Skip to content
Menu
Finabase Learn

Strategies · Long-term investing

Lump sum or instalments: every start month since 1993, tested

All at once or in 12 monthly parts: both rule sets tested on every SPY start month from 1993 to 2025, with the worst outcomes of each and the starts where each lost.

On this page
  1. The idea
  2. The rules
  3. What the backtest shows
  4. Where each choice lost
  5. Compared with its neighbours
  6. How Finabase does it
  7. Frequently asked questions
  8. Summary
  9. Check your understanding

Someone with a sum to invest, from a bonus, a house sale or an inheritance, has two simple choices. Put all of it into the fund today, or split it into equal parts and invest one part a month. The second is Dollar-cost averaging vs lump sum: how DCA works and what data show applied to a lump of cash, and it is the one most people prefer, because it avoids the worst outcome: investing everything the month before a crash.

Both are strategies with rules that leave nothing to judgment, so both can be tested. This lesson runs them from every possible starting month in SPY's history, 394 of them, and looks at more than who won. It shows how large the gap was each way, what the worst year looked like under each, and the starting months that made each choice the wrong one.

The idea

The two strategies end in the same place: fully invested in the same fund. They differ only in the first year, and only in one respect. The lump sum holds the fund for the whole year. The instalment plan holds, on average, about half fund and half cash.

So the question "lump sum or instalments?" is really the question "is the fund likely to do better than cash over the next year?" Stock prices have risen in most years, which is why they are held at all. That makes the lump sum the favourite before any data is examined. George Constantinides made the argument formally in 1979, and Vanguard's 2012 study found that a lump sum beat 12 monthly instalments about two times in three in the US, the UK and Australia.

The instalment plan buys something real with the return it gives up. If prices fall during the year it buys part of its position lower, so its loss is smaller. It is a way to hold less risk for a year. The test shows the price of that, and when it paid.

The rules

Lump sum

  1. On the last trading day of the month the money arrives, invest all of it in the fund.
  2. Hold. Do nothing else.

Twelve instalments

  1. Divide the money into twelve equal parts.
  2. On the last trading day of the month the money arrives, invest one part.
  3. On the last trading day of each of the next eleven months, invest one more part, whatever the price.
  4. Keep the parts not yet invested in cash.
  5. After the twelfth purchase, hold.

Rule 3 is the one that gets broken. A plan that pauses "until things calm down" is no longer this strategy; it is market timing, and it has no rules to test.

A worked example

Take $12,000 at the end of October 2007, when SPY closed at $154.65.

  • Lump sum. $12,000 buys 77.59 shares. Eleven months later, at the end of September 2008, SPY closed at $115.99 and the position was worth $9,000, a loss of 25.0%.
  • Instalments. $1,000 a month bought shares at twelve prices between $154.65 and $115.99. At the end of September 2008, the day of the last purchase, the shares were worth $10,296, a loss of 14.2%.

The instalment plan was 12.6% ahead. A lump sum the month before a bear market is the case it exists for.

Now take $12,000 at the end of January 2012, with SPY at $131.32. By the end of December 2012 SPY was $142.41. The lump sum had gained 8.4%; the instalments, bought at rising prices, had gained 2.7%. The lump sum was 5.6% ahead. Nothing dramatic happened that year, and that is the ordinary case.

What the backtest shows

The test, stated in full so it can be checked:

  • Data. SPY month-end closes from February 1993 to October 2026. Every month from February 1993 to November 2025 is a starting month: 394 in all.
  • Trading. Every purchase is at a month-end close. Each start is scored on the day of the twelfth instalment, eleven months after the first, when both are fully invested. From that day on the two hold the same fund, so the gap between them, as a percentage, never changes again.
  • Costs. A cost that is a percentage of the amount bought, such as a spread, is the same for both: each buys $12,000 of the fund. It cancels out of the comparison and is left out. A broker that charges a fixed fee per order would charge the instalment plan twelve times.
  • Left out. Dividends, interest on the waiting cash and taxes. Dividends favor the lump sum, which holds more of the fund during the year. Interest favors the instalments.

The figure follows one start, the worked example above, for four years.

SPY monthly from 2007-10: 12 monthly instalments against a lump sum, no dividends, no interest on cash $0.6 $0.8 $1 Growth of $1 Lump sum (buy and hold) 12 monthly instalments 0% −20% −40% −60% Fall from the peak 2008 2009 2010 2011
SPY monthly from October 2007: twelve monthly instalments against a lump sum. Dividends and interest on cash not included. Top: value of $1. Bottom: the fall from the highest value so far.
Measure, start October 2007 12 instalments Lump sum
Value after four years, per $1 $0.93 $0.81
Total return −7.1% −18.8%
Deepest fall from a peak −45.3% −52.2%

Notice the deepest fall. The instalment plan was fully invested by September 2008, so it took most of the fall to February 2009 anyway. Twelve months of instalments shortened its exposure to the bear market; it did not avoid it.

The second figure is the ordinary case, the start of January 2012.

