A refreshing example of the importance of proportions can be found at your local pub. While beer is generally made of four primary ingredients — malt, hops, yeast, and water — the specific combination of those core ingredients has an important effect on the end result.
A bright, grassy IPA gets its distinct flavor by leaning heavily on the hops, while a smooth, dark porter relies more on roasted malts. Even with the same foundation, the end experience is very different! So before passing judgment on beer as a whole, it’s important to try a few different things to see what you like. Everyone has different preferences, but there are lots of tasty combinations to suit all types of people.
Recently I received a nice email from Taylor Larimore, the “King of the Bogleheads” and author of the Three-Fund Portfolio. Taylor politely asked me to use a slightly different baseline definition of his portfolio with 80% stocks rather than the 60% I showed before. And more importantly, he requested that I clarify that there is actually no fixed allocation in the Three-Fund Portfolio. Instead, he believes each investor should determine their allocation to the three namesake funds according to their own personal situation.
I always strive to accurately reflect the ideas of portfolio authors, so naturally I’m happy to oblige. And beyond a simple recounting of his statement and update to the Three-Fund definition, I thought this would also be a good opportunity to explore Taylor’s point about variable allocations more thoroughly.
What exactly does it mean to choose Three-Fund percentages to best meet your needs?
Like mixing the same four ingredients in different proportions can make distinct beers, rearranging the same three funds in a portfolio really can create different investing experiences. So let’s order a full flight of Three-Fund data to help you identify your favorite.
How the Three-Fund Portfolio Works
Taylor Larimore is a disciple of Jack Bogle, so it’s no surprise that he promotes the same common-sense philosophy as the Vanguard founder who popularized the concept of the index fund. The Three-Fund Portfolio is thus built around the idea that the best asset allocation is one that is widely diversified, simple, and inexpensive. To achieve that goal, it focuses on three low-cost index funds that are staples among Bogleheads:
- Total Stock Market
- Total International Stock Market
- Total Bond Market
While the proportion of stocks and bonds in a Three-Fund portfolio is variable, Larimore does provide a very specific structure to the stock portion of the portfolio. Within your stock allocation, he recommends 80% domestic and 20% international. So if you have 50% stocks overall, that implies 40% domestic and 10% international. And if you have 80% stocks, it’s 64% domestic and 16% international.
Why an 80/20 split? Larimore sees it as a compromise between Bogle’s recommendation of no more than 20% international and Vanguard’s recommendation of no less than 20%. In his own words, “When 20% is the only percentage of stocks that these two expert sources agree on, I feel comfortable suggesting that 20% figure.”
So interestingly, while Larimore is completely on board with index funds that use market cap weight to track entire markets with the lowest possible costs, he is clearly NOT a cap-weight purist. After all, the US does not make up 80% of the global market. And just like Bogle, Larimore prefers a clear domestic bias over a global cap-weighted fund.
That mild disconnect between Three-Fund structure and popular global market indexing philosophy is a source of some debate, especially among Three-Fund fans outside of the United States. We’ll get to that in a little while, but for now let’s stick to the rules preferred by the author. Pick the percentage of stocks that best matches your appetite for risk, and place the rest in a broad intermediate bond fund. Allocate the stock portion between 80% domestic and 20% international. And stay the course.
The Effect of Stock Percentage on Three-Fund Performance
Comparing a portfolio with a floating allocation towards stocks against other fixed portfolios is a confusing game to play, which is why the Three-Fund Portfolio page sticks to a single allocation starting point. But that doesn’t mean we can’t crack open some calculations and look deeper at other options.
To explore the spectrum of possibilities within the Three-Fund universe, I looked at every combination of all 3 Three-Fund assets in 5% stock intervals while following the same rules set out by Larimore. For each version of the portfolio, I then referenced the inflation-adjusted returns since 1970 and calculated metrics like the average return, safe withdrawal rate, and ulcer index. By charting every version in one image, we can then learn a few things about each option.
Average Return
To start, let’s look at the average return.
Here you can see why many people argue that one should hold the highest percentage of stocks that they can tolerate. The average return of the Three-Fund Portfolio was linearly proportional to the percentage of stocks. So if your main goal is to maximize the average return, the more stocks the better.
Deepest Drawdown
Why would anyone ever want less than 100% stocks? Let’s look at the risk side of the equation by studying the deepest drawdown.
Now it starts to get more interesting. I bet some of you expected to see another straight line where increasing the percentage of stocks also always increased the risk. But it doesn’t work like that.
