Trading Costs Explained
Spreads, swaps, commissions, slippage, and last-look execution: the costs that sit between you and your trade, with worked examples and the math of what they add up to.
Most beginner trading material treats price direction as the whole game. It is not. Before direction matters at all, every trade has a price tag, and that tag compounds quietly across every position you ever open. A strategy that looks profitable on paper can be a money loser in practice simply because nobody priced the friction. This article lays out the four (often five) costs in retail forex trading, with worked examples and an honest note on the ones brokers prefer not to lead with.
The bid-ask spread: the first cost you ever pay
Every quote you see is really two prices. The bid is what your broker is willing to buy the base currency from you for; the ask is what they will sell it to you for. The gap between them is the spread, and you cross it the instant you open a position. (For a refresher on quotes, see How Currency Pairs Work.)
A typical EUR/USD quote might be 1.0849 / 1.0851. The bid is 1.0849, the
ask is 1.0851, and the spread is 0.0002, or 2 pips. If you buy at the
ask and could only sell back instantly at the bid, your position is already
worth 2 pips less than you paid. Direction has not entered the picture yet.
Translate that into money with pip value: on a single mini lot (10,000 units) of EUR/USD, one pip is worth $1, so a 2-pip spread is $2 paid every time you open a position. On a standard lot, it is $20. The spread sounds tiny because the unit (a pip) is tiny, but multiply it by lot size and trade frequency and it stops feeling tiny fast.
Spreads are not constant. They are tightest on heavily-traded major pairs during liquid hours, and they widen on exotic pairs, in quiet sessions, and around major news. A pair quoted at 0.8 pips during the London-New York overlap may be quoted at 3 pips fifteen seconds before a US inflation release. The broker is not gouging you in that moment; they are protecting themselves against the volatility they expect, and the spread is the price of that protection passed to you.
Swap and rollover: the cost of holding overnight
Every spot FX position carries an implicit short-term loan. You are borrowing one currency to hold another, and the two currencies pay different short-term interest rates. The daily settlement of those interest differences is called swap or rollover, and it is debited from or credited to your account at a fixed time each trading day (usually around 17:00 New York time).
The mechanics are simple enough. If you are long a currency that pays a higher overnight rate than the one you are short, you receive a small credit. If you are long the lower-yielding side, you pay. The size of the swap is small per night, but compounds: a position held for a month pays or receives roughly twenty days’ worth (weekends roll into one charge, usually Wednesday).
A couple of practical points beginners often miss:
- Wednesday is triple-swap night. FX settles spot-plus-two, so the Wednesday rollover covers Saturday and Sunday as well as Wednesday. The charge or credit is roughly three times the usual.
- “Swap-free” accounts. Many brokers offer a swap-free option, often marketed to traders whose religion forbids paying or receiving interest. These accounts typically replace swap with an equivalent fee under a different name, or with a wider spread. They are rarely free in any meaningful sense.
- Holding direction matters. A trade that has positive carry one week may have negative carry the next if a central bank changes course. Strategies that survive only because they pick up nightly carry are exposed to that policy risk in addition to the underlying price risk.
Commissions: the second cost (sometimes)
Brokers operate on two broad models, and the cost shows up differently under each.
Under a market-maker model, the broker is your counterparty: they quote you a price and take the other side. Their revenue is the spread itself, plus any positions they choose to internalise or hedge. The trade appears “commission-free” on your statement because the cost is already baked into the spread you paid.
Under an ECN or STP model, the broker passes your order through to a network of liquidity providers and takes a flat per-lot commission. The spread you see is closer to the raw inter-bank spread, but you owe a commission on top, typically something like $3–$7 per round-trip standard lot.
“Commission-free” sounds attractive until you do the arithmetic. A market-maker showing you a 1.6-pip spread on EUR/USD is charging the equivalent of about $16 round-trip on a standard lot. An ECN showing you a 0.2-pip raw spread plus $5 round-trip commission is charging about $7 round-trip on the same lot. The “free” account is more than twice as expensive. The right cost to compare is total cost per round trip, not either component alone.
Slippage: the cost that doesn’t appear on a statement
Slippage is the gap between the price you saw and the price you actually got filled at. It is rarely zero and it is rarely in your favour.
Two situations produce most of it:
- Market orders during fast-moving conditions. A market order says “fill me now at the best available price”. Between the moment your click reaches the server and the moment the order is matched, the market may have moved. The fill is at the new price, not the one your screen showed.
