Changing Expectations ≠ Changing Allocations
I recently had a discussion with a friend who had some extra money ready to invest. She asked me where to put it. When I asked about prior experience and expectations, she said that her current 401(k) investments “have been returning about 10 percent per year.” She was invested in a diversified U.S. stock and bond portfolio, with a roughly 65 percent allocation to stocks. Her recent experience taught her to expect a comfortable 10 percent return from her investments. And she was prepared to invest more heavily in stocks because of this recent experience.
This example may seem naïve to experienced investors. They should know that returns ebb and flow and volatility is the name of the game, especially when tilting a portfolio toward stocks. But I also recently met with a financial professional who exhibited some of the very same biases. We were having a discussion about market expectations. I said, “When the stock market is priced higher, you should reduce your return expectations.”
See the table below with the Shiller’s CAPE ratio (cyclically-adjusted price-to-earnings) and subsequent stock market returns from 1926 through 2014. The CAPE is only one measure of valuation — I am not arguing that it is the best, but it makes the point that higher valuations tend to be followed by lower subsequent returns over an average 7-year horizon. Since the CAPE and some other measures seem to suggest that the market might currently be on the high end of valuations, he immediately tried to figure out how reduced return expectations should manifest in his portfolio allocation.
These two conversations made me really think about how expectations and allocations are related. First of all, when I say that stock market return expectations should be reduced, I am saying nothing about what might happen in the next year. Although the data show that average 1-year returns are lower when the CAPE ratio is high, the actual correlation is fairly close to zero. There is still a good chance that 1-year returns could be positive. But longer-term return expectations — the ones with which my friend and the financial advisor are concerned — can probably be reduced when prices are higher.
Shiller’s CAPE ratio is but one of several measures that propose to model mean reversion in the stock market. Mean reversion asserts that higher-than-average returns tend to be followed by lower-than-average returns, and vice versa. The problem is that it is difficult to know what mean reversion looks like, in terms of both the duration and magnitude of these cycles, and whether the mean itself can be expected to change over time.
Considering these challenges, it is far from certain that you should change your stock exposure simply because stocks have had a good run. After all, the equity premium — the higher expected return you get from holding stocks instead of bonds — is still presumed to be positive even when stocks are highly valued.
By the same logic, current bond return expectations should also be reduced. We’re at the lowest interest rates in decades, and bond prices move inversely with interest rates. If interest rates move upward, we can expect bond prices to move down.
The mean reversion paradigm suggests that investors should reduce their portfolio return expectations, given that both stocks and bonds appear to be highly valued. However, to repeat, it is unclear what anyone should actually do with that information. If you’ve set a portfolio allocation with a long-term plan, say for retirement, you should stick with that plan unless your situation changes. Market return expectations should not change your situation, but you may want to prepare for years of underperformance as compared with the recent past.
What does this mean practically for financial planning? First of all, if you’re talking to a financial professional and they promise you 10, 8, or even 6 percent returns, you need to take ask some serious questions about the risk factors involved with that investment (that’s a nice way of saying what I really think).
If you’re working on your financial plan for the future, it still depends on your planning horizon. If you’re looking at your account balance today and you plan to retire 10 years from now, don’t use an input assumption of 10 percent growth. It may be unrealistic and it certainly isn’t conservative. A lower expectation about future growth may mean that you should consider saving more today. If you have a longer time horizon, 20 or more years, it’s probably reasonable to use long-term average return expectations. But you should recognize that your money will probably grow below trend over the next five to seven years, if history is any indication.
Finally, if you plan on retiring in the next few years and you’re excited because recent market performance has buoyed your portfolio, take it with a grain of salt. Remember that the money will still need to be invested during retirement, unless you buy an annuity which will reflect today’s low interest rates. Lower expected returns might mean that you should be more conservative with your withdrawal rate, lest you decimate your portfolio too quickly.
My friend with extra money to invest and the financial professional may want to reduce their expectations for the near future. That doesn’t necessarily mean they should invest in a different portfolio.