Glossary term
Tracking Difference
The gap between what an ETF actually returned and what its index returned over the same period, both measured as total return. Tracking difference = ETF return − index return. It captures every cost inside the fund and every income the fund earns on top, so it shows the true price of owning the ETF, which the expense ratio alone does not.
AI-generated — produced automatically by Closelook’s systems under this site’s editorial policy.
What it means
An index fund promises the index. Tracking difference measures how much of that promise arrived. Closelook computes it as the fund’s total return minus the index’s total return over one period: a calendar year, a trailing one-, three- or five-year window, or the fund’s life, annualised. Both sides are total returns. The fund side is the net asset value with every distribution reinvested, after everything that happens inside the fund. The index side is the variant the fund is contracted to follow, usually a net-return index for UCITS funds and a gross or total-return index for many US funds, never the price index.
A negative number means the fund delivered less than the index: that is its real cost. A positive number means the fund delivered more than the index: its income outran its expenses. Some data sites flip the sign so that the figure reads like a fee; check the convention before comparing tables.
What goes into it
On the expense side, everything that costs the fund money: the total expense ratio, trading costs when the index rebalances, withholding tax on foreign dividends, swap fees in synthetic funds, cash drag from distributions not yet reinvested, and sampling error where the fund holds fewer names than the index. On the income side: dividends and interest from the holdings, securities-lending revenue, swap income, and tax treatment more favourable than the index assumes. A net-return index deducts the maximum withholding rate; a fund that reclaims part of it earns back some of its fee.
Why it beats the expense ratio
The expense ratio is a promise about one cost line. Tracking difference is the outcome of all of them. extraETF’s tables show funds on the same index with expense ratios from 0.15% to 0.50% whose three-year tracking differences ranged from beating the index by 0.21% to lagging it by 0.25%. A fund can charge 0.50% and lag by 0.15%, which means it earned back 0.35% through lending and tax work. Two funds on the same index with the same fee can sit a quarter-point apart. The number to read is the difference, over several years, next to the fee.
How Closelook uses it
Tracking difference becomes the headline metric on Closelook’s ETF pages, shown per calendar year and annualised over one, three and five years, with the expense ratio beside it. Where only a price index or only the exchange price is available, no figure is shown rather than a wrong one. The related measure, tracking error, describes how tightly the fund follows day to day, not how much it delivered.
Common questions
- What is the formula for tracking difference?
- Tracking difference = ETF total return − index total return over the same period. The fund side is net asset value with distributions reinvested, after all costs and including lending income; the index side is the total-return variant the fund tracks.
- Is a negative tracking difference bad?
- In Closelook’s convention a negative figure means the fund delivered less than its index, which is its real cost. Most funds are slightly negative because fees exist. Some data sites report the same fact with a positive sign; read the convention before comparing.
- How is tracking difference different from tracking error?
- Tracking difference is the actual gap in return over a period. Tracking error is the standard deviation of that gap, a measure of how steadily the fund follows the index. A fund can have a large tracking difference with a tiny tracking error if it lags by the same amount every day.
- Why can an ETF beat its index despite its fee?
- Securities-lending revenue, withholding-tax reclaims that a net-return index does not assume, and swap income can add more than the expense ratio takes away. Over a full cycle those sources vary, so several years of data matter more than one.