1
Impact of Law Firms on
the IPO Process
Lan Huang
Thesis Supervisor: Jens Martin
September 2013 University of Amsterdam Amsterdam Business School Master in International Finance
2 Abstract
Based on the information asymmetry mechanism and the risk signal theory, this study investigates the relationship between legal expenses and IPO outcomes including underpricing and subsequent lawsuits to examine the impact of legal firms on IPO process. Analyzing samples of 7,246 offerings in all US public market from 1990 to 2012, we first find that legal firms are mainly compensated for their reputation, offer size and the complexity of the deal. Employing two-stage least square regression, we find that given the same offering size, complexity of the deal and industry risk, issuers can effectively reduce underpricing by paying larger fee to a legal firm of a given prestige. However, the impact of legal expenses on underpricing becomes insignificant after 2000; instead, legal expenses are significantly negatively correlated with the probability of subsequent lawsuits in the recession years including 2001-2003 and 2007-2008. These finding suggests the changing relationship between legal counsels and IBs and the changing role of legal counsels in the IPO process.
3
Table of Contents
1. Introduction ∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 4 2. Background and Literature Review∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 5
2.1 IPO Process∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 5 2.2 IPO Pricing Theory∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 6 2.3 Intermediaries Reputation Signal and Measurement of Reputation∙∙∙ 7 2.4 Other Risk Mechanism∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 8 2.5 Expert Compensation∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 9 3. Hypotheses and Methodology∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 10 4. Data ∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 14 5. Analysis of Results∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 15
5.1 Estimation of Legal Expenses ∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 15 5.2 Underpricing Observable Regression Result∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 17 5.3 Subsequent Lawsuit Observable Regression Result ∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 19 6. Conclusions ∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 20 References ∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 23 Figure and Tables∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 25 Appendix: Interview with Practicing Lawyer∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙ 43
4 Introduction
Private firms raise capital in a public equity market for the first time through an Initial Public Offering (IPO). During the IPO process, issuers have to engage experts including underwriters, accounting firms and law firms to prepare necessary legal document and market the IPO. Private firms prior to IPO are not required to disclose company information to the public. Therefore, there are information asymmetries among the participants. Market participants in the IPO process could be divided into two groups: informed participants and uninformed participants (Beatty et al., 1986). The IBs, accounting firms and law firms which are engaged to prepare legal documents and review financial data have access to detail information of the issuers prior to IPO, being classified as informed participants. Most of potential investors lacking of access to information of issuers are classified as uninformed participants. To make investment decisions, these potential investors rely on information disclosed in the preliminary prospectus and registration statement; moreover, they seek signals of the quality of issuers from these better-informed intermediaries such as their reputation (Heeley et al., 2007).
Issuers often face the trade-off between selecting the quality of intermediaries and the expert expense required. This thesis therefore aims to analyze how legal experts compensated in the IPO process and assess observable variables including the IPO underpricing and subsequent lawsuits to answer the following research questions:
Are the legal fee issuers pay well spent? Do issuers effectively signal their higher quality to the potential investors and reduce underpricing by engaging a more expensive law firm? Do issuers effectively reduce legal risks by engaging a more expensive law firm?
5
reputation, offering size, and the complexity of the case. After controlling these factors of legal expenses, we further find that issuers effectively reduce the underpricing by paying more legal expenses to legal firms especially before 2000. However, the relationship of legal expenses and underpricing become insignificant since 2001; instead, legal expenses are significantly reduce the probability of subsequent lawsuits in the recession years (2004-2007 and 2008-2009). This finding suggests a possible changing of the relationship between IBs and legal counsels in the IPO process after the dotcom bubble.
This study contributes to the current literature in four ways. First, we adopt a different ranking to measure the reputation of legal firms. It is based on the peer review in equity offering profession, thus represent the law firm’s reputation better than rankings based on total revenue or market shares. Secondly, we use legal expenses as main independent variable to provide a quantitative impact while current literature only examines the effect of reputation. Thirdly, we expand the current research on the factors of legal experts’ compensation by including the number of IBs involved as a proxy of the complexity level of the deal. Finally, we compare the effect of legal firms in hot periods and recession years and find an interesting turning point for both the factor of legal experts’ compensation and their impact on IPO outcome.
Background and Literature Review IPO Process
Private firms offer their ownership shares and raise capital in a public equity market for the first time through an Initial Public Offering (IPO). Required by The Securities Act of 1933, these issuers must register the offering securities with SEC. To prepare necessary legal document (the preliminary prospectus and registration statement) and market the IPO, issuers have to engage experts including underwriters,
6
accounting firms and law firms during the IPO process. The accounting firm is normally the first engaged expert, as early as three years before IPO, to perform financial statement audit and assure the issuers can obtain an unqualified audit opinion in the registration statement for IPO. It provides the underwriter the assurance of the issuer’s financial quality. The legal firms are engaged to ensure regulation compliance, represent the issuer and provide legal advisory to the issuer in the disclosure requirement of registration statement. The lawyers are usually the important drafter of the preliminary prospectus. Investment Banks (IBs) are engaged as underwriters to manage the offering and advise the issuers on the timing and pricing decision. In a “firm commitment” offering (97% of IPOs in US are firm commitment offerings), underwriters purchase all offering securities from the issuer and sell the securities to prospective investors at a specific price (offer price). Unlike the accounting firm and legal firms, the underwriters collect information of the issuer’s true value from prospective investors, insiders, auditors and legal firms to determine the offer price and perform marketing function of the IPO since they assume the risk of overpricing and low demand of the offering securities.
IPO Pricing Theory
Scholars has found an average initial IPO return of 16 percent by reviewing IPO transactions in US market in the 1960-1987 time period (Ibbotson et al., 1988). This large amount of return has attracted a great amount of literature examining the IPO process to rationalize the IPO underpricing.
Several authors have presented the key mechanism- information asymmetry- in IPO pricing. Private firms prior to IPO are not required to disclose company information to the public. There are information asymmetries among the participants. Baron (1980) argues that underpricing is due to the superior information of the
7
underwriters. Rock’s (1986) argument involves the “winner’s curse” theory that uninformed investors will win the deal (get allocated) most likely when the IPO is overpriced. Therefore, underpricing is to compensate the uninformed investors for the information asymmetry and induce them participate in the market.
Another related explanation proposed by Ritter in 1984 that the greater uncertainty of the offering securities’ true value, the greater cost the issuer has to pay (more underpricing) to induce the uninformed investors into the market. Megginson and Weiss (1991) also argue that underpricing is a compensation for increased risk. Underpricing could be explained as a function of risk involving in investing the IPO.
Intermediaries Reputation Signal and Measurement of Reputation
Market participants in the IPO process could be divided into two groups: informed participants and uninformed participants (Beatty et al., 1986). The IBs, accounting firms and law firms which are engaged to prepare legal documents and review financial data have access to detail information of the issuers prior to IPO, being classified as informed participants. Most of potential investors lacking of access to information of issuers are classified as uninformed participants. To make investment decisions, these potential investors rely on information disclosed in the preliminary prospectus and registration statement; moreover, they seek signals of the quality of issuers from these better-informed intermediaries such as their reputation (Heeley et al., 2007).
Existing empirical works have addressed that better investment banks are able to conduct diligent investigation of an issuer and have a greater influence on pricing (Carter & Manaster, 1990; Higgins & Gulati, 2003; Krishnan,Ivanov, & Masulis, 2011) and that better accounting firms are able to provide better audit quality and service to issuers and underwriters (Balvers et al.,1988; Michaely et al.,1995; Weber et al., 2003).
