Venture Capital Participation and the Performance of Chinese IPOs. Isaac Otchere* and Anna Vong

Venture Capital Participation and the Performance of Chinese IPOs Isaac Otchere* and Anna Vong Abstract Prior studies have identified the value-added potential of venture capitalist to their portfolio companies in the initial public offerings, the post-issue operating performance and the long-run stock market performance. We empirically examine this phenomenon in China, where venture capital (VC) has become an increasingly important source of funding for small and medium-sized enterprises. In this study, we do not find support for the certification role of the VC at the time of the initial public offerings (IPOs), as the underpricing level of the VC backed group is comparable to that of the non-vc backed group, and the IPO cost for the VC backed group is significantly higher than for their counterparts. By comparing the stock market performance and the operating performance of VC backed IPOs with a control sample of non-vc backed IPOs from 1991 through 2008 in China we find that the presence of VC improves both the long-term stock performance and the post-ipo operating performance. The post-issue performance differential is observed for the early years of the IPOs, when the value-added VC monitoring is present. Nevertheless, additional analysis indicates that VC reputation is associated with lower underpricing level, suggesting that reputable and, presumably, experienced VCs use their expertise and influence to reduce underpricing. Keywords: Venture capital; IPO; China Corresponding author: isaac_otchere@carleton.ca. Otchere is at the Sprott School of Business, Carleton University, Canada. Anna Vong is at the University of Macau, Macau. Email:annavong@umac.mo 1

1. INTRODUCTION The on-going economic reforms in China have gradually shifted the economy away from complete reliance on state-owned and collective enterprises to a mixed economy in which private enterprises play an important role both in promoting economic growth and creating employment opportunities. According to the 2006 Blue Book on Private Economy Development published by All-China Federation of Industry and Commerce (ACFIC) on September 22, the share of private businesses in the country s gross domestic product (GDP) first exceeded 50% in 2005 and has been expanding. The private sector accounted for 60% of fixed assets investments, while the state and the foreign enterprises were responsible for 30.6% and 9.4%, respectively. The number of private businesses has also grown from 3 million in 2004 to nearly 52 million in 2011, and the number of employees recruited by the private sector has skyrocketed from 47 million to 160 million in 2011. 1 All this demonstrate how important the private sector is in China s rapid economic growth. Since the late 1990s, the government has lifted up many legal and economic barriers 2, which hinder private sector growth and has, at the same time, introduced a number of measures to promote the availability of bank loans to the private sector. Despite these efforts, Brandt and Li (2003) and Ge and Qiu (2007) have noted that private and privatized firms have been discriminated against the formal loan market, and to expand and grow these firms have to rely on informal and expensive funding sources like trade financing. Allen et al. (2005) found that family connections and the personal reputation of the entrepreneurs are common ways for private enterprises in China to obtain financial capital in the form of trade credits and private credit. The difficulty of obtaining bank loans, coupled with the increasing demand for financial capital drives many private businesses in China, mostly small and medium-sized enterprises (SMEs) to seek alternative source of capital financing. Long-term equity financing from venture capitalists is an important source of long term finance. The importance of venture capital in contributing to the success of young firms has grown significantly, both in developed and in emerging 1 China s growth: Planning or Private Enterprise? Library of Economics and Liberty, August, 2012. 2 According to Firth et al. (2009), during the 15 th Congress of the Chinese Communist Party in 1997, approval was granted to banks to lend to private businesses, and in order to protect private property right, a constitutional amendment was approved in 2004, stating that gives equal status as public property. 2

economies. In recent years, venture capital investments worldwide have increased from 10% in 1991 to 22.7% in 2008 (Chemmanur, 2010). The growth of international venture capital investment in emerging economies rose from 8.7% in 1991 to 56% in 2008. Representing one of the world s fastest growing markets, China has also recorded a tremendous growth in venture capital investment. According to the China Venture Capital Research Institute (CACRI), the total amount of venture capital in China has grown from US$4.76 billion in 2003 to US$36.67 billion in 2008. The average amount of venture capital per institution in 2008 was US$191.96 million, which was 1.55 times what it was in 2007. Despite the sub-prime mortgage crisis in the U.S. and the slowdown of global economic growth, the venture capital industry in China still has achieved a considerable growth. Venture capitalists usually provide financing in stages corresponding to the development of the portfolio companies. Staged financing allows VCs to reassess their investments periodically and to withdraw if the net present value turns negative. Sahlman (1988) concludes that such financing practices motivate entrepreneurs to pursue behaviors that will maximize the value of their companies. Aside from contracting, VCs with concentrated equity positions in funded companies behave like large stockholders and are actively involved in managing and monitoring their portfolio companies. Bygrave and Timmons (1992) document that the value-added services provided by VCs range from coaching and guidance, networking for strategic alliances to attracting further capital. In addition, by retaining a significant ownership position after the investee firms become public, VCs can continue their monitoring role which further accelerates the growth of the companies. Although the role of venture capitalists in certifying and monitoring new IPO firms has been examined extensively in developed countries, relatively few studies have been conducted in emerging economies. One probable reason could be the weak regulatory system found in emerging markets, which leads to the slow development of the VC industry, and perhaps more importantly, the scarcity of data. Most of the IPO-VC based studies tend to be descriptive and very often surveys or case studies are used. Results are thus biased towards voluntary disclosure by the VCs. In China, the establishment of the Chinese Venture Capital Associations in 2002, which aims to promote the rights of entrepreneurs and laws on investor protection and disclosure, 3

