Is Commission-Free Trading Actually Free? what congress and 85,000 Trades Reveal About PFOF

A close-up, angled shot of a printed financial chart showing market trends and technical indicators, including a "14, SMA" (Simple Moving Average) label and price points. The image represents the complexities of stock execution and payment for order flow (PFOF) discussed in the article "Is Commission-Free Trading Actually Free: What Congress and 85,000 Trades Reveal About PFOF."
Stocks & Forex

Is Commission-Free Trading Actually Free? what congress and 85,000 Trades Reveal About PFOF

September 23, 2026

Is Commission-Free Trading Actually Free? What Congress and 85,000 Trades Reveal About PFOF

Five researchers ran 85,000 real trades through five different brokers at the exact same moment, just to see who got the better price. The gap they found was real. It just wasn’t where the “you’re being scammed” headlines told them to look.

No, “commission-free” trading isn’t strictly free. Brokers are often paid by market makers to route your order there, a practice called payment for order flow. But the popular story that this is secretly draining your account doesn’t hold up against the research, which found that which broker you use moves your execution price more than whether that broker takes the payment at all.

That finding is the whole reason this guide exists, and it’s laid out in full under what the research actually shows. First, the two things that make it make sense: what a trade used to cost, and what “routing an order” actually means.

The quick answer

If you’ve got thirty seconds, here’s everything that matters:

  • Payment for order flow (PFOF) is legal in the U.S. and disclosed under SEC rules. Brokers publish quarterly reports showing where they send orders and what they’re paid.
  • A peer-reviewed study of 85,000 simultaneous market orders found large differences in execution quality between brokers, but PFOF levels didn’t explain those differences.
  • Visible commissions fell from about $200 per trade in the 1980s and $40 in the 1990s to zero. Congress’s own researchers say the hidden costs being debated are generally not believed to be that large.
  • The EU’s PFOF ban became fully applicable in every member state on June 30, 2026. The U.S. relies on disclosure and a best-execution duty instead.

Where you are right now

1980s About $200 per trade Transparent cost

Printed on your confirmation. You knew exactly what you paid.

1990s About $40 per trade Transparent cost

Discount brokers cut fees. PFOF begins in the background.

The real history: what trading used to cost

Most arguments about payment for order flow start in the present, which makes $0 look like the baseline and any hidden cost look like a loss. The more useful starting point is what a trade actually cost before “free” existed.

According to a 2024 Congressional Research Service (CRS) brief prepared for members of Congress, retail commissions ran roughly $200 per trade in the 1980s, fell to around $40 in the 1990s, when PFOF first appeared, and reached zero at major brokers in the 2010s. For someone making 50 round-trip trades a year, the 1990s price alone meant about $4,000 in commissions before a single share moved in their favor.

That’s the Transparent cost era. You didn’t have to trust anyone about it; the number was on your confirmation slip.

Two details from that history tend to get distorted. First, PFOF was pioneered in the 1990s by Bernard Madoff, back when he had a strong industry reputation and long before his unrelated fraud conviction. CRS notes there was no indication his use of PFOF was illegal or unethical, though the association may have hurt the practice’s reputation. Scandal-style coverage often leans on the Madoff name; the practice and the fraud are separate things. Second, zero commissions didn’t arrive because brokers became charitable. Revenue from PFOF is one of the things that made them financially possible, which is exactly why it deserves scrutiny rather than either blind trust or outrage.

EraVisible commissionHidden cost (PFOF)
1980sAbout $200 per tradeNot applicable; PFOF not yet in use
1990sAbout $40 per tradePFOF begins, alongside commissions
2010s to today$0 at major U.S. brokersPossible, debated, and not reliably measured; generally believed to be smaller than past commissions
Then vs. now: the real cost of a trade. Source: Congressional Research Service, “Payment for Order Flow (PFOF) and Broker-Dealer Regulation,” IF12594, February 20, 2024.

How payment for order flow actually works

Here is the whole thing, without adjectives.

When you tap “buy” on 10 shares, your broker has to send that order somewhere to be filled. It can go to a public exchange like the NYSE or Nasdaq, or to a wholesaler, also called a market maker, which fills the order itself from its own inventory. That in-house fill is called internalization.

The wholesaler earns money on the spread, the small gap between the price it buys at and the price it sells at. Retail orders are attractive to wholesalers because they tend to be small and not driven by information that will move the price moments later. So wholesalers pay brokers for the right to fill them. That payment is PFOF. CRS describes it as generally a fraction of a cent per share, which adds up because of volume: the 12 largest U.S. brokerages earned a combined $3.8 billion from it in 2021.

