Track 8 · Arithmetic · lesson 1
Compounding, properly
12 min
Track 8 · Arithmetic · lesson 1
12 min
Seventy-two. Divide it by an annual rate and you get, near enough, the number of years until the thing doubles. Nine percent doubles in eight years. Six percent in twelve. Three percent in twenty-four.
It works on savings, on debts, on prices and on the number of users of anything. It is the only piece of financial arithmetic worth carrying around in your head, and the whole of this lesson is what it is hiding.
A balance growing at rate r for n years is multiplied by (1 + r) raised to the power n. That is the entire formula, and everything else in this track is a rearrangement of it.
The doubling rule falls straight out of it. Doubling needs (1 + r) to the n to equal 2, and for the rates people actually meet, n times r comes out at around 0.72. So n is about 72 divided by the rate expressed as a percentage.
It is an approximation. It is a good one between about four and twelve percent and it drifts at the extremes.
Here is where the rule stops being a party trick and starts being useful. Once you know one doubling time, you know all of them, because doublings do not add up — they stack.
Predict
The model below runs the formula directly. Twelve years are fixed. Two controls move: what goes in each year, and the rate.
Compound growth
Hypothetical model, not a forecast
Grows a starting balance plus a fixed yearly contribution at a constant rate.
It assumes
It ignores
Turn the yearly contribution down to nothing, so the starting balance is on its own. Then find the lowest annual return that doubles it inside the twelve years — you want a final balance above 20,000 and below 21,000.
Paid in at the end of each year, so it earns nothing in that year.
| Year | Balance | Total paid in |
|---|---|---|
| 0 | 10k | 10k |
| 1 | 12.3k | 12k |
| 2 | 14.7k | 14k |
| 3 | 17.1k | 16k |
| 4 | 19.6k | 18k |
| 5 | 22.2k | 20k |
| 6 | 24.9k | 22k |
| 7 | 27.6k | 24k |
| 8 | 30.5k | 26k |
| 9 | 33.4k | 28k |
| 10 | 36.4k | 30k |
| 11 | 39.5k | 32k |
| 12 | 42.6k | 34k |
Not there yet — keep moving the controls.
Six percent gets there, and a tenth of a point less does not. The rule was not close — it was exact to the precision the control allows, which is the sort of thing that makes an approximation worth memorising.
Now use it backwards, which is where it earns its keep.
A debt at 24% doubles in about three years if nothing is paid. Prices rising at 2% double in about thirty-six. A fee of 1% a year sounds unrelated to any of this, and a later lesson shows it is the same arithmetic wearing a different sign.
The rule does not care what the number describes. Anything that grows by a percentage of itself is running this formula, including the things you would rather were not.
Check
The exact answer uses the natural logarithm of two, which is about 0.693 — so seventy is the more accurate divisor for very small rates and for anything compounding continuously.
Seventy-two wins on two counts anyway. It is more accurate in the six to ten percent band where most of these questions live, because annual compounding is slightly slower than continuous. And it divides cleanly by 2, 3, 4, 6, 8, 9 and 12, which is the difference between a rule you use in your head and one you write down.