Venture Capital Syndication s Member Background, Organizational. Structure, and IPO Underpricing: Evidence from the GEM of China

Transcription

1 Venture Capital Syndication s Member Background, Organizational Structure, and IPO Underpricing: Evidence from the GEM of China Hao Xu 1* Difang Wan 1 Jin Xu 2 ( 1. School of Management, Xi an Jiaotong University, Xi an, China, Sloan Management School, Massachusetts Institute of Technology, Boston, USA, 02139) *Corresponding author at: Department of Organization Management, School of Management, Xi'an Jiaotong University; henry_hsu@126.com This work was supported by the NSFC (National Natural Science Foundation of China) under Grant No , No , and No About Authors Hao Xu is a Ph.D candidate at the Department of Organization Management, School of Management, Xi'an Jiaotong University, Xi'an, China. Difang Wan is a professor at the Department of Organization Management, School of Management, Xi'an Jiaotong University, Xi'an, China. Jin Xu is a post-doctor at Sloan Management School, Massachusetts Institute of Technology, Boston, USA.

2 Venture Capital Syndication s Member Background, Organizational Structure, and IPO Underpricing: Evidence from the GEM of China Abstract: Through manually gathering the data of venture capital syndication in firms listed in the GEM of China, we study the relation between syndication s member background, internal organizational structure and firm s IPO underpricing. This research shows that syndication scale, leading institution scale, and leading background scale can all effectively enhance firm s quality. As a leading institution or a leading background, the government cannot improve firm s performance. Foreign venture capital, however, is an effective investor, no matter whether it plays the role of a leading institution or a leading background. What is more, compared with the single background structure syndication, mixed background structure syndication makes more contributions to the firm. Keywords: Venture Capital; Syndication; IPO Underpricing; GEM; China 1. Introduction Venture capital investment originated from America in the 1940s, and its support for technological innovation activities is one of the main reasons of technological and economic progress (Gordon, 2002). In the venture capital industry, institutions tend to join together to form the syndication (Bygrave, 1987). In countries where financial markets are relatively developed, such as America and Canada, approximately 60% of the venture capital investment adopts the syndication mode. In Europe, about 30% of the the venture capital investment is made by syndication (Casamatta and Haritchabalet, 2007). Yet, in China this percentage is only about 20% and shows a downward trend (Wang Yuan, et al, 2010). The syndication of venture capital investment means two or more venture capital institutions, simultaneously or successively, provide the growth start-up firm with equity capital (Wilson, 1968), as well as management consulting and other value-added services in terms of corporate governance structure, overall business strategy, and human resource allocation (Hellmann and Puri, 2002; Davila and Foster, 2005). Venture capital syndication is the strategic alliance among venture capital institutions. Syndication s importance to the development of start-up firms lies in that

3 it helps different types of venture capital institutions realize resource complementation in the aspect of capital, human resource, and management experience, addresses the issue of insufficient resources confronting the single venture capital institution, and diversifies the potential risk for investors (Cochrane, 2005). Hochberg and Ljungqvist (2007) claim that venture capital syndication leads to better achievements for firms and investors. Venture capital institutions tend to form the syndication to create a network connection among institutions to share resources (Bygrave, 1987). Recently, syndication attracts the attention of scholars in the field of venture capital. A majority of studies mainly focus on the cause of syndication (Sahlman, 1990; Lerner, 1994), syndication contract structure (Hellmann, 2002; Cumming, 2008), syndication and firm s performance (Cochrane, 2005; De Clercq and Dimov, 2008), and syndication network structure (Sorenson and Stuart, 2001; Hochberg, et al, 2007, 2010). Venture capital institutions in the syndication may share different backgrounds -- private background, foreign background, and government background, and venture capital institutions with different backgrounds have distinct investment purposes and varied screen and regulatory effects on start-up firms. At the same time, few scholars have conducted the meticulous research regarding syndication s member background and organizational structure. We would like to make a contribution here. This paper is organized as follows. Section 2 provides the theoretical analysis. Section 3 describes variables and data. Section 4 presents empirical findings. Section 5 concludes the study. 2. Theoretical Analysis 2.1 IPO Underpricing and Venture Capital Syndication The issue of IPO underpricing is widespread in stock markets all over the world. Its reason is that investors fail to know the firm s true value because of asymmetric information and uncertainty, and the higher the degree of asymmetric information and uncertainty, the higher the degree of IPO underpricing (Beatty and Ritter, 1986; Rock, 1986). Among researches about venture capital s contribution to firm s quality, Wang, et al. (2003) believe that venture capital provides the start-up firm with not only funds,

