BJ Banner

Does Dollar Cost Averaging Really Work?

BJ Banner

One definition of dollar cost averaging (DCA) is taking the amount of money you have to invest now and, instead of investing right away, dividing it up and investing it in increments over a predetermined period of time. For example, let’s say you liquidated several actively managed funds, received a large bonus or inheritance and you wanted to invest $120,000 in index funds now. You can either plunge right in and invest the entire $120,000 all at once (IFA's advice) or you can wade in slowly and spread it out over a year, by investing $10,000 every month. This concept is also often referred to as wade or plunge into the market and is discussed in the video below:

 

 

Proponents of dollar cost averaging like to highlight that dollar cost averaging turns the volatility of the stock market into an asset for the investor. When prices are up, you buy less shares and when prices are down you get to buy more shares. But this only makes sense when referring to the other definition of DCA, which refers to benefits of regular deposits into a savings account or a 401k program. This is actually taking the plunge with each amount you have to invest, when you have it to invest. 

Because securities are priced every day and every month for a positive expected return, we would expect a wading in DCA strategy to underperform a lump-sum strategy more often than not (see the data table below the bars in chart below).


Some advisors also use dollar-cost averaging as a behavioral tool to help get their investors into the market in order to overcome the “inertia effect.” An object at rest tends to stay at rest and an object in motion tends to stay in motion. Getting investors off of the sidelines by slowly sticking their toes back into the water versus just jumping in will help them move towards their long term financial goals.

Beyond the behavioral arguments for dollar cost averaging, does the strategy actually work for investors who are trying to take advantage of volatility?

Analysis

We looked at monthly performance data provided by MoneyChimp going back to 1950 and examined two separate investment strategies. The first strategy is a lump-sum investment of $10,000 at the beginning (January 1st) of each calendar year into an S&P 500 index fund with an annual expense ratio of 0.2%. The second strategy is a DCA strategy that invests an equal portion ($833.33) at the beginning of each month over the course of the entire year. The balance that was not invested during each month earned 3% annualized interest (long term historical average of 1-month T-Bills). We then compared the year-end outcomes of each strategy.

Here is what we found.

  1. On average, the lump-sum investment strategy outperformed the DCA strategy 58% of the time over the 67 year time period from 1950-2016. See chart below.

 

  1. On average, the lump-sum strategy outperformed by 2.08% per year, which means that the investor had 2.08% more money through the lump-sum strategy versus the DCA strategy at the end of each year.
  2. The best year for using the lump-sum strategy was 1975.
  3. The best year for using the DCA strategy was 1962.
  4. In terms of statistical significance, the lump sum strategy outperformance has a t-stat of 1.96, which is close to the 95% confidence level.
  5. This statistical significance seems to be sample dependent. If we were to split the 67-year sample size into two approximately equal time periods, we see that there is no statistically significant difference between the lump-sum and DCA outcomes for the 34-year period from 1950 to 1983, although it is still positive. We do see a statistically significant difference in performance for the 33-year period from 1984 to 2016.
Time Period Mean Std. Dev Count T-Stat
1950-1983 1.24% 9.35% 34  0.78
1984-2016 2.94% 7.96% 33 2.12

Conclusion

Although dollar-cost averaging sounds like a great idea, it hasn’t been able to keep up with a simple lump-sum strategy from a simple empirical standpoint. More often than not, the lump-sum strategy has been a superior strategy, which is what we would expect. While we have observed a strong statistical significance of the lump-sum strategy’s outperformance, it seems to be sample dependent with most of the significance coming from the last 33 years. Nonetheless, there is no evidence that a DCA strategy provides a superior implementation strategy versus a lump-sum strategy. 

There are still behavioral reasons why an investor would want to implement a DCA strategy. As independent wealth advisors, our value is to help investors overcome these behavioral biases through education so that they can capture more of what the benefit the capital markets may provide.

Big picture: stick with a lump-sum strategy. Once you have determined your risk capacity it just makes more sense to plunge right in at that level of risk exposure.