at a lower price.Because the underwriter cannot directly gain from any price appreciation above the offer price on unsold securities,while bearing the full downside of any price fall,there is every incentive to fully distribute the securities offered. Based on the logic of the efficient markets hypothesis,beginning in 1982 the SEC began permitting publicly traded firms meeting certain requirements(basically,large firms)to issue securities without distributing a prospectus.Instead,SEC Rule 415 states that by filing a letter with the SEC disclosing the intention of selling additional securities within the next two years,a firm can sell the securities whenever it wants.Existing disclosures,such as quarterly financial statements,are deemed to be sufficient information to investors.The securities can be taken off the shelf and sold,in what are known as"shelf'issues.In practice,shelf issues are commonly done for bond offerings.Before selling equity,however,many firms prefer to hire an investment banker and conduct a marketing campaign (the road show),complete with a prospectus.From 1984-1992 there were virtually no shelf equity offerings,but they have enjoyed a resurgence since then (Heron and Lie(2003)). 1.3 The information conveyed by investment and financing activities Smith's classic 1986 survey article "Investment Banking and the Capital Acquisition Process,"focused on announcement effects associated with securities offerings and other corporate actions.These transactions can be categorized on the basis of the leverage change and the implied cash flow change.For example,calling a convertible bond (forcing conversion into equity)decreases a firm's leverage and reduces its need for cash flow to meet interest payments, and repurchasing stock increases leverage and uses cash flow.The studies that he surveyed found that leverage-decreasing transactions on average are associated with negative announcement effects if new capital is raised(such as with equity issues).Leverage-increasing transactions on average are associated with positive announcement effects if no new capital is raised (such as with a share repurchase).As Smith pointed out,these patterns are difficult to reconcile with traditional tradeoff models of optimal capital structure.The patterns are consistent,however,with informational asymmetries and agency problems being of importance. There are several problems with interpreting announcement effects.First,and most mechanically,in an efficient market the announcement effect will measure the difference between the post-announcement valuation and what was expected beforehand.If investors had a high likelihood of an announcement occurring beforehand,this updating element is small,and the announcement effect vastly underestimates the impact of the event.Second,any financing activity implicitly is associated with an investment activity,and any investment activity is implicitly associated with a financing activity.Corporate financing and investment actions invariably convey information about both of these activities,due to the identity that sources of funds uses of funds.For example,if a firm raises external capital,the firm is implicitly conveying the information that internal funds will be insufficient to finance its activities(bad news).It is also conveying the information that it will be investing more than if it didn't finance externally.This may be good or bad news,depending upon the desirability of the investment.So the announcement effect depends upon the relative magnitude of multiple implicit and explicit pieces of information. 6
6 at a lower price. Because the underwriter cannot directly gain from any price appreciation above the offer price on unsold securities, while bearing the full downside of any price fall, there is every incentive to fully distribute the securities offered. Based on the logic of the efficient markets hypothesis, beginning in 1982 the SEC began permitting publicly traded firms meeting certain requirements (basically, large firms) to issue securities without distributing a prospectus. Instead, SEC Rule 415 states that by filing a letter with the SEC disclosing the intention of selling additional securities within the next two years, a firm can sell the securities whenever it wants. Existing disclosures, such as quarterly financial statements, are deemed to be sufficient information to investors. The securities can be taken off the shelf and sold, in what are known as “shelf” issues. In practice, shelf issues are commonly done for bond offerings. Before selling equity, however, many firms prefer to hire an investment banker and conduct a marketing campaign (the road show), complete with a prospectus. From 1984-1992 there were virtually no shelf equity offerings, but they have enjoyed a resurgence since then (Heron and Lie (2003)). 