By Jack Miller, Head of Global Execution Services, Baird
When I lived in New York and my kids were very little, we were fortunate enough to live in a “doorman building.” These hard-working and omnipresent men and women were a critical part of the operations of the building and the experience of living there. They would greet you when you came and went, let you know when you had a package or delivery, make friends with your kids, and, yes, open the door. And, while they weren’t security guards per se, they did, much like flight attendants who are there “primarily for your safety,” make sure that anyone coming into the building had a reason to be there and kept away those with otherwise malicious intent.
The advertising executive and author Rory Sutherland talks about the “doorman fallacy.” A version of this involves an apartment building whose management laments, “Our expenses are out of control. Do we really need to pay someone to open the door? There have been no robberies at the building so why are we keeping so many ’guards’ on staff? This is a safe neighborhood!” So, the staff is released; but then the situation changes. The observation that is obvious in hindsight is that perhaps the door people played a role in defining the experience of living in the building, including safety, and the absence of bad outcomes was not evidence that there was no threat.
Rule 611, or the Order Protection Rule, is perhaps the defining component of Reg NMS, which in turn set the stage for our current US market structure. For those of us who came up in the industry after the implementation of Reg NMS, the pre-NMS world is imagined as kind of a Wild West, a dangerous world with absurdly wide spreads, SOES bandits and other nefarious characters, technology disparities, and privileged access to liquidity and prices that make today’s debates about fair access and level playing fields look “cute” by comparison.
Reg NMS was a new sheriff, and with it came its deputy, the OPR. First, all the exchanges need to connect to each other electronically. Then, if one exchange displays the “best price,” another exchange cannot accept a worse price. Instead it *must* route incoming orders to the exchange displaying the best price rather than executing against its own inferior price. If two market participants agree they *want* to trade at a different price (for example, in case of a block), they can still do that – an allowable “trade through” as long as they “sweep” the top of book, thereby satisfying the market maker or investor who had been willing to display the best price. Yes, we must now get into the messy business of defining what exactly qualifies as “protected,” but the point is that we now have a single national market system virtually unified through technology with a single national best bid and best offer (NBBO), and investors need not worry about getting a worse price because they chose the wrong trading venue.
These days it’s hard to find anyone who will make a full-throated defense of OPR. Indeed, it is imperfect and has led to many unintended consequences. Chief among them is the proliferation of protected venues. We now have 17 active protected exchanges in the U.S. and there are few barriers to adding more (indeed, there are more in the pipeline). Investors, brokers and other exchanges *must* access all of them which imposes a real cost on the industry (a broker dealer connecting directly to a single exchange runs easily into 6 figures annually, and an order of magnitude more when multiplied across all exchanges). Given that 10 of these exchanges maintain less than 1% market share, the market structure community at large is correct in questioning whether this state of affairs withstands a basic cost benefit analysis.
Other knock-on effects of OPR include the debates around access fees, rebates, tick sizes, and lot sizes. A well-defined NBBO must define exactly what is protected. How much of a given stock must be displayed to qualify for protection (i.e., the lot size)? How much must a market maker improve a price to establish a new best bid or offer (i.e., the tick size)? If exchanges are by definition interconnected, how do exchanges compete for order flow now that they can’t compete on the basis of unique access (i.e., the access fee and rebate debate)? These topics alone have consumed oceans of ink, but the point is they are premised on the idea that there exists some concept of an NBBO that serves as a backstop for best ex and a reference price for analysis.
Ironically, the NBBO is already not the de facto best price. In a retail trading context, investors expect (and usually receive) price improvement from the NBBO as a matter of course. Indeed, the much-maligned Order Competition Rule proposal set the assumption that investors *must* seek better pricing than the NBBO, which implicitly and paradoxically acknowledges that the NBBO is not the “best” price after all. The NBBO is also a poor indicator of prices for *size* that exceeds liquidity displayed on the inside (I have written about the limitations of the NBBO in my piece “One Size Doesn’t Fit All”, especially in the context of trading in institutional size where investors may willingly *choose* to execute at prices outside the NBBO).
Perhaps OPR is no longer necessary. After all, broker dealers are obligated to seek Best Execution for their clients. If better prices exist for clients, then agents handling their orders must execute at those prices whether there is an OPR or not, making OPR redundant. Furthermore, academic literature [1] has pointed out that a majority of order traffic is non-routable, thereby bypassing OPR protection. Said another way, modern routing technology and the interconnectedness of markets means we do not need to rely on OPR to access the best prices. If this is so, why do we need trade through protection in the first place? Perhaps a principles-based, rather than rules-based system is preferable or at least sufficient. The neighborhood has been cleaned up.
With all of these thoughtful and well-reasoned arguments in opposition to OPR, what arguments are left to its defense? This quickly becomes philosophical given we won’t and can’t know the full impact of modifications until they are implemented. Which, ironically, is perhaps the strongest argument in its defense. U.S. equity markets score better than any other equity market globally on most generally accepted measures of market quality, liquidity, and cost. Competition among venues – exchange and non-exchange – is robust. Given this, what problem are we fixing exactly, and are we willing to bear the cost of the unintended consequences?
Before we remove the doormen, it is worth considering the alternative reality? For all of its imperfections, the OPR does provide a framework for ensuring that investors can access the prices they see on their screens and liquidity providers have the opportunity to trade. Best execution is inherently in the eye of the beholder and no rules-based system can capture all perspectives, but OPR provides a backstop (or, at least, reference point) for quantifying it. I don’t know if U.S. market quality is because of or in spite of OPR, but we do have evidence that it is consistent with high-functioning markets, which is something we should protect carefully.
[1] https://www.nber.org/system/files/working_papers/w28515/w28515.pdf
