Dollar Costing Average

[lou]

Has any body used a dollar costing average investment strategy.

Dollar cost averaging means investing a fixed amount at fixed intervals of time. That's a sensible approach, for example, if it means committing yourself to investing a fixed amount of your salary every month toward your retirement.

However, some people also think you should dollar cost average a lump sum. For example, if you had $12,000 that you wanted to invest in a stock index fund, they would tell you to invest $1000 per month over a year, rather than investing the whole amount immediately. The rationale is that market volatility should then work in your favor, because you will automatically be purchasing more shares when the price is low, and fewer shares when the price is high.

As appealing as that theory is, its advantage looks like a myth, as this calculator shows. It uses market data to let you compare dollar cost averaging with lump sum investing for the start date you specify.

Link to calculator http://www.moneychimp.com/features/dollar_cost.htm

Each strategy wins at least some of the time, but after a few runs you'll see that DCA is the statistical "dog", losing about two times out of three.

Of course, dollar cost averaging will win if your start date falls right before a dramatic crash (like October 1987) or at the start of an overall 12 month slump (like most of 2000). But unless you can predict these downturns ahead of time, you have no scientific reason to believe that dollar cost averaging will give you an advantage.

So why do so many people persist in believing that this old dog really knows how to hunt? Maybe because it has a psychological appeal: if the market dips, people will be happy because DCA will be saving them money; and if the market goes up, people will be happy regardless.


-----------------------------------------------------------------------------

I did some further testing of the data. While DCA may perform worse more often it is it performs less worse. This preserves your capital making percentage increases more valuable. I have attached a PDF of my results. I have included to bank rates being 3% (on call rate) and 6% (being cost of debt finance).
The PDF shows a greater long term ROI when compared to lump sum investing.

3313

I have a couple of strategies that could work, and would like some feedback on.
1)DCA investing using your personal equity. Over a term of 12 months, then repeat.
2)DCA investing using margin facility or revolving credit mortgage. Over a term of 12 months, then repeat.
3)DCA investing per 1 or 2 above. However selling stake when price crosses 365 day moving average. Or some other TA trigger. If sell trigger is not triggered before the 12 month cycle hold until it does. Once stake is sold resume DCA investing cycle the following month.

Some other issues I could think of:
Brokerage fees could erode returns. However could look at automatic share purchase plans to avoid the broker where possible.
Tax implications of the strategy. Trader Yes? No? Why? Why not? Mitigating factors?

[Lizard]

Thanks Lou - that is interesting!

I started looking after a reasonably large sum of cash from the start of 2002 and chose to invest it into the market using DCA. I was pretty happy with the way it worked out - and looking at your table, it is obvious that I just picked a good year to do it! I used 3 month gaps rather than one month and I reckon the result would be similar to monthly for less work and fees.

I still operate this portfolio on a roughly 3 months schedule, just to keep the time spent on it down to a minimum.

[POSSUM THE CAT]

Lizard Better strategy with managed funds unless you have an awful lot of money as you do not get very much diversity.

[lou]

Thanks Lou - that is interesting!

I started looking after a reasonably large sum of cash from the start of 2002 and chose to invest it into the market using DCA. I was pretty happy with the way it worked out - and looking at your table, it is obvious that I just picked a good year to do it! I used 3 month gaps rather than one month and I reckon the result would be similar to monthly for less work and fees.

I still operate this portfolio on a roughly 3 months schedule, just to keep the time spent on it down to a minimum.

How have you found the returns? How big is the portfolio? What type of stocks have you invested in?

[lou]

Lizard Better strategy with managed funds unless you have an awful lot of money as you do not get very much diversity.

A managed fund or ETF will mean more diversity and less brokerage fees.
However one of the benefits of DCA is it will perform better the more volatile the stock is. Since individual stocks are more likely to be more volatile than the market it should provide better results.

[upside_umop]

Kiwisaver is basically a form of DCA.
I would like to start doing it more often but get too impatient! (apart from KS!)
It's a great way to minmise the risk on your overall entry price.

[shasta]

A managed fund or ETF will mean more diversity and less brokerage fees.
However one of the benefits of DCA is it will perform better the more volatile the stock is. Since individual stocks are more likely to be more volatile than the market it should provide better results.

There is also the power of compounding, should you find stocks that pay quarterly dividends with a DRP. If you dont require the dividend income, reinvesting into extra shares each quarter adds to your position.