1.3 The information conveyed by investment and financing activities Smith’s classic 1986 survey article “Investment Banking and the Capital Acquisition Process,” focused on announcement effects associated with securities offerings and other corporate actions. These transactions can be categorized on the basis of the leverage change and the implied cash flow change. For example, calling a convertible bond (forcing conversion into equity) decreases a firm’s leverage and reduces its need for cash flow to meet interest payments, and repurchasing stock increases leverage and uses cash flow. The studies that he surveyed found that leverage-decreasing transactions on average are associated with negative announcement effects if new capital is raised (such as with equity issues). Leverage-increasing transactions on average are associated with positive announcement effects if no new capital is raised (such as with a share repurchase). As Smith pointed out, these patterns are difficult to reconcile with traditional tradeoff models of optimal capital structure. The patterns are consistent, however, with informational asymmetries and agency problems being of importance. There are several problems with interpreting announcement effects. First, and most mechanically, in an efficient market the announcement effect will measure the difference between the post-announcement valuation and what was expected beforehand. If investors had a high likelihood of an announcement occurring beforehand, this updating element is small, and the announcement effect vastly underestimates the impact of the event. Second, any financing activity implicitly is associated with an investment activity, and any investment activity is implicitly associated with a financing activity. Corporate financing and investment actions invariably convey information about both of these activities, due to the identity that sources of funds = uses of funds. For example, if a firm raises external capital, the firm is implicitly conveying the information that internal funds will be insufficient to finance its activities (bad news). It is also conveying the information that it will be investing more than if it didn’t finance externally. This may be good or bad news, depending upon the desirability of the investment. So the announcement effect depends upon the relative magnitude of multiple implicit and explicit pieces of information
A substantial literature,dating back to the mid-1980s,documents that the market reacts negatively,on average,to the announcement of equity issues in the U.S..Convertible bond issues generally are greeted with a moderate negative reaction.Bond offerings have slightly negative reactions,and share repurchases are greeted with positive announcement effects.In the last decade,researchers have examined the long-run performance of firms following these events.The long-run performance evidence shows that in general the market underreacts to the announcement. Most of the literature on long-run performance has focused on relative performance,i.e., do issuing companies underperform a benchmark?Baker and Wurgler(2000),however,present empirical evidence that issuing firms display market timing ability.Using U.S.data on issues of debt and equity (IPOs and SEOs),they find that the fraction of external financing that is equity predicts the following calendar year's stock market return with greater reliability than either the market dividend yield or the market's market-to-book ratio.Baker and Wurgler's sample covers returns from 1928 to 1997.Interestingly,the fraction of equity issuance was highest in 1929,a year that included the October stock market crash.When the sample period is split in two, however,their results hold in both subperiods. If firms can successfully time their equity offerings to take advantage ofwindows of opportunity,"they have a time-varying cost of external capital.How should this affect a firm's investment and financing policies?Stein(1996)addresses this important issue,and concludes that the normative answer depends upon the interaction of two assumptions.The first assumption is whether differences in the cost of external equity reflect misvaluations or differences in equilibrium expected returns.The second assumption is whether managers are trying to maximize short-run firm value or long-run firm value.If one assumes that a low expected return occurs because a stock is overvalued,then managers should issue stock but not invest in low return activities if they are focused on maximizing the wealth of long-term shareholders.On the other hand,if one assumes that low expected returns are rationally being forecast by investors,then a firm should issue stock and use a lower hurdle rate in choosing its investments,much as the neoclassical model of optimal investing and financing would recommend The remainder of this chapter discusses securities issuance.In Section 3,the short-run and long-run reactions to various corporate announcements will be summarized.In Section 4, initial public offerings will be analyzed in detail,with substantial focus on contractual mechanisms.But first,detailed attention is given to firms conducting seasoned equity offerings 2.Seasoned Equity Offerings(SEOs) When a firm that is already publicly traded sells additional stock,the new shares are perfect substitutes for the existing shares.For these transactions,the academic literature tends to use the term seasoned equity offering(SEO),as contrasted with an unseasoned equity offering, an IPO.Practitioners generally use the term follow-on offering,especially if the equity issue is within several years of the IPO.SEOs are also referred to as secondaries,although secondary offering is a term that can mean either a follow-on offering or shares being sold by existing 7