shows that the venture capital industry is relatively young. Added to this regulatory development is the launch of the Small and Medium Enterprise Board (SME) in the Shenzhen Stock Exchange in 2004, which provides a valuable exit channel for venture capitalists. Since listed companies are obliged to disclose their backgrounds, operating, and financial data to the public, the VCs influence on them can thus be examined and explored extensively. Venture capital investments have grown over the past decade. It is thus an appropriate time to study and to understand the function of venture capital in developing their portfolio firms in China which represents one of the world s fastest growing markets for venture capital investment. In particular, the study aims to shed light on the impact of VC participation on the short and long-term financial performances of the investee firms, as well as the perception of investors of these mechanisms in China, a country with different institutional features that what is known in the Western countries. The investigation of the impact of VCs participation in investee firms in this emerging market can provide not only insights for practitioners and policy makers who have continuously amended and improved the legislation 3 to support the development of the VC industry, but can also extend the literature in an environment beyond the United States where the value-added potential of VC has been widely acknowledged. In this study, we first investigate the certification role of venture capitalists in Chinese IPO companies which were listed from 1991 to 2008 in China. If the certification provided by VCs is valuable, the extent of underpricing should be limited. To examine this, we employ the matched pairs methodology where each VC backed IPOs is matched with a non-vc backed IPO in the same industry and with a similar offering size. Furthermore, as most VCs may continue to hold significant equity positions after an IPO, the monitoring function should positively impact both the post-ipo operating performance and long term market performance. In addition to comparing the post-ipo operating and market performance for the two groups of IPO firms, cross-sectional regression analysis is conducted to ascertain whether the superior performance of VC backed IPOs, if any, is related to VC participation or to factors such as IPO offer size, offer price, age of an IPO firm, etc., which have been documented in the extant IPO literature. 3 Various favorable grants have been provided for venture capital investment in Shenzhen as well as in Guangzhou, and in order to encourage international venture capitalists, entry relaxation has been permitted by the Ministry of Foreign Trade and Economics. 4

Using a sample of 120 VC-backed IPOs and 120 matched non VC-backed IPOs, we find from the univariate analysis that underpricing of VC-backed IPOs is not statistically different from that of non-vc-backed IPOs. However, in a more econometrically robust regression analysis, we find that the presence of the VC leads to higher underpricing. The VC-backed IPOs outperform non-vc backed companies in the first year. Consistent with the findings of existing literature, we find that in China, both VC-backed and non VC-backed IPOs underperform in the long run. However, VC-backed IPOs after-market performance is better than that of non-vc backed IPOs; the difference in performance is as high as 18% for the first year. Nevertheless, the importance of VC backing seems to vanish after the first year of IPO, as beyond that time, the difference between the two groups is not statistically significant. The cross sectional regression also shows that the presence of a VC in the IPO firm significantly improves the stock market performance. The finding of an improved long-run stock performance is consistent with the previous results on post-ipo operating performance that VC-backed IPOs are better. Consistent with prior studies, we also observe that underpricing is negatively related to the long-term stock market performance. The coefficient of the underpricing variable shows that the higher the underpricing, the lower its three-month return; however, after the three months the effect of initial return on long-term performance vanishes. Other findings worthy of note are that the average investment duration of venture capitalists in their investee firms is 2.7 years, which is longer than the 2.1 years reported in Singapore. The mean VC equity stake holding after listing is 15.79% of the firm which demonstrates a fairly high institutional holding. With such a high ownership level, venture capitalists, in their own self-interest, would closely monitor their portfolios and such a monitoring role is crucial to the performance of the Chinese VC backed companies, as reflected in the better operating and financial performance. Both the post-ipo operating performance and long term after-market stock performance of VC backed companies are superior to that of non-vc backed companies. Nevertheless, the certification role of VCs is not supported because the underpricing level of the VC backed companies is comparable to that of the non-vc backed ones and, instead of having a lower IPO cost due to VC certification, the VC backed sample has higher cost. We also find that both VC-backed and non-vc-backed firms in China go public around the same time (5) years. This finding is not consistent with Megginson and Weiss (1999) who find that VC-backed IPOs go public at a significantly early stage in their life cycle than non VC-backed IPOs. 5

