On May 18, 2012 Facebook (FB) held its initial public offering (IPO) on NASDAQ, raising over $16 billion making it one of the largest IPOs in history. To the surprise of many investors, there was almost no underpricing, as the stock closed the first day of trading almost flat from its offer price. The IPO was described as not only disappointing but also detrimental to the broader market in the financial press. The “failed” FB IPO was blamed for causing everything from mutual funds’ declines in assets under management to significant increases in IPO underpricing, to subsequent canceled IPOs. We investigate the consequences of the busted Facebook IPO and dispel the notion that the pricing of all IPOs changed as a consequence.
Morgan Stanley (MS) was the “lead-left” underwriter with J.P. Morgan (JPM) in the second seat followed by Goldman Sachs (GS) and 30 other underwriters. The lead underwriters set the initial price range for the offering at $28-$35 per share offering 337 million shares. During the roadshow, nine days before the IPO, it was selectively verbally disclosed to some institutional investors that FB may not meet its projected revenue and earnings estimates. In spite of the selective disclosures, at the completion of the book-building process, the offering was still significantly oversubscribed. Institutional demand for FB shares exceeded five times the shares available. As a result, the lead underwriters decided to increase the offering range to $34-$38 per share with 421 million shares offered.
Facebook opened for trading at 11:30 a.m. slightly above the offer price of $38 per share. During the first sixty seconds of trading, over 158 million shares changed hands and the price ranged from a low of $42 to a high of $45. The high price for the day of $45 happened within the first minute of trading. The stock subsequently fell back to the offering price of $38 per share within 19 minutes of the open. Over the course of the day, Morgan Stanley was actively buying shares to provide price support and keep FB from falling below its offering price. It closed the first day of trading at $38.23. In the days and weeks following the IPO, without the price support of Morgan Stanley, FB continued to drop. By the beginning of September, FB had fallen over 50% and closed for trading at $17.73.
The average level of underpricing on IPOs in the United States has varied significantly from year to year, averaging 18% for the period 1980-2014. IPO failures are rare; few IPOs close at a price below the first day offer price because investment banks are allowed to provide price support following the IPO. Underpricing, or “money left on the table”, is an opportunity cost to the firm going public and is a wealth transfer from the issuing firm to investors who receive allocations of underpriced shares. The majority of IPO shares are sold to institutional investors who predominantly sell their holdings within the first year and fully realize the “money left on the table”. Thus the institutional clients of the investment banks are the primary beneficiaries of IPO underpricing.
Gondat-Larralde and James (2008) show that it is optimal for banks to enter into a repeated game with a coalition of investors to maximize the probability that its IPOs succeed. They show that banks set IPO prices and in effect, underpricing, in a manner that equalizes downside risk across all of its offerings. For investors to want to remain in the coalition the expected benefit across all offerings must be positive, and thus on average investors should be rewarded with underpricing. Thus there exists a repeated game between investment banks and institutional investors in the allocation of IPO shares. Investment banks in the IPO setting have a dual role; they work with issuing firms to raise capital at as high a price as possible, but must simultaneously keep their institutional investor clients happy by providing positive returns across the portfolio of IPOs they bring to market.
We examine the IPO market surrounding the Facebook offering and pay special attention to the response by the Facebook lead underwriters, Morgan Stanley, J.P. Morgan and Goldman Sachs to such a visible and costly failed IPO. The enormity of the Facebook offering relative to the entire portfolio of IPO activity makes it ideal for examination. The proceeds raised in Facebook ($16 billion) are equal to the aggregate proceeds raised in the 89 IPOs preceding Facebook that were taken public by the FB lead underwriters.
We find that prior to the Facebook offering, the IPO market was very active with deals pricing on average one every five days. Following Facebook, it was 41 days before another company launched an IPO.
When the IPO market resumed, IPOs were priced with significantly deeper discounts even after controlling for market trends and IPO characteristics. Average underpricing jumped from 12% pre Facebook to over 20% following the Facebook IPO. Deeper analysis shows this increase was solely concentrated in IPOs led by J.P. Morgan, Morgan Stanley and Goldman Sachs; the three lead FB underwriters. These banks increased average underpricing from 15% prior to Facebook, to 27% following Facebook. We find no difference in underpricing for all other investment banks surrounding the FB deal. We contend that investment bank loyalty to their institutional client base propelled the FB underwriters to underprice future IPOs to make up for the perceived losses incurred in the failed FB deal. Due to the enormity of the Facebook IPO, it required corrective pricing (significant underpricing) on many future IPOs to compensate for the lack of underpricing on Facebook.
Based on Facebook’s first day closing price, underpricing was $0.23 per share for aggregate “money on the table” of $97 million. The average level of underpricing by the FB lead banks prior to the Facebook offering was 15%. Based on this average level of underpricing, coalition investors would have expected $2.4 billion “money on the table” in the Facebook IPO. Thus the unexpected aggregate wealth loss (unexpected money on the table) was over -$2.3 billion. We find that it takes these 3 banks 85 IPOs over the course of the following 18 months to provide the expected level of “money on the table” to investors to make up for the perceived loss on Facebook.
Existing literature has not directly addressed what happens when underwriters fail to “deliver” the expected underpricing that market participants have come to expect. Our paper offers one example of a well-publicized “failure” and the resulting additional degree of underpricing that exists in future deals as evidence of the dynamic relationships that exist in the IPO market. Our paper is the first to find evidence that an additional factor may impact the extent of underpricing in a given deal. Specifically, following an “important” IPO event that failed to deliver first-day gains expected by institutional investors, the underwriters responsible for the “failed” IPO are associated with higher degrees of underpricing on their post-event subsequent IPOs.
Our findings highlight the inter-temporal relationship between investment banks and investors in making markets for newly public firms’ equity. Because these firms’ stock has not previously priced and traded, valuation uncertainty is high. One mechanism for compensating investors for helping to make a market is underpricing. Previous research has shown investors appear to demand 15% underpricing over time. The results in our paper help elucidate how underwriters deliver this average level of underpricing, especially when one large deal fails to deliver on the expectation.
Gondat-Larralde, C., and K. James, 2008, “IPO Pricing and Share Allocation: The Importance of Being Ignorant,” The Journal of Finance, LXII, No. 1, 449-478.
The preceding post comes to us from Laurie Krigman, Associate Professor in the Finance Division of Babson College, and Wendy Jeffus, Lecturer in the Finance Division of Babson College. The post is based on their article, which is entitled “IPO Pricing as a Function of Your Investment Banks’ Past Mistakes: The Case of Facebook” and available here.