How the Market Data Infrastructure Proposal Could Affect Free Trading Platforms

By: Tianna Larson


Robinhood has come under fire once again, this time for failing to properly disclose its payment for order flow practices. The trouble comes after recent enhancements to the order flow disclosure requirements, reflecting the SEC’s concern about the practice.[1] While controversial, order flow revenue supports most brokerage firms’ ability to provide free or low-commission trades to retail investors. In a crude sense, the practice reflects a steal-from-the-rich ethos. However, a recent SEC proposal[2] threatens the retail order flow internalization dynamic and could impact the continuity of free trading services.

Payment for order flow is a legal practice in which wholesalers (or “market makers”) pay brokers to route retail orders to the market makers, who then execute trades from their own inventory outside of the organized markets.[3] The market maker promises the broker that it will execute their orders at better prices than would be available on the public market. This is called “price improvement” and facilitates brokers’ compliance with their federal mandate to give customers “best execution.” Market makers supply liquidity by transacting in large volumes, and they profit by selling a stock for more than they purchased it and taking some of that spread (the “profit-spread”). They face a major adverse selection risk of selling to informed traders, who only buy (sell) when they know the stock’s value exceeds (is less than) the price of the market maker’s offer.[4] The risk of losing out to informed traders causes market makers to widen bid/ask spreads, which harms liquidity.[5] Importantly, retail investors are typically not informed and thus present little risk of adverse selection. If a market maker can ensure that it is trading only with retail investors, it will be willing to give them a better deal.[6]

Payment for order flow is one of the myriad aspects of market structure that could change under the SEC’s market data infrastructure proposal. No doubt, the proposal’s effect on retail order flow internalization would be indirect, and assessments about how the proposal would impact broker’s ability to provide free trading services are speculative. Nevertheless, the existing dynamic supporting free trades is at risk under the new proposal, and industry experts should explore this consequence further.

The two relevant rule changes are the new definitions of round-lot orders and protected quotations. Brokers transacting for customers are required to trade at the best available ask and bid price—the national best bid and offer (“NBBO”). The NBBO is calculated based primarily on round-lot quotation information. Round lots are a standard bundle of securities to be sold on exchanges, often in units of 100 shares or a multiple thereof. While current regulations require round-lot quotation information to be collected and disseminated, odd-lot information is not required to be displayed under SEC rules. Odd lots are orders in quantities less than a round-lot unit, and they occur more often with higher-priced shares.

The lack of odd-lot display harms price discovery because there may be more aggressively priced odd-lot orders that would not go into the NBBO. Accordingly, the SEC proposed to redefine round-lot sizes and “protected quotations” (quotes that are required to receive execution at a price equal to what is quoted on another exchange).[7] Widening the range of protected quotes and reducing the number of shares included in the definition of a round lot would likely lead to narrower NBBO spreads.[8] With a narrower spread, wholesalers have less opportunity to provide price improvement. Even if wholesalers are able to offer some price improvement in a narrower-spread environment, they will likely experience significantly reduced profits (because there is less profit-spread to keep). The Commission postulates that wholesalers may counteract the lower revenue per order by reducing any payments for order flow. In that case, the Commission suggests that “retail broker-dealers could respond by changing certain aspects of their business.”

The Commission’s postulation is vague. But given the extent to which brokers rely on order flow revenue to support their non-commission trading services, free trades may well be one aspect of business subject to change.

Could reduced order flow revenue be the end of free trades? That’s unclear. For one thing, the degree to which wholesales may reduce payment for order flow is ambiguous. Even market makers and brokers—experts in the best position to analyze these impacts—don’t know![9] For another thing, brokerages may be able to recoup many lost costs elsewhere. For brokerages like E*Trade and TD Ameritrade, order flow revenue can yield between 15 – 20% of the firms’ net revenue, though the percentage is far higher for Robinhood, whose order flow payments make up the primary revenue stream. But brokerages also accept rebates from exchanges to route orders to them, which could cushion the blow from reduced market maker compensation, and sometimes exchanges pay more than they charge to place trades. And, recent entrances of new exchanges are also likely to reshape equity market structure. Additionally, brokers use the allure of free trades to attract more customers, which invites more cash for brokers to sweep into their banking subsidiaries and generate interest. The profit from cash sweeping can contribute between 20 – 65% of a brokerage’s net revenue. Given these alternative revenue streams and the uncertain degree to which order flow payment will be reduced, it isn’t clear that brokerages couldn’t continue to provide free trades. Still, it is worth noting that a significant portion of broker revenue—the “profit engine in lieu of commissions”—could very well shrink under the proposed round-lot and protected quote definitions.

