Underpricing and Long-term Performance of IPOs under Regulation
Changes in China
Name: Jiacheng, Ma
Student ID: 10621458
Supervisor: Dr. R. Almeida Da Matta
MSc Business Economics, Finance track
University of Amsterdam, Amsterdam Business School
June 30, 2015
Table of Contents
Abstract ... 3
1. Introduction ... 4
2. Literature Review ... 6
3. Features of Chinese IPO market and Stock Exchange... 14
4. Data, Methodology and Summary Statistics ... 19
4.1 Data ... 19
4.2 Methodology ... 21
4.3 Summary Statistic ... 23
5. Empirical Results Analysis of IPOs Underpricing ... 27
5.1 Differences in Differences ... 27
5.2 Short-term Performance ... 32
5.3 Long-term Performance ... 35
5.4 Operating Performance... 38
6. Conclusion and Limitation ... 44
6.1 Conclusion ... 44
6.2 Limitation ... 45
Reference List ... 47
Abstract
This paper mainly studies changes of IPOs underpricing, short-term and long-term post-issuing stock performance and operating performance of 651 Small-and-Mid Enterprise Board IPOs under the regulation changes in China. In Small-and-Mid Enterprise Board, the relative change(Difference in Difference value) of IPOs underpricing is positive 52% from 2006 to 2012, whereas the relative change of IPOs underpricing is negative 23% between 2008 and 2010. This may due to the different market effects on IPOs underpricing in different market. Both in the short run and long run of stock performance, the buy-and-hold abnormal return decline significantly after regulation changes. Moreover, the long-term stock accumulative return could be explained by operating performance of the company.
1. Introduction
IPO underpricing, a phenomenon which the offering price is lower than the opening price, is common across 25 countries according to Loughran, Ritter, and Rydqvist (1994). They found that higher IPO underpricing in developing countries than in developed countries. Ritter (1984) provides convincing evidence that IPOs are underpriced.Researchers tried to explain this interesting phenomenon by using asymmetric information distribution among issuers, underwriters, and investors. Beatty and Ritter (1986) argue that the uncertainty of going IPOs influenced the level of underpricing, generally, it’s costly for investors to obtain information for those IPO firms with high uncertainty. As a result, IPO firms need to give investors a big discount to compensate their risk exposure.
With a more than 20-year development of Chinese stock market, it’s becoming more and more mature than ever before. Since the first IPO reform in 1994, Chinese investors have experienced seven different reforms until 2006. For these seven reforms, some of them were considered to be successful by the public, but some were not. Many past researchers’ findings have addressed how these reforms affected the IPOs underpricing before and after IPO regulation change. However, in these research papers, we could easily find these researchers only observe how regulations changes impact the whole market but ignore the effects to the market ingredients such as different stock board segments and different
industries.Sometimes a bull or bear market could also have significant effects on IPOs listing day return. Furthermore, time span is another factor we should take into consideration because IPO regulation changes have long-term and short-term effects to IPO underpricing.
In a vast of theoretical models, they try to explain the reasons why the offering price is extremely different from the opening price. The empirical findings are consistent with results in theoretical models. The current literature shows us the importance of economic and political context in affecting the IPOs pricing. In the process of a IPOs pricing and during the period of post-issuing trading, they all combine and balance economic and political factors and as well as the economy structure. A more profound understanding of how regulation changes affect IPOs underpricing in certain market comes to be necessary.
In this paper, our investigation was performed to studythe short-term and long-term performance and underpricing of IPOs firms in Small-and-Mid Enterprise Board Segment before and after IPOs regulation changes in 2009. The time span of the datawe use is from June 2006 to December 2012. The results report that IPOs underpricing is 52% higher after the IPOs reform in the time span from 2006 to 2012. However, when we only consider the time span from 2008 to 2010, the IPOs underpricing is 23% lower after the IPOs reform.The reason for this phenomenon is mainly because the IPOs underpricing in Main Board and Small-and-Mid Enterprise
Board are extremely different before 2009. We also investigate the short-term and long-term stock price performance in Small-and-Mid Enterprise Board before and after the regulation changes. We find that, before IPOs reforms, the after-issuing stock return outperform the market by 125.93% and 144.1& within a month and three years respectively; after IPOs reforms, the after-issuing stock return outperform the market by only 33.03% and 52.80% with a month and three year respectively. In Small-and-Mid Enterprise Board,operation performance is highly related to the long-term stock price performance.In our study, our results present that long-term accumulative return is significantly relative with operating performance. Overall, the purpose of this study is to provide a new perspective and new methodology when we consider IPOs-reform influence. It’s a new evidence to the vast literature relates to IPO underpricing.
In our study, the structure is as follows. Section 2 reviews the past literature. Section 3 presents the features of Chinese IPO market and Stock Exchange. Section 4 describes the Data, Methodology and Summary Statistics. Section 5 provides the empirical results of short-term and long-term underpricing and performance of IPOs. Section 6 is the conclusion and limitations.
2. Literature Review
these models is the information asymmetry which indicates that investors have more information than issuers have but different investors have different amount of information. Rock (1986) presents the Winner’s Curse Hypothesis and Welch (1992) builds Negative Casade Model. In Rock’s Winner’s Curse model, it supposes that investors have initiative in pricing IPO and are classified as informed investors and uninformed investors. There are no information communication between these two kinds of investors. When the informed investors discover the investment value of the new shares, they will be eager to subscribe and leave no chance for uninformed investors to subscribe these new shares. That’s how Winner’s Curse model formed. To attract the uninformed investors to subscribe, the price of new shares has to be underestimated so as to compensate the uninformed investors’ information asymmetry. Based on Welch’s Negative Casade Model, uninformed investors only subscribe the new shares if the new shares are in hot sale, and they don’t bid for the high-offering-price new shares as well as the informed investors. Chemmanur (1993) presents Information Collection Hypothesis which assumes issuers have more inner information about the companies than that investors have. The issuers expect the investors to acquire information by themselves, but it’s costly for investors to collect companies’ information so the issuers could only lower down the issuing price to attract external investors to collect information of companies. Finally the stock price in secondary market would be adjusted by companies’ information collected by investors. Therefore, IPOs underpricing is the compensation for investors who collect information, the blue chip companies could distinguish themselves with other
underperformance companies by using high IPOs underpricing.
Although the information asymmetry have some explanatory power of IPOs underpricing, some researchers states that IPOs underpricing is not always the compensation for information asymmetry in the market. Even the information is equivalent to each party in the process of IPOs, the phenomenon of IPOs underpricing still exists. Tinic (1988) points out that issuers could avoid the legal liability through the IPOs underpricing. Assuming that all the participants know the stock price of a IPOs on listing day could be 20 US dollar, but the issuer still issues new shares at 10 US dollar per share. By doing this, the issuer is not potential to be prosecuted. This is called Lawsuit Avoidance Hypothesis. In USA, the strict information disclosure system make the IPOs public information transparently, it’s possible for investment banks, accountants and issuers to face the lawsuit if they conceal any information. The investors could submit a case to the court if IPOs prospectus is not real or is hiding some company’s real information. To avoid the unnecessary lawsuit in the process of issuing new shares, underwriters and issuers always use IPOs underpricing to achieve this goal because only those investors who suffer the loss in IPOs would sue. On the other hand, Drake and Vetsuypens (1993) find that the IPOs underpricing could not protect IPOs firms from being sued, in the cases of IPOs lawsuit, majority of them are caused by high underpricing, whereas the rest of them are caused by low underpricing. It can be explained that the reason for this sort of phenomenon is the punishment afterwards.
Baron (1982) explains the IPOs underpricing by using principal-agent theory. In his theory, the assumptions are that issuers are at information disadvantage comparing with underwriters but not investors. To motivate the underwriters to sell the stocks, issuers have to allow the low pricing as they could not monitor the underwriters regardless of the cost. Nevertheless, Muscarella and Vetsuypens (1989) find when investment banks go public, their stock price is still in low pricing though there are not any scrutiny problems. This evidence doesn’t support or disprove Baron’s hypothesis because underwriters could show the market the underpricing is necessary cost to go public by lowering down their own shares price. Loughranand Ritter (2002) make a further investigation on principal-agent relation between underwriters and issuers. They think if underwriters have right to decide the placement of new shares, underwriters would not maximize the issuers’ profit by having the decision right. On occasion, underwriters would lower down the price and then distribute the new shares to clients. There’s evidence to show that underwriters benefit the clients through IPOs underpricing to exchange with clients’ correlated conditions. They also prove that if enterprisers know the listing stock price is higher than their expected price, these enterprisers could stand a higher underpricing. In other words, issuers don’t bargain with underwriters if issuers know that their fortune could grow rapidly recently.
etc. in 1999. When they study Japanese IPOs problems, he finds if IPOs company’s shares are cross-holding by number of financial institutions, the underpricing is always higher. They think this’s made by the potential interest conflict, the higher IPOs underpricing is a market reaction to complicated and non-transparent interest relations among number of Japanese business groups. One thing is deserved to be mentioned is that, nothing about interest conflict phenomenon could be found in US market, but in Asian countries like the Philippines, interest conflict problems exist in IPOs market. Sullivan and Unite (2001) study 104 IPOs in the Philippines between 1987 and 1997. They find that IPOs underpricing has no relationship with enterprise’s basic characteristics such as enterprise size, age, industry and so on. But if IPO firm belongs to an enterprise group and hire foreign underwriters, the underpricing would be higher. Sullivan and Unite consider this may be relative to the interest conflict as the Philippines stock exchange is monopolized by several large enterprise group and lack efficient legal supervision and management.
IPOs underpricing could also be explained by Underwriter Reputation Hypothesis. As the investment banks underwrites a lot of shares at the same time and have a number of potential clients, so investment banks are able to adjust the IPOs pricing to build up their own reputation and make more profits by virtue of reputation. The evidence from Beatty and Ritter (1986) state that the reputation of underwriters has a negative relationship with the IPOs underpricing, a highly credible underwriter will price IPOs close to its true value to keep reputation. Besides, Chemmanur and
Fulghieri (1994) find that the reputation acquisition could alleviate credibility problem by using a model which based on result of Beatty and Ritter (1986). Following the changing of IPOs market, the way of making profit through the investment bank reputation is also changing. Carter and Manaster (1990) study the American IPOs in 1980s. In their paper, they present the reputation of investment bank has become a signal of the risk relates to the firms go public. Investment banks with high reputation always refuse to recommend the young firms with enormous risk to go public. Therefore, the IPOs which are underwritten by high-reputation investment bank are always with less underpricing. But from studies of Cooney, Singh, Carter and Dark (2001) and Loughran and Bitter (2004), after 1990s, especially in “Internet Bubble”, the higher reputation of investment bank, the higher IPOs underpricing. It’s totally opposite to negative relationship between investment bank reputation and IPOs underpricing in 1980s. They states the reason for this phenomenon in 1990s is the looseness of underwriters’ recommendation standard. The underwriters recommend these young firms without any profitability to go public to occupy more market shares and earn more commission by giving up underwriting fee and giving a higher underpricing to clients.However, Tian and Zhan (2000) test the relationship between the reputation of underwriters and IPO underpricing in China, and find that underwriters’ reputation have no explanatory power on Chinese IPO underpricing. It implies that Chinese stock market has its own specific characteristics.
Seasoned Equity Offering Hypothesis is another model which explain IPOs underpricing. Allen and Faulhaber (1989) present IPOs underpricing is one of signal for company to show his intrinsic value. The company with high quality choose toissue stocks at under-real-value offering price to distinguish with those companies with low quality on purpose. By doing this, the high-quality company could deliver company’s information to investors so that the investors are willing to pay for a higher price of company’s seasoned equity offering in the future. But it’s difficult for the low-quality companies to pretend to be a good firm by using IPOs underpricing. If Seasoned Equity Offering does work, then the level of underpricing is negative with the frequency of seasoned equity offering.
Signaling Hypothesis is established in 1980s. Allen and Faulhaber (1989), Grinblatt and Hwang (1989) and Welch (1989) discover this Signaling Model successively. In this model, it assumes that investors could not distinguish the conditions of IPO companies because of the information asymmetry. The companies which issue stocks at lower price are always with bright future, though they have to suffer a relative loss as they issue stocks at high cost and low price, an additional issue in the future could compensate their loss in the IPOs. For the firms which could not take this strategy is always regarded as in bad operating performance so they could not suffer the loss made by IPOs underpricing.
new shares liquid, issuers lower down the price on purpose. By doing so, companies with high autonomy could disperse the ownership within the huge amount of new shares subscribers to increase the difficulty to be merged.
In the Market Climate Hypothesis, it presents that when the market is in rising period, the price risk and market uncertainty increase. It means that it would be much harder for investors to value the enterprise. To attract the investor to participate in IPOs, the issuers and underwriters would set up a higher underpricing than that in any other periods. Ritter (1991) and Aggarwal and Convoy (2000) verify that IPOs underpricing has significant positive relation with market climate. In another model which is called “Flow Effect Hypothesis”, the supporters think the investors’ final decisions would be affected by other investors surrounding and market condition in that moment. To mobilize investors’ enthusiasm and to build a better market atmosphere, underpricing could attract a number of investors and make the IPOs process successful.
We also need to consider the long-term performance of IPO firms. Ritter (1991) presents us a very good way in analyzing long-term abnormal return and cumulative abnormal return by using matched firms after first day IPO. In addition, we could use this benchmark-adjusted return as well in analyzing short-term stock returns. Ritter (1991) argues that investors are overoptimistic about the earnings potential of young growth companies. Usually, most of investors are guided by issuers’ financial report.
Teoh, Welch, and Wong (1998) reveal that the IPO firms with abnormal accruals would experience poor stock return performance in three year thereafter. It’s because accruals would inflate or deflate earnings and cash flow directly and therefore affect the firm value in the end. Jain and Kini (1994) and Mikkelson, Partch and Shah (1997) find that IPO firms would experience a worse post-issuing operating performance. They also indicate that managers would decorate the accounting reports as good as possible before going public to attract investments. In China, different industries have different accruals, especially some “hot industries” such as real-estate. In analyzing the factors which relate to IPO firms long-term performance, excluding these corporate-related factors such as sales growth rate, return on asset and so on, we have to consider industrial effect as well. In analyzing the IPO -firms long-term performance, motivated by Cheung, Ouyang, and Tan (2009), we should separate the whole data sample into two sub-samples base on important events of IPO reforms to identify the effects of regulatory changes.
3. Features of Chinese IPO market and Stock Exchange
In the past two decades, Chinese economy has been developing rapidly. However, Chinese capital market is still a young boy comparing with European and American capital market. The low market efficiency is another characteristic of Chinese capital market as it’s politically driven and putting state enterprise at a great advantage. The government try to balance the market orientation to make it to be free but still under
government’s control. Even though there are significant reforms in the regulation of stock market since 2001, permission from government is still important when company go IPOs.
Shanghai Stock Exchange is the oldest stock exchange in China, it dates back to the Republic of China and have been sleeping until 1990. The re-opening of Shanghai Stock Exchange became a milestone in modern Chinese capital market history. In 1991, the Shenzhen Stock Exchange was opened. However, the purpose of Chinese stock market establishment is different from that in the western world. In European and American stock markets, capital is directed to the companies which are in good operation and healthy financial condition. While in Chinese stock market, its establishment is to raise money for these state-owned enterprise and help them survive.
From 1990 onwards, private investors are investors are allowed to invest in shares. There are “A-shares”, “B-shares” and “H-shares” in China right now. For A-shares, it consists of three major board segments which are Main-Board Segment, Small-and-Mid Enterprise Board Segment and Growth Enterprise Board Segment. No matter in which stock exchange these A-shares are traded, only Renminbi (RMB) is allowed for the stocks trading. For B-shares listed in Shenzhen and Shanghai Stock Exchange are traded with foreign currencies. The face value of B-shares are valued in RMB, but Shanghai B-shares are traded in US dollar, whereas Shenzhen B-shares are
traded in Hong Kong dollar. H-shares are those company which shares incorporated in mainland China but listed and traded in Hong Kong Stock Exchange.
