ratio) + β2*Initial return+

β3*Volume Ipo / Volume Nasdaq + ε6

Where

TSR (Total share return) = (P t+ Total Div in t years)*/P0 t=1 and 3 year

Market-to-book ratio = (Asset－book value of equity +market value of equity after

IPO) / Assets

Initial return = (First-day closing price－offer price) / offer price

Volume Ipo / Volume Nasdaq = Average IPO daily trading volume / average Nasdaq

daily trading volume for 30 and 60 days

SECTION 3: Data and Sample Statistics

My sample consists of 488 IPO firms obtained from Hoovers, a D&B company which offers comprehensive data on companies, industries, market strategy and business deals. If some information was missing, I searched in a number of sources including the new issues database at Thomson financial securities data and the SEC electronic data-gathering and retrieval system (EDGAR). Some material information, such as daily stock price and volume, market-to-book ratio, etc., is quoted from either Yahoo Finance, the NASDAQ website or SEC filings.

The IPOs chosen from Hoovers went public from Jan 1998 through Dec 2003 in Amex, NASDAQ and NYSE and must have raised more than $20 million with an offer price greater than $5 per share. In order to strip away some companies with intense and huge capital or money supply, financial institutions (banks and savings),

6

Note to long-term model: Expected Sign Ln (market-to-book ratio) (－) Initial return (－)

real estate investment trusts (REITs), closed-end funds, and insurance firms were removed.

The available issue-specific data includes the offer size, industry, offer price, total IPO outstanding shares, the lead underwriters, daily stock price and trading volume, market-to-book ratio and the market index. An important index of the underwriter’s prestige ranking comes from the Carter and Manaster (1990) rankings for IPO underwriters. I employed the NASDAQ composite index as market index because it is broad-based and includes over 3,000 companies, more than most other stock market indexes. It measures all U.S. domestic and internationally based common stock types listed on the NASDAQ stock market.

Figure 1 presents the aggregate volume of new issues in each year from 1998 to 2003. Table 1 summarizes some characteristics of IPOs issued in the whole period and in the hot and cold markets. The average first initial return is 49.57% and the issue proportion is 25.28%. I also observed a relatively large volume of overpriced issues (74% of firms’ initial returns are smaller than the mean, and the median initial return is only 19.81%), such that the distribution is highly skewed which is consistent with the existence of a ‘winner’s curse’ in new issue markets. Some 86% of IPOs are underwritten by highly reputable investment banks, 39% of issuers are related to the internet industry and 76% of IPOs were issued in a hot market. The one-year performance of IPO firms over the NASDAQ composite index is 111.54% and the three-year return of IPOs over NASDAQ composite index, on average, is 114.57%. Inconsistent with other literature, I did not observe an average lower long-term return of IPOs relative to market index in my sample. Nevertheless, more than half of the IPO firms performed under the market index in both the first and third years (ratios are nearly 50% and 60% respectively and not included in Table 1).

SECTION 4: Empirical Results

Identification of Relatively Hot and Cold Markets

Table 3 defines the equity issue volume and industry distribution as two criteria for identifying hot and cold markets in each period. The total number of new issues is significantly high in the first three years (56,145 and 170 respectively) and relatively low in the subsequent periods (42, 37 and 38 respectively).

The industry sector distribution of IPOs in the six years is also representative. Table 3 reports on the ten industry sectors in each year and Table 4 describes the main industry title. The industry division derives from the Standard Industry Classification (SIC) codes that indicate the company’s business type. As we know, 1998-2000 was the internet-boom period, when dotcom and related industries all tried to go public to attract investors’ capital support. The number of internet and related IPO firms rose from 19 in 1998 to 75 in 1999 and stayed at 72 in 2000. But only one year later, both internet and the information system sectors experienced a cold winter worldwide. This depression lasted for at least three years and had a serious impact on the internet sector. In my sample from 2001 to 2003, the volume of dotcom and related IPOs fell dramatically each year to fewer than 10. More clear evidence is also found in Figure 2 and Figure 3, each industry presents the distinct trend in the period. Figure 2 presents the total amount of IPOs of each industry sector from 1998 to 2003: the dotcom and related firms issued the highest volume in this period, followed by the biopharmaceuticals, medical device and service industries. Figure 3 shows how the volumes of each industry change over time.

costs made these firms attract only a few investors. I have identified the first three years as hot markets and the subsequent three years as relatively cold markets. Table 5 presents the cyclical distribution of each industry in both hot and cold markets. Consistent with my analysis above, internet and biopharmaceutical companies have similar cyclical distributions (6.4 and 5.4), while medical device and service sectors have the same cyclical distribution (both are 3.5). IPO numbers in other industries changed little in either hot or cold markets. Figure 4 displays a comparison of industry distribution in hot and cold markets. Most sectors issue more equity in hot markets than in cold.

Testing Relationship between Three Factors and Short-term Return

of IPOs

Seven variables and the results are shown in Panel A of Table 6. As predicted, the issued proportion has a significant negative relationship with initial return. In hot markets with a high price, firms would like to issue higher proportions of equities. On the one hand they can raise more funds in this period, but on the other they left money on the table as a signal to investors that their firms were good quality and that high-growth opportunities could only come through lower levels of underpricing. It is the opposite situation in the cold market. Moreover, from the perspective of insiders (previous owners and managers), high underpricing may generate high price and oversubscription until the end of lockup expiration. Because the high first-day return attracts more investors for aftermarket purchases, insiders can exploit investors to maximize their personal wealth through secondary-offering shares retained on lockup expiration. More first-day return usually relates to more retained ownership and a low proportion sold to public investors on the IPO date.

a higher initial return, although the result is insignificant. It suggests that more return volatility over the market requires a high premium for bearing the risks.

However, underwriter reputation (score) in Panel A did not at first present the expected significant negative relationship to initial returns. One explanation is that in my sample, more than 86% of IPO firms are backed by highly prestigious investment banks (scoring 8 and higher); the difference between them is insignificant. But I prefer an alternative interpretation: underwriters were unwilling to price these offerings at a high level even if market would have accepted it. A majority of the firms in my sample (76%, 371 out of total 488) went public between 1998 and 2000 in the internet boom; about 45% of these firms were either in or related to the internet and information system industry. During the internet bubble, everyone was concerned about the true value of stocks. Underwriters were unwilling to make a higher offer prices even if the market was willing to pay because they wanted to avoid lawsuits and protect their good reputations if and when the share price dropped sharply. Prestigious underwriters are aware that on the IPO date, traders and other speculators will bid up the price to irrational levels, reputable underwriters were unwilling to price their offerings at the market level determined by these ‘noise traders’. However, if this was the case, I would expect that top-tier underwriters correlated to lower initial returns when hot factors were removed. So I re-examined the underwriter’s effect on underpricing in two sub-groups of relatively hot and cold markets. I use the cold market as a group where the internet bubble was removed.

score variables (-0.0054). It supports my interpretation above and in normal or cold markets without many noise traders: prestigious underwriters can and are willing to reduce the severe level of underpricing. That is why new issue firms would like to cooperate with top-tier underwriters even though they charge for a great deal in fees (at least in my sample, nearly 86% of IPO firms choose prestigious underwriters whose score is 8 or higher). Especially in cold or normal markets, firms can reduce issue costs and alleviate the problem of information asymmetry through the lower amount left on the table.

Effects of Issuer’s Capital Structure and Liquidity on Long-term

Underperformance

I examined the subsequent performance of IPO firms for one and three years relative to market performance and these results are presented in Table 7. In general, the three-year return model of new issues produces better results than that of the one-year return.

IPOs.

Initial return presents different results in the two periods: at first it has an insignificantly positive relationship with one-year stock returns. However, in accordance with my prediction, three-year stock performance shows a significantly negative correlation to initial returns. Why does the one-year return, as a proxy for long-term performance, have the opposite result to a three-year return? I consider it is because the majority of firms issued IPOs between 1998 and 2000, and one year later these firms (issued in 1998 and 1999) were still in hot markets and kept their high prices. Similar proof can also be found in the long-term return distribution. About 50% of firms underperformed relative to the market index after one year, while the same ratio for three years is more than 60% (ratio is not included in Table 1). Therefore, in my sample, due to these abnormal hot market effects, the one-year return is not a good proxy for measuring long-term underperformance relative to three years or longer periods. But in sub-group sample, the coefficients of initial returns are significantly different: it is negative (-0.1181) in the hot market but significantly positive (1.1809) in the cold market. It suggests that firms issued in the cold market did not take more advantage of timing on IPO date and the short-term return is consistent to the long-term performance.

In Table 7, I also did not observe the expected negative relationship between trading volume and one-year performance. However, the trading volume of IPO firms over the market presents negative relationship to the three-year performance using both 30 and 60 days volume. It supports the liquidity theory: when attached stock liquidity increases relative to the whole market, the required premium by investors is reduced.

SECTION 5: Conclusions and Recommendations

review of all the associated factors. The theory and literature are primarily based on information asymmetry between issuers and investors, as well as on financial behaviour.

Firstly, I documented a timing effect in my sample: there is abnormal new issue volume in hot periods relative to other periods. In distribution terms, there are proportionally more internet and high risk firms. I also tested the relationship between my five proxies for various characteristics of new issuers, and short and long-term performance. In the initial-return model, theories concerning issue proportion, underwriter’s reputation and divergence of opinion are all documented. And in the long-term return model (three years), all the variables presented the expected indications and the results confirm my prior theories: lower leverage requires lower returns, the abnormally high initial return are often accompanied by subsequent poor long-term return, and stock liquidity theory can explain the lower level of long-term return.

There are two main findings in my study. The first finding is that top-tier underwriters are not always associated with lower levels of underpricing. In some environments, such as during the internet boom, underwriters are unwilling to price offerings at high levels because they want to avoid lawsuits and protect their prestigious reputation if and when the bubble eventually bursts. The second finding is that the problem of information asymmetry is still an important consideration for firms choosing an IPO. It suggests that an equity issue is not always the last resort. When issue costs are lowered to compensate for information asymmetry, firms prefer to go public rather than employ other financing methods. In hot markets, more equity financing drives the issuer’s aftermarket capital structure, and the return required by investors is decreased because of the low leverage and its attached risks.