4 but also value-added services, thereby alleviating the asymmetric information between investors and the firm. Megginson and Weiss (1991) find that the venture capitalist holding shares plays a good certification role, and at the time of IPO, the fact that venture capitalists hold certain proportions of shares sends investors a signal that the company is of relatively high quality. Syndication is an effective mode of venture capital investment, which helps increase the return for the start-up firm (Cumming, 2008). Guo and Jiang (2013) assert that compared with non-vc-backed or non-syndication-backed firms, syndication-backed firms have better performance and more R&D investment. Brander, et al. (2002) find that the average yield of syndication projects is 35% to 39%, which is higher than that of individual venture capital institution s project (15% to 20% on average). Therefore, the higher the level of syndication s involvement, the better the performance of the start-up firm. Additionally, the unity of venture capital institutions integrate the knowledge and resources of each venture capital institution. Venture capital institutions are highly heterogeneous in terms of resources and abilities. Hence, the more institutions joining the syndication, the higher the degree of resource diversity the syndication obtain, and eventually syndication could provide more effective information for the firm. 2.2 Leading Venture Capital Institution and Leading Venture Capital Background In the syndication, there is usually one venture capital institution playing the leading role (Wright and Lockett, 2003). The leading institution of syndication refers to the one which is most active and has the highest level of involvement. It initiates the syndication, invites other venture capital institutions to join the syndication, and coordinates them. In the investment stage, the leading institution makes more efforts to incubate the firm and projects (Meuleman, et al, 2009). Hence, the involvement level of the leading institution influences firm s quality, thereby affecting IPO underpricing. The leading institution of syndication mainly comes from three backgrounds: government background, private background, and foreign background. Government-backed venture capital institutions are strikingly different with

5 non-government-backed venture capital institutions with regard to the fund source and the investment purpose. Funds of government-backed venture capital institutions are mostly from state-owned enterprises or governments of different levels, and the fundamental purpose of government participation in venture capital investment is for the overall national progress in science and technology rather than huge returns (Gompers and Lerner; 2000). Nonetheless, non-government-backed venture capital institutions finance the start-up firm so as to get huge returns. Since they need to accumulate the industry reputation for the next round of investment and improve the possibility of firm s IPO to exit, they tend to make greater efforts to screen and monitor projects during the incubation period, which makes their certification effects stronger. Venture capital institutions with the same background have homogeneous fund sources and investment purposes. The involvement level of each background can be obtained by gathering venture capital institutions with the same background in the syndication. We define the background which has the highest level of involvement as a leading background. Like the leading institution, the leading background affects firm s quality and IPO underpricing as well. 2.3 Syndication Structure Syndication s member may share different backgrounds. The syndication constituted by venture capital institutions with the same background is defined as single background structure syndication, and the syndication constituted by venture capital institutions with different backgrounds is defined as mixed background structure syndication. Venture capital syndication embodies not just the number and scale of venture capital institutions, but also the advantage of resource complementation among venture capital institutions. In the mixed-structure syndication, the diversification level of members is higher and the advantage of resource complementation is more conspicuous. 3. Data and Variables 3.1 Data

