1. What the 4% rule actually says
Quick Answer: The 4% rule says you withdraw 4% of your portfolio in your first year of retirement, then raise that dollar amount with inflation every year after. It was never a rule about taking 4% of a moving balance. Our investing guides use it the same way its author did, as a floor test, not a plan.
Financial planner William Bengen ran the numbers in 1994 across every 30-year retirement window in modern US market history. He was looking for the worst one. The rate that survived even that window became the “SAFEMAX.”
Two details get lost almost every time the rule is repeated:
- The percentage is set once. Retire with $1,000,000 and you take $40,000 in year one. In year two you take $40,000 plus inflation, whatever the market did.
- It was a worst-case answer, not an average one. Most historical retirees could have spent considerably more. Bengen’s own data shows the typical sustainable rate ran far above the worst case.
So the rule is a stress test that got mistaken for a spending plan. That distinction drives everything below.
Not sure which withdrawal number applies to you?
We publish the math behind every rate we quote, with no sponsored placements. Browse our investing guides →
The short interview below covers the same ground in Bengen’s own words before we get into the numbers.
2. What the research says the rate should be now
Quick Answer: The two most-cited 2026 sources land almost a full point apart. Bengen raised his own safe rate to 4.7%. Morningstar puts its base case at 3.9%. Both are defensible, because they answer slightly different questions: the same way dollar-cost averaging looks different on the way in than on the way out.
Bengen’s revision comes from better data and a wider portfolio. Adding small-cap, mid-cap, micro-cap and international stocks to the original two-asset mix lifted his SAFEMAX from 4.15% to 4.7%. He is still answering the historical question: what survived the worst window that actually happened?
Morningstar is answering a forward-looking one. Its 2026 base case of 3.9% for a 30-year retirement is built on simulated future returns and a 90% success target, up from 3.7% the year before.
| Source | Starting rate | How it was tested |
|---|---|---|
| Bengen, original study (1994) | 4.15% | Worst 30-year window in US history, large-cap stocks and intermediate government bonds |
| Trinity Study (1998) | Around 4% | Confirmed the finding across a range of stock and bond mixes and payout periods |
| Bengen, revised (2025) | 4.70% | Same worst-window method, seven asset classes instead of two |
| Morningstar 2026, steady spending | 3.90% | Simulated forward returns, 90% success target, 30% to 50% in stocks |
| Morningstar 2026, flexible spending | Up to 5.70% | Same simulation, but the retiree cuts spending after bad market years |
Source: DollarVisor compilation of published safe-withdrawal-rate research, August 2026. Each figure is reported as its own publisher states it.
Read the right-hand column and the disagreement mostly dissolves. Nobody is claiming 4% is broken. They are arguing about how much cushion a first-year retiree should buy.
3. What a bad first decade does to the math
Quick Answer: Losses in your first two years of retirement do far more damage than the same losses ten years in, because you are selling shares to fund withdrawals while prices are down. This is sequence of returns risk, and it is the single reason the 4% rule lands near 4% instead of near 7%.
The table models a $1,000,000 portfolio through an ugly opening: two years down 15% after inflation, then 4.5% real for the remaining 28. Withdrawals rise with inflation, so every figure is in today’s dollars.
| Year of retirement | Started at 3.9% | Started at 4.0% | Started at 4.7% |
|---|---|---|---|
| Year 0 | $1,000,000 | $1,000,000 | $1,000,000 |
| Year 5 | $627,000 | $622,000 | $586,000 |
| Year 10 | $558,000 | $546,000 | $462,000 |
| Year 15 | $472,000 | $452,000 | $307,000 |
| Year 20 | $366,000 | $334,000 | $113,000 |
| Year 25 | $233,000 | $188,000 | $0: ran dry in year 23 |
| Year 30 | $67,000 | $5,000 | $0 |
Illustrative scenario modeled by DollarVisor, August 2026. Assumes two opening years at −15% real, 4.5% real returns thereafter, inflation-adjusted withdrawals taken at the start of each year, and no fees or taxes. Not a forecast.
In this scenario a 4.0% start finishes with $5,000 left. A 4.7% start runs out seven years early. Same portfolio, same markets, 70 basis points apart.
