The Cost You Do Not See on a Zero-Commission Trade
This series examines execution quality in U.S. retail equity orders: what investors ask for, how brokers and trading venues handle those orders, and what customers actually receive. This opening chapter focuses on a deceptively simple question: if a broker charges no commission, what else determines whether a trade was good? The answer is not a hidden fee with one universal name. It is the full outcome of the order—its price, completion, timing, and the consequences of what happened to any shares that did not execute.
A useful way to read the evidence is to treat execution quality as a trade-off between immediacy and price control. A marketable order usually offers speed and a high probability of completion, but the investor accepts the available market price and spread. A nonmarketable limit order gives the investor more control over the maximum or minimum acceptable price, but the trade may wait, fill only partly, or never fill. The right comparison therefore cannot begin and end with the commission line.
A zero commission can still leave an economic bill
The most visible cost of a retail trade is the one that appears on the confirmation: a commission, if there is one. But a broker can charge zero commission while an order still has an economic cost. An immediately executable order may pay the bid-ask spread—the gap between the best displayed price to buy and the best displayed price to sell—or receive less price improvement than another order could have received. A limit order may avoid paying that spread on the shares it fills, yet impose a different cost if the market moves away before the order is completed.
Consider two investors who each submit an order for the same stock through a commission-free account. One needs the position now and accepts an immediately executable order. The other is willing to wait and posts a limit price. The first may receive a nearly complete trade quickly but surrender some price control. The second may obtain a better price on filled shares, but may end the day with fewer shares than intended. If the stock then moves away, the second investor faces a choice: abandon the trade, accept the changed price later, or cross the spread to finish it. Their confirmations may show the same commission—zero—while their economic outcomes differ in several dimensions.
That is why “cost” needs a wider definition. For a retail order, at least five questions matter:
- What price was obtained? Compare the execution price with a relevant benchmark, such as the midpoint of the best bid and offer when the broker received the order.
- How much of the intended trade was completed? A favorable price on a small filled portion does not answer whether the investor received the position they sought.
- What did the unfilled portion cost? A missed or delayed trade can have an opportunity cost when the market moves away from the limit price.
- How long did execution take? Speed matters differently for an urgent marketable order and a patient limit order, but it matters to both.
- What happened around the execution? The market’s movement after a trade can reveal whether the liquidity provider faced adverse price movement, although that measure is not the same as the provider’s total profit or the investor’s eventual return.
These questions describe an order lifecycle, not a broker slogan. The broker, market maker, exchange, or other venue is part of the path, but the label of the institution is not itself the customer outcome. A route to a market maker does not tell us whether the order filled at a favorable price; a route to an exchange does not, by itself, prove the opposite. The relevant evidence is the relationship between the order’s intention, handling, execution, and result.
The practical implication is modest but important: a fee schedule is a starting point for analysis, not its conclusion. Explicit charges are easy to see. Implicit costs require following what happened to the order.
The retail order—not the fee schedule—is the unit of analysis
The need for this broader view is not theoretical. The study Retail Limit Orders examines more than 27 million retail “top orders” handled by 19 active U.S. retail brokers, using FINRA Order Audit Trail System data from May 2020. The sample covers 300 exchange-listed stocks grouped by market capitalization. OATS contains the order’s receipt, routes, executions, cancellations, prices, and associated timestamps, which lets the researchers examine the order’s life rather than only the trades that eventually appeared in a transaction database.
That scope matters because nonmarketable limit orders are a meaningful part of retail activity. In the study’s sample, they account for about 25.5% of orders and 30% of submitted shares, although they account for 18.5% of executed shares. The difference is not a contradiction. A limit order can contribute to submitted volume without producing an execution, precisely because non-execution is part of the choice the investor made. Looking only at executed trades would systematically remove part of the decision’s consequence.
The study also makes a terminology distinction that is easy to miss. It uses “marketable orders” for market orders and marketable limit orders—orders executable when the broker receives them. It uses “limit orders” for nonmarketable limit orders. That distinction is essential when comparing outcomes. A limit price does not automatically mean patience, and a marketable limit order is economically closer to a market order at the moment of arrival than to a resting order waiting for the market to reach it.
The order itself is therefore the right unit of analysis because it connects intention to consequence. A customer may have wanted a complete position, a maximum acceptable price, an immediate exit, or merely an opportunity to buy if the market reached a chosen level. Public execution statistics cannot always identify that intention perfectly. But any serious comparison should at least ask whether the metric is evaluating price, completion, or the cost of not completing the order.
This is also why a single headline number can mislead. A fill rate measures how much was executed relative to the submitted quantity; it does not say whether the price was good. An effective spread measures the price paid or received relative to a benchmark for executed shares; it does not fully charge the unfilled shares for a market move. A speed statistic measures elapsed time for executions; it does not tell us whether waiting was an intentional way to avoid paying immediacy. A post-trade realized-spread measure can illuminate price movement after execution; it does not summarize the investor’s entire strategy.
The SEC’s Rule 605 framework is designed around this problem. The rule has required standardized monthly execution-quality reports from market centers since 2000, with the aim of making execution performance more comparable and encouraging competition on implicit as well as explicit costs. The SEC’s 2024 final amendments expand the reporting system to larger customer-facing brokers—those that introduce or carry at least 100,000 customer accounts—and add a summary report alongside more detailed data. The changes also update the treatment of order sizes, order types, nonmarketable limit orders, timing, and spread measures.
That expanded disclosure is valuable because retail execution has become more fragmented, automated, and varied than it was when the original framework was adopted. But it is not a universal score. The SEC itself warns that execution-quality reports do not, by themselves, establish whether a broker satisfied its legal duty of best execution. They also cannot erase differences in the securities, order sizes, customer populations, and order types handled by different firms. A summary can make information easier to read without making unlike order flow comparable.
What this series will keep separate
The central discipline is to keep several ideas apart even when they are related. Price asks whether filled shares executed favorably against a benchmark. Completion asks whether the intended quantity was filled. Opportunity cost asks what the unfilled quantity cost when the market moved. Speed asks how quickly executed shares were completed. Post-trade movement asks what happened after execution over a specified horizon. Handling context asks how order type, size, market conditions, broker, venue, and execution capacity shape the comparison.
The distinction between price and completion is especially important for limit orders. A limit order can receive a favorable benchmark comparison on its filled shares while still carrying an opportunity cost on shares that never filled. Conversely, a marketable order can receive price improvement while still being the wrong choice for an investor who could have waited for a better price. Neither observation is enough without the investor’s objective and the fate of the full order.
The evidence also needs a time label. The retail study is based on May 2020, and its sample is filtered to produce a coherent set of ordinary orders whose costs can be measured. It is broad evidence about the defined population, not a current broker-by-broker ranking or a guarantee that every retail order behaves the same way. The SEC release describes a final reporting framework and its implementation requirements; it does not, by itself, show what the amended reports will imply for a particular broker’s customers today.
Those boundaries do not weaken the subject. They make the subject more precise. The study provides a rare view of retail order lifecycles and finds that nonmarketable limit orders can perform well under the conditions it examines. The SEC framework supplies a public vocabulary for comparing execution quality across order types, sizes, speeds, and post-trade outcomes. Together, the sources support a better question than “Which broker has zero commissions?” They support the question “For the kind of order I am trying to place, what did the full lifecycle cost, what did it deliver, and how comparable is the evidence?”
The next step is to examine the first decision inside that question. A marketable order buys immediacy and a high probability of completion. A nonmarketable limit order buys price control while accepting uncertainty about whether and when the market will meet it. Understanding that exchange is the foundation for every later measure of retail execution quality.