6 This paper uses firms listed in the GEM of Shenzhen Stock Exchange before December 2013 as objects and studies venture capital institutions of various backgrounds in the syndication. It is our critical work to determine the venture capital syndication. The methodology we apply is as follows. (1) Determining venture capital institutions among firm s shareholders. (2) Selecting the firm with two or more shareholders that are venture capital institutions and defining this firm as a VC-syndication-backed firm. To determine the venture capital institution investing in the start-up firm, we manually distinguish the name of firm s top ten shareholders. We define shareholders whose names include Venture Investment, Entrepreneurship Investment, Venture Capital Investment, or Technology Investment as venture capital institutions. If these terms are not part of the shareholders names, we refer to the China Venture Capital and Private Equity Directory (2012 and 2013 edition) composed by Zero2IPO Database, and treat the company as a venture capital institution if it is recorded in the Directory. If the Directory does not record the institution, we utilize the search engine to search shareholder s official corporate website to check its main business, and identify it as a venture capital institution if it specializes in venture business. The shareholder will be excluded as an ordinary institutional investor, if it cannot be identified as a venture capital institution by the above means. Ultimately, we select firms with two or more shareholders that are venture capital institutions and define these firms as VC-syndication-backed firms. In this way, we finally get 184 VC-syndication-backed firm samples. To determine the background of venture capital institutions, we refer to the shareholder s Equity Property and divide the backgrounds of venture capital institutions into the government background, the Chinese private background, and the foreign background. Our data is from Wind Financial Terminal Database and Genius Financial Terminal Database, and we use Stata 10.0 to conduct the empirical analysis. 3.2 Variables

7 The IPO underpricing rate of the first day (underpricing_1st) is the main dependent variable in this paper, and its value is the rate of return on stocks on the first day of IPO. To ensure the robustness of the study, we also use five days IPO underpricing rate (underpricing_5th) as an alternative dependent variable, the value of which is the return rate of five consecutive trading days since IPO. This study designs four types of independent variables. (1) Syndication scale variables. The shareholding ratio of syndication (SynSR) is the sum of the shareholding ratio of all syndication members. Number of syndication s members (MemberNum) is the number of venture capital institutions in the syndication. (2) Leading institution variables. LeadInsSR is the shareholding ratio of the venture capital institution with maximum shareholding ratio in the syndication. GovLeadIns is the shareholding ratio of the government-backed leading institution. PriLeadIns is the shareholding ratio of the private-backed leading institution. ForLeadIns is the shareholding ratio of the foreign-backed leading institution. (3) Leading background variables. LeadBgSR is the shareholding ratio of the background with maximum shareholding ratio in the syndication. GovLeadBgSR is the shareholding ratio of government leading background. PriLeadBgSR is the shareholding ratio of private leading background. ForLeadBgSR is the shareholding ratio of foreign leading background. (4) Syndication structure variables. Syndications in 184 samples of this study comprise the syndication constituted by only government-backed venture capital institutions, the syndication constituted by only private-backed venture capital institutions, the syndication constituted by private-backed and government-backed venture capital institutions, and the syndication constituted by private-backed and foreign-backed venture capital institutions. The first two types are single background structure syndications, and last two types are mixed background structure syndications. GnGSyn, PnPSyn, PnGSyn, and PnFSyn indicate four kinds of syndications respectively.

8 In accordance with previous research experience and our research purpose, the control variables are as follows. (1) Dummy of high-tech firms (HitechDum), HitechDum=1 means the start-up firm is identified as a national, provincial, or municipal high-tech firm. (2) Stock offering size (OfferSize), its value is the logarithm of the product of issuing quantity and issuing price. (3) Time from the establishment of the firm to IPO (IPOTime), it is years the firm take to go public. (4) Debt asset ratio (DAR), its value is firm s debt asset ratio of the year before IPO. (5) Lot winning rate, it is the online issuing lot winning rate. 3.3 Summary Statistics Table 1 reports the results of summary statistics of syndication-backed firms in the GEM of China. The mean IPO underpricing rate of the first day (underpricing_1st) is 27.81%, which is lower than the average underpricing rate of the whole GEM (34.46% 1 ). This indicates that the syndication improves firm s quality and reduces firm s uncertainty and information asymmetry. Five days IPO underpricing rate (underpricing_5th) is 24.72% and this is lower than the IPO underpricing rate of the first day, which means that with time going on investors better understand the true value of the firm and the level of information asymmetry becomes lower. In 184 firm samples, the shareholding ratio of syndication (SynSR) is 26.94% on average. The maximum shareholding ratio of syndication is 99.99%, which shows that almost all of the shareholders of the firm are venture capital institutions. As for the number of syndication s member, every syndication consists of 3.26 venture capital institutions averagely. The mean shareholding ratio of the leading institution (LeadInsSR) and the mean shareholding ratio of leading background (LeadBgSR) are 16.12% and 24.60% respectively, which denotes that both the leading institution and the leading background have a relatively high involvement level in the syndication. 96% of syndication-backed firms are high-tech firms, and this percentage is higher than that of the whole GEM firms. Sample firms issue 676 million Yuan stocks on average and 1 The statistical data of the whole GEM is calculated by authors.

