In this series, we provide our readers with two distinct perspectives on the same topic — one from an academic, the other from a practicing financial advisor.
QUESTION: Is the 4% rule dead?
THE ADVISOR — JOE ELSASSER, CFP, PRESIDENT, COVISUM:
The 4% rule is not a rule at all, unless you mean rule of thumb. There have been other “rules” based on similar ideas. Famed portfolio manager Peter Lynch of the Fidelity Magellan Fund famously published a 7% “rule” in his 1995 article entitled “Fear of Crashing.” His argument was that stocks dramatically outperform bonds over time, so an all-stock portfolio, based on stocks with an average dividend yield of 3% and 8% return from capital gains would grow from $100,000 to almost $350,000 after 20 years, and would even survive a 25% crash, with an ending value of approximately $185,000.
Lynch’s argument, which represented much of the feeling of the time, was challenged by Bill Bengen, who is now known as the father of the 4% rule. Bengen changed the discussion by arguing first that a 20-year retirement period was not enough, extending his analysis to 30 years. Bengen used rolling 30-year periods, actual market returns and actual inflation to identify 4% as a safe withdrawal rate. His results showed that if you began drawing on a portfolio on the worst possible day in history and continue withdrawing an inflation-adjusted amount every year for 30 years, 4% was the highest amount you could sustain while still having money left at the end of the 30-year period. He called this withdrawal rate the Safemax.
In short, Bengen’s process for arriving at 4% was considerably more robust and used better assumptions than Lynch’s. As a result, it yielded a much lower safe withdrawal rate. Funny enough, Lynch’s rule was pretty close to a 4% rule, but he assumed that a 3% and growing dividend would always be present.
Now, the 4% is being challenged on all sides, in roughly the same way Bengen challenged Lynch and the prevailing “wisdom” of the day. Bengen chose a 30-year period, but life expectancy has been consistently increasing. Bengen used U.S. stock returns. Is that a fair assumption? Is it likely that the next 30 or 40 years of U.S. returns will parallel our economic history in which stocks climbed from 22% of the world’s total market capitalization to 54%? 1
Wade Pfau evaluated the 4% rule using a global dataset and found that not only is the United States’ experience not representative of the experience of other countries, but it would be dangerous to assume that the rule will work in the future. [PDF]
The practical aspects of implementing a 4% rule become even more tenuous. The 4% rule is a gross number and doesn’t account for fees or taxes, both of which take a significant bite out of most retirees’ portfolios. Also, few retirees actually follow a smooth withdrawal pattern in which a level real amount is withdrawn from the portfolio annually. They buy cars, go on trips early in retirement, pay for college for a grandchild. They may have some income sources like Social Security that don’t start until a little later in retirement, necessitating a higher withdrawal amount early in retirement. These issues are often referred to as the “lumpy” cash flow problem. Whenever there are spikes in an income need, the spikes represent additional risks, particularly if they occur early in retirement.
Although the 4% rule has come to be relied on by many in the financial services world, we need to remember that it was an iteration of prior thought based on better assumptions. New research is challenging those assumptions and coming to even more conservative results.
When we pair new research with the practical challenges of actually implementing a 4% rule, the result should be clear. The 4% rule is dead. In retirement, the stakes are simply too high to rely on a rule of thumb. Strong financial planners use strong tools to come up with reasonable “crash tests” based on the holdings of the portfolio. They consider lumpy cash flows, Social Security and other guaranteed income sources and account for taxes and fees to arrive at a reasonable withdrawal strategy.
1. Global Evidence on the Equity Risk Premium. Journal of Applied Corporate Finance, Vol 15, No 4, pages 27–34.
THE QUANT — RON PICCININI, PH.D., DIRECTOR OF PRODUCT DEVELOPMENT, COVISUM:
Yes, and financial planning software killed it. Other vestiges of a computer-less era are the rule of 72, printed tables of logarithms, stock quotes ending in eighths, and paper stock certificates. As Indiana Jones yelled to Panama Hat about the Cross of Coronado, “It belongs in a museum!”
What is more interesting about Bengen’s original study of the Safemax withdrawal rate is the framework of analysis and the variables to take into account. Let’s discuss a few:
- Life Expectancy: will the client draw on the portfolio exactly 30 years? If the client lives longer, the Safemax goes down.
- Asset Mix: the original study assumes a 60/40 portfolio.
- Interest Rates: They have never yielded so little, especially compared to the sample used in the Bengen study.
- How often is the portfolio rebalanced and at what cost? This will have a big impact.
- What if a different mix is used? This will have a bigger impact.
- What if the stock/bond mix changes along the way? Computers allow you to simulate that.
- Taxes: Bengen estimates the impact of taxes in his 2006 book. At a 35% effective tax rate, the original Safemax of 4.15% is reduced to 3.38%.
- Logic: Historical data shows that you haven’t died, yet it would be foolish to consider yourself immortal. Similarly, simply because a 4% Safemax “worked” on past U.S. data, it may not be wise to assume that it will work in the future. Indeed, if future returns resemble more closely the path of other industrialized nations, the 4% could be extremely risky. For example, Wade Pfau (2010) showed that the Safemax for Japan using the same period was 0.47%. Given the current demographic dynamics of the U.S., that is something to keep in mind.
In the age of computers, professional advisors will use financial planning software to find the proper withdrawal strategies, take into account the suitable variables, and stress test the plan. Eventually, it will be the case that 4% will be the proper Safemax for a client, but that will be the exception, not the rule.
— More from The Advisor and The Quant:
- How to Plan for Retiree Health Care Costs
- Social Security Timing: What Couples Should Consider
- How to Manage Sequence of Returns Risk in Retirement
Joe Elsasser, CFP, RHU, REBC developed Social Security Timing software for advisors in 2010. Through Covisum, Joe introduced Tax Clarity in 2016.
Based in Omaha, Nebraska, Joe co-authored “Social Security Essentials: Smart Ways to Help Boost Your Retirement Income.”
Renaud “Ron” Piccinini, Ph.D., came from France to America, finally settling in Omaha, Nebraska. He brings extensive experience in building world-class risk systems, supporting tens of billions of dollars in assets to Covisum. Previously, Ron co-founded PrairieSmarts, a software business. Ron wrote his dissertation on what are now known as “Black Swan Events.”