In China, investors can’t submit orders in the market directly. All the trading are under government’s supervision by two “Securities Companies”. One is the Brokerage Securities Company, their business is only in brokerage; another one is the Comprehensive Securities Company which provide additional services like the underwriting of IPOs. In addition, trading is also regulated by Shenzhen and Shanghai Stock Exchange. Daily volatility is restricted to 10% but to 5% in special stocks which were unprofitable in the past three years. To reduce the stock price daily volatility, transaction tax is another important instrument. Before 2005, the highest transaction tax could be 5% of the turnover; but after 2005, the transaction tax was set to 0.1%. Moreover, “T+0” system is implemented, it means that when you buy a stock, you can only sell it after the transaction day. The aim of this system is also to reduce the daily volatility.
Comparing with Shanghai Stock Exchange, Shenzhen Stock Exchange has more number of listed companies. Until the end of 2012, there are 1,512 companies listed in Shenzhen Stock Exchange, whereas there are 941 listed firms in Shanghai Stock Exchange. However, Shanghai Stock Exchange has market capitalization of 15,860 billion RMB which is five times larger than the market value of Shenzhen Stock Exchange (3,410 billion RMB).
Both two stock exchanges are regulated by a main governmental ministerial-level public institution which is China Securities Regulatory Commission (CSRC). The function of this institution is to set standard rules for stock exchange activities including trading, IPOs and so on. Under CSRS’s inspection, companies are necessary to meet CSRC’s requirements to go IPOs. In general, to go public, firms have to be profitable, have a “…complete and clearly defined organization” and in a “…sound
financial condition” (China Securities Regulatory Commission 2012). However, to be
listed in different board market, additional requirements are different. In Main-Board market, company’s common stock value must be at least 30 million RMB and the accumulated profits of the last three years must be at least 30 million RMB as well. Besides these, company’s accumulated cash flow of the last three years must be at least 50 million RMB and the value of intangible assets must be less than 20% of total assets. But for the companies listed in Small-and-Mid Enterprise Board market, starting from 2004, the only difference between Main-Board market and Small-and-Mid Enterprise Board market is shares value in Small-and-Mid Enterprise Board is around 100 million RMB (China Securities Regulatory Commission 2012). In the process of going IPOs, underwriters have to be verified by CSRC. When underwriters apply for an IPO in the stock exchange, the exchange would submit the application to the CSRC and wait for the decision. In 2001,a new system, Securities Offering Review and Approval System, is established because of the retirement of Administrative Quota System. For the old system, it lacks of flexibility as it is district
oriented, the idea of this system is balance new capital and IPOs quota to different provinces equally and give priority to important and strategic industries to go IPOs. Under the new system, the limitation of IPOs quota for industries and provinces disappears, but government can still determine what kind of firms could go IPOs, and the final decision is still under Public Offering Review Committee control.
Once company got IPO permission from Public Offering Review Committee, investors could subscribe new shares during the offering period. From May 2002, the amount of new shares which could be subscribed by secondary investors is based on their holding stocks market value. The new shares would be firstly allocated to strategic investors and legal persons and then investors could buy 1,000 shares with a holding stocks market value of 10,000 RMB in the secondary market. The new shares are used to be oversubscribed, in this case, the regulators would determine the amount of new shares for each investor based on the market value mechanism. But in a special case of undersubscription, the residual new shares would be sold be sold to the investors through on-line offering. The on-line offering are special for individual investors.
For the new shares offering to most investors, these shares are defined as tradable public shares, whereas non-tradable shares are held by state, domestic financial institutions and local government. Before 2006, the majority of a company’s shares are non-tradable shares. But since non-tradable shares reform in 2006, these shares
are as same as the tradable shares in the public. This reform is regarded as a significant improvement in Chinese capital market and enhances the efficiency and the liquidity of the whole stock market including the IPOs market.
From 2006 to 2009, Chinese stock market experienced both bull and bear market. But the step of IPOs reform didn’t stop. There are two different ways for target investors to get new shares: for institutional investors, they could get new shares from off-line subscription; for individual investors, they could only purchase new shares from on-line subscription. From 2009 onwards, another round of new reforms start, and they are focusing on Small-and-Mid Enterprise Board by setting a three-month frozen period for off-line new shares. The institutional investors who successfully subscribe new shares could not trade these shares within 3 month from the issuing day. Moreover, the category of target institutional investors is enlarged, more professional investors could participate into the issuing price setting and new shares off-line subscription. In general, reforms starting from 2009 make the IPOs market become more and more efficient and alleviate new shares underpricing.
4. Data, Methodology and Summary Statistics
4.1 Data
The data are retrieved from CSMAR database. The sample consists of IPOs in Main-Board market and Small-and-Mid Enterprise Board market from June 2006 to
December 2012. Our sample includes 147 IPOs trading on Main-Board market and 651 IPOs trading on Small-and-Mid Enterprise Board market. In our study, we classified the whole sample as two sub-samples based on the IPOs reforms. The first sub-sample contains 46 IPOs on Main-Board segment and 223 IPOs on Small-and-Mid Enterprise Board segment from June 2006 to December 2008. In the second sub-sample, it includes 101 IPOs Main-Board segment and 428 IPOs on Small-and-Mid Enterprise Board segment from January 2009 to December 2012. There are no firms going public from January 2009 to May 2009 in Small-and-Mid Enterprise Board.
The industries in stock market could be classified as six major categories which are Finance, Utilities, Real Estate, Conglometrates, Industries and Commerce. But because of the limited functions for Main-Board market and Small-and-Mid Enterprise Board market, Main-Board market is only for stocks in industries like Finance, Utilities, Real Estate, Industries and Commerce to go public, whereas Small-and-Mid Enterprise Board market consists of industries like Utilities, Real Estate, Conglometrates and Industries. Comparing the industries in these two board market, we could easily find only “Finance”, “Real Estate” and “Industries” are overlapped.
4.2 Methodology
When we compare the difference of IPOs underpricing before and after regulation changes, IPOs firms between June 2006 and December 2008 are the “using data” and the IPOs firms from January 2009 to December 2012 are the “master data”. By controlling the factors like industries, offering price, issuing number of shares and book value, we choose firms from “using data” and match them to “master data”.
The listing-day return is calculated as:
𝑅𝑒𝑡0 = 1 𝑛 𝑃𝑖0 𝑃𝑖𝑙 − 1 𝑛 𝑖=1
where Ret0 is the average return of the IPOs on the listing day (day 0), Pi0 is the
closing price of stock i on day 0, and Pil is the offering price of stock i. By adjusting the
listing return with market return, we have adjusted listing-day return as following:
𝐴𝑑𝑗𝑟𝑒𝑡0 = 1 𝑛 𝑃𝑖0 𝑃𝑖𝑙 − 𝑃𝑖,𝑚0 𝑃𝑖,𝑚𝑙 𝑛 𝑖=1
where Adjret0 is the average of the market-adjusted listing day returns of IPOs on day
0, Pi,m0 is the closing value which relates to Shanghai or Shenzhen A-share market
index and Small-and-Mid Enterprise Board market index on the listing day of stocki, Pi,ml is the closing value which relates to Shanghai or Shenzhen A-share market index
When we examine the short-term and long-term performance of new listing companies before and after 2009, we calculate the BHAR (buy-and-hold abnormal return) as following: 𝐵𝐻𝐴𝑅𝑇 = 1 𝑛 𝑛 𝑖=1 𝑅𝑒𝑡𝑖 ,𝑡 𝑇 𝑡 =0
where BHART is buy-and-hold abnormal return from t to T, Reti,t is the return of share
i at time t and t=0 equals to the listing day. We should consider the market effect to the BHAR, so the BHAR formula is adjusted as following:
𝐴𝑑𝑗𝑢𝑠𝑡𝑒𝑑_𝐵𝐻𝐴𝑅𝑇 = 1 𝑛 𝑛 𝑖=1 𝑅𝑒𝑡𝑖 ,𝑡 − 𝑅𝑒𝑡𝑖 ,𝑇𝑀 𝑇 𝑡 =0
where BHART is the adjusted buy-and-hold abnormal return from t to T, Reti,t is the
return of share i at time t,𝑅𝑒𝑡𝑖 ,𝑇𝑀is market return corresponding to Shanghai or
Shenzhen A-share market index and Small-and-Mid Enterprise Board market index at time t and t=0 equals to the listing day.