Above 40% stocks, increasing the stock percentage did indeed increase the deepest drawdown. So one can look at this chart to find the max drawdown they can handle and stop there. However, below 40% stocks, decreasing the stock allocation further actually increased the max drawdown again. That’s likely because you’re just shifting from concentration in one asset to concentration in another, although it’s also notable that the minimum was at 40% and not simply halfway.
Surprised? Real-world portfolio behavior doesn’t always match the simple models in your mind.
If your goal is to minimize drawdowns with your Three-Fund Portfolio, the data suggests that simply avoiding “risky” stocks altogether may not be the best choice. The ideal mix was about 40% stocks (32% domestic, 8% international) and 60% bonds.
15-Year Baseline Return
Next, let’s look at another accumulation statistic — the 15-year baseline return. This is the 15th-percentile 15-year real CAGR looking at every investing timeframe since 1970, and it’s my favorite measure of long-term returns that looks beyond the average to the lower end of outcomes.
Statistics using percentile calculations can sometimes get a little jagged because of how individual datapoints get filtered, which is why this chart is a little more bumpy. But the overall contrast with the average return chart is unmistakable, and the message is pretty important.
Nobody is guaranteed the average return, and once you exceeded about 40% stocks the more conservative baseline return that measures the below-average outcomes flatlined. If your goal is to maximize returns even during painfully long 15-year periods where markets don’t perform as hoped, the data indicates that the Three-Fund Portfolio worked best between 40% and 70% stocks.
30-Year Safe Withdrawal Rates
For the retirees out there, check out the numbers for traditional 30-year safe withdrawal rates.
The numbers are honestly pretty flat overall, which might seem counterintuitive to those who wrongly believe that withdrawal rates are driven by average returns and thus benefit from high percentages of stocks. There was also a clear drop-off in SWRs above 80% stocks, and the optimum was between about 50% and 80%. So if your main goal is to maximize withdrawal rates in retirement, keep that range in mind.
Ulcer Index
Finally, I want to look at one more metric called the ulcer index. This is a composite number that accounts for the depth, length, and frequency of all drawdowns, and is a nice measure of overall investing pain. I find the curve quite interesting. For reference, lower numbers are better.
Here you can see an impressively smooth bowl with a minimum between about 20% and 35% stocks. That’s notably a little different than the 40% ideal for the deepest drawdown alone, which reflects how the ulcer index measures other types of pain as well. And I also find it thought-provoking how the line is not linear, but accelerates as one increases the stock percentage to higher levels.
So if your main goal is to invest in a Three-Fund Portfolio that helps you sleep well at night — not only in temporary sharp drawdowns but also in long, slow times of lost decades of growth — the data suggests that you should target between 20% and 35% stocks.
Lessons for Investors Outside the United States
I mentioned earlier that there is a good amount of debate on the proper way to interpret the Three-Fund Portfolio outside of the United States. After all, it clearly is based in a degree of home US bias that arguably doesn’t apply in other countries. Should investors around the world follow the same advice to invest primarily in their domestic market, or think more broadly with a cap-weighted global stock fund?
By default, all of the Three-Fund international translations on Portfolio Charts follow the same structure as every other portfolio where “domestic” and “international” are always interpreted through the lens of the home country setting. But to expand on that, I’d like to compare other approaches.
Running the same numbers for every country I have on record is overkill, but as one example of the process let’s focus on the United Kingdom. I’m going to show the same charts as before, only instead of looking at just the default 80/20 guideline I’ll include three practical options.
- 80/20: 80% UK stocks, 20% global stocks
- 20/80: 20% UK stocks, 80% global stocks. Think of it as mostly global with a tilt towards the domestic market.
- MCW: 100% global stocks in a simple market-cap-weighted index fund
Which approach worked best? Let’s find out.
Annual Returns
I have to be honest that even I was surprised by the results here. There was really no significant difference in average return between the 3 Three-Fund options across the entire spectrum of stock percentage.
Perhaps that’s just a unique oddity of the UK market relative to the developed world as a whole, or it could also teach us something about the interconnectedness of global stock markets today where international diversification sometimes matters less than you think. It deserves further study, but for today the takeaway is still clearly that average returns in the Three-Fund Portfolio were proportional to the percentage of stocks, even in the UK.
Deepest Drawdown
Now we’re getting somewhere. The UK 80/20 version of the Three-Fund did not exhibit the same 40% hitch in the data, and also showed deeper drawdowns overall relative to less concentrated options. Also, the MCW version was most similar to the US data with an optimum around 35%, which makes sense when you realize that a global stock portfolio is most similar to a US stock portfolio because of the weightings involved.