8
As far as the quality of intermediaries do matter in their performance in IPO process, the potential investors tends to interpret the involvement of reputational intermediaries as the signals of higher firm quality and lower investment risk(Hambrick et al., 2008). There are more studies further proved that the prestige of underwriters provide signals of the quality of issuers to investors by analyzing the extent of IPO under pricing (Chen et al., 2008); the prestige of accounting firms associated with less IPO underpricing also signals the quality of an IPO (Balvers et al.,1988; Beatty, 1989).
Regarding the main functions of law firms in the IPO process, issuers not only rely on these legal counsels to ensure their compliance with legal and regulation requirement, but also to protect their interest against other participants such as IBs and prospective investors. Barondes (2002) supports the argument that law firms perform a gatekeeper function in IPO by the empirical finding on the negative relationship between the pre-IPO price adjustment and the law firm’s investment. Barondes, Nyce and Sanger (2007) later on find a positive relation between the pre-IPO price adjustment and the participation of prestigious issuer’s law firm and argue that the reputation of issuer’s law firm signal the quality of the issuer. However, the reason of positive pre-IPO price adjustment has not been proved. It might because that prestigious law firm performs better gatekeeper function and solve more problems beforehand, leading to less price adjustment in the IPO process or that the prestigious law firm performs better job protecting issuer’s benefit against IBs. Besides, Beatty and Welch (1996) did not find significant relation between the prestige of issuer law firm and underpricing.
Other Risk Mechanism
9
another two signal mechanism: the participation of venture capital and the retention percentage of pre-IPO shareholders.
Megginson and Weiss (1991) examined the venture-backed and non venture-backed IPOs between 1983 through 1987 and find that the underpricing is significant lower of venture-backed IPOs. They argue that venture capitalists perform monitor over these issuers providing a certification of the issuer’s quality. The risk for uninformed investor is lower; hence the underpricing is lower.
Leland and Pyle (1977) suggest that pre-IPO shareholders with greater private information of the issuer’s true value play as insiders in the IPO process and aftermarket selling. The greater percentage of shares retained by pre-IPO shareholders can signal the better quality and less uncertainty of the issuer.
Expert Compensation
There are fewer academic studies on the compensation of IPO participating experts. As Beatty and Welch (1996) suggested, the compensation for underwriters are significantly related to the offering size (the total offering proceeds) and the underwriters’ reputation and the compensation for auditor is mainly related to the complexity of the case. Regarding legal expert compensation, they find that in addition to the offer size, issuers pay significantly more to prestigious lawyers. However, they find a surprising result that legal firms are also paid more while the firm discloses more uses of proceeds on the prospectus which is opposite to their hypothesis that the disclosure of uses of proceeds on the prospectus reduce the potential legal liability for issuers and engaging law firms, and thus reduced legal expenses to cover the risk.
10
Hypothesis and Methodology Hypothesis
Underpricing has traditionally asserted as a function of issuers’ risk: the greater the issuer’s risk, the greater the underpricing (Cohen et al., 2005). As some examination mentioned earlier show, the issuer’s law firms bear gatekeeper function in the IPO process and can be signal of issuer’s quality. Hence, we assume that by engaging more expensive legal firms, the issuers could effectively signal the higher quality and lower risk of the firm to the underwriters in determining offering prices and to potential investors so that there will be less underpricing. In addition, as the better law firms are able to perform due diligence more independently and require the issuer to disclosure potential risks properly, the potential legal liability for issuers should be reduced at the same time. Hence, we assume that by engaging more expensive legal firms, the issuers could effectively reduce the likelihood of subsequent lawsuits.
Hypothesis 1:
Issuers can reduce the IPO underpricing by engaging more expensive legal firms.
Hypothesis 2:
Issuers can reduce the subsequent legal risk by engaging more expensive legal firms.
Methodology
To accurately investigate the effects of legal advisor (legal fee) on IPO outcomes (underpricing and subsequent lawsuits), we adopt two stage regression method. First, we employ ordinary least squares regression to estimate legal expenses by controlling the difference of offering characteristics. Secondly, we use the estimated legal
11
expenses from first stage to test its impact on underpricing by least squares regression model and subsequent lawsuit by probit regression model. To examine whether law firms act differently in hot and recession markets, this study further divides data into five sets based on the economic status: 1990-2000, 2004-2007 and 2010-2012 are hot markets and 2001-2003 and 2008-2009 are recession markets.
First Stage Least Square Regression Legal Expenses Estimate
Extended from Beatty and Welch’s (1996) finding that legal firms are compensated for the reputation, the work load and risk proxies, we formulate the legal expenses as the result of the law firm’s reputation, the complexity of the deal, offer size, service market demand and firm specific risk.
Legal Expense =c + β1*Law Firm Prestige + β2*Number of IBs Involved + β3*Offer
Size + β4*Number of IPOs + + β*Industry Control + ε
As Beatty and Welch (1996) shows that prestigious law firms charge significant more fees, Issuer law firm prestige (LFP) is included. Other than using Vault Guide
to the Top 100 Law Firm in terms of profitability or revenue of law firms (Barondes et
al., 2007; Okamoto, 1995), this study measure law firm prestige by the peer review ranking in equity offering section in US published by The Legal 500. (http://www.legal500.com/c/united-states/finance/capital-markets-equity-offerings# table_2809) Law firms classified as top tier, second tier, third tier and fourth tier are given the score of 1, 2, 3, and 4 respectively, while those which are not included in the top four tier list are given score of 5.
The complexity of the deal indicates the work load for both IBs and legal firms
and it could be measured by the number of IBs involved in the IPO process. Offer
size is measured as the total offering proceeds. Service market demand is included
and measured as number of IPOs of the offering year since legal service is expected to be more expensive in hot market. Firm specific risk is controlled by SIC first two
12
digit code.
Second Stage Regression
After controlling factors which account for the inherent difference of legal expenses in the first stage, we can further examine whether engaging more expensive law firms matters in the IPO outcomes in the second stage regression. Tow observable IPO outcomes (dependent variables) in hypothesis 1 and hypothesis 2 are Underpricing and Subsequent Lawsuits respectively. Underpricing is measured as the percentage from the offer price to the first closing price in the secondary market (Certo, 2003). For Subsequent Lawsuits, issuers that have been named in federal class action securities fraud lawsuits since the passage of the Private Securities Litigation Reform Act of 1995 are coded as 1 and others as 0.
Hypothesis 1: Underpricing = c + β1*Estimated Legal Expense+ β2*IB
Prestige+β3*IB Prestige2+ β4*NASDAQ Change+ β5*Retention + β6*Venture-Backed
(dummy) + β*log (Offer Size Control) +β*Year Control + β*Industry Control +ε Hypothesis 2: Subsequent Lawsuit =c + β1* Estimated Legal Expense + β2* IB
Prestige + β3*Retention + β4* Venture-Backed (dummy) +β*log (Offer Size Control)
+β*Year Control + β*Industry Control +ε
The independent variables are similar to those used in the existing literature examining the effect of intermediaries’ prestige on IPOs. The second stage models are formulated based on the risk mechanism.
