Updated: August 16, 2026
OPRS may receive compensation when readers open an account through partner links on this page. Our analysis is based on independent research, BBB data, and IRS publications.
The 4 percent rule sits inside almost every retirement conversation, yet the origin story usually arrives wrong. It came from one working financial planner and one university study, both published more than three decades ago. Reading what those two papers actually said, and what they did not, is the difference between using the rule as designed and using it as a bumper sticker.
The pages below trace the number back to its two original papers, explain what it was calibrated to measure, walk through the substantive critiques, and connect it to the newer dynamic methods that try to fix its most rigid parts. No market predictions. No fear framing. Only what the record says.
Where the 4 percent number came from: William Bengen’s 1994 paper
William Bengen was a working financial planner in California. He published “Determining Withdrawal Rates Using Historical Data” in the October 1994 Journal of Financial Planning. He wanted a simple answer to a client question: how much can a retiree spend each year from a portfolio without running out?
Bengen ran rolling 30-year windows through the historical U.S. market record back to 1926. He tested portfolios that mixed U.S. large-cap stocks with intermediate-term U.S. government bonds. Allocations ranged from 50 to 75 percent stocks. Each window started at a different year.
The rule tested was uniform. Withdraw a constant real dollar amount in year one. Adjust that same dollar amount upward each year for inflation. Do not adjust for the portfolio balance. Repeat for 30 years.
The output Bengen cared about was SAFEMAX. That is the highest starting withdrawal rate that would have survived every 30-year historical period without exhausting the portfolio. For the 50 to 75 percent equity mix, SAFEMAX landed just above 4 percent. Bengen rounded to 4 percent and labeled it a floor, not a target.
The 1998 Trinity Study: what Cooley, Hubbard, and Walz added
Four years later, three finance professors at Trinity University published “Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable” in the AAII Journal. Philip Cooley, Carl Hubbard, and Daniel Walz used a similar historical backtest, but they added two changes that shaped how the rule spread.
First, they swapped intermediate government bonds for long-term high-grade corporate bonds. That single change raised historical yield and lowered survival slightly during the worst inflation periods. Second, they framed the output as a success rate, not a binary pass or fail.
Their tables report the percentage of historical 30-year windows in which each starting rate survived at each stock or bond mix. At a 50 percent stock and 50 percent bond mix, a 4 percent starting withdrawal with inflation adjustments succeeded in about 95 percent of 30-year historical windows in their sample.
That success-rate framing is the reason the 4 percent number stuck. It gave planners a defensible number and a way to talk about probability rather than certainty.
What “4 percent” actually meant: the floor, not the median
Both papers are careful about a point that gets lost in headlines. The 4 percent number was calibrated to the WORST historical 30-year window in the U.S. record. It was not the median case, and it was not the expected case.
In most historical windows, a retiree who followed the rule strictly would have died with a portfolio much larger than the starting balance. In some windows, several times larger.
That framing matters. The rule was designed to protect a retiree who happened to start withdrawing right before a bad decade, such as 1966 or 1973. In an average or lucky decade the same rule leaves substantial unused portfolio. It was a conservative anchor, not a prediction of the future paycheck.
Bengen’s own updates in the 2020s
Bengen has continued to refine his own work in the years leading up to 2026. In interviews and follow-up writing, he has argued that the original 4 percent number understated what the historical record actually supports.
His updated portfolio uses a wider set of asset classes. It includes U.S. small-cap stocks, mid-cap stocks, and international equities alongside the original large-cap and bond mix. His more recent published analysis raises the historical SAFEMAX closer to 4.7 percent for the broader portfolio.
The mechanics have not changed. He still measures against the worst 30-year window. The higher number comes from added diversification, not from a rosier reading of the same data.
The takeaway is worth stating plainly. The man who first calibrated the number now believes 4 percent was too cautious for a diversified portfolio. The underlying method still relies on U.S. history.
How the SAFEMAX has moved across the three sources
The three numbers above sit side by side in the chart below. Each bar reflects the starting withdrawal rate that the underlying source treated as the historical safe floor for a retiree with a 30-year horizon.
The main critiques
The 4 percent rule has attracted careful critiques from every generation of retirement researchers since 1994. The critiques do not reject the arithmetic. They question whether the historical U.S. record is a fair guide to the next 30 years for today’s retirees.
