Is the 4% rule still safe in 2026? A re-examination

Bill Bengen's 4% rule, published in 1994, is now older than the average user reading this post. It was modeled on US data, US bonds, and a 30-year retirement. None of those assumptions hold cleanly in 2026.

What's changed since 1994

• Bond yields. 1994 had 30-year Treasuries at ~7%. 2026 hovers near 4-5%. Lower bond yields mean lower portfolio income across a 60/40 mix, mechanically reducing safe withdrawal.
• Equity valuations. The Shiller CAPE ratio sits well above its long-term average. Higher starting valuations historically correlate with lower forward 10-year returns.
• Longevity. A 65-year-old today has a meaningful probability of living past 95. Modeling 30 years undercounts tail risk; 35-40 is more honest.
• Inflation regime. 2021-23 reminded us inflation isn't dead. Sequence-of-inflation is as dangerous as sequence-of-returns and the 4% rule wasn't stress-tested for it.

What the updated research says

Morningstar's 2024 update settled on 3.7% as the "new safe withdrawal" for a typical 60/40 portfolio over 30 years. Researchers using Monte Carlo with 2026 valuations land between 3.3% and 3.8% for a 95% success rate. 4% is now closer to ~85% success.

What to do with this

1. Plan at 3.5%, not 4%. If you can retire at 4%, you can definitely retire at 4% — but planning at 3.5% and spending 4% in good years is safer than the reverse.
2. Use guardrails. Guyton-Klinger or Vanguard's dynamic spending model adds 50-100bps of safe withdrawal by giving you flexibility instead of a hardcoded 4%.
3. Model 35 years, not 30. A 60-year-old might retire 35-40 years. Run your simulator for the longer horizon.
4. Keep a cash buffer. Two to three years of expenses in cash + short bonds insulates you from selling equities into a 30% drawdown.

k25x's retirement simulator runs all of this — guardrails, custom horizon, current valuations — locally on your numbers.