IBs assume the risk of overprice and low demand of offering shares. The
prestige of IBs (IBP) demonstrating different level of confidence in the IPOs signals
the quality of the issuers to the market and greatly influences the underpricing (Carter at el., 1998; Certo, 2003; Loureiro, 2010). Besides, IBs participate as principal gatekeepers to control IPO risk (Kraakman, 1984). Its better reputation may indicate better quality and smaller probability of subsequent lawsuits thus included as a
13
variable. We follow the methodology introduced by Carter & Manaster to measure the prestige of IBs based on the frequency an IB has been listed on the top of managing underwriters since the most prestigious underwriter usually appears first as advertisement. We use the Carter & Manaster rating updated by Loughran and Ritter (http://bear.warrington.ufl.edu/ritter/ipodata.htm). We apply five sets of rating to our data according to the time period: 1992-2000, 2001-2004, 2005-2007, 2008-2009 and 2010-2011. The more frequently an IB appears on the top, the higher rating assigned to it on a 0-9.001 scale. The IB prestige squared is also included to account for any non-linear relationship.
According to Leland and Pyle, the percentage of common stock retained by pre-IPO shareholders reveals the quality of the firm. It may affect the underpricing and the probability of subsequent lawsuit and thus included as a variable. Retention
ratio is calculated as the fraction of number of shares retained by pre-IPO
shareholders over total shares outstanding after IPO.
According to Megginson and Weiss (1991), venture capital fund perform influential monitor function in the IPO process and provide assurance of the issuer’s quality. The participation of venture capital is expected to signal better quality (reduce underpricing) and reduce the fundamental risk (reduce probability of subsequent lawsuit) thus included as a variable. Venture-backed IPOs are coded as 1 and others as 0.
In additional to risk factors, scholars also find that issuers are more underpriced in hot market (Ljungqvist, at el., 2006). Hence, we take the market status into account in the underpricing model to exclude the hot market timing effect. The market status is measured as the NASDAQ return over listing day and the year of IPO finished is included as control dummy. We also include the proceeds of the IPO (logged) and
14
hot business effect.
Data
The primary data used in this study are collected from the Thomson One database. All IPO transactions including all security types in all US equity markets for the time period from 1990 through 2012 are analyzed in underpricing model (hypothesis 1). In subsequent lawsuit model (hypothesis 2), only data in the time period from 1997 through 2012 are included since the Private Securities Litigation Reform Act was put into effect in 1995.
After deleting IPOs without legal expenses, we have 7,264 samples for the time period from 1990 through 2012. We use these 7,264 samples to analyze legal expenses in the first stage regression. To mitigate the effect of outliner, IPOs with extreme high underpricing (0.02% percentile) are excluded from the sample. After deleting 367 samples without complete underpricing information, a total of 6,864 samples are analyzed in the underpricing model. There are 3,837 samples for the time period from 1997 through 2012 analyzed in the subsequent lawsuit model.
Descriptive Statistics
Table 1 demonstrates the summary statistics for IPOs samples in this study. The mean of legal expenses is $574 thousands. As the Figure 1 shows that 71% of legal expenses are low (under $500 thousand). 15% are between $500 thousand and $1 million. Only 14% are over $1 million. Regarding average legal expenses of each year covered by this study, table 8 shows an obvious time trend. Legal expenses are relatively low during the time period from 1990 through 2000 and increases substantially since 2001. The increase is even more obvious after inflating the expenses to the 2012 level by Consumer Product Indexes.
15
The Legal 500 and their average legal expenses. There is increasing premium with
law firm’s prestige.
The mean of offering size (Total offering proceeds) is $138 million, larger than other literature shows. Main reason is that this study includes ADRs, close-end funds and REITS deals and their offer size are relatively large.
The mean of underpricing is 18.4 percent which is consist with the IPO data provided by Jay R. Ritter. The mean of venture-capital dummy is 0.33, meaning that there are 33% of the IPOs are venture-backed. The mean of pre-IPO shareholder retention is 46%, which is lower than other literature shows. This is due to the retention rate of close-end funds and REITS are zero, which lower the average retention rate.
Correlations between independent variables are relatively not significant as Table 2, Table 3, and Table 4 present, suggesting that the regressions adopted in this study is unlikely to be affected by multi-collinearity problem.
Analysis of Results
The results of the estimation of legal expenses are presented in Table 5, Table 6, and Table 7.
Estimation of Legal Expenses
Model 1 in Table 5 presents the results of first regression in this study to examine the factors of IPO lawyer compensation. The empirical data suggest a statistically significant correlation between legal expenses with issuer law firm prestige, IPO proceeds, number of IBs involved and the number of IPO of issuing year with p-value under 0.001.
The positive correlation between legal expenses and IPO proceeds supports the hypothesis that issuer with larger offering size are capable to afford more expensive
16
law firms. The positive correlation between legal expenses and number of IBs involved supports the hypothesis that law firms charge issuers based on the complexity of the IPO and the work they have to perform.
There is a significant premium for issuer law firm reputation. The coefficient -0.085 indicates that an issuer has to pay an extra 85 thousands to engage a law firm with one tier better ranking regardless of the offer size, complexity of the deal, and firm specific risk. Law firm’s reputation could be priced in the market.
Given the supply and demand pricing theory, we would expect that price for legal service is higher in a hot IPO market while the demand of legal services is higher. However, legal expenses are significantly negatively correlated to the number of IPOs of the issuing year. A possible explanation is that issuers have better bargaining power in a hot market, so legal firms reduce the fee level in order to participate in as many IPOs as possible to maximize their profit. The correlation between number of IPOs of offering year with both IPO proceeds and number of IBs involved are all negative, indicating that smaller and less complicate issuers are tend to go public while the market is hot. These inherent characteristics of IPOs in hot market could be another possible explanation of the negative relation between hot market and legal expenses.
There is an interesting time trend of legal expenses as Table 8 shows. Average legal expenses are relative greater after year 2001even after adjusting Consumer Product Indexes. To further examine the impact of different offering timing, a late period dummy defined as IPOs during the time period from 2001 through 2012 is added in the regression and shows as statistic significant (Model 2 in Table 5). The cut-off point of legal expenses is coincident with the post dot-com bubble period and the enactment of Sarbanes–Oxley Act of 2002. A possible explanation is that the law firms enhance their due diligence during the IPO process and accordingly raise the fee
17
in reaction to the increased regulatory scrutiny and legal risk change.
The empirical data also show that legal fee for the following industries are more expensive as the coefficients are positive with p-value lower than 0.05: Agriculture &Livestock(code11) , Metals & Mining(code12) , Petroleum Pipelines(code46), Communication Services(48), Finance( 62), Insurance(63), Real Estate(65). The adjusted R-square is increased from 0.29 to 0.32 after the industry control added to the regression, indicating that the firm specific risk is an influential factor to determine legal expenses.
Comparing results for Hot 1, Hot 2, Hot 3, Recession 1, and Recession 2 in Table 5, the relations between legal expenses and IPO proceeds, Law Firm Prestige, and number of IBs involved are all consistent with hypothesis. However, the factor- number of IBs involved becomes less significant in the recession years. Possible explanation is that only big companies go public in recession years. The complexity of the deal and work load for the law firms do not differ much. Law firms charge mainly depends on the offer size. Another finding is that offer size becomes insignificant in the time period of 2010 through 2012.
