What Is a Safe Withdrawal Rate?
A safe withdrawal rate is the percentage of your portfolio you can take out each year without running out of money over your target horizon. The 4 percent rule is the standard baseline — withdraw 4 percent in year one, adjust for inflation after that, and historically the money lasted about 30 years. That means $40,000 a year from a $1,000,000 portfolio, or a $1,250,000 portfolio to support $50,000 a year. Retiring early usually calls for a more cautious 3 to 3.5 percent, which raises the target.
The short answer
Your safe withdrawal rate (SWR) is the share of your portfolio you can spend each year with a high chance the money outlasts you. The 4 percent rule is the benchmark: take 4 percent in the first year, then adjust that dollar figure for inflation annually.
Flip it around and it tells you what you need to save. Spending $40,000 a year means a portfolio of about $1,000,000; $50,000 a year means about $1,250,000. Choose a more cautious 3.5 percent and those targets rise to roughly $1,142,857 and $1,428,571.
Run your own numbers: the retirement calculator turns your target income into the portfolio you need, and the FIRE calculator does the same for early retirement.
Where the 4% rule comes from
Two pieces of research made 4 percent the default, and it is worth knowing exactly what each one tested, because the popular version of the rule is broader than the findings behind it.
Bengen (1994). Financial planner William Bengen took a portfolio of 50 percent US stocks and 50 percent intermediate-term Treasuries and ran it through every 30-year period beginning between 1926 and 1963, withdrawing a fixed percentage in year one and raising the dollar amount with inflation thereafter. A 3 percent initial rate survived at least 50 years in every period tested. A 4 percent rate lasted at least 33 years in every period. A 5 percent rate lasted only around 20 years for the worst-case retirees — those who began in the late 1960s and early 1970s, when a market decline and high inflation arrived together. Bengen also found that a stock allocation somewhere between 50 and 75 percent held up better than a more conservative one.
The Trinity study (1998). Philip Cooley, Carl Hubbard and Daniel Walz used S&P 500 returns and long-term high-grade corporate bonds from 1926 to 1995, tested five allocations from all-stock to all-bond, withdrawal rates from 3 to 12 percent, and payout periods of 15, 20, 25 and 30 years. Their contribution was the framing everyone now uses: rather than a pass or fail, they reported a portfolio success rate — the share of historical periods in which the money did not run out. On inflation-adjusted withdrawals they concluded that rates of 3 to 4 percent “continue to produce high portfolio success rates for stock-dominated portfolios,” while rates above 7 percent performed poorly across every payout period. Their analysis did not adjust for taxes or transaction costs.
Two details matter. First, the withdrawal is set as a percentage of the portfolio's value at retirement — after that you adjust the dollar amount for inflation rather than recalculating a percentage each year. Second, it was tested against roughly 30-year retirements, which is why an early retiree planning for 40 or 50 years often uses a lower rate.
Three things “safe” can mean
Most confusion about the 4 percent rule comes from three quite different claims sharing one number. Separating them makes the rule far easier to use honestly.
- A historical finding. “In the US market record from 1926 onward, a 4 percent inflation-adjusted withdrawal from a stock-and-bond portfolio would not have exhausted the money over 30 years.” This is a statement about the past, and it is testable. It is also a statement about one country's market history, over a period that included the strongest equity century on record.
- A planning rule. “Multiply your annual spending by 25 to get a target.” This is a piece of arithmetic derived from the finding — useful precisely because it is crude, since it gives you a number to aim at before you know anything about your future returns. It carries no probability with it at all.
- A forward-looking expectation. “I can safely spend 4 percent of my portfolio a year.” This is a forecast, and nothing in the research supports it as a certainty. Future returns, future inflation, your own time horizon, your tax position and your actual allocation are all unknown, and a rule fitted to one historical sample cannot promise anything about the sample you will personally live through.
