'Dark Pools' Threaten Wall Street

One of the fastest moving trends on Wall Street has flown under the radar of individual investors and, seemingly, the Securities and Exchange Commission: the rapid rise of "dark pools" stock trading arenas.

As the name suggests, dark pools lack transparency: They are used by institutional investors seeking to trade large blocks of stocks without creating the price wobbles that routinely accompany such moves. The trading is done away from the traditional exchanges, offering unprecedented anonymity.

Recently, more than 20% of all trades in New York Stock Exchange-listed stocks have been funneled through these dark pools, up from just 3% to 5% two years ago, according to NYSE figures.

Asked about the SEC's view of dark pools, a spokesman cited a recent speech by the head of the Division of Market Regulation, Erik Sirri, who said that "while the increasing use of hidden orders may be troubling," the SEC believes the new venues are available to all market participants. He suggested that the SEC is not in the position of favoring one market model over another. It would appear that until the trend toward dark pools has a measurable impact on investors, the SEC is willing to be simply an observer.

In the wake of the subprime mortgage debacle of the past two months, the importance of transparency is obvious. The packaging and repackaging of impossibly shaky mortgages into sophisticated investment products resulted in securities that were overly complicated and not well understood, and that ultimately went bust and brought down an entire industry.

An increasing number of these dark pools are popping up — more than 35, it is estimated — that seek to match large buy and sell orders without recourse to the traditional trading discourse on stock exchange floors. This is not to be confused with electronic trading, which is not new. The electronic communication networks such as Archipelago and Instinet that started up in the 1990s changed trading forever by essentially replacing the matching efforts of the specialist with computers. The ECNs report trading data in a traditional manner — publicly.

Dark pools, by contrast, do not alert the market makers that a sizeable order has been placed. Instead, by breaking the orders into smaller pieces, computer programs match and execute the different segments of the order in hundreds of separate transactions. This slicing and dicing is very similar to the complex structures banks used to divide up subprime mortgage debt.

All the major investment banks have dark pools operations, such as Block Alert, owned by Merrill Lynch, and ACE, which is part of Citigroup. There are also independent firms such as Liquidnet, which now ranks as one of the top 10 brokers. None comes close to Sigma X, which is owned by Goldman Sachs, some say.

"Goldman is a leader in the area. They're way ahead of everyone else," a financial services analyst with Punk, Ziegel & Company, Richard Bove, said. "They have spent hundreds of millions of dollars on systems. They have more people in IT than they have traders or investment bankers. They want to drive the price of trading to a level that will drive most firms out of business."

The attraction of these dark pools is clear. If an insurance company, for instance, wants to unload 1 million shares of Deere & Co., it traditionally would have to phone its broker, who would then have to contact the floor of the New York Stock Exchange, and in the process information about the institution's intent to sell would begin to leak out to other traders. Presumably, the shares would come under pressure, and the seller would end up receiving less money for his stock.

The problem is that in using the dark pools, there is little way for sellers to assess whether they have received the best possible execution on the order. Although by law the participants have to print the trade on one of the exchanges, the information is after the fact, and not especially revealing.

This is especially the case if the broker has "internalized" the order. This popular activity allows brokers to match the order within their own shops, operating beyond the vision of other dealers and outside the spotlight of the SEC. Though internalization has always taken place, the development of crossing networks has greatly expanded the in-house opportunities.

This has an impact on the average investor by shrinking the amount of trading that is being funneled through traditional channels, and which is available for filling the orders that such small investors place through their brokers. That is, it reduces liquidity and transparency in the marketplace for the small investor, while enhancing it for the big players. Less liquidity means less opportunity for "price improvement," wider spreads, and, ultimately, more costly executions.

Also, over time, internalization will mean a continued consolidation in the brokerage industry. Those firms such as Goldman Sachs that have giant order flows already are obviously best positioned to fill orders in-house. They profit off a spread between the price they paid for a stock and the price they charge the buyer. While initially they are likely to seek narrow spreads so as to attract business, if there is a fall-off in the number of competitors, the spreads will likely widen.

Not everyone is against the rise in dark pools activity. Dark pools "are providing a critical service with new technology," a former chief of the SEC, Harvey Pitt, who now heads Kalorama Partners, said.

Others argue that the private trading could augment opportunities for self-dealing. In an op-ed piece written several years ago, Mr. Pitt acknowledged that trading firms often put their own interests ahead of their clients'. Especially troubling, and the centerpiece of an SEC investigation into Knight Securities, was the practice of using client orders to enhance the profitability of the firms' proprietary trading desks. It is hard to imagine that increasing nonpublic trading improves matters.

Finally, as more investors rely on complicated algorithms to perform the market making functions that were once the province of the specialist, the transparency of the marketplace will be forever diminished. This will not be a problem until it is a problem — just as the inadequate pricing of risk in the CDO market was not a problem until it became a problem.