The trading cost nobody screenshots 

The trading cost nobody screenshots 

Spreads get advertised. Commissions get compared. Overnight financing gets discovered, usually in week three, usually by accident. 

Which cost do traders actually check? 

The spread, almost always, because it is the number brokers compete on and therefore the number they publish in large type. Commission comes second, because it appears as a line item and is easy to compare. Overnight financing comes a distant third, and for a lot of retail accounts it comes nowhere at all until it has already been charged. This is a ranking by visibility, not by size. For anyone who holds a leveraged position for more than a day or two, the order is frequently reversed, and financing quietly becomes the largest single cost of the trade. 

What does financing actually cost on a real position? 

Figures from a live funded account at an FCA-regulated UK broker in July 2026, on a standard EUR/USD lot, give a concrete answer. Holding the position long cost roughly GBP 7.10 a night. Holding the identical size short cost roughly 90p a night. Same instrument, same broker, same size, and a difference of nearly eight times depending on which way round you were facing. 

For context, the all-in cost of opening and closing that same lot inside the day, spread and commission together in liquid hours, came to about GBP 5.30. So a single night held long cost more than the entire round trip. Hold it a fortnight and the financing line dwarfs everything else on the statement. 

Two things pin those figures to their moment before anyone extrapolates from them. Swap rates track central bank policy, so the near-eightfold gap between the two directions belongs to the July 2026 rate differential rather than to EUR/USD, and it will narrow or flip as policy moves. The other half of the number is the broker’s own markup, which is a commercial decision and varies between firms. Funding an account cannot get round either constraint: a swap rate observed once is a swap rate observed once, and the site that funded this one is reporting a dated reading rather than a rate card. What generalises is the structure, which is durable, and the pounds and pence are an illustration of it. 

One live funded account at an FCA-regulated broker, a standard EUR/USD lot, July 2026. The nightly figures were recorded; the week and month rows are worked forward from them, with the month charged as 38 nights rather than 30 because four Wednesdays at triple rate add eight to it. 

Why is Wednesday different? 

Because spot foreign exchange settles two business days after the trade. A position held over Wednesday night is therefore settling on Friday, and rolling it means rolling over the weekend, so the financing for Saturday and Sunday is collected in one go. The charge is typically tripled. 

Nothing about this is hidden or improper. It follows directly from settlement conventions and every broker will confirm it if asked. 

But it is not in the advertising, and a trader who has budgeted for seven nights of financing across a week will find they have paid for nine. On the long side of the example above that is an extra GBP 14 nobody planned for. 

Why is the long side so much more expensive than the short? 

Financing on a leveraged position reflects the interest rate differential between the two currencies plus the broker’s own charge. When you are long a lower-yielding currency against a higher-yielding one, the differential works against you and the broker’s markup is added on top. Reverse the position and the differential works for you, though the markup is still deducted, which is why the short side cost 90p rather than paying you anything. 

That asymmetry is the part with practical consequences. A strategy that is structurally long a particular pair carries a running cost that the identical strategy run short does not. Backtests that ignore financing will overstate the long version and understate nothing, which is a fairly reliable way to generate a strategy that works on paper and bleeds in production. 

How does this interact with leverage? 

Directly, and this is the bit that catches people out. Financing is charged on the full notional value of the position, not on the margin you put up. Under FCA rules a retail client can take up to 30:1 on major currency pairs, so GBP 3,300 of margin controls roughly GBP 100,000 of notional exposure. The financing follows the GBP 100,000. Which means the ratio of nightly financing to the capital you have actually committed is far higher than most people assume. At the July 2026 rates above, a month of holding that long position cost around GBP 270 against margin of a few thousand pounds. That is a meaningful share of the account, accruing whether the trade goes anywhere or not. 

Is any of this disclosed? 

Yes, in the sense that swap rates are published, usually inside the platform, sometimes in a contract specifications table several clicks from anywhere a new customer would look. The information exists. What does not exist is any prominence, because financing is not a competitive selling point and no broker leads with it. 

The result is a predictable gap between what is technically disclosed and what is practically known. Every figure above exists because somebody deposited money, opened a position, held it and then read the statement, and this research, built from live accounts opened and funded with the reviewer’s own money rather than from published fee schedules, is the sort of exercise that has to be done from outside the industry because nobody inside it has a reason to do it. 

What should a trader do with this? 

Before you open a leveraged position you intend to hold, find the swap rate for that instrument in that direction, multiply it by the number of nights you expect to be in, add a night for every Wednesday, and put the result next to your intended profit target. If financing eats a visible share of the target, the trade needs to be shorter, smaller, or a different instrument. 

That is a two-minute calculation and it is the single highest-value habit in leveraged trading, precisely because so few people do it. Spread comparison is well served by the market. Financing comparison you have to do yourself. 

If you are already holding something, the same sum runs backwards and takes about as long. Find the swap line on last month’s statement, divide it by the margin you have committed rather than by the notional it controls, and you have the rate you are quietly paying to keep the trade open. Most people who run that once change something: the size, the direction, or how long they are prepared to sit in it. Four structural facts survive any change in rates, and they are the four to carry away. Financing is charged on notional. The two directions are priced differently. Wednesday costs triple. And none of it appears anywhere a new customer would look.