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

A rule of thumb with a paper trail

In 1994, a financial planner named William Bengen published an article in the Journal of Financial Planning that asked a practical question: if you retire with a portfolio of stocks and bonds, how much can you withdraw in the first year, then raise each year with inflation, without running out over 30 years? Looking at historical US market data, he found that a first-year withdrawal of roughly 4% had survived every 30-year starting point in his dataset. A few years later, a widely cited study by three finance professors at Trinity University examined a similar question and reached comparable conclusions, which is why it is often called the "Trinity study".

Those two sources gave us "the 4% rule". It is a guideline derived from past data, not a law of nature, and everything below is about understanding what it does and does not promise.

How the rule works in practice

The mechanics are simple. In year one you withdraw 4% of your starting portfolio. In every later year you withdraw the same amount, adjusted upward by inflation. You do not recompute 4% each year on the new balance. If you start with 1,000,000, your first-year withdrawal is 40,000. If inflation is 2.5%, the second year it is 41,000, the third 42,025, and so on.

Turn it around and you get the "25x" target: your portfolio needs to be 25 times your first-year spending (1 divided by 0.04). Some examples:

• Spending 30,000 a year: 750,000
• Spending 40,000: 1,000,000
• Spending 60,000: 1,500,000
• Spending 80,000: 2,000,000

Note that this works from spending, not income. Anything covered by other sources, such as a pension or Social Security, reduces the amount your portfolio has to supply, and the target shrinks with it. The retirement calculator lets you enter those income sources alongside your savings.

What it assumes

• A 30-year horizon. Someone retiring at 65 is planning for 30 years. Someone leaving work at 45 needs 45 or 50.
• A balanced portfolio, in Bengen's case a substantial stock allocation, since returns need to outpace inflation.
• US market history, which includes a strong long-run record. Other countries' histories were rougher in some periods.
• Fixed real spending, adjusted only for inflation, with no flexibility in bad years.
• Low costs. Fees come straight out of the withdrawal capacity. A portfolio charged 1% a year effectively starts from a lower safe rate.

Stress test 1: different returns

Take a 1,000,000 portfolio, 40,000 first-year withdrawal, 2.5% annual inflation, and try constant nominal returns. This is a simplified deterministic model, not a forecast. Balance after 30 years:

• 3% return: money runs out in year 27
• 4% return: about 65,600 left, then exhausted shortly after (year 31 if you keep going)
• 5% return: about 585,000 left after 30 years, and exhausted in year 38 if you extend to 40
• 6% return: about 1,327,000 left
• 7% return: about 2,367,000 left

The spread is huge. The same 4% start produces anything from failure to a doubled estate depending only on the average return. The rule is a summary of history, and it assumes returns land near the historical average in a particular sequence.

Stress test 2: the order of returns

Two retirees, both with 1,000,000, both withdrawing 60,000 in the first year (a deliberately aggressive 6%), raised by 2.5% each year, for 10 years. Both experience exactly the same ten annual returns, with an average of 4.5%: +20%, +15%, +10%, -10%, -15%, +5%, +5%, +5%, +5%, +5%. The only difference is the order. Retiree A gets them in that order. Retiree B gets them reversed, so the losses come first.

• Retiree A (good years first): about 701,500 left after ten years
• Retiree B (bad years first): about 585,200 left
• Difference: about 116,000, from order alone

This is called sequence-of-returns risk. When you are withdrawing money, early losses hurt more because you are selling assets at low prices to fund spending, leaving fewer shares to recover. A retiree who is still saving would actually prefer the bad years early, because they could buy cheaply. Withdrawers get the opposite.

Stress test 3: longer retirements

In a simple model with a constant 2.44% real return (5% nominal minus 2.5% inflation), the level initial withdrawal rate that exactly empties a portfolio is about 4.6% for 30 years and about 3.8% for 40 years. With zero real return it is 1/30, or 3.3%, for 30 years. Longer horizons push the sustainable rate down. This is why people planning early retirement commonly talk about 3.25% to 3.5%, which turns a 40,000 spending target into a portfolio of roughly 1,140,000 to 1,230,000 rather than 1,000,000.

How people use the rule well

• As a starting estimate, not a contract. It tells you whether a plan is in the right neighborhood.
• With flexibility. Cutting spending by 5 to 10% after a bad market year, or skipping an inflation raise, improves survival odds more than most other tweaks.
• With a cash buffer. Holding one to two years of expenses outside stocks reduces the chance of selling at the bottom.
• With an inflation sanity check. See how fast costs grow in your own budget using the inflation calculator.
• With a re-check every year or two, comparing actual balance against the plan.

To see how a working-years savings path leads to a target like 25x, try the 401(k) calculator with your own contribution and expected return, then compare the projected balance against your 25x number. The rule is a useful mental ruler. Treat its output as a range, not a promise, and this article is general education, not personal financial advice. For a plan that depends on it, a fee-only planner can test it against your real situation.