If you follow the financial news, you already know that Uber seems to have a better model for its ride-hailing service than its stock. After priced at an offer price of $45 the stock opened at $42, a rare rebuke for a well-known and so eagerly anticipated IPO (initial public offering). Four days later, it still trades under water. It looks like Professor Mihir Desai of Harvard got his wish when he titled his NYT OP-ED piece “Let’s Root for Uber’s IPO to Fail.”
Interestingly, Professor Desai expressed a wish for an outcome that goes against the average performance of IPOs, which usually gain in price immediately after they are offered to the public, a performance identified as underpricing. Setting a price for an IPO stock before the market has opened is a great challenge to the underwriting banks. That’s why they organize presentations of the IPO firm to investors around major centers of finance in the so-called roadshows. It is through this process that underwriters get a feel of the market’s receptivity of the new stock. In Uber’s case the investors had reportedly placed (subscribed for) buy orders three times the number of shares offered, a sign of strong demand. Despite pre-offering estimates about Uber’s stock value exceeding $100 billion and the oversubscription it garnered, the lead underwriter Morgan Stanley still decided to price the stock conservatively giving it a total market value of $76 billion. To everybody’s surprise, however, even that lower offer price (of $45) wasn’t conservative enough.
The market for IPOs, which along with new issues of already traded stocks comprises the new issues market, is one of the most challenging types of markets to analyze and understand. The hundreds of academic and practitioner papers written on the pricing of new issues can only tell us what we would expect to see on average. Much less can be predicted on a stock by stock basis. Thus, we know that on average IPO stocks are underpriced, that is, they are offered below the price the market sets when trading starts. And this pattern is a global one observed around the world. The main reason is that new stocks, no matter how known to the public, are surrounded by uncertainty that is perceived differently by the selling firm and the prospective investors. The firm cannot know with certainty where demand and supply for the stock will balance to produce a price. Investors, on their part, cannot know what the future prospects of the firm are because they can glean some but not all the information about the firm from the prospectus (i.e., the document that explains the operations and finances of the firm). When sellers and buyers have different information about a product, economists call this asymmetric information. (George Akerlof won the Nobel prize for analyzing such markets – an example of which would be the used cars market.) Therefore, to entice investors, especially those less savvy about investments, to buy the new stock, underwriters set the price below what they expect the market will be willing to pay at the offering. It’s like listing your home at a price below its estimated fair value.
But even with the roadshows and all their years of experience underwriters sometimes fail spectacularly. Facebook, for example, was offered at $38 in 2012. It closed the first day barely up at $38.23. Then it fell and stayed below its offer price for the next 420 days. Actually, the IPOs of Uber and Facebook defied the empirical evidence that stocks of firms with greater buzz about them are welcome by the market at much higher prices than their offer price. Exhibit A of this phenomenon are the IPOs of internet technology-based firms whose IPOs in the dot.com mania years of 1999 and 2000 yielded initial market prices in excess of 77% of their offer price. Unfortunately, publicity and anticipation did not work at all in that way for Uber.
The second thing we know about IPO stocks is that, on average, are poor investments for the first several years following their offer. This means, if you buy IPO stocks at their market price the day they are first offered and hold them for three to five years you will earn a lower rate of return than if you buy a portfolio that is broadly diversified, like the 500 stocks of the Standard & Poor’s Index. Again, this is a global phenomenon. However, what holds for the average IPO stock does not hold for every individual IPO stock. For example, after trading in losing territory for over 420 days, Facebook’s stock climbed over its offer price and never looked back. It is now traded at around $185.
The longer-term under-performance of IPO stocks is a combination of two factors. One is of behavioral nature. In forecasting the future, we have a cognitive bias toward putting too much weight on recent events. Investors look at the recent usually good news and performance of the IPO firm and extrapolate this good performance into the future ignoring the fact that eventually everything reverts to a more subdued normal level. This law of reversion to the mean echoes the everyday saying that what goes up eventually comes down. The other reason is that firm managers decide to have an IPO when news are still positive before negative developments sour market receptivity. This is to say, managers time the IPOs. When, subsequently, the negative news comes out investors sell the stock and the price falls. But again, not every IPO stock falls victim of the bias or the timing. Microsoft, Google, Apple, Facebook are examples of stocks that proved to be big winners for investors over the long haul.
The failure of the Uber IPO is noteworthy because the reputation of the lead manager (Morgan Stanley) and the co-managers (Goldman Sachs and Bank of America) as well as the financial backing of Uber by big venture funds should have added credibility of the offer price. But, as in elections, you need to count everybody’s choice (in this case the investors’) before you know what the outcome is. Perhaps the market is sensing that the gig economy in which Uber wants to operate is not settled enough to support reliable predictions. However, Uber has just scored a big victory in this regard. On Tuesday (5/14/19), the National Labor Relations Board determined that the Uber drivers are not employees and, hence, not eligible to unionize. As a result, the price has jumped by about $2. Still, however, the jury is out on how well Uber will do as the poster stock of the gig economy.
As for Professor Desai’s wish, did it come through? He wished for a broken Uber IPO, that is, one with a market price falling short of the offer price, as a rebuke to large venture capital funds, like the Softbank Vision Fund, which tend to provide plenty of financing to glamorous firms like Uber at the expense of other noteworthy startups. He apparently assumes that such funding bias is not driven by sound fundamentals and hence hurts the efficiency of the market in allocating capital to businesses. On one level, Prof. Desai’s wish came through. Months before the IPO, Uber’s shares were estimated to be worth between $100 and $120 billion. At current prices, they are worth closer to $70 billion. This means Uber’s financial backers were terribly off the mark regarding the value of this business when they sank money into Uber. Therefore, Prof. Desai got his wish. Will this chasten venture capitalists and make them more efficient in valuing startups? It remains to be seen. On another level, Uber’s broken IPO worked in favor of its backers and other shareholders. They sold new shares at $45 when shortly afterwards the market determined they were worth only $42. This means they sold overpriced shares! This was a gain to them. As more stories come out, we are learning that some dark clouds had already moved into Uber’s blue sky before the IPO. It is very plausible that influential Uber shareholders (like its venture capital backers) pushed a higher offer price on the underwriters, thus, contributing to the eventual overpricing.
As I said, when it comes to a specific IPO, it’s difficult to make predictions. As in other areas of economics and social sciences, we can recognize patterns that hold on average but can never accurately foretell how a particular case will break out. Dismal sciences indeed.
Let's Reason 