Academic: More equities near retirement

While many experts are calling for a more conservative investment approach for target-date funds, an Australian academic is arguing that in many cases equity allocations should be increased from current levels for participants approaching retirement.

At last month's joint Department of Labor and Securities and Exchange Commission hearing on target-date funds, also known as lifecycle funds, Michael Drew, a professor of finance and economics at Griffith University, Brisbane, Australia, testified that asset allocations should be based on the size of the participant's account, not on his or her age. Because the bulk of the growth in accumulated wealth takes place in the plan participant's later years, the funds should adopt a more aggressive asset allocation closer to the target retirement date, he argued — the opposite of the strategy employed by most target-date funds.

“The portfolio size effect shows that many of these funds de-risk at precisely the wrong time,” Mr. Drew said in an interview. “Looking back over the past 40 years, it doesn't matter what you do with the assets for the first 20 years, it's such a small amount of money. It's what you do with the largest amount of money that matters.”

Other target-date experts — well aware of pressures on Capitol Hill to restrict equity allocations for funds that are approaching their target date — disagreed.

Anne Lester, managing director and senior portfolio manager of the global multiasset group at J.P. Morgan Asset Management, New York,agreed that shifting to more conservative assets might not enable every participant to reach their retirement goals, but said employers and managers could add fiduciary risk by acting too aggressively. “Each fiduciary has to grapple with that question. We have always felt that the glide path needs to be conservative approaching retirement because the potential negative consequences of very aggressive investing near retirement are too high, with 2008 really illustrating that.”

In a recent paper, “Portfolio Size Effect in Retirement Accounts: What Does it Imply for Lifecycle Asset Allocation,” published in April in the Journal of Portfolio Management and co-authored by Anup Basu of Queensland University of Technology, also in Brisbane, Mr. Drew argued the widely accepted practice of decreasing an investor's exposure to risky assets on the basis of age is flawed.

Messrs. Drew and Basu created two hypothetical investors with a 40-year investment horizon. One contributed to a fund with a lifecycle strategy, in which the mix of assets gradually switched from a stock-and-bond mix to an all-bond-and-cash portfolio for the participant's final 20 years of employment. The second participant followed a contrarian strategy, starting in cash and bonds and after 20 years moving to a 100% stock allocation for the final two decades of employment. For each year, Messrs. Drew and Basu randomly chose a year between 1900 and 2004 of returns from U.S. stocks, bonds cash, doing this 10,000 times to produce 10,000 outcomes.

The result: The median final wealth of the contrarian investor was 12.5% higher than the median outcome for the lifecycle target-date fund participant. The lifecycle strategy outperformed only when the researchers compared the bottom decile of the respective groups (meaning these outcomes have a one in 10 chance of occurring).

“We're not advocating the contrarian strategy,” Mr. Drew explained. “We just wanted to demonstrate how irrational it is to de-risk too soon. Just because you have a birthday, you shouldn't compromise your long-term investment performance. The last thing financial planners would say today is to sell your stocks, but that's what's happening in many lifecycle funds.”

In his research, Mr. Drew also argued that it's very likely investors would select a lifestyle asset allocation model with the sole objective to minimizing the severity of a major economic downturn although, by doing so, they'd also limit their chances of greater wealth.

For example, he points out that if a plan participant met the 10th percentile outcome at retirement, he would be better off only by 8% following the lifecycle model than the contrarian model. If the plan participant met the 90th percentile outcome at retirement, the participant would be 55% better off in the contrarian model than in the lifecycle model.

Don Ezra, global director, investment strategy, Americas institutional in Russell Investments' New York office, counters that individuals are much more affected by the fear of loss than the possibility of doing better.

“I think most people are comfortable with the shape of the glide path,” he said. “Risk, after all, has a friend called pain. What's important is explaining to participants what it is the fund is actually trying to accomplish.”

Mr. Drew, who also sits on the investment committee of QSuper, the A$23 billion (US$18 billion) superannuation fund for Queensland government workers, said he has had discussions with money managers interested in his research.

He said he hopes the next generation of target-date funds will have a “feedback loop” that will take into account the fund's balance when making allocation decisions, and will have a concrete investment goal — say, 12 times pre-retirement salary. Other assets held outside of the fund would also affect the size of that target goal.

“If you're behind your target, you keep risk on the table,” he said. “You de-risk when you're ahead of that target. Likewise, if you have three great years, a feedback loop would take some of the risk off the table.”

Rod Bare, director of asset allocation strategies at Morningstar Inc., Chicago, said the retirement plan industry should continue to look for ways to add one or two more risk profile inputs to the portfolio allocation decision beyond age to keep the portfolio's risk profile in line with the participant's risk profile. “The refinements, however, will likely center around the strong medicine of suggesting higher contribution rates and less around holding equity levels higher for a bit longer.”

Other experts questioned whether target-date funds need much tinkering at all. Joseph Nagengast, a principal at researcher Target Date Analytics LLC, based in Marina del Rey, Calif., said it would be difficult to customize target-date funds based on each participant's account balance and target accumulation goal.

“We have millions of investors with no account balances,” he said. “What better than an age-based account, which is still better than investors doing it on their own? That's why target-date funds were created.”

Edward Moslander, director of pension products at TIAA-CREF, New York, agreed. “Target-date funds were created for those who are disengaged,” he said. “So we agree that feedback loops are important. That said, there are no silver bullets. Target date funds are the worst solution except for all the others.”