9 take 9.78 years to go public. The mean debt asset ratio (DAR) and the mean lot winning rate (LWR) are 38.65% and 1.13% respectively. INSERT TABLE 1 ABOUT HERE 3.4 Test Model According to the previous analysis, we use firm s IPO underpricing rate of the first day (underpricing_1st) and five days IPO underpricing rate (underpricing_5th) as dependent variables, and use indicators of syndication scale, leading institution, leading background, and syndication structure as explanatory variables to build the empirical model. Our basic regression model is: _ 1, _ 5, u n d e r p r ic in g s t u n d e r p r ic in g th N u m S R H ite c h D u m O ffe r S iz e IP O T im e D A R L W R Empirical Findings 4.1 Venture Capital Syndication Scale and IPO underpricing This study applies the cross-sectional data regression methodology to the analysis of data. To ensure the robustness of the empirical analysis, we use firm s first day IPO underpricing rate and five days IPO underpricing rate as dependent variables to perform the regression analysis. We find that the coefficients and significance levels of all independent variables and control variables are consistent in both two models. Hence, we mainly explain the results of the model with the dependent variable of first day IPO underpricing rate. Table 2 describes the effect of syndication scale on firm s IPO underpricing. In model 1 and model 2, both the shareholding ratio of syndication (SynSR) and the number of syndication s members (MemberNum) are negative at the 5% significance level. The two variables denote syndication scale and the above results show that the larger the scale of syndication, the higher the quality and the lower the uncertainty of the firm. Coefficients and significance levels of control variables are basically in line with previous studies. Dummy of high-tech firms (HitechDum) is significantly negative, because high-tech firms are supported by Chinese government and enjoy national tax incentives. Such firms have better prospects, compared with ordinary companies, and

10 therefore the their IPO underpricing rate is low. The lot winning rate (LWR) indicates investors sentiment. Its coefficient is negative at the 1% significance level, which means that passionate investors have a strong desire for investment, and as a result stock market shows irrational prosperity that ultimately gives rise to a high yield. INSERT TABLE 2 ABOUT HERE 4.2 Syndication s Leading Institution and IPO underpricing As is shown in table 3, the coefficient of the shareholding ratio of the syndication leading institution is negative at the 10% significance level. In the model with the dependent variable of five days IPO underpricing rate, although the coefficient is not significant, its significance level is very close to10%. This reveals that leading institution s involvement level increases firm s quality and decreases the uncertainty and information asymmetry. INSERT TABLE 3 ABOUT HERE Table 4 divides the backgrounds of leading institutions into government background, private background, and foreign background. In model 1 and model 2, government-backed leading institution (GovLeadIns) and private-backed leading institution (PriLeadIns) have no impact on the IPO underpricing rate. In model 3, the coefficient of foreign-backed leading institution s shareholding ratio (ForLeadIns) is significantly negative. Model 4 uses the shareholding ratio of all types of venture capital institutions to perform the analysis. We find that the coefficient of government-backed leading institution is still insignificant, and influence and significance levels of foreign-backed leading institution are highest. This indicates that, as the leading syndication institution, foreign-backed venture capital institution is the most efficient investor, because transnational venture capital institutions can expand the scale of venture capital, have stronger convening power, and overcome asymmetric information and agency problems (Sorenson and Stuart, 2001). INSERT TABLE 4 ABOUT HERE 4.3 Syndication s Leading Background and IPO Underpricing Syndication s leading background refers to the background which has the largest shareholding ratio. Like the leading institution, leading background also plays a role

