← →

Results of Dollar Cost Averaging for the TSX Composite

March 3rd, 2009 · 24 Comments

Most net buyers of equities do not invest a lump sum in the market. Rather, they invest their savings gradually over time. After reading a passage in The Intelligent Investor on how such a dollar cost averaging strategy performed during the Great Depression, I was curious to see how it would have performed in recent years.

To model the strategy, I made some assumptions: $1,000 in inflation-adjusted dollars is invested in the TSX Composite Index at the start of a calendar year. I used the annual real returns of the TSX Composite available on the Libra Investments website. Also, assume there are no transaction costs, taxes etc.

Here’s the surprising news: such a regular investing program would have produced a positive return for every rolling 10-year period starting in 1970. It is true that the 10-year period ending in 2008 produced the smallest profit in the sample but it was a profit nonetheless. What about 20-year rolling periods? The profits ranged from 50% to 150%. Investors would have at least doubled their money in real terms in investing over 30 years.

I would caution against reading too much into these results. In fact, as I’ll show in a future post, Canadian investors in the S&P 500 and MSCI EAFE fared a lot worse following a dollar-cost averaging strategy and not just in the recent past. The point is that, despite the daily dose of depressing news, Canadian investors could have easily obtained decent results even over 10-year periods. Unfortunately, I’m one of the investors who didn’t — chasing tech stocks in the initial years made sure my returns look worse. But whose fault is that? The market’s or the investor’s?

Bookmark:  

Buzz Up Digg Mixx
Print

Related Posts:

Tags: Investing

24 responses so far ↓

  • 1 Ray // Mar 3, 2009 at 3:04 am

    The benefit of dollar cost averaging is that you might buy at high points but you probably also buy at low points, as Benjamin Graham has shown in the intelligent investor and as you have shown, dollar cost averaging works pretty well if followed through. However the problem is that many don’t follow through with it, usually in times like these many investors stop putting the money in and losing out on low prices. The trick is to stick with it.

  • 2 What is Recession- Recession A Good Thing | Financial Highway // Mar 3, 2009 at 6:30 am

    [...] With stock markets at historic lows it would also make good financial sense to start investing in very stable companies, I wouldn’t recommend lump sums investments I would rather dollar cost average, small amounts on regular bases. This strategy has worked well in the past as Canadian Capitalist points out. [...]

  • 3 mfd // Mar 3, 2009 at 10:00 am

    I’m really interested in the future post regarding dollar cost averaging. Right now I’m dollar cost averaging a version of the sleepy portfolio.

  • 4 EconStudent // Mar 3, 2009 at 10:49 am

    CC: I am wondering if you have read Security Analysis. That is the most famous book that Graham wrote. I thought The Intelligent Investor was Security Analysis, but I think I am wrong.

    I am reading Margin of Safety right now. It is out of print, but you can find the pdf file online through google.

    What do you think of junk bonds? Is it the right to invest in them?

  • 5 Brian // Mar 3, 2009 at 11:19 am

    I think we are missing a key point of information here. What is the return if the lump sum was invested at the beginning of each 10 year period. Is that return greater, or less than the DCA return?

  • 6 Canadian Capitalist // Mar 3, 2009 at 11:19 am

    EconStudent: The Intelligent Investor is a different book by Ben Graham and is aimed at non-professional investors. Graham revised it extensively over time and I really like the version edited by Jason Zweig. I’ve tried reading Security Analysis many years back and it was tough sledding, so I gave up pretty quick.

    I don’t own junk bonds because they have risk-return characteristics similar to equities and highly correlated with stocks as well. I don’t know if you’ve read Unconventional Success. David Swensen classified junk bonds as “non-core” asset class.

  • 7 Canadian Capitalist // Mar 3, 2009 at 11:30 am

    Brian: Comparing DCA with average return on an initial lump-sum investment isn’t an apples-to-apples comparison. If I invest $1,000 over 10 years, the initial $1,000 would have been invested for 10 years, the next $1,000 for 9 years and so on. So, it is not really comparable to investing $10,000 in a lump sum in the beginning. What I think is useful is to know if investing regularly over the years is eventually profitable. At least for Canadian stocks, that has been the case in the past.

  • 8 Brian // Mar 3, 2009 at 12:22 pm

    CC,

    I would argue otherwise. You can compare both methodologies and they both have advantages and disadvantegs in different markets.

    In general, lumpsum investment is perferred in bull markets because money is invested for the full time, but in bear markets, DCAing is perferred because your purchases buys more units as the price goes down.

    But that’s a small argument, what matters is that our money is invested in the market one way or the other.

  • 9 Ray // Mar 3, 2009 at 12:26 pm

    Brain: I think you are right it would be nice to see how these two different strategies would compare, however i believe DCA reduces your risk as you buy at different points and average down when markets are down

  • 10 Four Pillars // Mar 3, 2009 at 1:23 pm

    I think to compare DCA vs lump sum you would have to account for the DCA money earning interest before it gets put into the market.

    ie Mr. DCA starts with $100,000 on Jan 1, 2000 – he puts $1k into the market and $99k into a high interest savings account and GICs. Each month he takes $1k from the cash pile and puts it into the market.

    Mr. LumpSum just puts $100k in to the market on Jan 1, 2000.

