There are many variants on dollar cost averaging, in this analysis I only look at one very narrow scenario.
This analysis is comparing investing all capital in one lump sum vs. investing 1/8 of the capital each quarter over two years, from 1980 through 2016. I.e., what would an investment be worth in March 1982 if it was put into the S&P500 as a lump sum in March 1980, vs. putting 1/8 into the S&P500 in each of March 1980, June 1980, …, December 1981. Then, what would an investment be worth in June 1982 if it was put into the S&P500 as a lump sum in June 1980, vs. putting 1/8 into the S&P500 in each of June 1980, September 1980, …, March 1982. Etc.
I did not include dividends from the stock already purchased, or investment income on the balance not yet invested, or taxes paid on these income sources. These can have a meaningful impact on results.
Also, I only did this quarterly (based on the closing price on the first trading day of each month), not daily, but I think it’s reasonably representative.
The results were:
average outcome: lump sum up 20.6%, spread out up 11.7% (lump sum 8.0% better, 1.206 / 1.117)
best outcome: lump sum up 76.8%, spread out up 43.0% (lump sum 23.6% better, 1.768 / 1.430)
worst outcome: lump sum down 43.8%, spread out down 38.1% (spread out 10.2% better, 0.619 / 0.562)
The average turned out to be better by putting it all into the market in one lump sum, but the variance in outcomes is also higher.
This is an analysis of past performance, but past performance is not a guarantee of future performance.
Comments / Questions: joseph AT StockMarketMovement.com