Comparing with some others’ work, we could not find anything relates to Difference in Difference Model. That’s why we useDiff-in-Diff Model (Difference in Difference Model) to test whether the regulation changes in 2009 is effective or not. Our Diff-in-Diff Model is:
𝑅𝑒𝑡𝑖0 = 𝛽0+ 𝛽1× 𝑆𝑀𝐸𝑖𝑡 + 𝛽2× 𝑇𝑖𝑚𝑒 𝑇𝑟𝑒𝑛𝑑𝑖𝑡 + 𝛽3× 𝑇𝑖𝑚𝑒 𝑇𝑟𝑒𝑛𝑑𝑖𝑡 × 𝑆𝑀𝐸𝑖𝑡
where 𝑅𝑒𝑡𝑖0 is the stock i first-day return, 𝑇𝑖𝑚𝑒 𝑇𝑟𝑒𝑛𝑑𝑖𝑡 is a dummy variable
which equals to 1 if the listing date of stock i belongs to the time from January 2009 to December 2012, 𝑆𝑀𝐸𝑖𝑡 is a dummy variable as well which equals to 1 if IPO firms
is in Small-and-Mid Enterprise Board market.
Furthermore, we use normal regression model to test the relations between buy-and-hold abnormal return of stock i and its changes in operating performance. The regression model is:
𝐵𝐻𝐴𝑅𝑖𝑡 = 𝛽0+ 𝛽1× ∆𝑅𝑂𝐴𝑖,−1 𝑡𝑜 𝑡+ 𝛽2× ∆𝐶𝐹𝑂𝐴𝑖,−1𝑡𝑜 𝑡 + 𝛽3× ∆𝑆𝐴𝐿𝐸_𝐺𝑖,−1 𝑡𝑜 𝑡 + 𝛽4× ∆𝐶𝐸𝑖,−1 𝑡𝑜 𝑡
The dependent variable is the one-year, two-year and three-year stock accumulative abnormal returns following the IPOs. ∆𝑅𝑂𝐴𝑖,−1 𝑡𝑜 𝑡 is the change of return on asset
in period -1 to t. ∆𝐶𝐹𝑂𝐴𝑖,−1𝑡𝑜 𝑡 is the change of cash flow on total assets in period
-1 to t. ∆𝑆𝐴𝐿𝐸_𝐺𝑖,−1 𝑡𝑜 𝑡 is the change of growth rate of sales in period -1 to t.
∆𝐶𝐸𝑖,−1 𝑡𝑜 𝑡 represents the change of capital expenditure in period -1 to t.
4.3 Summary Statistic
Table 1 presents the market-adjusted and unadjusted stock listing day return in Main-Board market and SME (Small-and-Mid) Board market. This table illustrates the
results of IPOs underpricing of the whole sample in each year from 2006 to 2012. The unadjusted average listing-day return for SME Board and Main Board are 78.5% and 49.3% respectively for the whole period. IPOs underpricing in SME Board is 19.2% higher than that in Main Board. In the both of these two board markets, they get to the peak value of underpricing in 2007 and then decline persistently. Even though we consider the market effect to IPOs underpricing, we could still get the same results of a declining underpricing from 2007, but listing-day return in SME Board market declined significantly since 2009 whereas listing-day return in Main Board market started to decline in 2008.
Table 2 presents descriptive statistics for the variables which are highly correlates to IPOs underpricing. In SME Board market, over the whole sample period, average offering price is 20.60 RMB and the median offering price is 18.00 RMB and both the mean and median offering price are increasing since 2006. However, even in the sample period, mean and median offering price in Main Board are 13.04 RMB and 10.00 RMB respectively, which are lower comparing with SME Board. It’s because Main Board is only for super large enterprises. The number of shares issued in SME Board is quite stable over time which fluctuates from 30 million to 40 million while the number of shares issued in Main Board it extremely large. However, in the Main Board, the less firms go public, the larger number of shares they issue. For the actual money raised in both the Main Board and SME Board, it’s moving constantly. The changing tendency of P/E Ratio is quite similar for both of the board markets, which declines in SME Board and in Main Board since 2009 and 2007 respectively.
Table 1: The IPOs’ under-pricing in China from 2006 to 2012
This table reports that the unadjusted and adjusted listing day return or underpricing of Small-and-Mid Enterprise Board and Main Board IPOs in China. The sample includes 651 Small-and-Mid Enterprise Board IPOs and 147 Main Board IPOs between June 2006 and December 2012. The unadjusted and adjusted underpricing of IPOs are defined as 𝑅𝑒𝑡0= 1 𝑛 𝑃𝑖0 𝑃𝑖𝑙 − 1 𝑛 𝑖=1 and𝐴𝑑𝑗𝑟𝑒𝑡0= 1 𝑛 𝑃𝑖0 𝑃𝑖𝑙 − 𝑃𝑖 ,𝑚 0 𝑃𝑖,𝑚𝑙 𝑛
𝑖=1 , respectively. 𝑃𝑖0is the closing price of stock i on day 0, 𝑃𝑖𝑙 is is the
offering price of stock i, 𝑃𝑖,𝑚0 is the closing value which relates to Shanghai or Shenzhen A -share market index and Small-and-Mid
Enterprise Board market index on the listing day of stock i, 𝑃𝑖,𝑚𝑙 is the closing value which relates to Shanghai or Shenzhen A-share
market index and Small-and-Mid Enterprise Board market index on the offering day of the stock i. Ret represents underpricing and SD stands for standard deviation.
Year
Small and Medium Enterprise Board Main Board
Unadjusted Adjusted
N Unadjusted Adjusted N
Ret SD Ret SD Ret SD Ret SD
2006 0.965 0.60 0.920 0.61 52 0.354 0.26 0.330 0.23 14 2007 2.090 1.12 1.983 1.11 100 1.319 0.92 1.227 0.77 26 2008 1.205 0.91 1.281 0.93 71 0.487 0.25 0.534 0.23 6 2009* 0.648 0.35 0.650 0.34 54 0.559 0.65 0.561 0.60 9 2010 0.451 0.48 0.449 0.46 204 0.315 0.33 0.313 0.30 27 2011 0.206 0.30 0.214 0.28 115 0.174 0.30 0.185 0.29 39 2012 0.282 0.86 0.276 0.83 55 0.386 0.56 0.383 0.54 26 Total 0.785 0.93 0.774 0.91 651 0.493 0.66 0.480 0.59 147
Table 2: Descriptive statistics for Small and Medium Enterprise Board and Main Board IPOs in China This table reports the descriptive statistics for Small-and-Mid Enterprise Board and Main Board IPOs.
Variable
Small and Medium Enterprise Board Main Board
Offering Price Number of Shares Issued (Millions) Actual Money Raised (Millions) Process Time
P/E Ratio Offering Price Number of Shares Issued (Millions) Actual Money Raised (Millions) Process Time P/E Ratio 2006 Mean 8.75 40.32 310.51 17.96 46.41 6.22 2252.20 8175.35 180.79 29.91 Median 7.45 32.60 258.70 18.00 43.74 4.83 850.00 3470.00 16.00 26.60 2007 Mean 11.30 36.46 390.91 15.25 83.17 12.11 1516.04 17322.82 20.54 104.74 Median 10.21 24.50 256.52 15.00 77.45 8.24 1000.50 10756.81 12.50 66.00 2008 Mean 12.16 36.37 423.72 13.78 54.63 9.43 1588.23 12716.45 12.83 52.63 Median 10.78 30.00 317.22 13.00 45.22 8.11 1462.67 9448.33 12.00 44.92 2009* Mean 23.56 35.44 784.52 10.26 70.62 10.77 2413.73 13902.74 13.11 51.92 Median 20.00 34.50 669.00 10.00 61.86 6.95 520.00 11114.93 13.00 45.16 2010 Mean 28.02 38.12 993.98 12.19 74.07 11.76 1694.99 6931.85 14.04 49.03 Median 25.84 30.00 765.35 13.00 63.59 9.25 281.50 2850.00 14.00 41.38 2011 Mean 25.96 34.81 886.04 10.03 47.19 19.59 253.64 2668.45 11.49 41.52 Median 23.00 33.00 709.80 9.00 44.18 18.80 100.00 1833.75 11.00 36.63 2012 Mean 17.99 38.67 635.00 10.16 32.51 10.81 182.83 1383.30 12.81 29.58 Median 18.00 34.38 524.80 10.00 28.47 9.05 100.00 961.34 13.00 21.84 Total Mean 20.60 37.09 717.79 12.58 62.64 13.04 1110.18 7474.94 29.96 51.94 Median 18.00 30.00 563.40 12.00 54.86 10.00 200.00 2550.00 13.00 39.90
5. Empirical Results Analysis of IPOs Underpricing
5.1 Differences in Differences
Table 3 shows the changes in average listing day return in our study. We presents our results by “Board” in columns (1), (2), (4), (5), (7), (8), (10) and (11), and by “Industries”. In our table, we investigate three industries as following: “Finance”, “Real Estate” and “Industries”. We also presents the differences in average listing day return between Main Board and Small-and-Mid Enterprise Board by the whole industries and different industries in columns (3), (6), (9) and (12). Row 3 of the table measures the changes in average listing day return between row 1 and row 2. Row 1 measures the average listing day return before 2009 and Row 2 measures the average listing day return after 2009. The relative listing day return (the “difference in difference” of the changes in listing day return) in the whole sample and sample period is 52%, with a t statistic of 1.76. Within the different industries, relative listing day return for “Finance”, “Real Estate” and “Industries” are 29%, -69% and 25% respectively. It’s because only these three industries are overlapped, so we can’t list out all the six industries.
