Does Dollar-Cost Averaging Work? Here's What the Numbers Say — Journal Report

By Derek Horstmeyer

Many investors follow the strategy of dollar-cost averaging to invest money in the stock market. But does it always deliver the most bang for the buck?

With dollar-cost averaging, an investor buys a fixed dollar amount of a position at regular time intervals — say, on the first of each month — because it allows you to buy more shares when the market is low and fewer when it is high. Over time, the strategy should lower your average cost per share, if purchases correspond to market cycles.

However, after testing how effective the strategy is, my research assistants Eray Tulun and Lilia Benrabia and I find that while dollar-cost averaging does well on an annualized basis versus a fixed-share strategy — where an investor buys a fixed number of shares or percent of stock at regular intervals — that isn't always the case. Specifically: Dollar-cost averaging overall outperforms a fixed-share strategy by 0.4 percentage point a year over the long run, but the dollar-cost strategy underperforms during down markets compared with a fixed-share strategy.

A million simulations

To study the issue, we set up a trading simulation to mimic the S&P 500 over the past half-century and ran one million simulations for each strategy using historical market performance as parameters in our simulations.

For our dollar-cost averaging strategy, we set up a portfolio where each year the investor allocated $100 to buying shares in the S&P 500. So, if the S&P 500 was priced at $100, then the investor bought one share in the stock index or two shares if it was priced at $50. (It should be noted that changing the time frame to a month in the above scenario yielded the same qualitative results.)

For our fixed-share strategy, we set up a portfolio where each year the investor would buy a fixed number of shares (or fixed percentage of their portfolio) of the S&P 500. But since the price is variable, some years you wouldn't be able to buy as many shares as you used to if the price of the S&P 500 had gone up and there would be cash left over if the price of the S&P 500 had gone down. We assumed the idle cash would be held in an interest-bearing account with a 5% rate of return.

Across all simulations, dollar-cost averaging outperforms the fixed-share strategy by about 0.40 percentage point on an annualized basis. Averaged over all market conditions over a 20-year horizon, we find that dollar-cost averaging delivers 6.93% in annualized returns while the fixed-share strategy delivers 6.53% a year in returns.

Up vs. down

But we also found that while the dollar-cost averaging strategy does well in up markets, it lags behind the fixed-share strategy in down markets.

For a market that goes up over a period of two-plus years, we found that the dollar-cost averaging strategy yielded a return of 23.57% a year while the fixed-share strategy returned 16.04% a year. That is a difference of 7.53 percentage points a year.

However, during a period where the market is lower over a period of two-plus years, we found that the dollar-cost averaging strategy yielded 4.39% a year while the fixed-share strategy yielded 6.03% a year. This is a difference of negative 1.64 percentage points a year for dollar-cost averaging versus fixed-share purchasing.

Finally, we tested what impact market volatility has on dollar-cost averaging versus fixed-share purchasing. To implement this, we tested our strategies over a 20-year horizon assuming volatility of 10% and volatility of 35%. Again the results favor the dollar-cost averaging strategy, with it showing a slight outperformance during a period of 10% volatility and doing much better during the high-volatility scenario.

Bottom line: Over the long run it is best to employ a dollar-cost averaging strategy, but adopting a fixed-share strategy could be worthwhile during an extended downturn.

Derek Horstmeyer is a professor of finance at Costello College of Business, George Mason University, in Fairfax, Va. He can be reached at reports@wsj.com.