The 4% Rule's Own Author Now Says 4.7%. Morningstar Says 3.9%.

The man who invented the 4% rule now says the number is 4.7%.^[1] Morningstar, running the same question through a different method, says 3.9% for 2026.^[2]

On a $1 million portfolio that spread is the difference between spending $39,000 a year and $47,000 a year. Eight thousand dollars, annually, for thirty years, from two credible sources answering one question.

Which tells you something more useful than either number: the answer is not a fact to look up. It is an output of assumptions, and the assumptions are where all the content is.

Where the Rule Actually Came From

Not from a committee, an institution, or an accumulation of evidence. From one paper: William Bengen, Journal of Financial Planning, October 1994.^[1] He took Ibbotson Associates data on US markets back to 1926, ran a 30-year retirement against every historical starting point, and asked what withdrawal rate would have survived the worst of them.

Note the boundaries of that question, because they travel with the answer whether or not anyone repeats them. US markets only. A 30-year horizon. Historical returns replayed, not forecast. A retiree with a 45-year horizon, or a portfolio outside the US, is outside what the study examined.

And here is the detail I find genuinely funny. Bengen's calculation produced 4.15%. It got rounded down to 4%, and the rounded number is the one that became a rule.^[1] Millions of retirement plans are anchored to a figure whose last significant digit was discarded for being untidy.

Why Its Author Raised It

Bengen kept working, and his later research put the historical maximum higher, at 4.7%, which he calls the Universal Safemax: the rate that would have survived the genuine worst case in the data, a retiree who hit high inflation and a poor equity market at once.^[1]

He has also said, pointedly, that early retirees clinging to 4% may be cheating themselves, spending less than the evidence supports and arriving at the end with money they could have used.^[3] That is a real cost, just an invisible one. Nobody writes an article about the retiree who died with an unnecessarily large balance and skipped trips they could have afforded.

Worth noting what he considers the central threat, because it is not a market crash: he has called inflation the retiree's greatest enemy.^[3] A crash you can wait out. A sustained inflation raises the amount you must withdraw every single year while simultaneously reducing what the portfolio is worth in real terms, which is why inflation is the variable that breaks withdrawal maths rather than merely denting it.

Why Morningstar Says Less

Morningstar's answer for 2026 is 3.9%, up from 3.7% a year earlier, and their construction is explicit: a 90% probability of still having money after 30 years, on a portfolio holding 30% to 50% equities with the rest in bonds and cash.^[2]

The methodological difference is the whole story. Bengen replays what actually happened. Morningstar substitutes forward-looking capital market assumptions, which is why the number moved up when their return expectations improved, and why it was lower when they were looking at depressed bond yields and elevated equity valuations.^[2]

Neither is wrong. They are answering different questions. “What would have survived history” and “what should survive a forecast” are not the same, and the second is only as good as the forecast. If you use a number, know which of the two it is.

The Part of the Paper That Actually Mattered

The 4% figure is the famous output. The durable contribution was the mechanism, and Bengen's paper is what introduced it to financial planning: sequence-of-returns risk.^[1]

Two portfolios can earn identical average annual returns over thirty years and end in completely different places, depending on the order of those returns. If the bad years arrive early, while you are also selling to fund your living costs, you are liquidating shares at depressed prices and permanently removing them from the recovery that follows. The same bad years arriving in year 25, after two decades of compounding, are survivable.

This is why a single withdrawal percentage is a crude instrument. It is trying to compress a question about the ordering of unknown future returns into one number chosen in advance. It also reframes what volatility means once you are withdrawing rather than accumulating: during accumulation, volatility is mostly noise you wait out, and during drawdown it is the thing that can end the plan.

What Moves the Answer More Than the Answer Does

Flexibility, by a wide margin. Every one of these studies assumes a fixed inflation-adjusted withdrawal regardless of conditions, which is what makes the safe rate so low. A retiree willing to trim spending in a bad year can support a materially higher starting rate, because the failure scenarios all involve continuing to withdraw fully while the portfolio falls. Rigidity is the expensive assumption, not the market.

Your actual horizon.Thirty years is the study's assumption, not your life. Retiring at 45 means planning for something closer to 50 years, where the safe rate is lower and the historical evidence is thinner because there are fewer non-overlapping 50-year periods to test.

Allocation. The equity weight drives the outcome, which is why Morningstar states theirs. Getting that decision deliberately right matters more than the third decimal place of a withdrawal rate, and it is the subject of asset allocation and rebalancing.

Taxes and account order. A 4% withdrawal from a traditional 401(k) and the same withdrawal from a Roth are not the same amount of spending money. Which accounts you draw from, and in what sequence, changes the real number materially, and the capital gains brackets are part of why.

There is also a way to attack the problem from the other side rather than by spending less. All of these rates are conservative because the horizon is unknown, and that specific risk can be transferred to a pool: what a SPIA actually buys, and why no bond ladder can match it.