Estimating Future Returns

One of the most important variables in establishing a financial plan is future returns. A few percentage points difference in portfolio return can make or break a plan. Historically, both stock and bond market returns vary significantly from year to year and even longer periods of time can have quite a bit of difference in returns. This creates a problem, as we need to plan today, but will not know what actual returns are until it is too late. As the Jedi Master Yoda said, “Hard to see the future is.” Yogi Berra concurs: “It’s tough to make predictions – especially about the future.” Some hard decisions we can simply avoid, sticking them in the proverbial “too hard pile.” Unfortunately, with portfolio returns we have to take a guess to figure out where our plan will take us. Fortunately, we can make an educated guess using reasonable assumption and math.

A Natixis study in 2021 found that US investors expected long-term stock market returns of 17.5%.[1] One famous radio show host likes to tout 12% returns from a “good growth stock mutual fund,” as if this was a guarantee for all time. This is nonsense. Other people assume that we can simply extrapolate recent returns out ad infinitum. This is a flawed approach because it ignores the change in valuation that is implicit in the historical return.

Return on investment can be broken down into three components. First, there is the current yield on the investment. For example, an investor may choose among an equity investment with a 3% dividend yield, a bond fund that pays a 3% interest rate or a rental house that yields 3%, net of expenses. In each case, the current yield component of the investment is 3%. Growth comes from an expansion in the earnings power of the underlying investment. Stocks represent ownership interest in companies. If the companies are reinvesting in their businesses, they should be able to grow earnings per share, in addition to passing along inflation. Real estate rents generally rise at roughly the rate of inflation, while rent increases could be higher or lower than inflation depending on the trends in the area. Bonds do not have a growth component. The third component of return is change in valuation. This can also be thought of as a change in the rate investors demand. For instance, if I buy a house to yield me 4%, net, and later find another investor willing to accept a 3% yield, I can sell the house for a significant premium to what I paid. (To earn a 4% yield, an investor pays 25x the annual net income. To earn a 3% yield, the investor pays 33x the annual net income.) The investor who paid 25x net earnings and sold for 33x net earnings made a 33% gain on change in valuation, in addition to whatever was earned on the ongoing earning and growth in earnings during the holding period. If net earnings grew by 2% annually, the investor earned 4% current yield, 2% growth, plus 33% on change in valuation. If the holding period was 5 years, the change in valuation averaged about 6% per year (allowing for a little compounding.) This totals 12% annual returns, half of which came from the valuation change. The valuation change did not alter the future cashflows of the house. It simly pulled earnings forward as the prospective investor paid a higher price for a dollar of future earnings (mathematically accepting a lower return.) The new investor will only earn a 3% current yield, so he shouldn’t expect to earn the same 12%. Further, to realize a gain on valuation adjustment, the next buyer will have to accept an even lower return than 3%. If the buyers that paid a price to earn a 3% yield also enjoys 2% annual net earnings growth, the return before change in valuation will be 5% per year. If he holds the property for ten years and sells it to an investor for a 2% yield, there will be a 50% gain on valuation, but if he sells it for a 4% yield like what the first buyer paid, there would be a 33% loss, which would wipe out about 6 years of earnings. The same logic applies to stocks. Returns have been boosted over the last couple of decades by an increase in valuation. This can’t go on forever, because a higher valuation implies a lower forward return. This is a mathematical axiom, not a theory.

Calculating future return truly is as simple as estimating the three components and making a calculation, but this is not necessarily easy. For individual bonds, it’s not hard – we know what the price is, the time to maturity and the amount due at maturity. For a collection of bonds, there isn’t a discreet ending date. The change in valuation component is based on the change in interest rates.

We can reasonably estimate the current yield on an investment. We can also make a reasonable guess about the growth rate in earnings. The hardest question is future valuation. Here, we have some choices. We can use a historical average valuation, assume the current valuation will hold, or make any other estimate we believe is reasonable. To get a range of outcomes we can test a few different assumptions.

We don’t know what the valuation rates will be at any particular point in time. Recently, massive money printing by the Federal Reserve and artificially low interest rates have pushed people to accept much lower prospective returns than historically, driving up asset valuation. Rampant speculation and euphoria over potential returns has also deemphasized current yield. Assuming this historically elevated multiple to hold indefinitely is unrealistically optimistic. Over time, things valuations tend to trend toward their mean. To estimate long-term returns, it is best to use a multiple somewhere around the historical average.

