What Long-Term Stock Returns Should I Assume in My Plan?
“If all you have is a hammer, everything looks like a nail.”
We’ve all heard that phrase, alongside the concept of having “the right tool for the job.”
I submit that many people in the Retirement Planning community (especially online in DIYer circles) do not have the right tools or mental models for including long-term stock market returns in their financial plans.
Example 1:
I saw this question on Reddit this week:
Howdy! Currently 26, I can FIRE at 45 if I get 8% returns. What percent returns for the stock market do you guys use planning for 20+ years out? The 100-year average is 10%, but most Retirement calculators have returns set at 6%. The difference between 6% and 10% is huge in terms of how much Money gets built up. I’m not sure if 8% is too hopeful or if that’s realistic for planning purposes. Cheers.
The first 3 answers came in, and I cringed a bit.
Answer 1: For long term like that I always used 7.2% to be pessimistic and because it makes the mental math really easy. At 7.2% your assets double every 10 years.
Answer 2: 7% is typically used (to include the average of 10% plus 3% for inflation). This gives you money in today’s dollars for reference. The money you will actually have will be higher (hopefully) in proportion to increased spending.
Answer 3: Well, consider the fact that the SP500 has returned 37% over the past year, and almost doubled (87%) over the past 5 years. Crashes do happen, as we’ve seen in 2008, 2020, and 2022, so you have to factor in that you may have a down year (~20% loss). But those are typically good opportunities to use any cash you may have (HYSA) to buy solid stocks that may be undervalued temporarily.
Let’s use these three answers to explain the right and wrong ways to think about future stock market returns.
Lesson #1: We Just Don’t Know –> Add Variance
One of the three answers above is better than the other two (Answer 2).
But we need to start our lessons today by addressing the fact that none of the answers highlighted that:
- We have no idea what future returns will be, and…
- While we can use past returns as an intelligent reference point, we should add significant levels of potential variance into those past returns.
I understand the desire for shortcuts…“Let’s just pick one number…how about 7% per year?”…
But such a shortcut blacks-out so much essential information. This article dives into some of those details: Average Returns vs. Actual Returns.
A better approach starts with varying the long-term average. Perhaps it’s 5%, 6%, up to 10% per year over the long run. That seems reasonable.
BUT! We must also vary the actual yearly returns that lead us to our long-term average. We must consider how the sequence of returns risk could affect our portfolio. More so, we must consider how multi-year negative shocks will affect our psyche.
Those realities are not captured when a smooth average is used. That’s a problem.
Lesson #2: How to Capture Inflation?
The second of the three answers above does a reasonable job of capturing the difference between nominal returns and real returns. This article dives into the details: Accounting for Inflation in Retirement and FIRE Planning
Much like varying our stock return assumptions, we must also vary our inflation assumptions.
Lesson #3: Pessimism and Ease
The first answer above reads: “For long term like that I always used 7.2% to be pessimistic and because it makes the mental math really easy. At 7.2% your assets double every 10 years.”
I get it. It’s a convenient short-hand.
But I have two rebuttals.
Most importantly, we need to go back to Lesson #1. Picking a single annual return doesn’t reflect reality.
I also want to address the idea of “intentional pessimism.” I understand the desire for including factors of safety. But I’m quite wary of overly conservative assumptions. Further reading: The Crushing Cost of Conservative Retirement Planning
Lesson #4: Lack of Perspective
Answer #3 above is, by far, the most dubious.
Yes, the S&P 500 has performed amazingly over recent years. If we included dividend reinvestment (which I don’t think our answerer did), recent performance has been:
- 1 Year: 37%
- 5 Year: 110%
- 10 Year: 256%
Amazing!
But our questioner is 26 years old, hoping to retire at 45, and then hoping to live another 30, 40, 50 years beyond. If history is to be our guide, looking at the most recent 1-, 5-, or 10-year timeline is simply inappropriate.
DIYers need to be wary of “the blind leading the blind.” That’s precisely what this answer is.
What Should You Do Instead?
What should you do instead of making these mistakes in long-term stock market projections?
Ben Carlson recently published this article (he was answering a question specific to a 22-year timeline). Take a look, and I’ll some comments below :
First, these are real returns. Inflation has already been factored in and subtracted out. That’s why the average is (coincidentally, haha) 7.2%.
But look at the variance! Many of the 22-year periods that ended in the 1980s saw less than 4% annualized real returns. Many of the periods ending in the 60s and the 90s/early 00s saw greater than 10% annualized real returns. That is a huge variance that’s not captured with the answer “just assume 7% per year.”
And going back to one of my old articles (Decades of Zero Return), we must also be aware of this chart:
While the long-term average is, indeed, our ~10% returns minus ~3% inflation = ~7% real returns, there have been multi-decade periods of zero return. Smooth averages gloss over that fact.
I mean…just imagine it! What would it be like to live through a 20-year period where your supposed 7% annualized return ends up actually being 0%?!
We must do better.
First, use tools and programs to your advantage. A couple weeks ago, I showed you the wonderful outputs from Engaging Data. That’s one of many online websites that can help visualize past market returns and/or illustrate levels of variance in future market returns.
If you’re comfortable in Excel or other spreadsheet software, start there! Build a 50-year retirement timeline, reference a single cell as your “Average Return” (which is easy to change), and then use a RANDOM function to vary each year’s return. It’s a little sticky, as stock market returns don’t follow a Normal distribution, nor a Uniform Distribution. But directionally, you’ll be better served.
Use Monte Carlo analysis. We use it every day at work to conduct retirement projections for our clients.
I don’t want to totally bash “using the average.” If I’m doing literal back-of-the-napkin math for someone, my go-to numbers are 9-10% for diversified stocks, 4-5% for diversified bonds, 3% for inflation. But I know the limits of such simple assumptions, and I know when not to use them. I urge you to do the same.
Thank you for reading! If you enjoyed this article, join 8500+ subscribers who read my 2-minute weekly email, where I send you links to the smartest financial content I find online every week. You can read past newsletters before signing up.
-Jesse
Want to learn more about The Best Interest’s back story? Read here.
Looking for a great personal Finance book, podcast, or other recommendation? Check out my favorites.
Was this post worth sharing? Click the buttons below to share!