Further analysis of the VC-funded IPOs indicates that VC reputation is associated with lower underpricing, suggesting that reputable and, presumably, more experienced VCs use their expertise and influence to reduce underpricing. Both VC and CEO controlling power affect underpricing i.e., the larger the VC and CEO holdings in the pre-ipo period, the lower the underpricing. This finding is consistent with the reputational capital theory, which asserts that VCs with large reputational capital at stake would use their expertise and experience to minimize underpricing in order to preserve their reputational capital. Also, VC investment duration is positively related to underpricing. This finding stands in sharp contrast to the grandstanding hypothesis especially in light of the finding that VC investment duration in China is longer than has been documented in other countries such as Singapore. The remainder of the paper is structured as follows: Section 2 presents a review of the relevant literature and our hypothesis. In section 3, we discuss the developments and trends that have affected the development of the VC industry in China. Section 4 describes that data and the methodology. Section 5 presents the empirical findings. In Section 6, we present further analysis of the performance of VC-backed IPOs. Section 7 concludes the paper. 2. Literature review and hypothesis development A myriad of research has been done on the post issue performance of IPOs. The literature has shown that after the initial burst on the first trading day (leading to perceived underpricing), the aftermarket returns of the average IPO have been sub-par. Ritter (1984) and Carter and Manaster (1990) among others document underpricing of IPOs. The widespread documentation of underpricing spawned a lot of research into why underpricing exists. Some researchers have examined whether having VCs participating in the issue might affect the short term and long term performance of the investee firm differently from what has been documented in the literature. Extant literature finds that VC-backed IPOs are less underpriced than non VC-backed IPOs and attribute the lower level of underpricing to the certification role of the venture capitalists. Whether or not venture capital participation signals information about the quality of the investee firm has also been debated in the literature. Some researchers have hypothesized that VC funding 6

confers advantages to the investee firms because of the large ownership stake of the VCs (who are also involved in the business) reduces agency costs (Belden et al, 2010). Other researchers have also suggested that VCs offer valuable expertise especially in strategic planning. This view suggests that better operating performance of the investee firm will depend on the continued ownership and participation of the VC. Another strand of literature suggests that VCs are able to certify new offerings, and that they choose to fund companies that have better chance of becoming successful in the post issue period. The VC s reputational capital at stake also acts as a disciplinary mechanism that helps keep underwriters from opportunistic behaviors that can damage their reputational capital. Beatty and Ritter (1986) find that underwriters that systematically misprice offerings tend to lose market share. Similarly, Booth and Smith (1986) assert that in the capital raising process the underwriter certifies that the price for the company s shares is in line with the internal information and its true value, with a more reputable underwriter able to provide a more trustworthy certification. Titman and Trueman (1986), in their signaling model, suggest that investment banks will attract high quality clients if they maintain a reputation for accurate pricing. They assert that high quality firms will pay a premium for a more accurate underwriter who can better certify that the firm is indeed of high quality. Entrepreneurs with favorable information about their firms value will choose high quality investment bankers than those entrepreneurs with less favorable information. Megginson and Weiss (1999) examine the impact of venture capital funding on the short-run pricing of IPOs and find that VC-backed IPOs are less underpriced than non VCbacked IPOs. They conclude that the presence of venture capitalist in the issue firms serve to lower the cost of going public which in turn help to maximize the net proceeds to the offering firm and the VC (if they use the IPO as a vehicle to harvest their investment). The finding that VC-backed IPOs are less underpriced is based on the hypothesis that the VCs certify that the issue price is fair and reflects all available information, including inside information. According to Megginson and Weiss (1999), the certification emanates from the VC s reputational capital at stake which needs to be protected. Since VCs fund and take firms public firms over time, their reputational with the market and the entrepreneurs will depend on their ability to get the issues priced right by revealing what the investing public needs to know. 7

Since VCs are able to certify the issue, their offerings will be more fairly priced. Failing this, their reputational capital will be damaged and their ability to obtain favorable IPO prices in the future will be hampered and the quality of their deal flow will be affected. Extant literature also shows that IPOs underperform and the underperformance has, in part, been attributed to underpricing in the short run and overpricing in the long run. VC s participation can help improve the performance of these firms. Long term performance of VC-backed firms could be better because, if as a result of the certification role and reputational capital at stake VCbacked IPOs are less underpriced, then the long run overpricing should not be observed (Belden et al, 2010) and thus VC-backed IPOs should perform better than non VC-backed IPOs. In terms of operating performance, Jain and Kini (1994) document a disappointing operating performance in the post IPOs period and attribute the results, among others, to window dressing prior to the IPO and timing the IPO. Drawing on Megginson and Weiss (1999) work, the importance of reputational capital to the VCs would suggest that this behavior of window dressing with the view to fooling the market would be avoided. It suggests that VC-backed IPO pre and post operating performance won t be significantly different (all things being equal) and the post performance will be better than that of the non-vc backed IPOs. The foregoing review suggests that since reputation is critical to the VCs, they are more likely to reveal and certify information about the true intrinsic value of their investee firms and are able to negotiate an offering price that reflects the intrinsic value of their investee firms, then the level of underpricing will be lower and therefore long term overpricing will be reduced. Thus, the market and operating performance will be better than those of the non VC-backed IPOs. The reasons to expect better performance is the management expertise that the VC brings to the investee firm. The Chinese IPO/VC market has increased significantly, the total amount of venture capital in China has grown from an amount of US$4.76 billion in 2003 to US$36.67 billion in 2008 (China Venture Capital Research Institute (2008). This outstanding growth in VC investment in China makes an examination of VC impact on IPO performance in a growing market in a country that is the second largest economy in the world a worthwhile. 8