7 A substantial literature, dating back to the mid-1980s, documents that the market reacts negatively, on average, to the announcement of equity issues in the U.S.. Convertible bond issues generally are greeted with a moderate negative reaction. Bond offerings have slightly negative reactions, and share repurchases are greeted with positive announcement effects. In the last decade, researchers have examined the long-run performance of firms following these events. The long-run performance evidence shows that in general the market underreacts to the announcement. Most of the literature on long-run performance has focused on relative performance, i.e., do issuing companies underperform a benchmark? Baker and Wurgler (2000), however, present empirical evidence that issuing firms display market timing ability. Using U.S. data on issues of debt and equity (IPOs and SEOs), they find that the fraction of external financing that is equity predicts the following calendar year’s stock market return with greater reliability than either the market dividend yield or the market’s market-to-book ratio. Baker and Wurgler’s sample covers returns from 1928 to 1997. Interestingly, the fraction of equity issuance was highest in 1929, a year that included the October stock market crash. When the sample period is split in two, however, their results hold in both subperiods. If firms can successfully time their equity offerings to take advantage of “windows of opportunity,” they have a time-varying cost of external capital. How should this affect a firm’s investment and financing policies? Stein (1996) addresses this important issue, and concludes that the normative answer depends upon the interaction of two assumptions. The first assumption is whether differences in the cost of external equity reflect misvaluations or differences in equilibrium expected returns. The second assumption is whether managers are trying to maximize short-run firm value or long-run firm value. If one assumes that a low expected return occurs because a stock is overvalued, then managers should issue stock but not invest in low return activities if they are focused on maximizing the wealth of long-term shareholders. On the other hand, if one assumes that low expected returns are rationally being forecast by investors, then a firm should issue stock and use a lower hurdle rate in choosing its investments, much as the neoclassical model of optimal investing and financing would recommend. The remainder of this chapter discusses securities issuance. In Section 3, the short-run and long-run reactions to various corporate announcements will be summarized. In Section 4, initial public offerings will be analyzed in detail, with substantial focus on contractual mechanisms. But first, detailed attention is given to firms conducting seasoned equity offerings. 2. Seasoned Equity Offerings (SEOs) When a firm that is already publicly traded sells additional stock, the new shares are perfect substitutes for the existing shares. For these transactions, the academic literature tends to use the term seasoned equity offering (SEO), as contrasted with an unseasoned equity offering, an IPO. Practitioners generally use the term follow-on offering, especially if the equity issue is within several years of the IPO. SEOs are also referred to as secondaries, although secondary offering is a term that can mean either a follow-on offering or shares being sold by existing
shareholders,as opposed to a primary offer where the issuing firm is receiving the proceeds. (And on the subject of ambiguous terms,this chapter will use public ownership to mean stock that is traded in the market,rather than government ownership.Private ownership is used to mean non-traded stock,rather than being owned by the private sector. 2.1 Announcement effects Numerous studies have documented that in the U.S.there is an announcement effect of -2%,on average,for SEOs.The most popular explanation among academics for this negative announcement effect is that of the Myers and Majluf(1984)adverse selection model.Myers and Majluf assume that management wants to maximize the wealth of its existing shareholders in the long run.At any point in time,however,the current market price may be too high or too low relative to management's private information about the value of assets in place.In other words, strong-form market inefficiency is being assumed.If management thinks that the current market price is too low,the firm will not issue undervalued stock,for doing so dilutes the fractional ownership of existing shareholders.If management thinks that the current stock price is too high,however,the firm will issue equity if debt financing is not an option.Rational investors, knowing this decision rule,therefore interpret an equity issue announcement as conveying management's opinion that the stock is overvalued,and the stock price falls.2 How this negative