The 4% Rule at FatFIRE Scale: Safe Withdrawal Rates

Does the 4% Rule Still Work for Early Retirees?

The 4% rule says you can withdraw 4% of your portfolio in your first year of retirement, adjust that dollar amount for inflation every year after, and expect the money to last 30 years. For a traditional retiree at 65, the rule has held up remarkably well: its creator, William Bengen, now argues the historically safe rate is 4.7%, not 4%. For a FatFIRE retiree leaving work at 45 with a 50-year horizon, the honest answer is different. The rule was validated on 30-year periods, and the math degrades as the horizon stretches. Over 50 to 60 years, historical safe rates fall to roughly 3.1% to 3.5%, and a rigid 4% withdrawal carries a meaningful chance of ruin if you retire into high valuations and refuse to adjust.

That is not a reason to abandon the framework. It is a reason to understand exactly where it came from, where it breaks, and which of the modern alternatives fits a large portfolio and a long horizon. This article covers the primary research, the failure modes, and a worked example for a $5M portfolio at age 45. When you want to stress-test your own numbers against thousands of simulated market sequences, the withdrawal calculator on this site runs Monte Carlo simulations with Guyton-Klinger guardrails and account-by-account withdrawal sequencing.

Where Does the 4% Rule Come From?

The 4% rule comes from William Bengen's 1994 paper "Determining Withdrawal Rates Using Historical Data", published in the Journal of Financial Planning. Bengen tested every rolling 30-year retirement starting from 1926, using a 50/50 portfolio of S&P 500 stocks and intermediate-term Treasuries, rebalanced annually, with withdrawals adjusted for CPI inflation. The highest initial withdrawal rate that survived every single historical 30-year period, including retirements beginning in 1929, 1937, and the late 1960s, was approximately 4%. Bengen called this worst-case rate SAFEMAX. He never called it the 4% rule; the nickname came later.

Two details from the original paper matter more than the headline number. First, 4% was the worst case, not the average. Most historical retirees could have withdrawn 5%, 6%, or more and died with money left over. The rule is calibrated to the unluckiest retiree in a century of data, which is why treating it as a hard ceiling leaves most people underspending. Second, the binding scenario was not a market crash. It was the retiree who quit in the late 1960s and immediately ate a decade of stagflation: mediocre nominal returns with high inflation relentlessly ratcheting up the required withdrawal. Bengen told CNBC in 2025 that inflation, not bear markets, is retirees' greatest enemy.

The Trinity study followed in 1998. Three Trinity University professors (Cooley, Hubbard, and Walz) reframed the question in terms of success rates rather than a single worst-case rate, testing withdrawal rates of 3% to 12% against rolling periods from 1926 to 1995. Their headline result: a 4% inflation-adjusted withdrawal succeeded in 95% of 30-year periods with a 50/50 portfolio and 98% with a 75/25 stock-heavy portfolio. The Trinity study is the source of the "95% success rate" framing that most retirement calculators still use.

Note what neither study tested: horizons beyond 30 years, taxes, fees, or any portfolio more complex than two US asset classes. Every one of those omissions matters at FatFIRE scale.

What Do the Original Researchers Say Now?

Bengen has spent three decades revising his own number upward. In 2006 he added small-cap stocks to the test portfolio and raised SAFEMAX to 4.5% for tax-advantaged accounts and 4.1% for taxable accounts. In his 2025 book, A Richer Retirement (Wiley), he re-ran the analysis with a portfolio diversified across several equity asset classes, including small and micro caps, and raised his worst-case historical rate to 4.7%, which he now calls the universal SAFEMAX. In his re-run of roughly 400 historical retirement scenarios, only one cohort, the 1968 retiree, actually required a rate as low as 4.7%. He has also said that most retirees who do not begin their retirement at extreme valuations could reasonably start at 5.25% to 5.5%.

Morningstar's annual State of Retirement Income research takes the opposite approach: forward-looking Monte Carlo simulation using current valuations and yield-based return expectations rather than raw historical data. Their recommended starting safe withdrawal rate has moved with market conditions: 3.3% in 2021, 3.8% in 2022, 4.0% in 2023, 3.7% in 2024, and 3.9% in the 2025 edition published December 3, 2025. That 3.9% assumes fixed real withdrawals, a 30-year horizon, a 90% success target, and, notably, an equity allocation of only 30% to 50%. Morningstar's same report shows that flexible strategies change the picture dramatically: a TIPS ladder supported 4.5% at the time of publication, and a guardrails approach supported a 5.2% starting rate on a 40/60 portfolio.

