Remarks before the SIFMA Equity Market Structure Conference

Thank you, Randy [Snook], for your kind introduction, and for accommodating my schedule by allowing me to deliver my remarks at the beginning of today’s conference. It is a special privilege for me to be able to speak at SIFMA’s market structure conference at this early point in my tenure as a Commissioner. In recent years, I thought it would be a refreshing change to focus on less controversial issues than, say, the Dodd-Frank Act. So it was that I turned to the Securities Acts Amendments of 1975, Regulation NMS, and market structure. And increasingly, I found myself puzzling over the strange role that the Commission plays in directing—and often determining—the evolution of U.S. equity market structure and wondering where it had lost its way. This morning, I’d like to share some initial thoughts on this history and how it might help us think more clearly about the market structure issues we currently face.[1]

It will surprise none of you, I’m sure, to hear me say that my thoughts on these issues are my own and that the views I express this morning may not reflect the views of the Commission or my fellow Commissioners. But I hope that you’ll hear me out anyway, because I think it’s critical, as today’s conference proceeds, to be willing to look at current challenges in our equity markets from the outside, as it were, not accepting our current regulatory framework as a given that determines and constrains the limits of future possible market structure reforms. I’ve long advocated retrospective review of regulation, but if retrospective review simply means looking back no further than specific recently adopted rules—and not a more searching review of the assumptions underlying those rules and the statutory and regulatory framework that gave birth to those rules—we risk finding ourselves unable to solve apparently intractable problems in our markets.

The Commission’s recently proposed transaction fee pilot, which I supported, is a case in point. It represents an important attempt to examine whether exchange transaction-fee structures incentivize broker-dealers to make order-routing decisions that conflict with brokers’ best-execution obligations to their customers. At the same time, however, it leaves unasked more challenging questions that go to the heart of equity market structure regulation, questions like the following:

• Is there any reason for the Commission to have any role in determining the fees exchanges can charge their members other than to counter incentives created by our own market structure rules?
• Do the key provisions of Regulation NMS distort the incentives of exchanges and broker-dealers in routing and executing transactions?
• In a world where communications technology continues to enhance the flow of information and reduce the costs of transparency, is there any justification for the Commission’s command-and-control approach to regulating how orders interact and how investors communicate in the equity markets?

Avoiding questions like these allows us as regulators to stay in our comfort zone—a safe space with lots of regulations for us to poke and prod when we are stressed out. We remain secure in our ability to control the levers that tweak market participants’ incentives, fine-tuning them to generate conduct consistent with our vision of an orderly, efficient market. Avoiding these questions also preserves the casual assumption that inefficiencies in our capital markets invariably reflect market failure that justifies another round of regulatory action to save market participants from themselves, or from each other.[2] Avoiding these questions ensures that we persist in our reliance on what Louis Jaffe described as “[t]he nostrum most approved” by a regulator “for the ills of a regulated industry”: “more regulation.”[3]

These tendencies aren’t new. Jaffe’s quip appears in a reflection on the career of the SEC’s second chairman, James Landis, published in the Harvard Law Review in 1964. In much of our economic life, however, we’ve moved away from the style of command-and-control regulation that still characterizes the Commission’s approach to regulating the structure of the equity markets. This morning, I’d like to spend a little time reflecting on why we find ourselves in this odd position.

Critics of this top-down approach frequently point to decisions the Commission made in adopting Regulation NMS in 2005. My boss at the time, Commissioner Paul Atkins, dissented from the Commission’s adoption of that rule, and it’s instructive to read the statement that he and Commissioner Cyndi Glassman issued in dissent: They understood the incentives the regulation would create and predicted many of the ways these incentives would distort market behavior, including the near-exclusive focus on price as the measure of execution quality, the limitation of competition and innovation, and, importantly, the proliferation of ancillary regulation, which would be required to mitigate the distortions introduced by the Commission’s own rules.[4]

It is beyond question that the adoption of Regulation NMS was a watershed event in the development of U.S. equity market structure. But, in my view, it’s helpful to look back thirty years prior to Regulation NMS, to the enactment of the Securities Acts Amendments of 1975, to understand why—perhaps uniquely among American marketplaces—every interaction of participants in the U.S. equity markets is carried out according to specifications either mandated or approved by the Commission.