The practical consequence: use the second meaning to set a target, treat the first as evidence rather than proof, and never rely on the third. No withdrawal rate is guaranteed safe — including 3 percent.
Portfolio needed by withdrawal rate
Your required portfolio is simply annual expenses divided by the withdrawal rate. Here is what that looks like across common rates and spending levels:
| Annual expenses | At 3% | At 3.5% | At 4% | At 5% |
|---|---|---|---|---|
| $30,000 | $1,000,000 | $857,143 | $750,000 | $600,000 |
| $40,000 | $1,333,333 | $1,142,857 | $1,000,000 | $800,000 |
| $50,000 | $1,666,667 | $1,428,571 | $1,250,000 | $1,000,000 |
| $60,000 | $2,000,000 | $1,714,286 | $1,500,000 | $1,200,000 |
| $80,000 | $2,666,667 | $2,285,714 | $2,000,000 | $1,600,000 |
Dropping from 4 percent to 3 percent raises the portfolio you need by a third — that is the real price of extra caution. Each rate corresponds to a multiple of expenses: 25× at 4 percent, about 28.6× at 3.5 percent, about 33.3× at 3 percent, and 20× at 5 percent.
Choosing your rate
There is no single correct number. The right rate depends mostly on how long the portfolio must last and how much flexibility you have:
- 5 percent — only reasonable for a short retirement horizon, or if you have other income and can cut spending in bad years.
- 4 percent — the classic baseline, tested against roughly 30-year retirements. A sensible default for a traditional retirement age.
- 3.5 percent — a common choice for early retirees who need the money to last 40 years or more.
- 3 percent — a conservative rate for very long horizons or for those who want a wide safety margin.
The FIRE community leans toward the lower end for exactly this reason. See FIRE number by expenses for how the same math produces an early-retirement target.
Why the order of returns matters
A withdrawal rate that looks safe against an average return can still fail, because withdrawals turn the order of returns into a permanent effect. Selling units to raise cash while prices are down removes shares that are not there to recover later. The clearest way to see this is to give two retirees the identical set of returns in opposite order.
Both start with $1,000,000, both take 4 percent ($40,000) in year one and raise it 3 percent a year for inflation, and both live through exactly the same ten annual returns: −20%, −15%, +5%, +10%, +15%, +20%, +25%, +10%, +8%, +12%. One gets them in that order; the other gets them backwards. Both therefore earn the same 6.07 percent a year annualised and withdraw the same $458,555 in total.
| Year | Withdrawal | Bad years first: return | Balance | Good years first: return | Balance |
|---|---|---|---|---|---|
| 1 | $40,000 | −20% | $768,000 | +12% | $1,075,200 |
| 2 | $41,200 | −15% | $617,780 | +8% | $1,116,720 |
| 3 | $42,436 | +5% | $604,111 | +10% | $1,181,712 |
| 4 | $43,709 | +10% | $616,442 | +25% | $1,422,504 |
| 5 | $45,020 | +15% | $657,135 | +20% | $1,652,981 |
| 6 | $46,371 | +20% | $732,917 | +15% | $1,847,601 |
| 7 | $47,762 | +25% | $856,444 | +10% | $1,979,823 |
| 8 | $49,195 | +10% | $887,974 | +5% | $2,027,159 |
| 9 | $50,671 | +8% | $904,287 | −15% | $1,680,015 |
| 10 | $52,191 | +12% | $954,348 | −20% | $1,302,259 |
Same returns, same withdrawals, same average — and a gap of $347,911 after ten years. The retiree who met the bad years first ends up with 27 percent less, and is now in a materially weaker position going forward: the eleventh year's inflation-adjusted withdrawal of $53,757 represents 5.6 percent of that portfolio, against 4.1 percent for the other retiree. One has drifted well above a 4 percent draw without ever choosing to; the other is still roughly on plan.
