ETF Tracking Error vs Tracking Difference: The Number That Really Matters

If you’re evaluating ETFs by expense ratio alone, you’re only reading half the label. I’ve seen this play out many times: two ETFs tracking the same index, nearly identical TERs, yet meaningfully different long-term results. The explanation is almost always in tracking error and tracking difference — two metrics that most investors skip over entirely.

Expense ratio matters. But it’s a stated cost, not the actual cost. What lands in your account depends on how precisely the ETF tracks its benchmark, and that’s what TE and TD measure. The distinction between the two is also widely misunderstood — even by experienced investors — so let’s get it straight from the start.

What Is Tracking Error?

Tracking error (TE) is the standard deviation of the daily return difference between an ETF and its benchmark index. It measures consistency — how steadily the ETF mirrors the index from one day to the next.

TE = σ(ETF daily return − Index daily return)

A lower TE means the ETF follows the index without large daily divergences. TE is typically expressed in annualized basis points (bp). One basis point equals 0.01%. A TE of 10bp means the annualized tracking volatility is approximately 0.10%.

The key thing to note: TE has no direction. It tells you how much the ETF wobbled around the index — not whether it ran ahead or behind. That’s what tracking difference is for.

What Is Tracking Difference — And Why It’s Different

Tracking difference (TD) measures the cumulative return gap between an ETF and its benchmark over a defined period, typically one year. It has a direction.

TD = ETF cumulative return − Index cumulative return

If a fund’s TD is −0.30% over one year and the index returned 10%, the ETF returned 9.70%. On a $10,000 investment, that’s $30 that didn’t make it into your account.

TD is usually negative — the ETF delivers slightly less than the index. But it can be positive, most commonly when securities lending income exceeds total operating costs. When TD is positive, the fund actually outperformed its own benchmark net of fees. This is more common than many investors realize.

TE and TD measure different things. A fund can have low TE (consistent day-to-day tracking) but still have a meaningfully negative TD (cumulative shortfall). The reverse is also possible. For long-term investors, TD is the more direct measure of real-world return impact — but both indicators together give you the complete picture. This distinction is the core value of this article.

Five Sources of Tracking Deviation

Why can’t an ETF perfectly replicate its index? Several reasons — and they push TD in different directions.

Securities lending is worth understanding because it often surprises people. Some broad-market ETFs generate enough lending income to offset their TER almost entirely — meaning a fund with a higher stated expense ratio can actually deliver a smaller real cost than a cheaper-looking competitor. For a deeper look at how fees compound over time, see How ETF Expense Ratios Erode Long-Term Returns.

The relationship between mutual funds and ETFs covers similar cost dynamics in a broader fund context if you want that comparison as well.

How to Interpret the Numbers

The ranges below apply specifically to broad-market index ETFs (S&P 500, global developed markets, total US market, etc.). Small-cap, thematic, and sector ETFs operate under different benchmarks. These are widely cited rule-of-thumb ranges used in practice, not formally standardised thresholds — comparing ETFs tracking the same index against each other remains more reliable than applying a fixed cutoff.

For TD, a useful reference point is the fund’s own TER. If TD is significantly more negative than −TER, something beyond the stated fee is bleeding return — often cash drag or high transaction costs during rebalancing.

⚠️ Important caveat: Small-cap ETFs, emerging market ETFs, and thematic ETFs routinely show TE of 20–50 bp or more. This isn’t poor management — it reflects the structural difficulty of precisely tracking illiquid, frequently-reconstituted benchmarks. Always compare a fund against peers tracking the same index, not against broad-market standards.

For practical guidance on putting all of this together in fund selection, How to Choose an ETF: A Checklist is a useful companion.

Dividend Withholding Tax and How Fund Structure Affects TD

For ETFs tracking foreign indices — US equities being the most common case for global investors — dividend withholding tax treatment is a meaningful and often overlooked driver of TD.

The benchmark index is calculated on a gross dividend basis (pre-tax). The ETF, however, only reinvests dividends after tax has been withheld at source. That gap flows directly into TD. How much is withheld depends on the fund’s legal domicile and any applicable tax treaty.

This structural difference is most visible when comparing US-listed ETFs (domiciled in the United States) with UCITS ETFs (typically domiciled in Ireland or Luxembourg). Both may track the same index with comparable stated fees, but their TD values often differ because of how each structure handles dividend withholding. US-based investors can access US-listed ETFs directly — VOO, VTI, and similar — with no regulatory barrier. EU investors cannot purchase US-listed ETFs due to PRIIPs regulations, so they use UCITS equivalents such as VUAA or CSPX instead.

