SEC Investor Advisory Committee Meeting
Securities and Exchange Commission
Investor Advisory Committee Meeting
Thursday, June 10, 2021
Panel Discussion Regarding Best Execution and its Role in post-NMS Market Structure
Panelists:
- JW Verret, Associate Professor, Antonin Scalia Law School (Moderator)
- Sal Arnuk, Partner & Co-Founder, Themis Trading LLC
- Paul Atkins, Former SEC Commissioner
- Tyler Gellasch, Executive Director, Healthy Markets Association
- Daniel Gray, Senior Special Counsel for Market Structure,
Division of Trading and Markets, U.S. Securities and Exchange Commission - Chester Spatt, PhD, former Chief Economist at the SEC, and
Professor of Economics at Carnegie-Mellon
Arnuk remarked that best execution is evolving and complex. He added that national best bid and offer (NBBO) does not account for a variety of market factors and is stale because there are two price speeds. He stated that banning payment for order flow (PFOF) would guide our markets to the best state of price and demand as well as eliminate market fragmentation.
Atkins asserted that Regulation National Market System (Reg NMS) has downsides, diminishes transparency, fuels trading volatility, and promotes flash crashes, none of which are good for markets. He also stated that Reg NMS and Rule 611 have not served market participants well and that the concept of best execution depends on many factors like certainty, speed, and size. He added that NBBO should no longer be used, and that Reg NMS concentrates market participants as well as hurts competition and innovation, so the SEC should question what that concentration means for the market. Atkins recommended that the SEC take a comprehensive view of equity market structure, including Reg NMS and a removal of Rule 611.
Gellasch agreed with Arnuk’s identification of the most pressing issues. He stated that today the system exists in which riskier trades can be made more easily than ever before and that investors are more diverse. He said the discussion of best execution is bigger than if retail investors are getting the best price on a stock and that it is about the impact on all investors. Gellasch commented that order routing incentives peg the broker’s fiduciary obligation directly against their customer’s best interest and that order routing creates segmentation. He stated that the attraction to particular venues means orders are not interacting.
Spatt stated that best execution is complex because of the innovation of financial technology and suggested that the ambiguity in the meaning of best execution means that the SEC wants best execution to develop organically. He added that NMS is at the core of the complexity of our trading system. Spatt suggested that the SEC should consider banning pricing tiers for rebates and, absent that, should implement much stronger disclosure requirements, which are important in the context of best execution. He added that another important best execution factor is access to data and that latency is not going away simply because markets are moving faster. Spatt stated that the underpinnings of best execution require further attention from the SEC.
Gray highlighted three SEC developments over the past year that impact best execution. He first spoke about the SEC adopting market data infrastructure rules resulting in round lot size reduction, odd lot exclusion from NBBO but inclusion in NMS market data, and quotes outside of NBBO being included in NMS market data. The second development he mentioned was the enforcement order with Robinhood regarding violation of a broker’s duty for best execution. The final development he mentioned was the expanded Rule 606 disclosure requirements including more detailed disclosure for certain types of orders.
Question & Answer
Mina Nguyen, Managing Director, Jane Street Capital, asked about PFOF as an incentive structure and moving retail trading onto the exchanges. Arnuk responded that banning payment for order flow in the dark and not doing it for institutional routers and exchanges will lead to unintended consequences. He added that when we craft specific rules, they end up being unusable in the market and that simple, principle based regulation works best. Arnuk concluded that removing the “maker-taker” system on the stock exchanges would reduce distortions in order routing in the market. Nguyen asked Gellasch about the role of zero commissions, and Gellasch commented that there is a disconnect between PFOF and pricing trades. Regarding best execution recommendations, Gellasch said there is no rule about what best execution means for investment advisors and that the SEC needs to enforce, in conjunction with FINRA, best execution as it exists now in FINRA rules and create an expectation of what best execution means for investment advisors.
Paul Mahoney, Professor of Law, University of Virginia School of Law, asked if markets are too fragmented and what the SEC should do. Arnuk said yes, the current systems are designed to segment order flow and that this is good for the exchanges but not good for investors. Gellasch added that because the exchanges are so similar, one might think they consolidate, but that does not happen because there is a value to the exchanges for having two places as opposed to one in which to trade. Spatt stated that the number of exchanges is a consequence of the structure of SEC rulemaking and that sometimes the structure of regulation changes the way markets are organized. He added that NMS directly leads to proliferation because it provides protection only at the top of the book and that the SEC needs to look at its rulemaking and encourage a simpler structure.
