Related
Portfolio Strategies
William Bengen’s latest research shows that adding asset classes allows for a higher withdrawal rate in retirement than 4%.
Charles Rotblut leads a class in AAII's new Essential Investing Video Course. Go to https://www.aaii.com/ves for more information and to subscribe.
William “Bill” Bengen is a retired financial planning practitioner. His groundbreaking research into retirement withdrawal rates led to what has been known as the 4% rule (raised to 4.5% and now 4.7%). Cynthia McLaughlin and I spoke to him about the updated insights into withdrawal rates in his latest book, “A Richer Retirement: Supercharging the 4% Rule to Spend More and Enjoy More” (Wiley, 2025).
—Charles Rotblut, CFA
Cynthia McLaughlin (CM): To start, you use the term SAFEMAX in your book. It’s a key concept that may not be familiar to all individual investors. Could you define it?
William Bengen: The SAFEMAX is a percentage applied to the first-year portfolio value that is used to determine the first-year withdrawal rate. It is not used in each successive year; instead, investors simply give themselves an inflation adjustment. The SAFEMAX is meant to represent the highest withdrawal rate you could safely take given the circumstances you’re facing when you retire.
SAFEMAX is William Bengen’s term for a safe maximum withdrawal rate. It represents the highest withdrawal rate you could safely take given the circumstances you’re facing when you retire, according to Bengen.
The 4.7% withdrawal rate is the current Universal SAFEMAX. It is the historical maximum safe withdrawal rate, assuming a 30-year time horizon and Bengen’s diversified seven-asset-class portfolio (55% stocks, 40% bonds and 5% cash). Different allocations or time horizons would change the SAFEMAX.
Charles Rotblut (CR): When we last spoke for the January 2018 AAII Journal article “Insights on Using the 4% Withdrawal Rule From Its Creator,” you had raised the SAFEMAX from 4.0% to 4.5%. Now, your latest book suggests that retirees can start with a 4.7% withdrawal rate. What led you to increase this rate?
Over time, I’ve gone through several cycles of research. During each cycle, I’ve made the analysis of safe withdrawal rates more sophisticated. In this case, I added more asset classes, which increased diversification and returns to the portfolio and therefore had a very beneficial effect on the withdrawal rate. That’s why it went up.
CR: In the past, the only asset classes you worked with were large-cap stocks, small-cap stocks and intermediate-term government bonds, right?
That’s right. That was the 4.5% SAFEMAX incarnation. I added four more asset classes a few years ago—U.S. micro-cap stocks, U.S. mid-cap stocks, international stocks and Treasury bills. So, we now have seven asset classes. It’s still not what most advisers would consider a well-diversified portfolio, but we’re approaching the limits of the value we can get by adding more asset classes. Each additional asset class adds less value in terms of raising the safe withdrawal rate.
CR: So, there’s a point of diminishing returns if more asset classes are added. For retirees, complexity is also an important consideration, particularly in terms of how easy it is to manage the portfolio.
True. Interestingly, there really is no correlation between investment returns and your ultimate safe withdrawal rate, which is surprising. You’d think that higher investment returns would automatically produce a higher withdrawal rate.
The Joker in the deck is inflation. If you have a period of rising investment trends with rising inflation, the rising inflation increases the withdrawals and offsets the benefit of the investment return. There are a lot of nonintuitive things happening in this world.
CR: Several people have been calling for a lower withdrawal rate, say 3.0% or 3.5%. Obviously, you’re going the opposite way. Why do you think the 4.7% withdrawal rate will work despite others’ concerns and calls for a lower withdrawal rate?
It’s important to recognize at the outset that the 4.7% rule is actually a worst-case scenario.
I studied 400 retirees retiring from January 1926 through the end of 2024, starting with the first date of each quarter. Out of those 400 retirees, the 4.7% rule was only applicable to one of them. The other 399 retirees all had higher starting withdrawal rates [Figure 1]. It may surprise people to know that the historical average across all 100 years is 7.0%. There are even folks who are lucky to get 8.0%.
For example, I estimate that the folks who retired during the Great Recession’s bear market, which bottomed in April 2009, would be able to support an 8.0% withdrawal rate from that low point because stocks were so cheap. We haven’t seen anything comparable since then.
I don’t see any justification for going lower than 4.7%. Even if I look at the current market, which is very expensive—and the more expensive the market, the lower the withdrawal rate—the lowest withdrawal rate I’d recommend is 5.25% or 5.50%. That’s still an increase over the 4.7% worst-case scenario, but it’s a lot less than the 100-year historical average of 7.0%.