Underpricing Observable Regression Result
Model 2 in Table 6 shows there is a negative relationship between legal expenses and underpricing with statistical significance. The result of this variable is consist with the hypothesis that engaging more expensive legal firms signals better quality of the issuer. After employing two-stage regression to eliminate the effect of law firm reputation, offering size, the complexity of the deal and industry risk, the coefficient of legal expenses on underpricing decrease from -2.03 to -6.12 at the 1 percent confidence level. It indicates that after adjusting factors of legal fee, the effect of engaging more expensive law firm becomes even significant such that paying an extra
18
$1 million could decrease underpricing by 6.12% or paying extra $ 163 thousands, issuers could decrease underpricing by1 %.
The results of the relationship between IB reputation and underpricing are significantly negative as hypothesis and other literature suggest. The coefficient -9.69 means that IBs with 1 score higher rating can reduce underpricing by 9.69%. The prestigious underwriters well signal the quality of the issuer and reduce the underpricing extent.
The result for NASDAQ return over the offering day proves that first day return of IPO is in significantly positive relation to the market performance, indicating that market status is a determinant of underpricing. The timing of IPOs matters.
There is a significantly positive correlation between underpricing and venture capital backed IPO dummy. The coefficient 7.3 means that the venture capital backed IPOs are averagely with 7.3% more underpricing. This result is surprisingly against the hypothesis that venture capital funds perform monitor function on their investing companies and their involvement signal lower risk of the issuers. A possible explanation for this unexpected result is related to the positive correlation between the involvement of venture capital and the involvement of more prestigious IBs. The venture capitalists tend to engage prestigious IBs to ensure the success of IPO in sacrifice of offering price, hence the underpricing is greater. The risk signal function of venture capital is not influential.
The positive correlation between underpricing and retention percentage is consist with the hypothesis that pre-IPO shareholders play as insiders in the IPO process and the percentage of ownership retained signals the quality of the issuer. However, the correlation is not significant. Explanation might that pre-IPO shareholders retain more and sell fewer shares, therefore care less about underpricing when negotiating with IBs on the offering price. The risk signal function of insides retention is not
19
influential.
Examining results for Hot 1, Hot 2, Hot 3, Recession 1, and Recession 2 in Table 6, it is obvious that these risk signal variables explain underpricing well in the hot periods but do not work in the recession periods. The impact of IB prestige on underpricing is the only factor which stays consistent across these models, indicating that underwriters play as the most important intermediate in determining offer price. However, the negative relation between the IB prestige and underpricing is less significant in recession years as well. The possible reason is that in recession years, only relative larger firms with better quality go for IPOs and they tend to engage more prestigious IBs to ensure the success of IPO and better analyst coverage. More prestigious IBs have comparatively larger market share in recession year. As Hoberg (2003)’s finding that IBs with greater market share have greater bargaining power in determining the underpricing level, IB prestige on underpricing is less influential in recession years. Moreover, as Table 6 suggests, issuers are able to reduce underpricing effectively by engaging a more expensive law firm before 2000 but not effectively anymore afterwards. There is a possible role changing for law firms in the IPO process after the dot-com bubble. Before 2000, the issuer law firm performs a more important role helping issuer bargain offer price against underwriters. After dotcom bubble, these intermediates become more risk awareness and are likely to participate as a package such that IBs are often cooperate with certain legal counsels. Hence, the impact of law firms on underpricing becomes less influential.
Subsequent Lawsuit Observable Regression Result
The result of the probit regression for all data (model 1 and model 2) in table 7 shows that neither legal expenses, IB prestige, the involvement of venture capital or pre-IPO shareholder retention has significant influence on the probability of
20
subsequent lawsuits. The only significant variable is IPO proceeds. This result is supported by the practice that law firms file lawsuit based on the damage level. The larger the offer size, the higher probability of the happening of subsequent lawsuit.
We further compare results for Hot 1, Hot 2, Hot 3, Recession 1, and Recession 2 in Table 7and find that basically both the legal expenses and the IB prestige are negatively correlated with the probability of subsequent lawsuits in these models, but the relationships are significant only in the recession years. These results indicate that in recession years, underwriters and legal firms perform a better gatekeeper function in the IPO procedure and keep risky issuers from the market. Besides, the variable IPO proceeds maintains significant across all models in table 7, indicating that the offering size is the most important determinant of the probability of the happening of subsequent lawsuit.
Conclusion
This study has set to answer how are legal experts in IPO compensated by analyzing the factors of legal expenses and further to answer what are legal experts’ impact on the IPO process by examining the relationship between legal expenses and two observable outcomes including IPO underpricing and subsequent lawsuits.
First, I have found that legal expenses is positively correlated with IPO proceeds and number of IBs involved, and negatively correlated with law firm’s reputation with statistical significance after controlling industry risk. As Beatty and Welch (1996) suggested that legal experts are mainly compensated for offering size and reputation, this study further prove that legal experts are also greatly compensated for the complexity of the deal. However, this study also finds an unexpected result that there is no significant relationship between legal expenses and hot market demand, but legal expenses significantly increase after 2001 instead. The most likely
21
reason is that the legal scrutiny requirement increase after the dotcom bubble after 2000. This also enhances the hypothesis that law firms are compensated for the demanding workload. In sum, legal experts in the IPO process are compensated mainly for their reputation, offering size, complexity and workload of the case.
Secondly, this study has found that there is a negative relationship between legal expenses and underpricing at a statistical significant level, which is supporting the IPO pricing theory hypothesis and Barondes, Nyce and Sanger’s finding that the participation of reputational issuer legal counsel can reduce the underpricing. However, this study does not find significant impact of legal expense on the probability of subsequent lawsuits. However, the impact of legal expenses on underpricing becomes insignificant after 2001. This study also finds that legal expenses are significantly negatively correlated with the probability of subsequent lawsuits in the recession years (2001-2003 and 2008-2009). These findings might be able to answer the question that given the company risk, complexity and offer size equal, what are issuers actually paying for and what are law firms’ impact in the process. Before 2000, issuers are able to reduce underpricing effectively by paying more legal expenses while these legal firms do not influentially change the potential legal risk, which indicates that legal firms function more in line with issuers’ benefit by assisting issuers negotiate offering price against underwriters and its gatekeeper role is not as significant as regulation designed before 2000. However, the role of legal counsels changes after 2001. It is very likely that underwriters and legal counsels participate in IPO as a package because of the risk awareness after dotcom bubble, thus the role of legal counsel as a gatekeeper keeping risky issuers from the market becomes influential, especially in the recession years.
The limitation for this study is that the examined time period is relative long so that using current ranking for legal firms might not accurate enough to account the
22
impact of their reputation. Secondly, issuers and participating experts might act differently in hot periods and recession periods; therefore, using same variables based on risk theory might not able to account the factors of underpricing across time. Finally, the legal expenses examined in this study include all expenses paid by the issuer including expenses to the issuer’s legal counsel and to IB’s legal counsel. Thus the impact of representing legal counsels from each side might be mixed.
For further research, we suggest a further examination on the relationship between offer size and legal expenses especially for the most recent three years since there are likely to be several caps of fees for different ranges of offer sizes other than a mere linear relation between offer size and legal expenses. To better understand the impact of legal firms, we suggest a further examination on the relationship between IBs and legal firms. It is also valuable to further examine if legal firms stay on board in the subsequent offerings.
23 Reference
Balvers, R., McDonald, B., & R. Miller. 1988. Underpricing of new issues and the choice of auditor as a signal of investment banker reputation. The Accounting Review, 63: 605-622.