A high yield quarterly dividend payer who has a DRP scheme (with a 2.5% discount) is ASX:HHY

Im sure there are others within the "Diversified Financials" sector

[lou]

Kiwisaver is basically a form of DCA.
I would like to start doing it more often but get too impatient! (apart from KS!)
Yes Kiwisaver is a form of DCA. However eventually what you are contributing compared to you entire holding will be relatively minimal.
I was suggesting taking it to the next level. Where you divest your holdings every 12 months(approx) and start DCA investing again.


It's a great way to minmise the risk on your overall entry price.
I think this is why this investment strategy would be well suited to margin investing.

[Lizard]

However one of the benefits of DCA is it will perform better the more volatile the stock is. Since individual stocks are more likely to be more volatile than the market it should provide better results.

I've always found it intriguing that investing a lump sum of money in a series of sequential trades will give a lower final return sum than would be estimated from the average result of the trades. The greater the spread or volatility between trade returns, the further the actual falls below the average result. I've tended to think this is the best argument for diversification between stocks (i.e. reduced volatility), but perhaps using DCA can also work by avoiding volatility. The question is how to apply it without having fresh cash or reducing overall returns by keeping cash out of the market.

Re your earlier questions, the portfolio in question only used DCA in one year to invest across the market as a whole, not individual stocks i.e. added about 5 stocks per quarter. Probably the fact the market was quite weak in 2002 was one of the reasons I did it that way (also because I can't analyse stocks fast enough to choose 20 stocks in a week!). I operate some arbitrary rules just to make it easier to manage the money and avoid me worrying about it every day and over-trading (since it's not mine) and because I've never had a good record for market timing or instinct. Basically the rules say I sell down stocks that are no longer wanted or in which holdings have become excessively large (and sometimes small) in the months of March, June, Sept and Dec. I buy new stocks in Jan, Apr, Jul, Oct. The other 4 months I can ignore it all together. There are a few other guidelines around why I choose what to buy and sell and I am not super-strict as to sticking to all the rules - the concept is a work in progress and obviously takes quite a number of years to decide how well it works and fine-tune! The portfolio operates as "balanced" with about half in fixed interest and property trusts, which is probably important to know if you are wondering where the cash goes to/comes from (all dividend and interest income is paid out monthly to beneficiary). One more reason for the rules was that, since I am also trading in shares on my own behalf, I needed to be clear around the reasons for share transactions within the portfolio to avoid accidentally turning it into a "trader" too.

[Lizard]

I was suggesting taking it to the next level. Where you divest your holdings every 12 months(approx) and start DCA investing again.
.

Since it seems that the downside to DCA results is the time out of the market, would be good to find a way to turn this "batch" process into a "continuous" process. Maybe one way would be to take two volatile shares and then "swap" the $ amount allocated to them say every three months.

[lou]

Since it seems that the downside to DCA results is the time out of the market, would be good to find a way to turn this "batch" process into a "continuous" process. Maybe one way would be to take two volatile shares and then "swap" the $ amount allocated to them say every three months.

This is good in theroy but if both your stocks are correlated then you won't see any benefits. You will just be swaping one dog for another.

If you could find two stocks that are negatively correlated you might be on to a winner. However bear markets tend to increase correlation. Which would screw you. "The only thing that increases in a bear market is correlation".

In regard to your comment the biggest downside to DCA is time spent not in the market. I would have to agree. That's why I think it will work well with margin lending as your equity will stay in the market and your margin ratio will change.

[Phaedrus]

Lou, Dollar Cost Averaging is based on the automatic assumption that it is impossible to advantageously time your market entries and therefore the very best you can do is to spread your entries out so as to average the good and the bad. I strenuously maintain that it IS possible to time your market entries such that they are, overall, better than average. Keep in mind here that all you have to do is beat the completely random entries occasioned by a DCA approach.

What I am talking about here is aligning your actions with prevailing market sentiment. In other words, if the overall market is rising there is little point in sitting on cash. Why delay buying because of some completely arbitrary schedule you have inflicted on yourself? Conversely, why would you want to buy into a falling market, when your target stocks are getting cheaper by the day?

It is very easy to monitor the market and doing so will not turn you into a crazed trader!

The chart below features a simple 200 day Moving Averge - a crude but effective device commonly used to monitor longterm market sentiment. When the Index is green (above the SMA200) buying is allowed - more than that, it is encouraged. This is no time to be sitting on cash, waiting to invest it some time in the future. When the Index is Red (below the SMA200) buying is prohibited. This is a good time to be sitting on cash and building up your funds. There is no point buying into a falling market.

Lou, I'm sure you can see that even this ultra-simple system will outperform any DCA buying by a handsome margin - and it is not hard to devise more profitable systems than this one.