In general, the results in Table 3 suggests that listing day return in SME Board is unaffected by the IPOs reform in 2009. In different industries, the listing day return is affected in different levels. In “Finance” and “Industries”, the relative listing day return is positive, it means that the IPOs is even more underpriced. But in “Real Estate”, the relative listing day return is negatively, the IPOs reform has impacts on
“Real Estate”.
The findings in Table 3 aren’t identical with our expectations that IPOs reform will reduce first trading day return. From Figure 1, we could easily find there is a huge increase in listing day return from 2006 to 2007 due to the bull market and following with a jump from 2007 to 2008 because of the stock market crash. Such an abnormal market phenomenon between 2006 and 2008 wascaused by government’s control, or it was politically driven. The results in Table 3 could be explained by abnormal market phenomenon. So that’s why when we consider the whole sample period, the IPOs reforms have no effect on SME Board, but instead we find that the IPOs in SME Board are more underpriced than ever before. However, when we shorten the time span like Figure 2 and Figure 3, especially in Figure 3, IPOs reform can really work in alleviating the underpricing in SME Board as the listing day return in SME Board decreases more than that in Main Board from 2008 to 2009.
The results in Table 4 give us evidence to verify our findings from Figure 1, 2 and 3, and they are also consistent with the results in Table 3. There are three models which are (ⅰ), (ⅱ) and (ⅲ). Row 3 is the interaction item “Time Trend*SME Dummy”, which is the “difference in difference” of the changes in listing day return. In Model (ⅰ) and (ⅱ), the coefficient of the interaction item are 0.52 and 0.84 respectively, which are significant at 5% significance level. But the coefficient of interaction item in Model (ⅲ) is negatively which is -0.23 and significant. It implies that IPOs reforms have effects on firms go public in SME Board in a period from 2008 to 2010. The abnormal market performance from 2006 to 2007 could disturb the results and be contradictory to our hypothesis.
Furthermore, Table 5 is the testing results of Difference in Difference Model in Table 4. In Table 5, the first three interaction items’ coefficient are significant with values of -0.63, 1.1 and 0.52 respectively, but the last three are all insignificant. The IPOs reform did have effects on new stocks’ listing day return in SME Board.
Table 3: Average Listing Day Return Before and After the IPOs Regulation Change in 2009
Standard errors are shown in parentheses. MB represents the Main Board and SME represents Small-and-Mid Enterprise Board. The sample consists of all observations available on listing day. “Before 2009” is as same as “before IPOs reform”. “After 2009” is as same as “after IPOs reform”. “Industry 1” stands for Finance. “Industry 2” stands for Real Estate. “Industry 3” stands for Industries.
All Industries Industry 1 Industry 2 Industry 3
Variable MB (1) SME (2) Difference (SME-MB) (3) MB (4) SME (5) Difference (SME-MB) (6) MB (7) SME (8) Difference (SME-MB) (9) MB (10) SME (11) Difference (SME-MB) (12)
1.Listing Day Return Before
1.65 1.24 -0.41 1.72 1.70 -0.02 0.35 1.16 0.81 1.37 1.19 -0.18 (1.41) (0.81) (1.63) (1.48) (0.78) (1.67) (0.16) (0.51) (0.53) (0.72) (0.81) (1.08)
2. Listing Day Return After
0.28 0.39 0.11 0.22 0.49 0.27 0.34 0.46 0.12 0.31 0.38 0.07 (0.44) (0.52) (0.68) (0.21) (0.37) (0.43) (0.26) (0.33) (0.42) (0.62) (0.54) (0.82)
3.Change in mean Listing Day Return
-1.37 -0.85 0.52 -1.51 -1.22 0.29 -0.01 -0.70 -0.69 -1.06 -0.81 0.25 (1.48) (0.95) (1.76) (1.33) (0.84) (1.57) (0.25) (0.73) (0.77) (1.01) (0.96) (1.39)
Table 4: Difference in Difference Model for IPOs Regulation Change in 2009 in Different Time Span
This table reports the results of Difference in Difference Model in different time periods. The standard errors are shown in parentheses. The independent variable is listing day return of stock i. Time Trend equals to 1 if IPOs date is after 2009. SME Dummy equals to 1 if stock listing in Small-and-Mid Enterprise Board. * and ** represents significance level at 10% and 5%, respectively. Standard errors are shown in parentheses.
2006-2012 2007-2011 2008-2010 Independent Variables (ⅰ) (ⅱ) (ⅲ) Intercept 1.65 2.00 0.46 (0.22) (0.30) (0.07) Time Trend -1.37 -1.77 -0.35 (0.23) (0.31) (0.08) SME Dummy -0.41 -0.65 0.62 (0.22) (0.31) (0.09)
(Time Trend)*SME Dummy 0.52** 0.84** -0.23**
(0.24) (0.31) (0.11)
Number of Observations 894 572 222
R-Square 0.289 0.35 0.27
Robust YES Yes Yes
Table 5: Specification Test For Difference in Difference Model
Replication of the Difference in Difference Model by changing Time Trend dummy. The independent variable is listing day return of stock i. The IPOs date which after certain year equals to 1. For instance, “After 2006 Dummy” equals to 1 when IPOs date is after 2006. SME Dummy equals to 1 if stock listing in Small-and-Mid Enterprise Board. SD stands for standard error s. * and ** represents significance level at 10% and 5%, respectively. “After 2008 Dummy × SME Dummy” is the baseline.
Independent Variables Coefficient SD R-Square After 2006 Dummy × SME Dummy -0.63** 0.19 0.0058 After 2007 Dummy × SME Dummy 1.10** 0.27 0.2014 After 2008 Dummy × SME Dummy 0.52** 0.24 0.2890 After 2009 Dummy × SME Dummy 0.36 0.22 0.2553 After 2010 Dummy × SME Dummy 0.16 0.22 0.1254 After 2011 Dummy × SME Dummy -0.02 0.31 0.0283
5.2 Short-term Performance
We examine the adjusted return of the IPOs in Small-and-Mid Enterprise Board over a period of 21 trading days which is almost one month following the listing day. Results are presented in Table 6, which Panel A consists of the firms listed before 2009 and Panel B is composed of the firms listed after 2009. In Panel A, new firms underperform the market as most of the AdjRet is negative in the first 21 trading days. The adjusted buy-and-hold abnormal return are all positive, starting at 154.53% and reducing to 125.93%. In Panel B, results are quite similar but there are seven days with positive AdjRet. The adjusted buy-and-hold abnormal return are all positive, starting at 40.3% and reducing to 33.03%. Comparing the results of Panel A and Panel B, we could find after the IPOs reform took place in 2009, the abnormal returns “before 2009” are extremely higher than the abnormal returns “after 2009”. Investors are eager to bid up the prices even the prices are unreasonable at the first trading day, and then with market correction of the bubble prices, the stocks’ return underperform the market return in following trading days. However, with less enthusiasm because of the IPOs reform, investors’ speculations have been weakened cause the listing day return to be reasonable and without too much price corrections in the following trading days.