CR: You’ve stated that the earlier a bear market occurs in retirement, the lower the maximum safe withdrawal rate is. You’ve also tied this idea to Robert Shiller’s cyclically adjusted price-earnings (CAPE) ratio. Could you explain this concept?
The idea behind that is if you have a very expensive stock market, you’re probably not far from a bear market. A major bear market early in retirement depresses the safe withdrawal rate because taking withdrawals when stocks are down really damages the portfolio. Similarly, if you choose to increase your withdrawal rates with inflation, high inflation rates force you to increase your withdrawals, which also damages your portfolio. High inflation damages the portfolio even more severely than the major bear markets.
CM: Why did you choose to use the CAPE ratio?
In the May 2008 issue of the Kitces Report, Michael Kitces published a chart showing the correlation between the SAFEMAX for each year of retirement and the CAPE ratio, or stock market valuation, at that starting time. It’s an amazing chart because the two lines seem almost perfectly correlated in the opposite direction. As stock market valuations go up, the safe withdrawal rate drops, and vice versa.
I was excited when I first saw that chart. However, after I did some work with it, I realized that it didn’t provide enough information to specify withdrawal rates to the extent I like to. It required adding inflation into the picture as a second variable to generate more reliable SAFEMAX estimates.
CR: Should retirees be able to financially endure bear markets as long as they stick with that 4.7% withdrawal rate?
Historically speaking, that withdrawal rate is for an ultraconservative person who doesn’t want to risk anything. That doesn’t mean we couldn’t encounter a period where we might go below it, but I don’t see that on the horizon.
CR: On other hand, let’s say someone is lucky enough to experience booming stock markets and low inflation. If they get through the first 10 years of taking withdrawals, is there a time at which they can look at how they’re doing and adjust their withdrawal rate upward? Perhaps they want to spend more later in retirement.
Yes. No type of financial plan, including a retirement withdrawal plan, is carved in stone. It can and should be flexible to make adjustments during its tenure. Typically, retirement withdrawal plans last for at least 30 years, if not more. That’s a long time. It would be highly suspicious if you didn’t have to make some changes in between.
I developed a methodology that allows me to pinpoint a withdrawal rate that’s appropriate for the environment in which a person retires, but it’s not perfect. Things can happen in the middle of retirement. Let’s say you retired in the late 1950s, and in the 1970s, you suddenly face tremendous inflation that you did not expect. You’re going to have to make an adjustment.
CR: Is there anything investors can use to determine when they might need to make an adjustment to their withdrawal rates?
Part of the retirement withdrawal plan I recommend is developing a template. It’s basically a forecast of your current withdrawal rate for each and every year.
To calculate future withdrawal rates, take your portfolio withdrawals and divide them by the value of your portfolio at the start of each year. In a normal plan, that withdrawal rate starts at about 4.7% and, in 30 years, becomes 100% because the plan assumes that you will take out all your money in the last year.
There should be a gradual increase in this curve of withdrawal rates. To make sure you’re on track, you can compare the estimated curve of withdrawal rates to your actual results. If you fall way above or below your estimate, it’s time to ask yourself if you need to make an adjustment. You won’t always need to make an adjustment, but it’s good to check.
CM: You’ve noticed a sharp increase in the odds of running out of money when the SAFEMAX rises above 5.25% for the 30-year period. Is this a result of having to rely too much on high returns?
Higher withdrawal rates deplete the portfolio faster. So, as you would expect, the higher your withdrawal rate, the shorter the period during which you’ll be able to safely make withdrawals.
CM: With inflation and stock valuations influencing withdrawal rates, are there any general observations you can give regarding factoring each element into an initial safe withdrawal rate?
In my latest book, I developed tables that are sorted by inflation regime—low, moderate or high inflation—and then by market valuation. They show what the SAFEMAX is for a given environment and allocation [Figure 2].
The safe withdrawal rates shown are not supposed to be 100% accurate. We’re dealing with relatively little data for current phenomena, and the markets are changing, so you have to be careful about expectations. Just because you have a safe withdrawal rate doesn’t mean it’s going to last for your whole retirement.
As we discussed earlier, you may need to make some adjustments, and they may not all be negative adjustments. You may actually be able to increase your withdrawal rate. For instance, people who retired into the huge bull market of the mid-1980s got a significant tailwind from investment returns, and they were able to achieve much higher withdrawal rates than I would have estimated based on circumstances at their retirement date.
CM: Your research on the SAFEMAX shows that the safe withdrawal rates are approximately even for stock allocations between 46% and 73% of the total portfolio. Could you elaborate?