announcement effect should be interpreted is a subject of debate.If a firm is issuing shares equal to 20%of its existing shares,a downward revaluation of 2%for the existing shares is a dollar amount equal to 10%of the proceeds being raised.If this 2%drop is viewed as a cost of an equity issue,then external equity capital is very expensive.On the other hand,if this 2%drop would have occurred when the basis for management's opinion regarding firm value was disclosed in some other manner,then the downward revaluation is not a cost of the equity issue for long-term shareholders.It is a cost only to those shareholders who would have sold their shares in between the equity issue announcement and when the negative news would have otherwise been impounded into the share price.If this is the case,then the negative announcement effect is mainly a matter of indifference to a firm where long-term shareholder wealth maximization is the objective,and external equity is not inordinately costly. As mentioned earlier,when a firm raises external equity capital,it not only conveys information about whether management thinks the firm is overvalued or not,but also suggests that something will be done with the funds raised.If the market interprets the equity issue as implying that a new positive net present value project will be undertaken,the announcement effect could be positive.On the other hand,if the market is concerned that the equity issue means that management will squander the funds on empire building,then the announcement effect could be interpreted as causally linked to the equity issue,in which case external equity is in fact very expensive.The rationale is that the additional equity resources are relaxing a constraint on management's tendency to engage in "empire-building,"or growth for the sake of growth.In other words,agency problems between shareholders and managers are intensified. 2 The Myers-Majluf predictions are very sensitive to the assumptions about the objective function of management, the portfolio rebalancing rules of investors,and the source of information asymmetries.Daniel and Titman(1995) discuss some of these issues in detail. 8
8 shareholders, as opposed to a primary offer where the issuing firm is receiving the proceeds. (And on the subject of ambiguous terms, this chapter will use public ownership to mean stock that is traded in the market, rather than government ownership. Private ownership is used to mean non-traded stock, rather than being owned by the private sector.) 2.1 Announcement effects Numerous studies have documented that in the U.S. there is an announcement effect of –2%, on average, for SEOs. The most popular explanation among academics for this negative announcement effect is that of the Myers and Majluf (1984) adverse selection model. Myers and Majluf assume that management wants to maximize the wealth of its existing shareholders in the long run. At any point in time, however, the current market price may be too high or too low relative to management’s private information about the value of assets in place. In other words, strong-form market inefficiency is being assumed. If management thinks that the current market price is too low, the firm will not issue undervalued stock, for doing so dilutes the fractional ownership of existing shareholders. If management thinks that the current stock price is too high, however, the firm will issue equity if debt financing is not an option. Rational investors, knowing this decision rule, therefore interpret an equity issue announcement as conveying management’s opinion that the stock is overvalued, and the stock price falls.2 How this negative announcement effect should be interpreted is a subject of debate. If a firm is issuing shares equal to 20% of its existing shares, a downward revaluation of 2% for the existing shares is a dollar amount equal to 10% of the proceeds being raised. If this 2% drop is viewed as a cost of an equity issue, then external equity capital is very expensive. On the other hand, if this 2% drop would have occurred when the basis for management’s opinion regarding firm value was disclosed in some other manner, then the downward revaluation is not a cost of the equity issue for long-term shareholders. It is a cost only to those shareholders who would have sold their shares in between the equity issue announcement and when the negative news would have otherwise been impounded into the share price. If this is the case, then the negative announcement effect is mainly a matter of indifference to a firm where long-term shareholder wealth maximization is the objective, and external equity is not inordinately costly. As mentioned earlier, when a firm raises external equity capital, it not only conveys information about whether management thinks the firm is overvalued or not, but also suggests that something will be done with the funds raised. If the market interprets the equity issue as implying that a new positive net present value project will be undertaken, the announcement effect could be positive. On the other hand, if the market is concerned that the equity issue means that management will squander the funds on empire building, then the announcement effect could be interpreted as causally linked to the equity issue, in which case external equity is in fact very expensive. The rationale is that the additional equity resources are relaxing a constraint on management’s tendency to engage in “empire-building,” or growth for the sake of growth. In other words, agency problems between shareholders and managers are intensified. 2 The Myers-Majluf predictions are very sensitive to the assumptions about the objective function of management, the portfolio rebalancing rules of investors, and the source of information asymmetries. Daniel and Titman (1995) discuss some of these issues in detail