3. Chinese VC industry: Developments and trends. China s economy has undergone significant structural changes, with the consequence that the country has achieved a growth rate of 8% per annum since 1980. In 2005, China overtook Britain and became the fourth largest economy in the world in terms of GDP, and in February 2011, it overtook Japan as the world s second largest economy (IMF, 2010). Needless to say, this super performance that has made the country an economic powerhouse reflects the sentiments expressed by James Wolfensohn, a former World Bank President who once said that the Chinese have accomplished in only 20 years what would take other countries two centuries to achieve. The country has attracted significant amount of capital from overseas. The Venture capital investments have featured prominently in the nation s march to prosperity. The venture capital industry in China was established in the mid 1980s when the government realized that the high tech sector held the key to unlocking the growth potential of the country. 4 This desire coincided with the establishment of the China New Technology Start-up Investment Company, a venture capital firm that went bankrupt in 1997. Despite the false start, the government and the private sector continued to support the industry. The establishment of China Direct Investment Funds (CDIFs) was another significant attempt by the government to boost investment but the CDIFs were largely invested in government-owned, state owned enterprises or township/village enterprises (TVEs) in China. Today, venture capital investments come from different entities such as the government, state-owned enterprises, private firms, non bank financial institutions, multinational corporations and foreign venture capital funds (Pukthuangthong and Walker, 2007). By 2000, foreign funded venture capital firms accounted for 8% of China s venture capital organizations. By the same year, there were 120 venture capital firms and 156 incubators in China, with Beijing, Shanghai and Shenzhen emerging as the centers of the venture capital industry in China (Xiao, 2002). The venture capital industry was given a boost with the establishment of the Chinese Venture Capital Associations in 2002 that aims to promote the rights of entrepreneurs and laws on investor protection and disclosure. Added to this regulatory development was the launch of the Small and Medium Enterprise Board (SME) in the Shenzhen Stock Exchange in 2004, which 4 Xiao (2002) and Pukthuanthong and Walker (2007) give an excellent comparative analysis of the venture capital industry in the US and China. 9

serves to provide a valuable exit channel for venture capitalists. With these concerted efforts aimed at promoting private investment, it is not surprising that there has been a surge in VC investments. By 2004, China had taken the third position as the country with the largest amount of venture capital investments. The country represents one of the world s fastest growing markets for venture capital investment. Pukthuanthong and Walker (2007) identify factors that have accounted for the growth to include robust economic growth, a (slow but) growing commitment to intellectual property right protection, a strong entrepreneurial culture, and educational system that directs a large number of students to engineering and business programs in China and the US universities. Despite these developments, a number of challenges affect the venture capitalist in China. There are significant cultural, institutional and regulatory hurdles that affect the VC s ability to monitor their investments. As a result of cultural factors, the relationship between venture capitalist and the investee companies is different in China than what obtains in the West. A number of differences have been identified in the literature that suggests that VC role of monitoring may not be the same as expected in the West. A thriving economy and entrepreneurial spirit are not the only necessary ingredients for the venture capital industry to thrive, but a sound legal system coupled with law enforcement and the availability of various intermediaries such as qualified lawyers who are well versed in local and international law and qualified accountants who can provide auditing and consulting services are needed. But in China, VCs may be unable to find qualified intermediaries who can provide these services efficiently (Pukthuanthong and Walker, 2007). Also, China does not have good accounting and reporting standards; there is a lack of credit rating agencies, and the weak legal system affects property rights. Intellectual property rights are either poorly enforced or not enforced at all (Liu, 2007). The level of managerial sophistication is also found to be comparatively low in China (Bruton and Ahlstrom (2003). Bruton et al (2004) find that effective monitoring by the VC is possible if the VC builds personal relationship with funded entrepreneurs. Whilst business relationship in the west is more formalized, in China and other Asian countries, the relationship is collectivist, in the sense that the entrepreneurs expect their business partners to provide them with personal and equity links and extended interconnected networks. Chinese VCs frequently maintain contacts with not only the CEO, but also with a wider range of top and middle level managers. Differences also exist in 10

