The Exchange Act defines “dealer” broadly to encompass any person that is “engaged in the business of buying and selling securities . . . for [its] own account,” unless it is not doing so as “part of a regular business.”[1] Drawing the line between dealers and other active participants in our markets has long been challenging, and those challenges have only increased as our markets have evolved over the decades. Today’s proposal seeks to bring some clarity to how this definition applies in markets where some of the key providers of liquidity are not registered as dealers and may not be subject to any Commission registration framework. It would do this by establishing a qualitative test and, solely for the government securities market, a quantitative trading threshold to determine whether such firms should be required to register as dealers.

While I have deep reservations about the breadth of today’s proposal, it addresses some important issues on which public comment will be valuable. I particularly appreciate the use of the rulemaking process to clarify the scope of the term “dealer.” Using the notice-and-comment rulemaking process to solicit public input to inform the Commission’s response to market and technological evolution, in lieu of extending ambiguous provisions of the law through enforcement actions against unsuspecting market participants, demonstrates a commitment to transparency and good government that should be the hallmark of all our work at the Commission. In addition, I appreciate the concerns expressed in the proposal about the largest firms making markets in U.S. Treasury securities sitting outside of the dealer framework.

To be clear, however, I am not convinced that the proposal gets the scope of its clarification of the dealer definition right, and I look forward to hearing what commenters have to say in response. In particular, I look forward to commenters’ views on the following questions.

1. What will the proposal’s effect be on liquidity in different markets? The market generally benefits when there are more, rather than fewer, liquidity providers. A more varied set of liquidity providers also benefits market resilience; when one type of liquidity provider is unwilling to step in, another may be able to fill the gap. A market in which all of the liquidity provision is concentrated in a handful of large dealers regulated on the traditional model—which the proposal seems to favor—may impair market liquidity without increasing market resilience.
2. Does the proposal articulate a sufficiently clear rationale—other than leveling the regulatory playing field—for requiring active traders who do not have any customers to register? What benefit would come from requiring active liquidity providers to register as dealers? The proposing release seems to take the position that some non-dealer liquidity providers are too highly leveraged and too risky for their own good and that of the markets,[2] and that dealer rules would be a handy tool to tamp down this leverage. One of the reasons our capital markets are so dynamic, however, is that we allow market participants to fail. Externalities matter, but the market is quite good at absorbing bad decisions by individual firms, albeit usually not at a price that the failed firm likes. Taking steps that seem designed to save market participants from themselves sends a terribly destructive message—the government is watching out for your risk-taking, so you do not have to. Markets are actually more resilient as a whole when individual firms know they can fail and that, if they do, nobody will be there to rescue them.
3. Is there something unique about the Treasury markets that makes dealer registration of large liquidity providers in that market appropriate? As the proposing release notes, not only have Treasury markets grown dramatically, but “technological advances have increasingly enabled certain market participants that are not registered as dealers to perform critical market functions, including liquidity provision, that once were primarily performed by regulated dealers.”[3] Would regulating these market participants as dealers help to address the types of market events we have seen over the past several years?
4. Should we create explicit exemptions from dealer registration for any categories of liquidity provider? Although the Securities Exchange Act defines “dealer” broadly, we also have broad authority to craft appropriate exemptions. For example, does it make sense to subject to the dealer regulatory framework proprietary market participants that do not have any customers and that add to market liquidity through their own regular trading activities? If so, what conditions, if any, should be placed on those exemptions?
5. Is the proposal’s approach to private funds, and the investment advisers that advise them, reasonable? Although the proposal appropriately excludes registered investment companies entirely from application of the rule, private funds and investment advisers that meet either the qualitative or quantitative test set forth in the proposed rule would be required to register as dealers. How many private funds and advisers would be captured by the proposed rule? Is the proposed rule more complicated than it needs to be to pull in funds and advisers that should be registered as dealers? Would it even be feasible for funds and advisers to register as dealers?
6. What lies down the road? While the obligations on dealers that would have to register under the proposal would be relatively limited initially—self-regulatory organization membership, net capital, transaction reporting, and recordkeeping and certain other requirements—experience suggests that, once registered with us, firms are likely to see their regulatory obligations grow over time. Once registered with us, firms become attractive candidates for additional regulation. Look no further than our recent proposal to vastly expand the requirements on private fund advisers for an example of this phenomenon at work.[4] What are the ancillary and future regulatory burdens that firms required to register as dealers under this rule should expect to incur?
7. How does this rule interact with other recent Commission proposals? The Commission has issued a large number of extremely complicated rules over the past several months. Do commenters have sufficient time and understanding of existing proposals to comment on how this proposal might be affected by—or might affect—other rules that have been proposed. To take one example, the proposal notes that the rule is designed to capture dealing activity wherever it occurs, including on Communication Protocol Systems, which were the subject of a proposal the Commission approved in January of this year.[5] What issues might this raise?

Thank you to the Division of Trading and Markets, the Division of Economic and Risk Analysis, the Division of Investment Management, and the Office of General Counsel for your work on the release and our many discussions about its contents.