The story of the great tumult that preceded the enactment of the ’75 Amendments has been told many times. The Paperwork Crisis of the late 1960s had raised questions about the ability of market infrastructures to respond to the increasing role of institutional investors in the market, as well as more generally about the market’s ability to perform its necessary role in fostering economic growth. The Commission responded to this crisis with a series of reports and statements in the early 1970s that articulated the case for a “strong central market system.”[5] The vision was of a market that leveraged composite quotation and transaction data to facilitate universal access and give individual investors access to “the best available price, no matter where it is being made.”[6]

One Commissioner at the time described this vision as representing a shift from taking “the industry as economics shaped it” to an “activist role” that would involve the Commission in “order[ing] the emerging forces in a rational manner.”[7] The 1971 Institutional Investor Study prepared by the Commission implied that this would be achieved via an approach that combined “appropriate regulation” and “the forces of competition.”[8]

There was, of course, as there often is, the suggestion that the phrase “appropriate regulation” would be interpreted more or less aggressively depending on the industry’s ability to fulfill the vision that the Commission had for it. Commissioner Sommer spelled this out for market participants in the final lines of a 1974 speech: The markets, he said, were in the midst of a revolution and when revolutionaries fall out with each other, “[t]hey often end with a strong man in the saddle.”[9] He expressed his hope that this result could be avoided, “because that strong man will be government, more of it than you will find comfortable once the revolution is over.”[10]

When finally enacted, the ’75 Amendments represented an ambiguous mandate to the SEC.[11] On one hand, they instructed the Commission to remove barriers to competition in the equity markets. On the other, they directed the Commission to manage that competition consistent with the public interest. The central mandate of what would become Section 11A of the Exchange Act—that the Commission “facilitate the establishment of a national market system”[12]—was capacious enough to accommodate either an industry-centered or a Commission-centered reading of the Amendments. And this ambiguity pervaded much of the commentary around the statute, with commentators dividing over whether the statute mandated industry-driven reform, or an SEC-directed restructuring, of the equity markets.[13]

Other parts of the statute, however, extended an unambiguous invitation to the Commission to involve itself in micromanagement of market infrastructures and market interactions. The exchanges saw their status as self-regulatory organizations confirmed, but they also saw their activity subject to extensive scrutiny by the Commission, which gained the authority to approve or disapprove SRO rule changes and to change or abolish those rules.[14] The Commission also was given “pervasive rulemaking power to regulate all organizations engaged in the business of collecting, processing, or publishing information relating to quotations for and transactions in securities.”[15]

Some might say that this generous grant of authority to the Commission was justified given evident market failures: After all, the equity markets had manifested obvious and severe deficiencies in the late 1960s and early 1970s; a small number of markets dominated equity trading and resisted the publication of quotes or trading data; and linkages between the markets “were primitive or even non-existent.”[16] There was, in this view, a compelling need for a robust regulatory response.

Yet, it’s not at all clear to me that the expansive role for the SEC mandated—or, depending on your view, implied—by the ’75 Amendments was warranted by the challenges confronting the equity markets at the time. Nor is it clear to me that such a role continues to be warranted in the very different markets that have developed over the intervening 43 years. Some have argued that, given the then-dominance of the New York Stock Exchange in particular, price competition would never have developed on its own and, accordingly, competition alone would not have forced exchanges to improve transparency and establish linkages with other markets.[17]

What this argument disregards, in my view, is a proper appreciation of the changes that were already occurring in the market in the early 1970s. Although the market for exchange services was indeed highly concentrated, with the New York Stock Exchange dominating that market, this dominance was due less to any so-called “natural genius”[18] of the Exchange than to a set of anticompetitive constraints imposed in its capacity as a self-regulatory organization that the Commission countenanced for over four decades.[19] These constraints included rules decreeing fixed commissions, prohibitions on off-board trading, and membership restrictions,[20] and together these constraints likely slowed the development of significant competitive alternatives.

