By Rob Bennett
Buy-and-Holders and Valuation-Informed Indexers have some sharp differences on how to calculate the safe withdrawal rate (the percentage that a retiree can take out of his portfolio each year with virtual certainty that the portfolio will survive 30 years, presuming the stocks will continue in the future to perform at least somewhat as they always have in the past). The Buy-and-Holders say that the safe withdrawal rate is always 4 percent; it makes sense to believe that the safe withdrawal rate is a single number if you believe that the market is efficient. The Valuation-Informed Indexers believe that, since valuations affect long-term returns (as Shiller showed in 1981), the valuation level that applies at the time the retirement begins must be taken into consideration to perform the calculation accurately and the safe withdrawal rate thus may drop to as low as 1.6 percent (in 2000) and rise to as high as 9.0 percent (in 1982).
I thought that in the interests of achieving a greater measure of comity with my Buy-and-Hold friends, it might be a good idea to advance some reasons why the two camps are not necessarily as far apart in their thinking as they sometimes appear to be. I came up with three reasons.
First, the safe withdrawal rate is a highly conservative number. The idea is to look at the worst-case scenario. While a calculation taking valuations into consideration really does produce a safe withdrawal rate of 1.6 percent for retirements that began at the height of the bubble, the fuller reality is that it would not be at all unreasonable for investors to plan retirements calling for withdrawal rates greater than the one that qualified as “safe” under the tough standard that applies pursuant to the conventional definition of the term.
The retirement calculator at my web site identified not just a “safe” withdrawal rate (the rate that has a 95 percent chance of working out, according to the historical return data) but also a “reasonably safe” withdrawal rate (the withdrawal rate that has an 80 percent chance of working out). In 2000, when the safe withdrawal rate was 1.6 percent, the reasonably safe withdrawal rate was 2.6 percent. Today, the safe withdrawal rate is 2.8 percent but the reasonably safe withdrawal rate is 3.4 percent.
There’s nothing “wrong” with an investor choosing to go with a retirement plan with only an 80 percent chance of working out rather than insisting on meeting the dictates of what constitutes “safe” according to most of the existing literature in the field. This is a judgment call that each investor must make for himself or herself. The job of those offering safe withdrawal rate calculations is to help investors to better inform their own choices, not to insist on the highest possible standards of retirement safety.
Second, investors might be notified that, if they elect to go with a weaker form of safety in their retirements than the ideal 95-percent-safe standard, there’s a good chance that they will learn early in their retirements if they are living through one of the return patterns that are likely to bring on retirement failure down the road a bit. It is retirements that take big hits in the early years that collapse, usually many years later. It is foolish to ignore bit hits because the retirements have not yet gone under; it is critical to pay attention to warning signs when there is still time to bring retirements up to a level of reasonable safety. But it should be reassuring for investors to know that, even retirements that are not developed to meet the highest safety standards, can become highly safe so long as they avoid for ten years or so the major hits that signal rocky waters ahead.
Third, at times when stock valuations are so high that reports of the accurately calculated safe withdrawal rate become depressing to aspiring retirees, it is usually possible to increase the safe withdrawal rate dramatically by moving a greater portion of one’s assets to non-stock asset classes. Most conventional safe-withdrawal-rate studies were developed in the 1990s, when stock prices were booming. It was fashionable at that time to suggest that even retirees should go with high stock allocations because the returns from stocks were viewed as being so much higher than the returns available from other asset classes. Calculations that take valuations into consideration tell a very different story.
At the top of the bubble, the safe withdrawal rate for an 80-percent stock portfolio was only 1.6 percent real. But Treasury Inflation-Protected Securities were at the time offering a guaranteed 30-year return of 4.0 percent real. That’s a safe withdrawal rate of 5.8 percent real! The safe withdrawal rate is larger than the return in the case of TIPS because the convention in safe-withdrawal-rate calculations is to assume that the portfolio value will be reduced to zero over the course of 30 year. It is smaller than the return in the case of a high-stock portfolio because heavy losses in the early years of a retirement can wipe out a counter-intuitively large amount of long-term gains and the odds of heavy losses taking place in the early years of a retirement are very high for retirements beginning at times of super-high valuations.
So lowering the stock allocation and increasing the TIPS allocation can increase the safe-withdrawal-rate dramatically. Many investors would not feel comfortable going with a 100 percent TIPS allocation even in the circumstances that applied in early 2000. But even a shift to a 50 percent TIPS allocation would have pushed the safe withdrawal rate up to 3 percent real. (and pushed the reasonably safe withdrawal rate up to 3.4 percent real). Investors are often shocked to hear that the safe withdrawal rate can drop to as low as 1.6 real (this truly is a hard-to-believe number when you consider that stocks earn an average long-term return of 6.5 percent real). It can come as a comfort for them to learn that they have the reasonably appealing option of increasing their TIPS allocation by 30 percent and thereby increasing their safe withdrawal rate by 1.4 percentage points (or by 1.8 percentage points if they are can live with a 20 percent chance that the retirement will fail).
The purpose of including valuation adjustments in safe-withdrawal-rate calculations is not to shock or depress investors. It is to inform them of the realities they face when seeking to put together a prudent retirement plan at a time of high valuations. Ignoring valuations is a reckless, head-in-the-sand approach to retirement planning, in my assessment. But I have seen many investors overreact to the story told by the valuation-adjusted numbers. Investors should be reassured that there is usually a way to construct at least a reasonably safe retirement plan even at times when stock prices have unfortunately gone through the roof.
Rob’s bio is here.