11 of leading the syndication. Table 5 reports that the shareholding ratio of the leading background (LeadBgSR) is negative at the 5% significance level, and this denotes that with the involvement level of syndication s leading background increasing, firm s quality becomes better and uncertainty becomes lower. INSERT TABLE 5 ABOUT HERE Table 6 distinguishes the types of syndication leading backgrounds. In model 1, the coefficient of government leading background (GovLeadBgSR) is not significant. In model 2 and model 3, private leading background (PriLeadBgSR) and foreign leading background (ForLead BgSR) are significantly negative. Model 4 uses all types of leading backgrounds to conduct the analysis, and government leading background is still insignificant. Combined with table 4, we find that no matter whether the government plays the role of a leading institution or a leading background, it cannot effectively improve firm s quality. Because the government s investment purpose is not for profit, its project screening and supervision levels are low. Additionally, no matter whether foreign capital plays the role of a leading institution or a leading background, it can always enhance firm s quality and alleviate firm s uncertainty and information asymmetry. INSERT TABLE 6 ABOUT HERE 4.4 Syndication s Organizational Structure and IPO Underpricing Syndication structures of the sample objects consist of four types, syndication constituted by only government-backed venture capital institutions, syndication constituted by only private-backed venture capital institutions, syndication constituted by private-backed and government-backed venture capital institutions, and syndication constituted by private-backed and foreign-backed venture capital institutions. The first two types are single background structure syndications, and last two types are mixed background structure syndications. Table 7 presents the influence of syndication structure on the IPO underpricing. In single background structure syndications, the coefficient of the syndication constituted by only government-backed venture capital institutions (GnGSyn) is insignificant, and the coefficient of the syndication constituted by only private-backed

12 venture capital institutions (PnPSyn) is negative at the 10% significance level, because private-backed venture capital institutions seek to maximize their profits and are better engaged in supervision than government-backed venture capital institutions. As for mixed backgrounds structure syndications, the syndication constituted by private-backed and government-backed venture capital institutions (PnGSyn) and the syndication constituted by private-backed and foreign-backed venture capital institutions (PnFSyn) are negative at the 10% and 5% significance levels respectively, and their absolute values are larger than those of single background structure syndications. This shows that in mixed background structure syndications, syndication members have a high degree of diversification and the advantage of resource complementation is more significant. The mixed background structure syndication can effectively improve firm s quality and reduce its uncertainty. Particularly, it can be noticed that the contribution of the syndication constituted by private-backed and government-backed venture capital institutions is far greater than that of the syndication constituted by only government-backed venture capital institutions or that of the syndication constituted by only private-backed venture capital institutions. INSERT TABLE 7 ABOUT HERE 4.5 Robustness Analysis First, this paper uses the IPO underpricing rate of the first day and five days IPO underpricing rate as dependent variables to build models, and finds that coefficients and significance levels of all independent variables and control variables are consistent in both two models. This has, to some extent, ensured the robustness of the results. Moreover, we use the winsorize method to adjust singular values and perform the regression analysis again, and eventually find no significant differences. 5. Conclusion We manually gather the data of venture capital syndication in firms listed in the GEM of China, and study the influences of syndication s scale, leading institution scale, leading background scale, and internal organizational structure on firm s IPO underpricing. We find that (1) the overall syndication scale, leading institution scale,

13 and leading background scale can all effectively improve firm s quality; (2) no matter whether the government plays the role of a leading institution or a leading background, it cannot enhance firm s quality or ease its uncertainty; (3) no matter whether foreign capital plays the role of a leading institution or a leading background, it is always an effective investor; (4) when it comes to syndication s structure, mixed background structure syndications contribute more to the firm than single background structure syndications. In the theoretical aspect, this paper contributes to the literature on syndication s member background and organizational structure. In the practical perspective, managers of start-up firms are supposed to raise the venture capital of syndication mode, and consider the effectiveness of venture capital investment in terms of syndication s member background and organizational structure. In addition, the empirical findings show that the practice is effective that Chinese government in recent years vigorously guides the transformation of venture capital structure and encourages non-government-backed venture capital institutions to fund the start-up firms.