Jennifer Marietta-Westberg, Principal, Cornerstone Research, asked that if the SEC needs more information before conducting a rulemaking. Atkins responded by referencing the SEC’s practice of taking a more holistic look at the information gathered by staff over time. Spatt asserted that the SEC potentially could do a much better job of facilitating access to market structure data by the academic community. He also noted his disappointment in regards to the Circuit Court ruling on the fees and rebate pilot. Spatt also said that the agency problem has been made more severe by the tiering of rebates and that academics do not have relevant information about differences in pricing across different market participants. He concluded that the SEC has access to that information but does not have the resources to study the data, asserting that academia does. Gellasch followed up by expressing disappointment that we are relying on private market participant studies for information on pricing tiers and that the SEC could take action there. He concluded that the SEC did require enhanced 606 disclosures but that they should revisit 605 and 606 to provide market participants the information they need to make informed decisions.
Brian Hellmer, Managing Director of Global Public Market Strategies, Wisconsin Investment Board, asked if it is more productive to revise the definition of best execution or to focus on eliminating conflicts of interest and encouraging transparency. Spatt responded that both directions are relevant and that a lack of clear definition has the advantage of allowing things to evolve organically but a disadvantage is the lack of clarity around the responsibilities of the parties. He added that a lot of this discussion is framed around a world where latency is not an issue and that there is a need to review what best execution means. Arnuk weighed in regarding legislating best execution adding that the SEC and the relevant committees are not going be able to come up with a good definition of best execution. Gellasch said that best execution is only valuable if it is enforceable but is too hard to litigate and regulate. He concluded that the SEC should focus on policies, procedures, practices, and eliminating conflicts of interest. Atkins followed up by adding that best execution should be principles based and that it is hard to have a precise best execution definition.
Verret asked about insufficient disclosure in terms of payment inducements for order flow, and Gellasch stated that, regarding order routing decisions, the system is complex but that having more transparency in a conflicted system is worse than just having a less conflicted system. Spatt agreed that a less conflicted system is needed but that there are various market participants that prefer a conflicted system. Arnuk made the point that they are looking at the wrong things when legislating order routing incentives and that the market needs to be simplified.
Verret also asked if the SEC can have a bipartisan discussion around the Trade Through Rule and its unpredictable consequences. Gellasch responded that the Order Protection Rule (OPR) is a backstop and that it makes no sense that the burden is on an exchange to route to a competitor. He questioned whether most market participants would be able to determine when and whether they are getting the best price on their order and that eliminating the OPR without an alternative will not protect most investors. Arnuk added that we should look at PFOF bans in other countries where it has been successful. Atkins agreed that it is good to look at other countries but that if PFOF is banned, something else will arise and the SEC needs to be careful when looking at best execution.
Theodore Daniels, Founder and President, Society for Financial Education and Professional Development, asked what the industry is saying about conflict of interest and best execution. Gellasch responded that the discussion has been around how orders are routed and needing more information about order routing to know if brokers are fulfilling their best execution obligations.
Panel Discussion Regarding Best Execution Issues Unique to Wholesale Brokers
Panelists:
- JW Verret, Associate Professor, Antonin Scalia Law School (Moderator)
- Doug Cifu, CEO, Virtu Financial
- Stanislav Dolgopolov, Chief Regulatory Officer, Decimus
- Hitesh Mittal, Founder and CEO, BestEx Research
- Larry Tabb, Director, Market Structure Research, Bloomberg, LLP
In his presentation, Dolgopolov outlined the background of the wholesaler market. He said the business model of wholesaling in equity markets has been relatively stable over the years. Another point he made was that typically, there is no direct customer-broker relationship between wholesalers and ultimate customers, but the duty of best execution can extend to multiple players. He said although FINRA is quick to be critiqued or blamed, they have created a very sophisticated framework. He continued with a comparison between the U.S. and Europe’s regulatory regime, pointing out that in Europe, the tick size does apply to wholesalers. Dolgopolov also highlighted various weaknesses as it relates to best execution in wholesaling: enforcement actions, overlapping regulations, zero-commission trades, rerouting practices, price improvement practices, odd-lots, front-running, use of different data feeds, and wholesalers operating as dark pools. He listed potential solutions, emphasizing that he is skeptical about banning PFOF and or zero-commission model, but is more supportive of a pass-through for PFOF, standardized disclosures, or reforming the tick size regime.