Beatty, R. P. 1989. Auditor reputation and the pricing of initial public offerings.
Accounting Review, 64:693-709.
Beatty, R. P., & Ritter, R. 1986. Investment banking, reputation, and the underpricing of initial public offerings, Journal of Financial Economics, 15: 213-232.
Barondes, R., Nyce, C. & Sanger, G. 2007. Underwriters' counsel as gatekeeper or turnstile: An empirical analysis of law firm prestige and performance in IPOs. Capital Markets Law Journal, 2: 164-190.
Barondes, R., 2002, Professionalism Consequences of Law Firm Investments in Clients: An Empirical Assessment, American Business Law Journal, 39:379-444. Baron, D., 1982. A model of the demand for investment banking advising and distribution services for new issues, Journal of Finance, 955-976.
Carter & Manaster, 1990. Initial Public Offerings and Underwriter Reputation. Journal of Finance, 45: 1045-1067.
Carter, R.B., Dark, F.H., & Singh, A.K. 1998. Underwriter reputation, initial returns, and the long-run performance of IPO stocks. Journal of Finance, 53: 285-311.
Certo, S.T. 2003. Influencing initial public offering investors with prestige: signaling with board structures. Academy of Management Review, 28(3): 432–446.
Chen, G., Hambrick, D.C., & Pollock, T.G. 2008. Puttin’ on the ritz: pre-IPO enlistment of prestigious affiliates as deadline-induced remediation. Academy of Management Journal, 51: 954-975.
Cohen, B.D. & Dean, T.J. 2005. Information asymmetry and investor valuation of IPOs: Top management team legitimacy as a capital market signal. Strategic Management Journal, 26: 683– 690.
Connelly, B.L., Certo, S.T., Ireland, R.D., & C.R. Reutzel. 2011. Signaling theory: a review and assessment. Journal of Management, 37(1): 39-67.
24
Journal of Financial Economics, 33: 3-56.
Heeley, M. B., Matusik, S. F., & Jain, N. 2007. Innovation, appropriability, and the underpricing of initial public offerings. The Academy of Management Journal, 50(1): 209–225.
Hoberg, 2003, Strategic Underwriting in Initial Public Offerings, Yale University Working Paper.
Ibbotson, Sindelar & Ritter. 1988. Initial Public Offerings. Journal of Applied Corporate Finance, 1: 37-45
Leland & Pyle. 1997. Informational Asymmetries, Financial Structure, and Financial Intermediation, Journal of Finance, 32 371-387
Ljungqvist, Nanda & Singh, 2006. Hot Markets, Investor Sentiment, and IPO Pricing. The Journal of Business 79(4)
Loughran & Ritter, J., 2004.Why Has IPO Underpricing Changed over Time? Financial Management, 33: 5-37
Loureiro, G. 2010. The reputation of underwriters: A test of the bonding hypothesis. Journal of Corporate Finance, 16(4): 516-532.
Megginson, W. L. & Weiss, K.A. 1991. Venture capitalist certification in initial public offerings. Journal of Finance, 46: 879-903.
Michaely, R. & Shaw, W.H. 1995. Does the choice of auditor convey quality in an initial public offering? Financial Management, 24(4): 15-30.
Okamoto, K.S. 1995. Reputation and the value of lawyers. Oregon Law Review, 74(15): 44-45.
Reinier H. & Kraakman, 1984. Corporate Liability Strategies and the Costs of Legal Controls, The Yale Law Journal, 93: 857–897
Ritter, J., 1984, The ‘hot issue’ market of 1980, Journal of Business 57, 215-240. Rock, 1986. Why New Issues Are Underpriced. Journal of Financial
Economics,15 :187-212
Weber, J. & Willenborg, M. 2003. Do expert informational intermediaries add value? Evidence from auditors in microcap initial public offerings. Journal of Accounting Research, 41: 681–720.
25
Figure 1: Legal expenses Distribution (in million)
This figure presents the distribution of the total 7,261 US IPO deals taken from the Thomson Financial’s SDC Platinum Equity database over a 23-year period from January 1, 1990 through December 31, 2012. Legal expenses smaller than 0.5 million (included) are classified as Low. Legal expenses range from 0.5 million (not included) to 1 million (included) are classified as Middle. Legal expenses greater than 1 million (not included) are classified as High.
26
Table 1 : Summary Statistics
This table reports the descriptive statistics for our sample of US IPO deals taken from the
Thomson Financial’s SDC Platinum Equity database. A total of 7,246 samples were made over a 23-year period from January 1, 1990 through December 31, 2012; among them, there is a total of 6,864 samples with complete information for the underpricing model. A total of 3,827 samples made over a 16-year period from January 1, 1997 through December 31, 2012 are analyzed in the subsequent lawsuit model.
Variable n Mean Min Max SD
Stage 1 :
Legal Expenses (million $) 7246 0.574 0.000 16.000 0.848
Issuer Law Firm Prestige ( ranking) 7246 4.466 1.000 5.000 1.201
IPO Proceeds (million $) 7246 138.083 0.013 18140.100 485.850
Number of IBs involved 7246 1.304 1.000 14.000 0.793
Number of IPO per year 7246 469.969 19.000 796.000 222.225
Stage2 underpricing model:
Underpricing (%) 6864 18.440 4.990 492.000 43.847
IB Prestige (Ritter rating) 6864 7.042097 0.000 9.001 2.407379
NASDAQ Return over day of IPO(%) 6864 0.590 -38.708 52.946 4.318
Retention (%) 6864 46.38417 0 99.93847 33.9066
Venture-backed (dummy) 6864 0.332 0.000 1.000 0.471
Stage2 subsequent lawsuit model:
Subsequent Lawsuit 3827 0.159 0.000 1.000 0.366
IB Prestige (Ritter rating) 3827 7.310745 0.000 9.001 2.404886
NASDAQ Return over day of IPO(%) 3827 0.516 -38.708 52.946 5.304
Retention (%) 3827 49.31345 0 99.75018 34.13597
27
Table 2 : Correlations
This table reports the correlations between independent variables in the stage 1 least square regression.
Stage1 Issuer Law Firm
Prestige IPO Proceeds
Number of IBs involved
Number of IPO offering year
Issuer Law Firm Prestige 1
IPO Proceeds -0.150 1
Number of IBs involved -0.229 0.302 1
Number of IPO offering year 0.123 -0.118 -0.408 1
Table 3 : Correlations
This table reports the correlations between independent variables in the stage 2 least square regression for the underpricing model.
Legal Expenses IB Prestige NASDAQ Return Retention Venture-backed
Legal Expenses 1
IB Prestige 0.286 1
NASDAQ Return -0.024 -0.018 1
Retention 0.138 -0.044 -0.005 1
28
Table 4 : Correlations
This table reports the correlations between independent variables in the stage 2 probit regression for the subsequent lawsuit model.
