Aftermarket volatility and underpricing of Canadian initial public offerings

Canadian Journal of Administrative Sciences, Sep 2002 by Vijay Jog, Liping Wang

Abstract

This paper investigates aftermarket volatility and underpricing of the Initial Public Offerings (IPOs) listed on the Toronto Stock Exchange during the period 19901999. Our results indicate that aftermarket volatility has increased significantly but the average degree of underpricing has not increased. The decomposition of the aggregate aftermarket volatility in market, industry, and firm-specific components reveals that the increase is mostly attributable to firm-specific risk. We also find that the volatility of IPOs is smaller than that of comparable (matched) seasoned stocks. Although our results show an overall positive relationship between IPO underpricing and aftermarket volatility, we find that this relationship has become considerably weaker in recent years.

Resume

Cet article etudie la volatilite apres-bourse et la sousevaluation des titres suite ii un appel public a l'epargne (APE) a la bourse de Toronto durant la periode 19901999. Nos resultats indiquent que la volatilite apresbourse a augmente de maniere significative, mais que le niveau moyen de la sous-evaluation quant a lui est reste intact. La decomposition de la volatilite totale apresbourse selon ses composantes de marche, d'industrie et celle propre aux firmes, indique que l'augmentation de la volatilite peut etre attribuee aux risques specifiques aux firmes. Nous trouvons aussi que la volatilite des APE est plus faible que celle des emissions d'actions comparabies deja listees. Bien que nos resultats montrent une relation positive globale entre la sous-evaluation des APE et la volatilite subsequente a l'emission, nous estimons que cette relation s'est considerablement affaiblie au cours des dernieres annees.

In almost all papers related to documenting and explaining IPO underpricing, aftermarket return volatility plays an important role.' There is a common impression that, in recent years, aftermarket volatility of Canadian IPOs has increased.2 The U.S. results document a noticeable increase in the degree of underpricing in recent years.3 In this paper, we empirically investigate the evidence using Canadian IPOs issued during the years 1990-1999, not only in terms of underpricing and total aftermarket volatility, but also relative to market and industry indices and matched seasoned stocks.4 We also investigate the relationship between IPO initial return (underpricing) and aftermarket volatility, to test whether the degree of underpricing (on a cross-sectional basis), as theoretically expected, has increased in concert with the increased IPO aftermarket volatility.5

Our results show that IPO aftermarket volatility has indeed increased more rapidly than either market/industry indices or comparable seasoned stocks. A decomposition of total return variance (our proxy for aftermarket volatility) among market, industry, and firm-specific components indicates that the main source of this higher IPO volatility is increasingly firm specific.

However, contrary to the U.S. results, we see no noticeable increase in the average degree of underpricing for Canadian IPOs. Although we find a positive relationship between IPO initial return and volatility, this relationship has become much weaker in the last five years. These results indicate that the IPO environment has become increasingly risky but the risk has not manifested itself in increased initial returns for investors as predicted by theoretical models (A la Rock, 1986).

The remainder of the paper is organized as follows. The next section contains a brief review of the literature regarding aftermarket volatility and its implications for IPO underpricing. Then we describe the hypotheses, data, and the research methodology followed in this paper. The empirical results are then presented and discussed. The final section contains conclusions and some suggestions for future research.

Background

In the past 20 years, numerous researchers have provided convincing evidence that, on average, IPOs are underpriced. For example, Ritter (1998) reports that for the more than 13,000 firms that went public in the U.S. during 1960-1996, an average new issue was trading at 15.8% higher than its offering price shortly after public trading started. In a recent paper, Loughran and Ritter (2002b) document a noticeable increase in the average degree of underpricing in the U.S. IPOs, with the highest increase recorded in IPOs during 1999 and 2000.6 However, it is also clear that this increase can be mostly attributed to the tech stocks issued as a result of the Internet bubble and is, thus, both time and period specific.7 Jog and Srivastava (1997/98) show an average underpricing of 8.26% based on their sample of 399 IPOs listed on the TSE during the period 1971-1995. Other research has also shown that IPO underpricing is an international phenomenon, although the degree of underpricing varies from country to country (Jenkinson, Ljungqvist, & Wilhelm, 1999; Loughran, Ritter, & Rydqvist, 1994).

A number of hypotheses have been offered to explain this observed underpricing in efficient capital markets. Of these, a model developed by Rock (1986) has received considerable attention. Rock's model is based on the existence of information asymmetry between two groups of investors: informed and uninformed investors. According to Rock, informed investors invest in information production and subscribe to IPOs only when they believe the equilibrium aftermarket price would be higher than the offer price. The uninformed investors, on the other hand, may subscribe to every IPO, as they are unable to distinguish a priori between underpriced or overpriced IPOs. Consequently, underpriced IPOs would be more commonly oversubscribed and overpriced IPOs would be undersubscribed. Because oversubscribed issues are subject to quantity rationing, uninformed investors may find that they are eventually allocated more overpriced issues. In other words, the uninformed investors face a "winner's curse": their expected return from allocated issues is less than their expected return from submitting their purchase orders. Realizing this, uninformed investors would no longer subscribe to IPOs unless the IPOs are, on average, underpriced. In order to keep uninformed investors in the IPO market, issuers must set their issue price lower than the expected aftermarket price. In a sense, underpricing can be considered as compensation to uninformed investors for their continuing participation. Beatty and Ritter (1986) build on this asymmetric information model introduced by Rock (1986) and show that the "degree of underpricing is directly related to the ex ante uncertainty about the value of the issue ... because as ex ante uncertainty increases, the winner's curse problem intensifies" (pp. 215-216). The uninformed investor will therefore demand higher average underpricing (more money to be "left on the table") as ex ante uncertainty increases. Using two proxies for ex ante uncertainty (number of uses of proceeds and the reciprocal of gross issue proceeds), they show that there is a positive relationship between ex ante uncertainty and expected underpricing.