ETF Tracking Difference: The Real Gap Between Your ETF and Its Index
Tracking difference is the actual gap between an ETF’s total return and its benchmark index’s return over a given period, usually slightly negative due to fund fees, trading costs, and withholding tax, and is a more useful real-world performance gauge than the more commonly quoted tracking error, which measures the volatility of that gap rather than its size.
Not financial advice. All figures for educational reference only. Data as at July 2026.
Key Takeaways
- Tracking difference is typically close to, but usually slightly worse than, an ETF’s expense ratio in a given year — for example a 0.30% expense ratio might produce a tracking difference of around -0.35% to -0.45% once trading costs and withholding tax are included.
- Tracking error measures how consistent the tracking difference is over time, essentially its statistical volatility, while tracking difference measures the actual cumulative return gap — the two numbers answer different questions and shouldn’t be used interchangeably.
- US-domiciled ETFs held by Singapore investors face a 30% US dividend withholding tax, which alone can create a wider tracking difference than Ireland-domiciled UCITS ETFs, which typically only face a 15% US withholding tax under the US-Ireland tax treaty.
- Physically-replicating ETFs, which hold the actual underlying stocks, and synthetically-replicating ETFs, which use swaps, can post noticeably different tracking differences depending on securities lending income and swap financing costs.
- A consistently positive tracking difference, where the ETF actually outperforms its index, is rare and usually comes from securities lending income offsetting fees rather than from any active stock-picking, since the ETF is designed to passively track the index.
What Is ETF Tracking Difference?
An ETF is designed to replicate the performance of a benchmark index as closely as possible, but it can never do so perfectly, because running the fund costs money and involves practical frictions the index itself doesn’t experience. Tracking difference is simply the resulting gap between what the ETF actually returned and what the index returned, over the same period.
It’s easy to confuse tracking difference with the more commonly advertised tracking error, but they measure different things. Tracking difference is a single cumulative number — “this ETF returned 0.4% less than its index over the past year.” Tracking error is a statistical measure of how much that gap bounces around day to day — a low tracking error simply means the ETF tracks the index consistently, even if it consistently trails by a meaningful margin.
For an investor deciding between two similar ETFs, tracking difference is generally the more directly useful number, since it reflects the actual cost of holding the fund versus simply owning the index, whereas tracking error mainly matters for very short-term or leveraged trading strategies.
How It Works for Singapore Investors
| Factor | Effect on Tracking Difference |
|---|---|
| Expense ratio | Direct annual drag, roughly equal to the stated TER |
| US dividend withholding tax (US-domiciled ETF) | 30% withholding on US dividends — significant additional drag for SG investors |
| US dividend withholding tax (Ireland-domiciled UCITS ETF) | Reduced to 15% under the US-Ireland tax treaty — smaller drag |
| Securities lending income | Can partially offset fees, sometimes producing a smaller-than-expected tracking difference |
| Trading and rebalancing costs | Small ongoing drag, larger for less liquid or higher-turnover indices |
Source: Standard ETF industry mechanics; US-Ireland double taxation treaty withholding tax treatment.
Tracking Difference Example
An ETF with a 0.07% expense ratio tracking a global equity index returns 9.85% over a year, while the index itself returns 10.00% over the same period.
- Raw tracking difference: 9.85% − 10.00% = −0.15%
- This is roughly double the 0.07% expense ratio alone, reflecting additional drag from withholding tax on dividends and minor trading costs.
- An equivalent UCITS-domiciled version of the same ETF, facing a lower 15% US withholding tax rate instead of 30%, might post a smaller tracking difference of around −0.10% over the same period, despite a similar headline expense ratio.
Advantages of Comparing Tracking Difference
- More honest cost comparison than expense ratio alone. Two ETFs with identical expense ratios can have meaningfully different tracking differences due to tax domicile and replication method.
- Reveals hidden costs. Withholding tax drag doesn’t appear anywhere in a fund’s stated expense ratio, but shows up clearly in tracking difference.
- Useful for long-term holders. Since tracking difference compounds over years, even a small annual gap matters more to long-term investors than short-term traders.
- Helps evaluate fund manager efficiency. A consistently smaller tracking difference than peers, for a similar index and fee, suggests better fund operations.
Risks and Limitations
- Historical, not guaranteed. Past tracking difference doesn’t guarantee the same gap in future years, since costs and tax treatment can change.
- Not always disclosed clearly. Some fund factsheets emphasise tracking error over tracking difference, making it harder for retail investors to find the more relevant figure.
- Comparing across different indices is misleading. Tracking difference should only be compared between ETFs tracking the same or very similar index, not different underlying benchmarks.
- Currency effects can distort short-term figures. For SGD-denominated comparisons of a foreign-currency ETF, currency movements can temporarily obscure the underlying tracking difference.
Tracking Difference vs Tracking Error
| Feature | Tracking Difference | Tracking Error |
|---|---|---|
| What it measures | Actual cumulative return gap vs index | Volatility/consistency of that gap over time |
| Typical unit | Percentage points per period (e.g. per year) | Standard deviation, annualised percentage |
| Best used for | Comparing real-world cost of holding the ETF | Assessing consistency of index-tracking quality |
| Can be positive? | Yes, rarely, via securities lending income | Always a positive number (it’s a measure of dispersion) |
| More relevant for | Long-term buy-and-hold investors | Short-term or leveraged/inverse ETF traders |
The Bottom Line
For Singapore ETF investors, tracking difference — not the more commonly quoted tracking error — is the number that actually tells you how much return you’re giving up by holding the fund instead of the index, and comparing it across similarly-indexed ETFs, especially US versus Ireland-domiciled versions, can meaningfully affect long-term returns.
Frequently Asked Questions
What is ETF tracking difference?
It’s the actual gap between an ETF’s total return and its benchmark index’s return over a given period, reflecting the real-world cost of holding the fund, including fees, trading costs, and withholding tax.
What's the difference between tracking difference and tracking error?
Tracking difference measures the actual cumulative return gap between the ETF and its index, while tracking error measures how volatile or consistent that gap is over time — they answer different questions.
Why does a US-domiciled ETF have a wider tracking difference than a UCITS ETF?
US-domiciled ETFs face a 30% US dividend withholding tax for Singapore investors, while Ireland-domiciled UCITS ETFs typically only face a 15% US withholding tax under the US-Ireland tax treaty, resulting in less drag on returns.
Can an ETF's tracking difference be positive?
Yes, though it’s uncommon — it usually happens when securities lending income earned by the fund offsets or exceeds its fees and other costs, causing the ETF to slightly outperform its index.
Is a low expense ratio the same as a low tracking difference?
Not necessarily — expense ratio is only one component of tracking difference, which also includes withholding tax, trading costs, and securities lending income, so two ETFs with the same expense ratio can have different tracking differences.
Where can I find an ETF's tracking difference?
Some fund providers publish it directly in factsheets or annual reports; otherwise it can be estimated by comparing the ETF’s published total return against its stated benchmark index’s total return over the same period.