Compounding: why your brain can't feel it
You can understand compounding perfectly and still not believe it. That gap is the single most expensive thing in personal finance.
7 min read
Compounding (returns earning their own returns) has a one-sentence definition. Your growth earns growth. Year two grows the money you started with plus everything year one added.
Everyone reading that understands it. Almost nobody trusts it enough to act on it. This lesson is about that gap, because the gap is where the money goes.
Your brain draws straight lines
Asked to extend a pattern, people extend it linearly. It is the default and it is usually right. A car at 100km/h covers 200km in two hours. Twice the hours, twice the distance. A job paying $40 an hour pays $400 for ten hours. This intuition works for almost everything you encounter.
It is completely wrong for compounding, and it fails in a specific direction: it under-predicts, and the error gets worse the further out you go. Which means the thing your intuition is worst at is precisely the long-term case that matters most.
Exaggerated cases, because they are the only ones you can feel
The doubling cent
One cent, doubling every day for 30 days. Guess the end figure before reading on. Almost everyone guesses somewhere in the thousands.
| Day | Amount |
|---|---|
| Day 1 | ? |
| Day 10 | ? |
| Day 20 | ? |
| Day 25 | ? |
| Day 28 | ? |
| Day 30 | ? |
Look at what happens between day 20 and day 30. Two thirds of the time has produced $5,242. The final third produces $5.36 million. More than half the total arrives in the last two days.
That shape is the entire lesson. Compounding is not a gentle upward slope. It is nearly flat for a long time and then it is not, and the flat part is where people conclude it is not working and stop.
The chessboard
One grain of rice on the first square, two on the second, doubling across all 64 (a little over 18.4 quintillion grains, 18,446,744,073,709,551,615, on the last square alone). The total exceeds global rice production for centuries. The first half of the board is a manageable pile. The second half is not describable in ordinary terms. Same shape again.
The real version
Take Amina. She invests $12,000 a year at an assumed 10%.
| After | She has contributed | She has | Growth as a share of the total |
|---|---|---|---|
| 5 years | $60,000 | $80,587 | 26% |
| 10 years | $120,000 | $210,374 | 43% |
| 15 years | $180,000 | $419,397 | 57% |
| 20 years | $240,000 | $756,030 | 68% |
| 30 years | $360,000 | $2,171,321 | 83% |
At five years she has put in $60,000 and has $80,587. Respectable, unremarkable, and exactly the point at which people decide investing is overrated. At thirty years, more than four out of every five dollars she holds are dollars she never earned and never deposited. They were produced by the money she deposited earlier.
Compare the two halves. Years 1 to 15 take her to roughly $419,000. Years 16 to 30 add about $1.75 million. Same person, same contribution, same return. The second half does more than four times the work of the first, and the only ingredient it had that the first half did not is that the earlier years already happened.
The shape of that is worth holding onto. Her contributions go in as a straight line, the same amount every year, year after year. Her balance does not move in a straight line at all: it tracks her contributions closely for the first several years, then separates from them, and the gap between the two lines keeps widening for as long as she leaves it alone. Everything in that gap is money nobody put in.
Run it yourself
What this does not settle
A fair objection: if a higher return compounds harder, is chasing the highest return the whole game? If one person assumes 10% and another believes they can get 20%, then on this lesson's logic the second person wins eventually.
The real numbers are more interesting than either answer, and they turn on how much of the higher return is real and how long the early years last. Savings rate vs return rate, the next module's opening lesson, runs that comparison with real figures, and it is worth reading before drawing conclusions from this one.
Check yourself
4 questions on this lesson. Nothing is recorded or sent anywhere.
1One cent doubling daily for 30 days ends at about $5.37 million. What is the point of the example?
The flat early stretch is where people conclude compounding is not working and stop, which is precisely when stopping is most expensive.
2After 30 years of $12,000 a year at 10%, roughly what share of Amina's balance is growth rather than her own contributions?
She contributed $360,000 and has $2,171,321. More than four in five dollars were produced by the dollars she deposited earlier.
3Why does human intuition consistently get compounding wrong?
Linear extension is right for almost everything else, which is why the intuition is so persistent. It fails worst on exactly the long horizons that matter most.
4Comparing Amina's first 15 years to her second 15, what does the lesson observe?
Roughly $419,000 in the first half against about $1.75 million in the second, on identical contributions and returns.