Legal Expenses IB Prestige NASDAQ Return Retention Venture-backed
Legal Expenses 1
IB Prestige 0.267 1
NASDAQ Return -0.011 -0.029 1
Retention 0.133 -0.015 -0.015 1
29
Table 5 : Result of Stage1 OLS Regression Legal Expenses Estimation
This study adopts two-stage regression to analyze the impact of legal expenses on underpricing and the probability of subsequent lawsuits. In first stage, an OLS regression is estimated for the legal expenses. This table presents first-stage OLS regression coefficients. The total 7,246 samples over a
23-year period from January 1, 1990 through December 31, 2012 are analyzed in Model 1 and Model 2. To examine the difference of legal
compensation over time, the data was further divided into five time periods based on the economic status as Hot 1, Recession1, Hot 2, Recession2 and Hot 3. Model 1 all Legal Expenses Model 2 all Legal Expenses Hot 1 1990-2000 Legal Expenses Recession 1 2001-2003 Legal Expenses Hot 2 2004-2007 Legal Expenses Recession 2 2008-2009 Legal Expenses Hot 3 2010-2012 Legal Expenses
IPO Proceeds (million $) 0.000305***
(0.000018) 0.000311*** (0.000018) 0.000384*** (0.000018) 0.000475*** (0.000084) 0.000742*** (0.000092) 0.001216*** (0.000322) 0.00006 (0.00005) Issuer Law Firm Prestige ( ranking) -0.084661***
(0.007278) -0.077887*** (0.007185) -0.047688*** (0.005931) -0.102737** (0.038189) -0.072329** (0.024038) -0.210026** (0.062892) -0.131924*** (0.037573)
Number of IBs involved 0.345832***
(0.012257) 0.286791*** (0.012724) 0.215682*** (0.026161) 0.214124* (0.092003) 0.279411*** (0.04239) 0.032367 (0.057167) 0.306765*** (0.03624)
Number of IPO offering year -0.000571***
(0.000042) -0.000044 (0.000055) -0.000144*** (0.000035) -0.005017 (0.006782) 0.002975* (0.001436) 0.007381 (0.012633) -0.001782 (0.008069)
Late Period( Dummy) - 0.431901***
(0.029312) - - - - -
Industry Controls Included Included Included Included Included Included Included
Constant 0.631685 (0.208788) 0.210716 (0.207685) 0.873138 (0.197506) 1.712531 (0.942975) 0.956034 (1.173104) 1.965832 (0.784008) 0.720968 (1.534616) N 7246 7246 5170 320 1098 111 547 R-squared 0.312538 0.33275 0.182092 0.45096 0.271493 0.52261 0.350012 Adjusted R-squared 0.305153 0.325488 0.17005 0.351319 0.227853 0.435345 0.269766
Numbers in parentheses are standard errors *** p<0.001 , ** p<0.01 , * p<0.05 , + p<0.10
30
Table 6 : Result of Least Square Regression-Underpricing Model
This table presents coefficients of underpricing model of OLS regressions and two-stage least square regression using the legal expenses estimated from first-stage regression. After excluding underpricing outliners and data without complete information, the total 6,864 samples over a 23-year period from 1990 through 2012 are analyzed in Model 1 and Model 2. To examine whether legal firms act differently over time, the data was further divided into five time periods based on the economic status as Hot 1, Recession1, Hot 2, Recession2 and Hot 3. (Numbers in parentheses are standard errors,*** p<0.001 , ** p<0.01 , * p<0.05 , + p<0.10)
All Hot 1 : 1990-2000
OLS(Model 1) Two-Stage Regression(Model 2) OLS(Model 3) Two Stage Regression(Model 4)
Legal Expenses (million $) -2.028712**
(0.752395) -6.114819** (1.987635) -3.368202* (1.58112) -17.07033*** (4.57017)
IB Prestige (Ritter rating) -11.10091***
(1.148796) -9.696844*** (1.116355) -12.99405*** (1.475773) -11.1781*** (1.434672) IB Prestige2 (Ritter rating) 0.909194***
(0.102554) 0.800011*** (0.099485) 1.067324*** (0.133606) 0.910882*** (0.129399)
NASDAQ Return over IPO day 0.698498***
(0.116442) 0.644894*** (0.113733) 0.887187*** (0.142827) 0.804898*** (0.139721) Retention (%) 0.056414*** (0.016881) 0.022033 (0.016441) 0.061153** (0.020051) 0.018013 (0.019643) Venture-backed (dummy) 9.946393*** (1.259644) 7.300594*** (1.227736) 10.7262*** (1.561742) 7.588995*** (1.524368) Log(IPO Proceeds) 5.017103*** (0.66684) 2.183595** (0.695165) 6.915742*** (0.91841) 4.101551*** (0.971505) Issuer Law Firm Prestige
( ranking)
1.275781**
(0.45486) -
2.680005***
(0.660342) -
Number of IBs involved -1.407996+
(0.825323) -
3.887233
(2.763041) -
Number of IPO offering year -0.024778***
(0.003299) -
-0.02266***
31
Late Period( Dummy) -21.44971***
(1.901987) - - - Year 1991 - -14.96775 - -17.75702+ Year 1992 - -15.48765+ - -18.63581+ Year 1993 - -12.76815 - -16.26878 Year 1994 - -14.13436 - -17.2607+ Year 1995 - -4.934783 - -9.244277 Year 1996 - -11.85438 - -16.6912 Year 1997 - -15.01989 - -19.01852+ Year 1998 - -5.681618 - -9.218878 Year 1999 - 29.5092** - 24.40198* Year 2000 - 22.21493* - 16.86288 Year 2001 - -8.885038 - - Year 2002 - -11.29397 - - Year 2003 - -11.4454 - - Year 2004 - -13.02751 - - Year 2005 - -12.96478 - - Year 2006 - -13.00114 - - Year 2007 - -13.67214 - - Year 2008 - -12.3476 - - Year 2009 - -14.59426 - - Year 2010 - -21.55512* - - Year 2011 - -19.85565* - - Year 2012 - -10.88613 - -
Industry Control Included Included Included Included
Constant 21.98581 33.26517 4.95637 45.92908
N 6864 6864 5063 5063
R-squared 0.118657 0.182557 0.126981 0.186069
32
Recession 1 : 2001-2003
Underpricing
Hot 2 : 2004-2007 Underpricing
OLS(Model 5) Two Stage Regression(Model 6) OLS(Model 7) Two Stage Regression(Model 8)
Legal Expenses (million $) -2.132544
(2.069732) -1.366245 (3.256863) -0.046796 (0.554766) -2.913342 (1.831149)
IB Prestige (Ritter rating) -16.87785*
(8.576028) -16.04873+ (8.611822) -3.522124* (1.621238) -3.680779* (1.623526)
IB Prestige2 (Ritter rating) 0.990913
(0.650609) 0.953925 (0.652123) 0.288663* (0.134657) 0.29747* (0.134425)
NASDAQ Return over IPO day -0.562672+
(0.331962) -0.530704 (0.329524) 0.729851** (0.229626) 0.738351** (0.230225) Retention (%) 0.111516+ (0.065794) 0.105427 (0.068783) 0.1658*** (0.03315) 0.165611*** (0.033026) Venture-backed (dummy) 5.182271 (5.304751) 3.823151 (5.242296) 9.356794*** (1.776285) 9.218809*** (1.77503) Log(IPO Proceeds) -0.674911 (2.278512) -0.843449 (2.293769) 2.495476** (0.882564) 3.231013** (1.049761) Issuer Law Firm Prestige
( ranking)
-0.998226
(1.459226) -
-0.409075
(0.454111) -
