This article was posted on August 1, 2002.
With some readers questioning "whether or not 4% is safe?" in the face of the continuing market decline, I thought it might be timely to look at some alternative withdrawal rates.
The Retire Early Study on Safe Withdrawal Rates assumes that the portfolio is depleted by the end of the pay out period. In other words, if you want your money to last 30 years, there's only one dollar left in the account after you make the withdrawal in the final year. You're broke in the 31st year.
Texas-based financial planner Jaye Jarrett published a a study in 1999 that looked at inflation-adjusted withdrawal rates where (1) the account finished the pay out period with the same amount of money it started with, and (2) where that ending balance was adjusted for inflation to have the same purchasing power as its starting balance. (In other words, if inflation averaged 3% per annum for 30 years, you'd need $2,427 for every $1,000 you started with to have the same spending power.) Jarrett also examined the traditional approach where the portfolio is depleted by the end of the period. The complete study is available on the web, click here. The table below shows the results for a portfolio of 75% S&P500/25% Intermediate-Term (5-Year) Government Bonds and a 30-year pay out period. Jarrett used the 1926-1998 Ibbotson database as his source for historical investment returns.
The most interesting result of the Jarrett study was how little you need to reduce your withdrawal rate to maintain the starting balance at the end of 30 years. Merely dropping your withdrawal rate from 4.00% to 3.79% was enough to finish the 30-year period with the same $1,000 you started with. Of course, just because this historical analysis shows you'll still have your $1,000 starting balance at the end of the 30-Year pay out period, it doesn't mean it won't fall below $1,000 during Years 2 through 29. The chart below shows the year-by-year balance for the worst case 30-Year period (1910-1940) using the Shiller 1871-2001 database. The minimum year end value was $723 in 1933.
Optimizing Asset Allocations for Safe Withdrawals
Keith Marbach at Jarrett's marketing affiliate Zunna, Inc. published a study in July 2002 that optimizes asset allocations for retirement withdrawals under various scenarios, (Click here to view the report.) He looked at both the standard inflation-adjusted withdrawals and withdrawing a fixed percentage of a portfolio while letting the annual withdrawal rise and fall with the portfolio balance.
For inflation-adjusted withdrawals, Marbach found that higher allocations of bonds slightly improved the safe withdrawal rates for the first two cases where 1) the portfolio was depleted at the end of 30 years, and 2) the original value of the portfolio was maintained at the end of 30 years. Surprisingly, for the third case, 3) where the portfolio value at the end of 30 years was increased to maintain spending power, a 100% stock portfolio was optimal. These results are summarized in the table below.
In cases where a fixed percentage of the portfolio was withdrawn each year, Marbach found that a 100% stock portfolio was optimal. Adding bonds to the portfolio actually reduced the maximum 100% survivable withdrawal rate. It also interesting to note that if you want to protect the ending value of your portfolio to maintain the same level of spending power, the "safe" withdrawal rate is a bit more than 4% (i.e., 4.16%.), though a withdrawal rate that high would leave your spending power diminished during some periods of high inflation. The table below details these findings.
What to conclude from these studies?
While many retirees want to "Die Broke" and spend it all before they expire, reducing your withdrawal rate by a less than a quarter point may go along way to smoothing the ride as you spend down the portfolio. There are worse things than leaving a large estate to a relative or worthy charity.
filename = moresafe.html
Copyright © 2002 John P. Greaney, All rights reserved.