Lump Sum vs Dollar Cost Averaging — Part 2
In Part 1 (Historical returns of Dollar Cost Averaging) of the dollar cost averaging study we looked at 5, 10 and 15 year returns of a dollar cost averaging investment study in the S&P 500 index with start dates from every year since 1920.
I now set out to do a study of the results to see how a lump sum investment would have done historically if one were to have invested a lump sum at different starting in each month beginning in 1920.
This would be for a scenario where someone gets a lump sum amount of cash, think for example inheritance, proceeds from a house, a sale of a business, etc.
I then recorded the 5 year, 10 year and 15 year returns for each starting year.
Results
Full spreadsheet of results
Histogram of Returns
Best 5 Year Return since 1990
Invested in December 1994:
- 5 Year Returns: 28%
- 10 Year Returns: 12%
- 15 Year Returns: 8%
Worst 5 Year Return since 1990
Invested in March 2004:
- 5 Year Returns: -6%
- 10 Year Returns: 7%
- 15 Year Returns: 8%
Best 10 Year Return since 1990
Invested in September 1990:
- 5 Year Returns: 16%
- 10 Year Returns: 19%
- 15 Year Returns: 12%
Worst 10 Year Return since 1990
Invested in March 1999:
- 5 Year Returns: -1%
- 10 Year Returns: -4%
- 15 Year Returns: 4%
Best 15 Year Return since 1990
Invested in November 1990:
- 5 Year Returns: 17%
- 10 Year Returns: 18%
- 15 Year Returns: 12%
Worst 15 Year Return since 1990
Invested in September 2000:
- 5 Year Returns: -2%
- 10 Year Returns: -1%
- 15 Year Returns: 4%
IRR Difference — Lump Sum vs Dollar Cost Averaging
Summary
The average returns of a lump sum strategy vs a dollar cost strategy are very similar.
What’s surprising is the chances of a 15 year period with a loss is slightly less with lump sum investing (0.1% vs 1.3% for dollar cost averaging).
Ignoring the averages, the data also shows that 45% of the time lump sum investment would have slightly worst results, the corollary of which is that 55% of the time lump sum investment would result in better returns. So it’s really close to coin toss of which strategy performs better.
Another interesting fact is that maximum drop in 5 years resulted in a -17% IRR in lump sum vs -31% IRR under dollar-cost strategy. These happened when investing in 9/1/1929 and 6/1/1927 respectively. Looking at the data since, the maximum drop in 5 years were -6% IRR in lump sum, vs -6% in a dollar-cost strategy, both of which resulted in respectable 15 year IRRs of 8%.
So whatever strategy you pick you need to be able to stomach a possible short term loss. So once again a horizon of 10 to 15 years is critical.
Next post
In the next post, we’ll look at how market timing might have made a difference in both strategies by looking at investing at decade highs, lows and averages.