The specific rates and treaty arrangements are beyond the scope of this article and change over time. The practical takeaway is this: when comparing ETFs on the same index with similar TERs, TD can diverge because of fund domicile and tax treatment, not just operational efficiency. TER comparison alone doesn’t capture this.

According to ESMA’s fund industry data, structural costs embedded in fund returns — including tax treatment differences — are a recurring theme in ongoing transparency and cost disclosure efforts across European markets.

How Much Does Annual TD Actually Cost Over Time?

The article’s examples so far show a single-year snapshot: TD of −0.30% costs roughly $0.30 per $100 invested in year one. But TD compounds silently. The table below works out what happens when a constant annual TD is applied over 10, 20, and 30 years — using a 7% per year gross index return as the baseline assumption. Starting value is set to 100 (units, not dollars — the result scales linearly to any currency amount).

Assumptions: index gross return 7% p.a.; TD is constant each year; figures are illustrative and computed arithmetically. Real-world TD fluctuates annually.

The compounding effect is the point. A TD of −0.30% feels immaterial over one year. Over 30 years it reduces terminal wealth by 61.5 units per 100 invested — the equivalent of roughly 8% of the index’s 30-year end value evaporating due to a gap that never appeared on the TER label.

The difference between −0.10% TD and −0.30% TD widens from 3.6 units at year 10 to 40.4 units at year 30. That is the cost of choosing an ETF on TER alone rather than on actual TD data — compounding magnifies what looks like a trivial annual gap into a material divergence by retirement.

• Add this step to your checklist: after finding two ETFs with similar TERs on the same index, look up their 1-year TD figures. A 0.20 percentage-point difference in TD — seemingly negligible in year one — widens to a gap of ~40 units per 100 invested (roughly 5% of the index end value) over 30 years at 7% gross returns.

Key Takeaways

The TE and TD checklist for ETF selection:

• Check the TER (expense ratio) — this is just the starting point
• Compare TD on a 1-year basis against other ETFs tracking the same index
• For broad-market ETFs, confirm TE is broadly below 15 bp as a rule of thumb (different benchmark for small-cap / thematic; peer comparison within the same index is more reliable than any fixed cutoff)
• Check whether the fund uses securities lending — if TD is low or positive, this is usually why
• Verify that the fund’s domicile (US-listed vs UCITS) is compatible with your regulatory and tax environment
• If TE is low but TD is significantly negative, investigate the specific source of the gap

ETFs look simple from the outside — just own the index, pay the fee, collect the returns. But “how precisely” and “at what real cost” the fund tracks its benchmark is what determines whether those returns actually reach you. Expanding your checklist from TER to TE and TD is the upgrade that separates a superficial choice from an informed one.

If you’re still building your foundation, What Is an ETF? covers the core mechanics. For a complete ETF selection framework beyond costs, How to Choose an ETF brings it all together.

For a broader view of how ETF costs compare to traditional mutual fund structures, Mutual Funds vs ETFs is a useful complement. If you’re weighing passive indexing against active management on performance grounds, Index Fund vs Active Fund Performance covers the long-run evidence.

Frequently Asked Questions

Q. What is the difference between tracking error and tracking difference?

Tracking error (TE) measures the volatility of the daily return gap between an ETF and its index — it captures consistency, not direction. Tracking difference (TD) is the cumulative return gap over a period, with a clear sign: −0.3% means the ETF returned 0.3 percentage points less than the index. Both measure different things and should be reviewed together.

Q. Can an ETF with a near-zero expense ratio still have tracking error?

Yes. Cash drag, sampling, and rebalancing transaction costs all generate tracking deviation independently of the stated fee. Securities lending can offset these — in some cases pushing TD into positive territory even for funds with non-zero TERs.

Q. Where can I check an ETF’s tracking error and tracking difference?

ETF providers often publish TE and TD on their fund pages. For comparison across funds, platforms like ETFdb, Morningstar, and JustETF (for UCITS funds) aggregate these figures. Use at least a one-year window for meaningful data.

Q. Why do small-cap ETFs tend to have higher tracking error?

Small-cap indices include many illiquid constituents, making precise execution at index prices difficult and costly during reconstitution events. Sampling is often necessary. The acceptable TE benchmark is structurally higher for small-cap indices than for large-cap broad-market ones.

Q. Is heavy securities lending in an ETF a risk worth worrying about?

Lending generates income that reduces real cost, but introduces counterparty risk. Most funds mitigate this with collateral. The risk isn’t inherently high, but verifying the fund’s lending policy and collateral quality through provider disclosures is a reasonable due diligence step.

← Back to all posts