One of the longest running debates in the field of investing is whether stocks are indeed long-run winners — a view most famously articulated by the Wharton School’s Jeremy Siegel and his book Stocks for the Long Run — or the quite opposite notion, most associated with Boston University financial economist Zvi Bodie, that time does not lessen the risk of holding equities.
Siegel’s argument implies that advisors should counsel clients to endure a portfolio’s ups and downs for the long-term gain they can expect, while Bodie argues that the concept of “time diversification” violates bedrock economic theories that there is no free lunch and, further, that equities do not exhibit a decreasing likelihood of loss over time.
The idea has significant implications for financial advisors managing investor portfolios. Siegel has long advocated stock-heavy portfolios, while Bodie has frequently, and passionately, warned investors against taking unnecessary risks, advocating portfolios made up of inflation-protected savings bonds (I-bonds) or Treasury inflation-protected securities (TIPS).
Now a significant new study, based on a more complete data set, weighs in on the Siegel side, though not without some sympathy for Bodie’s position.
The paper, called “Optimal Portfolios for the Long Run,” is written by Morningstar’s David Blanchett, Texas Tech University professor and Research Magazine contributor Michael Finke and The American College’s Wade Pfau, and has already struck a chord in the world of academic finance, having been downloaded 850 times (as of this writing), though published for just a month.
ThinkAdvisor spoke about the new research with Michael Finke, in Austin, Texas, for the Retirement Income Industry Association’s annual advisor conference.
In what way does your new study update Jeremy Siegel’s extensive research on stocks for the long run?
Siegel just used U.S. data. One of the most compelling arguments against the concept of time diversification is that it’s just data mining, and we have only a limited data set; and the U.S. might be different; it might be exceptional.
We looked at 20 countries and over 2,000 years of data collectively [that is, over 100 years of returns data for each country].
We estimated based on this big data set what the optimal portfolio should be given different time horizons. It’s one of the most basic questions of portfolio construction, and it’s amazingly understudied. And surprisingly there’s little agreement among academics about whether [time diversification] exists. And the best argument about why it shouldn’t exist is that it essentially gives long-run investors a free lunch. They get the higher return for less risk than short-run investors.
Economic theory, of course, says there is no such thing as a free lunch or everybody would line up for it. But you found otherwise?
The bizarre thing is if you have a 30-year time horizon, equities become less risky in terms of purchasing power than holding 1-year Treasury bills and rolling them over.
It’s not a 100% certain that time diversification is going to work — nobody knows what’s going to happen in the future. The best we can do is gather as much data as possible and make forward-looking estimates as to what an optimal portfolio should be based on the most thorough possible analysis of backward-looking data.
Does your study factor in varying investor goals and risk tolerances?
We assume different levels of risk aversion and we estimate what an optimal portfolio should be for those different investors at different holding periods.
Your average investor is a short-term investor; the average turnover of an equity is a year.
We have evidence from another study we have done that investor risk tolerance changes during a recession, so people are less willing to take risk during a recession. That means that as equity prices fall, that creates volatility in the short run. But when the market recovers, all of a sudden people’s appetite for risk increases and stocks go up. This creates short-term volatility but tends to be smoothed out over time.
What do your findings imply for financial advisors?
In essence, [time diversification] creates a free lunch for long-term investors and a source of value for advisors who can help clients maintain a risky portfolio.
And are there any other critical implications for academics and financial professionals?
Yes. If you believe in time diversification, you have to believe in market timing.
Stocks become less attractive in a bull market [when prices are high] and more attractive in a recession [when prices are low]. So sentiment is the backbone of time diversification; sentiment drives returns. If you look at risk tolerance tests given to investors, on average they’re more risk averse when the stock market is declining.
What do you think economists like Zvi Bodie would make of these findings?
Zvi Bodie would say that if risk is priced fairly and if markets are efficient then [this time diversification benefit] shouldn’t happen. You only get rewarded for taking real risk. But if markets are sentiment-driven, then you’re getting something for nothing.
As an economist, you don’t want to rely on any strategy that gives people a free lunch; if there is one, then everybody will take it.
And you found that long-term investors can be treated to a free lunch?
A certain percentage of the time, a long-term should lose out, but it happens so infrequently.
Meaning that they freak out and sell their portfolios?
So should advisors load their clients’ portfolios with stocks or avoid doing so because they’ll freak out?
Your average risk-averse worker is maybe not the ideal market participant to accept a lot of investment risk. If you think of the market as a way of allocating risk from those best able to bear risk and allocating safety to those who [most require] safety, then one would expect that a middle-class worker would not be among those best able to accept investment risk.
And if you believe in sentiment, then you may also have to recognize that these are among the workers most likely to move to cash during a recession.
But investors who have financial advisors and are more sophisticated are more likely to stick with [an equity-heavy portfolio].
The evidence suggests that you should tailor the equity allocation of a portfolio to the duration of the goal. But remember that nothing is ever certain.
You cannot portray risk as not risky, as Zvi Bodie [would affirm]. But it does appear that historically there has been a free lunch.
Check out these related stories on ThinkAdvisor: