We next analyze the fixed-price subscription following the auction for 77 hybrid IPOs in detail. The released information incorporates not only the public information that was generated before the auction period, but also the investor private information revealed in auctions. To investigate whether public information is the primary driver of investors’ demand for shares, we first abstract the element of public information from investor bids in auctions. As in Table 3, we use the market index return variable (Mkt_rtn) and the initial return variable (Ir_cipo) to reflect the market conditions, but we here calculate these two public information variables as of the auction’s beginning date. We also use the industry factor (Hi_tech) and the size of a firm (Ln_sale) to capture information on firm characteristics.
We regress the NQWP, the natural logarithm of oversubscription (Ln_os), and the natural logarithm of number of bids (Ln_nob) on the public information variables: the market index return prior to the auction period, the initial return of other contemporaneous IPOs, the high-tech dummy, and the size of a firm.
Table 9 presents the regression results, showing that investors indeed incorporate public information into their bids. Regression 1 indicates that both Ir_cipo and Hi_tech have a positive and significant impact on investors’ bidding prices of winning bids, suggesting that participants offer higher prices in hot issue markets and for high-tech firms; Ln_sale has a negative and significant impact on investors’ bidding prices, suggesting that investors submit lower bids for larger firms.
Place Table 9 about here
Regression 2 in Table 9 shows that the initial return variable has a very strong influence on the oversubscription of auctions. What is contrary to our expectation is that neither the market index return nor the high-tech dummy has a significant impact on investor oversubscription at auctions. This result is contrary to our earlier finding on investors’
oversubscription of pure fixed-price offerings that the market index return and the high-tech dummy have a positive and significant impact (Reg6, Table 4). In other words, the fixed-price offerings and auctions reflect a different relationship between oversubscription and public
information.
We posit that the insignificance of the market index return variable and of the high-tech dummy might be attributable to the participation of institutional investors and large individual investors in IPO auctions. Aggarwal, Prabhala, and Puri (2002) show that institutional investors have better information than retail investors, while Lee, Taylor, and Walter (1999) provide evidence that large investors have better information than small investors. Better-informed investors are in a better position at auction than uninformed investors, and they will definitely condition subscriptions on their private information. If informed investors are planning to flip their shares in the aftermarket, the public information on which they condition this action might be the initial returns of other contemporaneous IPOs, rather than market index returns and firm characteristics. The fact that institutional investors and large investors have an active role in Taiwanese IPO auctions presumably dilutes the influence of uninformed investors, resulting in an insignificant relationship between market index returns and oversubscription of auctions.
Another reason for the insignificant relationship between the market return and oversubscription of an auction is that in an auction the price is not set ahead of time, and the market return therefore may be less important. On the contrary, in a fixed-price offering the price is already set, and changes in the market return will therefore have influenced investors’
subscriptions.
To further investigate the influence of institutional investors and large investors on the oversubscription of auctions, we plot the histogram of allocation rates for 77 IPO auctions in Figure 4. The distribution surprisingly exhibits an almost reverse U-shaped distribution, in striking contrast to the U-shaped distribution of allocation rates we have observed in pure fixed-price offerings. We interpret this evidence as suggesting that herding is more likely to occur in fixed-price offerings, where investors are relatively homogeneous and uninformed, than in auctions, where investors are relatively diverse and some have better information than others.
Participants in our fixed-price offerings are exclusively individual investors, who are more subject to fads according to Lee, Shleifer, and Thaler (1991). In other words, investor characteristics are relevant to herding in IPO markets.
Place Figure 4 about here
Selling methods may be another important explanation for the presence of herding in
IPOs. Amihud et al. (2003) demonstrate that the distribution of allocation rates for 245 Israeli uniform-price auctions exhibits a U-shaped pattern.14 In discriminatory auctions, the winning bidders pay what they bid, and because the uninformed do not have any information advantage, they might not participate in the IPO market so as to avoid the “winner’s curse.” In either fixed-price offerings or uniform-fixed-price auctions, all winning bidders pay the same offer fixed-price, reducing the threat of the winner’s curse, and they become more aggressive in subscribing to IPO shares.
Our evidence is consistent with Chowdhry and Sherman (1996), who presents a theoretical model on the relationship between oversubscription and selling methods, and they conclude that extreme levels of oversubscription are more likely to occur in fixed-price offerings.
Another explanation of why investors avoid Taiwanese auctions is a longer delay between the auction and the IPO date than between the fixed-price offering and the IPO date.
Investors therefore expose themselves to more market risk in auctions than in fixed-price offerings.
Regression 3 in Table 9 shows that both Mkt_rtn and Ir_cipo have a positive and significant impact on the number of bids; this result is somewhat different from what is reported in Regression 2, where only Ir_cipo is significantly related to oversubscription. The underlying reason for the difference is that an oversubscription in auctions is equivalent to the quantity-weighted number of bids, where institutional bids and large bids are assigned a greater weight as they demand more shares. In the case of the number of bids, each individual bid is assigned an equal weight, and institutional investors and large investors are hence dominated by retail investors and small investors, who condition their subscriptions on the market index returns, resulting in a significant relation between number of bids and market index returns.
