by Rob Bennett
The Retirement Risk Evaluator is a second generation retirement calculator. The Risk Evaluator is the first retirement calculator to consider the effect of the starting-point valuation level in identifying the safe withdrawal rate. It is the first retirement calculator to get the numbers right (presuming that the research of Yale Professor Robert Shiller and others finding that valuations affect long-term returns is right).
The graphic below shows the default results for the new retirement calculator (the results that appear before the user enters his own inputs):
|Safe Withdrawal Rate||Reasonably Safe||Likely Success||Likely Failure||Almost Certain Failure|
The calculator identifies the safe withdrawal rate in four scenarios: (1) a case where the P/E10 level (P/E10 is the price of the S&P index over the average of its last 10 years of earnings) at the time the retirement begins is 8, a very low valuation level; (2) a case where the P/E10 level at the time the retirement begins is 14, the fair-value P/E10 level; (3) a case where the P/E10 level is 26, the sort of valuation level that applied for most of the time-period from 1996 through 2008; and (4) a case where the P/E10 level is 44, the highest P/E10 level of record (the one that applied in January 2000).
The safe withdrawal rate is defined as the withdrawal rate that stands a 95 percent chance of working out, presuming that stocks perform in the future at least somewhat as they always have in the past (the effect of valuations on long-term returns was determined through use of a regression analysis of the historical stock-return data). At the lowest valuation level examined, the Retirement Risk Evaluator identifies this number as 9 percent. At the highest valuation level, it identifies this number as 2 percent. At the fair-valuation P/E10 level, the number is 5.4 percent.
These numbers come as a shock to those saturated in the valuations-don’t-matter thinking encouraged by the Buy-and-Hold Model. For a retiree beginning retirement with a $1 million portfolio, a 2 percent safe withdrawal rate means getting by on spending of only $20,000 per year. Contrast that figure with the safe withdrawal rate that applied for the retiree invested 100 percent in Treasury Inflation-Protected Securities, which were paying a return of 4 percent real in January 2000. A return of 4 percent real translates into a safe withdrawal rate of 5.8 percent (the safe withdrawal rate is higher than the return because the conventional way of calculating the safe withdrawal rate assumes that the portfolio value will be reduced to zero over the course of 30 years). The TIPS investor could live on $58,000 per year, nearly three times the amount permitted the investor with an 80 percent stock allocation!
The comparison between the safe withdrawal rates that apply for 80 percent stock investors beginning their retirements at times of wildly different valuations is even more stark. An 80 percent stock investor who retired with $1 million in 1982 (when the P/E10 was 8 ) could spend $90,000 in every year of her retirement with a near-certain confidence that the retirement would survive 30 years. An 80 percent stock investor who retired with $1 million in 2000 (when the P/E10 was 44) could not enjoy full confidence in his plan if he were to spend more than $20,000 per year! The long-term penalty for investing heavily in stocks at times of high valuations can high indeed!
Another compelling way to come to appreciate the point is to consider the safe withdrawal rate that applies for a retiree who invests his entire portfolio in an asset class paying a return of zero. A zero-return asset class provides a safe withdrawal rate of 3.3 percent (because the standard assumption is that the portfolio will be reduced to zero over the course of 30 years). Yet the 80 percent stock portfolio for a retirement beginning in 2000 offered a safe withdrawal rate of only 2 percent! There were only a few months in the 13-year time-period from January 1996 through September 2008 when the P/E10 level dropped below 26. Yet the safe withdrawal rate for a retirement beginning at a time when the P/E10 level is 26 is 3.1 percent, slightly less than the safe withdrawal rate that applies for an asset class providing zero gains indefinitely!
Valuations matter. It is not possible to identify the safe withdrawal rate accurately without taking the valuations factor into account. Please review my Google Knol entitled “The First Retirement Calculator to Get the Numbers Right” for more background on and argumentation of this point.
Not all retirees insist on a 95 percent level of safety from their retirement plans. The Risk Evaluator also reveals to investors the withdrawal rates that are “Reasonably Safe” (these withdrawal rates have an 80 percent chance of working out), the withdrawal rates that yield “Likely Success” (these withdrawal rates have a 50 percent chance of working out), the withdrawal rates that yield “Likely Failure” (these withdrawal rates have only a 20 percent chance of working out) and the withdrawal rates that yield “Almost Certain Failure” (these withdrawal rates have only a 5 percent chance of working out). For example, for a retirement beginning at a P/E10 level of 14, the Safe Withdrawal Rate is 5.41, the Reasonably Safe Withdrawal Rate is 6.01, the “Likely Success” Withdrawal Rate is 6.61, the “Likely Failure Withdrawal Rate” is 7.51, and the Almost Certain Failure Withdrawal Rate is 8.41. Of course, no retiree is going to set up a plan calling for an Almost Certain Failure Withdrawal Rate. The purpose here is to help investors develop a sense of the price that is paid in terms of reduced safety in exchange for using a higher withdrawal rate. Understanding the nature of this tradeoff is the key to successful retirement planning.
The conventional wisdom is that going with a high stock allocation increases portfolio risk for retirees. It’s not necessarily so! The graphic below compares the safe withdrawal rates that apply at a P/E10 level of 8 (we may be seeing this P/E10 level again in the not too distant future) for retirees going with stock allocations of 20 percent, 40 percent, 60 percent and 80 percent.
At this low valuation level, the safe withdrawal rate increases with each upward move in the investor’s stock allocation, moving from 5.3 to 6.6 to 7.9 to 9.1. Investors who avoid stocks when stocks are selling for one-half of their fair value greatly diminish the safety of their retirement plans by doing so. It’s not too hard to understand why. At those prices, the odds of further price drops are small indeed. The Reversion to the Mean phenomenon is like a powerful magnet pulling valuations upward when the P/E10 level is 8. And the return on stocks is about 6 percent real even in years in which valuations remain stable. That’s a plenty high enough return to make for a very safe retirement!
That said, the conventional wisdom applies in spades when valuations are high. The graphic below compares safe withdrawal rates for different stock allocation percentages at the P/E10 level of 26 that applied for most of the time-period from 1996 through 2008:
At this valuation level, the safe withdrawal rate drops with increases in the stock allocation, moving from a high of 3.9 at a 20 percent stock allocation to a low of 3.1 at an 80 percent stock allocation. While this result is in accord with the conventional idea that adding stocks to a portfolio increases the risk associated with the portfolio, it does not offer support for the conventional explanation of why this is so. The conventional idea is that stocks are more risky because they provide higher returns (the theory is that investors are being compensated for being willing to take on added risk). The data shows the safe withdrawal rate dropping as the stock allocation goes higher (in cases in which the valuation level is high) because returns being generated are poor! Investors are not rewarded for taking on the added risk that comes with investing heavily in stocks at times of high valuations, according to the Risk Evaluator. They are penalized! We need to begin examining the concept of an Equity Risk Penalty.