Lump sum or spread it out?

Published 15 October 2026

Say you inherited €150,000 and you want to invest it. Do you put it all in at once, or a bit at a time over a few months? History has a fairly clear answer about which one usually ends up with more money. It also has a clear answer about when that choice hurts. Which of the two matters more is not something the numbers can decide for you: that depends on how you would carry a fall in the first months. This piece gives you both answers, in euros, for spreading over 3, 6, 12 or 24 months.

Two ways in, and one thing this is not

The first way is called investing a lump sum: the whole amount goes into the market on day one. The second has an awkward name, dollar-cost averaging, and Vanguard simply calls it cost averaging. You split the amount into equal parts and invest one part a month. Until the last part goes in, the rest waits on a savings account.

One thing this is not: putting part of your salary into a fund every month. That money does not exist yet, so there is nothing to invest at once. Spreading only becomes a choice when the whole sum is already in your account, after an inheritance, a bonus or the sale of a house. That is the situation this piece is about.

So the question is narrow, and it has a real answer. It just has two halves.

What usually happens: all at once comes out ahead

We took €150,000 and ran both routes through every month since the end of 2000 in a broad world index, MSCI World, counted in euros with dividends reinvested. The money that was still waiting earned the average rate Dutch banks paid on new savings deposits that month, according to the European Central Bank. Then we compared the two at the moment the spreading was done.^*

Investing it all at once came out ahead in 64% of the start months against spreading over three months, and in 68% against spreading over twelve. Against spreading over two years it was 77% (Figure 2). Run the same test on the American stock market since 1926 and you get almost the same picture: 62% to 73%.

The reason is not mysterious. Over long stretches, shares have earned more than cash more often than not. Vanguard counted it for the years 1976 to 2022: American shares beat cash 76% of the time.² Money that waits on a savings account is money that is not in the market during that time. In most months, that costs you something.

One detail makes these numbers easier to trust than they look. Once the spreading is done, both routes hold exactly the same investment. Whatever happens next happens to both in the same proportion. Vanguard put it plainly: "whichever portfolio has the most assets at that point will stay ahead forever."¹ So it does not matter whether you compare after one year or after ten. The winner was decided while the money was going in.

That is the first half of the answer. Most of the time, waiting costs money.

How much it costs, and why longer costs more

On average the cost is modest, and it grows with every month you wait. Spreading over three months left you €746 behind on average. Over twelve months it was €4,499, and over two years €10,112 (Figure 3). In the middle case, where half the start months did better and half did worse, the gap was larger still: €6,755 for twelve months.

Why the average sits below the middle case: in some months spreading won by a wide margin, and those months pull the average down. That is the second half of the answer, and it is where the rest of this piece goes.

You may have heard the opposite argument: spread it out and you buy more units when prices are low, so your average purchase price ends up below the average price. That part is true. Simon Hayley of Cass Business School looked at why the argument persists, and his summary is short: "DCA's lower average costs do not lead to higher expected returns."⁷ The money that waits is still money that is not invested.

One thing moves these amounts for everyone: the interest on the waiting money. A better savings rate makes spreading cheaper. Vanguard found the same: "as cash interest increases, LS's advantage diminishes, all other things being equal" (LS is their short for lump sum).²

When all at once goes badly

Figure 4 shows every start month separately. Above the line, investing at once came out ahead after twelve months; below it, spreading did. Most of the chart sits above the line. The biggest gaps below it all came for starts in 2001 and 2002, and in 2007 and 2008, in the run-up to a fall.

Against spreading over twelve months, the worst start for the lump sum was the end of August 2008. A year later the lump sum was worth €127,400. The same money spread over those twelve months was worth €166,885, a gap of €39,485. The lump sum was back at the €150,000 you started with after 19 months, slipped below it again, and stayed above it for good after 38 (Figure 5).

That is not the longest wait in the data. Someone who invested everything at the end of January 2001 had to wait 145 months, more than twelve years, before the lump sum was back at €150,000. Spreading barely changed that: spread over twelve months, the money was back above €150,000 for good after 133 months, and over two years after 129. It did spend less time below in all: 109 and 66 months. And now look at the mirror image. A start at the end of March 2009 gave the lump sum a lead of €39,094 over twelve months of spreading. The same size of gap, the other way round, and nobody knew in advance which month was which.

That is why the question you may have, how long until the lump sum caught up?, has a surprising answer: never. Once both are fully invested, the gap stays the same in percentage terms. The better question is the one this section answered: how long were you below what you put in?