SPY monthly from 2012-01: 12 monthly instalments against a lump sum, no dividends, no interest on cash $1 $1.2 $1.4 $1.6 Growth of $1 Lump sum (buy and hold) 12 monthly instalments 0% −10% Fall from the peak 2013 2014 2015 2016
SPY monthly from January 2012: twelve monthly instalments against a lump sum. The gap opens in the first year and stays: $1.48 against $1.40 after four years.

One start proves nothing either way, so here is every one of them.

Lump sum's lead over 12 monthly instalments, by starting month, SPY 1993–2025 −20% 0% 20% Lump sum ÷ instalments − 1, at the 12th instalment 1995 2000 2005 2010 2015 2020 2025
The lump sum’s lead over twelve instalments on the day of the last instalment, for each starting month, SPY 1993 to 2025. Above zero the lump sum was ahead.
Measure, 394 starting months Result
Lump sum ahead 298 starts, 76%
Average lead of the lump sum 4.0% of the amount
Median lead 5.2%
Average lead when the lump sum was ahead 7.7%
Average gap when instalments were ahead 7.2%
Largest lead for the lump sum 29.2%, start February 2009
Largest lead for instalments 29.5%, start May 2008
Starts that were below the amount invested Lump sum 21%, instalments 23%
Worst outcome Lump sum −44.0%, instalments −30.9% (both start March 2008)

Three findings matter more than the headline.

The sizes are symmetric. When the lump sum won it won by 7.7% on average, and when it lost it lost by 7.2%. The lump sum's edge comes from winning three times as often, not from winning bigger.

Instalments did not reduce the chance of a loss. They finished below the amount invested in 23% of starts, slightly more often than the lump sum's 21%. In a rising year the last instalments are bought near the top, and a small dip leaves them under water.

Instalments did reduce the size of the worst loss. The worst lump sum was down 44.0% after a year; the worst instalment plan was down 30.9%. That is the real benefit, and it is a benefit to the first year only.

By period, and with other schedules

Starting months Starts Lump sum ahead Average lead Worst start for the lump sum
1993 – 1999 83 92% 7.9% −4.4%, January 1994
2000 – 2009 120 57% −0.9% −29.5%, May 2008
2010 – 2019 120 79% 4.7% −5.4%, April 2011
2020 – 2025 71 82% 6.8% −13.3%, December 2021

The result depends on the decade. In the 2000s, which held two bear markets, the lump sum won only 57% of starts and was behind on average. In the other three periods it won about four times in five or more. An investor cannot know which kind of decade has begun, which is the whole difficulty.

Schedule Starts Lump sum ahead Average lead Worst outcome, lump sum Worst outcome, instalments
3 instalments 403 65% 0.7% −24.8% −13.8%
6 instalments 400 70% 1.8% −36.3% −20.1%
12 instalments 394 76% 4.0% −44.0% −30.9%
24 instalments 382 81% 8.5% −47.9% −40.2%
36 instalments 370 80% 13.0% −43.5% −41.8%

Each row is scored on the day of its own last instalment. The longer the schedule, the more return was given up on average: 0.7% for three months, 13.0% for three years. The protection did not grow with it. A three-year schedule had a worst outcome almost as bad as the lump sum's, because by the time it ended it had been mostly invested through a full bear market. Vanguard's title for this is exact: instalments mean taking the risk later.

Where each choice lost

Lump sums that lost to instalments. Every large loss for the lump sum is a start shortly before or during a bear market.

Start Lump sum after the 12th month Instalments Lump sum's lead
August 2000 −20.3% −7.1% −14.3%
October 2007 −25.0% −14.2% −12.6%
August 2008 −23.3% +7.3% −28.5%
December 2021 −14.2% −1.0% −13.3%

Instalments that lost to a lump sum. These are less memorable, because nothing went wrong: the cash simply waited while prices rose. The largest is the start of February 2009, at the low of the financial crisis. The lump sum gained 45.3% in eleven months; the instalments gained 12.4%. Many people began instalment plans in early 2009 because a lump sum felt reckless. That caution cost 29.2% of the amount.

Other ways the comparison misleads:

  • Overlapping starts. Neighbouring starting months share ten of their eleven months. The 394 starts are far fewer than 394 independent trials, and the bad ones come in clusters.
  • Interest on cash. With short-term rates near 5%, a year of instalments earns about 2.5% on the amount, since half of it sits in cash on average. That would have closed more than half of the 4.0% average gap in a high-rate year, and almost none of it from 2009 to 2021.
  • A fund that includes bonds. The test is all stocks. For a mixed portfolio the gap is smaller in both directions; Vanguard's study used a 60% stock, 40% bond mix.
  • Regular saving is not this choice. Investing part of each paycheck is not a decision to delay. The money is invested as soon as it exists.

Compared with its neighbours

Approach Average cost in this test Worst first-year outcome Main risk
Lump sum None −44.0% A crash in the first year
6 monthly instalments 1.8% of the amount −20.1% Prices rise while the cash waits
12 monthly instalments 4.0% of the amount −30.9% The same, for longer
Waiting for a fall, then buying Not testable: no rule Not testable The fall comes from a much higher price

How Finabase does it

Finabase has no tool that picks a schedule, and none of its tools says when to buy. Three are useful for seeing the mechanics.

