Managers, investors, the financial press, and other capital market participants often use categories to describe firms or their securities. Common examples include “bellwether,” “blue chip,” “tech,” “penny,” and “start-up.” Although these categories may increase a firm’s visibility, they can lead market participants to overlook individual firm characteristics and thus bias their evaluation of a firm.

In a recent study, we examine how firm categorization influences investor behavior. In particular, we examine how a firm’s status as a unicorn influences investor demand and retail investment activity at its initial public offering (IPO). We focus our analysis on unicorn IPOs primarily because “unicorn” has a clear meaning, stemming from its objective, well defined, and commonly known attributes, i.e., a venture-backed firm with a valuation in excess of $1 billion (Lee, 2013). Although the $1 billion valuation threshold was arbitrarily selected, market participants indicate that this amount now represents a “psychological threshold [to attract] potential customers, employees, and the press” (Griffin and Primack, 2015).

Our empirical analysis examines 1,050 U.S. IPOs filed between 2005 and 2018. We find that 52 IPO firms meet the unicorn criteria subsequent to November 2, 2013, i.e., the date that Aileen Lee published the “Welcome to the Unicorn Club” article in Tech Crunch (Lee, 2013). Although a unicorn’s clear boundary conditions reduce concerns that capital markets will either be unaware of, or disagree about, a firm’s unicorn status, they also present an empirical challenge because of the difficulty in forming a control sample comprised of firms necessary to isolate the effects of unicorn categorization on investors. To overcome this challenge, we exploit the fact that the unicorn label did not exist until November 2013. Because our sample begins in 2005, there are firms in our sample that were not categorized as unicorns during their IPOs even though they met the qualifications for being unicorns (e.g., Facebook, Zynga, LinkedIn, HomeAway, and Groupon). Including these firms as control group observations, we then use entropy balancing to reweight observations in the control sample to achieve covariate balance with our sample of unicorn IPO firms.

We predict that the positive appeal associated with calling an IPO firm a unicorn increases investor demand, particularly among retail investors. We test our prediction by examining the relation between unicorn categorization and two commonly used proxies for investor demand: IPO offer price revision and first-day returns. We find that, relative to our sample of control firms, unicorn status increases the price revision by 14.9 percent  and first-day returns by 12.8 percent, suggesting that the unicorn categorization effects are not only statistically significant, but are also economically meaningful. We provide further evidence that unicorn classification influences investor demand by finding a positive relation between unicorn classification and the proposed IPO valuation, the final IPO offer price, and the closing price on the firm’s first day of trading on the secondary market.

We next consider whether increased retail trader activity in unicorn IPOs may help explain our findings, using intraday TAQ data to identify retail investor trades. We find a significantly positive relation between unicorn categorization and the portion of shares purchased by retail investors on the first day. We also examine retail trader activity by examining the relation between unicorns and the residual from regressing first-day returns on offer price revision. Because retail investors do not participate directly in the book building process, the residual thus reflects a more direct measure of retail investor demand relative to first-day returns alone. Consistent again with categorization increasing retail investor participation, we find a positive association between unicorn categorization and this residual measure.

To understand better the mechanism behind our findings, we perform a path analysis of the relation between unicorn categorization and our two measures of retail investor demand, with firm visibility as a mediating variable. We choose firm visibility based on prior research that shows news coverage strongly influences retail investor purchasing decisions. We find that unicorn categorization is informative for understanding retail trade activity both directly and indirectly through the mediating effects of news coverage. These findings are consistent with the increased investor demand for unicorn IPOs being at least partially attributable to retail investors “who missed out in the subscription but heard the hype” (Mackintosh 2019) and suggest that increased firm visibility is one mechanism through which categorization affects retail trade activity.

Our primary focus is to understand how categorization affects investor demand and retail trade activity. However, we also conduct tests to assess whether categorization effects are rational. If the initial IPO investor demand appropriately reflects how unicorn status influences future cash flows, then unicorn status should not correlate with post-IPO abnormal stock returns. On the other hand, we would expect a negative relation if unicorn categorization led investors to shape inappropriately their initial investment decision. We thus examine post-IPO stock performance, finding evidence of a negative relation between unicorn categorization and post-IPO cumulative abnormal returns over the initial six-month period these firms trade. Furthermore, we find that the sensitivity of post-IPO stock returns to pre-IPO profitability is significantly higher for unicorn IPOs than other IPOs. These findings suggest that investors not only associate too much value with unicorn status generally, but that they also overlook readily available value-relevant information about individual members’ accounting performance. This is consistent with prior research that indicates categorization hides category members’ individual idiosyncrasies (that are relevant to pricing).

Our study provides evidence that unicorn categorization shapes investor demand and retail trade activity during IPO price formation. Our findings reveal also how categorization can act as an antecedent to news coverage, thus highlighting a mechanism through which firms, or their hired investor relations advisers, can enhance firm visibility. We also provide initial evidence that, relative to a control sample of other IPOs, unicorn IPOs experience greater initial investor demand and lower post-IPO stock performance. Although Gornall & Strebulaev (2019) provide evidence that unicorns tend to be overvalued as private companies, their study does not examine the post-exit performance of unicorns. Our study thus complements and extends that study, providing evidence that unicorn valuations continue to be elevated even after the IPO process has completed. Considering the increasing number of unicorns filing IPOs, our study provides timely, cautionary evidence to retail investors considering investments in unicorn IPOs.

This post comes to us from Professor Badryah Alhusaini at Arizona State University’s W.P. Carey School of Business and professors Bradley E. Hendricks and Wayne R. Landsman at the University of North Carolina at Chapel Hill’s Kenan-Flagler Business School. It is based on their recent paper, “Categorization effects in capital markets: Evidence from unicorn IPOs,” available here.