Number of IBs involved 1.987513
(3.548314) -
-1.121772
(0.821537) -
Number of IPO offering year 0.151384
(0.251809) -
0.000869
(0.027566) -
Late Period( Dummy) - - - -
Year 1991 - - - -
Year 1992 - - - -
Year 1993 - - - -
33 Year 1995 - - - - Year 1996 - - - - Year 1997 - - - - Year 1998 - - - - Year 1999 - - - - Year 2000 - - - - Year 2001 - - - - Year 2002 - -2.414295 - - Year 2003 - -4.437994 - - Year 2004 - - - - Year 2005 - - - -0.421699 Year 2006 - - - -1.417063 Year 2007 - - - -0.560159 Year 2008 - - - - Year 2009 - - - - Year 2010 - - - - Year 2011 - - - - Year 2012 - - - -
Industry Control - - Included Included
Constant 63.23204 77.32521 -6.163809 -8.538975
N 312 312 993 993
R-squared 0.075055 0.0718 0.21951 0.219812
Adjusted R-squared 0.044326 0.0441 0.171043 0.171364
Numbers in parentheses are standard errors
*** p<0.001 , ** p<0.01 , * p<0.05 , + p<0.10
34
Recession 2 : 2008-2009
Underpricing
Hot 3 : 2010-2012 Underpricing
OLS(Model 9) Two Stage Regression(Model 10) OLS(Model 11) Two Stage Regression(Model 12)
Legal Expenses (million $) 0.52286
(2.454803) 4.376843 (3.976866) -0.201072 (1.233754) -3.447588 (2.763524)
IB Prestige (Ritter rating) -4.209073
(6.741785) -2.619444 (6.602534) -8.501388*** (2.50971) -8.211927*** (2.468658)
IB Prestige2 (Ritter rating) 0.227688
(0.619617) 0.131213 (0.594622) 0.754897** (0.24819) 0.719165** (0.244379)
NASDAQ Return over IPO day -1.332602+
(0.697093) -1.239379+ (0.69263) 0.6295 (0.402487) 0.577711 (0.398719) Retention (%) 0.184554* (0.084028) 0.14652 (0.089628) -0.004952 (0.058176) 0.002854 (0.057405) Venture-backed (dummy) -4.949395 (7.062989) -5.415408 (6.805423) 5.455663 (3.496433) 5.333772 (3.452418) Log(IPO Proceeds) -0.74937 (2.714224) -3.280799 (3.046308) -1.113051 (1.668936) -0.175496 (1.64789) Issuer Law Firm Prestige
( ranking)
1.048343
(1.753425) -
0.427264
(0.866791) -
Number of IBs involved 0.682181
(1.67656) -
-0.45142
(1.056409) -
Number of IPO offering year 0.082625
(0.348273) -
0.270017
(0.201828) -
Late Period( Dummy) - - - -
Year 1991 - - - -
Year 1992 - - - -
Year 1993 - - - -
35 Year 1995 - - - - Year 1996 - - - - Year 1997 - - - - Year 1998 - - - - Year 1999 - - - - Year 2000 - - - - Year 2001 - - - - Year 2002 - - - - Year 2003 - - - - Year 2004 - - - - Year 2005 - - - - Year 2006 - - - - Year 2007 - - - - Year 2008 - - - - Year 2009 - 1.537764 - - Year 2010 - - - - Year 2011 - - - 1.591776 Year 2012 - - - 8.468982**
Industry Control - - Included Included
Constant 10.16766 23.81469 -20.03648 27.58042
N 91 91 405 405
R-squared 0.147239 0.154239 0.306838 0.320913
Adjusted R-squared 0.040644 0.071726 0.199893 0.218373
Numbers in parentheses are standard errors
*** p<0.001 , ** p<0.01 , * p<0.05 , + p<0.10
36
Table 7 : Result of Probit Regression- Subsequent Lawsuit
This table presents coefficients of subsequent lawsuit model of Probit regressions and two-stage probit regression using the legal expenses estimated from first-stage regression. The total 3,827 samples over a 16-year period from 1997 through 2012 are analyzed in Model 1 and Model 2. To examine whether legal firms act differently over time, the data was further divided into five time periods based on the economic status as Hot 1, Recession1, Hot 2, Recession2 and Hot 3. (Numbers in parentheses are standard errors,*** p<0.001 , ** p<0.01 , * p<0.05 , + p<0.10)
All Hot 1 : 1997-2000
Underpricing
Recession 1 : 2001-2003 Underpricing
Probit(Model 1) Two Stage
Probit (Model 2) Probit(Model 3)
Two Stage Probit
(Model 4) Probit(Model 5)
Two Stage Probit (Model 6)
Legal Expenses (million $) -0.017981 0.021685
(0.081378) -0.003438 (0.093193) -0.036878 (0.175918) -0.401838* (0.183599) -0.324929 (0.357206)
IB Prestige (Ritter rating) -0.074327 -0.062464
(0.059559) -0.019775 (0.104427) -0.023267 (0.104442) -0.312571 (0.449689) -0.306518 (0.464748)
IB Prestige2 (Ritter rating) 0.004839 0.003664
(0.00507) 0.002082 (0.008839) 0.002056 (0.008805) 0.012903 (0.033671) 0.011648 (0.034198) Retention (%) -0.000148 0.032883 (0.067374) 0.000014 (0.001131) 0.048379 (0.09329) -0.005293 (0.00421) -0.020111 (0.271847) Venture-backed (dummy) 0.01563 -0.000067 (0.00092) 0.049681 (0.094377) 0.000017 (0.001131) -0.023753 (0.266992) -0.005464 (0.003955) Log(IPO Proceeds) 0.142948 0.131524*** (0.033166) 0.057474 (0.05563) 0.069069 (0.057961) 0.117504 (0.106262) 0.051751 (0.117292)
Issuer Law Firm Prestige ( ranking) 0.073881 - 0.021275
(0.038086) -
0.070508
(0.072873) -
Number of IBs involved 0.0535 - 0.101294
(0.112712) -
-0.10379
(0.159252) -
37
Late Period( Dummy) -3.925451 - - - -
Year 1997 -1.076449 0.763278 - - - Year 1998 -2.23631 0.893251 0.108244 0.108166 - Year 1999 -1.554696 0.544553 0.10858 0.108055 - Year 2000 -2.565013 0.462437 0.124581 0.123864 - Year 2001 0.482599 0.830721 - - - Year 2002 0.444045 0.79076 - - -0.171382 0.218389 Year 2003 0.493234 0.930737 - - 0.003135 0.237894 Year 2004 1.235716 0.66064 - - - Year 2005 1.086031 0.639195 - - - Year 2006 0.99557 0.629067 - - - Year 2007 1.350527 0.642494 - - - Year 2008 -0.223908 0.479555 - - - Year 2009 0.304889 0.936634 - - - Year 2010 0.484808 0.535996 - - - Year 2011 0.338941 0.299527 - - -
Industry Control Included Included Included Included Included Included
Constant -2.925137 -2.015545 -0.512798 -0.30224 -5.071883 1.187565
N 3827 3827 1751 1751 320 320
Obs with Dep=1 609 609 273 273 70 70
38 Hot 2 : 2004-2007 Underpricing Recession 2 : 2008-2009 Underpricing Hot 3 : 2010-2012 Underpricing
Probit(Model 7) Two Stage
Probit (Model 8) Probit(Model 9)
Two Stage Probit (Model
10)
Probit(Model 11) Two Stage Probit (Model 12)
Legal Expenses (million $) 0.00578
(0.039278) -0.092601 (0.152871) -0.187988 (0.233907) -0.716104+ (0.401733) 0.020586 (0.071655) 0.239757 (0.160917) IB Prestige (Ritter rating) -0.128666
(0.12882) -0.108918 (0.127193) 0.511503+ (0.297159) 0.470656+ (0.285918) -0.097988 (0.154853) -0.15254 (0.156164) IB Prestige2 (Ritter rating) 0.010197