4. Thirty years is the wrong default for a lot of retirees
Quick Answer: Every published safe rate is tied to a time horizon, and almost all of them assume 30 years. If you retire at 55, or you are half of a healthy couple, 30 years is optimistic. Guaranteed lifetime income, including what an annuity does, exists precisely for that gap.
The Social Security Administration’s period life table for 2023 puts remaining life expectancy at age 65 at 18.12 years for men and 20.66 years for women. Those are averages, and averages hide the tail: a healthy 65-year-old couple has a real chance of one partner reaching 95.
Horizon changes the answer in a predictable direction:
- Retiring at 65 in average health. A 30-year plan is a fair stress test, and the 4% rule fits.
- Retiring at 55. You are planning for 40 years. Most researchers drop the starting rate by about half a point.
- Retiring at 72 in poor health. A 20-year horizon supports a higher rate, and holding to 4% may mean dying with money you meant to spend.
5. What a 4% withdrawal actually buys in your state
Quick Answer: $40,000 a year is not the same income in California as it is in Arkansas. Federal price data for 2024 shows a spread of about 27% between the most and least expensive states, a bigger swing than the entire 3.9%-to-4.7% research debate. At DollarVisor we run state numbers first for exactly this reason.
The Bureau of Economic Analysis publishes regional price parities, an index where 100 equals the national price level. The bars below convert a $40,000 withdrawal into what it would buy at national average prices.
| State | Price index | Relative buying power | $40,000 is worth |
|---|---|---|---|
| California | 110.7 | $36,100 | |
| Hawaii | 110.0 | $36,400 | |
| New Jersey | 108.8 | $36,800 | |
| National average | 100.0 | $40,000 | |
| Oklahoma | 87.8 | $45,600 | |
| Iowa | 87.8 | $45,600 | |
| Mississippi | 87.0 | $46,000 | |
| Arkansas | 86.9 | $46,000 |
Source: DollarVisor calculation using BEA regional price parities for 2024, released February 19, 2026. Bars are scaled to the highest buying power in the table.
Housing drives most of that gap. California’s rent index sits at 154.3 against a national 100; West Virginia’s is 54.2. Moving across that line changes your effective withdrawal rate without touching the portfolio.
6. Year-one income at each published rate
Quick Answer: On a $1,000,000 portfolio, the gap between the lowest and highest published rate is about $18,000 of first-year income. On a $250,000 portfolio it is about $4,500. The size of the pot decides your standard of living; the rate only decides the trim, which is why asset allocation during the saving years matters more.
| Starting rate | $250k | $500k | $750k | $1M | $1.5M | $2M |
|---|---|---|---|---|---|---|
| Steady, inflation-adjusted spending: no cuts required | ||||||
| 3.9% (Morningstar 2026) | $9,750 | $19,500 | $29,250 | $39,000 | $58,500 | $78,000 |
| 4.0% (the classic rule) | $10,000 | $20,000 | $30,000 | $40,000 | $60,000 | $80,000 |
| Higher rates: only safe if you will actually cut spending after a bad year | ||||||
| 4.7% (Bengen 2025) | $11,750 | $23,500 | $35,250 | $47,000 | $70,500 | $94,000 |
| 5.7% (flexible spending) | $14,250 | $28,500 | $42,750 | $57,000 | $85,500 | $114,000 |
Illustrative calculation by DollarVisor, August 2026. Rate multiplied by portfolio balance, before taxes and fees. Rates as published by their original sources.
Read down the $1M column and the whole 4% rule debate is worth $18,000 a year. Read across the 4.0% row and doubling the portfolio is worth $40,000. Saving beats optimizing, which is the case for starting early, even with something as small as a Roth IRA for a working teenager.
Your rate depends on the mix you actually hold.
Every safe-rate study assumes a specific stock and bond split, and most retirees do not know theirs. See how to check and rebalance yours →
7. Guardrails: the version of the rule that adapts
Quick Answer: Guardrails are the 4% rule with a feedback loop. You start higher, up to 5.7% in Morningstar’s 2026 work, in exchange for a written promise to cut spending when the portfolio drops past a set line. It is the same idea behind a bond ladder: decide the response before the stress arrives.
A simple version looks like this:
- Set your starting rate. Say 5% of a $1,000,000 portfolio, or $50,000 in year one.
