How Does Compounding Actually Work?

The short answer. Compounding is earning returns on your previous returns, so growth accelerates over time rather than accumulating in a straight line. £10,000 growing at 7% a year becomes about £19,700 after ten years, £38,700 after twenty and £76,100 after thirty. Almost all of that final figure is growth on growth, not the original sum — which is why time in the market matters more than the rate.

Why is compounding described as exponential?

Because each year's return is calculated on a larger base. In year one, 7% of £10,000 is £700. In year thirty, 7% is over £4,900 — the same rate, seven times the money, because it is being applied to everything the previous years produced.

The last decade of a thirty-year run adds more in absolute terms than the first two combined. This is the whole argument for starting early, and it is also why interrupting a long run is so expensive.

What destroys compounding?

1. Fees. A 1% annual charge does not cost 1%. Over thirty years it removes roughly a quarter of the final sum, because every pound taken is a pound that never compounds.
2. Withdrawals. Taking money out resets the base.
3. Large losses. A 50% fall requires a 100% gain to recover. Drawdowns are asymmetric, which is why avoiding catastrophe matters more than catching every rally.
4. Tax on the way through. Sheltering returns inside an ISA or pension keeps the compounding base intact.

Does compounding work with shares, which do not pay a fixed rate?

Yes, but unevenly, and the unevenness matters. Reinvested dividends compound directly. Retained earnings compound inside the business when management reinvests them at good returns — which is the actual mechanism behind long-term equity returns, and the reason return on capital is such a heavily watched metric.

What is different from a savings account is sequence. Shares do not deliver 7% every year; they deliver −20%, +30%, +5%, −8%. The arithmetic average and the compounded average are not the same number, and volatility drags the second below the first.

What rate should you assume?

Lower than the historical average, if you want to be safe rather than encouraged. Long-run global equity returns after inflation have been in the region of 5% a year. Planning on 10% because a recent decade delivered it is how people end up short.

What this cannot tell you

Real returns are not smooth, and the arithmetic above assumes they are. A sequence of poor years early in a withdrawal phase can produce a very different outcome from the same average return delivered in a different order.

Educational information, not financial advice. Every figure on this page is read from the source filings when the page loads rather than written into the article, so what you are reading is today's data and not a snapshot of the day it was published.