Two scholars affiliated with Boston College, Carroll School of Management, have published a study of initial coin offerings (ICOs), that has given a mixed verdict on the common contention that cryptocurrencies in general and ICOs in particular constitute a “bubble” akin to, say, 17th century tulips or the mortgage derivatives of a little more than a decade ago. It is also fair to say that, though carefully worded, the conclusions of the paper are shaded to the favorable rather than the unfavorable side of “mixed.”
The paper, by Hugo Benedetti and Leonard Kostovetsky, agrees with the proposition that the evidence is consistent with a bubble explanation, but consistency is a low bar for such an important hypothesis. The evidence is also consistent with an alternative explanation, the possibility that early investors are being rewarded for risk for investing in unproven pre-revenue platforms through unregulated offerings.
Thoughts on ICO Underpricing
The facts that elicit these dueling interpretations? There has been significant ICO underpricing in the dataset created by Benedetti and Kostovetsky. Average returns from ICO price to the first day’s opening market price, over a holding period that is an average of just over two weeks (16 days), is a healthy 179%.
These scholars observe that some IPOs don’t list their tokens within 60 days. Imputing a loss of 100% to the pre-ICO investors in such a case, Benedetti and Kostovetsky adjust for the returns of their asset class. That takes the 16 day return down to a still healthy 82%.
What happens after trading begins? According to these authors’ dataset, tokens continue to appreciate in price, so that the average buy-and-hold returns for the first 30 trading days come to 48%.
The paper references two key differences between ICOs and initial public offerings (IPOs), aside from the obvious difference in the asset making its “initial” appearance. First, “ICO firms are much younger and smaller, typically in the earliest stages of a firm’s life cycle.” Second, and relatedly, the firms involved “do not use an underwriter to help determine value and attract buyers.”
Despite these differences, one of the Boston College findings about ICOs is consistent with the IPO literature. In both markets, the start-ups sell their tokens or equity at a significant discount to the opening market price. The degree of underpricing is larger for the ICOs than it is for the IPOs, but this “is not surprising considering the entrepreneur’s lack of expertise in determining market demand …greater uncertainty about the value of a startup company … and the urgency of distributing tokens to allow the platform to function.”
Furthermore, the dataset indicates that these ICO returns on the initial pop have declined over time, which suggests that firms have learned from prior offerings, that the terra incognito is not so incognito any longer. This in turn suggests that the market is not a proper tulip-like bubble but simply a very risky uncertain venture which is losing its alpha (to some extent) as the uncertainties lessen.
Thoughts on Post-ICO Returns
The second of the Benedetti and Kostovetsky empirical findings, that which concerns the price move after the ICO, is strikingly at odds with the IPO literature. The tokens continue to generate abnormal positive returns in the days and weeks that follow from the first pop. Newly listed equities, on the other hand, show underperformance in the days after the IPO, a generalization documented in the literature going back to 1991 (Ritter).
Why this discrepancy? This “could be an indication of bubbles,” or even of outright scams, but there are other possible explanations. In particular, most ICOs have no lock-up period. As a consequence, the opening price already reflects the supply from insider sellers. With IPOs, on the other hand, insiders cannot sell right at the bell, which means the release of that supply after the lockup constitutes a downward pressure on the price.
In conclusion, Benedetti and Kostovetsky caution against a facile reliance on the tulip analogy. Although regulators “should continue to deter fraudulent activities, they need to be careful not to throw out the baby with the bathwater.”