Notably, the Commission already had the authority to address at least some of these issues prior to the enactment of the ’75 Amendments, as demonstrated by its abolition of fixed commissions between 1973 and 1975. And, what’s more, markets were already finding mechanisms—such as the evasion of fixed commissions through non-price competition—to get around some of them. Regulation that creates a framework in which competitive forces function properly and that keeps barriers to entry low—if judiciously applied—could have gone a significant way toward achieving the objectives for a national market system spelled out in the statute, such as “economically efficient execution,” availability of quotation and transaction information, best execution of customer orders, and so on.[21]

Fischer Black once noted, “financial . . . markets work, almost all the time, without the intervention of government.”[22] The ’75 Amendments encouraged the Commission to ignore this fact by giving it the tools to interfere, at will, “with the proper workings of the market.”[23] Congress asked the SEC to establish communication links between markets, even though participants in competitive markets always have aggressively sought out new technologies.[24] Congress expressed concerns about price disparities between markets, but arbitrageurs are ubiquitous where price disparities are pronounced.

Whether more aggressive moves by the Commission to discourage or prohibit anticompetitive restrictions would have addressed these challenges quickly enough for an impatient regulator and its even more impatient congressional overseers is an open question.[25] But remember: Even with the myriad Commission interventions into the details of market structure following the ’75 Amendments, a majority of the Commission felt that further intervention—and aggressive intervention at that—was required to realize the objectives of the National Market System a full three decades after those Amendments were enacted. It is entirely possible that our focus on perfecting our regulatory framework for market structure while neglecting barriers to entry and anticompetitive practices by dominant exchanges actually thwarted the entry of alternative sources of market information and impeded the creation of effective linkages between exchanges.

Given the unlikelihood of an imminent repeal of the ’75 Amendments, you may be wondering whether this historical reflection has a point. I believe it has several, actually.

First, this history counsels, I think, for humility in our estimation of regulators’ ability to micromanage how market participants communicate and trade with one another. It also counsels for humility in our assessment of the benefits of this micromanagement over the past 43 years. Outside the financial markets, our economy demonstrates on a daily basis that even markets that haven’t been comprehensively planned by a regulator can be wondrously effective in allocating resources to their best uses.

Our tortured dalliances with the Consolidated Audit Trail might serve as a reminder to be humble. Even the best laid CAT plans—plans for an enormous but seemingly fairly straightforward data-collection project designed by a committee of experts that was in turn designed by regulators—prove enormously difficult to execute, while the market builds systems of far greater complexity with silent efficiency. CAT, by design, assigns decision-making rights to entities who do not bear the full costs of those decisions. If that happened in the markets, we would call it failure, but we accept such a delinking on the regulatory side, with its unavoidable consequences.

Second, as we progress with further market structure reforms, we should be willing to consider eliminating rules that interfere with—or even foreclose—efficient methods of communication or market interactions rather than imposing additional rules that merely mitigate the effects of prior regulatory choices. The fact that we have a rule in place that addresses a particular market structure issue should not be interpreted as prima facie evidence that the rule was a necessary or appropriate response to market failure when it was adopted, or that it remains necessary or appropriate given changes in the market. In fact, we need to be open to considering whether the rule addressed a market failure at all, as opposed to a prior regulatory error.

Consider, for example, the order protection rule. This rule has exerted an outsized influence on the current shape of our equity markets, but the justification for the rule was never clear: As Commissioners Atkins and Glassman pointed out in their dissent, rationales for the rule were “a moving target,” shifting “from the protection of limit orders, to the need to increase market depth and liquidity, to the reduction of transaction costs for long-term investors and issuers.”[26] And whether it was actually necessary to achieve any of these objectives remains unclear.

