Investing a lump sum at once has beaten spreading it out about two times in three — but a monthly SIP from your salary is a different choice, and its real strength is that it keeps you investing

“Lump sum or SIP?” is really two questions, and most advice blurs them. The first applies when you already have a sum to invest — a bonus, an inheritance, a matured deposit, the proceeds of a flat — and must decide whether to put it in at once or feed it in over months. The second is how to invest money you have not earned yet: a fixed amount from each month’s pay, which is what a systematic investment plan (SIP) usually is. The evidence answers the two very differently.

With a sum in hand: investing at once usually wins

Vanguard compared the two approaches in 2023 using global shares from 1976 to 2022. Investing a lump sum immediately beat splitting it into three monthly parts in 68% of rolling one-year periods. Splitting it still beat leaving everything in cash 69% of the time, and investing at once beat cash 70% of the time¹. The reason is plain: over 1976–2022, US shares beat cash in 76% of periods and US bonds in 68%, so every month money waits is usually a month of return given up¹.

The longer the money is drip-fed, the more often investing at once wins: in the US the hit rate rises from 66.4% with a three-month split to 73.7% with a six-month split¹. Paying interest on the cash while it waits narrows the gap but does not close it — the lump sum still won 65% of the time for an all-equity portfolio¹. This is consistent with Vanguard’s own earlier work and with a long academic literature reaching the same conclusion¹.

By how much — and the third of the time it loses

The typical advantage is modest. Starting from $100,000, the median all-equity portfolio was worth $111,940 a year later if invested at once and $109,580 if spread over three months — about 2.2% more. In a 60/40 mix of shares and bonds it was 1.8% more, and in a 40/60 mix 1.2% more¹. The more of the portfolio is in shares, the more there is to gain from getting it working sooner.

The other third of the time matters too, and it is the third people remember. At the 5th percentile — roughly the worst year in twenty — the all-equity lump sum ended at $82,947 against $85,906 for the three-month split¹. Investing at once means accepting a wider range of outcomes: more often ahead, occasionally further behind. A deep fall straight after investing a life-changing sum is not hypothetical — US shares lost 38% in real terms in 1931 alone³ — and the regret of that experience is what makes many people hesitate.

Those last two per cent are Vanguard’s own median figures, about 2% for an all-equity portfolio after a year¹. In India, the usual way to spread a lump sum is a systematic transfer plan (STP) from a liquid or debt fund into an equity fund, which at least earns something on the waiting money. The same arithmetic applies: the longer the transfer runs, the more return is given up on average.

A SIP from your salary is a different decision

Vanguard is careful to separate the two meanings of cost averaging: its study is about money already available, not about investing a fixed amount from each paycheque¹. That distinction is the heart of the SIP question. If your savings arrive monthly, a monthly SIP is not holding money back; it is investing each instalment as soon as it exists. In the study’s terms, a salary SIP is much closer to the lump-sum strategy: money invested as soon as it is available, twelve times a year.

So the real alternative to a salary SIP is not a lump sum. It is saving the money up in a bank account and investing it later, whenever the market looks right — and that is where the evidence turns against investors.

What “rupee-cost averaging” does and does not do

A fixed amount buys more units when prices are low and fewer when they are high. Invest 1,000 a month at unit prices of 100, then 50, then 100: you buy 10, 20 and 10 units — 40 units for 3,000, an average cost of 75 a unit, below the average price of about 83. That is real arithmetic, and it is why a falling market is less frightening to someone with a running SIP. In a market that dips and recovers, like this one, the monthly buyer ends ahead: 40 units worth 4,000, against 30 units worth 3,000 for someone who put the whole 3,000 in at the start. In a market that only rises, the order reverses — and because markets have risen more often than they have fallen, the earlier money has usually done better. Averaging is a way of buying, not a reliable source of extra return.

Timing the market: the gap investors pay

Morningstar’s annual Mind the Gap study compares the return funds earned with the return the average dollar invested in them actually got, which depends on when investors bought and sold. Over the ten years to December 2024, the average dollar invested in US funds earned 7.0% a year against the funds’ own 8.2% — a gap of about 1.2 percentage points, or around 15% of the funds’ return, which Morningstar attributes to the timing and size of investors’ purchases and sales². Investors in allocation funds — all-in-one funds such as target-date funds, often held through retirement-plan menus — captured nearly 97%: a 6.3% a year dollar-weighted return against 6.5% for the funds². The measure has critics: a 2026 study in the Financial Analysts Journal, using the same 2015–2024 sample, put the cost of poor timing at only about 0.1 percentage points a year⁴. Both agree on the direction; they disagree on the size.

The long record shows why waiting for the right moment is so costly. After the 1929 crash, US shares took until February 1945 to recover; after the dot-com peak in March 2000 the fall and recovery took seven and a half years; after the 2008 crisis, four years⁵. Nobody announced those bottoms in advance, and an investor waiting on the side for a clearer signal usually missed the early part of each recovery.

The behavioural case for a SIP

The strongest evidence for automatic investing comes from behavioural economics rather than from market returns. When a large US employer switched its 401(k) plan to automatic enrolment, participation rose sharply even though nothing about the plan’s economics changed — people stayed with whatever happened by default⁶. Inertia, usually a problem, worked for them.

Richard Thaler and Shlomo Benartzi’s Save More Tomorrow programme used the same force deliberately: employees committed in advance to save part of future pay rises. In the first company to adopt it, 78% of those offered the plan joined, 80% were still in it after the fourth pay rise, and their average saving rate rose from 3.5% to 13.6% over 40 months⁷.

A SIP is the household version of the same design. It turns the decision to invest into the default and makes stopping the act that requires effort. It also removes the monthly temptation to judge the market — the stop-start buying and selling that Mind the Gap links to a wider gap between a fund’s return and its investors’. The trade-off is that a SIP keeps buying through a fall — the point of it, but uncomfortable — and that a SIP amount set years ago and never raised quietly falls behind a growing salary.

Putting it together

1. Money you already have and will not need for years: the evidence favours investing it at once, in the mix of assets you would hold anyway.
2. If a fall straight afterwards would make you abandon the plan, spread it over a few months — through an STP in India — and accept that this usually costs a little.
3. Money you will earn: invest it as it arrives, with a SIP or a standing instruction, and raise the amount when your pay rises.
4. Money needed within a few years is exposed to a fall whichever way it goes in — which is why short-term money is usually kept out of shares altogether.

This is information about what the research found, not financial or tax advice; the right choice for a particular sum depends on your plan, your other savings and how you would react to a fall. For why the order of returns matters most just before and after you stop working, see sequence of returns risk.