The safe withdrawal rate (SWR) is the percentage of your portfolio you can withdraw each year in retirement without running out of money. It's the engine behind every FIRE number calculation, and getting it right matters enormously. Too aggressive, and you risk depleting your savings. Too conservative, and you work years longer than necessary.
What the Trinity Study Found
In 1998, three professors at Trinity University โ Philip Cooley, Carl Hubbard, and Daniel Walz โ published "Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable," now known simply as the Trinity Study. They analyzed U.S. market data from 1926 onward and tested every historical 30-year retirement window.
Their key finding: a portfolio of 60% stocks and 40% bonds, withdrawing 4% in year one and adjusting for inflation each subsequent year, succeeded in 96% of all historical 30-year periods. The rare failures occurred in the most severe historical sequences โ retirees who had the misfortune of retiring just before major crashes.
Annual Income = Portfolio Value ร Safe Withdrawal Rate
Example: $2,000,000 ร 4% = $80,000/year in retirement income, adjusted for inflation annually. You increase the dollar amount each year by CPI โ the percentage stays fixed at your initial 4%.
How Retirement Length Changes Your SWR
The original Trinity Study tested 30-year windows โ appropriate for someone retiring at 65 and living to 95. FIRE retirees who leave work at 40 or 45 face 50-year retirements, and the math shifts meaningfully:
| Retirement Length | Retire At (Age) | Safe Withdrawal Rate | FIRE Number Multiplier |
|---|---|---|---|
| 20 years | ~75 | 4.5% | 22ร |
| 25 years | ~67โ70 | 4.2% | 24ร |
| 30 years | ~60โ65 | 4.0% | 25ร |
| 35 years | ~55 | 3.7% | 27ร |
| 40 years | ~50 | 3.5% | 28.6ร |
| 50 years | ~40 | 3.3% | 30ร |
A 40-year-old retiring today could need 50 years of portfolio support. At 3.3% SWR, a $60,000/year spending level requires $60,000 รท 0.033 = $1,818,000 โ significantly more than the 25ร ($1,500,000) figure the standard 4% rule suggests. This is why many FIRE planners in their 30s and 40s target a 3.3%โ3.5% withdrawal rate and build a larger safety buffer.
Portfolio Size vs. Annual Income: The Full Picture
Here's a complete reference table showing annual income available from portfolios at different SWRs:
| Portfolio Size | At 3.3% SWR | At 3.5% SWR | At 4.0% SWR | At 4.5% SWR |
|---|---|---|---|---|
| $500,000 | $16,500 | $17,500 | $20,000 | $22,500 |
| $750,000 | $24,750 | $26,250 | $30,000 | $33,750 |
| $1,000,000 | $33,000 | $35,000 | $40,000 | $45,000 |
| $1,500,000 | $49,500 | $52,500 | $60,000 | $67,500 |
| $2,000,000 | $66,000 | $70,000 | $80,000 | $90,000 |
| $2,500,000 | $82,500 | $87,500 | $100,000 | $112,500 |
| $3,000,000 | $99,000 | $105,000 | $120,000 | $135,000 |
Sequence of Returns: The Hidden Risk
The biggest threat to any safe withdrawal strategy isn't average returns โ it's sequence of returns risk. Two retirees can experience the exact same average annual return over 30 years and have drastically different outcomes, depending on when the bad years hit.
Retire in 2000 with $1,000,000? The dot-com crash in years 1โ3, combined with $40,000/year withdrawals while the portfolio is down 40%, permanently impairs the portfolio. The same retiree who had a great market in years 1โ5 before a crash is in a completely different position.
The full breakdown of sequence-of-returns risk โ and how to protect against it โ is covered in depth in our article on sequence of returns risk. The short version: cash reserves, a bond tent, and spending flexibility in early retirement are your best defenses.
A major market decline in retirement years 1โ5 is far more damaging than the same decline in years 15โ20. This is why early retirees should consider keeping 2โ3 years of expenses in cash or short-term bonds โ a withdrawal buffer that lets the equity portion recover without being sold at a loss. See also: the bridge fund strategy.
