ETF Tracking Difference vs Tracking Error: Measure Index Fund Quality

ETF Tracking Difference vs Tracking Error: Measure Index Fund Quality

বাংলায় পড়ুন

Learning objective

By the end of this tutorial, you should be able to distinguish an ETF's tracking difference from its tracking error, calculate simple versions of both and avoid confusing either measure with the ETF's market-price premium or discount to net asset value.

The distinction in one minute

An index fund tries to follow a benchmark, but an investable portfolio cannot copy a theoretical index perfectly.

Tracking difference measures the fund's actual return minus the benchmark return over a stated period:

Tracking difference = Fund total return − Benchmark total return

Tracking error measures how variable the periodic return gaps are, commonly as the standard deviation of the fund's periodic return minus the benchmark's periodic return:

Tracking error = Standard deviation of periodic active returns

Tracking difference asks, “How far ahead or behind did the fund finish?” Tracking error asks, “How consistently did it follow the benchmark along the way?” A fund can lag by a steady amount and still have very low tracking error.

1. Worked example

Assume an ETF's total return for a calendar year is 9.40% and its benchmark total return is 10.00%.

Its annual tracking difference is:

9.40% − 10.00% = −0.60 percentage points

The fund lagged the benchmark by 0.60 percentage points.

Now suppose its quarterly active-return differences were:

• Quarter 1: −0.10 percentage points
• Quarter 2: −0.20 percentage points
• Quarter 3: −0.10 percentage points
• Quarter 4: −0.20 percentage points

The average quarterly gap is −0.15 percentage points. The population standard deviation of these four gaps is 0.05 percentage points. A simple quarterly-to-annual scaling multiplies by the square root of four:

Annualised tracking error ≈ 0.05 × √4 = 0.10 percentage points

This fund has a negative tracking difference but low tracking error: it lagged fairly consistently.

Imagine a second fund with the same approximate average gap but quarterly active returns of +0.60, −0.90, +0.60 and −0.90 percentage points. Its path is much less consistent, so its tracking error is much higher even if its final annual lag is similar.

Real factsheets may use daily, weekly or monthly returns, sample rather than population standard deviation, and different annualisation conventions. Compare funds only after confirming the methodology.

2. Why an ETF can lag its index

Common sources of tracking difference include:

• Ongoing charges: management and operating costs reduce fund returns, while an index is usually calculated without the fund's expenses.
• Transaction and rebalancing costs: the fund must trade when the index changes.
• Sampling: a fund may hold a representative subset rather than every index constituent.
• Cash drag: subscriptions, redemptions or dividend cash can remain uninvested temporarily.
• Tax and withholding: the fund may receive dividends after deductions that differ from the index assumption.
• Corporate-action timing: treatment of mergers, rights issues and special dividends can differ.
• Index licensing and operational costs: these can affect net return.
• Securities lending: revenue can offset some expenses, but it introduces collateral and counterparty considerations.
• Derivative and collateral effects: synthetic or futures-based replication has financing, roll and counterparty features.

A small positive tracking difference is possible if securities-lending revenue, tax treatment or implementation effects more than offset stated charges. It should not be assumed to persist.

3. Why tracking error rises

Tracking error tends to increase when the fund's return gap changes unpredictably. Possible causes include:

• Partial replication in a broad or illiquid index
• Different trading times across markets
• Currency conversion and valuation timing
• Delayed index rebalancing trades
• Derivative basis and roll effects
• Cash flows and dividend-reinvestment timing
• Portfolio constraints, position limits or unavailable securities
• Market disruption or stale underlying prices

High tracking error does not automatically mean poor absolute performance. It means the fund has followed the benchmark less consistently over the measurement sample.

4. Use the correct benchmark return

Benchmark choice can change the conclusion. Check whether the fund compares itself with:

• A price-return index, which excludes dividends
• A gross total-return index, which reinvests dividends before withholding tax
• A net total-return index, which applies an assumed withholding rate
• A currency-hedged or unhedged index

Comparing a distributing ETF's price chart with a total-return index creates a false gap because cash distributions are missing from the fund series. Use total returns and include reinvested distributions consistently.

5. Tracking metrics are separate from premium or discount

Retail investors trade ETF shares at market prices. The fund also calculates net asset value, or NAV. The market price can sit above NAV, a premium, or below NAV, a discount. Investor.gov explains that authorised participants and the creation/redemption process generally help keep market price close to NAV, but the two can differ.

This market-price gap is not the same as tracking difference:

• Tracking difference: fund NAV total return versus benchmark total return over time
• Premium/discount: ETF market price versus NAV at a point in time
• Bid-ask spread: immediate cost between executable buy and sell prices

An ETF may track its index closely at NAV but still trade at a temporary premium or wide spread during stressed or asynchronous markets.

6. Low tracking error can hide a persistent cost

Consider a fund that trails its benchmark by almost exactly 0.40 percentage points every year. Its return gap is stable, so tracking error can be very low. The cumulative tracking difference, however, compounds against the investor year after year.

This is why ESMA research warns that tracking error measures volatility of the gap rather than the gap itself. Review both measures, the ongoing charge and realised total returns.

7. A practical ETF comparison process

1. Confirm both funds track the same exact index version.
2. Use the same currency and share-class hedging policy.
3. Compare NAV total returns, not an ETF market-price chart against an index.
4. Calculate tracking difference over one, three and five years where available.
5. Compare tracking error using the same observation frequency and annualisation method.
6. Review ongoing charges, withholding tax, replication method and securities-lending policy.
7. Inspect bid-ask spread, volume and premium/discount behaviour separately.
8. Check whether the recent sample includes a market disruption or index rebalance.
9. Consider fund size, closure risk, domicile, tax and platform costs.

Common mistakes

• Using “tracking error” to mean any performance gap.
• Comparing a fund with the wrong index variant.
• Ignoring distributions or using price returns instead of total returns.
• Ranking tracking error figures calculated from different frequencies.
• Assuming the lowest ongoing charge guarantees the smallest tracking difference.
• Mixing NAV performance with exchange market-price performance.
• Applying ordinary index-fund expectations to leveraged or inverse ETPs, whose targets usually reset daily.
• Treating a short performance history as reliable evidence of future replication.

Checklist

• Exact benchmark and return type matched?
• Same currency and hedging policy?
• NAV total returns used?
• Tracking difference measured over multiple periods?
• Tracking-error frequency and annualisation disclosed?
• Charges, tax and replication method reviewed?
• Premium/discount and bid-ask spread checked separately?
• Fund structure and liquidity risks considered?

Sources

• ESMA: Definitions of tracking error and tracking difference
• ESMA: Structural risks in the ETF sector
• Investor.gov: Exchange-traded funds
• Investor.gov: Mutual fund and ETF fees and expenses

This tutorial was prepared with AI-assisted research and writing under The Money Man editorial workflow. It is educational information, not personalised investment advice or an ETF recommendation. The return series is hypothetical and simplified.