TL;DR
At a 50-year horizon, the historical 95%-survival rate is approximately 3.25–3.5% for a 75/25–100/0 equity-heavy portfolio. Below that, almost every cohort survives; above it, sequence risk becomes the dominant failure mode.
The horizon that breaks the 4% rule
50 years is the planning horizon for someone retiring at 42 and expecting to live to 92. That's the heartland of the FIRE community. It's also the point where the safe withdrawal rate diverges sharply from the conventional 4%.
The mechanism is straightforward: a 50-year window contains roughly 1.5× as many bad-sequence starting points as a 30-year window, and far more compounding for whatever damage those sequences do early on. The Shiller record only contains around 100 fully-tested 50-year cohorts (the data ends in 1974 for full 50-year-from-now windows), but the worst ones are catastrophic for high withdrawal rates.
The historical numbers
Running every full 50-year window from 1871 to 1974 in the US Shiller data, with a 75/25 stock/bond allocation and rigid inflation-adjusted withdrawals:
- 3.0%: 100% historical survival
- 3.25%: ~98% historical survival
- 3.5%: ~93% historical survival
- 3.75%: ~82% historical survival
- 4.0%: ~72% historical survival
- 4.5%: ~50% historical survival
The 95% threshold sits between 3.25% and 3.5%. Most academic work rounds it to 3.5%, which captures the spirit of the test without being needlessly conservative.
The cohorts that broke higher rates
Four starting years dominate the failure list at 4%:
- 1929: classic Depression. Survives in nominal terms thanks to deflation, but the 1942–1965 bull market is what actually saves it.
- 1937: starts during recovery, but a sub-1942 second crash plus inflation in the 1940s erodes principal badly.
- 1966: the worst US cohort in the data. 16 years of low real returns at the start, no bull market until 1982.
- 1969: similar to 1966 with slightly better early years and a worse middle.
Notably absent: 2000. The dot-com bust + 2008 cohort hurts at 40-year horizons but hasn't been fully tested at 50 years yet — the 2000 cohort doesn't complete its 50-year window until 2050.
If your plan needs to survive 1966 (and any prudent FIRE plan does), the safe rate is closer to 3.5%. If you're confident the future will resemble the milder cohorts only, 4% is defensible.
What 0.5% does to the FIRE number
The difference between 3.5% and 4% as your assumption is substantial:
- At 3.5%: FIRE number = expenses × 28.6
- At 4.0%: FIRE number = expenses × 25
For a £40,000/year target: £1.14m vs £1.00m. The £140,000 gap represents roughly 1.5–2 years of additional accumulation for someone saving £75,000/year.
That's the explicit cost of moving from "I'll accept a 70% historical survival rate" to "I want 93%+". Most planners take the buffer. For context on this trade-off, see our 3.5% rule article.
How flexibility brings 4% back
If you'll commit to cutting 10–20% of discretionary spending when the portfolio drops materially below its real high-water mark, 4% becomes viable again at 50 years:
- Rigid 4% at 50 years: ~72% historical survival
- 4% with 15% guardrail cuts: ~92% historical survival
The flexibility recovers roughly two-thirds of the safe-rate headroom you'd otherwise give up by going from 30 years to 50 years. For most FIRE planners this is the right balance — a higher target rate, with explicit pre-commitment to cut spending when sequences turn bad. See our guardrails article for the mechanics.
What about 100% equities?
At long horizons, an all-equity allocation actually has a marginally higher historical safe rate than 75/25, because equities compound through inflation better. The trade is volatility — the worst-case drawdowns are deeper. Most FIRE planners are temperamentally better off with 75/25 + flexibility than 100/0 + rigidity, even if the math slightly favours the latter.
The pragmatic answer
For a 50-year FIRE plan:
- Conservative: 3.25% (30.8× expenses) — every historical cohort survives
- Balanced: 3.5% (28.6× expenses) — 93% historical survival, the modern default
- Flexible: 4.0% (25× expenses) with guardrails — historically robust with adaptive spending
Run all three against the Shiller record in our withdrawal survival tool. The output names the specific historical cohorts each rate would have broken in, which is a more useful answer than any single percentage.
Frequently asked questions
- Has any 50-year period survived 4.5%?
- Yes — most have, but not enough for a 95% survival benchmark. Roughly 50% of US 50-year cohorts survived 4.5% rigidly. Adding flexibility lifts that to 80%+.
- Is 3.25% too conservative?
- It's the rate that survived every 50-year US cohort since 1871. If you want zero historical failures, that's your number. If you can tolerate small adjustments, 3.5–3.75% is more practical.
- How does this interact with the state pension?
- Knock the state pension's present value off your required portfolio before applying the rate. For a 45-year-old UK planner, the state pension is worth roughly £180,000 of present-value buffer — that reduces the effective FIRE number meaningfully.
Stress-test your own FIRE plan
FIRE Wealth OS runs your savings rate and expenses against every historical market starting point since 1871. Free to use, no card required.