Fixed vs. Flexible Withdrawal Rates
The original SWR research assumes a fixed withdrawal: you set an amount in year one and adjust only for inflation, regardless of portfolio performance. This approach is psychologically simple but financially rigid.
Flexible withdrawal strategies โ also called dynamic withdrawal strategies โ allow you to spend more in good market years and pull back in bad ones. Research by Jonathan Guyton and William Klinger showed that flexible spending rules could support initial withdrawal rates up to 5โ6% while still maintaining portfolio survival, because retirees give up upside in bad years.
Practical options include:
- Guardrail strategy: Increase withdrawals 10% if portfolio is up 20%+ from your initial balance; cut 10% if it drops 20%+ below. Stays within a pre-defined corridor.
- Percentage of portfolio: Always withdraw the same percentage. In a $1M portfolio at 4%, you take $40,000 year one. If it grows to $1.1M, you take $44,000 in year two. Never runs out but income fluctuates.
- Floor-and-upside: Set a minimum spending floor covered by guaranteed income (Social Security, annuity), then pull discretionary spending from the portfolio only in good years.
For most FIRE retirees, a hybrid approach works best: plan conservatively at 3.5%, allow spending to flex up by 10โ15% in strong markets, and have a 2-year cash reserve to avoid selling equities in downturns. The 4% rule explainer covers the full landscape of approaches.
If you're retiring before 55, use 3.3%โ3.5% as your SWR for planning purposes. Yes, this requires a larger portfolio. But it also means you're very unlikely to run out of money even in the worst historical scenarios. The extra cushion usually results in dying with substantially more wealth than you started with โ which many early retirees consider a feature, not a bug, if they want to leave a legacy.
Where the 4% Number Actually Came From: Bengen's 1994 Research
Before the Trinity Study made "4%" a household phrase in the FIRE community, financial planner William Bengen did the original legwork. In 1994, Bengen published "Determining Withdrawal Rates Using Historical Data" in the Journal of Financial Planning. Instead of relying on average market returns โ which had led earlier advisors to recommend withdrawal rates as high as 7% โ Bengen tested every rolling 30-year retirement period in U.S. market history back to 1926.
His approach was simple but rigorous: for each possible retirement start year, hold a portfolio of 50-75% stocks and the rest in intermediate-term government bonds, withdraw a fixed percentage in year one, and increase that dollar amount for inflation every year after. Then check whether the portfolio survived the full 30 years. Bengen found that even in the single worst starting year in the historical record โ a retiree who began withdrawing in 1968, just before over a decade of high inflation and weak stock returns โ a 4.15% initial withdrawal rate survived the full three decades. He rounded this down to 4% and called it the "SAFEMAX" rate: the highest withdrawal rate that would have survived every historical sequence, not just the average one.
This is a critical distinction that gets lost when people casually cite "the 4% rule." It was never designed to be the rate that works on average โ it's the rate calibrated to the single worst 30-year stretch in nearly a century of data. In most historical starting years, a 4% withdrawal rate leaves retirees with far more money at the end of 30 years than they started with. Bengen's own later research found that a more typical outcome left the retiree's portfolio 2-3 times larger after 30 years, with the 4% figure representing a worst-case floor, not a typical result.
Common Mistakes People Make With Withdrawal Rates
The math behind safe withdrawal rates is straightforward, but the way people apply it in real life introduces errors that the original studies never accounted for. Here are the mistakes that show up most often:
- Treating the withdrawal rate as a spending ceiling, not a starting point. The SWR formula tells you your withdrawal amount in year one. Some retirees mistakenly recalculate 4% of their portfolio's current value every single year, which means spending swings wildly with the market โ down sharply after a crash, up sharply after a rally. The original studies assume you set the dollar figure once and then simply adjust it for inflation, not for portfolio performance.
- Ignoring investment fees. A portfolio charged 1% in annual fund and advisory fees behaves, for withdrawal purposes, like it's earning 1% less every year. A "safe" 4% rate calculated against index funds with a 0.05% expense ratio does not transfer cleanly to a portfolio paying 1-1.5% in advisor and fund fees. Retirees in higher-fee products should plan on a meaningfully lower SWR, often 0.5-1 percentage point lower, to reach the same historical survival odds.