14 References [1]. Wang Yuan, Zhang Xiaoyuan, and Liang Gui China Venture Capital Development Report. Economy& Management Press. (in Chinese) [2]. Beatty R P, and Ritter J R, 1986, Investment Banking, Reputation, and the Underpricing of Initial Public Offerings, Journal of financial economics, 15(1), [3]. Brander J A, Amit R, and Antweiler W, 2002, Venture-Capital Syndication: Improved Venture Selection vs. the Value-Added Hypothesis, Journal of Economics & Management Strategy, 11(3), [4]. Bygrave W D, 1987, Syndicated Investments By Venture Capital Firms: A Networking Perspective, Journal of Business Venturing, 2(2), [5]. Casamatta C, and Haritchabalet C, 2007, Experience, Screening and Syndication in Venture Capital Investments, Journal of Financial Intermediation, 16(3), [6]. Cochrane J H, 2005, The Risk and Return of Venture Capital, Journal of financial economics, 75(1), [7]. Cumming D, 2008, Contracts and Exits in Venture Capital Finance, Review of Financial Studies, 21(5), [8]. Davila A, and Foster G, 2005, Management Accounting Systems Adoption Decisions: Evidence and Performance Implications from Early-stage/startup Companies, The Accounting Review, 80(4), [9]. De Clercq D, and Dimov D, 2008, Internal Knowledge Development and External Knowledge Access in Venture Capital Investment Performance, Journal of Management Studies, 45(3), [10]. Gompers P A, 1996, Grandstanding in the Venture Capital Industry, Journal of Financial economics, 42(1), [11]. Gordon R J, 2002, Technology and economic performance in the American economy, National Bureau of Economic Research. [12]. Guo D, and Jiang K, 2013, Venture Capital Investment and the Performance of Entrepreneurial Firms: Evidence from China, Journal of Corporate Finance, 22, [13]. Hellmann T, 2002, Theory of Strategic Venture Investing, Journal of Financial Economics, 64(2), [14]. Hellmann T, and Puri M, 2002, Venture Capital and the Professionalization of Start-Up Firms: Empirical Evidence, The Journal of Finance, 57(1), [15]. Hochberg Y V, Ljungqvist A, and Lu Y, 2007, Whom You Know Matters: Venture Capital Networks and Investment Performance, The Journal of Finance, 62(1), [16]. Hochberg Y V, Ljungqvist A, and Lu Y, 2010, Networking as a Barrier to Entry and the Competitive Supply of Venture Capital, The Journal of Finance, 65(3), [17]. Lerner J, 1994, The Syndication of Venture Capital Investments, Financial management, [18]. Megginson W L, and Weiss K A, 1991, Venture Capitalist Certification in Initial Public Offerings, The Journal of Finance, 46(3), [19]. Meuleman M, Wright M, Manigart S, et al, 2009, Private Equity Syndication: Agency Costs, Reputation and Collaboration, Journal of Business Finance & Accounting, 36(5-6), [20]. Rock K, 1986, Why new issues are underpriced, Journal of financial economics, 15(1), [21]. Sahlman W A, 1990, The Structure and Governance of Venture-Capital Organizations, Journal of financial economics, 27(2), [22]. Sorenson O, and Stuart T E, 2001, Syndication Networks and the Spatial Distribution of Venture Capital Investments, American Journal of Sociology, 106(6), [23]. Wang C K, Wang K, and Lu Q, 2003, Effects of Venture Capitalists Participation in Listed Companies, Journal of Banking and Finance, 27(10), [24]. Wilson R, 1968, The Theory of Syndicates, Econometrica, 36,

15 [25]. Wright M, and Lockett A, 2003, The Structure and Management of Alliances: Syndication in the Venture Capital Industry, Journal of Management Studies, 40(8),

16 Table 1 Summary statistics: Syndication-backed firms of China s GEM. Obs Mean Min Max underpricing_1st (%) underpricing_5th (%) SynSR (%) MemberNum Lead InsSR (%) Lead BgSR (%) HitechDum OfferSize (100 million yuan) IPOTime (Year) DAR(%) LWR(%) Table 2 Syndication Scale and IPO Underpricing Dependent Variable: underpricing_1st Dependent Variable: underpricing_5th Model 1 Model 2 Model 3 Model 4 SynSR ** ** ( ) ( ) MemberNum ** ** ( ) ( ) HitechDum ** ** *** *** ( ) ( ) ( ) ( ) OfferSize ( ) ( ) ( ) ( ) IPOTime (0.6953) (1.0324) ( ) (0.2223) DAR ( ) ( ) ( ) ( ) LWR *** *** *** *** ( ) ( ) ( ) ( ) Constant *** ** *** *** (2.6067) (2.5311) (2.9812) (2.8964) Obs Adj R The t-values are in parentheses. ***, **, and * represent statistical significance at the 1%, 5%, and 10% levels, respectively. The same below.