As in Table 5 we use the fitted values and the residuals from regressions in Table 9 to capture public information and private information, respectively. In Table 10 we present the results of an analysis that relates the oversubscription of follow-on fixed-price offers to NQWP, Ln_os, and Ln_nob.
Place Table 10 about here
14 Amihud et al.’s report includes 37 fixed-price offerings and 245 uniform-price auctions. Of their 282 IPOs, 142 IPOs have an allocation rate of lower than 5% and 73 have an allocation rate of over 95%. Therefore, excluding the 37 fixed-price offerings from their IPO sample will not drastically change the U-shaped distribution of allocation rates; in other words, the distribution of allocation rates for 245 uniform-price auctions should also exhibit a similar U-shaped pattern.
Regression 1 in Table 10 shows that coefficients for the variables of Ln_os (t-statistic = 3.02), NQWP (t-statistic = 2.70), and Ln_nob (t-statistic = 3.41) are all positive and very significantly different from zero. This regression has an adjusted R-squared of over 59%, indicating that information released from the auction indeed has a very strong influence on investor demand for shares of follow-on fixed-price offerings.
In order to verify whether public information has a stronger influence than does private information on investors’ demand for shares of follow-on fixed-price offerings, we regress the oversubscription on the fitted values of and on the residuals of NQWP, Ln_os, and Ln_nob, respectively. Regression 2 relates the oversubscription of follow-on fixed-price offerings to the fitted values; this regression has an adjusted R-squared of over 40%, but none of the coefficients have t-values exceeding 2.0, suggesting an apparent collinearity.
Because the fitted values of Ln_os and of Ln_nob are highly correlated, we exclude the fitted values of Ln_nob and rerun the regression. Regression 3 shows the results; this regression has an adjusted R-squared of over 41%, and the coefficient of the fitted Ln_os is significant, but the coefficient of NQWP is not. The results suggest that when later investors subscribe to shares of subsequent offerings, they condition their purchase decisions more on earlier investor actions than on the revealed value of IPO shares.
Regression 4 relates the oversubscription of follow-on fixed-price offerings to the residuals of NQWP, Ln_os, and Ln_nob; this regression has an adjusted R-squared of above 19%, and there also is an apparent collinearity. We hence rerun the regression by excluding the residuals of Ln_nob. Regression 5, similar to Regression 3, shows that the coefficient of residual oversubscription is significant, but the coefficient of NQWP is not. This regression has an adjusted R-squared of 17.26%, suggesting that public information predicts much more of the variation of investors’ demand for shares than does private information.
VII. Conclusion
Our examination of information cascades in IPOs indicates that for Taiwanese fixed-price offerings the distribution of allocation rates exhibits a U-shaped distribution as implied by Welch (1992). Further evidence indicates that while the private signal is the primary driver of a negative cascade, the public signal is not only the primary driver of a positive cascade, but also outweighs the private signal. These results are not quite consistent with Welch’s (1992) model, which posits that asymmetric information is the primary driver of an information cascade.
Instead, the results are consistent with Draho (2001), who argues that public information acts as a coordinating device, because investors use it to form beliefs about the beliefs of other investors. Investors then will condition subscription decisions on their beliefs of other investors’ demands, hence creating an information cascade. In addition, we propose two explanations, flipping and the prospect theory, of why investors condition their subscriptions on market conditions. Finally, changes in market conditions influence investor subscription decisions since the prices have already been set in fixed-price offerings.
When we investigate whether asymmetric information is the primary driver of IPO underpricing, we find that the cascade dummy has a significant effect on the underpricing, but the effect of public information is even stronger. In short, we can rely on the evidence for assurance that asymmetric information is not the primary driver of IPO underpricing, even though not proof that public information is the primary driver of underpricing.
We also examine Benveniste and Busaba’s (1997) hypothesis of eliminating negative cascades through information spillovers. The distribution of allocation rates on follow-on fixed-price offerings suggests that information spills over from auctions to follow-on fixed-fixed-price offers. The evidence that most firms in our sequential hybrids have achieved a positive information cascade in follow-on fixed-price offers is consistent with Benveniste and Busaba (1997), who argue that information spillovers enable issuers to avoid the threat of a negative information cascade.
We also find that public information that is incorporated into earlier investor bids has a stronger influence on later investor demand for shares of follow-on fixed-price offers than does the private information that is incorporated into investors’ bids. Finally, we find that herding is more likely to occur in fixed-price offerings than in auctions. We interpret the evidence as suggesting that investor characteristics and IPO selling methods are related to herding in IPO markets.
References
Aggarwal, R. “Allocation of Initial Public Offerings and Flipping Activity,” Journal of Financial Economics 68 (2003), 111-135.
Aggarwal, R., N. R. Prabhala, and M. Puri. “Institutional Allocation in Initial Public Offerings:
Empirical Evidence,” Journal of Finance 57 (2002), 1421-1442.