- Forgetting one-time and lumpy expenses. A new roof, a child's wedding, a major medical bill, or a car replacement doesn't fit neatly into a smooth monthly withdrawal. Retirees who plan only for steady monthly spending are frequently surprised by these lump sums and end up withdrawing well above their planned rate in the years they occur.
- Not accounting for healthcare cost inflation specifically. General inflation (CPI) has historically run lower than healthcare cost inflation. A withdrawal plan indexed only to CPI can understate how much healthcare will actually cost by the time a retiree reaches their 60s and 70s, when medical spending typically rises as a share of the budget.
- Applying a 30-year-tested rate to a 50-year retirement without adjustment. As covered above, the original 4% figure was tested against 30-year windows. A 35-year-old retiring today may need their portfolio to last 55-60 years. Using 4% instead of the ~3.3-3.5% appropriate for that time horizon is one of the most common and most consequential mistakes early retirees make.
- Panic-selling during a downturn. The math behind every SWR study assumes the retiree stays invested through market declines and continues withdrawing from a diversified portfolio. A retiree who sells equities during a crash to "stop the bleeding" locks in the loss and permanently damages the portfolio's ability to recover โ this single behavioral mistake has historically done more damage to retirement outcomes than any reasonable choice of withdrawal rate.
Safe Withdrawal Rates and Taxes: The Piece Most Calculators Skip
Every SWR calculation above describes a gross withdrawal from the portfolio โ it says nothing about what you get to keep after taxes, and that gap can be substantial depending on which accounts the money comes from.
A withdrawal from a traditional 401(k) or IRA is taxed as ordinary income in the year it's taken. A retiree withdrawing $80,000/year from an all-traditional portfolio might owe $8,000-$12,000 in federal and state income tax, depending on their bracket and state, meaning their real spendable income is closer to $68,000-$72,000, not the full $80,000. A withdrawal from a Roth account, by contrast, is entirely tax-free (assuming the account is qualified), so the same $80,000 gross withdrawal leaves the full amount to spend. A withdrawal from a taxable brokerage account falls somewhere in between โ only the capital gains portion is taxed, and typically at lower long-term capital gains rates, so the effective tax drag is usually smaller than on traditional retirement account withdrawals.
This means two retirees with identical $2,000,000 portfolios and identical 4% withdrawal rates can end up with meaningfully different real-world spending power, purely based on how their money is split between traditional, Roth, and taxable accounts. A retiree who has thoughtfully built a mix of all three account types has far more flexibility to manage their tax bill each year โ pulling more from the Roth in a year they need extra cash without pushing themselves into a higher bracket, or harvesting capital gains at 0% during a low-income year.
Required Minimum Distributions (RMDs) add another wrinkle for anyone with meaningful traditional 401(k)/IRA balances. Starting at age 73 (under current law), the IRS requires a minimum withdrawal each year regardless of whether the retiree actually needs the money to live on. For someone who has been living primarily off taxable and Roth accounts, RMDs can force a withdrawal rate from the traditional account that's higher than their planned SWR โ which is one of the reasons Roth conversion ladders during the lower-income bridge years before RMD age are a popular strategy among early retirees with large traditional balances.
A Worked Example: The First Five Years of Retirement
Numbers become clearer with a concrete walkthrough. Consider a hypothetical retiree, Priya, who retires at 50 with a $1,600,000 portfolio (70% stocks, 30% bonds) and plans to withdraw 3.5% in year one โ $56,000, or about $4,667/month.
Year one: Priya withdraws $56,000. Suppose the market has an average year, and her portfolio (net of the withdrawal) ends the year at $1,650,000 after growth.
Year two: Priya doesn't recalculate 3.5% of her new balance. Instead, she takes her year-one dollar amount and simply adjusts it for inflation. If inflation ran 3% that year, her year-two withdrawal is $56,000 ร 1.03 = $57,680.