Table 6: The Short-term Performance of the IPOs in SME Board
This table presents the result of the short-term performance of companies after listing in Small-and-Mid Enterprise Board. Ret is the daily return on day t after listing day, which “t=1” stands for the listing day. Mktret is the market return on the corresponding day t, AdjRet is the market adjusted return on the corresponding day t, Mk_BHAR is the market buy-and-hold abnormal return, Unadj_BHAR is the buy-and-hold abnormal return of the IPOs, Adj_BHAR is the market adjusted buy-and-hold abnormal return of the IPOs. The buy-and-hold abnormal is defined as 𝐵𝐻𝐴𝑅𝑇=
1 𝑛 𝑛
𝑖=1 𝑇𝑡=0𝑅𝑒𝑡𝑖,𝑡 and the adjusted
buy-and-hold abnormal is defined as 𝐴𝑑𝑗𝑢𝑠𝑡𝑒𝑑_𝐵𝐻𝐴𝑅𝑇= 1 𝑛 𝑛
𝑖=1 𝑇𝑡=0𝑅𝑒𝑡𝑖,𝑡− 𝑅𝑒𝑡𝑖,𝑇𝑀 .
Panel A: IPO Firms in SME Board Before 2009
Day N Ret Mktret AdjRet Mk_BHA
R Unadj_BHA R Adj_BHA R 1 223 1.5457 0.0004 1.5453 0.0004 1.5457 1.5453 2 223 -0.0176 0.0061 -0.0237 0.0065 1.4957 1.4796 3 223 -0.0147 -0.0002 -0.0146 0.0063 1.4648 1.4457 4 223 -0.0040 0.0013 -0.0053 0.0076 1.4478 1.4269 25 223 -0.0086 0.0036 -0.0122 0.0113 1.4158 1.3898 6 223 -0.0015 0.0012 -0.0028 0.0126 1.4161 1.3864 7 223 -0.0063 0.0019 -0.0082 0.0146 1.4018 1.3665 8 223 0.0017 -0.0004 0.0020 0.0145 1.4040 1.3678 9 223 -0.0067 0.0023 -0.0091 0.0169 1.3886 1.3442 10 223 -0.0007 0.0050 -0.0057 0.0220 1.3857 1.3283 11 223 -0.0008 0.0029 -0.0037 0.0248 1.3881 1.3218 12 223 -0.0013 0.0000 -0.0013 0.0252 1.3858 1.3159 13 223 -0.0039 -0.0005 -0.0035 0.0251 1.3748 1.3079 14 223 0.0029 0.0033 -0.0005 0.0288 1.3798 1.3066 15 223 -0.0021 0.0015 -0.0035 0.0306 1.3763 1.2954 16 223 -0.0034 -0.0017 -0.0017 0.0293 1.3676 1.2903 17 223 0.0008 0.0013 -0.0004 0.0308 1.3698 1.2898 18 223 -0.0077 -0.0026 -0.0052 0.0288 1.3520 1.2776 19 223 -0.0049 -0.0009 -0.0039 0.0284 1.3392 1.2661 20 223 -0.0035 -0.0024 -0.0012 0.0265 1.3353 1.2662 21 223 -0.0011 0.0019 -0.0029 0.0281 1.3309 1.2593
Panel B: IPO Firms in SME Board After 2009
Day N Ret Mktret AdjRet Mk_BHAR Unadj_BHAR Adj_BHAR
1 428 0.4014 -0.0016 0.403 -0.0016 0.4014 0.403 2 428 -0.0058 0.0019 -0.0077 0.0004 0.3886 0.3879 3 428 -0.0101 0.002 -0.0121 0.0025 0.3703 0.3669 4 428 -0.0038 0.0002 -0.004 0.0027 0.3638 0.3597 5 428 -0.0033 0.0002 -0.0035 0.003 0.3600 0.3552 6 428 -0.0041 -0.0013 -0.0028 0.0016 0.3529 0.3500 7 428 0.0005 -0.0006 0.0011 0.001 0.3535 0.3517 8 428 -0.0027 0.0004 -0.0031 0.0015 0.3507 0.3483 9 428 -0.0035 -0.0004 -0.0032 0.001 0.3476 0.3448 10 428 0.0017 0.0003 0.0014 0.0012 0.3499 0.3462 11 428 -0.0032 -0.0021 -0.0011 -0.0007 0.3465 0.3451 12 428 0.0031 0.0012 0.0019 0.0005 0.3509 0.3481 13 428 0.0003 0.0011 -0.0008 0.0014 0.3497 0.3465 14 428 -0.0028 0.0001 -0.0028 0.0015 0.3461 0.3428 15 428 0.0013 -0.0003 0.0016 0.0013 0.3461 0.3439 16 428 -0.0019 -0.0023 0.0004 -0.001 0.3436 0.3445 17 428 -0.002 0.0015 -0.0035 0.0005 0.3383 0.3381 18 428 -0.0014 -0.0001 -0.0014 0.0004 0.3355 0.3355 19 428 0.0013 0.0001 0.0012 0.0004 0.3362 0.3355 20 428 -0.0003 0.0015 -0.0017 0.002 0.3359 0.3325 21 428 -0.0012 -0.0005 -0.0007 0.0015 0.3336 0.3303
5.3 Long-term Performance
After considering the short-term performance after issuing, it would be interesting to see how stocks perform in long-term during the post-issuing period. We study the stocks’ performance for 36 months after going public. We use 21 trading days as a month as well as in Section 5.2. The first month contains 21 trading days following the first trading day, the second month contains the days from day 22 to day 42 following the listing day, and so on. Panel A of Table 6 presents the results for IPOs in Small-and-Mid Enterprise Board before 2009. The market adjusted buy-and-hold return and market adjusted return in the first month are equal to these in the day 21 reported in Table 5 Panel A. It’s interesting to find that the buy-and-hold abnormal returns would decline first and start to increase around Month 16. At the end of Month 36, the adjusted buy-and-hold abnormal return of IPO firms outperform the market by 144.1%. In Panel B, which are IPOs in Small-and-Mid Enterprise Board, the results are quite similar to Panel B. The buy-and-hold abnormal returns decline first and grow at Month 13 and outperform the market by 52.8% in Month 36. Even though both IPOs long-term performance of Panel A & B outperform the market, yet cumulative abnormal returns of IPOs after 2009 underperform cumulative abnormal returns of IPOs before 2009. This can be due to the IPOs reform in 2009, but we think this phenomenon could also be correlative with companies’ operating performance.
Table 7: The Long-term Performance of the IPOs in SME Board
This table presents the result of the long-term performance of companies after listing in Small-and-Mid Enterprise Board. One month contains 21 trading days. Ret is the daily return on month t after listing day. Mktret is the market return on the corresponding month t, AdjRet is the market adjusted return on the corresponding month t, Mk_BHAR is the market buy-and-hold abnormal return, Unadj_BHAR is the buy-and-hold abnormal return of the IPOs, Adj_BHAR is the market adjusted buy-and-hold abnormal return of the IPOs. The buy-and-hold abnormal is defined as 𝐵𝐻𝐴𝑅𝑇=
1 𝑛 𝑛
𝑖=1 𝑇𝑡=0𝑅𝑒𝑡𝑖,𝑡 and the adjusted
buy-and-hold abnormal is defined as 𝐴𝑑𝑗𝑢𝑠𝑡𝑒𝑑_𝐵𝐻𝐴𝑅𝑇= 1 𝑛 𝑛
𝑖=1 𝑇𝑡=0𝑅𝑒𝑡𝑖,𝑡− 𝑅𝑒𝑡𝑖,𝑇𝑀 .
