Options Order Flow and PFOF: How Retail Option Orders Are Routed, Paid For, and Executed

Payment for order flow (PFOF) in options is money a large market maker pays a retail broker for the right to handle that broker’s customers’ option orders. Unlike stock orders, US option orders must execute on an exchange, so the market maker that buys the flow sends each order to an exchange auction or order book where it, or an affiliate, usually ends up taking the other side.

In short: options PFOF is much larger than equity PFOF, both per trade and in total. In Robinhood Securities’ Rule 606 report for the third quarter of 2025, options produced about 77% of the order-flow payments it disclosed, at an average of about $0.53 per contract, against roughly $0.18 per 100 shares for stocks. The buyers of that flow are a handful of firms (SEC staff and academic research put the top three at as much as 90% of options PFOF). The evidence on whether customers pay for it through worse prices is mixed and contested, and the SEC’s 2022 attempt to reform retail order handling was withdrawn in June 2025. As of November 2026 the Commission is studying options market structure but has not proposed rules specific to options PFOF.

This article assumes you know how the options market is organized (see how options markets work) and how bid-ask spreads appear on a chain (see how to read an options chain). It explains how the industry operates. It is not a recommendation for or against any broker.

A common misconception: “zero commission means it’s free”

Commission-free trading replaced an explicit fee with implicit ones: the bid-ask spread you cross, and whatever price improvement you do or do not receive. PFOF is the link between them. A market maker can afford to pay a broker for orders because retail orders are, on average, less informed than institutional ones, so trading against them earns more of the spread with less risk of being run over. Whether the customer is better or worse off than without PFOF depends on how much of that value comes back as price improvement, and that is the question the research below tries to answer.

How an option order moves

The consolidator model

SEC staff described the structure in “Roundtable on Options Market Structure: Supporting Data” (Division of Trading and Markets, April 9, 2026), prepared for the Commission’s April 16, 2026 options roundtable. The key points:

Exchange-only execution. US listed options trade only on exchanges, and the Options Clearing Corporation generally accepts for clearing only standardized options that traded on an exchange. Even Dash’s alternative trading system routes the orders it pairs to an exchange auction for execution.
Consolidators. Large market makers act as “consolidators” by buying individual investors’ order flow and choosing the exchange where each order executes, “perhaps based on where they (or an affiliate) are most likely to trade with the order as principal.” Staff describe this model as accounting for virtually all retail options orders.
Venue tools. Some exchanges let a consolidator direct an order to an affiliated market maker, which then receives a guaranteed allocation (staff give 40% as an example) if it is quoting at the best price. Others give lead market makers an entitlement to orders of five contracts or fewer at the best price. Many offer price improvement auctions, where the consolidator brings a customer order paired with its own contra-side and other participants get a short window to offer a better price.
Fragmentation. By 2025 no single options exchange had more than 20% market share, and 18 venues each had more than 1%.

Exchange fee models and “marketing fees”

Options exchanges compete for consolidator flow through their fee schedules. Two features matter.

First, auction economics. In “Retail Trading in Options and the Rise of the Big Three Wholesalers” (Journal of Finance, 2023), Svetlana Bryzgalova, Anna Pavlova and Taisiya Sikorskaya report that on most exchanges an execution by the wholesaler-affiliated market maker in a price improvement auction is charged about $0.05 per contract, while a competing market maker that responds and breaks up the paired order pays about $0.50, and the wholesaler receives a net rebate of about $0.30 per contract for bringing the order. They conclude that fees are “stacked against responders,” which makes many paired trades prohibitively expensive to break up.

Second, on-exchange PFOF. SEC staff note that some exchanges administer “marketing fee” programs: certain market makers pay a fee when they trade with customer orders, the exchange pools it, and the pool is disbursed to designated market makers who used it to attract order flow, often by paying brokers.

