LOADING

Type to search

Articles

Who’s More Competitive: Wholesalers or Exchanges?

By Hitesh Mittal, Founder & CEO, BestEx Research

Hitesh Mittal

The most controversial rule among the SEC’s recent proposals will likely be its Order Competition Rule.

Currently, wholesalers can buy order flow from retail brokers and provide a fill inside the National Best Bid and Offer (NBBO) without exposing the order to open market competition. In the proposed rule, Retail brokers or their wholesalers must submit orders to an eligible exchange for an auction, with the exception of orders that wholesalers can fill themselves at the midpoint price or better. When an exchange receives a retail order in these auctions, it will disseminate information about the order as well as the name of the originating retail broker so potential liquidity providers will know that it is a retail order. 

Institutional investors will clearly benefit from the ability to interact with the less toxic retail order flow that is largely captive to wholesalers today, but how will it impact the prices received by retail investors? The SEC believes that if institutional investors and other market makers are able to compete on exchanges to provide liquidity to marketable retail orders in these auctions, they will be willing to buy at a higher price than the National Best Bid or at a lower price than the National Best Offer. But retail investors already receive price improvement over the NBBO from wholesalers, so the question is whether the price improvement experienced in auctions will be even better than what retail investors receive today. 

This is really a question of whether institutional investors and market makers on exchanges are more or less competitive than wholesalers. Wholesalers claim to be highly competitive, citing price improvements over the NBBO on exchanges as evidence. 

But comparing the spread on exchanges–where order flow is more toxic on average–to the spread wholesalers provide to retail investors is not an apples-to-apples comparison. Market makers care much more about “realized spreads” than the NBBO at the time of a trade. The realized spread is the spread earned at the time of the trade, adjusted by the adverse selection cost (how much the midpoint price moves against an order after the trade). If a buy order is executed at $10.05 and shortly after the trade the price improves to $10.03, the trader experienced an adverse selection of two cents. 

In their proposal, the SEC includes analysis of the Consolidated Audit Trail (CAT) and data from 605 reports. We also provided an economic analysis in our 2021 paper “Payment for Order Flow: The Good, The Bad, and The Ugly”. While there are differences in the data sources, time periods, and methodologies between the SEC’s analysis and our earlier research, the results of the two analyses are remarkably similar.

To compare the competitiveness of different market structures, we can calculate the ratio of spreads earned to the adverse selection costs incurred. If this ratio is greater than one, a trader is making money because they have earned more in spread than they paid in adverse selection. If the ratio is less than one, a trader is losing money, earning less in spread than they’re paying in adverse selection. 

The SEC’s study found that the ratio of spread to adverse selection is less than one on exchanges (72%) and greater than one for wholesalers’ trades against retail investors (167%), making exchanges 2.3 times more competitive. Our study, referenced above, found that exchanges were 2.6 times more competitive than wholesalers.

The SEC’s proposed Order Competition Rule assumes that if retail flow remains segmented (and thereby adverse selection costs associated with providing liquidity to retail investors is unchanged), the spreads on exchanges will be even lower than what retail investors currently pay through wholesalers. 

Based on the assumption of segmentation, the SEC estimates that retail investors would experience a cost reduction of 57% trading in their proposed auction mechanism. The math is straightforward. The current adverse selection cost on retail order flow is 1.26 basis points, on average, and the spread to adverse selection ratio on exchanges is 72%. Therefore, the spread that retail investors would pay on exchanges is 72% of 1.26 basis points, or .91 basis points (.0091% of the value of the trade). Currently, retail investors pay 2.11 basis points on average, and hence a significant projected savings of 57%. 

An important question to circle back to is why the average ratio on exchanges would be less than one. A ratio of less than one implies that liquidity providers on exchanges are so competitive that they are willing to make less in bid-offer spread than the adverse selection they experience. 

The answer lies in the mix of investors providing liquidity on exchanges. Market makers who are taking risk by providing liquidity will want to ensure that their quoted spreads are higher than the adverse selection they experience. But institutional investors are also liquidity providers on exchanges because they use execution algorithms that place limit orders as a strategy to defray the cost associated with putting on a larger, longer-term position. For example, our own execution algorithms provide liquidity over 90% of the time when trading at or below a 5% participation rate. Since institutional investors’ trading costs are covered by the larger goals of their overall investment positions, they’re aiming only to reduce those costs on average. Considered in aggregate, the average ratio for all participants on exchanges is less than one. And that’s great news for retail investors who should be compensated for the low toxicity in their flow. 

Overall, we support the SEC’s proposal, as it increases competition and will reduce costs for both retail and institutional investors. It will minimize the forced intermediation in the marketplace, allow institutional investors to access non-toxic retail liquidity and provide retail investors even better prices than they do today–reducing their costs by 57%. For more on how the proposed auction mechanism will work, our suggestions for improvements, and the impact of this change on institutional investors, download the long-form of this paper. We encourage readers to participate in this debate and provide their feedback to the SEC via comment letters.