Home » Blog » Future Return Forecasts – Expecting the Unexpected

Future Return Forecasts – Expecting the Unexpected

Every year around this time I update my summary of expected return forecasts for stocks and bonds made by various financial companies.  I started doing this in 2018 for my own benefit, but I recently got a message from the blogger Froogle Stoodent saying that my forecast summary was a “public service”.  So, it’s nice to see that at least a few people share my interest in this topic.

I also use this information to update my regular Mindfully Investing article on expected future returns and risks.  But I probably won’t get to that for another week or so.

I’ve found over the years that predicting the future, at least in the world of investing, is a fool’s errand.  But paradoxically, mindful investors also know that building a forward-looking investing plan is one of the best ways to avoid costly investing surprises.  Specifically, an investing plan should include estimates of the returns you can reasonably expect from your specific investment portfolio.  Such forecasts help determine whether you’re investing goals, like retiring by a certain age, are realistic or not.

Unfortunately, individual investors often make overly optimistic assumptions about expected returns.  For example, the U.S. has one of the best long-term track records for stock returns in the world, with a historical average annualized return of about 9%.  But a recent survey by Natixis found that individual investors with more than $100,000 in investable assets were assuming future annual returns that are nearly twice that high (17.3%!) as shown in this graph.

And individual investors’ expectations have nearly doubled since 2017.  That’s most likely because an investment in the S&P 500 has produced a whopping 15.9% annualized return over the last four years.  It seems that investors expect the good times to keep rolling on forever.  But Buddhists like to point out that nothing lasts forever; every trend changes eventually.

The above graph also suggests that financial advisers’ expectations are more realistic at around a 6.7% annual return.  But as you’re about to see, that’s still pretty optimistic in light of the return forecasts available from the large financial firms.

2021 Forecasts

First, I should note that I found a lot more return forecasts this year because I discovered that the finance code word for expected return is “Capital Market Assumption.”  This is a good reminder to all readers that I’m a retired scientist just trying to figure all this stuff out myself.  I’m sure any Certified Financial Planner would have been aware of this jargon, but I was not (until now).  Regardless, I’ve noticed that these firms tend to run in a pack and, and except for a few outliers, adding more estimates did not drastically expand the overall range of this year’s forecasts.

This year, I also found a survey of investment firms by Horizon Actuarial Services that was incredibly useful.  The Horizon survey provides an overall average of the return forecasts from 39 firms, which includes some, but not all of the individual forecasts I found independently.

This plot shows long-term annual return expectations for various asset classes from 17 individual firms plus the Horizon average across 39 firms.  (Click on the image to enlarge it.)

All returns are on a nominal basis, meaning they are not adjusted for inflation.   I highlighted the Horizon survey overall average with a pink line connecting their forecast dots.  And as you would expect, those average forecasts fall in the middle of the pack for all the asset classes.  I also highlighted the forecasts from GMO with a red line because GMO is perennially the most pessimistic forecaster, and this year is no different.  In contrast, no firm was consistently the most optimistic across all asset classes.

There are too many estimates this year to discern each firm’s estimate in the above graph.  So, here’s a complete downloadable table containing all the forecast data I found, links to each firm’s forecasts, and some additional comments and notes.  (Click on the image to enlarge it.)

Click here to download an excel version of the table including the links to all the forecast sources.

As you can see from the “max. period” column, most of these forecasts cover the next 10 years, but some of the forecasts cover periods between 5 to 15 years.  The tan rows at the bottom of the table are all volatility (standard deviation) estimates, which I will discuss a little later in this post.

Each firm has its particular preferences for defining asset classes.  In most cases, “Non-US stock” estimates are for large-cap developed market stocks that exclude the U.S.  However, in some cases, the estimates might be for entire developed markets (including mid- and small-caps), global markets (that may include emerging markets), or other broad measures that include the U.S.  Similarly, some “U.S. Small Stock” estimates include mid-caps as well.

For bonds, I tried to collect estimates for 10-year government bonds or intermediate-term government bond funds wherever possible.  But in some cases, only “aggregate” bond fund estimates, which include some corporate bonds, were available from some of these sources.

2021 Themes

It seems that professional investors are assuming that the long-term annual returns for U.S. Large-cap stocks will be somewhere between about 0% and 7%, well below the 9% historical average.  The Horizon survey average for U.S. large-caps was toward the high end of this range at 5.8%.  However, most of the firms are expecting slightly better returns for U.S. Small-cap stocks, Non-U.S. stocks, and Emerging Market stocks.

Unfortunately, like last year, the expected returns for bonds are still dismal.  For U.S. bonds, forecasts of long-term annual returns ranged from about -1% to 3% with the Horizon survey average at a disappointing 1.2%.  Non-U.S. bond annual returns had a somewhat lower range from -2% to 2.5%, although the Horizon survey average was a bit higher at 1.4%.

Forecast Trends Over Time

Similar to prior years, I also examined how much the forecasts have been changing since I started tracking these forecasts in 2018.  This plot compares the 2021 forecasts to the 2018 forecasts for those companies where I found data in both years.

This graph needs some explanation.  The horizontal axis shows the 2018 return forecast, and the vertical axis shows the 2021 return forecast.  Each dot represents one company’s forecast for the asset class noted in the color legend.  The black “match line” indicates where dots would fall if there was no change in the forecasts between 2018 and 2021.  When dots fall above the match line, that means the 2021 forecast was higher than the 2018 forecast—the forecast went up.  When dots fall below the match line, that means the forecast went down.

