If you have $100,000 sitting in cash, should you invest it all right now or dollar cost average over the next three or six months?
In 1993, two financial planners set out to answer that very question. The results were surprising. They found one strategy that resulted in 46% higher returns or more than $46,000 based on an initial $100,000 investment.
We’re going to see what those researchers found. Then I'll share with you the results of my own study, which expands the data from 1871 to 2021.
Conventional wisdom is that dollar cost averaging is the way to go, but what does the data say? Let's find out.
Williams & Bacon (1993) Study
Most financial planners suggest that dollar cost averaging is a better way to get money to work. Proponents of the strategy argue that investing over the next few months would allow you to buy in at lower prices if the market were to fall between now and then.
This thinking is so widespread amongst the investment advisor community and financial planning experts that to question it would be crazy. In 1993, two financial planners did just that.
Richard Williams and Peter Bacon wrote an article in the Journal of financial planning called "Lump Sum Beats Dollar-Cost Averaging”. Their study looked at monthly rates of return for the S&P 500 index and 90-day US Treasury bills using data from 1926 to 1991.
They looked at three strategies.
- Strategy #1 assumed 100% of the portfolio was invested in the market on day #1.
- Strategy #2 assumed the funds were initially invested in 90-day US Treasury bills and then shifted an equal monthly installments into the S&P 500.
- Strategy #3 assumed the funds were invested in 90-day US Treasury and invested in the S&P 500 over a 6-month horizon.
Then, they looked at the returns of each strategy for every possible 12-month period from January 1926 through December 1990. In total, it was 780 different rolling year returns.
What they found was surprising.
Contrary to conventional financial wisdom, the lump sum investment outperformed between 60.5% and 64.5% of those 780 observations. Not only did it perform best most of the time, but it had the highest rate of return at 12.75% per year. The longer it took to get money invested, the worst returns got. The 3-month dollar cost average strategy produced 11.14% annually, while the six months dollar cost average strategy produced 9.97%. The 12 month dollar cost averaging produced just 8.5%.
The authors also isolated the years between 1950 and 1991 as well as 1970 and 1991. Their results were consistent across all time horizons and statistically significant at the 95% level in all but one (the 3-month strategy over the 1970-1991 period, in which rising inflation made it difficult for stocks).
The authors conclude:
"these results strongly support a lump sum investment strategy over a dollar cost averaging strategy."
If you had invested a lump sum of $100,000, your ending results after 12 months would have been $113,370. The dollar-cost averaging strategy ended with $109,630. That would be a difference of $3,740 after one year.
No big deal, right? Well, not if you stopped there. If you compound that at an average return of 13.37% per year, however, the difference between the two portfolios would be $46,008 after 20 years. That is 46% of the initial investment - all from spreading the investment over 12 months instead of 1.
The results of the 1993 study are clearly in favor of lump sum investing. Yet, most experts must have missed the memo. The argument for dollar cost averaging is as pervasive as ever.
What if we looked a little further back than 1926 and updated the numbers beyond 1991? Does their data still hold?
Expanding the Study
To find out, I did my own study using Robert Schiller's US stock market data. I looked at all of the different rolling 10-year time horizons from 1871 to May 2021. And I assumed a $1 million initial investment.
I tested four different strategies: a 12-month, 6-month, and 3-month dollar cost averaging strategy in addition to a lump sum strategy. Rather than looking out just 1 year, I decided to expand the time horizon to look out a full decade.
12-Month Dollar Cost Averaging
The first strategy I looked at was a 12-month dollar cost averaging strategy. Here, we would invest 8.3% of the $1 million into stocks right away and then in equal installments for the next 11 months. On average, this strategy would have grown $1 million to $2,547,944 over 10 years.
Some periods were better than others. If you had made your first investment in August 1919, your account would have grown to $6,206,725 by August 1929. Unfortunately, it would’ve then completely tanked during the Great Depression; but we’ll not go there.
In other times, the portfolio would have done worse. If you had made your initial purchase in February 1999 at the height of the tech bubble, you would have finished the decade near the bottom of the financial crisis in 2009. Your investment would have shrunk from $1 million to $659,061. Clearly, just because you dollar cost average over 12 months doesn’t make you immune to big losses.
If you compare the 12-month strategy to a lump sum strategy, things don’t look great. The lump sum outperformed the dollar cost average strategy in over 70% of rolling 10-year periods. The biggest advantage to the dollar cost averaging strategy was the period from July 1931 to July 1941. The 12-month dollar cost averaging strategy beat the lump sum by 5.9%.
The six month dollar cost averaging strategy still wasn’t as good as the lump sum, but it was a little bit better than the 12 month.
It outperformed lump sum 33% of the time. The biggest advantage to the six month dollar cost average strategy occurred from February 1932 to February 1942. A six-month dollar cost average strategy would have grown 8.9% over that time compared with 5.7% for the lump sum.