Measuring PFOF: what a Rule 606 report shows

Under SEC Rule 606(a), brokers publish quarterly reports listing where they route customer orders and the net payments they receive, separately for S&P 500 stocks, other stocks, and options, broken down by order type. These reports are the main public window into PFOF rates.

Worked example: Robinhood Securities, third quarter 2025

Robinhood filed its Q3 2025 reports with the SEC as exhibits to a Form 8-K on October 31, 2025. Robinhood Securities (RHS), the clearing broker, states that every options venue paid it on the same schedule:

Order type	Per-contract payment received by RHS
Single-leg, 1 to 100 contracts	$0.30, $0.38, $0.56, $0.70, $0.90 or $1.20, depending on the symbol’s average spread
Single-leg, 101 contracts or more	$0.15, $0.19, $0.28, $0.35, $0.45 or $0.60
Multi-leg	$0.30, $0.38, $0.56, $0.70 or $0.90
Index options	No payment

For stocks, RHS reports payments set as a fixed percentage of the quoted spread: 12.35% of the spread during regular hours. On a $0.01 spread that is $0.001235 per share, or about $0.12 per 100 shares.

Adding up the dollar amounts and per-unit rates in the report’s tables for July, August and September 2025 (our calculation; per-unit rates are reported in cents per hundred shares, which for a standard 100-share option means per contract):

Q3 2025, RHS	PFOF disclosed	Units implied by the rates	Average rate
Options	$307.9 million	about 577 million contracts	$0.53 per contract
S&P 500 stocks	$12.1 million	about 17 million round lots	$0.71 per 100 shares
Other stocks	$81.1 million	about 494 million round lots	$0.16 per 100 shares
All stocks	$93.2 million	about 511 million round lots	$0.18 per 100 shares

Options were 76.8% of the total. Three caveats: Robinhood says the reports present some, “but not necessarily all,” of the PFOF received; RHS passes 80% of this revenue to its affiliate Robinhood Financial, so the same money shows up in two reports; and the per-unit averages are weighted across order types, so they are approximations.

Same dollars, very different payment

Thomas Ernst and Chester Spatt (“Payment for Order Flow and Asset Choice,” NBER Working Paper 29883, 2022) made a point that the per-unit numbers understate: retail investors put similar dollar amounts into options and stocks, but options are far cheaper per unit. From Rule 606 data, they estimated typical PFOF at about 40 cents per 100-share options trade and about 20 cents per 100-share equity trade, and calculated that the same nominal investment generates about ten times as much PFOF in options.

With the Q3 2025 Robinhood numbers:

$1,000 in a $50 stock buys 20 shares. At the S&P 500 average of $0.71 per 100 shares, the broker receives about $0.14. At the all-stock average of $0.18, about $0.04.
$1,000 in options priced at $2.00 buys 5 contracts ($200 each). At the $0.53 average, the broker receives about $2.67; at the $0.56 tier, $2.80.

That is roughly 19 times the S&P 500 stock figure for the same outlay. Ernst and Spatt argue that this creates a second conflict of interest beyond routing: when PFOF is the main source of broker revenue, the broker has a financial incentive to encourage options trading rather than stock trading. The SEC staff report cites a 2025 version of their work for the same tenfold comparison.

How concentrated it is

Bryzgalova, Pavlova and Sikorskaya, using Rule 606 reports from the largest retail brokers, found that options PFOF exceeded equity PFOF by about 100% in every month of their sample: $2.4 billion from options versus $1.3 billion from equities in 2021. The top three options wholesalers (Citadel, Susquehanna, and Wolverine at the time) received a share that rose from 73% to an average of about 85%, peaking near 90% in the second quarter of 2021. SEC staff repeated these figures in April 2026 and added that options now account for about two-thirds of all PFOF. In Robinhood’s Q3 2025 report, Citadel Securities and Dash/IMC together handled about 70% of RHS’s non-directed option orders.

SEC staff also found that the number of broker-dealers reporting revenue from options market making in their FOCUS filings fell steadily from 2015 to 2021 before stabilizing, with the decline concentrated in firms with smaller net capital.

What the evidence says about execution quality

The case that customers pay

Bryzgalova, Pavlova and Sikorskaya (2023) built a measure of retail options trading from trades flagged as single-leg price improvement auctions in OPRA data from November 2019. They found that retail traders favor cheap, short-dated options: half of the retail trades in their sample were in options with less than a week to expiry, with an average quoted bid-ask spread of 12.6% and an effective spread of 6.6%. Assuming a 10-day holding period, they estimated that the aggregate retail portfolio lost $2.1 billion from November 2019 to June 2021, and that indirect trading costs (trade price versus midquote) came to $6.4 billion against about $900 million of direct costs. This is the core evidence summarized in why most options buyers lose.
Ernst and Spatt (2022) exploited changes in which market maker was the designated market maker for a given option at each exchange. They found that retail traders received less price improvement, and worse prices, from DMMs that paid PFOF to brokers. They also found that in stocks, PFOF was small and internalized retail orders received meaningful sub-penny price improvement, so the PFOF problem looks worse in options than in equities.
SEC staff (2000). The April 2026 report quotes the Commission staff’s 2000 special study on PFOF and internalization in the options markets: because specialists make the payments and are compensated through effective spreads, the growth of PFOF “could be expected, all other things being equal, to be accompanied by a widening of effective spreads.”

The case that it is more complicated

Market-wide spreads. The April 2026 SEC staff data show the median effective spread for equity options falling from 2.2% of the trade price in 2012 to 1.9% at the end of 2025, and for ETP options from 1.9% to 1.4%. Quoted spreads, however, widened across all but the 10 most liquid underliers, which staff call a sign of “potential deterioration in displayed market quality.”
Auctions do improve prices. In CAT data for December 2025, staff found that marketable individual-customer orders executed in single-leg auctions received price improvement 22% of the time in the tightest spread bin and 83% to 89% of the time in the widest bins. Orders executed electronically on the book received almost none (0% to 13%). The effective spread for marketable auction orders topped out at $2.98 in the widest bin, against $6.68 for electronic executions.
Disputed studies. A 2024 study (Huang and coauthors, “Some Anonymous Options Trades Are More Equal than Others”) argues that higher per-contract PFOF received by retail brokers may go with less volume-weighted price improvement. SEC staff note that responding studies, including a critique by Hosty and Billings and a June 2025 staff report by the Committee on Capital Markets Regulation, argue that its methods overstated costs and excluded much retail flow. Staff conclude that research on options PFOF and execution quality “is currently relatively limited.”
Limit orders change the picture. Lei Fu, Su Li, David Musto and Neil Pearson (SEC Division of Economic and Risk Analysis working paper, March 2025) found that customers trading 0DTE S&P 500 options with non-marketable limit orders were at the best bid or offer more than half the time in slightly out-of-the-money series, and executed at low cost. They also identified a sequencing problem in OPRA data that can make trades look worse than they were if quotes are not re-ordered.

The fair summary is that options PFOF clearly shapes routing and creates conflicts of interest, while the size of the cost to customers depends heavily on order type, venue mechanism, and measurement choices.

Complex orders

Spreads, straddles and other multi-leg strategies are a large part of retail options activity. In SEC staff’s December 2025 sample, electronic single-leg executions were the most common sale condition but only 39.7% of individual-customer dollar volume; the second and fourth most common conditions were multi-leg or complex orders, and single-leg auctions were third at about 20%.

Exchanges run complex order books in which a spread trades as one package at a net price, so the customer avoids “legging risk” (filling one leg and then seeing the market move before the other fills). Complex orders have their own priority rules, can trade against the individual leg markets, and can be exposed in complex auctions. They also carry PFOF: Robinhood’s schedule pays $0.30 to $0.90 per contract on multi-leg orders, so a four-leg iron condor of one contract per leg can generate several dollars of payment.

