ETF Tracking Error: Why Your ETF Might Not Match Its Index

What Causes Tracking Error?

Expense ratios are the largest and most predictable source of tracking error. An ETF with a 0.03% expense ratio will lag its index by approximately 0.03% per year purely from fees. Sampling is another major source — many ETFs do not hold every security in an index. An S&P 500 ETF might hold 500 stocks, but an international small-cap ETF might hold only a representative sample of 1,000 stocks out of 5,000. The sampling methodology introduces tracking error when the sampled stocks perform differently from the full index. Cash drag occurs because ETFs hold small amounts of cash for daily operations (dividends received but not yet reinvested, rebalancing proceeds). This cash earns near-zero returns while the market may be rising, creating a small performance gap. Securities lending can offset these sources — some ETFs actually outperform their indexes by lending shares to short sellers and keeping the lending revenue. The net tracking error is the combination of all these factors. More on how index funds manage tracking error.

Tracking Difference vs Tracking Error

Tracking difference is the raw performance gap between an ETF and its index over a specific period. If an ETF returned 9.8% and the index returned 10.0%, the tracking difference is -0.2% (the ETF underperformed). Tracking error is the standard deviation of those differences over time — a measure of consistency. An ETF could have a tracking difference of -0.2% every year (very predictable) or it could range from -0.5% to +0.1% (unpredictable). High tracking error (in the statistical sense) is worse because the deviation from the index is unpredictable, making it harder to rely on the ETF for precise index exposure. Most quality ETFs have tracking difference close to their expense ratio and tracking error under 0.10% annually. Always check both metrics on the issuer's website before choosing an ETF. Compare tracking error across ETFs.

How Securities Lending Affects Tracking Error

Many ETFs lend their portfolio securities to short sellers and earn lending fees. This revenue is paid back to the ETF and reduces the fund's expenses, which can offset or even exceed the expense ratio. Some popular ETFs have negative net expenses — meaning they generate more revenue from lending than they charge in fees, resulting in a positive tracking difference (outperformance relative to the index). For example, some broad market ETFs have securities lending revenue of 0.02% to 0.10% annually, which partially or fully offsets their expense ratio of 0.03% to 0.07%. However, securities lending introduces counterparty risk — if the borrower defaults, the ETF could face losses. The risk is small (lending is collateralized) but real. ETFs disclose their securities lending program in the prospectus, including the percentage of assets lent and the revenue generated. For most investors, securities lending is a net positive that reduces tracking error, but it is worth understanding how your specific ETF handles it. See how ETF lending compares to mutual funds.

Index Rebalancing and Reconstitution

Index providers periodically rebalance their indexes — adding, removing, or adjusting the weights of securities. The S&P 500 rebalances quarterly and reconstitutes annually. When this happens, the ETF must trade to match the new index composition. These trades incur transaction costs (commissions, bid-ask spreads, market impact) that create tracking error. Funds that track the same index but use different rebalancing methodologies may have different tracking error. Some ETFs use optimized sampling to minimize rebalancing costs, while others fully replicate the index (holding every security) to minimize tracking error. For large, liquid indexes like the S&P 500, full replication is common and rebalancing costs are minimal. For indexes with thousands of securities (like the CRSP US Total Market Index), sampling is necessary and rebalancing costs are higher. The tracking error from rebalancing is typically small (0.01% to 0.05%) but can spike in volatile markets when index composition changes significantly. Learn about rebalancing in bond ETFs.

Dividend Timing and Withholding Taxes

Dividend timing creates small but measurable tracking error. When companies in the index pay dividends, the ETF receives the cash but may not reinvest it immediately (cash drag). The index, by contrast, assumes dividends are reinvested continuously. This timing difference typically causes 0.01% to 0.05% in annual tracking error. For international ETFs, withholding taxes on dividends paid by foreign companies create additional tracking error. The index may assume no withholding taxes are applied, but the ETF must pay 15% to 30% withholding on dividends from foreign stocks. This can create persistent tracking error of 0.10% to 0.30% for international equity ETFs. Some international ETFs are more efficient at recovering withholding taxes through tax treaties, resulting in lower tracking error. Check the ETF's annual report for the "dividend income" line and compare it to the index's assumed dividend yield to estimate this component of tracking error. More on foreign tax credits for international ETFs.

How to Avoid High Tracking Error

Choose ETFs from reputable issuers with a track record of low tracking error. Vanguard, BlackRock (iShares), State Street (SPDR), and Charles Schwab are known for tight tracking on their core index ETFs. Avoid niche or thematic ETFs that use complex sampling methodologies or track custom indexes — these tend to have higher and less predictable tracking error. Check Morningstar's "Tracking Difference" metric on the ETF's quote page. Look for an annual tracking difference that closely matches the expense ratio (a tracking difference of -0.03% on a 0.03% ER fund is ideal). Avoid ETFs where the tracking difference significantly exceeds the expense ratio, as this indicates structural inefficiency. For funds that use securities lending, a small positive tracking difference (outperformance) is acceptable but should be stable over time. If you see wild swings in tracking difference from year to year, that is a red flag. Stick with large, liquid ETFs that have years of proven tracking performance. Direct indexing as an alternative to ETFs.

What is a good tracking error for an ETF?

A good tracking error for a passive ETF is under 0.10% annually for broad market equity ETFs, under 0.20% for international equity ETFs, and under 0.30% for bond ETFs. For the most popular S&P 500 ETFs (VOO, IVV, SPY), tracking error is typically 0.01% to 0.03%. For actively managed ETFs, tracking error is not applicable because they do not track an index.

Can tracking error be positive?

Yes. Some ETFs can outperform their benchmark index due to securities lending revenue, favorable dividend timing, or sampling methodologies that happen to beat the index. Positive tracking error is more common among ETFs than mutual funds because of the securities lending mechanism. However, positive tracking error is not guaranteed and can reverse in future periods. An ETF that consistently outperforms its index may be taking on additional risk that is not captured by the tracking difference calculation.

Does a higher expense ratio always mean higher tracking error?

Generally yes, but not always. A fund with a 0.50% expense ratio will typically have larger tracking error than a 0.03% fund, assuming similar tracking methodology. However, securities lending revenue can offset expense ratios, and efficient sampling can reduce tracking error below what the expense ratio suggests. Conversely, a low-cost ETF with poor sampling or inefficient trading can have higher tracking error than a slightly more expensive ETF with full replication. Always check actual tracking difference rather than assuming lower expenses equal lower tracking error.

How do I find an ETF's tracking error?

ETF issuers publish tracking difference data in the fund's annual and semi-annual reports. Morningstar provides "Tracking Difference" as a standardized metric on ETF quote pages. The ETF's website typically has a "Performance" tab showing cumulative and annual returns versus the benchmark. A quick method: subtract the ETF's total return from the index total return for the same period. For a more rigorous analysis, calculate the standard deviation of daily return differences between the ETF and the index over a rolling 12-month period. Most issuers also publish "premium/discount" and "tracking error" data in their ETF fact sheets.

Related Resources