This week Hewlett Packard (HPQ) agreed to buy Palm for $1.4 billion, marking the end of the independent existence of one of the highest flyers of the dotcom era. Not exactly a dotcom, Palm (PALM) was one of the great wacko dotcom blockbusters. In its initial public offering just over 10 years ago, Palm opened for trading at $165 a share, 334 percent higher than the $38 price that IPO insiders paid for the shares.
Palm, it's worth noting"”oh boy, how worth noting"”in the current environment was a Goldman Sachs (GS) deal. Then, as now, Goldman was the most prestigious investment bank and underwrote many of the hottest IPOs. Investment banks took a justified beating for the dotcom deals; virtually every one of the companies with those skyrocketing opening prices turned out to be a disaster for investors.
To this day, however, it seems to me that the crazy gyrations of those dotcom stocks have never been adequately explained. Palm is a perfect case in point. Notably, Palm did not go down from the sky-high $165 price over a series of months. It went down dramatically the first day, closing at $95. Those who got the shares for $38 did very well. But if you bought the shares in the morning for $165, you got killed by the afternoon.
The conventional explanation for this is that investment banks priced their IPOs low in the hope of funnelling money to favored investors and creating publicity that would lead to future, bigger stock offerings (and more fees). I've always wondered if this explanation was sufficient. The gains to institutional investors who got in at the IPO price were vastly bigger than the fees investment banks got on these deals.
On Palm, for instance, if Goldman and other underwriters got the standard 6 percent fee, the total for the IPO would have come out to about $60 million. A lot of money, but only a small fraction of the $1.5 billion that the investors lucky enough to get shares at $38 would have gained at the day's closing price.
The operations of investment banks' trading desks are a black box: What the traders are doing at any moment is a closely guarded secret, and even folks in other parts of the banking business aren't privy to their strategies. What the banks' trading desks were doing in those frantic early hours of the dotcom IPOs is something that, to this day, no one has looked at closely.
Did the banks make substantial gains on their IPO market making? Really, we still don't know. Palm is an especially interesting case in point here. The investment bank that underwrites a public offering is responsible for making a market in the stock. The underwriters of an IPO are also the only ones allowed to short the stock in its first three days of trading.
Don't jump to conclusions too fast. These days, the word "short" alone seems dangerous. There are good, though complicated, reasons for an underwriter to short a IPO"”it lets the bank (legitimately) stabilize the price by buying it back later. The very first share sold at the opening of an IPO is a share sold short by the lead underwriter or "bookrunner."
Still, the frenetic trading in a stock like Palm has always raised questions for me. Trading desks stand to make"”and sometimes lose"”money from big price swings, and a fall from $165 to $95 in one days certainly counts as "big." Did investment banks manage to take back in trading a meaningful part of the money they essentially gave away to institutional investors in those dotcom IPOs? Yes, in stock market terms this is a question about ancient history. But with the investment banks, and especially Goldman, yet again called to explain what they do to the public, it's ancient history that could still shed light on the present.
Read Full Article »