In a new research paper, a former U.S. Treasury official and a former World Bank economist suggest that a blend of mutual funds and income annuities is the best way to manage wealth and provide income in retirement.
Mark J. Warshawsky and Gaobo Pang put six payout methods through thousands of Monte Carlo simulations. Their findings show that, in terms of generating income and leaving a legacy, it makes sense to annuitize 25% of one’s savings at retirement.
Their detailed study, “Comparing Strategies for Retirement Wealth Management: Mutual Funds and Annuities,” appeared in the August issue of the Journal of Financial Planning. Both authors are consultants with Watson Wyatt Worldwide.
The most widely practiced decumulation technique—the 4% systematic withdrawal from mutual funds—emerges as the riskiest (in both the good and bad sense) strategy in the study. The variable annuity with a guaranteed 5% payout for life comes off as suboptimal, mainly because of high fees.
Pang and Warshawsky’s study tends to endorse the recommendations of many income-phase experts, including Moshe Milevsky and others, who have advocated strategies that combine income annuities with mutual funds in retirement.
It makes sense: If risky mutual fund investments maximize returns (they do) and life annuities maximize income (they do), shouldn’t retirees divide their assets among the two in a proportion that guarantees adequate lifetime income while satisfying the need for growth?
Ironically, several firms have offered such tools, but none of them have achieved mass-market success or received the investment in marketing support that the VA+GLWB has gotten. The VA’s advantage—perhaps a deciding one—may be that it comes in a succinct package.
A six-way horse race
Warshawsky and Pang chose six strategies for handling a $1 million investment at age 65, and ran them through Monte Carlo simulations using thousands of different performance scenarios and several different fee and asset allocation combinations. They assumed that all of the hypothetical investors would live to age 100, and that all of their money was in qualified accounts.
“We thought we’d look at the more conventional as well as the more exotic options, so we did a horse race to see their characteristics. It took a long time for us to do this,” Warshawsky told RIJ.
Here’s (roughly) what they came up with:
- A 4.5% SWiP from a balanced $1 million mutual fund portfolio. It paid $45,000 a year to start. Likely wealth at age 100 was about $800,000.
- A fixed single-premium income annuity (SPIA) paying out exactly $67,700 a year, with zero liquid wealth after age 65.
- A variable single-premium income annuity (SPIVA) paying up to $75,000 a year, depending on fees and asset allocation, with zero liquid wealth after age 65.
- A variable annuity with a guaranteed lifetime withdrawal benefit, paying $50,000 per year, with $1 million in liquid wealth at the beginning. Likely wealth at age 100 was about $50,000.
- A $250,000 SPIA paying about $18,000 a year and a $750,000 mutual fund portfolio with a 4.5% SWiP, paying a combined $52,000 to start. Likely wealth at age 100 was about $400,000.
- A phased conversion of $1 million in mutual fund assets to SPIAs between age 65 and age 75, with an average income of about $69,000 and zero liquid wealth after age 75.
As these numbers show, the SWiP strategy offered the best chance of accumulating greater wealth over time but potentially the least income because of its conservative payout rate. The all-annuity strategies offered higher income but no accessible wealth after age 65 or age 75.
The two blend concepts—the mutual fund/SPIA combination and the variable annuity with a GLWB—offered compromises between those two extremes. Of the two options, the mutual fund/SPIA hybrid produces higher average income and higher likely wealth in the long run.
The issue with the GLWB, as many have pointed out, is the costs. If the fees are high enough, they create so much drag on the underlying account balance that there’s no inflation protection for the income stream and little or nothing remains as a bequest when the owner dies.
“It’s a very clever product, with interesting and desirable risk-sharing properties. But the problem is cost,” Warshawsky said. “That was a problem when we first looked into this, and if the prices have gone up since then, it could be more of a problem.”
Warshawsky, who served as an Assistant Secretary of Treasury for Economic Policy in the George W. Bush administration, said that he and Pang were motivated to conduct the analysis by the obvious need among defined contribution plan participants for ways to convert savings to income at retirement.
“As companies move away from DB plans, something will be missing, and that’s a steady retirement income stream,” he said. “We’re not the only ones saying this. A lot of other individuals and companies believe the same.”
© 2009 RIJ Publishing. All rights reserved.
