Annual return (CAGR)
The steady yearly growth that would end at the same value: the compound annual growth rate.
3 min readLesson 5 of 57Interactive
In short
It makes periods of different lengths comparable. +50% over 5 years is about +8.4% a year, because each year grows on the last. Real years are never that smooth.
Why it matters
A total return on its own can't tell you whether an investment was any good, because it doesn't say how long it took. +60% is excellent over three years and disappointing over twenty. To compare results from periods of different lengths, you need to turn them into a yearly rate.
The compound annual growth rate, or CAGR, does that. It's the single steady yearly return that would have taken you from the starting value to the ending value in the same time. It's how funds, indexes and savings accounts are usually compared, and it's the number to hold up against your goals.
How it works
Divide the ending value by the starting value to get the growth factor. Then take the root for the number of years, which is the same as raising it to the power of 1 ÷ years, and subtract 1. The root undoes the Compounding: it finds the yearly rate that, multiplied by itself once for each year, gives the whole growth.
MarketLab measures years by the calendar: the time from the first to the last day, divided by 365.25 days. So a backtest that runs for 18 months uses 1.5 years, not a count of trading days.
CAGR is not the same as the average of the yearly returns. Averaging ignores the fact that each year builds on the last, so it flatters bumpy investments. A year of +50% followed by a year of −50% averages 0%, but it leaves you with 75% of your money.
CAGR = (ending value ÷ starting value)^(1 ÷ years) − 1
The steady yearly rate that grows the starting value into the ending value over that many years.
A worked example
$10,000 grew to $16,105 over exactly 5 years. Separately, another investment went +50% and then −50% over 2 years.
- Growth factor: $16,105 ÷ $10,000 = 1.6105, a total return of about +61%.
- Fifth root: 1.6105^(1 ÷ 5) = 1.10, so the CAGR is 10% a year.
- Check: $10,000 × 1.10 × 1.10 × 1.10 × 1.10 × 1.10 ≈ $16,105.
- The bumpy investment: 1.50 × 0.50 = 0.75. Its CAGR is 0.75^(1 ÷ 2) − 1 = 0.866 − 1 ≈ −13.4% a year, even though its yearly returns average 0%.
Result: The first investment grew at 10% a year. The second lost about 13.4% a year, which the simple average of 0% completely hides.
Try it
Watch money compound
$10,000 growing 7.0% a year for 20 years ends at $38,697 with compounding, against $24,000 without.
Final value
$38,697
Total return
+287.0%
Annual return (CAGR)
7.0%
The same every year here
Earned by compounding
$14,697
Growth on earlier growth
How to read it
- Below 0%
- The money shrank on average each year.
- About 2% to 4%
- Roughly what inflation has often been in calm decades, so in real terms little was gained.
- About 6% to 10%
- In the range that broad stock indexes have often delivered over long stretches, before inflation, though some decades were far lower.
- Above 15%
- Exceptional if it lasts many years. Over a short period or in a backtest, it's more likely luck or Overfitting.
In the real world
- Fund fact sheets show annualized returns over 1, 3, 5 and 10 years. Anything longer than a year is a CAGR, which is why you can line up a 10-year record against a 3-year one.
- Companies use CAGR too: annual reports and analysts talk about revenue or earnings growing at, say, 12% a year over five years.
- Financial planners use it to check goals. Doubling your money in 10 years needs about 7.2% a year, because 1.072 to the tenth power is roughly 2. The "rule of 72" (72 ÷ rate ≈ years to double) comes from the same maths.
Common mistakes
- Calling CAGR the return you got each year. It's a smoothed figure: real years might have been +30%, −15% and +12%.
- Annualizing a short period. +10% in three months works out to roughly +46% a year, but nobody should expect that to continue.
- Using an average of yearly returns instead. The average is always at least as high as the CAGR, and much higher when returns swing a lot.
- Forgetting the start and end dates. Starting at a market low or ending at a peak can make the same investment look far better.
See it in MarketLab
Test yourself
Optional: 4 quick questions on the words, using the idea, and how it plays out in real markets.
Loading your answers…