the role of the VC and the ability of the VC to monitor the investee firms. VCs in the US and the Western countries exert strong influence on the governance structure of their investee firms in their portfolios. Whereas VCs usually conduct their monitoring activities through membership of the firm s board of directors (Sahlman, 1990, Sapienza, 1992) which allows them to monitor the activities of the investee firm, Low (2002) asserts that VCs investments in China may not necessarily guarantee a board seat. Even when a board seat is earned, VCs will be considered outside directors and the influence of external directors in China is weak. This has serious implications for monitoring as the VC has to cultivate a personal relationship with the entrepreneurs in order to gain his trust, otherwise the VC will be considered an outsider and information needed to monitor the firm will be withheld from him (Pukthuanthong and Walker, 2007). Despite these challenges it must be mentioned that the government has put in place measures to enhance the prospects of the venture capital industry. For example, the Central government has promulgated a series of laws and regulations such as the Venture Capital law, which seeks to establish a proper legal framework for the venture capitalist industry. The government has also passed a legislation that makes it easier for foreign investors to enter the Chinese venture capital market by reducing the minimum required capital. In addition, the government has reformed its stock markets to ease the exit of the VC via the IPOs. These developments have laid the foundation for the sector to grow and provided the needed impetus for the development of the country. Given the findings in the literature, if VCs provide certification and value added role in China, then we would expect VC-backed IPOs to be less underpriced and perform better than the non VC-backed IPOs. 4. Data and Methodology 4.1. Data A data set consisting of VC backed IPO companies listed on both the Shanghai Stock Exchange and Shenzhen Stock Exchange 5 from 1990 to 2008 was obtained from the China Stock Market & 5 In 2004, the Small and Medium Enterprises Board (SME Board) was launched in the Shenzhen Stock Exchange, as a segment within the main board. It aims to provide a direct financing platform for small and medium enterprises. The SME board, nevertheless, serves as an important exit channel for those VC companies as evidenced by the 11

Accounting Research (CSMAR) database. Out of a total of 120 VC backed companies, 98 were listed on the Shenzhen Stock Exchange while 22 were listed on the Shanghai Stock Exchange. Out of the 98 companies listed on the Shenzhen Stock Exchange, 93 (95%) of them went public through the SME Board. In fact, the majority of VC backed IPO listings happened after 2004, the year the SME board was established. Table 1 contains the distribution of the VC backed companies during the study period. It appears from the statistics presented in Panel B that VCs in China do not primarily invest in high technology industry but rather, they appear to have a diversified portfolio. [Insert Table 1 here] In order to test the certification and monitoring hypothesis, we follow the Megginson and Weiss (1991) matched-pair methodology to construct a matching sample of non-vc backed companies. Each VC backed company is matched as closely as possible with an IPO company in the same industry and with a similar size but without VC support. The final sample consists of 114 VC backed and 114 non-vc backed companies. Table 1 Panel B shows the industry distribution of those companies. As shown in the table, the majority of the companies belong to the Machinery industry, followed by IT, Petrochemicals, Electronics, and Metals & Non-metals. These five industries account for over 67% of the sample. To examine the differences between VC-backed and non-vc-backed companies, we focus on firm and IPO transaction characteristics including firm s age, offer price, offering proceeds, flotation cost (total expense), underpricing, P/E ratio, total assets, total debt, debt ratio, profit growth ratio, return on assets and return on sales. The differences in offering and firm characteristics of the VC backed IPOs and a matched sample of non-vc backed IPOs are shown in Table 2. Both parametric and non-parametric tests are used to measure the significance level of the difference in means and medians. Most measures on the IPO characteristics are similar except for the offering proceeds and flotation costs. Consistent with Megginson and Weiss (1999) and Brav and Gompers (1997), we observe that VC-backed firms are on average, larger than non VC-backed firms (as measured by total assets), number of VC backed companies being listed there. 12

albeit the difference is not statistically significant. However, the offering proceeds are significantly larger than those of the non-vc backed firms. The direct cost of going public by VC backed firms is significantly higher than that incurred by non-vc backed firms. The finding of a higher flotation cost for VC backed companies in comparison to non-vc backed ones is consistent with that reported in Singapore, but the VC backed companies in China are significantly larger in terms of offering proceeds. That flotation cost is higher is inconsistent with the notion that because of the certification provided by the venture capitalists, the VC-backed firms can (hire more prestigious underwriters and) pay less, thus reducing the cost of going public. Perhaps the underwriters do not accept the certification of the VC. Both VC-backed and non-vc-backed firms in China go public around the same time (5) years. This finding is inconsistent with Megginson and Weiss (1999) who find that VC-backed IPOs go public at a significantly early stage in their life cycle than non VC-backed IPOs. 4.2. Methodology [Insert Table 2 here] After the construction of the matched sample, we examine the differences between VC backed and non-vc backed companies after the IPO in three major areas, namely, the short term performance, the post-ipo operating performance, and the long-term stock market performance after the IPO so as to ascertain the extent of the influence of VC on IPO companies in China. For the post-ipo operating performance analysis, we focus on three cash flows related measures. The first measure is the profit growth ratio, which is defined as the ratio of net profit of current year relative to that of last year. The second measure of operating performance is the return on assets (ROA), while the last one is the return on sales (ROS). These measures are based on the operating performance at the end of the last fiscal year before the IPO (Year - 1). With regard to the long term market performance comparison between VC backed and non-vc backed companies, we focus on both the buy-and-hold abnormal return (BHAR) and the cumulative abnormal return (CAR). The market return r mt is the return on either the Shanghai Stock Exchange composite index or the Shenzhen Stock Exchange composite index, depending on the Exchange where an IPO is listed. The initial return, defined as the difference between the closing price on the first trading day and the offer price divided by offer price, is excluded from 13

