ETF tracking difference is the number that tells you what an index fund actually cost you, and almost nobody checks it. Investors agonise over whether to pay 0.07% or 0.12% in fees, then never look at how far the fund’s return landed from its index. That is backwards. The TER is a price tag; the tracking difference is the receipt. Sometimes they match, sometimes the fund quietly beats its own fee, and sometimes a “cheap” fund costs more in practice than a dearer rival. This guide explains the three numbers people confuse, where the gap comes from, and how to check any fund yourself in five minutes.
ETF tracking difference vs tracking error vs TER
Three terms, constantly muddled, so let me separate them plainly.
TER (total expense ratio) is the annual fee the manager deducts from the fund’s assets for management, custody, administration and licensing. It is a published number — 0.07%, 0.20%, whatever the factsheet says. Crucially, it is a forecast of one cost, not a measurement of your outcome.
Tracking difference is the measurement. Take the fund’s total return over a period and subtract the index’s total return. If the index returned 10.0% and the fund 9.85%, the tracking difference is −0.15%. That figure bundles every cost and every offsetting income stream into one honest number. It is the true cost of ownership, and it can be smaller than the TER — occasionally even positive.
Tracking error is different again, and this is where most articles go wrong. It measures the volatility of the gap — how much the daily or monthly difference bounces around. A fund can have a large but steady tracking difference and a tiny tracking error. Long-term investors should care mostly about tracking difference; tracking error matters more to institutions on short horizons.
A taxi analogy helps. The TER is the fare on the price card. The tracking difference is what you actually paid after tolls and discounts. The tracking error is how erratically the driver drove. Your wallet only remembers the final bill.
Where ETF tracking difference comes from: the five moving parts
The realised gap is the sum of several forces — most pushing returns down, one or two pushing them back up.
- The TER itself. The baseline drag. All else being equal, a 0.20% fee should produce roughly a −0.20% tracking difference. All else is rarely equal.
- Securities-lending income. Funds lend out their shares to short sellers for a fee, and most of that revenue flows back into the fund. This income offsets costs and is the main reason some ETFs beat their own TER.
- Withholding-tax efficiency. Indices assume a standard tax haircut on dividends. A fund domiciled in a tax-efficient jurisdiction, or built with the right structure, loses less to withholding tax than the index assumes — and pockets the difference as outperformance.
- Sampling and replication choices. Funds holding every index constituent track mechanically; funds holding an optimised sample save on trading costs but accept small deviations in both directions.
- Cash drag and rebalancing costs. Dividends sit as cash before reinvestment, index changes force trades at unfavourable moments, and every transaction pays a spread. These frictions all nibble at returns.
Only one of these five is printed on the factsheet. The other four stay invisible until you measure the realised outcome — exactly why tracking difference deserves more attention than the fee. I cover the fee-versus-outcome argument in depth in my TER vs tracking difference guide.
The securities-lending twist
Lending deserves its own paragraph, because it surprises people most. When your ETF lends shares to a short seller, the borrower pays a fee and posts collateral, and between 60% and 100% of that fee is credited to the fund. On a large, in-demand portfolio, lending can generate several basis points a year — enough to cancel a big chunk of a low TER, or all of it.
This is how funds occasionally return more than their index: the index assumes no lending income, the fund earns some, and the gap turns positive. The trade-off is a small, collateralised counterparty risk. I unpack the mechanics and risk controls in my piece on securities lending.
Withholding tax and domicile
The second big offset is tax plumbing. When a fund receives foreign dividends, the source country withholds tax before the cash arrives, and how much depends on treaties. Irish-domiciled UCITS funds enjoy a US–Ireland treaty rate of 15% on American dividends, whereas the net-return indices most funds track assume a harsher 30% haircut. A US portfolio yielding around 1.3% therefore hands an Irish fund roughly 0.15–0.20% a year of “free” outperformance against its net index — before fees.
This is why Irish domicile dominates European ETF issuance, and why a Luxembourg fund on the same index can start each year at a structural disadvantage on US shares. The index did not change; the tax leakage did.
The synthetic edge on US equities
Synthetic ETFs push this one step further. A swap-based fund holds a substitute basket and receives the exact index return from a bank via a swap. Under US rules, swaps on certain major indices — the S&P 500 most famously — can pass through gross dividends with no withholding tax at all. That gives a well-run synthetic S&P 500 fund a head start of roughly 0.15–0.30% a year over the best physical rivals, which is why synthetics routinely top the tracking-difference tables for US equities. The cost is swap counterparty risk — collateralised and regulated under UCITS, but real. Physical purists pay a measurable price for their preference; whether it is worth paying is a personal call.
