What you will learn
- Separate a signed performance gap from its variability.
- Understand fees as one contributor to a tracking result.
Expense Ratio
Costs accrue daily and are built into the NAV, so reported fund returns are normally already net of them. Fund documents use different labels, such as total expense ratio, ongoing charges figure or, in Hong Kong key facts statements, ongoing charges over a year, and the costs they include can vary.
The ratio is charged every year on the whole balance, so small differences compound over time, and it is a known cost while returns are uncertain. It generally excludes brokerage, spreads, taxes and one-off sales charges, so the total cost of holding a fund can be higher.
Tracking Error
Each period, subtract the benchmark's return from the fund's return to get the active return, then take the standard deviation of those gaps. Annualize the result by multiplying the periodic figure by the square root of the number of periods per year, about √12 for monthly or √252 for daily data.
Tracking error measures how consistent the gap is, while tracking difference is the signed gap itself. A fund that lags its benchmark by a steady amount has near-zero tracking error but a negative tracking difference, and a fund whose gaps swing both ways can show a larger tracking error with a difference near zero.
Tracking Difference
For a passive fund the figure over a year is usually negative, mainly because of the expense ratio and trading costs. It can be partly offset by securities-lending income, and it is also affected by sampling of holdings, cash held for flows, timing differences and the tax treatment of dividends compared with the index's assumptions.
Tracking difference shows how far ahead or behind the fund ended up; tracking error shows how much the gap varied along the way. Compare like with like: total returns, over the same dates and in the same currency, and say whether NAV or market price was used.
Worked example: Steady lag versus changing gaps
Illustrative: Fund A returns exactly 0.10 percentage point less than its benchmark in every month of a fictional sample. The mean monthly difference is −0.10 percentage point, but the standard deviation of the monthly differences is zero. A low tracking error therefore does not mean that the fund matched the benchmark’s return. Fees may contribute to the gap, but are not its only possible cause.
Check your understanding
A fictional fund gains 5% and its benchmark gains 6% over the same period. What is the signed tracking difference?
Show answer and explanation
5% − 6% = −1 percentage point. That single-period difference is not tracking error; tracking error needs a time series of comparable periodic differences.