Why is simple Dollar Cost Averaging still the king of all investment strategies?
I analyzed the last 3 decades of stock market data to find the best DCA strategy
Thanks to the massive shout-out that Graham Stephan made while covering my analyses in his second video about our findings, we have added more than 2000 subscribers over the past week and we have blasted past the 9000 member mark!
Before going any further, I just want to say thank you to each one of you who took out the time to join us. If you haven’t joined the community, do consider subscribing
Now without further ado, let’s jump into this week’s analysis…
By now we have all heard the virtues of Dollar-Cost Averaging (DCA) and that you should never try to time the market. Basically, it has been repeated ad nauseam that
Time in the market beats timing the market
But what is interesting is that I could not find any research that has been done on the best way to do dollar-cost averaging.
Theoretically, there must be a better way than to randomly throw your hard-earned money once a month into SPY, right?
So in this week’s analysis, we will explore various methods to do DCA and see which one would end up giving you the best returns!
Given that dollar-cost averaging is about holding investments long-term, we need data, lots and lots of data! For this, I have pulled the adjusted daily closing price & Shiller P/E ratio of SPY for the last 30 years .
Now we have to devise different methods to do the Dollar-cost averaging that will maximize our long-term return. We will have different personas for reflecting different investment styles (all of them would be investing the same amount - $100 every month but following different strategies)
Average Joe: Invests on the first of every month no matter how the market is trending (this would be our benchmark)
Cautious Charlie: Invests in the market only if the Price to Earnings Ratio  is lesser than the last 5-year rolling average, else will hold Treasury-Bills 
Balanced Barry: Invests in the market only if the Price to Earnings Ratio is within +20%  of the last 5-year rolling average, else will hold T-Bills
Analyst Alan: Invests whenever the market pulls back a certain percentage from the last all-time high, else will hold T-Bills .
Given that we need to have some historical data before we start our first investment, I have considered the starting point to be 1st Jan 1994. So the analysis is based on someone who invested $100 every month since 1994. In all the above strategies, we will only hold treasury bills till the investment requirements are satisfied. I.e, in the case of Cautious Charlie, he will keep on accumulating T-Bills every month if the PE ratio is not within his set limit. Once it’s below the limit, he will convert all the T-bills and invest them into SPY.
Based on the time period of our analysis, we would have invested a total amount of $33,400 till now.
No matter what strategy we use, the most amount of returns were made by the Average Joe who invested every month no matter how the market was trending. A close second was Analyst Alan who accumulated money in T-Bills and only invested when the market dropped more than 1% from its all-time high.
The least amount of returns were generated by Cautious Charlie who only invested if the PE ratio was lesser than the last 5-year average (basically by trying to avoid over-valued rallies, he ended up missing on all the gains), followed by the Analyst Alan persona who waited for a 10% drop from ATH before investing.
There are some limitations to the analysis.
a. Tax on the gain on sale of treasury bills and transactions costs are not considered in the analysis. Both of these would adversely affect the overall returns
b. Since I am only using the monthly data for the P/E ratio and my SPY investments (due to data constraints), a much more complicated strategy involving intra-month price changes might have a better chance of beating the market (at the same time making it more difficult to execute).
c. While we have analyzed the trends using the last 30 years’ worth of SPY data, the overall outcome might be different if we change the time period to say 40, 50 or even 100 years.
I started off the analysis thinking that it would be pretty straightforward to find a winning strategy given that we are using nuanced strategies instead of randomly putting money in every month. I also checked for various time frames [5,10, 20 years] and various endpoints [Just before the covid crash, after the crash, before J-Pow, etc.]. In none of the cases did any of the strategies beat average Joe in the total returns.
Since this is an optimization problem, I am sharing all the data and my analysis in the hope that someone can tweak the strategy to finally give us that elusive risk-adjusted market-beating returns.
Till we find our King Arthur, all of us average Joes can rest easy knowing that there is no simple trick that can give you a better return than a vanilla DCA strategy.
Until next week….
 The data was obtained from Yahoo Finance API and longtermtrends.net. While the P/E ratio was available for the last 130+ years, the daily closing of SPY was limited to 30 years.
 We are using the Shiller PE ratio - this ratio divides the price of the S&P 500 index by the average inflation-adjusted earnings of the previous 10 years. This solves for the brief period in 2009 when the normal PE ratio went through the roof as the earnings of the companies fell drastically due to the financial crisis.
 We are holding treasury bills as it has the shortest maturity dates and does not have a minimum holding period unlike the T-Bonds
 The 20% cut-off is considered as it would be above one standard deviation from the historical trends
 The idea of investing after the market pullbacks is driven by the following report from JP Morgan which stated that 70% of the best days in the market happened within 14 days of the worst ones
If you found this insightful, please share it with your friends :)
How would you rate this week’s newsletter?