For a long time, investing for retirement followed a simple formula: Subtract your client’s age from 100, and put that percentage of his or her portfolio into stocks, with the rest in fixed income. Then along came target-date funds, with their automatic adjustments as the client ages, and made things even simpler.
But the notion that an older client should gradually move out of equities has come under fire recently. There is a growing debate over whether the old formulas still make sense, with new glide path expectations focused on actually raising the percentage of a portfolio in equities as a client ages.
The key argument came last September in a blog post by Michael Kitces, a partner at Pinnacle Advisors and writer on a broad range of issues regarding financial planning. Kitces was drawing on research from Bill Bengen on safe withdrawal rates, with the crucial insight being that retirees needed to plan based on worst-case investment scenarios.
Kitces built on this by putting forth the idea that it was helpful to think of “bucket strategies,” stages of retirement where different sources of income might be required. The first five years of retirement might be covered by cash reserves and the next five by fixed income. But after ten years, it would be helpful to have the growing assets that an equity portfolio can provide.
This argument had actually been made a long time ago. Back in 2001, a paper in the Journal of Financial Planning asserted that replenishing the buckets for cash and fixed income can reduce a client’s income. Retirement income is actually enhanced if the fixed income side is annuitized, and the rest of the assets are put into equities. That’s the research Kitces was building on.
Research Magazine picked up the ball from there, with an article in its March 2014 issue. The piece clarified that the glide path should, according to Kitces, increase from about 30 percent equities to 60 percent equities once the client is in retirement. It also laid out some of the basic math explaining why income needs to be more guaranteed through the early years of retirement. A loss to a retirement portfolio can be catastrophic early on, when avoiding risk is at its most crucial. Of course, there has been dissent against Kitces’ thesis. Leading the charge has been a writer and business professor named Moshe Milevsky. One of Milevsky’s arguments is that stocks are obviously riskier than bonds on a one-year time frame, so they must also be risker on a long-term basis.
Morningstar’s John Rekenthaler has also disputed Kitces’ findings, saying he hadn’t proved his thesis. Rekenthaler’s position: a glide path beginning at 10 percent in stocks and ending at 40 percent would not serve a retiree’s needs as well as a traditional declining glide path. The latter would start at 40 percent in stocks and decline to 10 percent, though both average out to 25 percent equities.
But Rekenthaler did allow for two important points behind the argument. He pointed out that if the retirees’ withdrawal rate is higher than the customary 4 percent that most advisors recommend — even if it’s just 5 percent annually — then an increased allocation to stocks makes an awful lot of sense. In his blog post, Rekenthaler called a 5 percent withdrawal “highly aggressive” and said it might make sense as “the starting point for a variable-withdrawal approach.”
Rekenthaler also suggest that a rising glide path makes sense for those who already have a substantial minority of their assets invested in equities. Morningstar’s calculations showed that when stocks approach 45 percent of a portfolio, the rising equity glide path begins to post a slightly better result than the declining glide path.
So what’s the optimal glide path? In the paper he wrote on the topic with Wade Pfau, a professor of retirement income at The American College, Kitces argues for a starting point in retirement of about 20 percent to 40 percent in equity allocation, with a gradual increase of 1 percent per year to a maximum 60 percent to 80 percent allocation to stocks by the end. Given that he has set the terms of this debate, that has to be the starting point for any further discussion.