Traditional wisdom dictates that a client’s allocation to equities should decline as he or she gets older, but in a session at the IMCA annual conference on Monday, Michael Kitces asked if that was more tradition than actual wisdom.
Kitces said there were two fundamental questions clients asked about safe withdrawal rates: how much they can spend without worrying about the market, and how much they need to save to spend a certain amount.
The average withdrawal rate is 6.5%, but as Kitces pointed out, using the average rate will by definition fail 50% of the time. You shouldn’t care about the average as much as the lows, he said: “How bad does it get if your timing is bad?”
Kitces pointed to several historical market events that led to the commonly accepted 4% safe withdrawal rate: the credit crash of 1907, the market crash that led to the Great Depression in 1929 and the dramatic “stagflation” of the 1970s. A retiree during those events could have safely withdrawn 4%, thus the 4% rule.
However, following those events, markets returned to similar levels each time, suggesting that the withdrawal rate just has to be low enough for retirees to get through the crisis before the markets return to normal, Kitces said.
Kitces suggested a strategy to create retirement income: at retirement, annuitize the bond portion — or part of it — of the client’s portfolio. The annuity provides income for the near term, while the rest of the portfolio continues to grow.
Under this strategy, a client’s equity allocation increases, but Kitces found that portfolios with higher equity allocations fared better than those with an allocation that decreased over time. He examined the effect of several factors: mortality credits in the annuity, sequence risk and rising equities.
He found that mortality credits actually hurt the client until he or she had been retired for at least 30 years, suggesting that the benefit was not from the annuity, but from the great allocation to equities.
He said that there was already 20 years of research that shows the traditional decreasing equities allocation is not as good as many people think it is. A 1996 paper by William Bengen found that declining equity reduces the maximum safe withdrawal rate, but at a 1% decline per year, it may be a reasonable tradeoff if it makes the client more comfortable.
A 2007 paper by David Blanchett tested more than 40 different glide paths and found that a static equity allocation did at least as well if not better than a declining equity glide path.
Kitces noted that neither of these papers tested a rising equity glide path, so he conducted his own study. He tested 121 glide paths with a 4% and a 5% withdrawal rate to measure the probability and magnitude of failure.
Using historical average capital market expectations, Kitces found that a declining equity glide path performed the worst. A static glide path performed better and the rising glide path performed best of all. In fact, he found that the ideal glide path started at about 30% of the portfolio allocated to equities, rising to about 60%.
He noted that he’s still researching how long the glide path should be, but that it should begin before retirement. If an advisor wants to adopt this strategy for a client who is already retired, Kitces said, “Do the same thing you would for a client who is over-risked.”
Kitces recommended that advisors who adopt a rising equity glide path start with half the allocation a client is comfortable with and move up. That way, they never have more equities than they’re comfortable with. It also means the equity allocation is smallest at the pivotal point—when the portfolio is largest and stands to lose the most.
Kitces tested the rising equity glide path against sequence risk, too, and found that a smaller allocation to equities earlier helped because if returns early on are bad, rising equities dollar-cost average into cheaper stocks; and if returns are good early on, it doesn’t impact how much money they client has to live on, only how much he or she can leave on. “You only die with quintuple your wealth, rather than septuple,” Kitces said.
He added that this strategy works best with clients who use a moderate withdrawal rate. He also warned that liquidating too much into cash can be too much of a drag on return.
Kitces continued, “Good results early on mean you’re so far ahead, a bear market has no effect. You already do dollar-cost averaging in the accumulation phase. It works just as well in retirement.”
This theory naturally brought up the question of target-date funds: Does this mean they’re a time bomb waiting to go off? Kitces said that depends on whether they are “to” or “through” funds. “To” TDFs that go flat on equities before retirement might be “OK,” he said: since they have already wound down their equity allocation to zero, you can still implement a suitable rising equity glide path.
“‘Through’ funds are more concerning,” Kitces said, because clients enter retirement with a too-aggressive allocation that declines when it should start more conservatively and increase.