Non-proportional thinking in IPOs

ABSTRACT

The fact that initial-to-final IPO price revision ratios are strong predictors of IPO underpricing is well known to researchers. Our study documents that the dollar amount of the price revision matters above and beyond the revision ratio, and dramatically so. Immediately following the IPO, market participants appear to see a positive signal in greater dollar amount revisions, even if holding price revision ratios equal. This extends Shue and Townsend (Journal of Finance, 2021) finding of market participants’ ‘non-proportional thinking’ to the IPO setting. It also implies that the dollar amount of price revision deserves attention from future IPO researchers, particularly when studying the determinants of IPO underpricing.

KEYWORDS:

• IPO
• underpricing
• investment banking
• behavioral finance

THE JEL CODES:

• G24
• G41

Acknowledgments

We are grateful to two anonymous referees, Pierre Mella-Barral, Timo Mandler, Arthur Petit-Romec, Miguel Urdanoz, Ivo Welch, and participants at the TBS brown bag seminar for their comments. Special thanks are due to a New York-based underwriter for his generosity with his time and insights as well as to Sammy Obeid who prompted discussions that led to this research project. All errors are ours.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ From our interview with an experienced New York-based underwriter: non-integer amounts are seen by investors as “lacking conviction”.

² From the same interview: “Bigger IPOs tend to have higher share prices … if we are doing a 100-million-dollar IPO and you set the price at 10, and 10 million shares, it’s just … we’d rather see that selling 5 million shares at a higher price. This is not a rule, it’s just kind of convention in what the market generally accepts.”.

³ From the same interview: “We’re often doing the reverse split in the week or two leading up to the IPO. Generally, you’re not finalizing valuation until right before launch, because market conditions can change. They might do the reverse split, but then we might still not be sure if it’s $14-$16, $13-$15, right? We’re making sure we have a round IPO price, and they’re going to have an irregular reverse split.”.

⁴ Starting in 2012, 20% (WilmerHale 2012).

⁵ In unreported results, we reproduce the regressions reported in Table 3 with the inclusion of a variable accounting for underwriter reputation that assigns to each IPO the Carter-Manaster rank (Carter et al. 1998) corresponding to its highest-ranked underwriter. This variable has a significant positive effect on IPO returns, but it does not undermine the economic and statistical significance of our results for dollar-revision dummy variables.

⁶ We thank an anonymous referee for suggesting that we consider firm EGC status. To construct the EGC dummy, we used pre-IPO company S-1 filings with the SEC.

⁷ A similar impact is observed by alternatively comparing the predicted value of the one-day IPO return for a $12 midpoint, $2 revision IPO versus a $18 midrange, $3 revision one, where the remaining variables are all set to their sample means. The predicted return is equal to 47% in the former case and 78% in the latter, resulting in 31% difference.