the analysis. The average market adjusted return on a portfolio of n stocks for event month t is the equally weighted arithmetic average of the market adjusted returns, and the cumulative abnormal return from event month q to event month s is computed as the sum of the average monthly market adjusted returns. The BHAR is the difference between the holding period return of company i and the market return is estimated as: T BHAR it = [ [ 1 + r it ] [ 1+ rmt] t= 1 The mean BHAR over a period T is: i= 1 T t= 1 n 1 BHAR T= BHAR it n 5. Empirical Results We begin the analysis by presenting evidence on the underpricing of both VC-backed and non VC-backed firms. Throughout the period of study, there are 149 VC companies conducting 261 investments in those 120 IPO companies. On average, an IPO firm receives funding of 2.18 times from VC companies which is higher than in Singapore (1.63), but lower than in the U.S (2.92). The descriptive statistics of VC involvement in the 120 IPO companies are shown in Panel A and underpricing of the IPO is presented in Panel B of Table 3. The mean VC investment duration, defined as the number of year between a VC s first investment and the company s IPO, is 2.70 years. For firms with multiple VCs, the mean investment duration can be as long as 3.13 years. The duration is longer than that documented for Singapore VC-backed companies of 2 years as documented by Wang et al, (2003) but is comparable to the 2.92 years documented for the US by Barry et al (1990). The average equity stake holding of 20.34% before the IPO and 15.79% after the IPO demonstrates a fairly high institutional holding. The decline in equity holding after the IPO may be largely a result of an increase in the number of shares issued to the public, rather than the sale of shares by the VC. Such large stakes are expected to motivate the VCs to closely monitor their investees subsequent to the IPO and at the same time send credible signals about their beliefs in the firm s prospectus. 14

We observe from Table 3 that both VC-backed IPOs and non-vc funded IPOs are underpriced on average by about 150%. Though we have hypothesized the importance of VC certification in the IPO performance, we did not find any statistically significant difference in underpricing between the VC and non-vc groups. The finding that the presence of VC could not help lower the initial underpricing is in contrast to the results reported in earlier studies but is similar to most of the recent findings. In sum, the results reported here are similar to those documented by Wang et al. (2003) in Singapore, but contrary to those reported by Megginson and Weiss (1991) and Jain and Kini (1995) in the U.S. Earlier, we found that VC and non VC-backed IPOs go public at the same age. [Insert Table 3 here] 5.2. Post-IPO Operating Performance The operating performance of VC and non-vc backed companies after the IPO are shown in Table 4. Year 0 denotes the year of the IPO while Year 1 and Year 2 are the first and the second year after the IPO respectively. The return on asset (ROA) shows that both groups suffer from a decline in performance compared to their pre-ipo year (Year -1) ROA. For the pre-ipo year, the VC-backed as well as the non-vc funded companies achieved about 12% returns, but in the year of IPO, their ROA drops by almost half to around 6%, and the situation remains unchanged in the following years, year +1 and +2. The evidence therefore suggests that both groups are good at timing their IPOs. The return on sales (ROS) also displays a similar falling trend, but the VC backed firms seem to perform a bit better, albeit the difference is not statistically significant. Lastly, the earnings growth rate, which measures the growth of net profits relative to the previous year, shows that VC backed companies perform much better, and this earnings performance difference is statistically significant at 10% for the second year after the IPO. The findings are in contrast to the results of Wang et al. (2003) for Singapore, where non-vc backed IPOs have better post-ipo performance, but are consistent with Jain and Kini (1995) results in the US where VC backed companies perform better after the IPO. [Insert Table 4 here] 15

5.3. Stock Market Performance To assess the differences in market performance between the VC backed and non-vc backed companies, we examine three-, six- and nine-month returns as well as long term one-, two- and three-year buy-and-hold abnormal return (BHAR) and cumulative abnormal return (CAR). The results are shown in Table 5. In general, firms with or without VC backing suffer from long-run underperformance. Although the two groups experience a decline in their after-market BHAR and CAR performance, the VC backed group performs comparatively better as reflected in the magnitude of the underperformance. The long-run underperformance of the VC backed group is statistically significant at 1% for just the three- and six-month returns but not so in other time windows. However, for the non-vc backed group, their underperformance is much more severe and is statistically significant at 1% for the three-, six-, nine-, twelve-, and fifteen-month windows. The differences in performance ranges from 5% to 18% for the first year and all are statistically significant. The finding of long-term underperformance of the IPOs sample is consistent with the existing literature, but our results show that when venture capitalists are involved in an IPO, the underperformance is significantly reduced, and the difference between IPO with and without VC backing can be as high as 18% for the first year. Nevertheless, the importance of VC backing seems to vanish after the first year of IPO, as beyond that period the difference between the two groups is not significant at conventional levels. Overall, the results confirm the monitoring role of venture capitalists as the results show that their involvement in IPOs improves the long-run after market performance. [Insert Table 5 here] 5.4. Cross-sectional Regression Analysis The aggregate (univariate) results presented in table 3 can conceal effects of certification or grandstanding. Therefore, to test the importance of the influence of VC backing on IPOs, we conducted cross-sectional regression analysis examining how (i) the IPO underpricing level, (ii) the change in operating performance between pre- and post IPO and (iii) the long term market performance are related to the VC involvement. In the regression, venture capitalists 16

