The 4% Rule: Where It Came From, What It Assumes, and When It Breaks

The 4% rule says that if you withdraw 4% of your portfolio in the first year of retirement, then raise that dollar amount by inflation every year, the money has historically lasted at least 30 years. It comes from William Bengen's 1994 study of US market history and was reinforced by the 1998 "Trinity" study.¹² It was built for a 30-year retirement. For a 40- or 50-year early retirement, current research points lower: Morningstar's 2025 estimate for a 40-year horizon is 3.3%.⁴

The rule is a starting point for one question: how big does a portfolio need to be before you stop working? It is not a spending plan you can switch on and ignore. The rest of this page covers what the original papers tested, what the rule looks like in dollars, and the situations where it gives the wrong answer.

What Bengen actually tested

Bengen was a financial planner in El Cajon, California. He took US return data from 1926 to 1992 and simulated a retiree starting in each year: withdraw a fixed share of the portfolio in year one, then the same dollar amount adjusted for inflation every year after, and count how many years the money lasted.¹

His base portfolio was 50% common stocks and 50% intermediate-term Treasury notes. The findings that became "the rule":

• A 3% first-year withdrawal never produced a portfolio life under 50 years in his data.¹
• At 4%, no historical starting year ran out of money in fewer than 33 years, and most lasted 50 years or longer.¹
• At 4.25%, the worst case dropped to about 28 years. He called 5% "risky" and 6% or more "gambling."¹
• He recommended a stock allocation between 50% and 75% and found that going below 50% stocks shortened portfolio life.¹

Two of his assumptions matter more than they look. First, he assumed all assets sat in tax-deferred accounts, so capital-gains taxes were not part of the model.¹ Second, he framed the minimum acceptable portfolio life as life expectancy plus 5 to 10 years, and he wrote that for a client aged 60 to 65 that usually works out to about 4%.¹ The number was calibrated for people retiring in their early sixties.

The worst starting years were not the Depression. Bengen identified the 1973–74 downturn, which combined a deep stock decline with high inflation, as the most damaging event in his data, and retirees who started in the late 1960s and early 1970s did worst at higher withdrawal rates.¹ That is sequence risk in its plainest form: bad returns plus rising prices in the first decade.

What the Trinity study added

In February 1998, three business professors at Trinity University in San Antonio, Philip Cooley, Carl Hubbard and Daniel Walz, published a follow-up in the AAII Journal. They used the S&P 500 for stocks and long-term high-grade corporate bonds for bonds, covered 1926 to 1995, and tested withdrawal rates from 3% to 12% over payout periods of 15 to 30 years, across allocations from 100% stocks to 100% bonds.²

Instead of "how many years did it last," they reported a success rate: the share of historical periods in which the portfolio was not exhausted. With inflation-adjusted withdrawals, they found 3% to 4% still produced high success rates for stock-heavy portfolios, while rates above 7% did poorly over every payout period.²

The paper is often quoted as proof that 4% is safe. Read in full, it says two things that pull in opposite directions. For stock-dominated portfolios over 30 years or less, the authors called 3% and 4% "exceedingly conservative."² And they wrote that young retirees "who anticipate long payout periods should plan on lower withdrawal rates than their older counterparts."² The study that popularized the 4% rule never claimed it held beyond 30 years.

The rule in dollars: a worked example

Example (hypothetical numbers, not a forecast): a $1,000,000 portfolio, a 4% first-year withdrawal, and 3% inflation every year.

Year	Withdrawal under the rule	How it is set
1	$40,000	4% × $1,000,000
2	$41,200	$40,000 × 1.03
3	$42,436	$41,200 × 1.03
4	$43,709	$42,436 × 1.03

After year one the percentage is never used again. Bengen was explicit that each later withdrawal is last year's dollar amount plus inflation, not a share of the current balance.¹ That is what makes the rule simple. It is also what makes it uncomfortable after a crash.

Same example, three different first years, with the year-2 withdrawal fixed at $41,200:

Market in year 1	Balance at start of year 2	Year-2 withdrawal	Effective withdrawal rate
Flat (0%)	$960,000	$41,200	4.29%
Up 15%	$1,104,000	$41,200	3.73%
Down 25%	$720,000	$41,200	5.72%

In the down-25% row, the rule tells you to keep spending as planned while your effective rate is already in the range Bengen called risky. Most people would cut back. If you know you would cut back, you are not really following the 4% rule; you are following a flexible rule, and you should plan with flexible-rule numbers (see the alternatives below).

How much you need at different rates

Dividing annual portfolio spending by the withdrawal rate gives the target portfolio. At 4% that is 25 times spending, as the example table shows.

Annual spending from the portfolio	At 4%	At 3.5%	At 3.3%
$40,000	$1,000,000	$1,142,857	$1,212,121
$50,000	$1,250,000	$1,428,571	$1,515,152
$60,000	$1,500,000	$1,714,286	$1,818,182

In the example table, moving from 4% to 3.3% raises the target by about 21%. For a household saving a fixed amount each year, that can mean several extra working years, which is why the savings rate behind the target matters as much as the rate itself. The savings-rate guide shows how those years add up.

