Tracking Error Explained: How Closely Funds Track an Index

Tracking error is the standard deviation of the difference between a portfolio’s returns and the returns of its benchmark. It measures how consistently a fund follows its index, not how much it lags on average. An index fund with a tracking error close to zero moves almost in lockstep with its benchmark every period. A fund with a high tracking error wanders, sometimes ahead and sometimes behind, even if it ends up in roughly the same place over time.

The number matters most for passive investors, who are buying a fund precisely because they want the index’s return and nothing else. It also shows up for active managers, where it works as a measure of how much risk they take relative to the benchmark.

How tracking error is calculated

Start with active return: the portfolio’s return minus the benchmark’s return for each period. If a fund returned 1.10% in a month and its index returned 1.15%, the active return for that month was -0.05%.

Collect those active returns over many periods, usually months, and take their standard deviation. That is the tracking error for the sampling period. To annualize a tracking error built from monthly data, multiply by the square root of 12. For weekly data, multiply by the square root of 52.

A simple illustration with four months of active returns:

Month	Fund return	Index return	Active return
1	2.00%	2.05%	-0.05%
2	-1.20%	-1.18%	-0.02%
3	0.80%	0.86%	-0.06%
4	1.50%	1.53%	-0.03%

The fund lagged every month, but by a similar amount each time. The standard deviation of those active returns is small, so the tracking error is low even though the fund always trailed. That distinction is why tracking error needs a companion metric.

Tracking error vs tracking difference

Tracking difference is the average gap between fund and benchmark returns over a period: the fund’s cumulative return minus the index’s. It tells you how much the fund cost you relative to the index. Tracking error tells you how predictable that cost was.

	Tracking difference	Tracking error
What it measures	Average shortfall or excess vs the index	Variability of that gap
Statistic	Mean (or cumulative difference)	Standard deviation
Typical sign for index funds	Slightly negative	Always positive
Main driver	Fees and costs	Sampling, cash, timing, taxes
Question it answers	”How much did I lose to the index?"	"How reliably does this fund track?”

A good index fund has both a small tracking difference and a small tracking error. A fund with a low tracking difference but a high tracking error got lucky in some periods and unlucky in others. A fund with a steady negative tracking difference and a low tracking error is simply charging its fee in a very predictable way, which is usually what you want.

What causes tracking error

Even a fund that tries to own exactly what the index owns cannot match it perfectly. The index is a calculation on paper; the fund has to trade real securities with real money.

• Fees. The expense ratio is deducted gradually from fund assets, which mostly shows up as tracking difference. Because it is steady, it adds little tracking error on its own.
• Sampling and optimization. Broad indexes can contain thousands of securities, some of them illiquid. Many funds hold a representative sample instead of every constituent. The sample behaves like the index on average but not exactly.
• Cash drag. Funds hold some cash for redemptions and receive dividends before they can reinvest them. In a rising market that cash holds the fund back; in a falling market it cushions it.
• Rebalancing and reconstitution. When an index adds or removes securities, the index changes instantly at a stated price. The fund has to trade, paying spreads and sometimes moving the price against itself.
• Withholding taxes. Funds that hold foreign stocks may lose part of each dividend to withholding taxes that the index methodology treats differently.
• Securities lending. Lending holdings to short sellers earns fees, which can partly offset costs and even push tracking difference positive.
• Pricing and timing. If a fund values holdings at a different time than the index, or uses fair-value pricing for foreign markets that closed hours earlier, daily returns can diverge and later reconverge.

These effects explain why two funds tracking the same index can show different numbers, a point worth checking when comparing index funds and ETFs.

Ex-post vs ex-ante tracking error

The calculation above is ex-post, or realized, tracking error: it uses historical returns. Portfolio managers also estimate ex-ante tracking error, a forward-looking figure produced by a risk model. The model compares current holdings with benchmark holdings and uses estimated volatilities and correlations to predict how far apart they could drift.

Ex-ante tracking error is a planning tool. A mandate might say a manager may run up to a certain tracking error budget, and the manager sizes positions so the model’s estimate stays inside it. Ex-post tracking error is how investors check whether that happened.

Tracking error for active managers

For an index fund, lower is better. For an active fund, tracking error is a dial, not a defect. A manager who wants to beat the benchmark has to hold something different from it, and any difference creates tracking error.

That is where the information ratio comes in: annualized active return divided by tracking error. It is a cousin of the Sharpe ratio, but it measures reward per unit of risk taken relative to the benchmark rather than relative to cash. A manager who adds a little return with very low tracking error can have a better information ratio than one who adds more return with a lot of tracking error.

A very low tracking error on an active fund that charges active fees is a warning sign. It can indicate closet indexing: a portfolio that hugs the benchmark while charging as if it were doing something different. The debate over whether active managers can consistently beat their benchmarks after fees sits close to the efficient market hypothesis, and tracking error is one of the tools used to test it.

Limitations to keep in mind

Tracking error depends on the sampling frequency and window. Daily data can exaggerate noise from timing differences that wash out over a month. A short window can miss a stressful period when tracking breaks down.

It also treats deviations above and below the benchmark the same way. A fund that beat its index by a wide but variable margin shows high tracking error, just like one that missed by the same pattern. Always read tracking error alongside tracking difference, and alongside measures like beta when the benchmark itself is in question.

The takeaway

Tracking error is the standard deviation of a fund’s active returns, and it answers how reliably a portfolio follows its benchmark. For passive funds, pair it with tracking difference: you want a small, steady shortfall that roughly equals the fee. For active funds, tracking error measures how far the manager strays from the index, and the information ratio tells you whether that straying was worth it.