Panel A: IPO Firms in SME Board Before 2009
Month N Ret Mktret AdjRet Mk_BHAR Unadj_BHAR Adj_BHAR 1 223 -0.0011 0.0019 -0.0029 0.0281 1.331 1.259 2 223 -0.0006 0.0014 -0.002 0.0488 1.291 1.189 3 223 -0.006 -0.0005 -0.0055 0.0642 1.195 1.087 4 223 0.0006 -0.0003 0.0009 0.0906 1.208 1.069 5 223 0.0004 -0.0003 0.0007 0.146 1.295 1.061 6 223 -0.0014 -0.0004 -0.001 0.173 1.255 1.036 7 223 -0.0049 -0.0019 -0.003 0.194 1.24 1.022 8 223 -0.001 -0.0009 -0.0002 0.232 1.234 0.996 9 223 0.0034 0.0014 0.002 0.264 1.26 1.006 10 223 0.004 0.0049 -0.0009 0.274 1.262 1.049 11 223 0.0057 0.0053 0.0004 0.272 1.254 1.047 12 223 -0.0019 -0.0007 -0.0012 0.297 1.21 0.992 13 223 0.0048 0.0018 0.003 0.305 1.232 1.021 14 223 -0.0014 -0.0017 0.0003 0.289 1.21 1.044 15 223 0.0037 0.0013 0.0024 0.256 1.208 1.076 16 223 0.003 0.0009 0.0021 0.269 1.275 1.100 17 223 0.0028 0.003 -0.0002 0.258 1.268 1.109 18 223 0.0002 0.0017 -0.0015 0.223 1.27 1.145 19 223 -0.002 0.0004 -0.0025 0.193 1.233 1.136 20 223 -0.0045 -0.0043 -0.0002 0.161 1.21 1.157 21 223 0.0005 0.0009 -0.0004 0.145 1.222 1.151 22 223 0.0012 0.0008 0.0004 0.141 1.262 1.175 23 223 0.0014 -0.0003 0.0016 0.159 1.297 1.178 24 223 0.0034 0.0022 0.0012 0.189 1.39 1.215 25 223 0.0033 0.0026 0.0007 0.239 1.536 1.259 26 223 0.0016 0.0022 -0.0006 0.293 1.635 1.245 27 223 0.005 0.0027 0.0024 0.357 1.777 1.272 28 223 0.0021 0.0008 0.0012 0.402 1.856 1.278 29 223 0.0021 0.0016 0.0004 0.43 1.961 1.316 30 223 -0.0002 0.001 -0.0012 0.461 2.034 1.329 31 223 0.0041 0.0014 0.0027 0.491 2.092 1.332 32 223 0.003 0.0015 0.0015 0.515 2.125 1.35 33 223 0.0031 0.0005 0.0027 0.545 2.214 1.38 34 223 0.0034 0.0026 0.0008 0.576 2.296 1.402 35 223 -0.0004 -0.0009 0.0005 0.585 2.34 1.4 36 223 0.0018 -0.0006 0.0024 0.607 2.435 1.441
Panel B: IPO Firms in SME Board After 2009
Month N Ret Mktret AdjRet Mk_BHAR Unadj_BHAR Adj_BHAR 1 428 -0.0012 -0.0005 -0.0007 0.0015 0.3340 0.3300 2 428 -0.0011 -0.0010 -0.0001 0.0047 0.3260 0.3200 3 428 -0.0006 -0.0003 -0.0003 0.0102 0.3090 0.2950 4 428 0.0011 0.0002 0.0009 0.0126 0.3090 0.2930 5 428 0.0007 -0.0006 0.0014 0.0130 0.3040 0.2910 6 428 -0.0002 -0.0008 0.0006 0.0105 0.3120 0.2990 7 428 -0.0008 -0.0001 -0.0007 0.0112 0.3050 0.2910 8 428 -0.0011 0.0000 -0.0010 0.0078 0.2930 0.2830 9 428 0.0007 0.0006 0.0001 0.0077 0.2970 0.2890 10 428 -0.0021 -0.0007 -0.0014 -0.0071 0.2650 0.2720 11 428 0.0002 0.0002 0.0001 -0.0150 0.2520 0.2620 12 428 0.0003 0.0000 0.0003 -0.0213 0.2290 0.2470 13 428 -0.0009 -0.0010 0.0001 -0.0257 0.2110 0.2380 14 428 0.0007 0.0005 0.0002 -0.0345 0.2040 0.2370 15 428 -0.0001 -0.0001 0.0000 -0.0496 0.1870 0.2410 16 428 0.0016 0.0009 0.0007 -0.0635 0.1760 0.2520 17 428 0.0003 -0.0004 0.0007 -0.0762 0.1750 0.2670 18 428 0.0004 -0.0006 0.0009 -0.0905 0.1610 0.2670 19 428 -0.0002 0.0008 -0.0010 -0.1040 0.1560 0.2810 20 428 0.0009 0.0006 0.0003 -0.1120 0.1480 0.2890 21 428 0.0001 -0.0003 0.0004 -0.1220 0.1410 0.2970 22 428 0.0007 0.0005 0.0002 -0.1250 0.1590 0.3160 23 428 -0.0004 -0.0006 0.0001 -0.1340 0.1620 0.3360 24 428 0.0022 0.0002 0.0020 -0.1350 0.1680 0.3440 25 428 0.0018 0.0007 0.0011 -0.1330 0.1800 0.3560 26 428 -0.0009 -0.0014 0.0005 -0.1320 0.2120 0.3880 27 428 0.0009 -0.0001 0.0009 -0.1300 0.2300 0.4000 28 425 -0.0001 0.0002 -0.0003 -0.1260 0.2590 0.4190 29 422 0.0011 0.0000 0.0012 -0.1150 0.2750 0.4330 30 414 -0.0002 -0.0005 0.0002 -0.1170 0.2850 0.4500 31 407 0.0011 0.0005 0.0006 -0.1190 0.2980 0.4730 32 404 0.0013 0.0015 -0.0002 -0.1090 0.3230 0.4860 33 396 -0.0005 0.0010 -0.0015 -0.1090 0.3270 0.4950 34 387 0.0014 0.0016 -0.0002 -0.1070 0.3440 0.5200 35 374 0.0022 0.0000 0.0022 -0.1150 0.3080 0.5030 36 370 0.0021 0.0004 0.0016 -0.1060 0.3440 0.5280
5.4 Operating Performance
Table 8 presents the results of the changes of post-issuing operating performance surrounding the issuing year such as operating return on assets (ROA), cash flow over assets (CFOA), sales growth rate (Sale_G) and capital expenditure growth rate (CE_G) in Small-and-Mid Enterprise Board. We also measure the changes in an event window from one year before IPOs to three years after listing. Panel A is for IPO firms listed between 2006 and 2009. From the results of Panel A, only operating return on assets declines, all the other three variables perform better and better after listing. Panel A is for IPO firms listed between 2009 and 2012. It’s interesting that the results of Panel B are quite similar to the results of Panel A. Operating return on assets decline over three years since IPOs comparing with the -1 year (the year before IPOs), whereas cash flow over assets, sales growth rate and capital expenditure growth rate increase significantly. Obviously, excluding the operating return on assets, the other operating performance measurement factors are highly performing over the next three years after listing. It indicates that the buy-and-hold abnormal returns are positively relative with the operating performance from listing day onwards. Under the similar operating performance, the cumulative abnormal returns in different time span in Tale 7 are quite different, firms went public before 2009 experienced a much higher buy-and-hold abnormal returns than those firms which went public after 2009. Operating performance could partially explain stock-outperform-market return, but in a case that reduction of stocks cumulative returns with similar operating performance, IPOs reform is the extremely important political factor to set up the reasonable offering price.
Table 9 contains the results of cross-sectional regressions of post-issuing stock returns on operation performance. Panel A reports the results of firms listed in Small-and-Mid Enterprise Board before 2009. In Panel A, three-year buy-and-hold abnormal return is positively related to capital expenditure growth rate, two-year and three-year buy-and-hold abnormal returns have positive relations with sales growth rate, and one-, two- and three-year buy-and-hold abnormal returns are positive with operating return on assets. In Panel B, it’s as similar as that in Panel A. One-, two- and three-year buy-and-hold abnormal returns are all positively related to operating return on assets and sales growth rate. Hence, the changes of operating performance is essential in adjusting stocks prices over the long run. It also verifies that Chinese stock market are not purely speculation-driven.