If we have current earnings yield, a reasonable assumption for long-term growth, a time horizon and an assumed exit valuation, future returns becomes a math problem. As an example, the current earnings yield of the S&P 500[2] is 5%. The long-term median is 6.72%[3]. This yield can be returned to shareholders or invested in growth. On average, earnings should inflate as selling prices are pushed upward, so the earnings yield is a real yield. Long-term real (inflation-adjusted) earnings growth has averaged 2.5%.[4] An optimist might conclude that if long-term growth is 2.5%, stocks should earn 7.5%, but that double counts reinvestment, which is necessary for earnings growth. The only yield the investor gets to keep is dividend yield[5], which is currently 1.6%. Adding the current dividend yield of 1.6% to the long-term growth yield of 2.5% gets a 4.1% real total return. Realistically, if companies are earning a 5% yield, growth plus payouts should be at least 5%. They could simply return all of their earnings if there aren’t opportunities to reinvest at a better rate. The 5% earnings yield should be the real return - if valuations stay where they are. If stocks reverted to their long-run median multiple, they would lose 26% of their value. If that happened over thirty years, it would be a drag of about 1% per year. We also face the problem of using one year of earnings in our earnings yield denominator, as economic cycles and other factors can move this number around quite a bit from year to year. Current profit margins are elevated, driving up corporate earnings. Since 2000, earnings have grown at more than double the rate of sales. Further, last year seems to have been well above trend due to pent-up demand from COVID. If we normalize earnings by applying the average margin since 2000, earnings would fall 26%. If we take the forty-year trend-line, earnings would be even a little lower than that. Putting this all together to make a reasonable guess at thirty-year returns, we start with the 5% earnings yield, reduce it for 25% to normalize earnings, generating a 3.75% return, and then subtract 1% annually for valuation adjustment back to average. This gets to 2.75%. We can add back inflation. Long-term inflation is anybody’s guess, but if we assume 2.5%, we get a 5.25% return for stocks. If valuations stabilize at today’s level, we could expect 6.25%. Obviously, there are some assumptions involved, but this gives us a reasonable estimate for planning purposes. This illustration applies to the S&P 500 as a whole, as of now, but the methodology is relevant for valuing any asset or asset class, as long as it has a yield.

Summary – rather than just picking made up numbers for expected returns, planners should use reasonable estimates based on current earnings yields and long-term valuation averages. While the projected results will not be as exciting, the reasonable assumptions will drive a plan that is likely to work if executed upon. Investors counting on above average returns from above average starting points may be disappointed in the outcomes of their plans.

[1] (Goodsell, 2022)

[2] This is a reasonable proxy for the US stock market, as it is the 500 largest companies by market capitalization.

[3] Stock market data from www.multpl.com

[4] I use a very long historical time period to minimize the effects of starting and ending points, as earnings are volatile. Even so, current earnings appear well above trend, which is enough to exaggerate the long-term trend. For instance, if we use the latest 70-year period before COVID, 1949-2019, the growth rate is 2.5%, but if we use 2021 as an ending point it is 3%. Other factors could cause the future to look different than the past. For instance, population growth has slowed. The baby boomer generation would have been a tailwind to economic growth for decades. Globalization, the development of much greater financial markets, four decades of falling interest rates, and globalization all likely contributed to higher growth. While one could fine-tune growth assumptions, I believe 2.5%-3% is a good starting point for estimating future returns for planning purposes. Adding in the 1.6% dividend yield gets us close to the 5% yield I use in the calculation.

[5] This is not an endorsement of dividend investing over total return investing, which is a different topic.

Bibliography

Goodsell, D. (2022, July 23). 2021 Global Survey of Individual Investors: The Next Normal. Retrieved from Natixis Investment Managers: https://www.im.natixis.com/us/research/2021-natixis-global-survey-of-individual-investors#:~:text=Individuals%20surveyed%20may%20report%20average,a%20result%20of%20the%20pandemic.

Disclaimers:

The views expressed are the views of Jacob Rothman on behalf of Rothman Investment Management, LLC through the period ending July 25, 2022, and are subject to change at any time based on market and other conditions. No client or prospective client should assume that any such discussion serves as a substitute for personalized advice from Rothman Investment Management, LLC.

The information herein should not be considered a recommendation to purchase or sell any particular security. The securities and strategies discussed herein are meant to be examples of Rothman Investment Management investment approach but do not represent an entire portfolio or the performance of a Fund or Strategy and in aggregate may only represent only a small percentage of the portfolio holdings. It should not be assumed that any of the securities discussed herein were or will prove to be profitable, or that the investment recommendations or decisions made by Rothman Investment Management in the future will be profitable. Further, nothing in this letter should be taken as financial advice.

The benchmark for US Large Capitalization stocks is the S&P 500, a market capitalization weighted index containing the 500 most widely held companies. 

Past performance is not indicative of future results. Therefore, no current or prospective client should assume that future performance of any specific investment, investment strategy or product made reference to directly or indirectly in this letter or indirectly via a link to an unaffiliated third party web site, will be profitable or equal the corresponding indicated performance levels. Different types of investments involve varying degrees of risk, and there can be no assurance that any specific investment will either be suitable or profitable for a client or prospective client’s investment portfolio. Historical results for investment indices and/or categories generally may not reflect the deduction of transaction and/or custodial charges, nor the deduction of an investment management fee, the incurrence of which would have the effect of decreasing historical results.