participation is represented by a dummy variable that takes on a value of 1 if a VC is involved and 0 otherwise. The variable natural logarithm of the flotation costs is used to examine the role of the certification done by VCs. We also include control variables that are commonly examined in the IPO literature. Variables such as the IPO firm s age, the offer price, the natural logarithm of the offering proceeds and the natural logarithm of the total assets are generally adopted as proxies for measuring the riskiness of an IPO. Young and small firms are assumed to be more risky and are expected to be more underpriced. Following prior studies, we include in our study price earnings ratio, debt ratio, pre-offering return on assets, percentage of shares retained and the natural logarithm of the CEO age as additional control variables. Previous studies have found that market sentiments influence underpricing, therefore we include market returns one month before an IPO as another explanatory variable. Finally, several dummy variables are employed. These include the main board dummy, the Shanghai Stock Exchange dummy and the high-technology dummy. We examine the effect of venture capitalists participation and the IPOs characteristics on the performance (IPO underpricing, change in operating performance between pre- and post IPO, and long term market performance) of IPOs by estimating the following model: Performance = α + β 1 + β 1 VC Participation +β 2 Log(Flotation Costs) + β 3 Log(Proceeds) +β 4 Log(Assets) + β 5 Age + β 6 Offer Price + β 7 Price/Earning ratio + β 8 Debt Ratio + β 9 Pre Offering ROA+ β 10 Log(CEO Age) + β 11 Shares Retained + β 12 PreIPOMarket Return + β 13 Main Board Dummy + β 14 Shanghai Exchange Dummy + β 15 High-Technology Dummy + ε 1 5.4.1. Regression Results Table 6 reports the cross-section regression results using the initial returns as the dependent variable. Specification 1 examines the unadjusted initial excess returns while model 2 considers the impact of the market condition on the first trading day using the market adjusted initial returns. We find that the VC participation variable is positive and significant in both specifications. The finding of a significant positive coefficient on the VC participation variable is 17

consistent with some of the recent VC studies, according to which VC-backed IPOs are more underpriced. The level of PE ratio positively influences the underpricing level, suggesting that investors pay attention to the information content of this accounting value. The positive coefficient of CEO age suggests that the more experience the CEO, the higher the underpricing. The negative sign on Proceeds is consistent with existing literature that suggests the greater the size, the lower the risk and consequently, the lower the initial return. Finally, judging from the significance level, the market sentiment before an IPO is the most influential factor in explaining the level of initial underpricing. [Insert Table 6 here] The results of the regression using with the change in operating performance between pre- and post IPO are presented in Table 7. We find the VC participation variable is highly significant suggesting that the presence of VC positively influences the profitability of the investee firms. As major stockholder, holding on average 16% of the firm s shares after the IPO, the VCs continue to advise and monitor their investee firms, thus leading to a better post-ipo return. This finding confirms the importance of the monitoring function of VCs and is consistent with that of Jain and Kini (1995), who observe that VC backed IPOs exhibit relatively superior post-issue operating performance than non-vc backed IPOs. Similarly, Campbell and Frye (2006) also find that the monitoring level is higher for IPOs with VC backing. Such high level of monitoring transforms to a better post issue performance. The significant negative coefficient on underpricing contradicts the signaling hypothesis of Jain and Kini (1994) but is consistent with the findings of Krigman et al. (1999) and Coakley et al. (2007). While the age of CEO impacts operating performance negatively, the percentage of shares retained by the insiders positively influences the performance. Lastly, the Shanghai Exchange Dummy variable is significantly positive suggesting that the IPOs listed on the mail exchange perform better in the post-issue market. [Insert Table 7 here] In Table 8, we report the regression results using 3-months, 6 months, and 1-year after market performance as dependent variables. All the three specifications show that the VC participation variable has a significant positive impact on the after-market returns. Consistent with the operating performance results, the presence of a VC in the IPO firm significantly improves the 18

stock market performance. Judging from the magnitude of the influence, the importance of VC seems to increase over time. The stock return differential between the two groups of IPOs of 5% for the three-month time window rises to 15% and 20% for the six-month and one-year time window, respectively. The finding of an improved long-run stock performance is consistent with the results of Brav and Gompers (1997) and Campbell and Frye (2006) that VC backed IPOs perform better. Similar to previous studies, we also observe that underpricing explains the stock market performance. The coefficient of the underpricing variable shows that the higher the first day return of an offering is, the lower its three-month return becomes, but after the first three months, we find that the initial returns do not significantly affect stock returns. Finally, the significant positive coefficient on the main board dummy suggests that IPOs listed on the main board perform better in the long run than those listed on the secondary board. The high-tech dummy is significantly negative. Given that the whole sample underperformed, this result suggests that the performance of the high tech firms is worse than that of non-high tech firms. [Insert Table 8 here] 6. Further analysis of the performance of VC-backed IPOs in China We provide further analysis of the VC-backed IPOs in China with the view to generating further insights into the value added function of the VC in the investee firms. To do this, we test whether VC and CEO characteristics and the interaction of these characteristics on the underpricing and the after-market performance of the investee firms. Specifically, we examine whether VC prestige power (reputation), VC ownership or controlling power and VC s structural power affect the investee firms underpricing and post issue performance. The VC s ability to influence the investee firm could depend a lot on the CEO of the funded firm. Therefore, we examine the VC characteristics in conjunction with CEO characteristics such as CEO ownership or controlling power, CEO education and CEO duality. In line with Chahine and Georgen (2010), we measure VC prestige power (reputation) as the number of IPOs the VC has been involved with previously. As Sorenson and Stuart (2001) argue, a VC may acquire prestige power from its past successful experience in the capital market. VC controlling or ownership power is measured by the percentage of the firm s shares owned by the VC prior to the IPO. Ideally, VC s structural power should reflect the position of 19