- Stop-loss triggers. A stop is a market order in disguise. When price reaches your stop level, the broker fires a market order, and that order is filled wherever the next available liquidity sits. If the gap is large (a news release, a weekend gap), the fill can be meaningfully worse than your stop level.
Slippage tends to be asymmetric: filled-better is rare, filled-worse is common. That is not a conspiracy. It is a consequence of how market depth behaves at moments of stress: the bids and offers nearest the current price are consumed first, leaving only worse prices behind.
The honest way to think about slippage is as a tax on uncertainty. If your strategy depends on getting filled at exactly your stop or exactly your target, you do not really have a strategy that includes the friction the market imposes.
Last-look execution: a structural quirk
This one rarely comes up in beginner material, but it is part of how retail FX execution actually works. Last look is a practice in which the liquidity provider receiving your order has a brief window (often under 100 milliseconds) to accept or reject it after seeing it. If the market has moved against the provider in that window, they can decline the fill.
The academic literature on this is direct: under last look, the effective spread you pay is wider than the quoted spread, because the provider is free to reject the trades that would have lost them money but obliged to fill the ones that go their way (see Cartea, Jaimungal, and Walton, Foreign Exchange Markets with Last Look, arXiv:1806.04460). It is a one-way option in favour of the venue, and its cost is built into your fills in a way that is invisible on your statement.
You will rarely see “last look” mentioned on a broker’s marketing page. What you will see is “no last look” prominently displayed by venues that have chosen not to use it. That distinction tells you the practice is common enough to be worth advertising the absence of.
Putting it together: the cumulative bill
The single-trade numbers feel small. The aggregate does not.
Consider an account that opens one mini-lot trade every weekday for a hundred trading days, pays a typical retail spread cost of 1.5 pips per round trip (so $1.50 per trip), and breaks even on direction by sheer coincidence: every winning trade exactly cancels every losing trade. The account does no worse and no better than the market.
The equity curve looks like this:
The account has lost 15% of its starting capital without a single losing directional bet. The losses are entirely the cumulative spread. Bump the trade frequency to several positions per day, or trade a wider spread, and the curve steepens accordingly. is roughly what an account looks like during the early weeks of unaware over-trading.
This is the part of the math beginners most often discount. A 1.5-pip spread sounds harmless. A $1.50 per trip cost sounds trivial. Four trades a day for a year, at one mini lot, is a thousand trips and a thousand and a half dollars of friction, without anyone losing on a single position. On a $5,000 account, that is 30% of starting capital paid in costs before direction even gets a vote.
What to actually compare when looking at brokers
The marketing comparisons you will find on broker review sites tend to optimise for headline numbers: tightest spread, lowest commission, biggest bonus. None of those, on its own, is the right thing to look at. A small checklist that is more useful:
- Total cost per round trip on the pair you actually trade. Spread plus commission plus any extras, in money, not pips. Get a real fill on a demo account during the hours you intend to trade and time-stamp the spread.
- Execution quality, not just spread. Slippage statistics if the broker publishes them; whether the broker uses last look; whether they are an A-book (pass-through) or B-book (internalise client losses) shop, or some mix.
- Regulation and capital adequacy. What jurisdiction the broker is regulated in (FCA, ASIC, CySEC, ESMA-member, NFA, FINMA, etc.), and the negative-balance protection rules that apply there.
- Swap rates on the pairs and direction you intend to hold.
- Withdrawal mechanics. How long withdrawals take, what they cost, and whether there are restrictions on withdrawing profits that do not also apply to deposits.
The biggest cost in retail trading is usually not the broker. It is the trader’s own activity level. A broker with a slightly wider spread but better execution and stronger regulation is the better choice for almost everyone, because the gap closes quickly when you measure cost-per-trip rather than headline-spread.
The takeaway
Forex trading has four routinely-charged costs (spread, swap, commission, slippage) and a structural quirk in retail execution (last look) that behaves like a fifth. None of them is large in isolation. All of them compound, and the dominant variable in how much you actually pay is how often you trade. Comparing brokers on headline spread alone misses most of the picture; the right number is total cost per round trip on the specific pairs you intend to trade, weighed against execution quality and regulation.
The arithmetic above should make one point unmissable: if your strategy does not have a clear, measurable edge larger than your friction, the friction will eat the account on its own. That makes risk management and position sizing the first place to look, not the last.