Market conditions and the optimal IPO allocation mechanism in China

Abstract

This paper investigates the impact of market conditions of market return and volatility on choosing an IPO mechanism, using data of 942 IPOs on either Shanghai or Shenzhen stock exchanges of China from 1994 to 2003. We find, on average, the issuers are more likely to have their IPOs offering and listing during times of high market return and low market volatility. The fixed price procedure of the secondary market proportional offering is optimal in minimising the underpricing and cross-sectional variation of the first day returns. The bookbuilding procedure is optimal in counteracting adverse conditions created by low market profitability, high market volatility and uncertainty induced by the time ‘gap’ from offering to listing. By comparing the advantages between the secondary market proportional offering and bookbuilding procedures, the latter is preferred.

Introduction

Fixed price, bookbuilding and auctions are three primary allocation mechanisms¹ employed in initial public offer (IPO) markets around the world. In the fixed price case, the issuer and underwriter set an unchanged (fixed) price prior to the IPO. In the bookbuilding scenario, the issuer and underwriter negotiate a price range and then solicit potential investors who nominate both the price they are willing to pay and their desired subscription number of shares. Accordingly, the offer price comes from a process of price discovery, related to demand at various prices. Finally, in the auction form of IPO allocation, the issuer and underwriter set a minimum acceptable price and the final offer price is determined by considering the investor bids.

In this paper, we investigate the IPO market in China, with a special emphasis on the allocation mechanism and the impact of market conditions. While it is true that China's IPO markets have witnessed a growth in the bookbuilding approach over recent years, since the establishment of China's stock market in the early 1990s, the fixed price method has been the dominant allocation mechanism. Indeed, it is quite interesting to note that the fixed price mechanism employed in China's IPO market has manifested in several different forms and their patterns of usage have varied over time. Specifically, the different variants of fixed price IPO allocation observed in China include: private placement; subscription warrants; saving linkage offerings; online fixed price offerings; and secondary market proportional offerings. A private placement variant was the unique method used before the establishment of Shanghai and Shenzhen stock exchanges. Limited and unlimited lottery forms were widely used in the first four years after the markets opened. Thereafter, saving linkage offerings, online fixed price offerings, and offerings proportional to the existing secondary market holdings have been employed.

An increasing number of studies focus on IPO allocation mechanisms. Benveniste and Spindt (1989) suggest that the American bookbuilding mechanism is efficient, since it encourages investors to reveal their beliefs about the issuer's value. The auction mechanism introduces asymmetrically informed investors to reveal what they know to the underwriter, and in turn, allocation priority will be given to an underwriter's regular investors. Welch (1992) shows that the fixed price mechanism used in some European countries can cause ‘informational cascades’. That is, investors will tend to update their beliefs about the value of issued shares by observing the investment choice of previous investors.

Some studies argue for a relation between allocation mechanisms and the magnitude of IPO underpricing. Leleux and Paliard (1996) show that the initial returns are significantly higher for firms allocating IPOs through the fixed price mechanism than for those using an auction-like approach. Ljungqvist et al. (2003) compare data on bookbuilding and public offer IPOs for a large number of countries. They find bookbuilding is substantially more expensive than public offers and that it cannot reduce underpricing. Derrien and Womack (2003) show that the auction mechanism is associated with less underpricing and lower variance of the first day returns. Also, IPO auctions tend to incorporate more information from recent market conditions into price than do other mechanisms.

A range of other studies explores the factors that influence the determination of taking a specific IPO allocation mechanism. Benveniste and Busaba (1997) conclude that firms with more price uncertainty, or a greater concern for risk, are more likely to prefer a fixed price offering. Alternatively, firms with greater future capital needs are more likely to prefer the bookbuilding mechanism. Vandemaele (2003) finds that the choice for an auction-like procedure appears to be positively related to firm valuation uncertainty at the IPO. The likelihood of choosing an auction-like procedure decreases as the reputation of the associated investment bank increases, whereas, the likelihood increases with the number of secondary shares sold by venture capitalists and investment banks.

Su and Fleisher (1999) show, using a sample of 308 of China's IPOs issued in 1987–1995, that underpricing is on average larger under the various lottery mechanisms than it is in the auction setting. Ti (2003) tests 354 Chinese IPOs issued in 1999–2002, showing that the fixed price approach exhibits lower odds-adjusted initial returns than do IPOs that use bookbuilding.²

In this paper, we study a sample of 942 IPOs offered in China over the period from 1994 to 2003 on either the Shanghai or Shenzhen stock exchanges. Initial analysis of variance (ANOVA) on the first day returns of IPOs cannot distinguish a statistical difference between using the fixed price and bookbuilding mechanisms. Accordingly, we seek further insights into IPO allocation mechanisms by partitioning the fixed price method into six distinct sub-groups. Moreover, we investigate the impact of market conditions (as reflected by return and volatility) on IPO underpricing across the different fixed price allocation procedures, and compare this to similar analysis for the bookbuilding procedure. We furthermore explore the determinants of choosing an IPO allocation procedure on China's stock market. Our study provides robust evidence on the optimal characteristics of the secondary market proportional offering and the bookbuilding procedures. To the best of our knowledge, the findings reported herein have not been previously documented in the literature on China's stock market.

The remainder of this paper is organised as follows. Section 2 interprets the evolution of China's IPO allocation mechanisms and new classification of fixed price procedures. Section 3 investigates the impact of market conditions on IPO underpricing and the efficiency of IPO procedures in the Chinese setting. Section 4 analyses the determinants of choosing an IPO allocation procedure. Section 5 concludes this study.