So the two most-cited authorities currently disagree by nearly a full percentage point (4.7% vs 3.9%), and both are defensible. Bengen answers "what survived the worst market sequence in recorded US history with a diversified portfolio?" Morningstar answers "what has a 90% chance of surviving the next 30 years given today's valuations and yields?" Neither answers the FatFIRE question, because both assume a 30-year retirement.

What Is Sequence-of-Returns Risk?

Sequence-of-returns risk is the risk that poor market returns early in retirement permanently impair a portfolio you are actively withdrawing from, even if long-run average returns turn out fine. When you sell shares into a drawdown to fund fixed spending, those shares are gone; the portfolio that remains is too small to fully participate in the recovery. Two retirees can experience identical 30-year average returns and end up with wildly different outcomes purely because of the order in which those returns arrived.

The definitive work on how long this danger window lasts comes from Michael Kitces. His analysis of historical safe withdrawal rates found that the correlation between a retiree's first-year return and their eventual 30-year safe withdrawal rate is only about 0.21. The correlation with the first decade of annualized real returns is roughly 0.8, peaking around year nine or ten, then falling to about 0.43 for full 30-year returns. In plain terms: a single crash at the start of retirement is survivable. A mediocre first decade is what kills portfolios. The 1966 to 1982 stretch, not October 1929, is the nightmare scenario.

This has a direct implication for early retirees: a 45-year-old faces the same dangerous first decade as a 65-year-old, but with 20 extra years of withdrawals stacked on the far side of it. The exposure to a bad sequence is not proportionally larger; the consequences of one are. This is the core reason long-horizon safe rates sit below 30-year safe rates, and it is also why every serious modern withdrawal method is dynamic: they all, in different ways, cut withdrawals when the sequence turns hostile instead of blindly ratcheting spending up with inflation into a falling market.

How Does the 4% Rule Change Over a 50-Year Horizon?

Over a 50-year horizon, the 4% rule stops being conservative and becomes a coin you would rather not flip. The most thorough public research on long-horizon withdrawal rates is Karsten Jeske's Safe Withdrawal Rate Series at Early Retirement Now, which extends the Bengen-style historical simulation back to 1871 and out to 60-year retirements, computed monthly. His findings, summarized across the series and in ChooseFI's analysis of the research: a 4% withdrawal from a 50/50 portfolio succeeded in about 95% of historical 30-year periods but only about 65% of 60-year periods. The failure rate roughly septuples when the horizon doubles.

Three structural findings from that research matter for anyone planning a 40-plus-year retirement:

• Historical failsafe rates fall with horizon but do not collapse. For equity-heavy portfolios, the rates that survived every historical start date land around 3.5% or lower for 40 to 60-year horizons, with roughly 3.25% at the 50 to 60-year end. The extra 20 to 30 years of retirement costs you about 0.5 to 0.75 percentage points of withdrawal rate, not half your spending.
• Long horizons demand more equity, not less. Over 30 years a bond-heavy portfolio can limp to the finish line. Over 50 years, inflation compounds too long for bonds to carry the load; Jeske's simulations show 75% or higher equity allocations paired with withdrawal rates of 3.5% or below offering the strongest success rates across all horizons. Morningstar's 30 to 50% equity recommendation is built for a 30-year problem and is arguably dangerous for a 50-year one.
• Final-value targets matter at this scale. Standard success metrics count a portfolio that ends at $1 as a win. Most FatFIRE retirees want capital preservation, whether for heirs, optionality, or psychological comfort. Requiring the portfolio to retain even 50% of its real value at the end of the horizon shaves the sustainable rate further. Decide which game you are playing before you pick a number.

For context on how withdrawal strategy fits into the broader sequence of decisions between your first savings milestone and full independence, see the financial independence hub.

When Does the 4% Rule Actually Fail?

The 4% rule fails in one specific and repeatable scenario: retiring at high equity valuations immediately before an extended period of poor real returns, with inflation doing more damage than the crash itself. Every historical failure or near-failure cohort fits this pattern. The 1929 retiree faced a deflationary collapse and mostly survived 4% because falling prices reduced required withdrawals. The retirees of 1965 through 1969 are the ones who actually broke the rule: they retired at then-elevated valuations into fifteen years of stagflation, and they are the reason SAFEMAX is 4.x% instead of 5.x%.

