Exploring Dollar Cost Averaging Versus Other Strategies
This week, one article that was pretty heavily discussed was on the concept of dollar cost averaging (DCA).
The article states that the returns, based on the recent historical results of dollar cost averaging into exchange traded funds (ETF) is not too good.
A better implementation would be to dollar cost average only when the price of the STI ETF is below the average.
“I took monthly closing prices of the ETF over the 9 years and found that the average price is $2.977. I would then do a dollar cost average strategy only when the ETF is below this average price.”
An even better implementation is to double down on the dollar cost average when below the average.
However, this strategy had a problem, it means that my capital of $10,900 is not fully invested over the 9 year period. I tested another tweak to the strategy which I call “Double-Up Below Avg NAV Strategy” after observing that the total invested amount is approximately halved. Instead of investing $100 each time, you would double up the investment to $200 each time the monthly closing price trades below the average price of $2.977.
I think there is something flawed in that.
The Advantage of Dollar Cost Averaging
There are much debate about dollar cost averaging versus investing in a lump sum.
There is a debate because, we do not know what the future holds.
The market at this point:
- might be at an all time high
- could head much higher
- could stagnate for 5 years.
Dollar cost averaging has purpose:
To average out your money by making use of the volatility so that you don’t invest at the high.
Thus you will not suffer from psychological trauma when you cannot come to terms with an underwater portfolio.
It is implemented because the future is uncertain, you do not know the direction of how the market will go.
There is also the conspiracy theory that, financial institution know most people do not have a lump sum to invest thus dollar cost average creates a plan so that you can regularly contribute and then they can earn from it.
In general, most people have to start their wealth building from somewhere with 0 capital. Thus, to a certain extent, dollar cost averaging is rather valid.
Why Dollar Cost Averaging is Generally Worse Off Performance Wise
Dollar cost averaging when compared to lump sum is worse off because, in general, markets eventually end up higher.
When you spread your money out over x period, you have less money in the market at work, compared to lump sum, which has the full sum at work.
If your financial assets pay dividends, the lump sum are likely to earn more dividends over a 10 year period then dollar cost averaging where the money will only be as big as the lump sum at the end of the 10 year.
For both lump sum and dollar cost average, if at the end of 10 years, the price is underwater, the dollar cost average portfolio is better since the average cost is lower, but it is still underwater. You suffer less psychological damage. But you are still underwater.
If the market moves sideways, the result is almost the same.
Thus, where the market ends up at the end is the determinant of results.
When Dollar Cost Averaging Shines
Lump sum will be worse off if there was a 50% drop, and the markets take a long long time to recover back. Markets like 1929 is a good example.
The above price chart is that of Dexus Group a good performing industrial REIT in Australia. I know, its not an ETF, but you will realize if you have a Australian REIT ETF the aggregate price is not too different.
If you lump sum invest at that point, you gonna go through that psychological trauma of sitting through 10 years of not recovering to the high, seeing your dividends cut drastically.
This looks no different then the STI ETF from 2007 to 2017.
I happened to read Actuary on FIRE’s post on Dollar Cost Averaging, and he presented some good data points on Dollar Cost Averaging.
You will see, most of the time, a lump sum strategy works better after studying many historical iterations. This is because generally markets over 10 years head higher.
However you have some periods where the market over 10 years did not head higher. The great depression and 70s Stagflation are 2 of them.
How many periods was the conclusion drawn?
Now the writers recommendation is flawed because he back tested one sample period, and conclude that dollar cost average does not work that well.
My sensing is that if you wish to make that kind of solid conclusion, you had better test through many periods.
I wouldn’t call DCA overrated just based on 1 ten year period.
In Actuary on FIRE’s study, he studied data from 1802 – 2015 (213 years). Not just that, he went through dollar cost average over 2 years to 30 years.
How do you effectively invest below the average?
Using that same period, he concludes if you invest below the average more, your results will be better.
I cannot fault the logic that if you invest below the average prices and when the price moves above the average, your results would be immensely better than a periodic investment regardless of price.
The question is how do you know what is “average”?
For the STI, that is if the market is always oscillating like the past 20 years. However, if the price trend is generally upwards, how do you implement this?
If you implement this strategy, and only invest in the STI ETF if it is below say, 2.70, what happens if it breaches 3.40 and stay up for the next 5 years?
The average might always be lagging, meaning in a rising market, it might be a long time before you revisit the average price.
Your capital stays out, and your capital is not in the market, and thus your capital is not participating in the growth.
In 2014, I profile a personal research article by a guy name Gaspar Fierro.
Mr Fierro tries to pit a buy and hold investor versus one of the best systematic market timer there is.
His results was shown in the table below:
The result surprised Mr Fierro, Ben Carlson and a lot of people. The market timer didn’t beat the buy and hold guy by a lot.
This is done on a lot of different markets (which is good!)
Why was the result so close? It is likely because the market timer couldn’t always invest as the market kept rising.
The buy and hold investor have more capital and also more opportunities to collect the dividend.
The writer’s invest at below average price is somewhat similar to the market timer’s situation.
Different Testing Duration Matters
Actuary on FIRE also pointed out that different duration, will have different conclusions whether lump sum wins out dollar cost averaging.
If the duration is short, or less than 20 years, there are probability dollar cost averaging would do better. I believe this is because there are more instances where the market either stagnates or returns are more volatile and heading nowhere after a large price drops.
It is interesting that I have done the same dollar cost average profile of the STI ETF, but ending end of last year (2017).
The article ends in 2016 instead.
The difference one year makes is big because 2017 was a good year. The 2.6% returns would have become closer to 5.5%.
Value Averaging Can be Challenging As Well
So if implementing based on average price is challenging what about if we average in based on value?
We can compute the PE or price earnings of the market and buy only if its below a particular PE.
In the retirement planning world, they did consider changing the portfolio allocation based on the CAPE ratio or the cyclical adjusted price earnings.
However, if you are using the CAPE ratio to determine staying invested, what happens if, since the 1990s, the CAPE ratio have been above average 95% of the time. And there have been massive gains.
When do we get invested?
The conclusion is that probability and return wise, lump sum investing is better.
However, you may just be afraid that when you start investing, you are investing at that peak.
Our psychological threshold are all different. And our capital as well.
If you are near the start of your wealth accumulation, don’t worry about this. If you put in $10,000 and its all your net worth now, and there is a 50% shave, you know that you will have $190,000 you will earn and put away for wealth building in the next 15-20 years.
Having unrealized paper losses of $5,000 is not going to kill you in the grand scheme of things.
However, if you are a 55 year old that is closer to retirement (65 years old), you might wish to hedge against the few rare scenarios where lump sum investing lose out.
You might wish to split your $400,000 in 3 batches and invest them over 3 years. For sure you might be super unlucky that the peak takes place in the fourth year. But that first 2 batches might have grown significantly (due to the market run up) that when the plunge happens, things are not so bad.
Dollar cost average still has a place because we do not know what the future holds, and to do a systematic value averaging and investing below the peak can be rather challenging.
To get started with dividend investing, start by bookmarking my Dividend Stock Tracker which shows the prevailing yields of blue chip dividend stocks, utilities, REITs updated nightly.
Make use of the free Stock Portfolio Tracker to track your dividend stock by transactions to show your total returns.
For my best articles on investing, growing money check out the resources section