What spreading buys you

Spreading does not buy a higher return. What it buys is fewer and smaller shocks in the first months. Over twelve months, the lump sum was worth less than €150,000 at the end in 23% of the start months, and the lowest value we saw was €90,905. Spread over twelve months, that happened in 20% of the months, and the lowest value was €110,139 (Figure 6). Vanguard found the same pattern in American data since 1926: over twelve months, a lump sum lost value in 22.4% of the periods and the spread version in 17.6%.¹

There is a second thing it buys, and it is harder to put in euros: less regret. Invest everything the week before a crash and it feels like your mistake. Follow a schedule and the market did it. The economist Meir Statman wrote about exactly this in 1995. His subtitle says it best: "Dollar-cost averaging may not be rational behavior, but it is perfectly normal behavior." A fixed schedule, he wrote, takes away part of the feeling of responsibility, and "is especially helpful for investors who are concerned about their exposure to regret."⁵

That is a real benefit. For someone who would lie awake over a fall like the one in 2008, or sell in a panic, paying for calm can be worth it. What matters is knowing that it is a price, and roughly how big.

€150,000, spread over 3, 6, 12 or 24 months

Here is the whole picture in one table. It is not a recommendation for one term or another. It shows how each choice turned out in the past, so you can see what you would be choosing.

Read it from left to right. The longer you spread, the more often investing at once would have done better, and the more you gave up on average. Both extremes get bigger too, because a longer spread keeps more money out of the market, in a fall and in a recovery. The dates are the end of the month in which everything went in at once.

Two things matter for you personally. A higher interest rate on your waiting money makes spreading cheaper. And the share of your wealth matters: Vanguard notes that the choice weighs less if the sum is "a small fraction of an investor's overall wealth."²

So which one? Your risk tolerance decides, in two places

Researchers do not agree on what follows from all this, and it is worth seeing why. The theory has been clear since 1979, when George Constantinides showed that, in his model, a fixed spreading schedule can always be beaten by a better-designed plan.⁶

Others look at the investor rather than the formula. Cho and Kuvvet, in the journal of the American financial planners' association, conclude that "highly risk averse investors should rationally prefer DCA to lump-sum investing" (DCA being spreading).⁸ Vanguard's 2023 model points the same way: its cautious investors who also strongly dislike losses were better served by spreading over three months. It also says that "most investors—particularly those without significant aversion to loss—should invest a lump sum immediately."² And Morningstar argues the opposite of what most people assume: measured by how far results can end up apart, the lump sum is the less risky choice.³

Our reading: they disagree less than it seems, because risk tolerance works in two places. The first is the one this piece is about: how fast you go in. The second comes earlier: how much of the €150,000 you put in shares at all. Vanguard's 2012 study ends on exactly that point. If an entry plan makes you uncomfortable, "it may imply a low willingness to take risk in general," and then it is worth looking again at your mix of shares, bonds and savings.¹

So the question to ask yourself is not only at once or spread? It is: if this money were worth a third less a year from now, what would I do?

And whichever you choose, Vanguard puts the choice in proportion: "whether LS or CA is chosen will make only a marginal difference compared with permanently keeping a cash allocation."²

What the studies say, side by side

You will often read that a lump sum wins "two-thirds of the time." That is roughly right, but every study measured something slightly different. Figure 7 puts them next to each other, each with its setup.

Three things stand out. Every row lands above half. The closest to a euro investor, Vanguard's figures for world shares in euros from 1998 to 2022, sit close to ours: 66.4% for three months, 67.9% for six.² And the longer the spread, the higher the number: about 90% for three years in Vanguard's 2012 study, and nine out of ten when the money was dripped in over ten years in Morningstar's.¹³

The numbers differ for dull reasons. Some studies let the waiting cash earn interest and some do not; some mix shares with bonds; some measure after one year and some after ten. None of that changes the direction.