Year three: a market downturn. Suppose the market drops 15% this year. Priya's portfolio, after the prior year's growth and withdrawal, was around $1,690,000 entering year three; a 15% decline brings it down to roughly $1,436,500 before that year's withdrawal. Priya still withdraws her inflation-adjusted amount (around $59,410 assuming another 3% inflation year), leaving her portfolio around $1,377,000 heading into year four. This is the exact moment sequence-of-returns risk becomes real โ an early-retirement downturn combined with continued withdrawals shrinks the portfolio at the worst possible time, before it's had years to compound and recover.
Year four and five: If the following two years see a market recovery โ say +18% and +12% โ Priya's portfolio can recover substantially even while she continues withdrawing her (now larger, inflation-adjusted) annual amount. This is the pattern the historical studies actually test: not smooth, steady growth, but exactly this kind of uneven sequence of gains and losses, and the SWR figures already account for stretches like Priya's year three.
The lesson from this walkthrough isn't that downturns don't matter โ they clearly do, especially early in retirement. It's that a properly calibrated SWR (3.3-3.5% for a 45-year retirement horizon like Priya's, rather than the standard 4%) is specifically sized to survive sequences like this one without requiring Priya to cut her spending or go back to work.
How Withdrawal Rates Interact With Social Security and Other Income
Nearly every SWR study models a retiree living entirely off portfolio withdrawals. In practice, most retirees eventually layer in Social Security, and some also have pensions, rental income, or part-time consulting work. These income sources change the withdrawal math in an important way: they reduce the amount that needs to come from the portfolio, which effectively lowers the retiree's real withdrawal rate once that income begins.
Consider a retiree spending $70,000/year with a $1,750,000 portfolio โ a 4% withdrawal rate on the full amount needed. If that retiree also expects $28,000/year in Social Security starting at age 67, then before age 67 the full $70,000 must come from the portfolio (a 4% rate), but after age 67 only $42,000/year needs to come from the portfolio โ a 2.4% rate against the same $1,750,000 balance (ignoring any portfolio growth or decline in the interim, which would change the percentage further). This is why many early retirees structure their plan as a "bridge" period at a higher withdrawal rate before Social Security or a pension begins, followed by a much lower, highly sustainable rate afterward. Modeling this transition accurately โ rather than assuming one flat withdrawal rate for the entire retirement โ is one of the most valuable things a full retirement calculator can do beyond the basic SWR formula.
Rebalancing and Its Effect on Withdrawal Sustainability
The historical SWR studies assume the retiree periodically rebalances back to their target stock/bond mix โ typically annually. Without rebalancing, a strong multi-year bull market gradually pushes a portfolio's stock allocation higher than intended, which increases both expected returns and volatility. A portfolio that drifted from 60% stocks to 80% stocks during a rally carries meaningfully more downside risk heading into the next downturn than the retiree originally planned for.
Annual rebalancing back to target also has a practical side benefit for withdrawals: in years when stocks have outperformed, rebalancing means selling some stocks and buying bonds โ which is also a convenient moment to raise cash for the next year or two of withdrawals. In years when stocks have underperformed, rebalancing means buying stocks while they're relatively cheap, funded by trimming the bond allocation, rather than being forced to sell depressed stocks to cover living expenses. This is part of why a disciplined rebalancing schedule and a sensible cash buffer work together rather than being separate strategies โ each makes the other easier to execute consistently, especially in a year when doing so runs against instinct.
A practical rebalancing rule that many retirees find easier to stick to than a rigid calendar date: rebalance when an asset class drifts more than 5 percentage points from its target, checked once a quarter. This "threshold" approach means quiet years require no action at all, while a year with a large market move triggers a rebalance automatically, without requiring the retiree to predict or time anything. Combined with a cash buffer covering 1-2 years of spending, this keeps the portfolio close to its intended risk level without constant manual intervention.
This article is for educational purposes only and does not constitute financial advice. MyFIRE is not a registered investment advisor. Always consult a qualified fee-only CFP before making retirement decisions.
Calculate your exact FIRE number with MyFIRE
Free retirement planner with bridge fund calculator, Monte Carlo simulation, and AI insights. See exactly when you can reach financial independence.
Start planning โ it's free โ