Equity market microstructure is complex and technical. This blog seeks not to predict outcomes or offer solutions; rather, it merely identifies how one of the many changes offered in the 600-page proposal could impact a narrow market activity. But, assume the SEC is right that payment for order flow will be substantially reduced, and assume that such a dent to brokers’ profit engines would hinder their ability to provide free trades. It would be important to consider the market impact of losing free trade services. Those platforms lower the entry barriers to a potentially lucrative activity and increase the number of retail investors in the market. Would removing a population of retail traders harm overall liquidity?[10] Or, would efficiency be improved by taking unsophisticated and speculative investors out of the market?[11] Should we protect investors’ access to the market regardless of economic reasons, because doing so is fair? Those questions, and others, should be considered before the proposed round lot and protected quote rules take effect.


[1] Payment for order flow has critics. They argue that payment for order flow creates a conflict of interest for brokers. Additionally, critics have argued that the system fragments the markets, creating a system where retail traders enjoy better spreads while institutional investors suffer wider spreads.

[2] The rule changes discussed in this post appear at pages 406–11 of the proposal. Market Data Infrastructure, Securities Exchange Act Release No. 88216 (Feb. 14, 2020), 85 FR 16726 (Mar. 24, 2020) (File No. S7-03-20) (Proposed Rule).

[3] Paul G. Mahoney, Equity Market Structure Regulation: Time to Start Over 19 (Va. L. & Econ. Research Paper No. 2020-11),

[4] Merritt B. Fox, Lawrence R. Glosten & Gabriel V. Rauterberg, Informed Trading and Its Regulation, 43 J. Corp. L. 817, 829 (2018). Informed traders include counterparties who trade lawfully on some informational advantage as well as those trading unlawfully on the basis of material non-public information. Id. In any manner, the result is that market makers are disadvantaged by the information asymmetry.

[5] See Andrew Verstein, Mixed Motives of Insider Trading, 106 Iowa L. Rev. __ (forthcoming 2020) (manuscript at 25–26), (proposing a principle that would enhance liquidity by causing narrower market maker spreads).

[6] Market makers profit on the bid/offer spread, keeping a portion of the profit-spread which is not given to the investor as price improvement. The ratio of the percent of the spread that goes to market makers versus the percent that goes to the client is a measure used to determine execution quality.

[7] Redefining round lots to include smaller sizes appears to be the path of least resistance, given that many of the Regulation NM rules refer, directly or directly, to round lots.

[8] Empirical research has found that odd lots tighten true spreads, especially for higher-priced stocks.

[9] Notably, Virtu Financials’ comment to the proposal requested a pilot program to further understand the impact to order flow internalization and expressed fear about the inability to provide price improvement. It further anticipated a “dramatically alter[ed] flow of orders to exchanges . . . .” While Fidelity expressed concern that these rules will make broker-dealer routing decisions more difficult,, neither it nor other brokers commented to express concern about reduced order flow revenue.

[10] Empirical research has found that retail investor attention improves stock liquidity. See Wenxuan Hou & Rong Ding, Retail Investor Attention and Stock Liquidity, 37 J. Int’l Fin. Markets Institution & Money (2015),

[11] Cf. Merritt B. Fox, Lawrence R. Glosten & Gabriel V. Rauterberg, Stock Market Manipulation and Its Regulation, 35 Yale J. on Reg. 67, 88 (2018) (describing how fad or fashion-driven noise trading can push stock prices away from that which reflects its fundamental value).