That failure profile is worth taking seriously right now. The Shiller CAPE ratio stood at roughly 41 in August 2026, the second-highest level in history after the 2000 dot-com peak. High CAPE does not predict a crash, but across 150 years of data it has reliably predicted below-average real returns over the following decade, which is precisely the window sequence risk lives in. Jeske's valuation-conditional analysis found that historical failures of the 4% rule cluster almost entirely in retirements that began when CAPE was elevated; when CAPE was below 20, 4% was essentially never the binding constraint, even over long horizons.

Being precise about what "failure" means also deflates some of the doom. Failure in these studies means depleting the portfolio to zero while blindly following the rule: never cutting spending, never earning another dollar, never adjusting anything for decades while the account balance visibly collapses. No FatFIRE household behaves that way. A $5M retiree whose portfolio drops to $2.8M in year six does not keep raising withdrawals with CPI; they trim, they adjust, and the "failure" becomes a decade of somewhat lower spending instead of ruin. The realistic risk at this scale is not eating cat food. It is being forced into spending cuts you did not plan for, at the worst possible time, because you anchored to a number calibrated for a different problem. The methods below exist to make those adjustments systematic instead of panicked.

Which Withdrawal Method Should You Use?

No single method dominates; each trades simplicity, starting income, and spending stability differently, and the right choice depends on how much of your budget is discretionary. Here is the current landscape, using rates from each method's own primary research:

Three observations for the $5M+ reader. First, the spread between the most conservative defensible answer (roughly 3.0% from the CAPE rule at current valuations) and the most aggressive (5.6% Guyton-Klinger) is $130,000 per year on a $5M portfolio. The method choice is worth more than most investment decisions you will ever make. Second, every method built after 2000 is dynamic. The research consensus has quietly converged on the idea that spending flexibility, not a lower fixed rate, is the efficient answer to sequence risk; Morningstar's own report shows flexibility raising sustainable starting rates by more than a full percentage point. Third, guardrails and amortization methods are only as good as your willingness to execute the cut when it triggers. If your baseline spending has no discretionary layer, a 10% cut is a lifestyle event, and you should start from a lower fixed rate instead.

The Guyton-Klinger mechanics deserve one paragraph of precision, since the method gets summarized sloppily. The 2006 paper defines four rules. The capital preservation rule: if your current withdrawal rate (this year's dollars divided by the current balance) rises more than 20% above your initial rate, cut the dollar withdrawal 10%; suspended in the final 15 years of the plan. The prosperity rule: if it falls more than 20% below the initial rate, raise the withdrawal 10%. The withdrawal rule: skip the inflation adjustment in any year following a portfolio loss when your withdrawal rate is above its initial level. The portfolio management rule governs which assets fund the withdrawal, selling winners first and avoiding equity sales after down years. With all four rules and at least 65% equities, the paper supported initial rates of 5.2% to 5.6% at 99% success over 40 years. Those are the exact rules implemented in the withdrawal calculator, so you can watch how often the guardrails would have triggered for your specific numbers rather than trusting the summary statistics.

What Does This Look Like for a $5M Portfolio at 45?

Take the canonical FatFIRE case: age 45, $5M liquid portfolio, 50-year planning horizon to age 95, no pension, Social Security treated as a bonus rather than a plan. Here is what each framework actually prescribes in year one:

The honest synthesis: the floor-and-flex answer. The historical record says $162,500 of inflation-adjusted spending was bulletproof at this horizon; call the region around $150,000 to $165,000 (3.0 to 3.3%) the floor that never has to move. The guardrails research says starting at $230,000 to $260,000 is defensible if, and only if, the top $60,000 to $100,000 of that spending is genuinely discretionary: travel, gifting, toys, the second property's operating costs. Structure the budget so the floor covers everything contractual and the flex layer absorbs the guardrail cuts, and you capture most of the higher starting income while keeping worst-case outcomes boring. What you should not do is split the difference to a flat $200,000 and never look at it again, which combines the low income of the conservative methods with the tail risk of the aggressive ones.

Two adjustments before any of those numbers become real. First, they are gross of taxes, which the next section addresses. Second, they assume zero future earned income. Kitces' sequence-risk finding cuts both ways: because the first decade determines nearly everything, even modest earnings early on are disproportionately valuable. A 45-year-old who covers $50,000 of a $200,000 budget with consulting income for the first five years is not withdrawing 4%; they are withdrawing 3% during exactly the window where withdrawal rate matters most. Run your own version of this table, with your accounts, allocation, and income assumptions, in the withdrawal calculator before committing to a number.

How Do Taxes Change Your Real Withdrawal Rate?