To invest money without even so much as considering prevailing market sentiment (the most important factor of all) is, in my opinion, folly.

http://i602.photobucket.com/albums/tt102/PhaedrusPB/AllOrds200ma49.gif

[lou]

Loved that post Phaedrus.

Timing the market correctly will give better results. I fully agree.

I will put like this:
Worst choice timing the market incorrectly.
Better choice buy and hold
Better choice still DCA
Best choice timing the market correctly

However Phaedrus, your graph is now green. You have none of your equity is invested do you invest everything now, do you wait for the next dip and invest when the market goes back into green, or would you use DCA to get into the market then follow your normal investing rules.

[Phaedrus]

Personally, Lou, I would not use any of those approaches. I would time my entries using conventional technical analysis as applied to each individual stock. Overall, my aim being to buy rising stocks in strong markets and totally avoid buying falling stocks in weak markets.

The Weekend Herald has an article by Diana Clement (who she?) entitled "Busting 10 Myths about Creating Wealth". http://www.nzherald.co.nz/opinion/news/article.cfm?c_id=466&objectid=10718031 According to her, it is impossible to successfully time markets.
"New and overly exhuberant investors sometimes try to time markets............You can sometimes "time" markets once. The reality is that you have lucked in. If your long-term financial plan is based on timing, you'll lose out bigtime, sooner or later".
What nonsense! It is apparent that this silly woman has little understanding or appreciation of trends. They can persist for years and to get yourself positioned on the wrong side of a major trend is to lose money.

If you buy downtrending stocks in a downtrending market, you really WILL "lose out bigtime" - and sooner rather than later! You are fighting the market.
Buying only uptrending stocks in an uptrending market greatly improves the chances that you will "luck in".

Select your stocks randomly and it really is a matter of pure blind luck whether you have made a good choice or not. Fundamental analysis enables you to skew the odds in your favour.
Time your entries randomly and it really is a matter of pure blind luck whether you have made a good entry or not. Technical analysis enables you to skew the odds in your favour.

Lou, I don't want to hijack this thread, so I will bow out now. All I wanted to do was challenge the very basis of DCA - the erroneous assumption that advantageous timing of stock entries is totally impossible and therefore not worth attempting.

[lou]

The Weekend Herald has an article by Diana Clement (who she?) entitled "Busting 10 Myths about Creating Wealth". http://www.nzherald.co.nz/opinion/news/article.cfm?c_id=466&objectid=10718031 According to her, it is impossible to successfully time markets.

"New and overly exhuberant investors sometimes try to time markets............You can sometimes "time" markets once. The reality is that you have lucked in. If your long-term financial plan is based on timing, you'll lose out bigtime, sooner or later".

What nonsense! It is apparent that this silly woman has little understanding or appreciation of trends. They can persist for years and to get yourself positioned on the wrong side of a major trend is to lose money.

If you buy downtrending stocks in a downtrending market, you really WILL "lose out bigtime" - and sooner rather than later! You are fighting the market.
Buying only uptrending stocks in an uptrending market greatly improves the chances that you will "luck in".

The article itself is more aimed at the general pleb rather than any one who has a interest saving and investing. I do like the bit about not investing in you car. A 10-15 year car with low k's is a far better option if you want to save/preserve your capital.



Lou, I don't want to hijack this thread, so I will bow out now. All I wanted to do was challenge the very basis of DCA - the erroneous assumption that advantageous timing of stock entries is totally impossible and therefore not worth attempting.
Thank you for you input, a good honest critique is what I started this thread for.

[Corporate]

Good thread, thanks Lou and Phaedrus!

[CJ]

Dollar cost averaging only works if you buy a stock that goes down. If you buy a stock in an uptrend and sell before it goes into a down trend, then there is no need to average. Therefore from a trading perspective, it is not a good tool. I wouldn't expect Phaedrus to use it.

For buy and hold investors, it makes sense. It works well for Kiwisaver, especially since we have just gone through a big dip. though it would still be best to dump cash in when the price is low and take it out when it is high.

[RRR]

DCA wouldnt have been a good idea (for example) if one decided to invest in Nikkei index in 1989 when Nikkei index was 30,000. Now nikkei is 9000. It doesnt work always.

[karen1]

I think I might have upgraded myself, or is that downgraded, from a Newbie to a General Pleb! Lou and Phaedrus, thank you for what I have found to be the most educationally informative thread I have come across in a long time. I've never been a fan of DCA, having had a reasonable grasp on the concept. My thoughts run quite simply along the lines that one can turn figures around to mean anything you want them to mean. Simply for the exercise I have applied the DCA theory in retrospect to a few stocks over the years, always concluding in either a "good move" or "bad move" answer, depending on how the figures pan out.

