Tuesday 22 December 2009

Dollar Cost Averaging - the Good Truth and the Bad Myth

Investopedia defines Dollar Cost Averaging (DCA) as "The technique of buying a fixed dollar amount of a particular investment on a regular schedule, regardless of the share price. More shares are purchased when prices are low, and fewer shares are bought when prices are high. ... Dollar-cost averaging lessens the risk of investing a large amount in a single investment at the wrong time." Part of DCA so-defined makes good investment sense and part of it is bad, faulty thinking.

The Bad, Faulty Logic
The idea that you can lower risk and get either the same return or a better risk-adjusted return by spreading your security purchases out instead of making one lump sum investment is wrong. The reason it is wrong is that markets more often go up than down, just over 50% of the time day to day, with the percentage rising steadily as the interval goes up to weeks, months, years and decades. The sooner you are invested in the market, the more the odds favour making more money. That doesn't mean you would never be better off doing DCA - if markets head down then it can pay off. But if you do DCA then it means you think the odds are more that it will go down till you are finished putting in money as you end up buying more low. Most of the time, you will end up buying higher as prices move up, i.e. probably some "buy low" and more "buy high".

Both in practice and in theory, studies demonstrate, as referenced in the excellent Wikipedia article on DCA, that DCA does not benefit an investor who has the choice of investing today or tomorrow.

If you are worried about the market tanking or need the money within a few years, you are far better off just keeping it in cash.

The Good Investment Sense
Following a regular, sustained automatic program of saving money and investing it whether monthly, quarterly or annually is a great strategy since it overcomes the fundamental problem with so many people (been there myself) of just not getting round to setting aside the money.

Investing the money when it is available adheres to the basic principle of not trying to outguess or time the market. In many cases, recommendations on various investment websites to do DCA mean only automatic periodic investment to avoid market timing, a smart strategy. They do not assume the presence of a lump sum investment alternative, such as when an inheritance is received. When a large lump sum is at issue, DCA makes no sense.

Far better is to split up the lump sum immediate investment and parcel it according to your target portfolio allocations (see previous posts on Asset Allocation and A Written Investment Policy). That way the risk is spread amongst a number of types of assets and holdings. This is a good way to lessen the anxiety of making the wrong choice, which is likely at the root of DCA's enduring popular appeal as a supposed risk reduction tactic despite the debunking it has already received.

Along this latter line of thought, famous finance professor Kenneth French puts forward, in this short video on Dimensional Fund Advisors, the idea that DCA can help the investor. He says the economic argument is clearly against DCA but the avoidance of an over-reaction to a big market drop after a single large investment lends some support to DCA. Since the series of DCA investments will not have as much shock effect as one big downward move, a kind of emotional risk diversification is apparently achieved. The investor is less likely to permanently and entirely turn away from the market.

Transaction costs are an issue for the regular saving investor. Mutual funds do not charge such fees but buyers of ETFs or individual stocks must pay commissions so small purchases may entail significant relative costs, even at low discount broker rates - e.g. a $10 commission on a $200 trade is a 5% burden that must be recouped in the rise of the investment. There is no hard and fast optimal percentage at which to buy since markets may move slowly or quickly, but in my view it makes sense to accumulate enough contributions to lower the transaction cost to a maximum - 1% is my rule of thumb.

The bottom line: if you are contemplating doing dollar cost averaging, make sure you are doing it for the right reason.

Disclaimer: this post is my opinion only and should not be construed as investment advice or recommendations.

3 comments:

Patrick said...

I once had a lump sum to invest that was pretty big by my standards. I split it into two and invested them a month apart. This served to limit my downside if the market tanked right after my first purchase (since my average price would be the harmonic mean of the two days' prices). To me, this this was worth the loss of one month's returns on half the lump.

Maybe this advantage was less logical and more psychological: I don't think I could face the possibility that my entire lump would decline in value right after I invested it. But if half of it declined, I could at least pat myself on the back with one hand while I invested at bargain prices with the other hand.

As it turned out, it didn't matter: the whole process was finished in first quarter 2008, several months before the market tanked!

CanadianInvestor said...

Patrick, your story seems to fit with the French explanation of dollar cost avergaing's appeal.

We would certainly be better off being able to avoid 2008-size downturns but then we only can know after it has happened. Right now, today, there are some predicting a double dip market fall while others are predicting slow and erratic upwards movement and still others continued strong growth. Which is right? - I don't know, we'll only know in a year or two or three.

Damien (the Bite Size Idea Guy) said...

The whole philosophy of dollar-cost averaging is based on two premises: (1) that the stock market is a good long-term investment, meaning in the long-run the market will go up and (2) that it is impossible to predict what the market will do in the short-term. If you believe both of these premises, then DCA is a good strategy for you. Read more about it here: http://www.bitesizeidea.com/bsi/dollar-cost-averaging-fancy-name-simple-investment-strategy