When the Seattle Seahawks had possession of the football at the New England Patriots’ half-yard line with a minute left in Super Bowl XLIX, their coach had to choose between a safe but predictable play or a risky, less-predictable play—at a point in the game when the wrong move could prove fatal.
In hindsight, the emerald-clad Seahawks chose wrong and lost the game—although no one knows for sure if the risk they took was foolish or calculated, or if the observant Patriot defenders simply read an “unpredictable” play and adjusted brilliantly.
Retirees and near-retirees who are entering their own personal retirement “red zone” and choosing a portfolio “glide path” are a little like the Seahawks on February 1. At a time when a financial mistake might throw them for a big loss, they have to decide whether to pass (read: stay invested in equities) or run (shift toward bonds) during retirement.
(Of course, for retirees and football teams alike, it depends partly on whether they’re leading or trailing and by how much—but let’s assume they’re somewhere in between.)
In an article in the February issue of the Journal of Financial Planning, Morningstar’s David Blanchett calculates that most retirees would be better off raising their allocation to bonds during retirement than raising their allocation to stocks—especially if they have a lot of savings and if they hope to leave a bunch of money to charity or heirs.
Sounds reasonable. But in finance as in football, reasonable people can disagree. Readers of Blanchett’s new paper might notice that it refutes a 2013 Journal article by Michael Kitces and Wade Pfau, in which those two friends and colleagues of Blanchett promoted a rising equity glide path in retirement.
“Yes, I think it’s a fair interpretation,” Pfau told RIJ when asked if the new Blanchett paper runs counter to his paper with Kitces. “As a general rule of thumb, he concludes in favor of declining glide paths, at least from a starting point of today’s market conditions.”
The Kitces-Pfau article attracted a lot of attention 18 months ago. It contradicted the conventional wisdom that says you should hold your “age in bonds.” It argued that, as a default strategy, equity allocations should briefly drop during early (to minimize exposure to sequence of returns risk in the “red zone”) and then rise steadily.
“The optimal equity exposure for a portfolio over an accumulation/decumulation lifetime may look less like a slow and steady downward slope, and more like the letter U, in which the stock allocation is the lowest at the point when lifestyle spending goals are most vulnerable to absolute losses in wealth (the retirement transition itself), but greater in both the earliest years and also the latest,” they wrote.
That paper inspired Blanchett to test the equities-in-retirement conundrum using a lot more variables than Kitces and Pfau used. “If I ran their analysis with their assumptions, I’d probably get the same results. But I used different assumptions. I assume that interest rates will increase over time. I use a utility model. I incorporate a bequest preference, where they focus on income for life,” he told RIJ.
“Where they tested only two scenarios, I tested over 6,000. And I found that a decreasing equity allocation was best most of the time. But I also found that an increasing glide path was best in about 10% of the scenarios. And I left out things like allocations to annuities. There isn’t one single glide path that’s good in all cases,” he added.
Blanchett considers his results highly robust. He ran 1,000 Monte Carlo simulations (some people might consider 1,000 too few) on the outcomes of various combinations of nine different glide paths, three different average equity allocations over retirement (20%, 40% and 60%), three withdrawal rates (3%, 4% and 5%), as well as different scenarios for inflation, life expectancy, shortfall risk, and importance of residual wealth.
Blanchett also acknowledged the elephant in the room—the Fed’s zero interest rate policy—and assumed that bond yields will gradually rise in the years ahead.
Which brought him to a conclusion that was different from Pfau’s and Kitces’, though not absolute. “Glide paths where the equity allocations decrease during retirement,” Blanchett’s paper says, “appear to be more efficient when compared to the other three changing glide paths considered, as well as a constant equity glide path.” The strategy of rapidly decreasing equity allocations in retirement turned out to be success in 75.2% of the scenarios considered—the best results of the nine glide paths he tested.
Blanchett tested three different rapidly declining equity paths over a 40-year retirement (80%-40%, 60%-20% and 40%-0%) and three slowly declining equity paths (70%-50%, 50%-30%, and 30%-10%). The 30%-10% and 40%-0% paths were the safest, especially when the withdrawal rates were 4% or 5%. The relatively least-safe path was an increase of equity allocation to 60% from 20% over 40 years, with a 5% withdrawal rate.
Rising equity glide paths were more beneficial than decreasing equity glide paths for certain types of retirees, however. “The increasing fast glide path will become considerably more attractive for retirees… where there is a large Social Security benefit [relative to income needs], a higher withdrawal rate, and a higher initial equity allocation,” the paper said.
It would be “considerably less attractive… where there are lower nominal returns or a bequest preference,” it continued. “An analysis that focuses on metrics like the probability of success, which ignores bequests entirely, is likely to find the ‘increasing fast’ glide path more optimal… [It does well in scenarios that with] “no bequest preference and low withdrawal rates.”
These outcomes are based on an assumption of rising rates. “The relative benefit of a decreasing equity glide path can at least be partially attributed to the return model used for the analysis, which directly takes into account today’s low bond yields but assumes yields eventually drift higher over time,” Blanchett wrote. He also assumes that the equity risk premium over bonds will stay in its historical range and not widen.
Wade Pfau concedes that his paper with Kitces made different assumptions about bonds. “Michael and I did look at three different sets of capital market expectations, two of which did [assume] lower bond yields. But we didn’t use capital market expectations that allow bond yields to rise over time, which is something that David and I are both now using in our newer work. So that could be a factor. I should try re-running the analysis with the newer capital market expectations I’ve been using. I wouldn’t want to conclude that this is the reason for the difference without testing it.”
What if the U.S. enters a Japanese-like era of negligible yields? That wouldn’t be good, Blanchett says. “If interest rates stay this low, then [most] recommendations are too optimistic. We’re at a unique place for the risks of stocks and bonds. There’s never been a time when yields were so low and stocks were so high,” he told RIJ, adding that cash is no refuge. “If you’re in cash right now, you’re actually earning a negative two or three percent a year. That’s destructive of wealth.”
© 2015 RIJ Publishing LLC. All rights reserved.
