Numerous online financial content creators claim that stocks can be expected to return an average of 10 to 12 per cent a year. This belief is misguided, and can lead to some questionable advice.
Assumptions about expected stock returns can affect how much people decide to save, how they allocate their assets, and how they choose between alternatives such as investing or paying off debt. Small differences in expected returns can make big differences in financial decisions.
Before debunking it, it’s important to understand the origin of the idea that stocks return 10 to 12 per cent. Going back to 1950 through 2023, U.S. stocks have delivered a nominal – before inflation – return of 11.32 per cent, as measured by the Fama/French Total U.S. Market Research Index, or 11.43 per cent, as measured by the S&P 500.
For the 20 years ending December, 2023, the total U.S. market returned an annualized 9.81 per cent, and the S&P 500 returned 9.69 per cent. The genesis of those often quoted 10-per-cent or higher returns is easy to see in recent U.S. data.
An important point is that you can’t buy groceries with nominal returns; we need to look at real returns. Take the 15 years ending April, 1985, as an example for why this matters: The U.S. stock market returned a nominal annualized 10.58 per cent, but inflation ran at 7.05 per cent. The real return, which is what matters to investors, was tiny.
The real return on U.S. stocks from 1950 through 2023 was 7.63 per cent, and 7.16 per cent for the 20 years ending December, 2023. A real return above 7 per cent is still exceptional, even for the U.S. market. From 1900 through 1950, U.S. stocks returned a real annualized 5.57 per cent.
Context for the difference in returns between these two periods matters. From 1950 through 2023, U.S. stock valuations increased dramatically. Valuations are the closest thing to gravity in financial markets, and high valuations suggest lower expected returns. Looking at history with no context can be misleading.
Some research on U.S. stock returns has suggested that good old-fashioned luck has played a meaningful role. Disasters that could have happened, and have happened to other countries, simply did not take place in the U.S.
Investors learning about the safety of the U.S. market has driven down expected returns, which has resulted in the rising valuations of U.S. stocks. Together, good luck and valuation increases explain about 2 per cent of the historical U.S. equity risk premium for the period 1920 through March, 2020.
The U.S. market has historically been a great place to invest, and it is still an incredible market for many reasons, but that is not a secret. For realized returns to be high in the future, there will need to be more good luck, more rising valuations or some combination – and valuations are already high.
Netting out the 2-per-cent contribution from luck and learning, the real return on U.S. stocks 1920 through 2020 – the period examined by the paper – is 5.28 per cent, a figure much closer to pre-1950 U.S. stock market returns and, as we will see next, global stock returns.
The magnitude of U.S. stock returns has been high enough to be deemed a puzzle, known as the equity premium puzzle. Knowing that the U.S. is an outlier, one of the ways that researchers have tried to resolve the equity premium puzzle is by looking at historical data outside of the U.S. market.
Global real stock returns from 1900 through 2023 were 5.16 per cent annualized. Research drawing on data for 38 developed markets extending as far back as 1890 for some markets uses block bootstrap to simulate developed market returns and finds a median real 5.28 per cent for international stocks and 4.78 per cent for domestic stocks.
That often cited 10-per-cent return for stocks based on the post-1950 period is roughly equivalent to a 7-per-cent real return in the historical data. That is about 2 per cent higher than unbiased estimates of U.S. expected returns, U.S. equity returns before 1950 and global stock returns spanning 1890 through 2023.
At PWL Capital, we have to estimate expected returns to give people financial advice. Our approach starts with the global historical real return from 1900 through 2023, removes the return attributed to valuation changes and then accounts for current valuations.
Following this process gives a real expected return of 4.62 per cent, or a nominal 7.24 per cent assuming 2.5 per cent expected inflation – a number clearly much lower than 10 per cent.
I don’t want to crush the dreams of people banking on 10-per-cent returns to meet their goals, but counting on returns that match the best historical period for the best performing stock market is likely to lead to bad long-term outcomes.
Benjamin Felix is a portfolio manager and head of research at PWL Capital. He co-hosts the Rational Reminder podcast and has a YouTube channel. He is a CFP® professional and a CFA® charterholder.