What newer research says

Morningstar publishes an annual estimate of the highest "safe" starting withdrawal rate. Its 2025 edition, The State of Retirement Income: 2025 (published December 3, 2025), uses forward-looking return and inflation assumptions rather than history, and defines safe as a 90% probability of money remaining after 30 years of inflation-adjusted withdrawals.³

Morningstar 2025 finding	Figure
Base case, 30 years, 90% success	3.9% (30% to 50% in stocks)3
Previous editions, same base case	3.3% (2021), 3.8% (2022), 4.0% (2023), 3.7% (2024)3
35-year horizon	3.5%4
40-year horizon	3.3%4
Same 30-year test using historical returns, 50/50 portfolio	4.4%4
Guardrails spending method, starting rate	5.2%4
Best flexible methods tested, starting rate	up to 5.7%4

Two readings of that table hold at once. The base-case number moves with bond yields and stock valuations, so it shifts from year to year. And history alone still supports a figure a bit above 4% for 30 years. The 3.9% versus 4.4% gap is mostly the difference between "assume the future looks like 1926–2024" and "assume lower stock returns over the next decade."⁴ Neither is a guarantee. Note also that Morningstar's base case points to lower stock allocations (30% to 50%) than Bengen recommended, because its return assumptions are less generous to stocks.

When the 4% rule is the wrong tool

The rule answers one narrow question well. These are the cases where it answers the wrong question.

Your retirement may last 40 to 50 years. Both founding papers judged success over 30 years or less, and the Trinity authors told young retirees with long payout periods to plan lower.² Morningstar's 40-year figure of 3.3% is a more honest starting point for someone stopping work at 45.⁴

Your spending is not flat. The rule assumes the same real spending every year. Early retirees often spend more in the first decade on travel, a mortgage that is not yet paid off, or children still at home. Health insurance before Medicare is the big one. Morningstar's 2025 report estimated that for a 62-year-old withdrawing 3.5% ($35,000) from $1 million, health coverage could take roughly a third of the withdrawal.⁴ In 2026 the federal subsidy rules for that coverage also got less generous. The early-retirement health coverage guide covers that piece.

You will adapt anyway. If a 25% drop would make you cut spending, a fixed-dollar rule overstates your risk and understates what you could spend at the start. Plan with a flexible method and write down its rules in advance.

Much of your money is in taxable accounts, or you pay high fees. Bengen modeled tax-deferred accounts with no capital-gains drag.¹ The 4% is gross. Taxes and any advisory or fund fees come out of it.

Your portfolio is not your main income. If a pension or Social Security will cover most fixed costs later, the portfolio has a different job: bridging the years before those start. A bridge is a shorter, front-loaded withdrawal problem, not a level 30-year one. Morningstar's 2025 report found flexible methods work better when paired with substantial guaranteed income.⁴

Your income before retirement is irregular. The rule is about drawing down. It says nothing about how stable your earnings are while you build the portfolio, or how part-time or freelance income after you leave a job changes the math. The non-W2 income guide covers how that income is taxed.

Alternatives that adjust to markets

Method	How it works	Trade-off
Lower fixed rate (3.3%–3.5%)	Same mechanics as the 4% rule, smaller start	Predictable income; you may die with a large unspent balance
Guardrails (Guyton-Klinger)	Start higher. If the current withdrawal rate drifts 20% above the starting rate, cut spending 10%; if it drifts 20% below, raise it 10%4	Higher starting income; you must accept real cuts in bad years
Fixed percentage of current balance	Withdraw the same percentage of whatever the portfolio is worth each year	Never runs out on paper; income swings with the market

In Morningstar's own guardrails example, a $1,000,000 portfolio that loses 30% in year one forces the year-2 withdrawal down from a scheduled $40,928 to $36,835.⁴ That is the cost of the higher start. A household whose fixed bills already eat most of the withdrawal has less room to take that cut, and should lean toward a lower fixed rate instead.

A workable decision rule: add up the expenses you could not cut in a bad year (housing, insurance, food, debt payments). If that floor is close to your planned withdrawal, use a low fixed rate. If the floor is well below it, a guardrails method lets you spend more early with a cut you can absorb.

Common mistakes

One mistake is treating 4% as a yield. The rule spends principal; it does not require the portfolio to earn 4%. Another is recalculating 4% of the new balance every year, which quietly turns the rule into the fixed-percentage method. That method behaves very differently in a bad decade. A 62-year-old and a 42-year-old should not start from the same number. And the costs between leaving work and Medicare or Social Security belong in the plan as line items, not in "miscellaneous."

Short answers

Is the 4% rule still valid in 2026? For a 30-year retirement it is still in the range current research supports: Morningstar's 2025 base case is 3.9%, and its historical-returns version is 4.4%.³⁴ For longer horizons, plan lower.

Does the 4% rule include Social Security? No. Both the original studies and Morningstar's estimates cover portfolio withdrawals only, and Morningstar states that its base case excludes Social Security and other non-portfolio income.³

More on planning the years before 59½ and 65 is on the early retirement hub.