Investor demand for IPOs and aftermarket performance: Evidence from the Hong Kong stock market

Abstract

In this study, we examine the relation between pre-offering demand and aftermarket performance of IPO firms in the Hong Kong stock market. We find that IPOs with high investor demand realize large positive initial returns but negative long-run excess returns, while IPOs with low investor demand realize negative initial returns but positive long-run excess returns. This result suggests that (1) pre-offering demand for IPOs is at least partly driven by investors’ over- or underreactions to information about firms’ post-issuance prospects, and (2) while high- and low-demand IPOs are not priced at their intrinsic values in early aftermarket trading, eventually their true values are reflected in their pricing.

Introduction

On average, initial public offerings (IPOs) of common stocks earn abnormally high initial returns (Ibbotson, 1975, Ritter, 1984, Loughran et al., 1994), but significantly underperform the market in the long run (3–5 years) (Aggarwal and Rivoli, 1990, Loughran and Ritter, 1995). These IPO return anomalies are observed in stock markets around the world. According to Ritter (1991), the short-run underpricing/long-run underperformance phenomenon represents an unresolved mystery in the IPO literature.¹ Although these anomalies have prompted numerous academic researchers to investigate the causes of IPO underpricing, several puzzles remain. For instance: why are some IPOs underpriced while others are not, and why do underpriced IPOs underperform after they become seasoned? Moreover, little attention has been paid to the differences in long-run performance between those IPOs that are under- or over-priced.

In this study, we address these gaps in the literature by characterizing the relation between IPO firms’ pre-offering demand and aftermarket performance. In particular, we are interested in the correlation between an IPO's demand and its initial stock return as well as its long-run return. Since investors’ assessments result in downward-sloping demand curves for IPO stocks, investor demand should affect IPO performance.

Focusing mainly on the causes of IPO underpricing, the theoretical work of Rock (1986), Aggarwal and Rivoli (1990) and Chowdhry and Sherman (1996) provides predictions regarding this demand-performance relation. According to Rock (1986), informed investors with superior information have the ability to distinguish between “good” and “bad” IPOs. Hence, informed investors subscribe only to high quality issues, leading to high demand (demand of informed investors plus uninformed investors) for good IPOs and low demand (demand of only uninformed investors) for bad IPOs. Rock's hypothesis implicitly suggests, therefore, that high-demand IPOs exhibit relatively higher returns both during the first days of trading and in the long run. Chowdhry and Sherman (1996) also posit a positive relation between investor demand and underpricing of IPOs, arguing that a severely underpriced IPO will attract a large number of investors who seek to exploit the resulting short-run profit opportunities. Their model suggests that high-demand IPOs experience a relatively large positive return on the first post-IPO trading day, but that the difference in post-issuance performance between high- and low-demand IPOs occurs only in the short run, with mispricings potentially corrected rapidly in opening-day trading. Finally, Aggarwal and Rivoli (1990) argue that IPO underpricing may be positively related to long-run underperformance as a result of investor irrationality (see also Rajan and Servaes, 2002, Ljungqvist et al., 2007).

A number of empirical studies also find evidence consistent with a relation between investor demand and IPO performance. Hanley (1993) demonstrates that the relation between an IPO's offer price and preliminary filing range predicts the direction of initial stock returns in US stock markets. Cornelli and Goldreich (2003) find that oversubscription for an IPO is positively correlated with aftermarket returns. Kandel et al. (1999) document a positive relation between IPO demand schedules and abnormal returns on the first trading day for a small sample of Israeli IPOs.² Overall, the above studies all indicate that pre-offering demand for IPOs plays a nontrivial role in the pricing of these IPOs the first trading day. However, it is worth pointing out that there is virtually no direct empirical evidence on the relation between the level of investor demand and the long-term performance of IPOs.

To study the relation between pre-IPO investor demand and post-IPO performance, we employ a unique data set for IPOs between 1993 and 1997 from the Stock Exchange of Hong Kong (SEHK), which provides oversubscription information at IPO offer prices.³ The Hong Kong IPO market is a suitable market for our study for two reasons. First, McGuinness, 1992, McGuinness, 1993 reports that IPOs in Hong Kong earn abnormal returns in the first day of trading and significantly underperform the market in the long run, as is the case in other markets. Thus, the results are largely generalizable to other markets of interest. Second, it is well known that investor demand for IPOs is fairly volatile in the Hong Kong stock market. For instance, some hot IPOs are oversubscribed by as much as 1000 times the number of shares offered, whereas, some cold IPOs have to be postponed or even cancelled because of undersubscription. These large variations in subscription ratios provide us an excellent setting in which to study the relation between IPOs’ investor demand and aftermarket performance.

Not surprisingly, we find evidence consistent with a strong relation between IPOs’ pre-offering demand and both short- and long-run post-issuing performance. The IPOs with high investor demand realize large positive initial returns but negative longer-run excess returns, while the IPOs with low investor demand realize negative initial returns but perform relatively well in the longer run.

These results cannot be explained by the information asymmetry hypothesis or the underpricing hypothesis. Although these two hypotheses predict a positive relation between investor demand and IPO initial returns, neither hypothesis can explain the observed differences in long-run performance between high- and low-demand IPOs. However, our empirical results are consistent with the speculative bubble hypothesis. Investor demand for an IPO is largely driven by investors’ overoptimistic or overpessimistic reactions to pre-offering information about an IPO's prospects. Consequently, both high- and low-demand IPOs are not priced at their intrinsic values in early aftermarket trading, but eventually their true values are reflected in their pricing.

The remainder of this paper is organized as follows. Section 2 provides a brief description of the data. Section 3 presents the results for the relation between IPO investor demand and short-run performance and Section 4 presents the results for the relation between IPO investor demand and long-run performance. Section 5 provides a discussion of the results. Finally, Section 6 presents the concluding remarks.