How I Made $5000 in the Stock Market

The Surprising Truth About ‘Safe’ Retirement Withdrawal Rates

Sep 29, 2025 03:00:00 -0400 | #Retirement

(Dreamstime)

Key Points

A number of seemingly insignificant details can sabotage an otherwise successful retirement financial plan.

That’s the lesson I draw from a detailed analysis of Bill Bengen’s new book, A Richer Retirement: Supercharging the 4% Rule to Spend More and Enjoy More. Bengen, of course, is the MIT-trained rocket scientist turned financial planner who, in the early 1990s, came up with what’s been called the Bengen Rule. According to it, if the stock and bond markets perform no worse in the future than they have over any 30-year stretch since the mid-1920s, retirees each year can safely withdraw an inflation-adjusted 4% of what their portfolio was worth when they retired. Bengen calls this the Safemax rate.

Even if you haven’t read Bengen’s new book, you probably have read that he now believes the Safemax rate is 4.7%. But what you may not be aware of is how sensitive Bengen’s upwardly revised rule is to the seemingly innocuous assumptions that underlie his calculations.

Surprisingly, the time of year when a retiree withdraws that year’s allotment is crucial. To illustrate, consider a 30-year retirement that began on New Year’s Day 1969. That year was one of the worst possible times over the past century in which to retire, due to high inflation and mediocre stock and bond market returns over the subsequent decade. I calculate that a retiree who didn’t take his first 4.7% withdrawal until the end of 1969 would still have money in his retirement portfolio in 2004—35 years later. But if he instead took out his annual allotment at the beginning of each year, his portfolio would have run out of money in 1995, just 26 years into his retirement.

Yes, this seemingly insignificant withdrawal timing decision caused a nine-year difference in when retirement funds were depleted. Bengen calculated this difference another way: A New Year’s Day 1969 retiree who withdrew his annual allotment at the end of each December could have withdrawn an inflation-adjusted 4.95% a year and not run out of money over the subsequent 30 years, he told Barron’s via email. In contrast, an otherwise identical retiree who instead withdrew each Jan. 1 could only have withdrawn 4.48% a year to have his portfolio last 30 years. In other words, the timing of annual withdrawals led to a nearly half an annualized percentage point difference in the Safemax rate. (Bengen says that the 4.7% rate that he uses in his book is an average of these 4.48% and 4.95% rates.)

What About Bonds?

The performance of intermediate-term Treasury bonds also plays an outsize role in Bengen’s calculations. He relied on the IT bond series in the Ibbotson Associates database, which was acquired by Morningstar in 2005. He would reach significantly different conclusions if he had relied on other well-respected bond databases.

To illustrate, consider the retiree mentioned above who begins retirement at the beginning of 1969 but doesn’t take his first withdrawal until the end of that year. As mentioned above, Bengen calculates that this retiree each year could have withdrawn an inflation-adjusted 4.95% and not run out of money until 1999, 30 years hence.

In contrast, I calculate that this retiree’s portfolio would have been depleted two years earlier, in 1997, had his IT bonds performed as reported in the database maintained by the Center for Research in Security Prices at the University of Chicago. The portfolio would have run out even earlier, in 1994, if the assumed IT bond returns were from a database maintained by Aswath Damadoran, a finance professor at New York University’s Stern School of Business.

These bond return differences will never be resolved once and for all, according to Edward McQuarrie, an emeritus professor at the Leavey School of Business at Santa Clara University. That’s because there is no one right answer to how intermediate-term Treasuries performed in long-ago decades. The U.S. Treasury didn’t begin regularly issuing five-year or 10-year notes until 1976, for example; for years prior, databases are forced to use other bonds of different maturities and rates to interpolate how a five- or 10-year note would theoretically have performed—and such interpolation can only be an estimate.

Other Factors

McQuarrie found other seemingly minor factors that can make all the difference between withdrawal success and failure. One is transaction costs, which are relevant because inexpensive stock and bond index funds weren’t created until the mid-1970s and 1980s, respectively, and before then it would have nearly impossible—and quite expensive—to construct an index fund on one’s own. Another is so-called sequence risk: Rearrange the order of annual returns over a 30-year period, and a retirement financial plan that lasted 30 years becomes a failure. Yet another is the volatility of an asset’s returns: Even if the stock market’s average future return is same as history’s, for example, an increase in volatility could transform previously successful retirement financial plans into failures.

“A substantial withdrawal rate that can be supported on one set of historical data can be shown instead to fail, if small reductions in return are made to the record, or small increases in inflation added,” McQuarrie concluded. “Likewise, small increases in volatility or a small rearrangement of the sequence of returns, or even taking withdrawals monthly rather than annually, may suffice to convert a successful record into a failed one.”

TIPS Ladders

Hope is not lost, however. There is a way to lock in a guaranteed inflation-adjusted withdrawal rate nearly as high as Bengen’s updated Safemax rate. The key is to invest in a bond ladder comprised of U.S. Treasury inflation-protected securities, or TIPS, that mature at regular intervals over the next 30 years. This approach hasn’t received the research attention it deserves because TIPS were only introduced around the turn of the century and therefore there is relatively little history available for study.

TIPS are identical to regular Treasury securities except that the interest rate they pay is indexed to the consumer price index. According to the website TIPSLadder.com, with a 30-year TIPS ladder you can lock in a guaranteed 4.6% inflation-adjusted retirement withdrawal rate between now and 2055. (This withdrawal rate changes daily as TIPS rates change.)

Note carefully, however, that with a TIPS ladder you are forfeiting the possibility that your retirement portfolio will perform much better than history’s worst case. If that were to happen, of course, your portfolio would support a higher annual withdrawal rate, a retirement that lasts longer than 30 years, and/or leave assets to heirs. With the stock market currently extremely overvalued by almost any historical measure, however, and with the threat of higher inflation in coming years, conservative retirees and near-retirees may be willing to forfeit that upside possibility in return for sleeping more easily at night.

Mark Hulbert is a regular contributor to Barron’s. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at mark@hulbertratings.com