Participants at a CFTC roundtable debated elements of proposed Regulation Automated Trading (“AT”). The participants discussed: (1) amendments to the proposed definition of “Direct Electronic Access” that would be consistent with and in furtherance of Regulation AT’s proposed registration regime; (2) quantitative measures to establish the population of AT Persons; (3) an alternative to imposing direct CFTC pre-trade risk control and development, testing and monitoring standards on AT Persons; (4) AT Persons’ compliance with elements of the proposed rules when using third-party algorithm or systems; and (5) source code access and retention.
Two CFTC commissioners issued statements pertaining to the meeting. CFTC Commissioner Sharon Y. Bowen identified “the positive impact” the proposed regulation will have on market stability and emphasized that the roundtable itself was intended to encourage discussion of stakeholders’ concerns. Nonetheless, she stressed that “time is of the essence” and that the CFTC “owe[d] it to stakeholders and end-users” to finish a strong and effective regulation on automated trading this calendar year to give “market participants and consumers increased confidence that algorithmic trading is properly regulated and that our markets are properly functioning.” Commissioner Bowen stated: “In the last few weeks, first at the Market Risk Advisory Committee Meeting and subsequently in individual meetings with stakeholders, I have heard increasing anxiety about the state of algorithmic trading from end-users.”
CFTC Commissioner J. Christopher Giancarlo asserted that although the proposal is a “well-meaning attempt by the [CFTC] to catch up to the digital revolution in U.S. futures markets,” Regulation AT has a “seemingly broad scope, hazy objectives and several significant inconsistencies.” He described the proposal’s requirement that proprietary source code be accessible to the CFTC and the Justice Department as “notorious” and stated that it “should come as no surprise that law abiding businesses are very concerned with the prospect of handing over highly valuable, proprietary business source code to the CFTC.” Commissioner Giancarlo emphasized that the proposal is a missed opportunity to respond to the emerging challenges of algorithmic trading and concluded that:
It is time to formulate and establish well-considered policy responses to the digitization of contemporary markets and then take action in a deliberate and thorough manner to enhance market liquidity, safety, and soundness.
Lofchie Comment: The differing viewpoints of CFTC Commissioners Giancarlo and Bowen serve to illustrate three separate, recurring issues in financial regulation: (i) the desirability of speed vs. care; (ii) the demands of the government for information; and (iii) the significance of the debate itself.
As to the issue of speed vs. care, both Commissioners agree on the desirability of formulating rules to govern automated trading. The argument that Commissioner Bowen advances seems fairly weak; i.e., the CFTC must deal with public “anxiety.” Mitigating consumer anxiety should not be the basis for justifying regulation. Financial regulation should have more measurable goals.
Commissioner Bowen’s argument that rules can be justified as public “confidence” measures – that having enough rules in place makes markets work better just because they are there – should be challenged. The counter to this argument is that too many rules undermine public confidence. In fact, we have thousands of rules; the government manufactures them far faster than industry can keep up. That is why the CFTC is forced to issue hundreds of no-action letters; market participants simply can’t keep up. Rules once adopted are not readily liberalized or withdrawn. Generally, rules reproduce like rabbits and tighten like boas. The CFTC would be better off not adopting still more rules in the absence of a more tangible outcome than anxiety reduction.
As to the demand of the government for information, the government is constantly requiring more information without formulating any plan as to how it will receive, store, compare or use that information. Where is the plan regarding algorithms? How is information such as the proprietary source codes to be protected? Will market participants be indemnified by the government if their information is “stolen” in any way? (Of course, they will not, but it is still something to be considered.)
Finally, whether one agrees or disagrees with Commissioner Bowen’s or Commissioner Giancarlo’s points of view, it is good that the governance structure of the CFTC (as well as the SEC) creates the opportunity for public disagreement. The opportunity for such a debate does a great deal to keep government just a little bit more honest and, in some instances, predictions of the dissidents are unhappily realized (e.g., the dissents over the SEC’s adoption of Regulation NMS). This governance structure stands in contrast to the CFPB. The absence of any means of dissent is one of the great flaws (but not the only one) with that agency.
The President’s party has now the majority vote in any agency (whether it is a majority of one to zero or three to two) and, therefore, the existence of the opportunity for dissent does not imply that the wheels of government must stop turning, only that sometimes the turns should be made a bit squeakier.