بازده اولیه IPO شامل چیست : شواهد از بازار چین

We studied the IPO price and long-term performance in China after the adoption of the book-building pricing mechanism. Using comparable firm value, we separated the IPO initial returns into pre-market deliberate underpricing and aftermarket overpricing. This separation enables us to clearly test different theories regarding high IPO initial returns. We find little evidence supporting the classic information theory on IPO underpricing but strong evidence supporting the behavioral arguments regarding IPO overpricing. Even though the results are specific to the Chinese market, we find some general results on what composes and drives IPO initial returns that have been lacking in the IPO literature.

China's stock market experienced an explosive growth in 2006 and 2007 and, as a result, became the fourth-largest capital market in the world. Together with the market bull-run came intensified IPO activities with a record amount of money raised and number of firms listed. More importantly, this was the time period in which the book-building method was adopted in setting IPO prices, which is the most popular pricing mechanism used worldwide. Previous studies on China's IPO market largely neglected the fact that Chinese IPO offer price is tightly controlled, and sometimes actually fixed, by the regulator. There is a large literature on IPOs, particularly on the US stock markets, regarding what determines IPO initial returns and the long-term performance. Several rational theories, which are mostly information asymmetry-based, have been proposed in the literature to explain why firms leave money on the table by setting the offer price low. There is a general agreement that firms and underwriters encourage participation and price discovery by offering IPO shares at a discount to fair value. Without that inducement, an investor may wait to buy shares in the aftermarket and if all investors wait, the IPO would fail. For example, according to Rock (1986), underpricing is necessary to induce uninformed investors to participate in IPO offerings when faced with adverse-selection by informed investors. Benveniste and Spindt, 1989 and Benveniste and Wilhelm, 1990 focus on the critical role that informed investors play in setting the offer price. Underpricing enables the issuing manager to reward sophisticated investors who share their information through a larger allocation1. The empirical challenge to rational theories of underpricing comes from the following observations. First, there is some evidence supporting the view that the closing price on the first day may not reflect fair value. A long-term investor who buys shares of a firm right after it goes public may realize abnormal negative risk-adjusted returns. Ritter, 1991 and Loughran and Ritter, 1995 provide empirical support for this observation. However, Brav et al., (2000) argue that the long-run underperformance of IPO may be due to insufficient correction for risk. They found that IPO firms have long-run returns that are similar to non-issuing firms matched by firm size and book-to-market ratios. Loughran and Ritter (2000) nevertheless point out that size and book-to-market matching schemes may miss significant market anomalies. Second, using a sample of more than 2000 IPOs during 1980–1997, Purnanandam and Swaminathan (2004) find that on average the offer price substantially exceeds the corresponding intrinsic value computed using multiples of firms in the peer group of the issuing firm. Furthermore, overvalued IPOs have large first-day returns but low long-run risk-adjusted returns. This suggests that investor sentiment and irrational over-optimism may drive the IPO market price as well as the offer price. There is still no consensus on whether IPO initial return represents rational underpricing or irrational sentiment in the US market or both. The ambiguity puts empirical studies in an awkward situation. When a factor is found to be associated with IPO initial returns, we are not sure which theory should be used to identify this factor. For example, IPO firm size has a negative relationship with IPO initial return. The rational theory suggests that size is a proxy for information asymmetry. Small firms suffer more from asymmetric information; therefore, investors demand more underpricing. Behavioral finance theory may argue that investor over-optimism bids up small firm prices, which are much easier to manipulate. China's market has experienced the largest IPO initial return in the world, with an average easily exceeding 100% for most years. This magnitude and variation may provide the opportunity to separate the deliberately underpriced and optimistically overpriced components in an IPO initial return. If the separation is possible, it will lead to a much more conclusive empirical analysis. In order to achieve this separation, we need to estimate the fair value (or intrinsic value) for an IPO issue. Following common practice in the literature, we use comparable firm P/E ratio to determine the fair value of the IPO. We find that the IPO offer price is less than its intrinsic value, which in turn is less than the IPO first-day market price. Therefore, China's IPO initial returns seem to exhibit both deliberate underpricing and irrational overpricing. We conduct further regressions on both the underpricing and overpricing components to determine which factors drive them. Because we have a clear separation, we can explain underpricing based on rational theory and overpricing based on behavioral theory, therefore avoiding the ambiguity problem often encountered in IPO research. We find little empirical support for rational theory but strong support for behavioral theory. Furthermore, only the overpricing component determines IPO underperformance in the long run. Another contribution of our work is that we clarify China's IPO phenomena in comparison to other markets discussed in the literature. There is a substantial amount of research studying IPO underpricing in China that relates it to the general IPO literature, such as Mok and Hui, 1998, Su and Fleisher, 1999 and Chang et al., 2008. Much of the China IPO research uses data before 2006, when the dominant IPO pricing scheme was not market-oriented. Therefore, the empirical results from these studies are not directly comparable to the general IPO literature in which the offer price is determined by the market, mostly through book-building processes. When the IPO offer price is arbitrarily set by the regulator, it is hard to derive any economically meaningful results. We therefore limit our IPO sample to the period during which book-building is adopted as the only pricing mechanism in China. The remainder of the article is organized as follows. Section 2 is a brief description of how China's IPO offer price has been historically set. Section 3 describes sample data and the research methodology. Section 4 reviews and interprets our empirical results regarding IPO initial returns and long-term returns. We conclude and suggest future research directions in Section 5.

