Jay Ritter, a professor of finance at the University of Florida who specializes in IPOs, says banks like Goldman knew full well that many of the public offerings they were touting would never make a dime. "In the early Eighties, the major underwriters insisted on three years of profitability. Then it was one year, then it was a quarter. By the time of the Internet bubble, they were not even requiring profitability in the foreseeable future."
Goldman has denied that it changed its underwriting standards during the Internet years, but its own statistics belie the claim. Just as it did with the investment trust in the 1920s, Goldman started slow and finished crazy in the Internet years. After it took a little-known company with weak financials called Yahoo! public in 1996, once the tech boom had already begun, Goldman quickly became the IPO king of the Internet era. Of the 24 companies it took public in 1997, a third were losing money at the time of the IPO. In 1999, at the height of the boom, it took 47 companies public, including stillborns like Webvan and eToys, investment offerings that were in many ways the modern equivalents of Blue Ridge and Shenandoah. The following year, it underwrote 18 companies in the first four months, 14 of which were money losers at the time. As a leading underwriter of Internet stocks during the boom, Goldman provided profits far more volatile than those of its competitors: In 1999, the average Goldman IPO leapt 281 percent above its offering price, compared to the Wall Street average of 181 percent.
How did Goldman achieve such extraordinary results? One answer is that they used a practice called "laddering," which is just a fancy way of saying they manipulated the share price of new offerings. Here's how it works: Say you're Goldman Sachs, and Bullshit.com comes to you and asks you to take their company public. You agree on the usual terms: You'll price the stock, determine how many shares should be released and take the Bullshit.com CEO on a "road show" to schmooze investors, all in exchange for a substantial fee (typically six to seven percent of the amount raised). You then promise your best clients the right to buy big chunks of the IPO at the low offering price — let's say Bullshit.com's starting share price is $15 — in exchange for a promise that they will buy more shares later on the open market. That seemingly simple demand gives you inside knowledge of the IPO's future, knowledge that wasn't disclosed to the day trader schmucks who only had the prospectus to go by: You know that certain of your clients who bought X amount of shares at $15 are also going to buy Y more shares at $20 or $25, virtually guaranteeing that the price is going to go to $25 and beyond. In this way, Goldman could artificially jack up the new company's price, which of course was to the bank's benefit — a six percent fee of a $500 million IPO is serious money.
Goldman was repeatedly sued by shareholders for engaging in laddering in a variety of Internet IPOs, including Webvan and NetZero. The deceptive practices also caught the attention of Nicholas Maier, the syndicate manager of Cramer & Co., the hedge fund run at the time by the now-famous chattering television asshole Jim Cramer, himself a Goldman alum. Maier told the SEC that while working for Cramer between 1996 and 1998, he was repeatedly forced to engage in laddering practices during IPO deals with Goldman.
"Goldman, from what I witnessed, they were the worst perpetrator," Maier said. "They totally fueled the bubble. And it's specifically that kind of behavior that has caused the market crash. They built these stocks upon an illegal foundation — manipulated up — and ultimately, it really was the small person who ended up buying in." In 2005, Goldman agreed to pay $40 million for its laddering violations — a puny penalty relative to the enormous profits it made. (Goldman, which has denied wrongdoing in all of the cases it has settled, refused to respond to questions for this story.)