ETF Tracking Error vs Tracking Difference: How Closely Your Index Fund Follows

An index fund or ETF makes one promise: give you the index’s return, minus a small fee. That’s it. So the only real question when you compare two Nifty 50 funds is how well each one keeps that promise, and that is what ETF tracking error and its sibling, tracking difference, are built to answer. Most investors look at the expense ratio and stop there. We think that’s half the job.

The two numbers sound alike, and fund factsheets often print them side by side, so they get mixed up all the time. They measure different things. One tells you how much return you lost to the index. The other tells you how wobbly the fund’s path was around the index. A fund can look good on one and poor on the other, and the worked examples below show exactly how.

What ETF tracking error actually measures

Tracking error is a measure of consistency. Investopedia defines it as the standard deviation of the difference between a portfolio’s returns and its benchmark’s returns. In plain words: take the fund’s return each day, subtract the index’s return that day, and see how much those daily gaps jump around.

In India, SEBI has pinned the definition down for passive funds. Under the SEBI Master Circular for Mutual Funds dated March 20, 2026 (paragraph 4.5.4), tracking error is the annualised standard deviation of the difference in daily returns between the underlying index (or goods, in the case of gold and silver ETFs) and the fund’s NAV, based on the past one year of rolling data. For a fund that is less than a year old, it is worked out on whatever data exists.

Notice what tracking error does not tell you. It says nothing about whether the fund ended up ahead or behind. A fund that trails the index by exactly the same tiny amount every single day has a tracking error close to zero, even though it is steadily losing ground.

Tracking difference: the number that hits your returns

Tracking difference is the other half. The same SEBI master circular defines it as the annualised difference of daily returns between the index (or goods) and the NAV of the ETF or index fund. Put simply, it is the gap in return over a period. If the index gave 12% over a year and the fund gave 11.6%, the tracking difference is roughly 0.4 percentage points against you.

This is the cost you actually feel. Expense ratio is a big part of it, because the index has no fees and the fund does. But it is not the whole story. Cash sitting uninvested, the cost of buying and selling when the index changes, and the gap between when dividends are earned and reinvested all show up here too.

Tracking difference tells you how much you lost to the index. Tracking error tells you how unpredictable that loss was. You want both small, but the first one is the one that comes out of your pocket.

What SEBI requires funds to disclose

The rules here came in through SEBI’s circular on development of passive funds dated May 23, 2022, and they now sit in the 2026 master circular. The main points, as the master circular states them:

• Daily tracking error disclosure. All ETFs and index funds, including debt ones, must disclose tracking error on a daily basis on the websites of the AMC and AMFI.
• Monthly tracking difference disclosure. Tracking difference must be disclosed on the AMC and AMFI websites every month, for periods of 1 year, 3 years, 5 years, 10 years and since the date of allotment of units.
• A 2% cap on tracking error. For ETFs and index funds other than debt ones, tracking error should not exceed 2%. If unavoidable force majeure events beyond the AMC’s control push it past 2%, the matter has to be brought to the trustees’ notice along with any corrective action.
• A 1.25% cap on tracking difference for debt funds. For debt ETFs and index funds, the annualised tracking difference averaged over one year should not exceed 1.25%, with the same trustee-reporting requirement if it does.

Mint’s report on the 2022 circular summarised the same two limits when they were announced. The practical upshot for us as investors is simple: the numbers are public, updated often, and comparable across fund houses. There is no excuse for picking a passive fund on the expense ratio alone.

One thing to keep in mind. The 2% cap is a ceiling, not a target. A large Nifty 50 fund sitting anywhere near that cap would be a red flag, not a pass.

Worked example 1: computing tracking error from daily returns

Let’s do the arithmetic once by hand so the number stops being a black box. The fund and returns below are made up for illustration. Here are five trading days for an imaginary index and an imaginary fund that tracks it:

Step one: the average daily difference is (-0.02 – 0.03 – 0.05 + 0.01 – 0.03) ÷ 5 = -0.024 percentage points. So on a typical day in this sample, the fund lagged by a shade over two basis points.

Step two: the standard deviation of those five differences (using the sample formula) works out to about 0.0219 percentage points per day. That is the daily tracking error.

Step three: annualise it. SEBI’s definition asks for an annualised figure. A common convention is to multiply a daily standard deviation by the square root of the number of trading days in a year; using 252 days for this illustration, 0.0219 × √252 ≈ 0.35%. Each AMC follows its own disclosed methodology, so treat this step as a demonstration of the idea rather than an official formula.

Step four: look at what you lost. Compounding the five index returns gives about +1.600% for the week; compounding the fund’s returns gives about +1.479%. The fund is behind by roughly 0.12 percentage points over five days. That gap is the tracking difference for the period, and it is what tells you the fund is running consistently a bit slow.

A real calculation uses about a year of daily data, not five days, but the steps are the same.