The U.S. sample is small and lucky
Wade Pfau, Michael Kitces, and David Blanchett have tested the same rule against long-run data from other developed economies. In samples from Germany, Japan, the United Kingdom, and France, the historical safe withdrawal rate falls well below 4 percent. The 20th-century U.S. was an unusually favorable investment environment.
Low starting yields and high valuations reduce forward rates
When bond yields are historically low and equity valuations are historically high at the start of retirement, expected 30-year returns are compressed. The SEC’s investor bulletin on the 4 percent rule makes this point in plain language. Past returns cannot be assumed forward.
Sequence-of-returns risk sits behind the whole rule
The 4 percent floor exists precisely because a retiree who happens to hit a deep bear market in the first five years is far more vulnerable than one who hits it later. The OPRS reference on sequence-of-returns risk walks through why identical average returns can produce very different outcomes.
A constant real dollar withdrawal ignores real retiree behavior
Real retirees do not draw the same inflation-adjusted amount every year regardless of markets. Most spend less in a bear market by choice. Modeling that behavioral flexibility raises safe rates above 4 percent in most academic simulations. The rigid rule is a conservative assumption, not a description of behavior.
Thirty years may understate longevity for young retirees
A 60-year-old couple has a meaningful probability that at least one spouse lives past age 95. A 30-year horizon leaves that tail unfunded. For an early retiree the appropriate horizon is longer, and the safe rate has to fall accordingly.
How dynamic methods respond to these critiques
Later researchers have proposed several methods that keep Bengen’s insight (calibrate to bad markets) while addressing the rigidity critique. The OPRS reference on dynamic withdrawal versus fixed-rate methods compared covers the mechanics in detail. Three families matter most.
Guardrails methods were formalized by Jonathan Guyton and William Klinger in 2006. They start at a higher rate, often 5.0 to 5.5 percent. Cut and raise rules trigger when the current withdrawal rate strays outside a corridor. The Guyton-Klinger guardrails method reference decodes the exact rules.
RMD-based methods use the IRS Uniform Lifetime Table divisor as the withdrawal rate each year. The divisor shrinks with age, so the rate rises naturally as the horizon shortens. The portfolio cannot exhaust by construction, though the paycheck can fall in bad years.
Floor-and-ceiling methods take a percentage of the current balance each year but clip the result inside a corridor. The floor is a minimum real spending level. The ceiling is a maximum. The method trades some historical survival probability for a paycheck the household can actually budget around.
None of these methods repeals the 4 percent rule. They inherit its worst-case discipline and add behavioral flexibility. Each involves a different trade-off between paycheck stability, historical portfolio survival, and required discipline in a bad market year.
What this history means for a retiree deciding today
The 4 percent rule is best read as a benchmark, not an instruction. It answers one specific question. What is the highest constant real withdrawal rate that has survived every 30-year period in the U.S. historical record for a 50 to 75 percent equity portfolio? The answer is close to 4 percent.
That benchmark is useful. It tells a retiree whether a proposed spending plan is meaningfully aggressive or meaningfully conservative relative to what the historical record supports. It does not tell a retiree what will happen next.
Every substantive critique of the rule points in the same direction. The real world involves longer horizons for early retirees, less generous forward returns than the 20th-century U.S. average, sequence risk in the first decade, and actual behavioral flexibility that the rigid rule ignores. Dynamic methods try to price those factors in.
The version of Bengen’s rule that actually works for a household is rarely 4 percent applied literally. It is the underlying discipline (calibrate to bad markets, plan for a 30-year floor, revisit yearly) applied through whatever method fits the household’s income floor, tolerance for a variable paycheck, and behavioral bandwidth.
Sources cited
- William P. Bengen, Determining Withdrawal Rates Using Historical Data, Journal of Financial Planning, October 1994. The original paper that defined SAFEMAX using rolling 30-year historical windows in U.S. large-cap stocks and intermediate-term U.S. government bonds.
- Philip L. Cooley, Carl M. Hubbard, and Daniel T. Walz, Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable, AAII Journal, February 1998. The Trinity Study, which added corporate bonds and success-rate framing to the Bengen methodology.
- U.S. Securities and Exchange Commission, Investor.gov, Updated Investor Bulletin: The 4 Percent Rule and Other Strategies for Sustainable Withdrawals. The SEC’s plain-language overview of the safe-withdrawal-rate literature and its forward-looking limits.