The regulatory picture as of November 2026

The December 2022 package

On December 14, 2022, the SEC proposed four equity market structure rules. Two are relevant here:

The Order Competition Rule (proposed Rule 615 of Regulation NMS, Release No. 34-96495) would have required many individual-investor orders to be exposed in open auctions before a wholesaler could internalize them. It covered NMS stocks only. The release itself contrasted the two markets: stocks are order-driven, while “the listed options market is a quote-driven market in which professional market makers dominate liquidity provision,” and the Commission said it drew on its experience with existing options auctions in designing the rule.
Regulation Best Execution (Release No. 34-96496) would have created a Commission-level best-execution rule covering all securities, options included, with extra documentation and review requirements for “conflicted transactions,” including those involving PFOF.

On June 12, 2025, the Commission withdrew both proposals, along with 12 other pending rulemakings. Any future attempt would require a new proposal and a new comment period.

A third piece of the package survived. Amendments to Rule 605, which requires monthly execution-quality statistics, were adopted in 2024, and the SEC later moved the compliance date to August 1, 2026 (Release No. 34-104147). Rule 605 covers NMS stocks, not listed options, so it does not produce options execution statistics.

The 2026 options roundtable

The SEC’s April 16, 2026 Roundtable on Options Market Structure, and the staff data report prepared for it, signal that options are on the agenda. The report says the core options regulatory framework has “remained largely unchanged” since the Commission’s 2004 concept release on competitive developments in the options markets, and frames its purpose as supporting discussion of whether current rules fit today’s market. Separately, a June 2026 proposal (Release No. 34-105655) would rescind the trade-through rule and the locked and crossed quotation provisions for NMS stocks; options have their own order-protection regime under the Options Order Protection and Locked/Crossed Market Plan, which that proposal does not rescind. As of this writing we have not found a Commission rule proposal specific to options PFOF.

Outside the US

An SEC staff working paper (Boulton, Shohfi and Walz, DERA, January 2025) notes that the European Union agreed in June 2023 to a general ban on PFOF, and lists Australia, Canada, Singapore and the UK as jurisdictions that have acted to curb it. Canadian investors trading US-listed options through a Canadian dealer are therefore in a different regulatory setting from US retail customers, a point covered in how options markets are regulated.

What to look at as an informed customer

None of this tells you which broker to use, but the public data let you see how your orders are handled:

Read the broker’s Rule 606 report. It shows which consolidators receive your options orders and what they pay per contract by order type.
Know the spread you are crossing. On a $0.80 bid and $1.00 ask, a market order to buy one contract at the ask costs $10 more than a fill at the midpoint ($0.10 x 100). That is about 19 times a $0.53 PFOF payment, which is why execution quality matters more than the payment itself.
Order type matters. The SEC staff and DERA evidence both show limit orders and auction executions achieving far better prices than marketable orders filled on the electronic book.
Wide, cheap, short-dated options are where costs concentrate. The 12.6% average quoted spread in Bryzgalova, Pavlova and Sikorskaya’s short-dated sample is the clearest illustration.

How market makers price and hedge the flow they buy is covered in how options market makers hedge.

The short version

Retail option orders in the US are mostly bought by a few large market makers, who pay brokers per contract and route each order to an exchange auction or book where they are likely to trade with it. Per contract and in total, options PFOF dwarfs equity PFOF: in Robinhood Securities’ Q3 2025 report, options were about 77% of disclosed payments at about $0.53 per contract. Academic work finds that retail options traders face wide spreads and lose money on average, and that PFOF-paying market makers give less price improvement, while SEC staff data show auctions do deliver real price improvement and other studies dispute how large the costs are. The SEC’s 2022 proposals to force auctions for stock orders and to write a best-execution rule were withdrawn in June 2025, and as of November 2026 options market structure is under review rather than under new rules.