Using about 200 China IPOs, we find that IPO initial returns have both underpricing and overpricing components. The separation of underpricing and overpricing components allows us to retest propositions from classic rational theory and behavioral theory that explain IPO initial returns. We find little support for the former theory and strong support for the latter theory in China's market. Although the findings are to some extent China-specific, this exercise helps clarify previous empirical IPO research on which factors determine IPO initial returns and through which mechanisms they exercise these impacts. None of the variables related to rational theory is significant with the expected sign. Offer price adjustment is significant but not positively correlated with underpricing as predicted in Benveniste and Spindt (1989). However, due to the immature development of the book-building process in China, offer price adjustment may not relate to information asymmetry and rational theories at all. We also find that positive pre-market returns do not lead to higher underpricing as predicted by proxy theory. Rather, it reduces underpricing. This indicates that issuer and underwriter take the window of opportunity opened by IPO issuance to maximize the offer price when investor sentiment is high. However, positive market sentiment strongly increases overpricing. Other variables related to investor sentiment that is widely used in recent literature, namely, individual investor demand and trading volume, also have a positive impact on IPO overpricing. Additional support for behavioral arguments comes from IPO long-term returns. We find that IPOs on average underperform in the long run. More importantly, only the overpricing component, rather than the underpricing component, of the IPO initial return predicts IPO long-term underperformance. We understand that our results may be driven by some special features in China's market. In fact, as summarized by Yong (2007), most past studies on Asian IPOs deal with factors unique to Asian markets. Nevertheless, we believe our method has broad implications for future IPO research. The fact that rational theories have little power in explaining the IPO return in China's market does not mean that they are not relevant. It is possible that evidence for rational theories is distorted by the underdeveloped pricing mechanism in China's IPO market. It would be more valuable if we can separate pre-market underpricing and post-market overpricing in the US market and test different theories, since it is the US market that gave rise to different IPO underpricing theories. In addition, some of the other puzzling empirical results can be reexamined. For example, the literature has mixed results on the impacts of underwriter reputation4 as well as the presence of ventured capital5 on IPO initial returns. Again, if the impact of underwriters and VCs are separately analyzed along underpricing and overpricing components, a more consistent set of empirical results may be reached. Our results are also affected by the evolving book-building mechanism in China. Ritter and Welch (2002) point out that recent IPO research focuses on book-building and share allocation. Ma and Faff (2007) study the IPO allocation methods in China, but the results are largely confined within the fixed-price system. When the book-building process in China finally converges to the world standard, further research should re-examine rational theories in the context of China's market. For behavioral theory, an analysis of detailed data on individual investor transactions can help better identify investor sentiment and overpricing. Finally, further research should conduct long-term return analysis more carefully, including the use of Fama–French three-factor modeling. The difficulties of regressions on long-term stock returns are widely understood, and so far, there are no perfect econometric tools to address them.6 Such studies will depend on the further development of more comprehensive econometric methods as well as the availability of longer-term databases on China's market.