  • Chart. SPY's full history, linked above, draws monthly bars. Any start month in the tables can be found on it, with the twelve closes that followed.
  • Portfolio. A portfolio records each purchase with its date and price and shows the average cost, which is the number an instalment plan changes.
  • Backtest and Strategy. These tools replay indicator signals and if/then rules on one symbol. They cannot run this comparison: Backtest buys a fixed $10,000 on each signal and charges no cost, and Strategy replays two years, which holds at most 13 starting months.

This page's numbers differ slightly from the test in Dollar-cost averaging vs lump sum: how DCA works and what data show, which reinvests dividends and scores each start one month later. That version has the lump sum ahead in 79% of starts against 76% here. Dividends are the difference, and they favor the side that holds more of the fund.

Frequently asked questions

Is a lump sum better than dollar-cost averaging?

More often than not, in this test. On SPY from 1993 to 2025 a lump sum was ahead of 12 monthly instalments in 76% of starting months, by 4.0% of the amount on average. It was behind by 12% or more for starts just before the 2000, 2008 and 2022 bear markets.

Does investing in instalments lower the risk of losing money?

It lowers the size of the worst loss, not the chance of one. The worst first year was −30.9% for instalments against −44.0% for a lump sum, but instalments finished below the amount invested in 23% of starts and the lump sum in 21%.

How long should an instalment plan run?

The test gives no best length, only a trade-off. Three instalments gave up 0.7% on average and twelve gave up 4.0%. Beyond a year the cost kept rising while the worst outcome stopped improving: 36 instalments had a worst case of −41.8%.

What about waiting for the market to fall before investing?

That is a different strategy, and it cannot be tested, because it has no rule for when to give up waiting. SPY closed at a new month-end high in about one month in three over this period, so the wait is often for a fall that starts from a much higher price.

Summary

A lump sum and a twelve-month instalment plan differ only in how much of the fund they hold during the first year. Because SPY rose in most years from 1993 to 2025, the lump sum was ahead in 76% of 394 starting months, by 4.0% of the amount on average. The instalment plan won when prices fell during its year, by about as much as it lost the rest of the time, and its worst first year was a loss of 30.9% where the lump sum's was 44.0%. It did not make a loss less likely, and longer schedules cost more without protecting more. The result leans on one fund in a period that rewarded being invested, leaves out dividends and interest, and in the 2000s the two choices came out about even.

Check your understanding

4 questions on this lesson. Each answer comes with the reason, and nothing is scored.

  1. A lump sum and a 12-month instalment plan end up holding the same fund. What is the one real difference between them?

    1. The instalment plan gets a lower average price
    2. During the first year the instalment plan holds about half fund and half cash on average
    3. The lump sum pays more in trading costs
    Show the answer

    B. During the first year the instalment plan holds about half fund and half cash on average After the last instalment both hold the same thing. The whole comparison is a year of holding less of the fund, which helps if prices fall that year and costs if they rise. The average price is lower only when prices fell.

  2. In the SPY test the lump sum was ahead in 76% of starting months. When it was ahead its lead averaged 7.7%; when behind, 7.2%. Where does its advantage come from?

    1. It wins by much more than it loses
    2. It wins about three times as often, by about the same amount as it loses
    3. It avoids the bear markets
    Show the answer

    B. It wins about three times as often, by about the same amount as it loses The gains and losses are nearly the same size. The edge is frequency: prices rose in most twelve-month periods, so holding the fund for the whole year was the better side of the bet three times in four.

  3. Instalments finished below the amount invested in 23% of starts and the lump sum in 21%. What did instalments actually improve?

    1. The chance of ending the first year with a loss
    2. The size of the worst first-year loss: −30.9% against −44.0%
    3. The average result
    Show the answer

    B. The size of the worst first-year loss: −30.9% against −44.0% They lost money slightly more often, because in a rising year the last instalments are bought near the top. What they cut was the depth of the worst case, since part of the money was bought lower during a fall.

  4. Stretching the plan from 12 to 36 instalments raised the average return given up from 4.0% to 13.0%. What happened to the worst outcome?

    1. It improved in proportion
    2. It barely improved: −41.8% for 36 instalments against −43.5% for the lump sum
    3. It disappeared
    Show the answer

    B. It barely improved: −41.8% for 36 instalments against −43.5% for the lump sum A three-year plan is mostly invested by the time a bear market that starts in its second year arrives. A longer schedule moves the risk later; it does not remove it, and the cost of waiting keeps growing.

Sources and further reading

  1. Vanguard Research, “Dollar-cost averaging just means taking risk later” (July 2012) — lump sum beat 12-month instalments in about two-thirds of periods (60/40 portfolio: US 67% over 1926–2011, UK 67% over 1976–2011, Australia 66% over 1984–2011)
  2. George M. Constantinides, “A Note on the Suboptimality of Dollar-Cost Averaging as an Investment Policy,” Journal of Financial and Quantitative Analysis 14(2), 1979
  3. SEC, Investor.gov — Dollar Cost Averaging (glossary)
  4. Finabase chart data — SPY monthly bars, February 1993 to October 2026; rules and results computed by Finabase