This is why a single expected-return assumption is a weak test of a withdrawal plan, and why the historical studies ran every starting year separately instead of using an average. It is also why the standard defences are all about the early years specifically: holding a cash or short-bond buffer to avoid selling equities into a fall, keeping some flexibility to trim spending, and not carrying the most aggressive allocation into the first few years of retirement.
Withdrawals are taken at the start of each year and the return applied to the balance afterwards. The return sequence is illustrative, not historical; its purpose is only to isolate the effect of ordering, since both columns hold everything else identical.
What the rule does not cover
The 4 percent rule is a guideline built on historical averages, not a guarantee. Three things it does not handle on its own:
- Sequence-of-returns risk. A steep market fall in your first few years of retirement does far more damage than the same fall later, because you are selling into weakness.
- Taxes and fees. Withdrawals are gross; tax and investment costs come out of what you take.
- Changing spending. Real retirements are lumpy — health costs, one-off purchases and lifestyle changes rarely follow a smooth inflation-adjusted line.
Most retirees handle this with flexibility: trimming withdrawals in bad years is one of the most effective ways to make a portfolio last.
The math behind it
The relationship works in both directions:
At a 4 percent rate, dividing by 0.04 is the same as multiplying by 25 — which is where the 25x rule comes from. A $50,000 spending level divided by 0.04 gives $1,250,000. Note this is an annual convention: withdrawal rates and the 25x rule are defined as yearly figures, unlike the monthly compounding used elsewhere on this site for growth projections. To see how contributions grow toward the target, try the future value calculator.
Assumptions
- Annual withdrawals, annual convention. The rate is applied once a year to the starting portfolio value, then adjusted for inflation — this is the standard definition, not the monthly compounding used for growth math elsewhere on the site.
- A roughly 30-year horizon. The 4 percent figure was tested against that length; longer retirements warrant a lower rate.
- A diversified stock and bond portfolio. The historical research assumed a balanced allocation, not cash or a single asset.
- Expenses include tax. Use total real spending, since withdrawals are gross of tax and fees.
Research & sources
The two papers behind the 4 percent rule are both short and readable, and both are more careful than the rule that came out of them:
- William P. Bengen, “Determining Withdrawal Rates Using Historical Data”, Journal of Financial Planning, October 1994 — the origin of the 4 percent figure, using 30-year windows beginning 1926–1963 on a 50/50 stock and intermediate-Treasury portfolio. Bengen has since revisited the number upward in later work as he added asset classes and refined the data.
- Philip L. Cooley, Carl M. Hubbard and Daniel T. Walz, “Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable”, AAII Journal 20(2), February 1998, pp. 16–21 — the “Trinity study”, source of the portfolio-success-rate framing. Return data from Ibbotson Associates, Stocks, Bonds, Bills, and Inflation 1996 Yearbook.
- U.S. Bureau of Labor Statistics, CPI-U, U.S. city average, all items (series CUUR0000SA0) — the index an inflation-adjusted withdrawal is normally raised by each year.
Method for the figures on this page. Every portfolio target is annual expenses divided by the withdrawal rate, carried at full precision and rounded for display; the multiples quoted (25×, 28.6×, 33.3×, 20×) are 1 divided by the rate. The sequence-of-returns table is computed directly from the stated returns and withdrawals rather than taken from any source. Nothing on this page is a projection of your own results. Last verified 8 September 2026.
Frequently asked questions
The bottom line
A safe withdrawal rate turns a portfolio into an income, and back again into a savings target. Four percent is the baseline — $1,000,000 supports about $40,000 a year — while early retirees planning for decades often prefer 3 to 3.5 percent and accept the larger target that comes with it.
Work out what you need with the retirement calculator, or read how much money you need to retire for the full picture.
Disclaimer: This page is for general educational purposes only and is not financial advice. The 4 percent rule is a historical guideline, not a guarantee; safe withdrawal rates vary with markets, inflation, taxes, fees and how long a retirement lasts. Consider speaking with a qualified financial professional before making decisions about your own money.