Investing Showdown: Stocks or Bonds for Ultimate Asset Growth?
If you’re an investor, you have probably made a decision about how to allocate the percentage of stocks vs bonds in your portfolio at some point.
This is known as asset allocation and it’s an important concept for increasing and preserving the value of the assets in your portfolio over time.
A common approach is to choose a ratio based on your risk tolerance and your age and adjust the level of stocks and bonds you own accordingly.
Typically, this sort of approach results in a split of stocks and bonds represented by two numbers that total 100%.
So, a 60/40 portfolio would contain 60% stocks and 40% bonds while an 80/20 portfolio would contain 80% stocks and 20% bonds.
And even though those are generally good approaches, I’m going to show you why you might want to base your own asset allocation on a completely different factor.
The Data
Let’s jump right into the data so I can begin to explain why I think there’s a better way to approach asset allocation.
The first thing I want to show you is a table summarizing the performance of investing periods that were one to ten years in duration, going back to 1928.
In summary, the range column on the left contains the durations of each investment period.
To the right of that are two investment types.
The first is a portfolio that mimics the S&P 500 as it would have been all the way back to 1928. The lowest performing period (MIN), highest performing period (MAX), average rate of return, and the number of periods that ended positively or negatively are all listed for each period.
Further to the right is the same information for a blended portfolio of treasuries, corporate bonds, and muni’s.
Now let’s look at this same data summarized in a graph:
Each clustered column in this graph represents all the one year, two year, three year, and so on, periods of performance for the S&P 500 (in red) and a mix of bonds (blue) since 1928.
For example, the very first column on the left illustrates the Min and Max values of all the one-year periods since 1928.
In this case, the high for stocks was 52.56% while the best year for bonds was 24.15%. Naturally, the longer each period is, the smaller the range of outcomes from min to max because it’s averaging the data over a longer period of time.
The line going across the chart illustrates the number of periods that ended above zero (or gained value) vs the number that lost value.
The lines on the top are the number of periods that gained value, while the lines on the bottom are the losers.
One nuance to point out is that there is one less period for each year you add to the length of an evaluated range.
So, there were 97 1-year periods since 1928, but only 95 3-year periods because the first 3-year period wasn’t complete until the end of 1930.
There are 88 10-year periods represented here.
Volatility vs. Risk
Before we move on, I want to highlight a common error that investors make when they see data like this.
That is, that they mistake volatility for risk.
Here’s what I mean by that.
Obviously, the value ranges for bonds is far more stable or less volatile than that of stocks. Bonds just don’t jump around in price nearly as quickly or drastically.
But that doesn’t necessarily mean that bonds are lower risk.
If you own a portfolio completely full of junk bonds, you may have very stable yields for a time, but if the issuer of those bonds goes bankrupt, then your yields are gone along with the issue price of the bond.
It is absolutely true that volatility is a component of risk, but it is not the only factor. Do not assume that bonds are always less risky than stocks. That is simply not true.
We will come back to this idea frequently as we continue.
Observations:
Based on this data, here are five quick observations I’ve made that I want to highlight before summarizing several conclusions I feel this data is leading us to.
1) It is risky to own stocks for five years or less.
If you had owned a basket of stocks for any 1-to-5-year period going back to 1928, you would have lost money 6.5% to 26.5% of the time.
I realize that 6.5% may not seem like a high probability but given the potential impact of a delayed or less-exciting retirement, it would be a lot to risk.
I, for one, would not take a 1 in 15 chance of my retirement being delayed when the fix is to simply derisk my portfolio.
Or maybe your kid will have to take out student loans or you won’t get to buy the house you’ve been saving for.
These would be costly outcomes if markets suddenly turned south.
2) It’s risky to own bonds for less than two years.
Consistent with the idea that bonds are not always “low-risk”, there is a 10.6% chance (11 out of 97 periods) that your bonds would lose value if held for one year or less.
However, unlike stocks, bonds would have only lost money 1% to 3% of the time if you held for 2-4 years and never for any period longer than 4 years.
Now, you may be asking yourself, if I can’t own stocks for five years or less and I can’t own bonds for less than two years, what do I hold for periods of zero to two years?
I will address that in just a bit.
3) The risk/reward tradeoff begins to strongly favor stocks 7 years.
Of the 91 periods that were seven years or longer, only one lost value (that being the years leading up to the 2008 financial crisis).
Even in that case, the loss was only 1.6%, so I’d hardly classify that as a catastrophic loss.
Also, in the 7, 8, and 9 year periods that lost value during the 2008 financial crisis, if you held on to your investment for just one more year (through 2009) you would have been back to enjoying positive returns in your portfolio.
4) The risk/reward for bonds is strongest in the 2-to-5-year range.
In the 2-to-5-year range, only 1% to 3% of the periods for bonds ended negatively.