Worked example 2: why you need both numbers

Now take two imaginary Nifty 50 index funds over the same year. Again, these are illustrations, not real funds.

• Fund A: the index returned 12.0%, Fund A returned 11.6%. Tracking difference about -0.40 percentage points. Tracking error a very low 0.10%.
• Fund B: the index returned 12.0%, Fund B returned 11.9%. Tracking difference about -0.10 percentage points. Tracking error a much higher 0.60%.

Fund A is beautifully steady but expensive: it hugs the index day by day while losing about 40 basis points a year, most likely to fees. Fund B kept more of the index’s return, but its daily path wandered more. Maybe it held more cash at times, or bought index changes at awkward prices.

Which is better? For a long-term SIP investor, Fund B’s smaller tracking difference is usually worth more, as long as its tracking error is not a sign of something sloppy that could flip into a big negative year. That is why the multi-period tracking difference disclosures matter. If Fund B shows a small gap over 1, 3 and 5 years, the higher wobble was noise. If the gap swings wildly from one period to the next, be careful.

In rupee terms the gap is easy to see. On ₹5 lakh, a 0.30 percentage-point difference in annual return is about ₹1,500 in the first year alone, and that difference compounds every year you stay invested.

A low expense ratio is a promise. A low tracking difference over three and five years is proof that the promise was kept.

What pushes an index fund off track

Investopedia lists the usual culprits, and they apply to Indian funds as much as anywhere:

• Fees. The index pays no expenses, the fund does. Investopedia calls expense ratios the most prominent cause of tracking error.
• Cash drag. Indexes hold no cash. Funds receive dividends and fresh money, and there is a lag before it is invested.
• Index changes. When stocks enter or leave an index, the fund must trade, which costs money, and it may not get the same prices the index used.
• Sampling. When an index has thinly traded stocks, a fund may hold a representative sample instead of every stock, which adds deviation.
• Volatility and liquidity. Illiquid holdings with wide bid-ask spreads make tracking harder. Investopedia notes that sector, international and dividend ETFs tend to show higher tracking error than broad equity and bond ETFs.

For Indian investors, this means a sector or smaller-cap index fund will normally carry a higher tracking error than a large Nifty 50 fund. Compare like with like.

The ETF-only problem: market price vs NAV

Here is a trap that tracking error does not catch. SEBI’s tracking numbers are calculated on NAV. But when you buy or sell an ETF on the exchange, you pay the market price, which can sit above NAV (a premium) or below it (a discount).

Say an ETF’s day-end NAV is ₹250.00 and it closes on the exchange at ₹247.20. That is a discount of ₹2.80, or 1.12% of NAV. If you sell at that price, you take a hit that never shows up in the fund’s tracking error. Investopedia notes that premiums or discounts as high as 5% have been seen, particularly in thinly traded ETFs.

SEBI has a safety valve for this. Under paragraph 4.5.3 of the master circular, investors can go straight to the AMC to redeem ETF units, for transactions up to ₹25 crore without exit load, if the ETF’s closing price has been at a discount of more than 1% to day-end NAV for 7 continuous trading days, if there have been no quotes on the exchange for 3 consecutive trading days, or if the total bid size has averaged less than half the creation unit size daily over 7 consecutive trading days. Most retail investors will never need it, but it is good to know it exists.

The practical habit: check the ETF’s indicative NAV and its bid-ask spread before you place the order, and prefer limit orders over market orders on thinly traded ETFs.

A quick checklist before you pick a passive fund

1. Shortlist funds tracking the same index, so the comparison is fair.
2. Pull the tracking difference for 1, 3 and 5 years from the AMC or AMFI website. Smaller is better, and consistency across periods matters.
3. Check the tracking error. For a broad large-cap index, it should be well below SEBI’s 2% ceiling.
4. Compare expense ratios, but use them to explain the tracking difference, not to replace it.
5. For ETFs, look at trading volume and spreads, and how far the market price tends to sit from NAV.
6. Remember that the fund tracks the index’s risk too. A tight-tracking fund still falls when the index falls. Our guide to risk management in trading covers how to size positions around that.

If you want to compare an active fund against its benchmark instead, the logic is related but the question changes, and our explainer on the information ratio picks it up from there. For specific passive products, see how the same tracking ideas apply to a gold ETF in India.

The short version

Tracking difference is the return you gave up. Tracking error is how erratic that gap was. SEBI makes funds publish both, tracking error daily and tracking difference monthly, and caps tracking error at 2% for non-debt index funds and ETFs. When two funds follow the same index, the one with the smaller and steadier tracking difference over several years has usually done its job better, whatever its expense ratio says. And if it is an ETF, remember the price you trade at matters as much as the NAV.

Disclaimer: This article is for education and information only and is not investment advice or a recommendation to buy or sell any security or fund. The funds, returns and prices in the worked examples are fictional illustrations. Please read the scheme documents and do your own research or consult a SEBI-registered investment adviser before making investment decisions.