Case study: judging young, ultra-cheap funds like WEBN and SPYL
Now to the question that probably brought you here. Searches for “WEBN tracking difference” have exploded, and rightly so. Amundi’s WEBN charges roughly 0.07% for an all-country world portfolio; State Street’s SPYL charges around 0.03% for the S&P 500. Fees that low invite healthy suspicion: does the discount survive contact with reality, or does sloppy tracking claw it back?
The honest answer: the early evidence is encouraging, but the records are short. WEBN, launched in June 2024, has tracked its Solactive benchmark tightly, with a realised gap close to — and in some periods better than — its fee. The low-cost toolkit is working for it: Irish domicile, lending on a multi-billion-euro book, and a cheap Solactive licence. The full profile is in my WEBN factsheet, and its place among rivals in the cheapest all-world ETF comparison.
SPYL tells a similar story on a narrower index. Tracking 500 large, liquid US stocks is the easiest job in indexing — full replication, minimal rebalancing, deep markets — so a competent issuer can run it at 0.03% and still land within a whisker of the index. SPYL’s early figures have done exactly that. Details live in my SPYL factsheet, alongside the wider field in the cheapest S&P 500 UCITS ETF round-up.
Two caveats keep me honest. First, a year or two of clean tracking is a promising start, not proof; the acid tests are a bear market, a heavy rebalancing year and thin lending demand. Second — the part cost-obsessed investors resist — a fund with a higher TER can still track better. A 0.12% synthetic S&P 500 fund with gross-dividend treatment can beat a 0.03% physical fund on realised returns. The table below shows the concept.
TER vs realised tracking difference: illustrative concept examples
| Fund profile (illustrative) | Headline TER | Typical realised tracking difference | Why the gap differs from the fee |
|---|---|---|---|
| Ultra-cheap physical all-world (WEBN-style) | ~0.07% | ~−0.05% to −0.10% | Lending income and Irish treaty rate offset part of the fee |
| Ultra-cheap physical S&P 500 (SPYL-style) | ~0.03% | ~0.00% to −0.05% | Easy index to replicate; small offsets nearly cancel the fee |
| Synthetic S&P 500 fund | ~0.05–0.12% | Often positive vs net index | Swap passes through gross US dividends, no withholding drag |
| Established physical world fund | ~0.20% | ~−0.10% to −0.20% | Scale, lending and tax efficiency beat the headline fee |
| Sampled emerging-markets fund | ~0.18% | ~−0.20% to −0.50%, variable | Sampling, local taxes and trading frictions add hidden drag |
Every figure above is illustrative, chosen to show the mechanics rather than any fund’s current numbers. Realised results vary year by year — check the live data before acting.
How to check any fund’s ETF tracking difference yourself
You do not need a Bloomberg terminal. Three free routes cover it.
1. justETF’s chart comparison. Open the fund page, set the chart to a multi-year window, and compare the fund against its benchmark or a long-established rival tracking the same index. The gap between the two lines over three or five years is a direct read on relative tracking. Once a year is enough.
2. The issuer’s own documents. Every UCITS ETF publishes fund-versus-index return tables in its factsheet, annual report and KID. Find the calendar-year table showing “fund return” next to “index return” and subtract. Annual reports also disclose lending revenue and the share of the portfolio on loan — worth a skim.
3. The index provider’s data. MSCI, FTSE Russell and Solactive publish official index returns. Pull the net-return figure for a calendar year and set it against the fund’s NAV return for the same year. Compare like with like: net-return index versus fund total return, same currency, same dates. Mismatched currencies are the commonest way people frighten themselves with phantom underperformance.
One warning for young funds: a few months of data is noise. Rebalance spreads, launch-phase inflows and short windows can flatter or smear a new fund. Judge WEBN, SPYL and their peers on full calendar years, and expect a fair verdict to take three of them.
Learning ETFs the slow way?
The free 12-page Starter Kit gets you to a first, sensible portfolio faster: the 7-filter fund shortcut, the 3-question broker matrix, and the 7 costliest beginner mistakes.
My take
After eleven years of watching European ETFs, my rule is simple: the TER decides my shortlist, the tracking difference decides my purchase, and the tracking error decides nothing for a long-term holder. Fee wars have pushed headline costs so low that the real differences now hide in the plumbing — lending policy, domicile, replication method — and tracking difference is the only number that captures all of it at once.