A number of empirical studies have documented cross-sectional patterns in the equity issue announcement effect.In general,these results show that there is less of a negative reaction when a firm can convince the market that there is a good reason for issuing equity,and there is a more negative reaction when good motivations are not obvious.Jung,Kim,and Stulz (1996) report that firms with a high q(market value-to-replacement cost),reflecting good investment opportunities,have an announcement effect that is insignificantly different from zero.Choe, Masulis,and Nanda (1993)document that the announcement effect is less negative when the economy is in an expansionary segment of the business cycle,when there may be less adverse selection risk. Korajczyk,Lucas,and McDonald (1991)report that the announcement effect is less negative if it follows shortly after an earnings report,at which time there is presumed to be less asymmetric information.Houston and Ryngaert(1997)provide direct evidence that adverse selection concerns explain part of the negative announcement effect.They study bank mergers, where common stock is the dominant means of payment to the shareholders of target banks. Some merger agreements specify that the target shareholders will receive a fixed number of shares in the acquiring bank(a fixed ratio stock offer),and other merger agreements specify a variable number of shares that add up to a fixed dollar amount(a conditional stock offer).If target shareholders are concerned that the acquirer is offering overvalued stock,the conditional stock offer provides protection against price drops.Consistent with adverse selection concerns, the announcement effect is-3.3%for fixed ratio stock offers,but only-1.1%for conditional stock offers. In general,studies find that larger issues have more negative effects.One problem with interpreting the relation between issue size and announcement effects is that if there is an unusually negative reaction,the issue size may be cut back by the time the deal is completed Since existing empirical studies do not take this endogeneity into account,the empirical estimates of the effect of issue size on the announcement are subject to a simultaneous equations bias.This bias results in an underestimate of the magnitude of the effect of issue size on the stock price.Thus,academics undoubtedly underestimate the degree to which the demand curve for a stock is negatively sloped. On the issue date,SEOs are,on average,sold at a discount of about 3%relative to the market price on the day prior to issuing(Corwin(2003),Mola and Loughran(2002)).Mola and Loughran report that the size of this discount has grown over time,and that there has been an increasing tendency to set the offer price at an integer.For example,in recent years a stock trading at $31.75 would very likely be priced at $30.00 or $31.00,whereas in the 1980s it would have been more likely to be priced at $31.00 or $31.50. 2.2 Evidence on long-run performance The long-run performance of SEOs has been the subject of a number of studies,all of which find that firms conducting SEOs typically have high returns in the year before issuing. For example,Loughran and Ritter(1995)report an average return in the year before issuing of 72%.During the five years after issuing,however,the returns are below normal (about 11%per 9
9 A number of empirical studies have documented cross-sectional patterns in the equity issue announcement effect. In general, these results show that there is less of a negative reaction when a firm can convince the market that there is a good reason for issuing equity, and there is a more negative reaction when good motivations are not obvious. Jung, Kim, and Stulz (1996) report that firms with a high q (market value-to-replacement cost), reflecting good investment opportunities, have an announcement effect that is insignificantly different from zero. Choe, Masulis, and Nanda (1993) document that the announcement effect is less negative when the economy is in an expansionary segment of the business cycle, when there may be less adverse selection risk. Korajczyk, Lucas, and McDonald (1991) report that the announcement effect is less negative if it follows shortly after an earnings report, at which time there is presumed to be less asymmetric information. Houston and Ryngaert (1997) provide direct evidence that adverse selection concerns explain part of the negative announcement effect. They study bank mergers, where common stock is the dominant