This statement is also available here as an Adobe PDF.

The Securities and Exchange Commission (SEC) has recently published its “Concept Release on Equity Market Structure” (January 14, 2010) and has also proposed new rules related to equity trading. The SEC’s concept release emphasizes that market structure should favor long-term investors over short-term investors. Consistent with prior policy, the SEC is particularly focused on protecting small investors who invest directly in stocks and less concerned with minimizing the transaction costs of mutual funds in which small investors invest indirectly in stocks.

But focusing on long-term investors and trying to protect the direct retail trader reflects a naïve view of the interests of investors and traders. Some investors, particularly small retail investors, can trade with virtually no market impact. Other investors, primarily institutional investors, are rightly concerned about the impact their orders might have on market prices. How investors react to the rules of the market that the SEC sets will depend in part upon the information upon which they are trading, the size of their orders, and the potential market impact, and investors respond differently to SEC rules depending upon their diverse needs.

The SEC has proposed that actionable indications of interest (IOI) transmitted to dark pools should be displayed to all within the National Market System (NMS). Actionable IOIs are orders to buy or sell that the order originator desires to communicate to participants in dark pools--electronic-matching facilities provided by a broker-dealer to a limited group of investors. Institutional investors use such orders to limit the potential for front running. In front running, another trader uses knowledge of an impending large order to trade ahead of that large order, resulting in poorer execution. A trader with a large order may want to display his order to a limited set of investors to reduce front running to obtain a better execution. If the SEC were to force the display of actionable IOIs, investors with large orders may choose other ways to execute them without display, negating the intent of the SEC with possible deleterious effects on the market.

As the SEC Concept Release recognizes, institutional investors quite sensibly are often adverse to displaying their intentions to the marketplace. To avoid displaying the full amount of their order, they have in the past used the upstairs market, where a broker-dealer can discretely market their orders to reduce price impact. In the aftermath of Regulation NMS (2005), these investors have altered their trading strategies and now frequently break their order into small “child” orders. Partly as a consequence, the average execution size on the New York Stock Exchange is now under 300 shares. Other times some institutional investors have channeled orders of larger size to dark pools. One dark pool reports that the average size of such trades is approximately 50,000 shares.

An institution’s optimal order execution strategy depends upon the information underlying the motive for the trade. If an institution anticipates quick changes in price, the institution will trade aggressively. In other cases, the institution will exhibit patience to minimize price impact. The actions of these institutional traders reflect their fiduciary responsibility to their beneficiaries. It is thus inappropriate for the SEC to dictate the trading strategy as well as the display requirements of the institutional order, unless it can demonstrate strong externalities associated with the display. Forcing the display breaches the property rights of the investors.

As with any change in market structure, Regulation NMS has had both positive and negative effects. It has dramatically increased competition among trading venues, as evidenced by the shift in market share to electronic markets. For example, the New York Stock Exchange (NYSE) is no longer the dominant trading facility for NYSE-listed stocks. The Regulation NMS clearly has reduced spreads and trading costs for the small trades associated with retail investors. The labor-intensive upstairs telephone market has been supplanted by low-cost electronic markets, such as the dark pools.

However, an unintended consequence has been the development of high-frequency trading, now roughly fifty percent of the overall volume. High-frequency trading takes advantage of latency (differences in the timing of flows of information to market participants) and co-location (locating trading computers closer to the market computer to obtain faster transmission). Even though such trading has become a political lightning rod in Congress, the SEC, to its credit, has not yet adopted restrictions on such trading. It has inquired in its Concept Release as to whether such trading enhances liquidity and whether it has increased or decreased short-term volatility. Just as with its 2005-2006 study of the up-tick rule for short selling, the SEC should facilitate the availability of data to evaluate various practices.

As noted, the SEC has shown a strong bias towards protecting the small direct retail investor. This bias is misplaced as the majority of retail investors invest through mutual funds. Moreover, the trades of small retail investors and institutional investors are fundamentally different and should not be viewed as interchangeable. Furthermore, direct retail investors with relatively less information are often attractive parties with whom to trade, as they are more likely to trade at a fair price and not exploit the other side of the trade. Consequently, competition gives them more favorable pricing than potentially informed institutions. In this case, price discrimination benefits the small direct retail investor. Indeed, this is as it should be and lies at the heart of most academic models of financial trading.

Lastly, the SEC’s historical approach to the corporate and municipal bond market has been completely different than its approach to stock trading. It has only recently required post-trade transparency through the publication of trade prices, but the bond market still lacks pre-trade transparency through the publication of consolidated quotes. In this type of market, the direct retail customer is at a substantial disadvantage. Despite the generally lesser price volatility of bonds in comparison to stocks, especially for high-grade bonds, retail spreads in the bond market are substantially larger than in the equity market and the institutional bond market. Rather than trying to determine whether it could improve the functioning of the equity market, the SEC should devote more of its energy to improving the efficiency of the bond market.