Skip to content
Find an ETF

What it actually costs to hold an ETF (advanced)

Part of Costs & fees Advanced

Three cost numbers, three different jobs — and the fund that looks cheapest on any one of them alone isn’t always the cheapest to actually hold.

Why the headline fee is a floor, not an answer

The yearly running cost of the fund, shown as a % of your money. Lower is cheaper. More → is the number every factsheet leads with, and it’s a real, guaranteed, every-single-year cost. But it’s a floor, not a ceiling: it’s what the fund promises to charge for running itself, not everything that actually drags on your return once the fund is out trading in the real market. Two funds with an identical 0.20% TER can end up costing you meaningfully different amounts once you look one level deeper.

All-in cost — the first layer down, and its own blind spot

The all-in yearly cost of owning the fund — the headline fee (TER) plus trading and other running costs rolled in. A truer picture of what you actually pay than the TER alone. More → rolls the fund’s own internal trading and running expenses into the TER, and it usually sits a little above it — rarely by a lot, for a plain index fund. It’s a genuinely truer number than TER alone. Its own blind spot: it’s still an internal figure. It says nothing about what it costs you to actually buy and sell the fund on an exchange — the The small gap between a fund’s buy price and its sell price at any moment. Crossing it — buying, then later selling — is a real, if tiny, cost on top of the fund’s own yearly fee. More → you cross each time you trade, which sits entirely outside both TER and all-in cost.

Tracking difference — the one that can run either way

Here’s the one that trips up a pure cost-column comparison: How closely the fund’s actual return has matched the index it targets — a small gap is normal; a bigger one means it strayed further from what it is meant to copy. More → isn’t a fee at all — it’s what actually happened once the fund traded in the real world for a year. A fund can post a tracking difference worse than its own TER (sampling drag, cash buffers, tax on dividends received) or, less intuitively, a tracking difference better than its TER — some funds earn back a little by lending out their holdings, which can partially offset the running cost. A lower-TER fund with a persistently worse tracking difference than a slightly higher-TER peer isn’t automatically the cheaper one to actually hold.

Reading the three together

The practical version: when two funds track the same index, don’t stop at the TER column. Check the all-in cost next — if it’s a lot wider than the TER gap between the two funds, that’s worth a second look at what’s driving it. Then check tracking difference over a few years, not one — a single year can be noisy, but a fund that’s persistently a little wider than its own TER (or its peers') is quietly costing more than the label suggests, and one that’s persistently tighter (or even ahead) is delivering more than its label promises. None of the three alone tells the whole story; together they’re the closest thing to an honest answer.

What we still can't show you

To be straight about the gaps: our tracking-difference read compares a fund against others tracking the same index — an honest peer comparison, not a licensed index feed, so it’s "tighter or wider than similar funds," not an official tracking-error figure against the real index series. We also don’t yet show a fund’s actual bid-ask spread (a real, per-trade cost that sits outside all three numbers above) — that needs a live quote feed we don’t currently carry. Read the three signals above as the most complete honest picture we can build from what we store today, not a claim that nothing else affects your real cost.

🤔 Two ETFs track the same index. Fund A has a 0.10% TER; Fund B has a 0.15% TER but a tracking difference 0.08 percentage points tighter than Fund A’s, three years running. Which single number tells you the real cost story?

Common questions

Is a fund with a better (more negative-favourable) tracking difference than its TER too good to be true?
Not necessarily — securities lending income is a real, if modest, offset some funds use, and it’s disclosed in fund reports. Persistent, multi-year outperformance vs TER is a fair sign a fund runs efficiently; a one-off year of it is more likely noise or a timing quirk.
Should I always pick the fund with the best all-in cost?
It’s the strongest single signal of the three, but check tracking difference over a few years too before deciding — this is background to inform your own comparison, not a rule to apply blindly, and it’s never a recommendation of a specific fund.