Dollar Cost Averaging

Update (2-8-08): After this column was originally posted, I received feedback from a few readers explaining that some of the advice in this column might be unfair to those who make small investments over a period of time based on the assets they have on hand (a worker contributing monthly to a 401(k), for example). If an investor can only make investments proportional to their monthly income, he/she really doesn't have any other choice but to engage in DCA-style investment, so I'm not sure the advice in this column really speaks to them. The aim of my column was to make the point that those who have significant non-invested assets and make a conscious decision to engage in DCA are probably leaving money on the table over the long-term.

________________________________________________________________________

As we close the book on another year and start a new one, it's time to think about one of the most interesting periods of the calendar for individual investors. In the first few months of the year you may find yourself, like many investors, flush with cash. Maybe it's a year end bonus (lucky you), a tax refund (shame on you), or the harvest of capital losses (smart on you). If you're like the typical investor, you face a decision – put that money to work immediately or reinvest it over time. In this column, we'll take a look at a popular method of investing called dollar cost averaging (DCA).

Question: Let's start with the basics. What is dollar cost averaging?

Answer: Dollar cost averaging (DCA) on its simplest level just means spreading an investment out over a defined time period. For instance, let's say you have $10,000 to invest. If you wanted to utilize dollar cost averaging, you might invest $1,000 a month for ten months. When you invest in a 401(k), 403(b) or other defined contribution plan, you are engaging in dollar cost averaging if you have a specific dollar amount deferred from every paycheck. In this method, the investor generally doesn't have a choice as most 401(k) investors rarely have the income necessary to fund their 401(k)s all at once.

Question: Why do people use DCA? Why is the technique so popular?

DCA is believed to result in more shares over time with a lower degree of risk. In essence, the theory holds that a DCA investor will acquire additional shares at lower prices and fewer shares at higher prices, thereby decreasing overall cost basis.

DCA is constantly lauded by financial advisors despite a trove of academic research indicating otherwise. The truth is DCA is a bad idea. DCA is almost never mentioned when the market is going up. If you decided to DCA during a period where the stock market advanced 20%, there would be a significant opportunity cost associated with seeking to avoid investing during a bad period. Those funds you invested later in the cycle would have been much, much better off had they just been invested as a lump sum at the beginning. In fact, DCA is really only effective when the market is going down. Since the market has generally trended upward over longer periods of time, a DCA investor has to either get lucky and dollar cost average during just the right time or be extraordinarily good at predicting market downturns. I wouldn't bank on either of those options happening consistently.

Question: What about those mathematical examples I see so often "proving" that DCA is effective?

Answer: A recent study by John G. Greenhut at Texas A&M included the following endnotes:

"A recent search on MSN came up with 250 sites discussing dollar cost averaging, with more than half providing a table or some numerical illustration of the ‘cost benefit' of DCA*."

Question: How much more effective is lump sum investing than dollar cost averaging?

Answer: As I referenced earlier, there is a very large amount of academic research proving DCA is less effective over the long-term than lump sum investing. One of the best studies was conducted by Michael Rozeff at University of Buffalo-SUNY in 1994**. Dr. Rozeff compared the results of a single lump sum investment in the S&P 500 between 1926 and 1990 versus the returns achieved when those investments were averaged over the entire year. In two-thirds of the years, the lump sum investment outperformed the averaged investment. More importantly, over the entire time period, the lump sum investor's annual average returns were 1.1% higher than the person using dollar cost averaging. Let's look at the value of that difference over a 20 year time period:

2008 Funds (Lump) 2008 Funds (DCA)
$500,000 $500,000

2027 Funds (@ 10% annual return)2027 Funds (@ 8.9% annual return)
$3,363,749 $2,751,234

That's a difference of over $610,000. Pretty significant.

My past columns will tell you I'm a big believer in letting the capital markets work for you. Not only is dollar cost averaging ineffective, it also requires more work on the investor's part and complicates what should be a far simpler process. If you have a lengthy time horizon and some patience, it's almost always better to invest your assets as soon as you can and let the market do the heavy lifting.

*John G. Greenhut, "Mathematical Illusion: Why Dollar Cost Averaging Doesn't Work," Journal of Financial Planning (2006).

**Michael S. Rozeff, "Lump Sum Investing versus Dollar Averaging," Journal of Portfolio Management, (1994).