11.4 Measures of Returns
Key Takeaways
- Absolute return is the simple percentage change and ignores the holding period entirely.
- CAGR annualises a return over a period and assumes a single investment and a single redemption.
- XIRR is the correct measure where cash flows occur at multiple dates, such as an SIP.
- Rolling returns show consistency across many overlapping periods and are more honest than trailing returns.
- Annualising a return earned over less than a year is prohibited and produces misleading figures.
Absolute or Point-to-Point Return
The simple percentage change in value, ignoring how long it took.
Absolute Return = (Ending Value - Beginning Value) / Beginning Value x 100
Example. INR 2,00,000 grows to INR 2,54,000.
Absolute return = (2,54,000 - 2,00,000) / 2,00,000 = 27.00%
That 27% is meaningful over 18 months and unimpressive over nine years. Absolute return is therefore used only for periods of less than one year, where annualising would be misleading.
Compound Annual Growth Rate
CAGR expresses the return as an annual rate, accounting for compounding.
CAGR = [ (Ending Value / Beginning Value) ^ (1 / n) ] - 1
where n is the number of years.
Example. INR 2,00,000 grows to INR 3,60,000 over 5 years.
Ratio = 3,60,000 / 2,00,000 = 1.80
CAGR = 1.80 ^ (1/5) - 1
= 1.1247 - 1
= 12.47% per annum
Note the arithmetic trap: the absolute return is 80%, and dividing by five gives 16%, which is wrong. Simple division ignores compounding and overstates the annual rate. CAGR of 12.47% is the correct figure.
CAGR's limitation is decisive: it assumes one investment at the start and one redemption at the end. It cannot handle intermediate cash flows, which makes it the wrong tool for any systematic investment.
Extended Internal Rate of Return
XIRR computes the annualised return where cash flows occur on multiple, irregular dates. It finds the single annual rate at which all cash flows, discounted to their own dates, equal the final value.
This is the correct measure for an SIP, because each instalment has been invested for a different length of time.
Why CAGR fails on an SIP. An investor contributes INR 10,000 monthly for three years — INR 3,60,000 in total — and the value is INR 4,50,000.
- Treating this as INR 3,60,000 invested for three years growing to INR 4,50,000 gives a CAGR of about 7.7%.
- That is wrong, because the final instalment was invested for one month, not three years. The average money-weighted holding period is roughly half the total period.
- XIRR gives the correct annualised figure, which in this case is materially higher than 7.7%.
A distributor computing SIP returns with CAGR understates the investor's actual return, sometimes substantially. Spreadsheet XIRR functions accept the dated cash flow list directly, and AMC and RTA statements report SIP returns on an XIRR basis.
Trailing Returns
Trailing or point-to-point returns measure from a date in the past to today — the familiar one-year, three-year and five-year figures.
Their weakness is end-point sensitivity. A five-year trailing return measured just after a rally looks excellent; the same scheme measured a month later after a correction can look ordinary. Nothing about the scheme changed.
Rolling Returns
Rolling returns compute the return over a fixed window across many overlapping starting points — for example every three-year period beginning on each business day over the last decade.
The output is a distribution rather than a single number, showing the best, worst, average and consistency of three-year outcomes. This is far more informative than one trailing figure:
| Measure | What it tells you |
|---|---|
| Trailing 3-year return | The outcome for one specific investor who happened to start on one specific day |
| Rolling 3-year returns | The range of outcomes across every starting day, and how often the scheme delivered above a threshold |
A scheme with a strong trailing return but a wide rolling distribution has been inconsistent; the trailing figure simply caught a favourable window. Rolling returns are voluntary disclosure, and their presence in a fact sheet is itself a small signal about the fund house's candour.
Annualised versus Absolute: the Rule
Returns for periods of less than one year must be shown in absolute terms and must not be annualised.
Annualising a short period compounds a short-run outcome into a fictional annual figure. A scheme returning 8% in three months would be presented as roughly 36% annualised, which implies a sustainability that three months of data cannot support. SEBI's norms, covered in the next section, prohibit this.
Choosing the Right Measure
| Situation | Correct measure |
|---|---|
| Single investment, single redemption, over a year | CAGR |
| Single investment, held under a year | Absolute return |
| SIP, STP, SWP or any multiple cash flows | XIRR |
| Assessing consistency of a scheme | Rolling returns |
| Comparing a scheme to its benchmark | Same measure and same period for both |
The last row matters as much as the others: a comparison is only valid when both series use the same measure over the same period, computed on a total return basis for the benchmark — which is the subject of the next chapter.
An investment of INR 5,00,000 grows to INR 8,00,000 over 4 years. What is the approximate CAGR?
Why is CAGR the wrong measure for the return on a three-year systematic investment plan?
What advantage do rolling returns have over trailing returns?