Tracking error and tracking difference: how to measure ETF replication quality

Tracking error and tracking difference are two distinct measures of how closely an ETF follows its benchmark index, and confusing the two leads to different conclusions about fund quality. Tracking difference measures the cumulative return gap; tracking error measures the consistency of that gap. An ETF that reliably underperforms its index by 0.15% every year has low tracking error but a meaningful tracking difference. Understanding both metrics is essential for evaluating ETF replication quality.

What tracking error and tracking difference are

Tracking difference is the simplest of the two measures. It is the difference between the ETF's total return and the index's total return over a given period, typically expressed on an annualised basis. If an ETF returns 9.85% over a year and its index returns 10.00%, the tracking difference is −0.15 percentage points. A positive tracking difference means the ETF outperformed the index; a negative tracking difference means it lagged.

Tracking error—sometimes called active tracking error or annualised tracking error—is the annualised standard deviation of daily or periodic return differences between the ETF and its index. It measures consistency: how stable or volatile the deviation is, rather than how large it is in total. A low tracking error means the ETF's performance relative to the index is predictable from day to day. A high tracking error means the gap fluctuates significantly, even if the average deviation is small.

How they are calculated

Tracking difference is straightforward: total ETF return minus total index return over the measurement period. Most fund factsheets and ETF data providers publish annual tracking difference figures. The key variables are the measurement period, whether distributions are included in the ETF return, and whether the index return assumes gross or net dividend reinvestment—the correct comparison treats both on the same basis.

Tracking error is calculated from a time series of daily return differences. For each day, compute the ETF return minus the index return. Then calculate the standard deviation of this series and annualise it by multiplying by the square root of 252 (the approximate number of trading days in a year). A tracking error of 0.05% per year is typical for a large, well-managed equity ETF tracking a liquid index. Tracking errors above 0.50% suggest significant operational challenges or replication choices that introduce meaningful variability.

What the evidence shows

Analysis of European ETF data by providers including Morningstar and TrackInsight consistently finds that tracking difference and tracking error are only loosely correlated with TER. Some ETFs with higher stated costs achieve better tracking than lower-cost competitors because of superior securities lending programmes or more efficient replication. Conversely, ETFs tracking illiquid or complex indexes often exhibit high tracking error regardless of their stated cost, because the underlying market's trading costs introduce significant day-to-day variability in replication quality.

Tracking difference tends to be the more useful metric for investors comparing long-term holding costs, because it captures the actual return gap experienced. Tracking error is more useful for understanding intraday and short-term replication behaviour—it matters more for investors who trade the ETF frequently or who use it to implement short-term tactical positions.

Limitations and trade-offs

Tracking difference figures are sensitive to the choice of measurement period. An ETF's tracking difference over the past 12 months may look favourable or unfavourable depending on when dividends were paid, how the index reconstituted, and whether there were unusual market events during the period. Rolling three-year tracking differences provide a more stable picture than single-year figures.

Tracking error can be artificially low for a fund that consistently underperforms its index by a stable amount—the error is low because the gap does not fluctuate, but the investor is still losing ground to the benchmark each year. Neither metric on its own gives a complete picture; both should be reviewed alongside the TER and the fund's securities lending disclosure.

Note that ETF tracking error—which measures the deviation between an ETF and its index—is conceptually distinct from portfolio-level tracking error, which measures the deviation between a portfolio and its benchmark. The same term is used for both, but the context determines the reference. This article covers ETF-level tracking error only.

ETF tracking error in pfolio

When selecting ETFs for a pfolio portfolio, tracking difference and tracking error provide useful quality checks alongside the TER. Well-tracked ETFs impose lower implicit costs through replication drag and contribute more predictably to the portfolio's intended exposure. ETFs available in pfolio are listed in the Assets section. For portfolio-level performance relative to a benchmark, pfolio's analysis tools are available at Portfolios.

Related articles

• ETFs explained: what exchange-traded funds are and how they span every asset class
• Total expense ratio: how ETF costs affect long-term portfolio returns
• Physical vs synthetic ETFs: how replication methods affect cost and risk
• Volatility in investing: how to measure and manage portfolio risk