The market is concentrated. In 2022, brokers sent more than 90% of marketable retail stock orders to just six wholesalers, and the top three, Citadel Securities (41%), Virtu Financial (26%), and G1 Execution Services (16%), handled about 83% between them, according to CRS.

The guardrails that already exist

A broker taking PFOF isn’t free to send your order wherever pays best. Three sets of rules apply:

  • Best execution. FINRA Rule 5310 requires brokers to seek the most favorable terms reasonably available for customer orders and to regularly compare venues. Price, speed, and the chance of price improvement all count. Marketable orders routed to wholesalers must be filled at the national best bid or offer or better.
  • Rule 606 (routing and payments). Brokers publish quarterly reports showing which venues received their customers’ orders and how much net payment they received, broken out by order type, for stocks and for options.
  • Rule 605 (execution quality). Market centers publish monthly statistics on how well orders were actually filled. Under 2024 amendments, larger brokers with 100,000 or more customer accounts must also report, as of the compliance date of August 1, 2026, which has now passed.

A fourth rule, Rule 607, requires brokers to describe their PFOF arrangements when you open an account and annually after that. It’s the fine print most people scroll past.

What the research shows, not what the headlines say

Policymakers have been debating two questions: does PFOF create a Hidden cost by getting retail trades filled at slightly worse prices, and if so, how big is it? Here’s what the main lines of evidence say, grouped by which way they lean.

The 85,000-trade study

The most direct test comes from five finance professors, Christopher Schwarz, Brad Barber, Xing Huang, Philippe Jorion, and Terrance Odean, who opened six brokerage accounts across five brokers and placed the same market orders in all of them at the same moment. Their paper, “The ‘Actual Retail Price’ of Equity Trades,” was published in The Journal of Finance in 2025.

CRS summarized the working-paper version as concluding that PFOF does not appear to harm price execution. That’s accurate, but the full findings are more interesting, and they cut in more than one direction:

  • Brokers differed a lot. The average round-trip cost, not counting commissions, ranged from about 0.07% of the trade’s value at the best account to 0.46% at the worst.
  • PFOF didn’t explain the differences. Brokers receiving more PFOF weren’t systematically the ones with worse prices.
  • The wholesalers did. The same wholesalers gave different prices to different brokers for identical trades.
  • The disclosures missed it. The authors reported that these differences couldn’t be predicted from the regulatory reports available at the time, and they estimated that a single basis point of cost across all retail trading is worth about $2 billion a year.

So the study doesn’t say “PFOF is harmless” and it doesn’t say “PFOF is a scam.” It says execution quality is real, varies a lot, and wasn’t driven by the variable everyone argues about.

The rest of the evidence, side by side

  • Leans toward “retail investors came out ahead” One study cited by CRS found the shift to PFOF-funded zero commissions lowered retail investors’ overall trading costs. Another found zero commissions improved overall market quality.
  • Leans toward “there’s a real cost” That same market-quality research found retail investors received less price improvement per share. Separately, after the United Kingdom effectively banned PFOF in 2012, one study found that the share of retail trades executed at the best quoted prices rose sharply.
  • The regulator’s estimate, and its critics When the SEC proposed a rule in late 2022 requiring many retail orders to be auctioned before a wholesaler could fill them, its economic analysis estimated retail investors could save about $1.5 billion a year. Multiple commentators challenged that estimate, and the SEC formally withdrew the proposal in June 2025 without adopting it.

The Net effect

On the one point the evidence does line up, CRS is direct: even under the larger estimates of possible hidden costs, overall retail trading costs have clearly fallen, because hidden execution costs are generally not believed to be anywhere near the $40 per trade that commissions used to cost.

That’s the honest conclusion. PFOF is legal, disclosed, and debated among economists. It may cost some traders something, the size is contested, and even the larger estimates leave most retail investors better off than they were in the commission era. What the 85,000-trade study adds is a warning against the lazy version of either story: the thing most worth checking is your broker’s actual execution quality, not just whether it takes PFOF.

The Robinhood settlement, precisely

On December 17, 2020, the SEC charged Robinhood Financial LLC with two things: repeatedly failing to disclose, in customer communications including its website FAQ, that payments from trading firms were its largest revenue source, and failing to meet its duty to seek the best reasonably available terms for customer orders. The revenue-disclosure conduct ran from 2015 to late 2018. Separately, from October 2018 to June 2019, Robinhood’s website claimed its execution quality matched or beat competitors, statements the SEC found false and misleading.

Robinhood agreed to pay a $65 million civil penalty and to retain an outside consultant to review its best-execution practices. It settled without admitting or denying the findings. The SEC’s order found that inferior execution prices had cost customers about $34.1 million in aggregate, even after accounting for the commissions they didn’t pay. CRS notes an analysis cited in the SEC’s order estimating the execution loss at about $5 per order for orders over 100 shares and about $15 for orders over 500 shares, during that period.