Amihud, Y., S. Hauser, and A. Kirsh. “Allocations, Adverse Selection and Cascades in IPOs:
Evidence from Israel,” Journal of Financial Economics 68 (2003), 137-158.
Benveniste, L. M., and W. Y. Busaba. “Bookbuilding vs. Fixed Price: An Analysis of Competing Strategies for Marketing IPOs,” Journal of Financial and Quantitative Analysis 32 (1997), 383-403.
Benveniste, L. M., W. Y. Busaba, and W. Wilhelm. “Information Externalities and the Role of Underwriters in Primary Equity Markets,” Journal of Financial Intermediation 11 (2002), 61-80.
Benveniste, L. M., and W. J. Wilhelm. “A Comparative Analysis of IPO Proceeds under Alternative Regulatory Environments,” Journal of Financial Economics 28 (1990), 173-207.
Biais, B., P. Bossaerts, and J. C. Rochet. “An Optimal IPO Mechanism,” Review of Economic Studies 69 (2002), 117-146.
Biais, B., and A. M. Faugeron-Crouzet. “IPO Auctions: English, Dutch, … French, and Internet,” Journal of Financial Intermediation 11 (2002), 9-36.
Bradley, D. J., and B. D. Jordan. “Partial Adjustment to Public Information and IPO Underpricing,” Journal of Financial and Quantitative Analysis 37 (2002), 595-616.
Chemmanur, T., and H. M. Liu. “How Should a Firm Go Public? A Dynamic Model of the Choice Between Fixed-Price Offerings and Auctions in IPOs and Privatizations,” Working Paper, Boston College. (2002) .
Chowdhry, B., and A. Sherman. “International Differences in Over-subscription and Underpricing of IPOs,” Journal of Corporate Finance 2 (1996), 359-381.
Cornelli, F., and D. Goldreich. “Bookbuilding: How Informative is the Order Book?” Journal of Finance 58 (2003), 1415-1444.
Daniel, K. “Discussion of ‘Why Don’t Issuers Get Upset About Leaving Money on the Table in IPOs?’ ” Review of Financial Studies 15 (2002), 445-454.
Daniel, K., D. Hirshleifer, and A. Subrahmanyam. “Investor Psychology and Security Market Under- and Over-reactions,” Journal of Finance 53 (1998), 1839-1886.
Derrien, F., and K. L. Womack. “Auction vs. Bookbuilding and the Control of Underpricing in Hot IPO Markets,” Review of Financial Studies 16 (2003), 31-61.
Draho, J. “The Coordinating Role of Public Information in Hot Market IPOs,” Working Paper, Yale University. (2001)
Draho, J. “The Timing of Initial Public Offerings: A Real Option Approach,” Working Paper, Yale University. (2000)
Kahneman, D., and A. Tversky. “Prospect Theory: An Analysis of Decision Under Risk,”
Econometrica 47 (1979), 263-291.
Lee, C., A. Shleifer, and R. Thaler. “Investor Sentiment and the Closed-end Fund Puzzle,”
Journal of Finance 46 (1991), 29-48.
Lee, P. J., S. L. Taylor, and T. S. Walter. “IPO Underpricing Explanations: Implications from Investor Application and Allocation Schedules,” Journal of Financial and Quantitative Analysis 34 (1999), 425-444.
Lerner, J. “Venture Capitalists and the Decision to Go Public,” Journal of Financial Economics 35 (1994), 293-316.
Ljungqvist, A., V. Nanda, and R. Singh. “Hot Markets, Investor Sentiment, and IPO Pricing,”
Working Paper, New York University. (2001).
Loughran, T., and J. Ritter. “Why Don’t Issuers Get Upset about Leaving Money on the Table in IPOs?” Review of Financial Studies 15 (2002), 413-443.
Lowry, M., and G. W. Schwert. “Is the IPO Pricing Process Efficient?” Journal of Financial Economics 71 (2004), 3-26.
Ritter, J., and I. Welch. “A Review of IPO Activity, Pricing, and Allocations,” Journal of Finance 57 (2002), 1795-1828.
Schultz, P. “Pseudo Market Timing and the Long-Run Underperformance of IPOs”, Journal of Finance 58 (2003), 483-518.
Sherman, A. “Global Trends in IPO Methods: Book Building versus Auctions,” Working Paper, University of Notre Dame. (2003)
Subrahmanyam, A., and S. Titman. “The Going Public Decision and the Development of Financial Markets,” Journal of Finance 54 (1999), 1045-1082.
Thaler, R., and E. Johnson. “Gambling with House Money and Trying to Break Even: The Effect of Prior Outcomes on Risky Choice,” Management Science 36 (1990), 643-660.
Welch, I. “Sequential Sales, Learning, and Cascades,” Journal of Finance 47 (1992), 695-732.
White, H. “A Heteroskedaticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedaticity,” Econometrica 48 (1980), 817-838.