That observation applies exclusively to the worst-case scenario—the individual who retired in October 1968. That was one of the first charts I ever created. I put an updated version of it in my book, and that’s when I realized that I was dealing with a phenomenon that had a lot of unexpected aspects to it.
When I originally did the research, I thought that if you increase your stock allocation, your withdrawal rates should go up because you have higher returns. But bear markets intrude. As it turns out, if you keep your stock allocation between about 46% and 73%, you get about the same withdrawal rate no matter what the stock allocation is within that range, which is kind of fascinating.
This shows you the dynamics of interplay between stocks and bonds in a portfolio, particularly during a bear market. If you go too high in stocks, bear markets can really damage your portfolio and reduce your withdrawal rate. Going too low in stocks has a similar effect—not getting enough oomph in the returns reduces your withdrawal rate.
CM: Beside allocating to bonds, you suggest allocating around 5% to money market funds. Since high cash allocations create a drag on returns, is this close to the upper limit?
The higher your money market allocation, the lower your withdrawal rate because it’s a low-returning asset. Even though it has very little volatility, which is great, it doesn’t give you very much return.
You have to find a level of cash you’re comfortable with—ideally somewhere between 5% and 10%. I wouldn’t go much beyond that because I think you’d just be cutting off your nose to spite your face.
I recently read an article about a study into asset allocation in retirement. The authors came to the conclusion that the best allocation is 100% in stocks during retirement—more specifically, 37% in international stocks and the remaining 63% in U.S. stocks. That really surprised me, but they may have a point.
If you take a look at the worst-case scenario, the 4.7% rule, that happened in 1968. It so happens that international stocks did very well during that period. That covered the worst-case scenario and raised all our goal rates.
The question I have is: Who is going to have the courage to have a 100% stock portfolio when the market is down 40% or 50%? Will they be able to stay with their game? I think it’s safer to have a more balanced portfolio and a potentially lower withdrawal rate but know that you’re not going to be scared out of your position when the time comes.
CM: You mentioned that small- and micro-cap stocks boosted the SAFEMAX. Which asset class surprised you the most in terms of its effects?
When I added small caps to the portfolios I tested, they raised the withdrawal rate from about 4.1% to 4.5%, which is a pretty big bump for just one asset class. Micro caps are very similar, but since we already have a small-cap stocks allocation, they don’t provide quite as much of a boost as the small caps originally did.
Additionally, there are asset classes that I haven’t included, such as Treasury inflation-protected securities (TIPS), commodities, precious metals, real estate, emerging markets stocks and so forth—all of which you’ll see in a well-diversified portfolio. I think they’re probably all worthwhile. However, I don’t think they’re going to have a major impact on the withdrawal rates that I’ve developed up to this point. Once again, it’s an issue of diminishing returns.
CM: Thinking about the Financially Independent, Retire Early (FIRE) movement, those who are interested in retiring early may want to weight more heavily to stocks. When is the best time to start moving to a more conservative portfolio? And, when doing so, would you suggest that investors use the 55% stock/45% bond allocation that you discuss in your book?
I suggest that you start about five years before retirement. That’s long enough to cover any bear market that might occur. As far as allocation goes, take a look at where you are and where you want to be at the beginning of retirement. If you want to get to a 60% stock/40% bond allocation and you’re currently using an 80% stock/20% bond allocation, you might want to reduce your stock allocation gradually over the next five years to bring it in line.
You can also do it all at once, but, in my opinion, the longer you can stay in stocks during the accumulation phase, the better. I hate getting out of stocks when you’re trying to accumulate money because it’ll just cost you returns.
Portfolio Strategies
Portfolio Strategies
Portfolio Strategies
DAVE G from TX posted about 1 year ago:
DAVE G from TX posted about 1 year ago:
DAVE G from TX posted about 1 year ago:
BARRY J from TX posted about 1 year ago:
VICTOR S from NC posted about 1 year ago:
ROBERT A from NC posted about 1 year ago:
GARY J from CA posted about 1 year ago:
BARRY J from TX posted about 1 year ago:
JAMES N from VA posted about 1 year ago:
SHEILA A from CA posted about 1 year ago:
BARRY J from TX posted about 1 year ago:
BARRY J from TX posted about 1 year ago:
KEVIN V from NC posted about 1 year ago:
KEVIN V from NC posted about 1 year ago:
GARETH D from MN posted 12 months ago:
ROBERT R from IL posted 11 months ago:
DAVE S from CA posted 11 months ago:
You need to log in as a registered AAII user before commenting.
Log InCreate an account