means of payment to the shareholders of target banks. Some merger agreements specify that the target shareholders will receive a fixed number of shares in the acquiring bank (a fixed ratio stock offer), and other merger agreements specify a variable number of shares that add up to a fixed dollar amount (a conditional stock offer). If target shareholders are concerned that the acquirer is offering overvalued stock, the conditional stock offer provides protection against price drops. Consistent with adverse selection concerns, the announcement effect is –3.3% for fixed ratio stock offers, but only –1.1% for conditional stock offers. In general, studies find that larger issues have more negative effects. One problem with interpreting the relation between issue size and announcement effects is that if there is an unusually negative reaction, the issue size may be cut back by the time the deal is completed. Since existing empirical studies do not take this endogeneity into account, the empirical estimates of the effect of issue size on the announcement are subject to a simultaneous equations bias. This bias results in an underestimate of the magnitude of the effect of issue size on the stock price. Thus, academics undoubtedly underestimate the degree to which the demand curve for a stock is negatively sloped. On the issue date, SEOs are, on average, sold at a discount of about 3% relative to the market price on the day prior to issuing (Corwin (2003), Mola and Loughran (2002)). Mola and Loughran report that the size of this discount has grown over time, and that there has been an increasing tendency to set the offer price at an integer. For example, in recent years a stock trading at $31.75 would very likely be priced at $30.00 or $31.00, whereas in the 1980s it would have been more likely to be priced at $31.00 or $31.50. 2.2 Evidence on long-run performance The long-run performance of SEOs has been the subject of a number of studies, all of which find that firms conducting SEOs typically have high returns in the year before issuing. For example, Loughran and Ritter (1995) report an average return in the year before issuing of 72%. During the five years after issuing, however, the returns are below normal (about 11% per
year,according to Table 1 below).Partly this is due to"market timing,"and partly it is due to abnormal performance relative to a benchmark.The conclusions regarding abnormal performance are hotly debated,and sensitive to the methodology employed and the sample used Figure 1 illustrates the evidence regarding average annual returns in the five years after issuing. The numbers show that,for 7,760 SEOs from 1970-2000,the average annual return in the five years after issuing is 10.8%.Nonissuing firms of the same size(market capitalization)have average annual returns of 14.4%.Therefore,relative to a size-matched benchmark,issuers underperform by 3.6%per year for five years. Table 1 Mean percentage returns on SEOs from 1970-2000 during the first five years after issuing First Second Geometric Six Six First Second Third Fourth Fifth Mean monthsmonths year year year year year years 1-5 SEO firms 6.7% 1.5% 9.4% 3.6% 10.9% 14.7% 15.9% 10.8% Size-matched 6.1% 7.0% 14.0% 12.9% 14.4% 15.3% 15.5% 14.4% Difference 0.6% -5.5% -4.6% -9.3% -3.5% -0.6% 0.4% -3.6% Number 7,502 7,475 7,504 7,226 6.603 5.936 5,188 7,760 SEO firms 7.4% 2.2% 10.6% 5.3% 12.3% 14.2% 14.2% 11.3% Style- 5.4% 5.6% 11.5% 13.6% 15.6% 16.9% 15.9% 14.7% matched Difference 2.0% -3.4% -0.9% -8.3% -3.3% -2.7% -1.7% -3.4% Number 6,638 6,622 6,638 6,289 5,711 5,123 4,448 6,638 All averages are equally weighted.For the first year,the returns are measured from the closing market price on the issue date until the sixth-month or one-year anniversary.All returns are equally weighted average returns for all seasoned equity offerings(SEOs)that are still traded on Nasdag,the Amex,or the NYSE at the start of a period.If an issuing firm is delisted within an event year,its return for that year is calculated by compounding the CRSP value-weighted market index for the rest of the year.The matching firm is treated as if it delisted on the same date. and its return for the remainder of the year is calculated using the CRSP VW index.Thus,once an SEO is delisted, by construction there is no abnormal performance for the remainder of the year.For the size-matched returns,each SEO is matched with a nonissuing firm having the same market capitalization (using the closing market price on the first day of trading for the SEO,and the market capitalization at the end of the previous month for the matching firms).For the style-matched returns,each SEO is matched with a nonissuing firm in the same size decile (using NYSE firms only for determining the decile breakpoints)having the closest book-to-market ratio.For nonissuing firms,the Compustat-listed book value of equity for the most recent fiscal year ending at least four months prior to the SEO date is used,along with the market cap at the close of trading at month-end prior to the month of the SEO with which it is matched.Nonissuing firms are those that have been listed on the Amex-Nasdag-NYSE for at least five years,without issuing equity for cash during that time.If a nonissuer subsequently issues equity,it is still used as the matching firm.If a nonissuer gets delisted prior to the delisting(or the fifth anniversary,or Dec.31,2001), the second-closest matching firm on the original IPO date is substituted,on a point-forward basis.For firms with multiple classes of stock outstanding,market cap is calculated based using only the class in the SEO for the SEO. For nonissuing firms,each class of stock is treated as if it is a separate firm.The sample size is 7,760 SEOs from 1970-2000 when size-matching is used,excluding SEOs with an offer price of less than $5.00,ADRs,REITs, closed-end funds,and unit offers.All SEOs are listed on CRSP for at least 6 months,and after Nasdag's inclusion, are listed within six months of going public.Returns are measured through December 31,2001.For partial event- years that end on this date,the last partial year is deleted from the computations.In other words,for an SEO that issued on March 15,2000,it's first-year return is included,but not the second-year return. 10
10 year, according to Table 1 below). Partly this is due to “market timing,” and partly it is due to abnormal performance relative to a benchmark. The conclusions regarding abnormal performance are hotly debated, and sensitive to the methodology employed and the sample used. Figure 1 illustrates the evidence regarding average annual returns in the five years after issuing. The numbers show that, for 7,760 SEOs from 1970-2000, the average annual return in the five years after issuing is 10.8%. Nonissuing firms of the same size (market capitalization) have average annual returns of 14.4%. Therefore, relative to a size-matched benchmark, issuers underperform by 3.6% per year for five years. Table 1 Mean percentage returns on SEOs from 1970-2000 during the first five years after issuing First six months Second six months First year Second year Third year Fourth year Fifth year Geometric Mean years 1-5 SEO firms 6.7% 1.5% 9.4% 3.6% 10.9% 14.7% 15.9% 10.8% Size-matched 6.1% 7.0% 14.0% 12.9% 14.4% 15.3% 15.5% 14.4% Difference 0.6% -5.5% -4.6% -9.3% -3.5% -0.6% 0.4% -3.6% Number 7,502 7,475 7,504 7,226 6,603 5,936 5,188 7,760 SEO firms 7.4% 2.2% 10.6% 5.3% 12.3% 14.2% 14.2% 11.3% Stylematched 5.4% 5.6% 11.5% 13.6% 15.6% 16.9% 15.9% 14.7% Difference 2.0% -3.4% -0.9% -8.3% -3.3% -2.7% -1.7% -3.4% Number 6,638 6,622 6,638 6,289 5,711 5,123 4,448 6,638 All averages are equally weighted. For the first year, the returns are measured from the closing market price on the issue date until the sixth-month or one-year anniversary. All returns are equally weighted average returns for all seasoned equity offerings (SEOs) that are still traded on Nasdaq, the Amex, or the NYSE at the start of a period. If an issuing firm is delisted within an event year, its return for that year is calculated by compounding the CRSP value-weighted market index for the rest of the year. The matching firm is treated as if it delisted on the same date, and its return for the remainder of the year is calculated using the CRSP VW index. Thus, once an SEO is delisted, by construction there is no abnormal performance for the remainder of the year. For the size-matched returns, each SEO is matched with a nonissuing firm having the same market capitalization (using the closing market price on the first day of trading for the SEO, and the market capitalization at the end of the previous month for the matching firms). For the style-matched returns, each SEO is matched with a nonissuing firm in the same size decile (using NYSE firms only for determining the decile breakpoints) having the closest book-to-market ratio. For nonissuing firms, the Compustat-listed book value of equity for the most recent fiscal year ending at least four months prior to the SEO date is used, along with the market cap at the close of trading at month-end prior to the month of the SEO with which it is matched. Nonissuing firms are those that have been listed on the Amex-Nasdaq-NYSE for at least five years, without issuing equity for cash during that time. If a nonissuer subsequently issues equity, it is still used as the matching firm. If a nonissuer gets delisted prior to the delisting (or the fifth anniversary, or Dec. 31, 2001), the second-closest matching firm on the original IPO date is substituted, on a point-forward basis. For firms with multiple classes of stock outstanding, market cap is calculated based using only the class in the SEO for the SEO. For nonissuing firms, each class of stock is treated as if it is a separate firm. The sample size is 7,760 SEOs from 1970-2000 when size-matching is used, excluding SEOs with an offer price of less than $5.00, ADRs, REITs, closed-end funds, and unit offers. All SEOs are listed on CRSP for at least 6 months, and after Nasdaq’s inclusion, are listed within six months of going public. Returns are measured through December 31, 2001. For partial eventyears that end on this date, the last partial year is deleted from the computations. In other words, for an SEO that issued on March 15, 2000, it’s first-year return is included, but not the second-year return