You have a sum of money to invest — an inheritance, a bonus, the proceeds of a property sale. Should you invest it all today, or spread it out over the next year to avoid buying at a peak? The second approach is called dollar-cost averaging (DCA). It feels safer. The question is whether it actually leads to better results.
How we tested it
Using Robert Shiller's monthly S&P 500 data with dividends reinvested, we compared two approaches for every possible 12-month window since January 1871:
- Lump sum: invest everything in the first month.
- Dollar-cost averaging: invest one-twelfth at the start of each of the next 12 months, holding the rest in cash.
We then compared the two portfolios at the end of the 12 months. To keep the test simple, uninvested cash earns nothing. That assumption favours the lump sum slightly; we discuss it below.
The results
| Windows starting from | Number of 12-month windows | Lump sum ended ahead | Median advantage of lump sum |
|---|---|---|---|
| 1871 | 1,818 | 71.5% of the time | +4.7% of the amount invested |
| 1926 | 1,158 | 73.7% of the time | +5.7% |
| 1950 | 870 | 76.4% of the time | +6.0% |
| 1990 | 390 | 79.0% of the time | +5.9% |
In roughly seven out of ten historical windows, investing everything at once left you with more money a year later. The typical advantage was around 5% of the amount invested. Vanguard's widely cited 2012 study of US, UK and Australian markets reached a similar conclusion, finding that lump-sum investing came out ahead about two-thirds of the time.
Why the lump sum usually wins
The explanation is simple. Stocks have risen in most years: measured with this data, the S&P 500's 12-month total return was positive in about 72% of all rolling 12-month periods since 1871. When the market usually rises, money waiting in cash usually misses out. DCA delays your investment, and delay is costly when the expected return is positive.
When DCA wins — and by how much
DCA comes out ahead when the market falls during the averaging period. The biggest wins happened during the Great Depression. For an investor starting in August 1931, a lump sum lost 40.2% over the next 12 months, while spreading the money over the year finished 4.2% up, because most of the purchases happened at much lower prices. Starting in August 1937 — the beginning of a steep bear market — the lump sum lost 21.3% while DCA gained 11.3%.
So DCA works as a form of insurance: on average you give up some return, and in exchange you avoid the worst outcomes from investing everything just before a crash.
What about interest on cash?
In our test the cash waiting to be invested earns nothing. In reality you would hold it in a savings account or money-market fund. With interest on the cash, DCA's results improve a little, especially when interest rates are high, and the lump sum's advantage narrows. The overall pattern — lump sum ahead most of the time — has held in studies that include interest, but the gap is smaller.
So which should you choose?
- If you are focused on expected outcome, the evidence favours investing a lump sum promptly, in a portfolio whose risk level you are comfortable holding.
- If a fall right after investing would make you sell in a panic, DCA can be worth its cost. A strategy you can stick with beats a theoretically better one that you abandon. Averaging in over a short period, such as three to six months, gives up less return than twelve.
- If you are investing from your salary, this isn't really a choice. Investing a fixed amount every month as you earn it is simply putting money to work when you have it — and it is the best thing you can do. Use the monthly contribution slider in our calculator to see how regular contributions add up.
Whichever approach you choose, the bigger decision is the overall level of risk — the mix of stocks, bonds and cash — rather than the timing of the first purchase. Getting that right determines how deep your drawdowns can be; see why a 50% loss needs a 100% gain.
Notes
Data: Robert J. Shiller, U.S. Stock Markets 1871–Present, via datasets/s-and-p-500. Monthly average prices with dividends reinvested, through June 2023. No taxes, fees or interest on cash. The 12-month comparison measures both portfolios at the end of month 12. Calculations are our own.
This guide is educational and not investment advice. Past performance does not guarantee future results.