Compounding — The Mechanism, Without the Sales Pitch

Money grows. If that growth is taken out, the next period of growth applies to the original amount again. If it is left in, the next period applies to the original amount plus what was added, so the base is larger every time.

Repeat that for long enough and the amount being added each period becomes larger than anything you contributed, because it is growth on growth rather than growth on your deposits. That is the entire idea, and it is arithmetic rather than a strategy.

In a mutual fund it happens inside the scheme. Gains are reflected in the value rather than paid out, which is what the growth option does and why our page on growth versus IDCW matters here. Choosing a payout option means taking part of it out, which is a legitimate choice and is not the same holding.

Everything above assumes two things, and both are frequently missing in real households.

Time. The effect is small over a few years and becomes substantial only over long periods, because it depends on repetition. Somebody with a three-year horizon does not have a compounding story, whatever the brochure says, and money for a three-year goal should not be in a market-linked holding at all.

Not interrupting it. Withdrawing along the way resets the base. Stopping instalments during a fall does the same to future contributions. The mechanism is uninterrupted growth, and every interruption removes some of it, as our page on rupee cost averaging discusses.

Which is why the practical advice on this site is so unexciting. The things that protect compounding are having a buffer, keeping the instalment sustainable, and not reacting to bad years.

Every article on this subject contains a figure: this much a month becomes that much in twenty years. We do not publish one anywhere on this site and it is worth explaining why.

That figure is produced by assuming a rate. Change the assumed rate slightly and the answer changes enormously, because the mechanism amplifies the assumption along with everything else. So the number is not a forecast; it is somebody choice of input, presented in rupees.

The harm is that households plan around it. A figure seen once becomes the expected outcome, and when reality differs the investor concludes something went wrong, when in fact nothing was ever promised.

Our SIP calculator will do the arithmetic on any assumption you supply. That is a different thing from us telling you what to expect, and the difference is the whole point.

The same mechanism applies to what you owe, and it is the strongest argument on this site for clearing expensive debt before investing.

An unpaid balance grows on the same principle, with the charge applying to a base that has already grown. On a credit card rolling over, that happens quickly and relentlessly, and it is happening while you are hoping for the other version to work in your favour elsewhere.

Clearing that debt has a certain outcome. No investment offers certainty of any kind, which is why the order matters. Our page on asset allocation puts this at the top for the same reason.

The effect does not arrive evenly, and understanding that prevents a common conclusion at the wrong moment.

In the early years the base is small, so growth on the base is small, and most of what you see in the account is simply the money you put in. It looks like a savings box because in arithmetic terms it is still behaving like one.

The later years are where the character changes, because by then the accumulated growth is large enough that growth on it exceeds the fresh contributions. Nothing switches on. The same mechanism has been running throughout; it only becomes visible once the base is big enough.

Which is why people abandon this in year two or three, when it feels like nothing is happening, and why the households that do well are simply the ones still there in year twelve. Our post on the first year of a SIP covers that stretch.

Cost compounds too, and against you.

An annual charge is deducted every year from a base that is meant to be growing, so the amount taken grows as well. Over long periods a small difference in the annual charge produces a difference in outcome that surprises people, which is why our page on the expense ratio treats a fraction of a percent as worth attention.

The same applies to charges incurred by trading in and out. Somebody switching schemes every eighteen months pays exit load and tax repeatedly on money that never left the market, and each of those is subtracted from the base, as our page on switching between schemes explains.

Four things, and none of them involves selecting anything.

• A buffer, so an ordinary emergency does not become a redemption. See building an emergency fund.
• An instalment you can sustain through a bad year rather than a larger one you abandon.
• Matching money to dates, so nothing has to be withdrawn early.
• Raising the amount as income rises, which our page on the step-up SIP covers and which does more than any selection decision.

Notice that three of those four are about not being forced to sell. That is the connection between this page and our page on risk and volatility: most of what threatens compounding is an interruption rather than a market.

Starting earlier helps, and we would rather say that plainly than illustrate it with a number that pretends to know the future. If you want to work out an amount and a horizon that you can actually hold to, get in touch.