After a couple of years of slightly increasing stock forecasts, this year most of these companies shifted their stock return expectations downward relative to 2018 as shown by the blue, red, yellow, and brown dots.  However, some of the stock forecasts went up slightly, and a few companies have decreased their forecasts for particular types of stocks more substantially.

Most of the bond forecasts have also shifted downward slightly as shown by the green and purple dots.  Lower expected returns for bonds are not surprising given that bond yields decreased in 2019 and 2020 and remain at less than half of the peak yield seen in 2018.  This has been mostly due to the Federal Reserve reducing base interest rates to essentially zero in response to the COVID pandemic.  Given that bond returns come mostly from yield, historically low yields do not bode well for long-term bond investors.

2021 Risk (Volatility) Forecasts

Many of the forecasts I found for the first time this year included volatility estimates, which gave me a lot more volatility data to work with this year.  Routine stock and bond volatility, as measured by the standard deviation of returns, is the most common measure of risk.  However, mindful investors know that volatility is a poor measure of long-term investing risk.

Regardless, here’s a cross plot of volatility forecasts versus long-term annual return forecasts from the various firms that provided both estimates for any of the asset classes.

The colored ovals around the dots help show where the various asset classes are grouped.  In most cases, Non-U.S. and Emerging Market stocks are predicted to have higher volatility than U.S. large-cap stocks, which is the pattern typically seen in historical data.  But some firms predicted substantially lower foreign stock volatility over the next few years.  And almost all the firms predicted higher volatility from U.S. Small-cap stocks as compared to U.S. Large-cap stocks, which again, is consistent with the historical data.  And, notably, some firms predicted that U.S. Small-caps will suffer from both high volatility and relatively low returns.

The Uncertainties

Given that it’s impossible to predict the future exactly, we should pay attention to the uncertainties surrounding the forecasts.  For example, here’s a Vanguard graph showing the 25th and 75th percentile bands around their forecasts for the 60/40 stock/bond portfolio (top), U.S. stocks (bottom left), and Non-U.S. stocks (bottom right).

The solid lines show the actual returns from the period 2010 to 2020, while the dotted lines show the forecasts including from present-day until 2031.

Focusing on the bottom left graph for U.S. stocks, we can see that the actual returns over the last 10 years were almost continually above the 75th percentile estimate.  As of 2020, the 13.4% annual return was way above the 75th percentile forecast.  Conversely, the bottom right graph for Non-U.S. stocks shows the actual returns being way below the 25th percentile estimate for the last few years.  This is yet another lesson that although these return forecasts are helpful, it’s prudent to expect the unexpected.

It’s also notable that the band from 25th to 75th percentiles covers only 50% of the distribution.  So, the entire range of uncertainty with these estimates is much greater than shown in these graphs.  For example, two years ago Vanguard reported that their 5th and 95th percentiles ranged across 12 percentage points!  Although many of the other firms don’t include such uncertainty estimates, it’s fair to say that the uncertainties with all these forecasts are likely similar to Vanguard’s findings.

Conclusions

At the start of this post, I noted that individual investors had wild expectations for future returns, while their advisers’ expectations were tamer.  However, on balance I’d say that the advisers’ expectations of 6.7% were still very optimistic considering that most advisers don’t often recommend a portfolio of 100% stocks.

For example, U.S. pension funds on average currently hold more than 50% in “fixed income” (mostly made up of bonds and bond funds) in their portfolios.  Using the average Horizon Actuarial forecasts from above, here’s the return math for a portfolio containing 50% U.S. bonds and 50% U.S. stocks:

  • U.S. Large-cap stocks – 5.8% annual return x 50% portfolio = 2.9% return.
  • U.S. Bonds – 1.2% annual return x 50% of portfolio = 0.6% return.
  • Total return – 3.5% for the combined portfolio.

These same pension fund managers report that on average they expect to achieve a 6.2% annual return over the long term as shown in this graph.

The data I summarized today suggests that only Emerging Market stocks have much potential to exceed 6.2% annual returns.  So, the big question is: where do these professional fund managers expect their outsized returns will come from?

Checking these forecasts against my own investing plan tells me there’s no need to revise my return assumption of about 5% annual because my portfolio is 100% stocks with 40% invested in Non-U.S. stocks.  But if you’re one of those optimistic investors who are hoping for annual stock returns of 17% for the next decade, you should be seriously downsizing your expectations and revising your investing plan accordingly.

4 comments

  1. This is great work!

    Nice catch on the inconsistency between what fund managers are expecting, and the returns that forecasters are predicting.

    As John Galbraith said, “The only function of economic forecasting is to make astrology look respectable.” He is, of course, right on.

    Your ‘wisdom of the crowds’ approach here is probably the best forecasting tool we have–which is why your work here truly is a public service 🙂 And, of course, you can bet that history will show these forecasts to be woefully inadequate.

    • Karl Steiner says:

      I agree that the future will likely reveal something significantly different than these forecasts. And we have no way of knowing in advance whether the forecasts are way too low or way too high. On the other hand, if enough people (or monkeys) make a forecast, one of them is bound to be right!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.