17 Table 3 Leading Institution and IPO Underpricing Dependent Variable: underpricing_1st Dependent Variable: underpricing_5th LeadInsSR ( )* ( ) HitechDum ( )** ( )*** OfferSize ( ) ( ) IPOTime (0.5292) ( ) DAR ( ) ( ) LWR ( )*** ( )*** Constant (2.6784)*** (3.0428)*** Obs Adj R Table 4 The Background of Leading Institution and IPO Underpricing Dependent Variable: underpricing_1st Dependent Variable: underpricing_5th Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 GovLead Ins PriLead Ins ForLead Ins Hitech Dum Offer Size IPOTime DAR LWR Constant (0.1703) ( ) ( ) * ( ) ( ) * ** ( ) ( ) *** ( ) ** (-2.479) ** ( ) ** ( ) ( ) ( ) ( ) ( ) (0.5493) (0.5329) (0.3280) (0.2582) ( ) ( ) ( ) ( ) -9.33*** -8.34*** -8.93*** -7.42*** ( ) ( ) ( ) ( ) 147.2*** 143.3*** 154.4*** 150.6*** (2.7065) (2.6458) (2.8634) (2.8013) *** ( ) ( ) ( ) ( ) -8.81*** ( ) 169.3*** (3.0767) ( ) ( ) ( ) ** ** ( ) ( ) *** *** *** ( ) ( ) ( ) * ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) -8.14*** -8.35*** -7.14*** ( ) ( ) ( ) 166.2*** 176.6*** 174.1*** (3.0243) (3.2444) (3.1973) Obs Adj R

18 Table 5 The Shareholding Ratio of Syndication s leading background and IPO Underpricing Dependent Variable: underpricing_1st Dependent Variable: underpricing_5th LeadBgSR ( )** ( )* HitechDum ( )** ( )*** OfferSize ( ) ( ) IPOTime (0.6305) ( ) DAR ( ) ( ) LWR ( )*** ( )*** Constant (2.6873)*** (3.0500)*** Obs Adj R Table 6 Syndication Leading Background and IPO Underpricing Dependent Variable: underpricing_1st Dependent Variable: underpricing_5th Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 GovLead BgSR (0.4960) ( ) (0.3026) PriLead * * BgSR ( ) ( ) ForLead * ** BgSR ( ) ( ) Hitech *** ** *** ** *** Dum ( ) ( ) ( ) ( ) ( ) OfferSize ( ) ( ) ( ) ( ) ( ) IPOTime (0.5096) (0.6063) (0.3078) (0.3253) ( ) DAR ( ) ( ) ( ) ( ) ( ) LWR -9.38*** -8.31*** -8.91*** -7.57*** -8.84*** ( ) ( ) ( ) ( ) ( ) Constant *** *** *** *** *** (2.7026) (2.6469) (2.9090) (2.8512) (3.0662) ( ) * ( ) ( ) ** ** ( ) ( ) *** *** *** ( ) ( ) ( ) * ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) -8.00*** -8.28*** -7.09*** ( ) ( ) ( ) *** *** *** (3.0186) (3.3236) (3.2720) Obs Adj R

19 Table7 Syndication Structure and IPO Underpricing Dependent Variable: underpricing_1st Dependent Variable: underpricing_5th GnGSyn (1.0643) (0.4022) PnPSyn ( )* ( ) PnGSyn ( )* ( ) PnFSyn ( )** ( )* HitechDum ( )** ( )*** OfferSize ( ) ( ) IPOTime (0.6380) ( ) DAR ( ) ( ) LWR ( )*** (2.9177)*** Constant (2.5314)5** ( )*** Obs Adj R