Two things are true at once. The case shows the risk critics describe is real: a broker can let PFOF arrangements and execution quality drift against customers. It also shows that existing U.S. rules gave the SEC the tools to act. Robinhood’s chief legal officer said at the time that the settlement related to historical practices that did not reflect the company as it then operated. The case is a documented piece of history, not evidence about any broker’s current execution.

On revenue, CRS reports that Robinhood collected $974 million in PFOF in 2021, about half its revenue that year, and that about 75% of its revenue in 2020 came from PFOF. Those figures are dated; for current numbers, the primary source is the company’s own SEC filings.

Does this affect options traders more?

It can, and the reason is structural rather than sinister.

Options spreads are generally wider than stock spreads, and there are thousands of contracts with different strikes and expirations, many of them thinly traded. Wider spreads mean more room for market makers to earn money, so PFOF on options is generally paid per contract and at higher rates than PFOF per share on stocks. That also means brokers with large options businesses tend to earn a larger share of their PFOF from options. Options exchanges have their own payment arrangements as well, so the routing picture is more layered than it is for stocks.

There’s a disclosure gap worth knowing about. Rule 606 routing reports do cover options, including what the broker was paid. But Rule 605, the execution-quality report, applies only to stocks; the SEC confirmed as recently as April 2026 that it doesn’t apply to listed options. So an options trader can see where orders went and what the broker earned, but has much less standardized public data on how well those orders were filled.

If you trade options actively, the practical takeaway is modest: the spread you cross on each trade matters, limit orders give you more control than market orders on wide-spread contracts, and your broker’s Rule 606 report will show how its options routing and payments compare with its stock routing.

The 2026 divide: U.S. disclosure vs. EU prohibition

Two of the world’s largest retail markets have now landed in opposite places, and both positions are defensible on their own logic.

The European Union prohibits it. Article 39a of MiFIR, added by Regulation (EU) 2024/791, bars investment firms acting for retail clients from receiving any fee, commission, or non-monetary benefit for routing client orders to a particular venue. It entered into force on March 28, 2024, and applied immediately in most member states. A transitional clause let countries temporarily exempt existing domestic practice; only Germany used it. That exemption expired, and the ban has applied across all 27 member states since June 30, 2026. The EU’s reasoning is that the payment itself creates a conflict of interest, whether or not a given trade was executed well.

The United States permits it, with disclosure. The SEC proposed an order-competition auction rule and a new Regulation Best Execution in December 2022. On June 12, 2025, it formally withdrew both, saying any future action in those areas would require a new proposal. What remains is the framework described above: FINRA’s best-execution rule, Rule 606 routing disclosure, and the expanded Rule 605 execution reports whose compliance date arrived on August 1, 2026.

Elsewhere. CRS lists the United Kingdom, Australia, and Canada as jurisdictions that have banned PFOF. The UK’s effective ban dates to 2012, and Canada does not permit it for orders in Canadian-listed securities.

ApproachWhat it requiresCurrent status (September 2026)
United States: disclosure and best execution PFOF permitted. Brokers must seek best execution (FINRA Rule 5310), publish quarterly routing and payment reports (Rule 606), describe PFOF terms to customers (Rule 607), and, for larger brokers, publish monthly execution-quality reports (Rule 605). In effect. 2022 auction and best-execution proposals withdrawn June 12, 2025. Amended Rule 605 reporting took effect on its compliance date of August 1, 2026.
European Union: prohibition Firms acting for retail and opt-in professional clients may not receive third-party payments for routing or executing client orders at a particular venue (MiFIR Article 39a). In force since March 28, 2024. Germany’s transitional exemption, the only one used, ended; the ban applies EU-wide from June 30, 2026.
U.S. disclosure approach vs. EU prohibition. Sources: CRS IF12594 (Feb. 2024); SEC withdrawal notice, Release 33-11377 (June 2025); Federal Register, 90 FR 47552 (Oct. 2, 2025); Regulation (EU) 2024/791. Status verified September 23, 2026.

What a ban doesn’t do is make trading free. EU brokers still need revenue, and they are replacing PFOF with spreads, commissions, subscriptions, or their own trading venues. Whether European retail investors end up paying more or less overall is an open empirical question, and it’s too early for good data.

United States: let it happen, make it visible

  • Theory: competition plus disclosure plus a best-execution duty can manage the conflict.
  • Tool: quarterly routing reports and monthly execution statistics.
  • Trade-off: zero commissions are easier to sustain, but you have to look to see the cost.