On the young ultra-cheap funds, I am cautiously positive. The structural ingredients for good tracking are present, the early numbers back them up, and the issuers are giants with reputations to protect. But I hold my enthusiasm to the same standard I ask of you: check the realised gap once a year, and let the fund keep earning its place.
Common mistakes
- Treating the TER as the total cost. It is one input among five. The realised tracking difference is the bill you actually paid.
- Confusing tracking difference with tracking error. One measures the size of the gap, the other its wobble. Watch the first.
- Judging a young fund on six months of data. Short windows are noise. Full calendar years, ideally three, give a fair verdict.
- Comparing against the wrong index variant. Gross versus net return, or mismatched currencies, creates phantom gaps that panic people needlessly.
- Assuming the cheapest fund always tracks best. Synthetic structures and lending income mean a pricier fund can deliver a better realised result.
- Never rechecking. Lending revenue, spreads and fund sizes drift. A yearly five-minute check on justETF keeps your choice honest.
Key takeaways
- TER is the published fee; tracking difference is the realised gap between fund and index returns; tracking error is the volatility of that gap.
- Five forces drive the gap: the fee, lending income, withholding-tax efficiency, sampling, and cash drag plus rebalancing costs.
- Lending income and the Irish 15% US treaty rate can offset much of a fund’s fee; synthetic S&P 500 funds can beat the net index outright.
- WEBN and SPYL show encouraging early tracking that supports their ultra-low fees, but two years of data is a start, not a verdict.
- A higher-TER fund can genuinely track better than a cheaper one; judge on realised outcomes, not price tags.
- Check any fund via justETF chart comparisons, the issuer’s fund-versus-index tables, and the index provider’s official returns — yearly is enough.
FAQ
Q: What is ETF tracking difference in simple terms?
A: The fund’s total return minus the index’s total return over the same period. If the index made 10% and the fund 9.9%, the tracking difference is −0.1%. It is the truest measure of what the fund cost you.
Q: Is tracking difference the same as tracking error?
A: No. Tracking difference measures the size of the return gap; tracking error measures how much that gap fluctuates. A fund can have a steady lag (low error) that is still expensive (large difference), or vice versa.
Q: Can an ETF beat its own index?
A: Yes, modestly. Lending income and better withholding-tax treatment than the net index assumes can push the fund above its benchmark. Synthetic S&P 500 funds do this fairly reliably against net-return indices.
Q: How is WEBN’s tracking difference so far?
A: Encouraging — the realised gap has stayed close to its roughly 0.07% fee, at times better, helped by lending income and Irish domicile. The record is about two years old, so keep checking annually.
Q: Is SPYL’s 0.03% TER too good to be true?
A: No. The S&P 500 is the easiest major index to replicate, so tiny fees are sustainable, and SPYL’s early tracking has been tight. Judge it on full calendar years, as with any young fund.
Q: Why do synthetic ETFs track US indices better?
A: Swaps on qualifying US indices pass through gross dividends without withholding tax, while physical funds lose at least 15%. That saving usually outweighs the swap fee, at the cost of collateralised counterparty risk.
Q: Where can I check a fund’s tracking difference for free?
A: justETF’s comparison charts, the fund-versus-index tables in the issuer’s factsheet and annual report, and official index returns from MSCI, FTSE Russell or Solactive. Morningstar is a useful cross-check.
Q: Does domicile really affect returns?
A: Yes. Irish-domiciled funds pay 15% withholding on US dividends under treaty, while net-return indices assume 30%. On a US-heavy portfolio, that is worth roughly 0.15–0.20% a year before fees.
Q: Should I switch funds over a few basis points of tracking difference?
A: Rarely. Selling an established position can trigger capital-gains tax that dwarfs years of the saving. Redirecting new contributions to the better-tracking fund captures most of the benefit.
Q: How many years of data make a tracking record trustworthy?
A: At least three full calendar years, ideally including one rough market. Shorter windows are dominated by launch effects, rebalancing luck and measurement noise.
Cited sources: justETF and Morningstar. Check current TER, AUM and holdings on the issuer factsheet — figures change.
Not personalised financial advice. Verify your situation with a qualified adviser. Operated by Aurum Digital Marketing LTD (UK), Companies House 15203665.
How much does domicile actually cost? Model the US dividend withholding drag of an Irish versus a Luxembourg fund with the free withholding tax and domicile calculator.
European UCITS/ETF investor with 11+ years building a global dividend and index portfolio, and founder of Financial Expert Class. I write every guide here from first-hand experience investing across EU brokers and tax regimes — not theory. More about me →