When a SIP Beats a Lump Sum, It Is Usually Bad News
We ran both against every stock we track, at every horizon. A lump sum won in about nine cases out of ten over twenty years — and the stocks where the SIP won turn out to be the ones that did badly. That is the finding, and it is the opposite of how SIPs are sold.
The short version
- Over 20 years, a lump sum beat a monthly SIP in 90.2% of the 215 stocks we could measure. The SIP won in 9.8%.
- Over one year the result flips: the SIP won in 58.1% of cases. Short windows are where a SIP has an edge, not long ones.
- The stocks where the SIP won are the ones that did badly. At 5 years their average lump-sum return was -3.0%, against +281.8% where the lump sum won.
- So "my SIP beat a lump sum" is mostly a statement about the stock, not about the strategy — and it is rarely a real choice anyway.
India puts more than ₹30,000 crore a month into SIPs, and the case for them is usually made with the word “averaging” and no numbers at all. We have the numbers. For every stock we track, we ran a monthly investment and the identical total invested once at the start, over every holding period the data supports. The lump sum won most of the time — and the exceptions turn out to be more interesting than the rule.
What we measured
For each stock and each horizon, two backtests on the same price history. The first buys at the close of every month for the whole period. The second takes the same total amount and buys once, at the first of those prices. Both are priced on the adjusted close, so dividends are reinvested in both, and both ignore brokerage and taxes.
A stock only appears at a horizon it has enough history for, which is why the count falls as the period lengthens. These figures are computed live from the same data behind every stock page here, so they move as the market does.
The lump sum won almost every time
| Holding period | Stocks | SIP won | Avg SIP XIRR | Avg lump-sum XIRR |
|---|---|---|---|---|
| 1 year | 484 | 58.1% | 20.87% | 10.65% |
| 3 years | 434 | 16.8% | 11.17% | 17.63% |
| 5 years | 399 | 20.3% | 19.71% | 19.25% |
| 10 years | 318 | 15.1% | 20.69% | 17.54% |
| 15 years | 289 | 12.5% | 20.11% | 18.12% |
| 20 years | 215 | 9.8% | 18.35% | 16.00% |
At 20 years the SIP won in 9.8% of the 215 stocks with a long enough record. At one year it won in 58.1%. The pattern runs one way and it is not subtle: the longer the period, the more reliably investing everything at the start came out ahead.
The arithmetic behind that is dull and inescapable. Money invested in month one compounds for the entire period. Money invested in the final month compounds for a month. A SIP is, by construction, a way of having most of your money invested for less than the full period — so in any stretch where prices rose, it must finish behind.
Where the SIP won, the stock had done badly
The more useful question is which stocks made up the minority. Splitting each horizon by who won and looking at what the lump sum returned in each group answers it immediately.
| Holding period | Where the SIP won | Where the lump sum won |
|---|---|---|
| 1 year | -10.7% | +36.9% |
| 3 years | -20.1% | +98.9% |
| 5 years | -3.0% | +281.8% |
| 10 years | +189.7% | +773.0% |
| 15 years | +358.4% | +2691.8% |
| 20 years | +865.4% | +4717.4% |
At 5 years, the stocks where the SIP came out ahead had an average lump-sum return of -3.0%. The ones where the lump sum won averaged +281.8%. That is not a small difference at the margins; it is two different populations.
A SIP did not beat a lump sum because monthly investing is clever. It beat a lump sum in the cases where the stock went nowhere — which is to say, in the cases you would not have wanted to be in at all.
Why this happens
Spreading purchases out only helps if later purchases are cheaper. In a stock that fell and stayed down, every instalment after the first buys more units for the same money, and the average cost ends up below the starting price. In a stock that rose steadily, every instalment after the first buys fewer units, and the average cost ends up above it.
So the SIP-versus-lump-sum comparison is really a question about the shape of the price path, and the answer is decided before you choose a method. Over long periods most surviving Indian stocks rose, which is why the lump sum wins 90.2% of the time at 20 years.
What this does not mean
- Survivorship applies, and it cuts toward the lump sum.Only companies still in today's Nifty 500 are here. The ones that collapsed are absent, and those are exactly the cases where spreading purchases out would have looked better.
- This measures single stocks, not funds. A SIP into a diversified index fund is a different proposition from a SIP into one company, and nothing here should be read across to it.
- Both legs ignore costs and taxes. A SIP incurs more transactions, so real outcomes would tilt slightly further toward the lump sum than shown.
- Timing is invisible here. Every lump sum is invested on the first day of the window. Someone who invested a lump sum a month earlier or later got a different answer, and this does not measure how much of the gap was luck.
How to use this
The practical reading is narrow and worth keeping narrow. If you have money arriving monthly, invest it monthly — that is what a SIP is for, and this data says nothing against it. If you have a sum already, the historical record says that waiting to feed it in gradually usually cost money, because the market you were waiting to buy into was rising while you waited.
And if you are looking at a SIP that outperformed a lump sum in the same stock, treat it as a signal to look at the stock rather than to congratulate the method. You can check any of this directly: every stock has its own SIP backtest showing both legs side by side, and the rolling-return pages show what each holding period actually produced. The calculations are on our methodology page.
HELP CENTER
Frequently asked questions
Everything you need to know about this page and the data presented.
Is a SIP better than a lump sum investment?+
Measured across every stock we track, no — a lump sum finished ahead in roughly nine of ten cases over twenty years, and the gap widens with the holding period. The reason is arithmetic rather than strategy: money invested at the start compounds for the whole period, while a SIP has most of its money invested for less than that. But the comparison is rarely a real choice, because a SIP invests what you earn each month whereas a lump sum requires already having the full amount.
When does a SIP beat a lump sum?+
When the price fell during the period. Spreading purchases out only helps if later purchases are cheaper, so a SIP wins in stocks that went sideways or down. In our data the stocks where the SIP won had a materially worse lump-sum return than the ones where it lost — at five years, roughly minus three percent against nearly plus three hundred. A SIP outperforming is therefore usually a symptom of a weak holding rather than evidence of a good method.
Does this mean I should stop my SIP?+
That is not a question this research answers, and we do not make recommendations. Nothing here argues against investing monthly — for money that arrives monthly, a SIP is simply what investing looks like. What the data contradicts is the specific marketing claim that monthly investing beats deploying a sum you already hold. Those are different claims and only the second one is tested here.
Do these figures include dividends?+
Yes. Both the SIP and the lump-sum leg are priced on the adjusted close, which folds in dividends and stock splits, so dividends are treated as reinvested throughout. Demergers are not credited, because our data source does not adjust for them, and brokerage and taxes are not modelled on either leg.
Is this affected by survivorship bias?+
Yes, and in a direction worth naming. Only companies still in today’s Nifty 500 are measured, so businesses that collapsed are absent — and those are precisely the cases where spreading purchases out would have looked best. If anything, including them would make the SIP look somewhat better than these figures suggest, though not enough to reverse the pattern at long horizons.