Guess I'm old fashioned enough to want to know what I paid for something, pure and simple. Mind you, DCA looks very tempting for a long held stock which isn't very healthy right now. Would be like balm on the wound.

[Snoopy]

I have a couple of strategies that could work, and would like some feedback on.
1)DCA investing using your personal equity. Over a term of 12 months, then repeat.
2)DCA investing using margin facility or revolving credit mortgage. Over a term of 12 months, then repeat.
3)DCA investing per 1 or 2 above. However selling stake when price crosses 365 day moving average. Or some other TA trigger. If sell trigger is not triggered before the 12 month cycle hold until it does. Once stake is sold resume DCA investing cycle the following month.

Some other issues I could think of:
Brokerage fees could erode returns. However could look at automatic share purchase plans to avoid the broker where possible.
Tax implications of the strategy. Trader Yes? No? Why? Why not? Mitigating factors?

Lou, I am not sure that I follow your pdf. Are you saying that if you had invested $10,000 as a lump sum in 'something'(?) on 1st January 1999 then on 1st January 2009 you would have only $8.895? Does your spreadsheet result not depend on your particular data set? If you really wanted to compare DCA vs Lump Sum shouldn't you at the very least compare ten ten year periods (it doesn't matter if they overlap): 1999 to 2009, 2000 to 2010, 1998 to 2008, 1997 to 2007 etc. to remove any bias of your starting point?

I don't know what the actual returns will be over the next ten years for your proposed invesdment. But I would venture to suggest that they will almost certainly not be according to your assumed example. I think your example is just one point on a data set. Interesting, but you will have to do at least ten times as much data processing as that to come up with something predictibly useful. Trying to be constructively critical here!

SNOOPY

[Snoopy]

I have a couple of strategies that could work, and would like some feedback on.
1)DCA investing using your personal equity. Over a term of 12 months, then repeat.
2)DCA investing using margin facility or revolving credit mortgage. Over a term of 12 months, then repeat.
3)DCA investing per 1 or 2 above. However selling stake when price crosses 365 day moving average. Or some other TA trigger. If sell trigger is not triggered before the 12 month cycle hold until it does. Once stake is sold resume DCA investing cycle the following month.

Some other issues I could think of:
Brokerage fees could erode returns. However could look at automatic share purchase plans to avoid the broker where possible.
Tax implications of the strategy. Trader Yes? No? Why? Why not? Mitigating factors?

I am not sure that many investors would look at DCA over such a short time horizon as twelve months. Using personal equity is a possibility, but just trading in and out would add around 2% of your portfolio value to your costs each year. That is more expensive than many fund managers for a start.

I would say using a margin lending facility is out. This will likely increase your trading frequency and you may be forced to sell on market 'noise'. Using a revolving credit mortgage would be a possibility as that gives you more control on when you sell.

I see where you are coming from in using a 365 moving average trading price as an indicator with a time frame of 12 months. However I am not sure that you fully appreciate the statistical nuances of doing this. For example the movement of share prices may not be at all future predictibly evenly distributed about the 365 day average. Probably 95% of the time your method will work. But the 5% of the time it doesn't work will likely ruin you. I am talking about times that you can't get out no matter what your stop loss is for liquidity reasons.

Share purchase plans are always good for avoiding brokerage, but it is unlikely you will be able to Doller Cost Average purchase evenly across all of your investments.

Also remember that the tax man does not treat traders and investors equally. For capital gains you have to make 30% or so more in capital gains than the investor just to make the same return after tax. That is a big handicap to overcome before you even start.

HTH

SNOOPY

[lou]

Lou, I am not sure that I follow your pdf. Are you saying that if you had invested $10,000 as a lump sum in 'something'(?) on 1st January 1999 then on 1st January 2009 you would have only $8.895? Does your spreadsheet result not depend on your particular data set?
Yes it does depend on your data set. The data set that I used was the S&P500. It was used through a calculator so it was a bit of a pain as I had to input everything manually. If you know where I could get a data set in excel format for a large exchange I would love to get my hands on it.




If you really wanted to compare DCA vs Lump Sum shouldn't you at the very least compare ten ten year periods (it doesn't matter if they overlap): 1999 to 2009, 2000 to 2010, 1998 to 2008, 1997 to 2007 etc. to remove any bias of your starting point?
I could of done that but it would of required more work. I did give the yearly return of each method so a comparison could be done that way.



