ETF Tracking Error vs. Tracking Difference: How to Read the Return Gap
An ETF's cumulative return gap versus its index and the volatility of that gap are different measures. Learn why low tracking error can coexist with a persistent cost-related shortfall.
Tracking difference is the return gap between an ETF and its benchmark over a specified period. Statistical tracking error measures how much periodic return differences fluctuate. An ETF that consistently lags slightly can have low tracking error and still have a negative cumulative tracking difference. Check the definition and observation frequency in each disclosure.
One measure shows direction; the other shows variability
If tracking difference is defined as ETF return minus index return, a negative figure means underperformance. Tracking error follows a statistical formula, such as the annualized standard deviation of daily or monthly return differences. The measure answering how far a fund lagged on average is different from the one answering how irregularly it deviated. Judging management quality from one number can miss persistent costs or rare large deviations. Also check whether a disclosure uses the term tracking error simply to mean a return gap.
Align currency and dividend treatment first
A price index excludes dividend reinvestment, while total-return indexes can differ in their pre-tax or after-tax assumptions. Comparing an ETF's return with distributions reinvested against a price index can make dividends look like excess performance. Differences in currency hedging, valuation cutoff, or base currency can create gaps unrelated to management skill. When building a data table, fix the start date, end date, currency, and distribution treatment in the first row. Trying to align these conventions only after downloading the numbers makes it easy to mix incompatible time series.
Identify causes on the dates when the gap widens
Cash holdings, index constituent changes, taxes, trading costs, and sampling methods can affect return differences. Check whether dates when the gap widened coincide with scheduled index reconstitutions or large distributions. Distinguish a one-off event followed by a return to the normal range from recurring costs that keep widening the cumulative shortfall. Do not assume past outperformance will persist. Factors that partly offset costs, such as securities-lending income, can also vary by period.
Worked example
Hypothetical example: Suppose the monthly return differences over three months are −0.02 percentage points each month for A, and −0.30, +0.20, and +0.04 percentage points for B. Both have an arithmetic average of −0.02 percentage points, but A's gap is constant while B's fluctuates substantially. The sum of these monthly gaps should not be called the exact cumulative tracking difference. To compare actual cumulative performance, compound the ETF's monthly returns and the index's monthly returns separately, then compare the results.
What to check
- Have you checked the tracking error formula, observation frequency, and measurement period?
- Have you aligned the index's dividend and tax treatment?
- Have you distinguished NAV returns from market-price returns?
- Have you considered both the cumulative gap and the volatility of periodic gaps?
Sources and primary materials
This is educational material about research methods. For an actual company decision, verify the filings and primary sources for the relevant period.