the VC in a syndicate. Because of lack of data, we consider the position of the VC in the top 10 shareholders as a measure VC structural power. Our measure of VC structural power is a dummy equal to 1 if the VC is in the top 10 shareholders and 0 otherwise. In addition, we measure VC participation in the issue as the percentage of VC s holdings sold during the IPO. The percentage of holdings sold during the IPO can affect the pricing of the issue as the VC would not like to leave too much money on the table; therefore, we expect that VC s participation in the IPO will lead to less underpricing. The impact of VC syndication is also examined as prior studies (including Hochberg et al. (2007) show that VC syndication has a positive impact on firms performance as the syndicate creates a network that facilitates the sharing of resources, information and contact, of which the latter is very important in China. Also, it is reasonable to surmise that a syndicate of VCs will be able to push for board representation of their members than VCs which act as the sole provider of venture capital in the portfolio firms especially those that may have difficulty dealing with powerful CEOs (Chahine and Geotgen, 2010). However, it is also reasonable to surmise that VC syndication can have negative effects on the performance of the investee firm as it may find it difficult to take decision in a decisive and efficient manager. We include VC duration with the view to examining the grandstanding hypothesis. If the grandstanding hypothesis hold in China, we would expect VC duration to be significantly negatively related to underpricing as VC s who have not been affiliated with the portfolio firm for long but want to take the firm public early would underprice the issue significantly. We define VC Duration as number of years from the date of investment to the IPO date. Prior studies also suggest that a VC s involvement in the investee firm and their ability to monitor and add value will depend on the CEO s controlling power (Hermalin and Weisback, 1988). We measure CEO s controlling power as the shares owned by the CEO as a fraction of the shares outstanding. We also examine the effects of CEO Education and its interaction with VC characteristics on the performance of the investee firms. VC s may be less likely to take actions that enhance their own interest if the CEO is highly educated or is powerful. Hermalin and Weisback (1988) show that VCs are less likely to sit on the board of, or get involved with, firms that have powerful CEO, those with substantial holding, highly educated CEO, and those where the CEO also is the Chairperson. We examine the impact of CEO educational level and CEO Duality and their effects on underpricing and performance of VC-backed firms. 20

6.1 Further analysis of underpricing of VC-backed IPOs In line with Chahine and Georgen (2010) we conjecture that the pricing of the IPO is the result of the balance of power within the investee firm, specifically between the VC and the CEO. It is possible that powerful VCs would use their skills and power in the pricing process to reduce underpricing. We conjecture that the higher the VC and CEO holdings prior to the IPO, the lower the level of underpricing. Lerner (1994) also finds that the more experienced the venture capitalists, the more proficient they are in the timing of IPO, thus reducing the level of underpricing. Similarly, VC s controlling power (pre-ipo holdings) can affect the underpricing of their investee firm. According to the signaling model of Grinblatt and Hwang, 1989, the percentage of insider holding signals to the market the amount of private information possessed by the insiders. We examine whether VC holding is related to underpricing. We examine the effect of VC characteristics on the performance of the investee firm by estimating the following model: Performance i, j =α 0 + α n VC _Char i, j + α n CEO _Char i, j + VC *α n CEO _Char i, j K n=1 n=d+1 n=f+1 D F H + α n Firm _Char i, j + u i, j n=h+1 The specific equation we estimate is IPO Performance = α + β 1 + β 1 VC Participation +β 2 VC Syndicate + β 3 VC Controlling Power +β 4 VC Prestige + β 5 VC Structural power + β 6 VC duration + β 7 CEO Education + β 8 CEO Controlling Power + β 9 CEO Duality+ β 10 CEO Participation+ β 11 PreIPOMarket Return + β 12 Age + β 1 Growth + β 13 Debt Ratio + β 14 Log(Proceeds) + β 15 Price/Earning ratio + ε 1 where IPO performance is underpricing or aftermarket stock market performance, VC participation is the percentage of VC s holdings sold during the IPO, VC Syndicate is the number of VCs in the investee firm. VC Controlling power is the VC share ownership prior to the IPO. In line with Chahine and Goergen (2010) VC Prestige power is the number of IPOs the VC has been involved with previously. VC s structural power is a dummy variable equal to 1 if the VC is a top 10 shareholders, and 0 otherwise, and VC duration is the time from year of VC investment to the IPO date. CEO ownership power is the percentage of the shares owned by the 21