Every rate in this article is pre-tax, and at FatFIRE scale the gap between gross withdrawal and spendable income is large enough to change the plan. Kitces modeled the question directly: for a retiree paying 25% ordinary income and 15% capital gains rates, a roughly 4.1% pre-tax safe withdrawal rate becomes about 3.4% on an after-tax basis, a drag of 0.6 to 0.7 percentage points, rising toward a full percentage point in top brackets. On a $5M portfolio, that is $30,000 to $50,000 a year that exists in the calculator but not in your checking account.

The drag is not uniform across account types, which is why the same headline rate supports very different lifestyles depending on where the money sits:

• Roth accounts have zero drag: a 4% withdrawal is 4% of spending.
• Traditional 401(k)/IRA dollars are taxed as ordinary income on the way out; at a 30% effective rate, a $200,000 withdrawal spends like $140,000. Large deferred balances also carry an embedded liability that arrives on a schedule you do not control: required minimum distributions currently begin at age 73 under SECURE 2.0.
• Taxable accounts sit in between, and better than most people assume. Only the gain portion of each sale is taxed, losses can be harvested against gains, and long-term rates top out at 23.8% federal (20% plus the 3.8% net investment income tax). Kitces notes the drag here is partially self-mitigating: taxes are only owed in the scenarios where the portfolio actually appreciated.

For an early retiree, the years between the last paycheck and RMDs are the single largest tax lever in the entire plan. Someone retiring at 45 with a significant traditional balance has nearly three decades of low-bracket years in which to convert deferred dollars to Roth at rates that will never be available again once RMDs and Social Security stack up. The withdrawal-sequencing decision (which account funds which year, and what gets converted along the way) routinely swings lifetime after-tax wealth by six to seven figures at this portfolio size, and it interacts with everything else in this article: conversions are cheapest in exactly the down-market years when guardrail methods are cutting withdrawals. Sequencing and location strategy get full treatment in the retirement planning hub; the practical point here is that your withdrawal method and your tax plan are one system, not two.

What Should a FatFIRE Retiree Actually Do With the 4% Rule?

Use it as a diagnostic, not a prescription. If your planned spending is under 3% of your portfolio, every framework in this article, from Morningstar's cautious Monte Carlo to Bengen's aggressive re-run, says you are done; the remaining questions are tax and estate questions. If you are between 3% and 4% on a 50-year horizon, you are in the zone where method choice, equity allocation, valuation levels, and spending flexibility genuinely determine the outcome, and a static rule is the one tool that cannot help you. Above 4.5% at age 45, you are relying on either continued earned income or the specific kindness of the return sequence, and you should know that is the bet you are making.

The research stack, compressed to five load-bearing facts: the 4% rule was the worst case over 30 years, not the expectation (Bengen 1994); its own author now puts the 30-year worst case at 4.7% while the most conservative major research house says 3.9%; stretching the horizon to 50-plus years pulls the historical failsafe down to roughly 3.25 to 3.5%; the first decade of returns decides nearly everything (Kitces); and systematic flexibility buys back most of what the long horizon takes away (Guyton-Klinger). Everything else is implementation.

Sources

• Bengen, W. P., "Determining Withdrawal Rates Using Historical Data", Journal of Financial Planning, October 1994
• Cooley, P. L., Hubbard, C. M., Walz, D. T., "Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable" (the Trinity study), AAII Journal, February 1998
• Bengen, W. P., A Richer Retirement (Wiley, 2025); summarized in Advisor Perspectives, Forbes, and CNBC
• Bengen on 5.25 to 5.5% for typical retirees: ChooseFI analysis
• Morningstar, The State of Retirement Income: 2025 Edition (December 3, 2025); year-by-year history via Financial Advisor Magazine and Keil Financial Partners
• Jeske, K., The Safe Withdrawal Rate Series, Early Retirement Now (2016 to present); long-horizon findings summarized at ChooseFI and CXO Advisory; valuation analysis in Part 3; CAPE-based rules in Part 18
• Kitces, M., "Understanding Sequence of Return Risk", Nerd's Eye View
• Kitces, M., "The Impact of Taxes on the Safe Withdrawal Rate", Nerd's Eye View
• Guyton, J. T., Klinger, W. J., "Decision Rules and Maximum Initial Withdrawal Rates", Journal of Financial Planning, March 2006
• Total Portfolio Allocation and Withdrawal (TPAW) and Amortization Based Withdrawal, Bogleheads wiki
• Shiller CAPE ratio, current level: multpl.com
• IRS, Publication 590-B, Distributions from Individual Retirement Arrangements (RMD age under SECURE 2.0)
• William Bengen, Wikipedia (2006 revision to 4.5% tax-advantaged / 4.1% taxable)