Why Target-Date Funds Might Be Inappropriate for Most Investors

That’s the finding of new research that the National Bureau of Economic Research began circulating in December. The study was conducted by Jonathan Parker, a professor at the Massachusetts Institute of Technology; Aaron Goodman, a Ph.D. candidate at MIT; and Victor Duarte and Julia Fonseca, assistant professors at the University of Illinois at Urbana-Champaign.

Though not all target funds follow the same algorithms for reducing equity exposure over time—known as “glide paths”—they do follow largely similar ones. For investors 40 or so years from retirement, for example, typically around 90% of portfolio investments are allocated to equities. The equities allocation declines gradually to around 50% for investors at retirement age, and to as low as 30% for those well into their retirement years.

Many assume that the proper glide path for a target-date fund is a simple matter of extrapolating historical return and volatility data into the future. In fact, however, dozens of additional factors are relevant when determining whether a given glide path is optimal for a particular investor. Analyzing the interactions between those factors turns out to be “very, very complicated,” Prof. Duarte says.

To illustrate, say you lose your job during an economic recession and stocks plunge. Depending on how many years you have until retirement, the losses in your portfolio might matter less than your loss of income. Thus, determining proper investment allocations needs to take into account not just the stock market’s historical return profile but also your age and the vulnerability of your job to the business cycle. That is difficult enough with just a few variables, but becomes almost impossibly complex when incorporating many additional factors.

Retirees Should Take Risks

Average recommended equity-exposure level in retirement portfolio

What artificial intelligence determines to be optimal

Typical target-date fund

What artificial intelligence determines to be optimal

Typical target-date fund

What artificial intelligence determines to be optimal

Typical target-date fund

What artificial intelligence determines to be optimal

Typical target-date fund

What artificial intelligence determines to be optimal

Typical target-date fund

To overcome this complexity, the researchers turned to a type of artificial intelligence known as deep learning. As Prof. Parker describes it, they used historical data to set up a model of the “game of life”—or, in other words, the “risky economic environment in which people earn, save and invest over their lives.” The AI program they created played this game of life over and over again, and with each successive round, it “learned” what, in light of different variables, were better and worse ways to invest a retirement portfolio. The computer eventually gained insights that up until now were beyond our reach. “No other modeling of equity allocation up until now has taken into account nearly as many variables,” Prof. Parker says.

Though the primary insight of this modeling is that one size doesn’t fit all, the research did reach one conclusion that does apply to all of us on average: The typical glide path used by target-date funds is too conservative starting at the age of 50. In contrast to an equity exposure level that drops to 50% by retirement age and to as low as 30% during retirement, the average recommended equity exposure in the researchers’ model never falls below 60%.

This average “masks considerable variation between different investors,” Prof. Fonseca says. Depending on individual circumstances, the model sometimes calls for equity allocations that are much lower than 60%—just as on other occasions it recommends much higher allocations. Among the myriad relevant factors to take into account when determining the proper equity allocation in a retirement portfolio, the researchers found that the following three play the biggest roles:

• Wealth. An investor with greater net worth is able to take on more equity risk. On average for the very wealthiest of investors, for example, the model found that the optimal equity allocation is close to 100% even during retirement. In contrast, the average recommended equity allocation for those with the least amount of wealth is around 20% during retirement years.

• State of the business cycle. Other things being equal, investors should have higher equity allocations in their retirement portfolios during economic recessions than during expansions. While this might seem counterintuitive, it is consistent with the basic contrarian insight that we should “buy when the blood is running in the streets,” a quote widely attributed to a member of the Rothschild banking family.

• Market valuation. To determine whether the stock market is overvalued or undervalued, the researchers’ model used the market’s dividend-to-price ratio—the higher the ratio, the cheaper stocks are. Based on the model’s findings, investors should have higher equity allocations when the ratio is higher. On average across all ages, the optimal equity allocation when the dividend ratio is near the high end of its historical range is more than 50 percentage points greater than when the ratio is near the low end of that range.

The researchers acknowledge that the dividend-price ratio isn’t a perfect market valuation indicator. For many years now it has painted a far too pessimistic picture of equities, for example. Prof. Fonseca adds that, in this case, the perfect may very well be the enemy of the good, especially since there is no perfect stock-market valuation indicator. The benefits of avoiding the sometimes poor returns following overvalued markets can outweigh the forgone gains during those times the market nevertheless performs well.

In any case, Prof. Parker hopes that this new study will not be interpreted as being overly critical of target-date funds. He says that these funds represented a big improvement over what came before. The point of this new research is that “we can do a lot better. It can’t be optimal to be average.”

Mr. Hulbert is a columnist whose Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at reports@wsj.com.