The arithmetic of a falling market
Investor behaviour7 min read

By the numbers
- rise needed after −40%
- +66.7%
- the SIP at the bottom
- −17.2%
- the bounce that broke even
- 20.8%
On this page
Every investor knows that markets fall. Very few have done the arithmetic of what a fall costs, which is why the same mistake repeats in every correction: people sell at the bottom, not because they are stupid, but because nobody showed them the table.
Here it is. No forecasts, no view on where the market is going, just the arithmetic of losses and the behaviour that arithmetic rewards.
The asymmetry, in one table
| The fall | The rise needed to get back to where you were |
|---|---|
| −10% | +11.1% |
| −20% | +25.0% |
| −30% | +42.9% |
| −40% | +66.7% |
| −50% | +100.0% |
| −60% | +150.0% |
| −75% | +300.0% |
Read the last three rows slowly. A fund that falls 50% has to double to recover. A fund that falls 75% has to quadruple. There is no cleverness that gets around this; it is the consequence of percentages being calculated on a smaller number after the fall.
The asymmetry runs the other way too, which is why the table is not an argument against equity. A 50% gain is undone by a 33% fall. Gains accumulate on a bigger base than losses subtract from. Over long periods, that is the whole reason equity works.
What a fall does to a lumpsum
Put ₹12,00,000 into an index at 100 the day before a fall to 60. At the bottom you have ₹7,20,000 — a loss of ₹4,80,000, exactly 40%. When the index climbs back to 100, you have ₹12,00,000 again and no gain. The fall cost you nothing permanent, but it cost you time: the years spent getting back to zero are years your money did not compound.
That is the lumpsum experience: honest, symmetrical, and entirely dependent on when you arrived.
What a fall does to a SIP
Now the same ₹12,00,000, but invested ₹10,000 a month through a fund that falls from 100 to 60 over twelve months, then recovers to 100 over the following fifteen.
| ₹10,000 a month through the fall | |
|---|---|
| Invested by the bottom | ₹1,20,000 |
| Units bought | 1,656 |
| Value at the bottom (index 60) | ₹99,366 |
| Loss at the bottom | −17.2% |
| Average cost per unit | ₹72.46 |
Two numbers in that table do the arguing. The index fell 40%; the SIP fell 17.2%. And the average cost per unit was ₹72.46 — not ₹100, and not ₹60.
Why? Because the SIP did not own the whole ₹1,20,000 at the top. It owned ₹10,000. The rest arrived on the way down, at 95, 90, 85, 75, 65, 60 — buying more units with each instalment as the price fell. By the bottom the SIP held 1,656 units, and its break-even price had fallen from ₹100 to ₹72.46.
The number that changes everything
That break-even price is the most under-appreciated number in investing. The SIP does not need the index to get back to 100. It needs the index to get back to ₹72.46.
From 60, that is a rise of 20.8%.
The index needs a 66.7% rise to break even. The SIP needs 20.8%. Same fall, same fund, same dates — a quarter of the recovery.
This is not a trick of the illustration. It is what a SIP does mechanically: it converts a fall into a lower average cost. The investor who understands this waits for a 21% bounce. The investor who does not is still waiting for a 67% one, which is why so many people describe a market that has recovered as still broken.
Stopping at the bottom
Suppose the investor stops the SIP at the bottom — the instinct that feels like protecting capital — and holds what they have. When the index returns to 100, their 1,656 units are worth ₹1,65,610: a gain of 38% on the ₹1,20,000 invested.
Suppose instead they keep the SIP running through the whole recovery. They end with 3,486 units worth ₹3,48,557 on ₹2,70,000 invested: a gain of 29.1%.
The stopped account shows the higher percentage, which is the sort of detail that makes this argument slippery. It ends with ₹1,82,947 less. Percentage flatters small sums; rupees decide retirements.
A fairer comparison: the continuing investor put ₹1,50,000 more in. Had the stopped investor parked those fifteen instalments in a fixed deposit at 6%, the fund account plus the deposit would be worth about ₹3,20,975 — still roughly ₹28,000 behind, with the gap having narrowed. The continuing SIP wins, but by less than the headline. What it really bought was units, and the units are what the recovery paid for.
Now the worst version. The investor sells at the bottom, locks in the ₹20,634 loss, and waits for "confirmation" before buying back. To re-enter at 100 they buy 993 units instead of the 1,656 they already held. The loss is now permanent, and it cost them the recovery they were waiting for.
Choppy markets: the drag nobody quotes
Falls are easy to see. The quiet erosion of a choppy market is not.
| Year | Return | Value of ₹1,00,000 |
|---|---|---|
| 1 | +20% | ₹1,20,000 |
| 2 | −20% | ₹96,000 |
| 3 | +20% | ₹1,15,200 |
| 4 | −20% | ₹92,160 |
Four years, an average annual return of exactly 0%, and a loss of 7.8%. Repeat that two-year block five times — ten years — and you are 18.5% down.
This is volatility drag: the gap between the average of the returns and the return you actually experience. It is why a fund's "average annual return" is a weaker number than its compounded return, and why a smoother path to the same average is worth real money.
How long recovery takes
The rise needed is certain. The time it takes is not, because it depends on the return the market delivers afterwards. In years, at three growth rates:
| The fall | 8% a year | 10% a year | 12% a year |
|---|---|---|---|
| −20% | 2.9 | 2.3 | 2.0 |
| −30% | 4.6 | 3.7 | 3.1 |
| −40% | 6.6 | 5.4 | 4.5 |
| −50% | 9.0 | 7.3 | 6.1 |
| −60% | 11.9 | 9.6 | 8.1 |
None of these columns allows for a second fall on the way, which is the normal case rather than the exception. A 50% fall at 8% a year is a nine-year recovery — and nine years is a large fraction of a retirement.
That is the honest case against a high equity allocation late in life, and it has nothing to do with market timing. If the money is needed in year five, a 50% fall is not a paper loss; it is a shortfall.
What this means before the fall
Four things, in the order they matter.
Decide your horizon before the fall, not during it. The arithmetic only works if you can leave the money alone long enough. An emergency fund is what makes that possible; without one, the first job loss forces the sale at the bottom.
Automate the instalment and then do not touch it. The SIP's advantage comes from buying through the fall, and that only happens if the instruction is running when the fall is happening. Every study of investor returns finds the same gap between fund returns and investor returns, and it comes from this.
Do not check the portfolio more often than you act. A monthly statement is enough. A daily one trains you to react to noise, and the reaction is almost always the expensive one.
Write the plan down while you are calm. One page: the goal, the horizon, the monthly amount, and the sentence "I will not sell because the market fell". The version of you who wrote it is more reliable than the version reading a red screen at 11pm.
The last word
The arithmetic of a fall is not a reason to avoid equity. It is the reason to enter it monthly, for a long time, with money you do not need soon — and then to leave it alone exactly when leaving it alone feels worst.
A 40% fall does not destroy a good plan. It tests whether the plan was good. The units bought at the bottom are the reward for passing the test, and they are only bought by people who are still buying.
Questions this raises
- How long does it take to recover from a 50% fall?
- At 8% a year, about nine years. At 10%, about seven and a half. At 12%, about six. That is the arithmetic alone, with no further fall in between — which is why the time horizon you choose before a fall matters more than the fund you choose.
- Should I stop my SIP when the market falls?
- The arithmetic says no. A SIP that continues through a fall owns units bought at the low prices, so it needs a much smaller recovery to break even than the index does. Stopping converts a paper loss into a permanent one and gives up the cheap units that would have repaired it.
- Why is my SIP down less than the index?
- Because you did not invest all your money at the top. Your average cost is the total invested divided by the total units bought, and the later instalments bought at lower prices. In the example here, an index down 40% left the SIP down 17.2% — the fall was real, but the average rupee entered well below the peak.
- Is a fall a buying opportunity?
- Only if the money was already meant to be invested for a long time. The arithmetic of recovery favours buyers, but it says nothing about whether the fall is over — a 20% fall needs a 25% rise, while a further 20% fall needs 56%. If your horizon is ten years, buying monthly through it is defensible. If your horizon is two, the arithmetic is not on your side.
- What is volatility drag?
- The gap between the average of yearly returns and the compounded return you actually experience. A fund that returns +20%, −20%, +20%, −20% has an average annual return of zero and a cumulative return of −7.8%, because each percentage is calculated on a smaller number after the loss. The more volatile the path, the wider the gap.