In his presentation, Tabb said retail investor orders generally do not go to exchanges, they are filled by market makers over the counter (OTC) and noted that the majority of OTC shares are sent by retail. Tabb said the vast majority of retail flow is executed by six wholesalers. He said although the market is concentrated, it is still very competitive. Tabb continued to explain that the compensation model for retail brokers occurs via PFOF. He said investors do receive price improvement. He then explored the question of how retail investor orders can be priced better, to which Tabb said it is a function of supply and demand as larger order sizes have more market impact and push prices. He said while this market structure is serving retail well, it is not the best. Tabb said the market is segmented as there are 271 venues to get an order filled, the process of negotiating this framework is very complex, there is too much traded OTC, and that PFOF creates perceived and possible real conflicts. However, he said PFOF should not be banned because it subsidizes the cost of providing zero commissions. Tabb explained that if retail brokers did not route orders to wholesalers, they would need to build out trading decks which is expensive and creates other conflicts. His first recommendation was to increase transparency by updating the 605 Reports, improving client reporting, and improving data timeliness. His final suggestions were to improve the odd-lot process, tighten spreads, and make best execution demonstrable.
Mittal discussed his paper on PFOF and what impact the wholesaler market structure has on market participants and retail investors. From the retail investor perspective, Mittal said they receive better spreads than they would at an exchange. He explained that moving orders to the exchanges would reduce the spread, but provide less benefit to investors than they receive from price improvement with wholesalers. Mittal discussed the concerns around information and trading ahead, but said that wholesalers are entitled to use the information about order flow to predict the directing of future order flow. However, he warned that continued information asymmetry is likely to drive other market makers out of the business and decrease competition, and added that more information creates higher spreads which benefits the market makers. Mittal said a PFOF ban would not help, and that these issues can only be improved through market structure reform. He said the major issue is rooted in the asymmetry of rules as different participants face different requirements. As an example, he said there is a lack of obligation to check prices with multiple dealers on an order-by-order basis because it does not apply to retail brokers in equity markets. He also said fair access obligation only applies to alternative trading systems when they cross a certain volume threshold, but that it is not applicable to single-dealer pools.
Cifu said the current market structure provides meaningful and measurable benefits to retail investors. He listed many of the benefits such as significant price and size improvement, superior order handling and execution quality, robust broker competition, competitive and resilient execution ecosystem. He presented data to show that wholesalers fill orders at prices better than the NBBO. Cifu said it is incorrect to think that bid offer spreads are widened by the wholesaler model and would narrow if executions took place on the exchange. He said in their control and test groups they found the spread actually widens on the exchange. Cifu expressed his belief that there is no lack of competition for retail order flow, noting that there are eight major wholesalers.
As for possible solutions, Cifu said he supports updating Rule 605 because it could result in more accurate measurements that reflect the price improvement and execution quality received by retail investors.
Question & Answer
Marietta-Westberg asked for specific recommendations for brokers on the issue of price improvement. Tabb said it would be great for brokers to display their price improvement, and also strike the NBBO in an email back so the investor so they can see the pricing.
Jamila Abston, Partner, Ernst & Young, asked for further explanation on the information asymmetry issue. Tabb said there are many venues which people can go through for orders, so there is segmentation within the exchanges, and getting rid of PFOF or forcing orders away from the six wholesalers is not an adequate fix. Cifu said the wholesalers have been compared to Facebook but that it is false, and believes the retail brokers do a good job managing client data. Cifu said every time they execute a market order there is a print to the TRF, meaning price improvement statistics are out there and available to other sophisticated market participants.
Hellmer said that price improvement shows who benefits from the current situation and said there is an agreement that the price improvement calculation is flawed. He asked why there were differing viewpoints that reforming this would either hinder or help price improvement. Cifu said they recut all of their 2020 data for the tightened spread by odd-lots, and lost 3 percent of their price improvement in 2020. He added that if they did a marketable order by the odd-lot, they get no credit for improving odd lot orders. Mittal said there is price improvement versus the NBBO spread, but that his point assumes the bid offer spread will remain constant if all retail flow went to the exchanges. In which case, Mittal said if flow goes to the exchanges, the bid offer spread will go down by 25 percent.
For more information on this hearing, please click here.
For an archive of past SIFMA hearing coverage, please click here.