Recently, a flurry of articles have discussed the pricing of the LinkedIn IPO, whether the investment banks screwed up and underpriced the IPO, or whether the IPO was priced correctly. These articles have piquéd my interest, and after some research into IPO pricing, I believe that I have come up with a reasonable framework for shorting some IPO first day pops.
First, let me present some research. Clearly, reasonable people can differ in their valuations of an IPO. Equally clearly, there is a simple way to resolve this to the benefit of the company doing the IPO. Just let the people with the highest valuations buy into the IPO. In other words, hold a Dutch auction. Why then do nearly all companies seeking to go public do so via book-building by an investment bank? The answer is that obtaining the highest possible price for the stock is not what the company’s owners necessarily want. If you sell the stock to the highest bidders in the market, then by definition, the stock has nowhere to go but down in the short term. This is undesirable to IPO investors, and surprisingly, also to the company’s owners. Why? Because IPO investors look dimly upon owners cashing out a large percentage of their stock at IPO, and most investment banks insist that owners sell only a portion of their ownership, with the remaining shares subject to a lock-up period. Accordingly, IPOs usually offer only 5-20% of the total equity, with more than 80% of equity still in the hands of insiders after the IPO. If the owners intend to sell more of their shares in the future after the lock-up, they want to create upward momentum for the stock price, and so they underprice the stock to ensure that result. Investment banks spin this behavior as the owners looking for “high quality buy-and-hold” shareholders, and are willing to lowball the stock price to attract such shareholders. This is a variation of the signaling hypothesis behind IPO underpricing. In essence, it means that the company is telling investors that it will soon be so profitable the money raised in the IPO is immaterial. The main point of the IPO is to “diversify the shareholder base”.
Another way of looking at this problem is that there are 3 parties involved in an IPO : the new investors, the company, and the previous owners. The motivations of the investors and the company are clear. The investors want a low price, and the company a high price. The motivations of the owners are conflicted. To the extent that the owners intend to hold and manage the company for the long term, they will want to obtain as high a price as possible to bring more cash into the company’s coffers, and perhaps sell a small portion of their own holdings to achieve a higher standard of living and modest diversification of their wealth. To the extent that the owners want to entirely cash out of their company in the short term, they will want to sell as few shares as possible (often over the objections of the underwriting bank, whose fees are a percentage of the total sum raised) to the lowest bidders, thereby ensuring frenzied bidding for the limited float, and saving the highest bidders for themselves in the future. The degree of the initial underpricing is an indication of the willingness of the owners to screw the company, which is indicative of their views on holding the company for the long term. In other words, the higher the first day IPO pop, the more likely the company is fundamentally flawed.
Now, although numerous academic studies suggest that investing in companies at IPO leads to underperformance in the long run (notably Loughran and Ritter 1995), shorting ALL IPO first day pops is likely to be a dangerous strategy. Firstly, there are structural factors above and beyond the signaling that can account for the first day pop, such as small retail investors not being able to access IPOs and thereby being forced to buy at a higher price in the open market. For example, even the Google IPO, which was performed by Dutch auction with 31 underwriting banks to reach as wide an investor base as possible, popped 17% on the first day. Secondly, SOME IPOs actually work out spectacularly well in the long term. If you are short, say, Microsoft, Dell, Google, or Walmart on the day of their IPO and held that position for 10 years, you would be a very poor man indeed. In fact, some insiders may not even realize the full potential of their own company. Paul Allen, for example, has said in his biography that he would have cashed out of Microsoft early on when he fell out with Gates, had Gates only offered him a more reasonable price. The stinginess of Gates ended up saving Allen a huge fortune in the long run. That said, if even insiders fail to see a company’s potential, the market is likely to only slowly come round to the bright prospects, and there should be ample time to close a short position at a modest loss. Most IPOs of successful companies take years for their success to become apparent.
In summary, my new framework for shorting IPOs is to short if there is a greater than 50% pop on the first trading day, and holding the position for 6-9 months until the expiration of the lockup period. If the theory is correct, the stock should deteriorate appreciably once the lockup period expires and insider selling floods the market. Since most lockups last for 180 days, it would take some time for the decline to be fully baked in. My framework is supported by academic studies suggesting that highly underpriced IPOs are lower in quality, and that extreme pops lead to long term underperformance. The LNKD IPO fits these criteria, having popped more than 100% on the first day. If you had shorted LNKD on the first day, you would be sitting on a 20% gain at this time. However, by the time I have finished my research and thoughts on shorting IPOs, the margin of safety for LNKD has declined substantially and the window of opportunity has probably passed. However, I am looking forward to the IPO of Groupon, a company with numerous competitors whose insiders are selling. Time will tell whether I end up shorting Groupon.