(0.010799) 0.00754 (0.010657) -0.046985 (0.03027) -0.046725+ (0.028737) -0.00255 (0.01563) 0.003203 (0.015786) Retention (%) -0.001739 (0.002623) -0.023911 (0.139995) -0.00588 (0.008094) 0.318294 (0.457985) 0.00362 (0.003643) 0.104121 (0.222759) Venture-backed (dummy) -0.056702 (0.143076) -0.001623 (0.002594) 0.390035 (0.469623) -0.004811 (0.00785) 0.122329 (0.221956) 0.00354 (0.003593) Log(IPO Proceeds) 0.307955*** (0.069343) 0.315116*** (0.090158) 0.341193* (0.170491) 0.488269* (0.240938) 0.173547* (0.07422) 0.20571** (0.076851) Issuer Law Firm Prestige ( ranking) 0.109247**
(0.037243) -
0.126656
(0.121198) -
0.073515
(0.065013) -
Number of IBs involved -0.029669
(0.0642) -
-0.012001
(0.075626) -
0.162898**
(0.061812) -
Number of IPO offering year - - - -
Late Period( Dummy) - - - -
Year 1997 - - - -
39 Year 1999 - - - - Year 2000 - - - - Year 2001 - - - - Year 2002 - - - - Year 2003 - - - - Year 2004 - - - - Year 2005 -0.068285 -0.070316 - - - - Year 2006 -0.065603 -0.05949 - - - - Year 2007 -0.073364 -0.058084 - - - - Year 2008 - - - - Year 2009 - - 0.319006 0.353802 - - Year 2010 - - - - -0.327816 -0.326515 Year 2011 - - - - -0.78225 -0.749449
Industry Control Included Included Included Included Included Included
Constant -2.28003 -1.753371 -3.422634 -2.44381 -1.797944 -1.330683
N 1098 1098 111 111 547 547
Obs with Dep=1 190 190 24 24 52 52
Pseudo R-squared 0.101618 0.092697 0.303832 0.30047 0.186614 0.171253
Numbers in parentheses are standard errors
*** p<0.001 , ** p<0.01 , * p<0.05 , + p<0.10
40
Table 8 : Number of IPO per year and Time trend of Legal Expenses
Year Number of IPO
Total Offer Proceeds(million $)
Average Legal Expenses per IPO($)
Adjusted Average Legal Expenses per IPO ($) 1990 19 641 341,053 600,253 1991 248 13,374 301,366 509,308 1992 564 39,667 279,593 458,533 1993 796 65,627 302,819 481,482 1994 560 36,707 319,727 495,576 1995 445 29,064 338,261 510,775 1996 792 70,644 330,565 482,625 1997 564 61,114 407,975 583,404 1998 327 35,410 402,852 568,021 1999 516 74,558 462,530 638,291 2000 349 57,696 539,957 718,142 2001 110 40,144 866,722 1,126,738 2002 116 25,986 713,310 913,037 2003 95 24,848 941,753 1,177,191 2004 283 66,802 993,833 1,212,477 2005 260 54,053 1,027,050 1,211,918 2006 247 55,007 1,000,232 1,140,265 2007 308 84,659 1,151,305 1,277,949 2008 47 10,425 1,151,489 1,232,094 2009 64 32,007 1,579,486 1,690,050 2010 175 43,475 1,360,149 1,428,156 2011 190 34,087 1,264,272 1,302,200 2012 186 48,533 1,366,418 1,366,418 This table presents the number of IPO, total offer proceeds and average legal expenses per IPO for our sample of total 7,261 US IPO deals taken from the Thomson Financial’s SDC Platinum Equity database over a 23-year period from January 1, 1990 through December 31, 2012. Adjusted Average Legal Expenses per IPO ($) is legal expenses adjusting Consumer Price Indexes announced by United States Department of Labor
41
Table 9 : Top Legal Firms List and Average Legal Expenses for Each Tier
This table presents the list of top four tiers legal firms in equity offering sector published by The Legal 500. IPOs engaged by legal firms other than firms listed above in ur total 7,261 US IPO deals taken from the Thomson Financial’s SDC Platinum Equity database over a 23-year period from 1990 through 2012 are classified as Other Legal Firms. The average legal expenses per IPO for each tier are calculated accordingly.
(http://www.legal500.com/c/united-states/finance/capital-markets-equity-offerings#table_2809)
Average Legal Expenses
(million)
Top Tier- ranking as 1 1.17
Cleary Gottlieb Steen & Hamilton LLP
Davis Polk & Wardwell LLP
Latham & Watkins LLP
Simpson Thacher & Bartlett LLP
Skadden, Arps, Slate, Meagher & Flom LLP
Second Tier- ranking as 2 0.98
Debevoise & Plimpton LLP Cravath, Swaine & Moore LLP
Fried, Frank, Harris, Shriver & Jacobson LLP
Gibson, Dunn & Crutcher LLP
Hogan Lovells LLP Kirkland & Ellis LLP Paul Hastings LLP
Paul, Weiss, Rifkind, Wharton & Garrison LLP Proskauer Rose LLP
Shearman & Sterling LLP Sidley Austin LLP Sullivan & Cromwell LLP Weil, Gotshal & Manges LLP
Third Tier - ranking as 3 0.85
Baker Botts L.L.P.
42 Covington & Burling LLP
Dechert LLP Goodwin Procter LLP
Morrison & Foerster LLP O'Melveny & Myers LLP Vinson & Elkins L.L.P. White & Case LLP
Fourth Tier - ranking as 4 0.69
Andrews Kurth LLP
Bracewell & Giuliani LLP DLA Piper
Jones Day Mayer Brown Morgan Lewis
Orrick, Herrington & Sutcliffe LLP Reed Smith LLP
Seward & Kissel LLP
Sutherland Asbill & Brennan LLP WilmerHale
Other Law Firms - ranking as 5 0.48
43
Appendix: Interview with practicing lawyer
A. The role of legal firms in the IPO process includes four key functions:
1. Advise company to get ready for IPO, including legal entity restructure, corporate governance and etc.
2. Draft prospectus 3. Perform due diligence
4. Protect client’s legal right against other participants in the transaction There are at least two sets of legal counsels participating in the IPO process with one representing issuer and one representing underwriting IBs. Sometimes, there is one more legal counsel representing shareholders. The issuer’s legal counsels perform all of the four functions mentioned above. The IB’s legal counsels verify the
prospectus drafted by issuer’s counsel, perform due diligence for IBs and protect IB’s benefit in the transaction.
From a practitioner’s view, investors do not pay as much attention to the law firm’s prestige as to the reputation of underwriting IBs. Besides, the priority goal for the issuers is to get the IPO deal done. There are only approximately 50-70 law firms capable for IPO engagement. The main role for legal counsels is to complete required procedures and does not significantly influence the IPO outcomes such as
underpricing.
IBs usually cooperate with certain legal counsels. Top banks are very likely to correlate to top legal firms. And law firm’s prestige stays consistent over time. B. Legal Expenses
The legal expenses increase significantly after 2000, mainly because of the risk awareness after the dotcom bubble and the accounting fraud of listed companies. In recession years, only more qualified companies go IPO. There should be a