Compare that to stocks where 6.5% to 16.7% of periods ended negatively and you can understand why the stability of bonds is favored in this window.
5) Year 6 is where bonds really begin to lose their luster compared to stocks.
The average rate of return for six-year periods is 5.21% for bonds and 11.96% for stocks. That’s more than double the return for only 3.2% greater historical risk.
For many of us, especially those of us in the building stage of our portfolios, this trade off is worth the move toward stocks for money that isn’t needed within the next six years.
Takeaways:
“Gee, Curt. This data is great, but what are you suggesting this means for our asset allocation?”
I’m so glad you asked.
Let’s walk through three takeaways based on the information we’ve been reviewing thus far.
1) A ratio asset allocation may not be optimal
I’ve already hinted at this a bit, but setting asset allocation to a given ratio may or may not be suitable.
For example, assume your portfolio is valued at $2,000,000 and you plan to withdraw $80,000 each year for annual living expenses.
A 70/30 portfolio would be suitable in this case because 30% or $600,000 of the assets that are in bonds would cover 7.5 years of living expenses. It’s a little conservative, but not out of line.
But what if your portfolio value was $5,000,000? Do you really need 25 years of living expenses in bonds?
Or what if the portfolio value was $800,000? Only three years of the portfolio value in bonds seems very risky to me. (Honestly, you just should keep working instead of drawing down a portfolio at 10% annually.)
Yes, in many cases asset allocation ratios will work just fine, but they are too general to be a one-size fits all strategy.
2) Asset allocation should be based on the timeframe of your investment
For me, this is the primary takeaway from this data.
If your goal is to optimize or maximize your portfolio’s value, then you should approach asset allocation based on how many years you have between now and when you need to liquidate the investment and use the money.
From 0 to 2 years, cash or equivalents are the prudent place to keep your money.
From 2 to 5 years, bonds have typically been a good choice, but this could change based on the shape of the bond curve at a given time.
For example, as I’m writing this in February of 2024, yields for 6-month T-bills are earning 3 basis points more than 5-year Treasury notes.
It’s hard to make the case that you should go buy a 5-year note when you can earn a little more with a 6-month T-bill.
Granted, this is a bit unusual, but it is something to watch as you set your own asset allocation.
Here is a scatterplot to illustrate this point…
This first scatterplot captures the performance of stocks and bonds in all 2-5 year long periods since 1928.
A couple of things should stand out immediately.
First, stocks are obviously more volatile than bonds. The range of reds is much greater than that of blues on the graph.
The second point is that the good years for stocks dramatically outpace those of bonds. The reds run much higher in the graph.
However, you’ll also note that there are plenty of periods where stocks lost value (44 total on this chart).
There is enough downside risk here to persuade me to choose an investment other than stocks for a period of 2-5 years.
Now, let’s look at a similar chart for stocks and bonds owned for periods of 6 to 10 years.
You’ll note that there is still greater volatility in stocks (though it is much lower) and that the highest highs for stocks far outpace bonds.
But you should also note that there are very few red dots that land below zero in this chart. There are only 6 periods that lost value compared to 44 from the 2-5 year scatterplot.
My conclusion from this chart is that it would be wise to choose stocks for periods of 6+ years in length because the potential gains outweigh my concerns about periods of negative returns.
So, in summary, for investment periods of zero to two years, stay in cash, savings, CDs, or money markets.
For investment periods of 2-5 years, check out the yield curve and invest where it makes sense (short term debt or bonds).
For investment periods of 7 years or more, invest in stocks. (If it’s 6 years, flip a coin.)
But the next graph might persuade you to invest in stocks.
3) The impact of returns over time
You may have taken note of the higher returns of stocks and questioned whether or not the potential for higher returns is worth it?
To evaluate this, I built a graph illustrating the performance of two portfolios with an identical annual contribution rate of $5,000 annually for 40 years.
The key difference is that one is completely invested in bonds, and the other one in stocks.
As you can see, the stocks did slightly better.
I used the average rates of return for a 7-year period to run this scenario. In that case, bonds earned 5.23% annually while stocks averaged 11.97%.
The final total after 40 years?
$4,261,379.28 for stocks and $672,222.76 for bonds. That’s a difference of $3,589,156.52.
Given this drastic difference in long-term performance, I would argue that it’s much riskier to invest in bonds for 7 or more years.
The risk being, that you won’t have enough money to meet whatever goals you have.
Wrap Up
None of this is meant to criticize the use of asset allocation by ratios. For most people, it’s a very suitable way to guide investment choices and force healthy rebalancing as these non-correlated asset types ebb and flow with market forces.
However, they are not really the optimal way to arrange your portfolio, assuming you want it to grow as much as possible.
For more about investing, check out our investing page or our investing playlist on YouTube.