Sources

1. Anatoly Shtekhman, Christos Tasopoulos and Brian Wimmer, "Dollar-cost averaging just means taking risk later", Vanguard research, July 2012 (read via a copy hosted by Indexa Capital). Vanguard sells index funds. Our source for: 67% (US, 60% shares and 40% bonds, spread over 12 months, measured after 10 years; p. 3); approximately 90% for 36 months (p. 3); "whichever portfolio has the most assets at that point will stay ahead forever" (p. 4); 22.4% and 17.6% of 12-month periods with a loss (p. 6); "it may imply a low willingness to take risk in general" (p. 6).
2. Megan Finlay and Josef Zorn, "Cost averaging: Invest now or temporarily hold your cash?", Vanguard, February 2023 (vanguard.com). Vanguard sells index funds. Our source for: US shares beat cash 76% of the time, 1976–2022 (p. 2); "as cash interest increases, LS's advantage diminishes" (p. 5); "most investors—particularly those without significant aversion to loss—should invest a lump sum immediately" (p. 2); investors with strong loss aversion preferring cost averaging (p. 7, Figure 5); "a small fraction of an investor's overall wealth" and "only a marginal difference compared with permanently keeping a cash allocation" (p. 8); 66.4% and 67.9% for world shares in euros, 1998–2022 (p. 9, Figure 6).
3. Tom Lauricella, "The dollar-cost averaging myth: why lump sum investing usually wins", Morningstar, November 2019 (morningstar.com.au; the same text as morningstar.ca). A report on a study by Maciej Kowara and Paul Kaplan, "Dollar-Cost Averaging: Truth and Fiction", which we could not find in full. Morningstar sells investment data and ratings. Our source for: "nine out of 10 times" over ten-year periods, US large shares, 1926–2019; a lump sum "presents a lower risk profile".
4. Matt Stucky, "Is dollar-cost averaging better than lump-sum investing?", Northwestern Mutual, January 2025 (northwesternmutual.com). An insurer and wealth manager. Our source for: 75% (all shares) and 80% (60% shares, 40% bonds), spread over 12 months and measured after 10 years. The page does not say which years it covers.
5. Meir Statman, "A Behavioral Framework for Dollar-Cost Averaging", The Journal of Portfolio Management 22(1), Fall 1995, pp. 70–78 (doi). Our source for: the subtitle (p. 70) and "is especially helpful for investors who are concerned about their exposure to regret" (p. 74).
6. George M. Constantinides, "A Note on the Suboptimality of Dollar-Cost Averaging as an Investment Policy", Journal of Financial and Quantitative Analysis 14(2), June 1979, pp. 443–450 (doi). Our source for: a fixed spreading schedule is beaten by a better plan (Propositions 1 and 2, pp. 445 and 447).
7. Simon Hayley, "Dollar Cost Averaging: The Role of Cognitive Error", Cass Business School, version of 28 January 2015 (City, University of London). A working version; the published version may differ. Our source for: "DCA's lower average costs do not lead to higher expected returns" (p. 3).
8. David D. Cho and Emre Kuvvet, "Dollar-Cost Averaging: The Trade-Off Between Risk and Return", Journal of Financial Planning 28(10), October 2015, pp. 52–58 (fpa). Our source for: "highly risk averse investors should rationally prefer DCA to lump-sum investing".
9. MSCI, MSCI World Index, net total return in euros, end-of-month levels, December 2000 to August 2026, the months we used (msci.com).
10. European Central Bank, MFI interest rate statistics: Netherlands, households, deposits redeemable at notice of up to three months, new business, January 2000 to August 2026 (ecb.europa.eu).
11. Kenneth R. French, Data Library: Fama/French 3 Factors, monthly, US stock market return and one-month Treasury bill rate, July 1926 to August 2026 (dartmouth.edu).
12. * ProofOfReturns, own calculation, October 2026, on sources 9 to 11. For every month we invested €150,000 at once, or in 3, 6, 12 or 24 monthly parts with the rest earning the savings rate of that month (sources 10 and 11), and compared the two at the end of the spreading period, when everything, interest included, was invested. All shares, no bonds; no costs and no tax on your side; the world index is in euros, the US market in dollars. It describes what happened, not what will happen.

How we checked. Verified 15 October 2026. Every figure from sources 1 to 8 was pulled from its source document twice, independently, and the two readings were compared cell by cell against the document itself. Our own calculation was checked by hand against the raw index levels for the August 2008 example. Four things a careful reader should know. First, Vanguard's 2023 study calls one pair of results an "average", while its own table places them on the middle (median) row; we do not use that pair. Second, the note under Figure 3 of Vanguard's 2012 study describes its differences the other way round from how its text and other numbers read them; we use only figures that do not depend on that. Third, we know Kowara and Kaplan's Morningstar study only through Morningstar's own report of it. Fourth, the studies in Figure 7 differ in market, period, mix, interest on cash and spreading term, so their percentages should not be averaged into one number. Sources 1 to 4 come from firms that sell investment products or advice: two from an index fund manager, one from a data and ratings firm and one from an insurer. Dutch public guidance (the AFM, Wijzer in geldzaken, Nibud) does not, as far as we found, say anything about spreading a lump sum over time. Think we got something wrong? Tell us at info@proofofreturns.com. We correct in public.