On average the six-month dollar cost average strategy grew $1 million to $2,607,236. The best performance came from April 1949 through April 1959 when $1 million grew to $6,360,930. The worst performing occurred during the great depression. $1 million invested over the six months starting March 1929 would have shrunk $669,806 by March 1939.
3 Month Dollar Cost Averaging
If you dollar cost average over three months, the results get slightly better - outperforming lump sum 36.5% of the time.
The biggest outperformance was investing the three months starting February 1932 ending February 1942. The three month dollar cost average strategy returned 8.2% versus 5.7% for the lump sum.
On average, $1 million grew to $2,637,952 with the best again coming in 1949 to 1959 when $1 million grew to $6,639,998. The worst performance occurred in the great depression with $1 million shrinking to $669,111.
With all three of these studies, the lump sum is superior between 63.5% and 70.7% of the time. My result is consistent with the original 1993 study.
On average a $1 million invested lump sum in any single month going back to 1871 would have grown $2,659,113 ten years later. That would be an average outperformance of about $22,000 compared with even the best dollar cost average strategy.
However, we do see that the lump sum strategy can result in worse performance if you invest right before a bad market.
For example a lump sum made in July 1929 right before the Great Depression would have shrunk to $662,675. If you would have spread those dollars out over the next three months investing 33% and July 33% in August and 33% in September, your $1 million investment would have shrunk to ‘only’ $669,111 by July 1939. So your downside would have been approximately $6,500 less following the dollar cost average strategy.
Lump Sum > Dollar Cost Averaging
Based on the 1993 study and my expanded study, it is clear that the numbers do not support dollar cost averaging at all.
And we shouldn’t be all that surprised. If dollar cost averaging were superior to lump sum investing, then it would make sense to sell all of your stocks right now and re-invest them over the next few months. No one would recommend doing that, so why do most financial experts recommend dollar cost averaging? If you're regularly saving monthly, then you have to dollar cost average; however, if you had a bunch of money to invest right now, history suggests getting it to work as fast as possible.
While the lump sum strategy is superior most of the time, the difference between lump sum and dollar cost averaging strategy is really not all that much over the course of 10+ years. $22,000 on an initial $1 million investment is a lot, of course; however, it’s not like you’re going to be running out of money in retirement over it. So, if your time horizon is at least that long (and it should be), the most important thing is that you invest. Period.
If you can’t stand the thought of investing a lump sum now, it would be better to make a less than optimal choice and dollar cost average then to not make a choice at all and sit in cash for 10 years.
But if you want to do the most logical thing, lump sum is clearly the best option in most environments. Dollar-cost averaging is not an effective strategy for either increasing returns or reducing risk.
A Better Way Than DCA
The biggest reason people want to dollar cost average is because they are afraid of stock market downside. What if you invest today and it is the absolute market top? I did a detailed study of this in another article (and video). The results were surprising.
If you are not OK with your portfolio dropping by 50% in any given year (which happens about once every 20 years), then you should consider whether owning some high-quality bonds would be a good idea. Choosing the right asset allocation is going to be a much better way for you to reduce risk than to try to do it via dollar cost averaging.
For example, let’s say that you were worried about stocks dropping 50%, but you would be more OK with a 30% drop in your portfolio.
According to this chart from Vanguard the worst rolling 12-month returns of a 70/30 stock and bond portfolio have been -30.7%. Compared to the -43.1% one year return of a 100% stock portfolio, that reduction in volatility may help you feel better about your portfolio.
If you’re worried about the market, choosing a less aggressive allocation is the best way to diversify that risk; dollar cost averaging may do that in certain periods, but the odds are not in your favor. Stocks may just as easily go up over the next three months, in which case you would be buying at increasingly higher prices.
Statistically that is far more likely that stock prices are higher in a year than lower. Just keep that in mind before you decide to take most financial experts' advice to dollar cost average. Over time, it is a losing proposition.
Disclaimer: This is entertainment only, not investment advice. All opinions expressed are my own. Any stocks or ETFs mentioned may be owned or taken a position within the next 48 hours. Neither the information nor any opinion expressed it so be construed as a solicitation to buy or sell a security of personalized investment, tax, or legal advice. This is prepared for informational purposes only. It does not address specific investment objectives, or the financial situation and the particular needs of any person who may receive this report. The information herein was obtained from various sources. Dividend Growth Machine LLC does not guarantee the accuracy or completeness of information provided by third parties. The information in this report is given as of the date indicated and believed to be reliable. Dividend Growth Machine LLC assumes no obligation to update this information, or to advise on further developments relating to it. If you want to run your own study, you can do so using this calculator.
Subscribe to Nathan Winklepleck, CFA
Subscribe to the newsletter and unlock access to member-only content.