What is the 4% rule, exactly? Bengen, Trinity, and the real math

Quick answer

The 4% rule says you can withdraw 4% of your retirement portfolio in year one and adjust that dollar amount for inflation each year afterward, and a 50–75% stock portfolio historically survived 30 years of withdrawals in ~95% of historical periods. It’s where the FIRE-community 25× annual expenses target comes from (1 ÷ 0.04 = 25).

For retirements longer than 30 years — most early-FIRE cases — a 3.25–3.5% rate (28.6–33× expenses) is the more defensible starting point.

The “4% rule” is the most-quoted number in FIRE math. It’s also one of the most-misunderstood. The original research doesn’t say you can safely withdraw 4% forever. It says that in roughly 95% of historical 30-year periods, a portfolio of 50–75% stocks and 25–50% bonds would have survived a 4%-of-starting-balance, inflation-adjusted annual withdrawal.

That sentence packs a lot of qualifiers. This guide unpacks them.

Where the 4% rule comes from

Two papers anchored the current FIRE math:

Bengen (1994). William Bengen, a financial advisor, looked at every rolling 30-year retirement window from 1926 onward in US data. He tested withdrawal rates against a portfolio of stocks and bonds, asking: at what initial withdrawal rate (adjusted for inflation each year) does a portfolio survive 30 years without going to zero, even in the worst historical sequences?

His answer: roughly 4%. That became “the SAFEMAX” — the historical worst-case sustainable withdrawal rate. Bengen later refined this upward to about 4.5% with different assumptions and longer data.

Trinity Study (1998). Three Trinity University researchers — Cooley, Hubbard, and Walz — followed up with a similar analysis using more granular failure-rate data. They published tables of success rates across stock/bond mixes and withdrawal rates. The 4% rate at a 50/50 stock-bond portfolio had ~95% historical success over 30 years. At 5%, success dropped to ~70%. At 6%, success was a coin flip.

Both papers used the same dataset (US stocks/bonds, 1926–1995) and roughly agreed: 4% is the floor that survived nearly every historical sequence over 30 years.

What “the rule” actually says

Stripped to essentials:

Starting withdrawal: 4% of your portfolio at the moment you retire.
Subsequent withdrawals: that same dollar amount, adjusted upward each year for inflation.
Portfolio mix: 50–75% stocks, balance in bonds (or equivalents).
Horizon: 30 years.
Failure threshold: portfolio reaches zero before year 30 ends.

A FIRE target of “25× your annual expenses” is just the 4% rule inverted: if you spend $40,000/year, you need $1M, because 4% of $1M is $40k.

What the rule doesn’t say:

That 4% works forever
That 4% works for retirements longer than 30 years
That 4% accounts for fees, taxes, or healthcare inflation outpacing general CPI
That the historical worst case is the worst case
That international or future returns will match the US 1926–1995 baseline

Why the rule isn’t 5% or 3.5%

It almost was. The Trinity data shows that at a 75/25 stock-bond mix:

3.0% withdrawal — 100% success across all 30-year windows
3.5% — ~99% success
4.0% — ~95% success
4.5% — ~89% success
5.0% — ~70% success
6.0% — ~50% success

The 95% / 100% threshold is somewhere between 3.5% and 4.0%. Bengen picked 4% partly because the failures were concentrated in a handful of bad sequences (retiring just before the Great Depression, or just before the 1973–74 stagflation). The other 95% of retirees who followed it would have ended the 30 years with more than they started with — sometimes vastly more.

This is the asymmetry: most people retiring on the 4% rule die rich. A small minority run out of money. Whether you’d accept that risk depends on:

Whether you have backup income (Social Security, pension, part-time work)
Whether you can flex spending in down years
Whether your retirement is shorter or longer than 30 years
Whether you assume returns going forward will match the past

The arguments for lower than 4%

Several researchers have argued the rule should be 3.0–3.5%:

Wade Pfau — uses current valuations (CAPE ratios, bond yields) and finds the rule’s 95% success rate may be optimistic when starting from today’s market. His safe rate is closer to 3.0–3.5% for current entrants.

Big ERN (Karsten Jeske) — long-running blog series running the Trinity math over longer horizons and international data. For 50- and 60-year retirements (which many FIRE practitioners are targeting), the safe rate falls to ~3.25%.

Morningstar — has periodically updated SWR projections downward, citing lower expected returns. Recent estimates have been in the 3.3–3.8% range for a balanced portfolio over 30 years.

These analyses are credible. They make different assumptions than Bengen’s original study. If you’re retiring at 40 and planning to live to 95, “30-year horizon” doesn’t apply to you, and 3.0–3.5% is a better starting point.

The arguments for higher than 4%

The original Bengen rule was conservative on purpose. Several refinements argue for 4.5–5%:

Bengen himself — bumped his SAFEMAX to 4.5% with broader asset classes (small-cap value, mid-cap, international). The added diversification raised the historical floor.

Guyton-Klinger flexible spending — a set of rules (the “guardrails”) that adjust withdrawals upward in good years and modestly downward in bad ones. Starting at ~5% with these rules has comparable success rates to a constant 4%.

Variable percentage withdrawal (VPW) — withdraw a percentage of current balance rather than the inflation-adjusted starting amount. Mathematically can’t run out (it asymptotes toward zero), but income fluctuates with the market.

Early retirees with optionality — if you can return to part-time work in a bad sequence, 5% might be defensible because you’ve effectively bought sequence-of-returns insurance.

