Tracking error and standard deviation in passive funds

Both are standard deviations. Both are quoted as annualised percentages. They measure completely different things, and confusing them is the most common error in reading a passive fund.

Standard deviation measures how much the fund moves. Tracking error measures how much it moves differently from its index.

Standard deviation: absolute volatility

The annualised standard deviation of the fund’s own returns — on this site, computed over a trailing three years of daily returns.

It answers: how bumpy was the ride? A fund with 18% standard deviation had returns dispersed widely around its own average; one at 9% was calmer.

Two limits worth holding on to:

• It is symmetric. A 4% up day and a 4% down day contribute identically, which is not how anyone experiences them. That asymmetry is why Sortino exists.
• It is not the same as risk. A fund can be perfectly placid while quietly concentrating into one sector or into weak credit. Placid is not safe; it is only placid so far.

For an index fund, standard deviation is essentially the index’s volatility. That is the point — you signed up for the market’s ride, and a low-volatility index fund would mean it was not tracking.

Tracking error: deviation from the index

The annualised standard deviation of the difference between the fund’s return and its benchmark’s.

It answers: how tightly does it follow? Near zero means it shadows the index almost exactly. Rising values mean it drifts around it.

For a passive fund, low tracking error is the definition of doing the job well. For an active fund, it is a measure of how much the manager is betting against the index — and it becomes the denominator of the information ratio, which prices whether that betting paid.

The distinction that actually costs money

Tracking error is not the same as tracking difference, and this is where index-fund investors lose real return.

• Tracking difference — how far the fund’s return has lagged the index over a period. Mostly the expense ratio, plus cash drag from money waiting to be invested, plus the costs of index rebalancing.
• Tracking error — how consistently it deviates, day to day.

A fund can have very low tracking error and a persistent tracking difference: faithfully lagging by the same 0.4% every year. It tracks beautifully, and it costs you 0.4% annually forever.

Tracking difference is what shows up in your returns. When comparing two index funds on the same index, that is the number that decides, not tracking error.

What causes each

Tracking error rises with:

• Cash holdings for redemptions, which do not move with the index.
• Rebalancing lag when the index reconstitutes.
• Sampling — some funds hold a representative subset rather than every constituent.
• Flows. Large inflows or outflows force trading at prices the index does not pay.
• Securities lending and corporate action timing.

Standard deviation, by contrast, is driven almost entirely by what the fund holds.

How to use them

Choosing an index fund: compare tracking difference first (it is your actual cost), then tracking error, then the expense ratio. Check R² is essentially 1.00 against the correct index. Also check the fund is large enough and liquid enough that flows do not distort it.

Assessing an active fund: a very low tracking error against the benchmark, combined with a beta near 1 and negligible alpha, is a closet index fund — index performance at active fees. Meanwhile a high tracking error is neither good nor bad on its own; it says the manager is taking active risk, and the information ratio says whether it worked.

Pitfalls to avoid

• Using the words interchangeably. They are different measures of different things.
• Assuming low tracking error means low cost. It means consistent tracking, which is compatible with consistently lagging.
• Comparing tracking error across different indices. A Nifty 50 fund and a smallcap index fund face different replication difficulty.
• Comparing standard deviation across categories. A liquid fund and a small-cap fund are not on the same scale.
• Reading either over one year. Both are window-dependent; include a stressed period.

Key takeaway

Standard deviation is how much it moves; tracking error is how much it moves differently from the index. For a passive fund, the number that actually reaches your returns is neither of those — it is tracking difference, the cumulative lag. Check that first, use tracking error to judge replication quality, and use standard deviation to know what ride you signed up for. Formulas and windows are on the methodology page.