So if you’re a UK investor interested in minimizing the depth of Three-Fund drawdowns, the data suggests that investing 35% in a single market-cap-weight global stock fund, with the rest in UK gilts, has historically been the sweet spot.
15-Year Baseline Return
This one is a little harder to interpret, as the lines cross in a few places and one could prefer any of them depending on the situation. However, the larger trend of baseline returns leveling out at higher stock percentages generally holds true in the UK as well. So if you’re looking for the best return even in the not-so-good long-term timeframes on record, the data indicates that going overboard with stocks may not help your cause.
Safe Withdrawal Rates
Here’s a result that I bet may surprise some people. While the stock optimum was still in the 75% range similar to the US, check out the Three-Fund stock strategy it was associated with. The 80/20 approach with mostly UK stocks was the best, while a global MCW fund loaded up on US stocks was the worst. What’s up with that?
As I’ve talked about before, safe withdrawal rates are really interesting and often counterintuitive. Because they focus on the worst cases of purchasing power within a specific market, they’re also more sensitive to things like local inflation and exchange rates than some other metrics. And as a result, divorcing the engine of your portfolio too much from your local economy may have unexpected consequences. So while international diversification is often a positive thing, investing local as a retiree may actually be more valuable than you think.
Ulcer Index
Here you can see the familiar bowl shape. Unlike the US where the ideal stock percentages to minimize investing pain were under 35%, in the UK is was somewhere between 35% and 60%. And interestingly, it depends on your stock approach.
If you’re the type to stick to the classic Larimore 80/20 guidance in the UK, then the data suggests that about 60% stocks (48% UK and 12% global) has been the ideal. But if you prefer your asset allocation to lean more towards global markets, then something closer to 35-40% might make sense.
That may feel backward from your impulse, as you may believe intuitively that diversifying your stock allocation beyond your home borders should allow you to increase the stock percentage for the same amount of risk. But that may not be the case in every country, and the interplay between global stocks and local bonds is also more sophisticated than you realize in a regularly rebalanced portfolio.
Smart diversification is simple to do, but sometimes complicated to explain.
Find Your Favorite Three-Fund Portfolio
After staring at multiple iterations of the Three-Fund Portfolio both inside and outside the United States, I admit that the the data complexity can start to feel overwhelming at a certain point. So let’s step back and look at the big picture.
What are the most important lessons from the exercise?
First, I think it’s fair to say that Taylor Larimore is right. Depending on your own needs, goals, and personality, it makes perfect sense to adjust the Three Fund Portfolio to the situation. So don’t just stop at the high-level summary on the Three-Fund page or in any of the comparison charts. Take the time to tinker with the numbers for yourself. Fixating on precisely perfect percentages may not be the best use of your time, as it’s entirely possible that they will shift with more data. But getting in the right risk and return ballpark for your general needs is always a productive exercise.
Speaking of tinkering, I think it’s important to point out that every country is different. While exploring all country options is too much for one article, there are some great free tools that allow you to do the same analysis for yourself. Fire up the Portfolio Matrix, and you can quickly calculate the same metrics for any combination of domestic stocks, international stocks, and bonds in a dozen different countries. So don’t just settle for stale allocation assumptions that may not apply to you. I’ve got you covered with way more numbers than you may even know exist.
And finally, I think an analysis like this reinforces a common theme that pops up often on Portfolio Charts — the unintuitive nature of historical returns. All too often, smart people become fixated on their favorite mental framework for how they believe portfolios should work and neglect to verify how they provably behaved. Maybe you thought increasing the stock percentage always increases drawdown risk, or that higher levels of international diversification surely increases safe withdrawal rates. But once you actually check the record, the data shows that it’s not always true. That’s why I believe resources like Portfolio Charts are so valuable. Theory is powerful and is absolutely something worth learning, but data keeps us honest and incentivizes new ideas to step forward.
Speaking of new ideas, I’ll go ahead and ask what I’m sure everyone is thinking. What’s the best version of the Three-Fund Portfolio?
From the data I’ve seen, anything with stocks above 80% or below 20% starts to fade a bit in several metrics. But between a range of 20% to 80% stocks, I think it’s totally up for debate. As Taylor Larimore might say, everyone is different and it’s reasonable to adjust the allocation to your situation.
To jump start that process, I’ve provided a bunch of reference data for comparison and a number of charts in this article to help broaden your mindset. And maybe most importantly, I also offer all of the tools you need to explore the topic and come to your own data-driven conclusion.
So don’t wait for me to tell you the answer. Jump in and work it out for yourself. In the process, you may discover not only a great portfolio but also a new way to think about investing.
Cheers!
Join the conversation