- Set an upper and lower rail. A common pair is 20% either side of that rate, measured against the current balance.
- Cut at the lower rail. If withdrawals reach 6% of the current balance, trim spending by 10%.
- Raise at the upper rail. If withdrawals fall to 4% of a grown balance, give yourself a 10% raise.
The catch is behavioral. Guardrails only pay off if you actually cut, in the year it hurts most. A retiree who ignores the rail in a down market has quietly taken a 5.7% fixed withdrawal, which no research supports.
8. What the 4% rule leaves out
Quick Answer: The rule is a gross-withdrawal figure. It says nothing about taxes, fund fees, Social Security timing or the fact that real retirement spending is lumpy rather than smooth. Withdrawals from a traditional 401(k) are ordinary income, and taxable-account sales trigger capital gains tax.
Four omissions matter most:
- Taxes come out of the 4%, not on top of it. A $40,000 withdrawal from a traditional IRA might leave $33,000 to spend. Roth withdrawals do not.
- Fees come straight off the safe rate. A portfolio paying 1% a year in fund and advice costs is really running a rate about a point higher than the one on paper.
- Social Security changes the portfolio’s job. If benefits cover $30,000 of a $60,000 budget, the portfolio funds only the other half.
- Spending is not a straight line. Most retirees spend more in their sixties, less in their late seventies, then more again on health care. The rule assumes a flat line.
9. How to set your own starting rate, in order
Quick Answer: Start with the 4% rule, then adjust in five steps for horizon, costs, guaranteed income and flexibility. Most people land between 3.5% and 5%. Keep a cash buffer outside the calculation so a bad year does not force a sale, the same logic as an emergency fund during your working years.
- Write down your real horizon. Use your retirement age and a plausible age 95, not a default 30 years.
- Adjust the base rate for it. Add about half a point for a 20-year plan, subtract about half a point for a 40-year plan.
- Subtract your all-in investment costs. Total your fund expense ratios plus any advice fee, and take that off the rate.
- Size the portfolio’s actual job. Subtract Social Security, pensions and annuity income from your budget. The rate covers only the remainder.
- Write your reaction rule now. Decide what triggers a spending cut and how deep. Without it, you cannot claim the higher rates.
10. The verdict
Quick Answer: Yes, the 4% rule still works in 2026, as a sizing tool. It tells you that a $60,000 portfolio-funded budget needs roughly $1.5 million, and that number has held up across three decades of research. It fails only when people treat it as a guarantee they can stop thinking about.
Use the 4% rule to answer “how big does the pot need to be.” Then set your actual withdrawal from your own horizon, your own costs and your own willingness to adjust. The published 2026 range is 3.9% to 4.7%, and where you land inside it is a decision about cushion, not a fact about markets.
11. Frequently Asked Questions
1. Is the 4% rule still safe in 2026?
It is still a reasonable starting point. Published 2026 research puts the safe starting rate between 3.9% for steady inflation-adjusted spending and 4.7% under the historical worst-case method. Where you land depends on your time horizon, your investment costs and whether you will cut spending after a bad year.
2. Does the 4% rule work for early retirement?
Not without adjustment. The 4% rule was calibrated for a 30-year retirement. Retiring at 50 or 55 means planning for 40 years or more, and most researchers lower the starting rate by roughly half a percentage point for that horizon. Guardrails matter more the longer the plan runs.
3. Is it 4% of the starting balance or 4% of the current balance?
The starting balance. You take 4% in year one, then raise that dollar amount with inflation each year regardless of what the portfolio does. Taking 4% of the current balance each year is a different strategy: it never runs out, but your income falls sharply in bad markets.
4. Does the 4% rule account for taxes and fees?
No. It is a gross withdrawal from the portfolio. Taxes on traditional 401(k) and IRA withdrawals come out of that amount, and fund fees reduce the return the rule assumes. Subtract your all-in investment costs from the published rate and plan your budget on the after-tax figure.
Working out what your own safe rate should be?
Tell us your retirement age, your state and roughly what the portfolio has to cover, and we will point you to the guide that shows the math. No sponsored placements, no sales calls.
This article is information, not financial, tax or legal advice. Rules change and your situation is your own. See our disclaimer.