Dollar Cost Averaging: Several years ago you had an analysis about whether dollar cost averaging (DCA) was a good way to make deposits into an IRA account. Can you update it? The analysis was shown in the fourth quarter 2002 issue, a little over seven years ago. Most DCA articles I have seen show regular deposits over an extended period of time. Because the authors are almost always proponents of it, they pick a period that includes a severe market drop and ends with stocks at a high level, which will make DCA look like a great investment tactic. One problem with this type of analysis is that the results are highly dependent on the period chosen. Also, if one starts with a substantial amount to invest, due to an inheritance or a receiving a large amount of income for example, the starting time and the amount of the initial investment will strongly affect the analysis.
Because of those problems, I used a different approach that avoids them. I took the point of view of those who are going to make their annual contributions to an IRA account. The choice is between investing all of it right away or dividing the investment over the course of the year. I based my analysis on investing it all at the end of January or 25% of it then and another quarter at the end of April, July, and October. The last transaction includes the earnings in a money market fund of the cash awaiting investment. I made annual comparisons between the two approaches starting in 1975, which was when IRAs were first available. I used the total return of the S&P 500 index as the investment and assumed money markets would pay a little less than the current T-Bill rate.
The results of the analysis are not at all surprising, but some of the details are interesting. In short, in years when stocks go up, it is better to invest all right away, but in years when they go down, DCA is better. Since stocks tend to go up more years than not, over the longer term, it is better to make the entire investment early in the year, and that is the simplest way to do things. During the 35 year period, 1975-2009, stocks were down only seven times, or 20%. The January-to-January periods that correspond to the investment times saw one more down period.
Over the 35 year period, the advantage to investing all right away averaged 1.4% a year with a median of 1.6%. With the current contribution limits of $5,000 for those under 50 and $6,000 for those at least that age, that works out to less than $100 a year, not enough to get excited about one way or the other. Not surprisingly, years with the largest moves in stock prices saw the largest differences. There were six over 10%. Three were advantageous to not doing DCA (by 10.1% in 1995, 10.0% in 2003, and 11.2% last year) and three favored DCA (by 13.8% in 2001, 13.9% in 2002, and 12.2% in 2008).
My analysis assumed that once the investments were made they would not be sold for the remainder of the year (and even longer). Of course, that is not how I think investments should be managed, which I have written about numerous times. Regardless of the timing of the deposits, my client and personal IRAs are managed using models that are designed to avoid most of the down periods and take advantage of the bulk of rising markets. Accordingly, I always make my full deposits and invest them according to the models as soon as I can.