Cure for excessive Wall Street compensation: Price wars

By Steven Pearlstein
Friday, August 20, 2010

Normally, the big investment banks would be salivating over the prospect of a $15 billion stock offering by General Motors, which by rights would generate $450 million in fees. But a funny thing happened on the way to this bonanza. Goldman Sachs -- convinced it was never going to be named lead underwriter because of its legal problems and its close ties to Ford -- decided to have some fun at its rivals' expense. Tossing aside Wall Street's most sacred commandment -- Thou Shall Not Undercut the Fee Structure -- Goldman offered to do the job for 0.75 percent of the stock sale, a quarter of the normal fee.

Sure enough, Goldman rivals Morgan Stanley and J.P. Morgan were offered the business. But the U.S. Treasury, which owns most of the stock that will be sold through the offering, insisted that the winners do it at the Goldman price. Given that the deal would almost certainly give them bragging rights as the top underwriters for the year, and given the public outcry that would have ensued if they had balked at giving a discount to the very same taxpayers who had just bailed them out, the Morgan twins decided this was an offer they could not refuse.

This story, first reported by Bloomberg News and confirmed by several government and Wall Street sources, goes a long way in explaining why so many people on Wall Street get paid so much more than everyone else. A handful of established firms control access to global financial markets and use this power to extract monopoly-like profits and funnel them to their executives and employees.

The reason for the lack of price competition is pretty simple: The banks know that if they start offering big discounts, all their rivals will be forced to do the same. In the end none of them would gain a competitive advantage, but all of them would wind up with less money. The only winner from a price war would be their customers.

This kind of cozy competition only works in markets where it is virtually impossible for upstart firms to gain a foothold by offering a lower price.

Size is part of it: An investment bank has to be big enough and have strong enough relationships with thousands of institutional investors and money managers to be able to market large stock and bond offerings in a matter of days.

Given the huge sums of money that are raised through these offerings, it may seem silly for a company to try to save a few million dollars in fees by eschewing the established players, who invariably claim they are so much wiser and more experienced that they will be able to get an extra 50 cents for each share of stock or shave the interest rate on a bond by an eighth of a percentage point. There's no way to prove they are right, but there's no way to disprove it either -- and surely nobody's ever been fired for going with Goldman Sachs.

So what we have is a suspicious consistency in IPO fees: 6 to 7 percent for deals of less than $200 million, 4 to 5 percent for moderate priced deals and 3 to 4 percent for those over $1 billion. Most of the fees go to the lead underwriters, with the rest scattered among the losers of each beauty contest. The fee pot is divided in three: Twenty percent is an underwriting fee for the banks' guarantee that they will buy the entire issue even if nobody else will. That was once an important consideration, but in today's markets, the investment banks ensure that no IPO goes forward unless the entire offering is pre-sold. As a result, the underwriting fee is now a pure windfall.

Sixty percent of the fee represents a sales commission, which also turns into something of a game, particularly when it involves highly desired shares of well-known companies such as GM. Since all the major investment banks have relationships with all the major institutional buyers, it's hard to say which sales force actually makes the sale. So once Fidelity or TIAA decide how many shares they want to buy, they can divide the commission among the investment banks any way they choose. In the end, the division has more to do with relationships and favors and thinly disguised kickbacks than how much work is done or how much skill is involved.

The final 20 percent is the management fee, which goes exclusively to lead underwriters. This is for helping to prepare the prospectus, organizing the 10-day road show to market the issue to prospective buyers, keeping the order book and advising the company on the offering's price and size. For a big deal such as the GM offering, it might involve incredibly intense work by as many as 30 professionals for as long as four months -- let's say, generously, 30,000 hours of work. On a $15 billion IPO, that works out to $3 million per banker, or $3,000 per hour worked.

In theory, it should be possible to raise money without paying such a hefty toll to Wall Street's gatekeepers. In some countries, new stock issues are sold through auctions that eliminate the need for large sales forces and their commissions. A few high-tech companies, including Google, successfully used such auctions here, but Wall Street has somehow managed to convince corporate directors and executives that the renegades didn't get full price for their offerings.

Moreover, now that banking and investment banking are done under the same roof, companies might worry that their reward for squeezing investment banking fees will simply be higher costs for their loans. It is probably no coincidence that the same firms that are underwriting GM's stock issue have also agreed to provide the carmaker with a $5 billion revolving credit facility, one that reportedly involves relatively high fees.

There's nothing about excessive compensation on Wall Street that can't be cured by some old-fashioned price competition -- reduce the fees and the trading profits and the bonuses will follow. All it would require is a determination by major companies to drive as hard a bargain on behalf of their shareholders as the Treasury has now done on behalf of taxpayers.

© 2010 The Washington Post Company