Dollar-cost averaging: what the average-cost effect really delivers

Basics8 min read · updated 8 September 2026 · 5 sources

Dollar-cost averaging (DCA, also called the cost averaging effect) describes the fact that a savings plan with a fixed amount buys more units when prices are low and fewer when prices are high. As a result, the average purchase price is below the average of the prices. This is always true, and yet it is no reason for higher returns. This article shows why, with figures and studies.

How cost averaging works

With a savings plan, you invest the same amount every month, say $100. If the price is low, you get more units for $100; if it is high, fewer. The average price per unit is therefore the harmonic mean of the prices, and mathematically this is always below or equal to the simple average (the arithmetic mean).

For comparison: if you buy a fixed number of units every month (say 10 units), you pay exactly the average price of $10.125 on average. Compared with this approach, the fixed amount really does have an advantage. That is the true core of the cost averaging effect.

For advanced readers: harmonic and arithmetic mean

That the average purchase price never exceeds the average price follows from the inequality between the harmonic and arithmetic means. The two are only equal if the price does not move. The more prices fluctuate, the larger the gap: with prices of $10 and $9, the savings plan pays $9.474 against $9.50; with $10 and $5, $6.667 against $7.50. The effect is therefore purely a phenomenon of fluctuation. It says nothing about whether the investment is worthwhile.

Why the effect brings no extra return

The catch lies in the comparison. The real alternative to a savings plan is not buying a fixed number of units, but a lump-sum investment: investing all the money right at the start. Whether the savings plan then does better depends only on how the price develops, not on the cost averaging effect.

Because stock markets rise more often than they fall in the long run (otherwise there would be no equity risk premium), the money not yet invested in a savings plan sits, on average, in the lower-yielding investment. The savings plan shifts risk into the future; it does not eliminate it. That is why experts speak of the cost averaging myth.

On average: one year staggered versus investing at once

With an assumed return of 6% a year, the expected difference can be calculated directly. $12,000 invested immediately is worth $12,720.00 on average after one year. If instead you invest $1,000 at the start of each month, you have on average only just over half the money in the market and end up with $12,386.53. That is an expected $333.47 less. This is the price for the lower risk of having invested everything shortly before a crash.

The reverse effect with withdrawals

In retirement, the calculation is reversed. If you withdraw a fixed amount every month, you sell more units when prices are low and fewer when they are high. With the same prices as above (10 · 8 · 12.5 · 10), a total of 40.5 units are sold for 4 × $100. The average selling price is $9.877 and therefore below the average price of $10.125. Here, then, fluctuation works against the investor. That is why the first years of a withdrawal phase are especially delicate, as the articles on sequence-of-returns risk and withdrawal rate show.

What the research shows

Constantinides (1979) showed formally that investing in equal installments is inferior to an optimal strategy: an investor who is allowed to react to new prices can replicate any savings plan rule and improve on it. Equal installments therefore throw information away instead of using it.

Rozeff (1994) compared lump-sum investing with staggered investing over twelve months. If the risk of both approaches is held equal, the lump sum delivers 1 to 4 percentage points more return depending on company size; with the S&P 500, it came out ahead in 40 of 65 years. His recommendation for risk-averse investors: rather invest a smaller portion in equities at once and put the rest in safe investments, instead of staggering the whole amount.

Behavioral economics explains why the savings plan is nevertheless so popular. Statman (1995) describes a behavioral framework: staggered investing reduces the regret after buying just before a crash and helps with self-control. The foundation is the prospect theory of Kahneman and Tversky (1979): losses hurt more than gains of the same size give pleasure. And Benartzi and Thaler (1995) show that checking frequently makes equities seem subjectively riskier. A savings plan that runs automatically takes this pressure off.

When a savings plan is still the right choice

For most investors the question does not even arise: they do not have a large sum, but save part of their income every month. In that case, the savings plan is simply the natural way to invest, and indeed immediately, as soon as the money is there. The advantage lies in the regularity, not in the average price.

• Ongoing income: invest whatever is left over each month. This is not staggering, but investing at the earliest possible moment.
• Discipline: a standing order stops you waiting for the “right moment”. Why timing rarely works out is shown in market timing.
• Large sum and nerves: if you are investing an inheritance and could not bear a crash right after buying, you can stagger. You should know that this costs you return on average, and set a fixed plan with an end date (say 6 to 12 months).
• Watch the costs: savings plans are cheap at many brokers; with fixed order fees per purchase, small contributions can become expensive.

The alternative to staggering

If you have a large sum and are afraid of getting in, there is a second option that does better in the research: controlling risk through the allocation rather than over time. Instead of spreading 100% equities over twelve months, you might, for example, invest 60% in equities and 40% in bonds or instant-access savings straight away and keep this mix. The risk is smaller from the start, but all the money is working immediately. This is exactly what Rozeff (1994) recommends to risk-averse investors. Which mix suits you is shown in the article on the 60/40 portfolio.

Cost averaging is no protection against losses: if the market falls for good, the savings plan loses too. It has merely invested later and at lower prices.

Common misconceptions

• “The savings plan buys more cheaply, so it earns more.” It buys more cheaply than the average price, but not more cheaply than a lump sum at the starting price when the market rises.
• “With a savings plan, timing does not matter.” That is not true for each individual contribution; the total return depends heavily on how the markets perform in the years when your wealth is large. This is called sequence-of-returns risk.
• “Volatility is good for savers.” Volatility creates the effect, but at the same time lowers the geometric return (see volatility drag). Less volatility with the same average return would be better.

How Portfolio Lab calculates it

In the Backtest you can set up a savings plan monthly, quarterly or annually, optionally with a fixed execution day, fees per execution and a spread, as at a broker. The returns shown are time-weighted: they show the portfolio's performance without the effect of the contributions. To compare with a lump sum, run the same period once with initial capital and once with a contribution and compare the final values. The details are in the backtest methodology.

This article explains a method; it is not investment or tax advice. Worked examples are simplified and rounded. All terms in the glossary.