European Union: remove the payment entirely

  • Theory: the incentive itself is the conflict, regardless of outcomes.
  • Tool: a flat prohibition under MiFIR Article 39a.
  • Trade-off: the conflict is gone, but brokers must charge in other ways, some of which may be just as hard to see.

What this isn’t

This guide is about how “free” trades get paid for. It isn’t a broker ranking, and it isn’t about trading rules or account types. If you’re building a fuller picture of how retail trading actually works, two related reads cover other parts of it. For how the old day-trading minimum changed, see The $25,000 Day Trading Rule Is Gone. For what “funded” prop-firm accounts really are, see Your “Funded Trading Account” Probably Isn’t Real Money.

FAQ

Is commission-free trading a marketing myth?

Partly. You don’t pay a visible commission, but at brokers that accept PFOF, market makers pay the broker to handle your orders. The cost moved from a line on your statement into the execution price, where it’s harder to see and its size is genuinely debated. “Free” is better read as “paid differently.”

What is payment for order flow, explained simply?

It’s a small payment a market maker gives your broker for the right to fill your order. The market maker earns money on the gap between buying and selling prices and shares a slice with the broker. It’s typically a fraction of a cent per share for stocks.

How do free trading apps make money?

PFOF is one source. Others commonly include interest earned on customers’ uninvested cash, margin lending, securities lending, and paid subscription tiers. The mix differs by broker and is described in each public broker’s SEC filings.

Does PFOF actually hurt retail traders?

The research hasn’t settled it. Some studies find retail investors’ total costs fell under PFOF-funded zero commissions; others find less price improvement per share or better prices after the UK’s ban. The 85,000-order study found big execution differences between brokers that PFOF didn’t explain.

How much does the average retail trader lose or save from PFOF?

No reliable single figure exists. As an illustration only, the 85,000-trade study’s range of roughly 0.07% to 0.46% round-trip cost works out to about $3.50 to $23 on a $5,000 round trip, compared with about $80 in commissions for the same round trip at 1990s prices. Costs for your own orders depend on the broker, stock, size, and timing, and the study’s results came from its own sample of trades.

Is PFOF legal in the United States?

Yes, as of September 2026. It’s subject to FINRA’s best-execution rule and SEC disclosure requirements under Rules 605, 606, and 607. The SEC’s 2022 proposals that would have reshaped it were withdrawn in June 2025.

Why did the EU ban PFOF while the U.S. kept it legal?

The EU treats the payment itself as a conflict of interest and prohibits it under MiFIR Article 39a, now applicable in all member states. U.S. regulators have relied on disclosure, competition among venues, and the broker’s duty of best execution to manage that conflict instead.

What’s the difference between SEC Rule 605 and Rule 606?

Rule 606 tells you where a broker sends orders and what it’s paid for them, published quarterly. Rule 605 tells you how well orders were actually filled, such as price improvement and speed, published monthly. Rule 606 covers stocks and options; Rule 605 covers stocks only.

How do I read a Rule 606 report?

Look for your broker’s quarterly report, usually linked in a website footer or disclosures page. It lists venues by the percentage of orders each received, split into market orders, marketable limit orders, non-marketable limit orders, and others, with the net payment received for each. Check the S&P 500 stocks, other stocks, and options sections separately.

Was Robinhood fined for PFOF?

Robinhood Financial paid $65 million in December 2020 to settle SEC charges that it misled customers about PFOF being its largest revenue source (2015 to late 2018) and made false claims about its execution quality (October 2018 to June 2019), while failing to meet its best-execution duty. It neither admitted nor denied the findings. PFOF itself was not ruled illegal.

Does PFOF affect options trades differently than stock trades?

Generally yes. Options PFOF is paid per contract and typically at higher rates, reflecting wider spreads. Options also lack the standardized Rule 605 execution statistics that stocks now have, though Rule 606 still discloses options routing and payments.

Is a broker that doesn’t take PFOF automatically better?

Not automatically. The 85,000-trade study found that PFOF levels didn’t explain which accounts got better prices. A broker without PFOF may charge in other ways or still route to wholesalers. Execution quality is something to check, not something to assume from a single policy.

Sources

This article is for educational purposes only and is not investment, legal, or tax advice. It does not recommend any broker, and nothing here should be read as a judgment on any firm’s current execution quality. Payment for order flow rules and disclosure requirements change over time; verify current rules at sec.gov and review a specific broker’s own Rule 605 and 606 reports before drawing conclusions about that broker.

Leave Comment

Your email address will not be published. Required fields are marked *

Reach the Editor
AdvoraHQ

AdvoraHQ Editorial

Online

Welcome to AdvoraHQ. We decode complex financial concepts—from tax strategies to market investing—using strictly primary sources and deep research.

Got a specific question, a topic request, or feedback on our research? We'd love to hear from you.

Email the Editor