I don't know what the actual returns will be over the next ten years for your proposed invesdment. But I would venture to suggest that they will almost certainly not be according to your assumed example. I think your example is just one point on a data set. Interesting, but you will have to do at least ten times as much data processing as that to come up with something predictibly useful. Trying to be constructively critical here!If I had a better a data set I could do something more conclusive. However I think you will find a similar pattern back in time. It comes down to the principal of protecting your capital. It is better to have a consistent return of 10%pa than a average return of 10%pa.



I am not sure that many investors would look at DCA over such a short time horizon as twelve months. Using personal equity is a possibility, but just trading in and out would add around 2% of your portfolio value to your costs each year. That is more expensive than many fund managers for a start.Yes the trading cost could be prohibitive.



I would say using a margin lending facility is out. This will likely increase your trading frequency and you may be forced to sell on market 'noise'. Using a revolving credit mortgage would be a possibility as that gives you more control on when you sell. As you are only investing in stages those sell signals would only happen at the end of your 12 month cycle. Where you could sell and start the process over again.



I see where you are coming from in using a 365 moving average trading price as an indicator with a time frame of 12 months. However I am not sure that you fully appreciate the statistical nuances of doing this. For example the movement of share prices may not be at all future predictibly evenly distributed about the 365 day average. Probably 95% of the time your method will work. But the 5% of the time it doesn't work will likely ruin you. The 365 day MA was just an example. It could be swapped out for any other indicator you would prefer.



I am talking about times that you can't get out no matter what your stop loss is for liquidity reasons. That is a risk for everybody. Phardus was suggesting that DCA is inferior as it was possible to time the market however this would be more of an issue for his strategy than DCA.



Share purchase plans are always good for avoiding brokerage, but it is unlikely you will be able to Doller Cost Average purchase evenly across all of your investments. It would be if you were investing in an index. If you want to do individual share purchase plans it would take a bit of research.



Also remember that the tax man does not treat traders and investors equally. For capital gains you have to make 30% or so more in capital gains than the investor just to make the same return after tax. That is a big handicap to overcome before you even start.Correct. However trading does allow certain tax deductions, depending on your portfolio size this could mitigate some or all of your tax liability.

[winner69]

lou - yahoo off all places has historical prices of all indices. Dial up the index or the individual stock and down the left side has Historical Prices

here's the S&P500 lot
http://au.finance.yahoo.com/q/hp?s=%5EGSPC&a=00&b=3&c=1950&d=03&e=25&f=2011&g=d

[Phaedrus]

Use DCA to "step" into a rising stock or market, and you underperform. You would be better off investing more, earlier.

Use DCA to "step" into a falling stock or market, and you underperform. You would be better off doing nothing.

[Financially dependant]

Use DCA to "step" into a rising stock or market, and you underperform. You would be better off investing more, earlier.

Use DCA to "step" into a falling stock or market, and you underperform. You would be better off doing nothing.

How about using DCA to step into a bottoming market (as buy signals start firing) and DCA to step out of a topping market (as sell signals start firing)....my 2 cents worth.

[Phaedrus]

As Lou said, "Dollar cost averaging means investing a fixed amount at fixed intervals of time" (totally ignoring the prevailing market trend). If you are buying on Buy signals and selling on Sell signals, you are not using DCA.

[CJ]

Agree with Phaedrus - DCA only works for 'set and forget' investing. Choose a good fund manager and your done. On average you will make the average returns of the market.

If you want to do better, you have to either choose a manager that beats the market or do it yourself and hope you can see something the average fund manager cant see (or is constrained due to the size of their portfolio).

[Snoopy]

Dollar cost averaging only works if you buy a stock that goes down. If you buy a stock in an uptrend and sell before it goes into a down trend, then there is no need to average. Therefore from a trading perspective, it is not a good tool.


DCA will still work if you buy a share that only goes up. Your first investment will be your greatest, in terms of number of shares bought. Then as the share price rises you will buy ever decreasing numbers of shares. Towards the end of your investment period, with the share still going up, you will buy your smallest number of shares for the money. The overall effect is that you have bought more shares when the price is lower. That is all DCA sets out to achieve.

I do agree with one point in your post CJ. Dollar Cost Averaging will be a sub optimal trading strategy, if you can second guess the market.

I think the whole point of DCA is that you don't have to think about the market at all. Whatever the market does you will buy more shares when the share price is lower and less shares when the share price is higher, and this result is guaranteed no matter what the market does.

SNOOPY