Table 8: The Changes in Operating Performance of IPO Firms Surrounding the Issuing Year This table presents the mean change or growth in operation performance. Year 0 is the fiscal year when the firm goes public. SME Board stands for the Small-and-Mid Enterprise Board. Panel A: IPO Firms in SME Board Between 2006 and 2009
Measure of Operating Performance
Year Relative to the IPO Year
From -1 to 0 From -1 to 1 From -1 to 2 From -1 to 3
Operating Return on Assets
Mean Change (%) -4.28 -5..95 -7.04 -6.34
Number of
Observations 223 223 223 223
Cash Flow Over Assets
Mean Change (%) 13.80 7.21 4.61 2.78
Number of
Observations 223 223 223 223
Sales Growth Rate
Mean Change (%) 32.00 64.79 95.20 167.61
Number of
Observations 223 223 223 223
Capital Expenditure Growth Rate
Mean Change (%) 49.99 141.62 240.73 362.38
Number of
Observations 223 223 223 223
Panel B: IPO Firms in SME Board Between 2009 and 2012 Measure of Operating
Performance
Year Relative to the IPO Year
From -1 to 0 From -1 to 1 From -1 to 2 From -1 to 3
Operating Return on Assets
Mean Change (%) -7.24 -8.86 -9.11 -9.57
Number of
Observations 428 380 281 85
Cash Flow Over Assets
Mean Change (%) 22.96 16.09 9.05 5.39
Number of
Observations 428 380 281 85
Sales Growth Rate
Mean Change (%) 21.89 51.21 94.81 151.48
Number of
Observations 428 380 281 85
Capital Expenditure Growth Rate
Mean Change (%) 34.84 101.99 242.17 396.84
Number of
Table 9: Cross-sectional Regressions of Post-issuing Stock Returns on Operation Performance
This table presents the regression results of the post-issuing stock price performance on the post-issuing operating performance. The dependent variable is one-, two- and three-year buy-and-hold abnormal return after listing. ∆𝑅𝑂𝐴𝑖,−1 𝑡𝑜 𝑡 is the change of return on asset
in period -1 to t. ∆𝐶𝐹𝑂𝐴𝑖,−1𝑡𝑜 𝑡 is the change of cash flow on total assets in period -1 to t.
∆𝑆𝐴𝐿𝐸_𝐺𝑖,−1 𝑡𝑜 𝑡 is the change of growth rate of sales in period -1 to t. ∆𝐶𝐸𝑖,−1 𝑡𝑜 𝑡
represents the change of capital expenditure in period -1 to t. * and ** represents significance level at 10% and 5%, respectively. Standard errors are shown in parentheses. Panel A: IPO Firms in SME Board From 2006 to 2009
Regression results for one-year buy-and-hold abnormal returns
Variable Intercept ΔROA ΔCFOA ΔSales_G ΔCE_G R-Square
Coefficient 1.63** 7.04** 0.113 SD (0.18) (2.04) Coefficient 1.27** -0.85 0.005 SD (0.13) (0.78) Coefficient 1.00** 0.32 0.040 SD (0.13) (0.21) Coefficient 1.20** 0.01 0.0004 SD (0.10) (0.02) Coefficient 1.44** 6.49** 0.22 0.01 0.133 SD (0.19) (2.01) (0.18) (0.02) Coefficient 1.06** -0.46 0.31 -0.01 0.043 SD (0.15) (0.74) (0.21) (0.03)
Regression results for two-year buy-and-hold abnormal returns
Variable Intercept ΔROA ΔCFOA ΔSales_G ΔCE_G R-Square
Coefficient 1.53** 2.03* 0.014 SD (0.13) (1.07) Coefficient 1.39** 0.10 0.0001 SD (0.09) (0.69) Coefficient 1.12** 0.29** 0.054 SD (0.11) (0.10) Coefficient 1.32** 0.03 0.008 SD (0.10) (0.02) Coefficient 1.22** 1.42 0.26** 0.01 0.061 SD (0.16) (1.07) (0.11) (0.02) Coefficient 1.07** 0.52 0.28** 0.01 0.057 SD (0.13) (0.69) (0.11) (0.02)
Table 9: Cross-sectional Regressions of Post-issuing Stock Returns on Operation Performance (Continue)
Regression results for three-year buy-and-hold abnormal returns
Variable Intercept ΔROA ΔCFOA ΔSales_G ΔCE_G R-Square
Coefficient 6.26** 2.83** 0.051 SD (1.52) (0.17) Coefficient 2.42** 0.62 0.002 SD (0.12) (0.97) Coefficient 2.27** 0.10 0.018 SD (0.16) (0.08) Coefficient 2.21** 0.06** 0.040 SD (0.13) (0.02) Coefficient 2.59** 7.11** 0.03 0.07** 0.107 SD (0.18) (1.45) (0.07) (0.02) Coefficient 2.09** 1.38 0.05 0.06** 0.052 SD (0.15) (1.08) (0.07) (0.02)
Panel B: IPO Firms in SME Board From 2009 to 2012
Regression results for one-year buy-and-hold abnormal returns
Variable Intercept ΔROA ΔCFOA ΔSales_G ΔCE_G R-Square
Coefficient 0.35** 1.50** 0.029 SD (0.06) (0.49) Coefficient 0.25** -0.17 0.002 SD (0.05) (0.19) Coefficient -0.08 0.59** 0.183 SD (0.05) (0.12) Coefficient 0.20** 0.02 0.003 SD (0.04) (0.02) Coefficient -0.003 0.67 0.58** -0.01 0.190 SD (0.08) (0.46) (0.13) (0.02) Coefficient -0.08 0.02 0.61** -0.02 0.185 SD (0.06) (0.17) (0.12) (0.02)
Table 9: Cross-sectional Regressions of Post-issuing Stock Returns on Operation Performance (Continue)
Regression results for two-year buy-and-hold abnormal returns
Variable Intercept ΔROA ΔCFOA ΔSales_G ΔCE_G R-Square
Coefficient 0.31** 2.55** 0.076 SD (0.06) (0.48) Coefficient 0.09** -0.20 0.002 SD (0.04) (0.25) Coefficient -0.25** 0.35** 0.286 SD (0.07) (0.08) Coefficient 0.05 0.01 0.005 SD (0.04) (0.01) Coefficient -0.07 1.72** 0.33** -0.002 0.320 SD (0.09) (0.44) (0.08) (0.005) Coefficient -0.24** 0.04 0.35** -0.008 0.288 SD (0.08) (0.21) (0.08) (0.005)
Regression results for three-year buy-and-hold abnormal returns
Variable Intercept ΔROA ΔCFOA ΔSales_G ΔCE_G R-Square
Coefficient 0.84** 5.21** 0.113 SD (0.19) (1.10) Coefficient 0.41** -1.22 0.026 SD (0.14) (0.91) Coefficient -0.41** 0.49** 0.508 SD (0.17) (0.14) Coefficient 0.13 0.05* 0.056 SD (0.12) (0.03) Coefficient -0.10 2.51** 0.48** -0.01 0.536 SD (0.23) (1.03) (0.17) (0.02) Coefficient -0.37** -0.03 0.51** -0.02 0.512 SD (0.15) (0.40) (0.16) (0.02)
6. Conclusion and Limitation
6.1 Conclusion
This paper mainly investigates the changes of underpricing, short-term and long-term performance of 651 IPOs in Small-and-Mid Enterprise Board before and after the IPOs reform in 2009. We find that there is a large underpricing of IPOs in Small-and-Mid Enterprise Board before 2009, but declines significantly after IPOs reform on Small-and-Mid Enterprise Board. It’s transparently that the result in this paper ispartiallyconsistent with the results in post studies, the average return of Small-and-Mid Enterprise Board IPOs after 2009 on the listing day is smaller than that before 2009. When we take market climate (such as bull and bear market effects) into consideration, government’s control has huge effects on reducing IPOs underpricing as the dif-in-dif value in events window from 2008-2010 is -0.23 which means that it significantly alleviates the underpricing.
Furthermore, we also find the evidence of both IPOs before 2009 and IPOs after 2009 in Small-and-Mid Enterprise Board are overacted in the first month but IPOs after 2009 outperform the market by 33.03% whereas IPOs before 2009 outperform the market by 125.93%. At the end of the third year after listing, IPOs before 2009 outperform the market by 144.1% and IPOs after 2009 outperform the market by 52.8%. Therefore, the regulation changes also have impacts on short-term and long-term stock price performance, which set up a more reasonable offering price after 2009. We also find that the long-term buy-and-hold abnormal returns are positively related to operating return on assets and sales growth rate. This indicates