What the rule assumes about you

The 4% rule presumes a specific retiree:

US-based. Other countries have lower historical equity returns. Vanguard’s global SWR estimates for non-US investors are typically 0.5–1% lower.
Tax-aware. The rule talks about gross withdrawals. If your money is in a traditional 401(k), 25% of each withdrawal goes to taxes — you need a higher gross to net the same.
Held the portfolio. Most failures came from retirees who panicked and sold during downturns. The rule assumes you’ll stay invested through 30% drawdowns.
Inflation-aware. The rule adjusts withdrawals by general CPI. If your spending categories (healthcare, certain services) inflate faster, your real purchasing power declines.
Without late-life cost shocks. Long-term care, medical events, supporting children. The smooth 4% trajectory doesn’t model these.

If any of these don’t fit, the right SWR for you is different from 4%.

A reasonable framework

The rule isn’t useless. It’s a benchmark — a starting point that tells you when you’re in the right ballpark.

For a 30-year retirement at standard ages (60–65 start): 4% is defensible. You probably won’t run out. You’ll likely end with more than you started.
For an early-retirement 50-year horizon: drop to ~3.25–3.5%. Sequence risk is too unforgiving over half a century.
With flexible spending and a willingness to earn part-time: 4.5% with the Guyton-Klinger guardrails is reasonable.
In years following a severe market drop: don’t increase your withdrawal in real terms. Hold steady or trim. This single behavioral rule covers most of the gap between 4% and 5%.
With significant Social Security or pension income coming later: you can use a higher rate now because that income reduces sequence risk during the front half of retirement.

Connecting it to your FIRE number

The 4% rule defines the target. If your annual expenses are E:

25× E = needed portfolio at 4% SWR (the standard)
~29× E = needed at 3.5%
~33× E = needed at 3.0%

A household spending $60k/year:

$1.5M for 4% SWR
$1.71M for 3.5% SWR
$2.0M for 3.0% SWR

The extra $500k of target represents the “safety margin” for a longer-than-30-year retirement or a more conservative withdrawal assumption.

Use the Standard FIRE calculator to plug in your expenses against your chosen SWR. The default is 4%; flex it for your situation.

What to do with this

The 4% rule is a floor, not a ceiling, for most 30-year retirements. For longer retirements, treat it as a ceiling instead. For variable spending or part-time income availability, treat it as a midpoint.

The number itself matters less than the underlying behavior:

Pick an SWR that fits your horizon and risk tolerance
Don’t increase your withdrawal in real terms during downturns
Plan for healthcare and long-term care separately
Use the 4% rule as a sanity check, not a vow

Compounding gets you to FIRE. The 4% rule helps decide when you’re there. The next risk — sequence of returns — is what the rule was specifically calibrated to survive, and the next post in this series walks through why early bad years are so much worse than late ones.

Go deeper:

How to Calculate Your FIRE Number — the full framework the 4% rule plugs into.
Sequence-of-returns risk — why the same average return can produce wildly different outcomes.
Compound interest, the formula behind FIRE — how saving rate turns into a portfolio.
Savings rate: the one number that decides time-to-FIRE — the upstream lever.

Use the Standard FIRE Calculator with your chosen SWR (default 4%, flex to 3.5% for very long retirements) to see your number.

Educational content, not financial advice. The 4% rule is calibrated on US historical data and 30-year horizons; future returns and longer horizons may behave differently. Consult a fee-only fiduciary before retiring.

Frequently asked questions

What is the 4% rule in simple terms? +

The 4% rule says that if you withdraw 4% of your retirement portfolio in year one and adjust that dollar amount for inflation each year afterward, a 50–75% stock portfolio historically survived 30 years of withdrawals in ~95% of historical periods. It's where the FIRE-community '25× annual expenses' target comes from: 1 ÷ 0.04 = 25.

Who came up with the 4% rule? +

Financial advisor William Bengen, in his 1994 paper 'Determining Withdrawal Rates Using Historical Data.' The Trinity Study (Cooley, Hubbard, Walz, 1998) re-ran similar tests with broader portfolio mixes and confirmed the result. Both used US data going back to 1926 and 30-year retirement horizons.

Does the 4% rule still work in 2026? +

It's still the most commonly cited starting point, but several researchers (Wade Pfau, Big ERN, Morningstar) argue current valuations and bond yields justify a safer 3.0–3.5% rate, especially for retirements longer than 30 years. The 4% number itself isn't broken; the assumptions behind it (30-year horizon, US-only returns) just don't fit every FIRE retiree.

Is 3.5% or 4% safer for early retirement? +

For a 40–50 year early-retirement horizon, 3.25–3.5% is the more defensible starting point. The original 4% study was calibrated to 30 years; longer horizons compound sequence-of-returns risk in ways the original math doesn't fully address. The cost is a larger target portfolio: 28.6× expenses at 3.5% versus 25× at 4%.

What does the 4% rule assume? +

A 30-year retirement horizon, a 50–75% stock / bond portfolio in US assets, that you stay invested through downturns, that you don't have major late-life cost shocks, and that you take inflation adjustments by general CPI. It also assumes pre-tax accounting — if your savings are in a traditional 401(k), part of every withdrawal goes to income tax, which the rule doesn't model.

How does the 4% rule connect to a FIRE number? +

The FIRE number is simply 1 ÷ SWR × annual expenses. At a 4% safe withdrawal rate (SWR), that's 25× expenses. At 3.5%, 28.6×. At 3%, 33×. See the full [FIRE number guide](/blog/how-to-calculate-your-fire-number) for the rest of the framework.

← All posts