Safe Withdrawal Rate by Age: How Your Retirement Horizon Changes the Math

Safe withdrawal rate drops as your retirement horizon lengthens because there's more time for a bad sequence of returns to destroy your portfolio. The Trinity Study tested 30-year horizons and validated 4%. Extensions by Wade Pfau, Michael Kitces, and others tested 40, 45, and 50-year horizons against the same historical data and found that the safe rate drops to roughly 3.5% at 40 years and 3.25% at 50 years. The math is straightforward: longer horizons mean more inflation-adjusted withdrawals taken against the same initial portfolio, and more opportunity for early-retirement market crashes to deplete the base. The specific numbers used here are approximate — exact rates depend on asset allocation, starting CAPE ratio, and willingness to use dynamic strategies.

Bengen's 1994 paper and the 1998 Trinity Study both focused on 30-year horizons — the average retirement length for someone retiring at 65. As the FIRE movement grew in the 2010s, early retirees began questioning whether 4% was still safe for 40-50 year horizons. Wade Pfau's 2010 paper 'An International Perspective on Safe Withdrawal Rates' was one of the first to systematically extend the analysis to longer horizons. Michael Kitces's blog has published extensive work on horizon-adjusted SWRs. The Trinity Study's 2011 update also briefly addressed longer horizons, finding that 4% had meaningful failure rates beyond 30 years.

Horizon-based SWR tables are point estimates that hide a lot of variation. The exact safe rate depends on: (1) starting portfolio allocation (75/25 tends to be optimal; extreme allocations on either side increase risk), (2) CAPE ratio at retirement (high valuations suggest lower forward returns), (3) expected cost-of-living inflation (retirees' inflation basket differs from CPI), (4) whether you use dynamic strategies (willing to cut spending in downturns can raise safe rate by 50-100bps), and (5) country/currency (non-US portfolios have different historical safe rates). Also: 'age' is a proxy for 'horizon' — a 65-year-old with genetic longevity markers for reaching 100 should probably use a lower rate than the table suggests. The age-based SWR is a starting point, not a final answer.

• “4% for 30-year horizons” — Bengen (1994), Trinity Study (1998)
• “~3.5% for 40-year, ~3.25% for 50-year” — Pfau (2010), Kitces various
• “Dynamic strategies raise safe rate 50-100bps” — Guyton & Klinger (2006), Journal of Financial Planning

• Bengen (1994), 'Determining Withdrawal Rates Using Historical Data', Journal of Financial Planning (accessed 2026-04-17)
• Cooley, Hubbard, Walz (1998), 'Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable', AAII Journal (accessed 2026-04-17)
• Pfau & Kitces (2013), 'Reducing Retirement Risk with a Rising Equity Glidepath', Journal of Financial Planning (accessed 2026-04-17)

Last verified: 2026-04-17