The SEC charged Robinhood Financial with misleading customers about its largest revenue source and with failing to satisfy its duty of best execution.
Between 2015 and late 2018, its customer communications, including the FAQ pages on its own website, omitted that it was paid by trading firms for routing customer orders to them.
One of the firm's central selling points was that trading was commission free. The order found that because its payment-for-order-flow rates were unusually high, customer orders were executed at prices inferior to those available from other brokers.
The harm the SEC calculated: $34.1M, and this is the number that matters, even after accounting for what customers saved by not paying commissions.
The firm also stated, between October 2018 and June 2019, that its execution quality matched or beat competitors. It paid $65M to settle, without admitting or denying the findings.
Every customer could see the commission. It was zero, prominently.
No customer could see the execution price they would have received somewhere else.
The cost did not disappear when the disclosed component went to zero. It moved into the component nobody is shown, and it was larger there.
Ask someone what a trade costs and they will quote you the trading fee, because it is the number on the screen. It is usually not the largest component and sometimes it is not even close.
This lesson counts all of it, and it is the lesson behind this course's lab.
Seven components of a round trip
1. The deposit. Card purchases carry the worst rates by a wide margin. Bank transfers are usually free or close to it and slower.
2. The spread. F107-02's book. You buy at the ask and sell at the bid, so a round trip pays the gap twice. On a venue that quotes you a single price with no visible book, the spread is where the cost lives and you cannot see it.
3. The trading fee. The disclosed number. Often the smallest item on this list.
4. Slippage. F107-02's arithmetic, and it scales with your size against the book rather than with the amount you are trading in absolute terms.
5. The withdrawal fee. Charged by the venue to move an asset out, frequently a flat amount, which makes it a punitive percentage on small withdrawals.
6. The network fee. Paid to the chain. F101-05's arithmetic: it is priced by computation and does not scale with your amount, so it is trivial on a large transfer and dominant on a small one.
7. The off-ramp. Selling back to money you can spend, which pays 1, 2, 3 and 5 again in the other direction, plus whatever the fiat conversion costs. This is the one people forget entirely when they estimate their costs, and it is often the largest single item.
Illustrative rates to demonstrate the method. Never read these as current pricing.
You buy $1,000 of an asset, move it to self-custody, move it back, and sell.
On the way in:
Card deposit fee at 2.5 percent: $1,000 x 0.025 = $25.00 Remaining to trade: $975.00 Spread, half of a 0.20 percent quoted spread: 975 x 0.001 = $0.98 Trading fee at 0.40 percent: 975 x 0.004 = $3.90 Withdrawal fee, flat: $5.00 Network fee to receive: $2.00
Cost in: 25.00 + 0.98 + 3.90 + 5.00 + 2.00 = $36.88
On the way out, on the remaining $963.12 of value:
Network fee to send back: $2.00 Spread again: 961 x 0.001 = $0.96 Trading fee at 0.40 percent: 961 x 0.004 = $3.84 Fiat withdrawal fee: $5.00
Cost out: 2.00 + 0.96 + 3.84 + 5.00 = $11.80
All-in round trip:
36.88 + 11.80 = $48.68, which is 4.87 percent of the original $1,000
Now the number that should change your behaviour. The break-even move:
The asset must rise 4.87 percent before you are level. Not before you profit. Before you are back where you started.
Two variations worth running yourself.
Replace the card with a bank transfer, taking the deposit fee to zero:
Cost falls to about $23.68, or 2.37 percent. One decision halved the total cost of the entire exercise.
Do the same round trip with $100 instead of $1,000. The flat items, $5 withdrawal, $2 network, $5 fiat withdrawal, $2 network again, total $14 and do not shrink. The percentage components scale down. All-in lands near 17 percent.
That is the real finding, and it is why the lab exists. Costs that are flat are percentages in disguise, and the percentage is set by your size. Learners doing F107-L with $20 routinely find 3 to 8 percent, and sometimes far more.
Where the cost hides when the fee is zero
The autopsy generalises, and the generalisation is the most useful thing in this lesson.
A business does not stop needing revenue when it stops charging a fee. So when a venue advertises zero commission, the correct response is not gratitude, it is a question: then where?
The usual answers:
- A wider spread. The single most common, and invisible on any interface that quotes one price rather than showing a book.
- Order routing. The autopsy. You are the product being sold to whoever fills your order.
- The conversion rate. Especially on fiat pairs and especially on a card purchase, where the exchange rate applied is a cost with no label.
- The withdrawal fee. Frequently set well above the actual network cost.
None of these are hidden in the sense of being concealed. They are hidden in the sense that no interface will ever compute them for you, which is functionally the same.
Two venues, one charging 0.40 percent and one charging nothing, are not comparable on that number, and choosing on it is exactly the error the SEC order describes.
The comparison that works is boring and takes ten minutes: move the same small amount through both, end to end, and measure what you get back. That number contains every component including the ones nobody discloses.
F107-L is this exercise, and it is a computational lab rather than a reading exercise for a reason. The figure you personally compute is worth more than any fee schedule, because the fee schedule is the part that was already visible.
What you can actually reduce
Deposit method. The largest single lever for most people. Bank transfer instead of card.
Order type. F107-02. A limit order on a liquid pair removes slippage and can capture the spread rather than paying it.
Frequency. Every component is charged per round trip. Halving your number of round trips halves this entire lesson, and it is the only cost reduction that also reduces the F106 risk surface.
Batching withdrawals. Flat fees make ten small withdrawals ten times the cost of one.
Chain and timing. Network fees vary by orders of magnitude between chains and by time of day on congested ones.
What you cannot reduce: the spread on the fiat off-ramp, most of the time, and the fact that the whole round trip is denominated in a currency you have to get back to.
Fees are small enough that they do not matter for a long-term holder.
Broadly right, and the reasoning behind it is wrong in a way that misleads people at the start.
For someone buying once and holding for years, a 5 percent round trip amortises across a long period and is genuinely a minor consideration next to what the asset does.
Two corrections.
It is a threshold, not a drag. The round trip is paid in full at the beginning, so the position starts underwater by that amount. A 5 percent all-in cost means the first 5 percent of any gain is not yours, and F102-05's warning applies: this is arithmetic, not a claim that any gain will occur.
Beginners are not long-term holders yet. People learning tend to make many small transactions, and the worked example shows small size is exactly where flat fees dominate: 4.87 percent at $1,000 and around 17 percent at $100. Somebody experimenting with $50 several times over a month can pay a quarter of their capital in costs while believing they are testing a strategy.
So the honest framing is that fees matter enormously at the start, when amounts are small and activity is high, and matter little later. That is the reverse of when most people pay attention to them, and it is why this lesson is in the Freshman level.
Count seven components, not one: deposit, spread, trading fee, slippage, withdrawal fee, network fee, and the off-ramp on the way out. The disclosed fee is often the smallest, and a venue charging nothing has moved the cost into the spread, the routing or the conversion rate, which is what the SEC found cost customers $34.1M net of the commissions they saved. Compute your own all-in round trip as a percentage and read it as a break-even move, and notice that flat fees are percentages in disguise: about 5 percent on $1,000 and around 17 percent on $100. Then reduce what you can, starting with the deposit method and the number of round trips.