References

Alti, A., (2004). How persistent is the impact of market timing on capital structure? Working paper (University of Texas)

Aggarwal, R., Leal, R. and Hernandez, L.(1990). The after market performance of initial public offerings in Latin America, Financial management, pp42-54 Balvers, R., McDonald, B. and Miller R. (1988).Underpricing of new issue and the

choice of auditor as a signal of investment banker reputation, The accounting

review Vol.63 p605-709

Bayless M. and Chaplinsky S. (1996).Is there a window of opportunity for seasoned equity issuance? Journal of finance, Vol.51 No.1 p253-278

Block, S. and Stanley, M. (1980).The Financial Characteristics and Price Movement Patterns of Companies Approaching the Unseasoned Securities Market in the Late 1970s, Financial Management, Vol.9 p30-36

Brennan and Franks (1997).Underpricing, ownership and control in initial public offerings of equity securities in the UK, Journal of financial economics, Vol.

45 p391-413

Brennan, M.J., Subrahmanyam, A., (1996). Market microstructure and asset pricing: on the compensation for illiquidity in stock returns, Journal of financial

economics Vol.41, p441-464

Carter, R.B. and Manaster, S. (1990). Initial Public Offerings and Underwriter Reputation, Journal of Finance, Vol.45, Issue 4, p1045-1068

Datar, V., Naik, N., Radcliffe, R., (1998).Liquidity and stock returns: an alternative test, Journal of financial markets, Vol.1 p203-219

Draho, J., (2001).The coordination role of public information in hot market IPOs, working paper (Yale University)

Diether, K., C. Molloy, and A. Scherbina, (2002).Differences of opinion and cross-section of stock returns, Journal of Finance, Volume 57, 2113-41

Franzke, S. A. (2003).Underpricing of Venture-Backed and Non Venture-Backed IPOs: Germany’s Neuer Market, CFS Working Paper, No.2001/01, Center for Financial studies, Frankfurt.

Houge.T. Loughran.T. (2001).Divergence of opinion, uncertainty, and the quality of initial public offerings, Financial management (Winter 2001) p5-23

Jog, V.M., & Riding, A.L.(1987).Underpricing in Canadian IPOs, Financial Analysis

Journal, November-December, 48-55.

Lee Charle M.C. and Swaminathan B (2000).Price momentum and trading volume,

Journal of finance, Vol. 55 p2017-69

Lee,P.J.,S.L.Taylor, and T.S. Walter. (1996a).Australian IPO Pricing in the short- and long-run, Journal of Banking and Finance, Vol.20, p1189-1210

Loughran, T. and J. Ritter (1995).The new issues puzzle, Journal of Finance V 50. No

1, pp23-51

Loughran, T. and J. Ritter (2002).Why has IPO underpricing changed over time, AFA 2003 Washington, DC Meetings

Loughran, T. and J. Ritter (2002).Why don’t issuers get upset about leaving money on the table in IPOs? The review of Financial Studies 15:2, 413-443

Loughran, T. and J. Ritter (2004).Why has IPO underpricing changed over time?

Financial Management, autumn 2004, pp5-37

Malcolm Baker and Jeffrey Wurgler (2002).Market timing and capital structure,

Journal of finance, Vol57, No.1, p1-32

Miller,E.,(1977).Risk uncertainty ,and divergence of opinion, Journal of finance,

Vol.32, No.4, p1151-68

Pagano,M., Panetta F, Zingales A.L (1998).Why do companies go public? An empirical analysis, Journal of finance, Vol.53, No.1, p27-64

Pastor L. and Veronesi P. (2003).Stock prices and IPO waves, Working paper (University of Chicago)

Philip J.L., Stephen and Terry (1996).Australian IPO pricing in the short and long run,

Journal of banking &finance, Vol. 20 p1189-1210

Economics, Vol.15, Issues 1-2, January-February 1986, Pages 187-212

Ritter, J.R. (1998).Initial Public Offerings, Contemporary Finance Digest, Vol 2., No.1,

p5-30, modified version.

Ritter, J.R. (1987).The cost of going public, Journal of Financial Economics, Vol.19,

p269-281

Ritter, J.R. (1991).The long-run performance of initial public offerings, Journal of

finance,Vol. 46. No.1, p3-27

Yan Gao, Connie X. Mao, Rui Zhong (2006).Divergence of opinion and long-term performance of initial public offerings, Journal of Financial Research,

pp113-129

Appendix 1: Figures

Figure 1: Numbers of IPO Included in the Test Sample

Figure 3: Comparison of Industry Distribution from 1998 to 2003

Industry Distribution in Each Year

0 10 20 30 40 50 60 70 80 1998 1999 2000 2001 2002 2003 1*** 2*** 3*** 4*** 5*** 6*** 7*** 8*** .com