But it is clear that the rule distorts the markets in a number of ways. It pushes broker-dealers to prioritize price above all else in their execution decisions, regardless of the needs or preferences of their customers. It magnifies problems with the Securities Information Processor feeds by increasing their importance. It induces brokers to subscribe to exchanges’ private data feeds in addition to the public tapes. And notwithstanding fragmentation of the market among a wide range of exchanges and other platforms, it pushes all trading facilities toward increasing homogenization, and restricts exchanges, in particular, to distinguishing themselves from each other in ways that don’t necessarily enhance issuer and investor choice or serve their interests.[27]

Given the lack of a compelling justification for the rule and its questionable effects on the market, I would suggest that we should start considering whether it’s worthwhile to continue focusing on regulatory tweaks to constrain market behavior that is driven primarily by the incentives created by the order protection rule. Shouldn’t we instead begin exploring alternatives that provide investors with sufficient information and flexibility to determine, in light of their own trading objectives, whether their broker is capable of executing their trades in a manner consistent with best execution, again, in light of the investor’s objectives?

Third, we need to be sensitive to how statutory and regulatory requirements may be distorting—or inhibiting—competition, whether through barriers to entry that deter new entrants or through requirements that prevent exchanges and market participants from developing novel responses to investor and issuer needs. Take, for example, a feature that has been fundamental to market regulation in this country from the very beginning: self-regulation. The Commission’s long-standing reliance on self-regulatory organizations has a certain logic to it – and indeed a certain appeal for someone like me, who tends to be skeptical of government’s ability to respond nimbly to the needs of rapidly changing markets. SROs are, by their very nature, closer to the markets, and seem better positioned to incorporate information about market developments into their regulatory framework in a timely manner.[28]

But commentators have long noted that self-regulation of an industry can impede competition within that industry, even through apparently beneficial provisions such as codes of conduct.[29] And it’s worth considering whether these concerns about the unavoidably anticompetitive effects of self-regulation become even more problematic in a world where exchanges have demutualized and increasingly see themselves as competitors to at least some of their members. Among other things, when SROs are for-profit corporations and compete with at least some of their member broker-dealers, does it make sense for them to play a uniquely prominent role in developing and administering national market system plans? I have no answers here yet. Only questions. But I think these are critical questions that I encourage all of you to continue asking both today and going forward.

Finally, we need to recognize that a national market system is not a goal to be achieved. Rather, it should be a continuously developing system that is agile enough to respond quickly to new technological developments and, more importantly, to changing investor and issuer needs. Recognizing that we are overseeing dynamically evolving markets that are capable of processing and acting on enormous quantities of information—quantities that are far beyond the ability of all of the financial regulators in the country combined to process, much less understand—should check the temptation to think we can regulate our markets to perfection. If we can resist this temptation, we have a better chance of avoiding the inflexible uniformity that utopian schemes unfailingly produce.

Investors and issuers have a wide range of needs and preferences in providing and obtaining capital, but the ’75 Amendments and Regulation NMS impose a particular vision on how market participants should communicate and interact with each other, restricting the ability of trading platforms to innovate in ways that could address these needs and preferences. Exchanges are now forced to compete to attract order flow rather than to provide innovative solutions to issuer and investor needs. Larry Harris has argued that the resulting uniformity across exchanges means that listing decisions become decisions about the marketing image an issuer is trying to promote rather than the unique characteristics of the market itself.[30]

To take one example, consider the challenges that thinly-traded stocks face in our highly regulated and carefully calibrated national market system. I am eagerly looking forward to the roundtable that our Chairman and our Director of Trading and Markets, Brett Redfearn, described in their speeches in Chicago last week.[31] I think it’s well past time to consider possible alternatives, including the recommendation in the Department of the Treasury’s Capital Markets Report that the SEC consider allowing issuers to suspend unlisted trading privileges in their shares.[32]

But I also think it’s worth taking a step back and reminding ourselves how remarkable it is that the SEC needs to spend resources attempting to identify a regulatory solution to an issue like this. Markets solve this kind of problem through private ordering all the time. It’s what markets do, and do remarkably well, unless they are prevented from doing so, for example, under force of law. I’m convinced, on this issue at least, that trading facilities, issuers, and investors could find a solution, if we would only allow them the freedom to do so.

I apologize for serving you up some depressing history with your morning coffee, but I hope that you will approach the rest of the conference with the lessons we can draw from that history in mind. I commend to you the upcoming remarks of Brett Redfearn. It’s wonderful to have him as our Director of Trading and Markets. Having spent years in the thick of markets, he knows that they work. Enjoy your conference. I am happy to take some questions, but might point you to the markets for the answers.