An index is a perfect measurement: it moves by definition. An ETF is a product trying to reproduce that measurement, and the reproduction is never exact. The difference between the two is tracking error.
Tracking error is not a flaw to eliminate; it is the normal cost of running a product that follows a rule. Understanding it is how you compare two funds honestly.
Tracking error measures how closely a fund's return matches its index. It exists because a product carries fees, may sample instead of holding everything, handles currency differently, and rebalances on its own schedule. Two funds tracking the same index can have different tracking errors, and a smaller one usually reflects a better-run or lower-cost vehicle.
What Tracking Error Is
Tracking error is the deviation of a fund's return from its benchmark index. If the index rises 10% and the fund rises 9.5%, the fund trailed by half a point that period.
It is usually measured as the volatility of that deviation over time, or as a simple cumulative gap. Either way, it answers the same question: how close does the product come to the rule?
| Year | Index return | Fund return | Deviation that year (fund − index) |
|---|---|---|---|
| Year 1 | +12.0% | +11.6% | -0.4 pts |
| Year 2 | +5.0% | +4.7% | -0.3 pts |
| Year 3 | -8.0% | -8.3% | -0.3 pts |
| Year 4 | +20.0% | +19.5% | -0.5 pts |
| Year 5 | +3.0% | +2.7% | -0.3 pts |
The fund trailed the index in all 5 years, with deviations consistently between 0.3 and 0.5 percentage points — a sign that this fund's tracking is stable, just stably a little behind.
-0.36% is roughly 36 basis points — about the same order of magnitude as a low-fee fund's annual expense ratio. A deviation this consistent and this small usually points to a vehicle running steadily, not to a problem that showed up in any single year.
Why the Gap Exists
The fund pays fees that the index does not have, which drags the return below the benchmark. Some funds sample the index, holding a representative subset instead of every member, which can drift from the full basket.
Currency treatment and rebalancing schedules add more small differences. None of these are signs of failure; they are the mechanics of running a product.
| Source | How it works | Typical direction |
|---|---|---|
| Fees | Management costs reduce the return | Fund trails index |
| Sampling | Holding a subset instead of everything | Small drift |
| Currency | Hedging or conversion differences | Depends on method |
| Rebalancing | Different timing of adjustments | Small drift |
How to Read Tracking Error
When comparing two funds, tracking error is a comparison of vehicles, not of strategies: the index is the same, so the difference is in how each product runs.
A persistently smaller tracking error usually reflects a lower-cost or better-run vehicle. A larger one may signal higher fees, heavy sampling, or operational differences worth understanding.
A fund can trail its index slightly and still be doing exactly what it was designed to do. The question is whether the gap is reasonable for the vehicle's costs and method, not whether it is zero.
When Tracking Error Matters
It matters when you compare funds or explain a gap between an ETF and its index headline. The index is a benchmark; the tracking error is the vehicle's toll.
It matters less for short, casual comparisons and more for long-term holding, where costs and tracking differences compound into a meaningful gap.
Frequently Asked Questions
Can an ETF ever match its index exactly?
Not in practice. Fees, sampling, currency, and rebalancing always create some difference, however small.
Is a bigger tracking error always worse?
Usually, but not automatically. It depends on why it exists: high fees are a real cost, while some sampling differences are a deliberate design choice.
How do I compare two funds?
Check both the index rule and the vehicle terms: fees, tracking method, currency, and reported tracking error. The same index with different vehicles can deliver different results.