    This might be tough to analyse – you’d have to know the various interest rates involved.

  • 11 Thicken My Wallet // Mar 3, 2009 at 2:22 pm

    To dovetail on Four Pillars point, practically speaking, how many people have a lump sum to invest which is equal to the aggregate of DCA over a period of time?

    Brian is right- if you time your entry correctly lump sum investing can yield more than DCA but I find it to be one of those classic economist findings since its premised on accurate market time and sums of cash most people don’t have to invest- the former being difficult for a retail (and most professional) investors and the latter being mostly out of reach for most.

  • 12 ETF2X // Mar 3, 2009 at 3:33 pm

    There is no absolute answer to this issue. If you want to read more about the likelihood of DCA beating all-at-once investing, go to http://www.gummy-stuff.org/ and use the site search engine for find articles on dollar cost averaging that Peter (aka Gummy) has written.

    Scroll down the site’s main page a little, enter “dollar cost averaging” in the Google search window, click on the button for http://www.gummy-stuff.org and then click on “Search”.

    Fred

  • 13 NN // Mar 3, 2009 at 3:36 pm

    Well, if we come to the conclusion that accurate market timing can enhance returns, it has been a wasted effort. The whole point of DCA is that it eliminates the timing decision.

  • 14 Aleks // Mar 3, 2009 at 4:47 pm

    “To dovetail on Four Pillars point, practically speaking, how many people have a lump sum to invest which is equal to the aggregate of DCA over a period of time?”

    Yeah, lump sum investing is generally more profitable than staying in cash and dollar cost averaging because the market tends to return more than cash. You are better off investing as much as you can as early as you can.

    The thing about DCA for most people is that they aren’t sitting on a pile of cash, they are diverting some of their income into the market instead of keeping it in cash. So functionally, both approaches are the same: investing right away rather than putting it off. If you get a lump-sum, invest it all right away; if you get paid every two weeks, invest every two weeks.

  • 15 brad // Mar 3, 2009 at 5:56 pm

    I am pretty much forced to use lump-sum investing for a good chunk of my RSP contribution each year, because my employer gives me a retirement “bonus” every year that comes as a lump sum in one of my February paychecks. No taxes are withheld from that sum, so I either have to apply it all at once to the previous year’s RSP contribution by the early March deadline or else spread it out over the coming year. I’ve always elected to apply it as a lump sum. Right now with the market so low that seems to be a wise choice, but in other years it might not make as much sense.

  • 16 Ray // Mar 3, 2009 at 6:00 pm

    Brad: If you wanted to DCA the bonus, you could buy money market funds with it and than set up an automatic PAC on monthly or biweekly bases. You would contribute on time for RRSP deduction and you could benefit from DCA, thats if you wanted to do DCA.

  • 17 bob // Mar 3, 2009 at 8:23 pm

    try value investing instead of dollar cost averaging..

    http://www.studyfinance.com/jfsd/pdffiles/v13n1/marshall.pdf

  • 18 Thicken My Wallet » Blog Archive » The end of buy and hold? // Mar 4, 2009 at 5:07 am

    [...] course, as Canadian Capitalist pointed out, a buy and hold strategy utilizing dollar cost averaging for a broad based index does not yield bad results over the long [...]

  • 19 Traciatim // Mar 4, 2009 at 9:36 am

    Bob, not sure if you are going to be back, but in the very first value example they use the chart on first glance would seem to indicate the rate of return on value averaging at 33% vs dollar costing at 32%, a small margin in favour of value averaging.

    However, the actual dollar value ends up with 135 shares owned with a value of $16 a piece for a total capital of $2160 for the value average technique and 235 shares owned with a value of $16 a piece for a total capital of $3760. I’m not so sure I see the advantage in a rising market, but I think i have to study the paper more as I’ve only skimmed through getting the premise and looking at the charts.

    It would seem that a yearly value averaging technique would be a perfect match for a discount brokerage like Share Owner Investments if you picked say 10-20 companies you really would like to own for a long time and value average in to them yearly (or even once a quarter). Your transaction costs would be really low, you would have free DRIPing, the account interface is really simple and easy to manage, and they pre-screen companies that are available making it mostly a pick of large stable companies. I see a strategy formulating for my future.

  • 20 The January Effect- Weekly review | Financial Highway // Mar 6, 2009 at 12:26 am

    [...] Canadian Capitalist Looks at Results of Dollar Cost Averaging for TSX [...]

  • 21 Friday Links | The Canadian Finance Blog // Mar 6, 2009 at 6:28 am

    [...] Canadian Capitalist shows the results of dollar cost averaging for the TSX composite. [...]

  • 22 Book Winner, Twitter, Rate Drop and Weekend Reading | Million Dollar Journey // Mar 6, 2009 at 7:31 am

    [...] Canadian Capitalist gives us the results of dollar cost averaging the TSX composite. [...]

  • 23 Patrick // Mar 6, 2009 at 10:25 am

    For those of you discussing whether dollar-cost averaging reduces risk, the surprising answer is “not really”. Here’s a post from my blog where I crunched the numbers.

  • 24 Dollar Cost Averaging | Financial Highway // Mar 11, 2009 at 3:31 am

    [...] Capitalist did a review of DCA for the TSX Composite over the past few [...]

Leave a Comment