TL;DR
The Trinity Study (Cooley, Hubbard, Walz 1998) showed 4% inflation-adjusted withdrawals survived 95%+ of US 30-year periods with a 50/50–75/25 stock/bond mix. Pure bond portfolios failed badly. Pure stock portfolios were riskiest in the short term but safest over 30+ years.
What the Trinity Study actually was
In 1998 three professors at Trinity University in Texas — Philip Cooley, Carl Hubbard and Daniel Walz — published "Retirement Spending: Choosing a Sustainable Withdrawal Rate" in the AAII Journal. It's the single most-cited paper in the FIRE world, and most people who quote it have not read it.
What they did: for every overlapping period in US data 1926–1995, they ran simulations of various withdrawal rates against various stock/bond allocations and reported the success rate — the percentage of historical periods where the portfolio lasted the full horizon.
The headline tables
For a 30-year horizon with inflation-adjusted withdrawals:
| Allocation | 3% | 4% | 5% | 6% | |---|---|---|---|---| | 100% stocks | 100% | 98% | 80% | 62% | | 75/25 | 100% | 100% | 82% | 59% | | 50/50 | 100% | 95% | 76% | 51% | | 25/75 | 100% | 71% | 27% | 5% | | 100% bonds | 100% | 20% | 0% | 0% |
A few things jump out:
- All-bond portfolios are terrible for inflation-adjusted withdrawals, even at 4%. Inflation eats them alive.
- All-equity beats balanced at the 30-year horizon and most withdrawal rates. The conventional wisdom that retirees should be heavily in bonds is empirically wrong for long horizons.
- The "safe" rate is allocation-dependent. At 4%, you can go 50/50 to 100/0 and survive. At 5%, you need at least 50% equities to even have a chance.
What gets misquoted
Three common errors:
-
"4% always works" — only at the 30-year horizon and only for a 50% or higher equity allocation. At 40 years the 4% success rate drops to ~85% even for an all-equity portfolio. At 50 years it's closer to 75%.
-
"4% means you can withdraw $40k from a $1M portfolio forever" — no, it means $40k in the first year, then adjusted up with inflation each subsequent year. That's a critical detail.
-
"The Trinity Study proved the 4% rule is safe" — it showed 4% survived in the US historical record. It did not prove anything about the future, and it did not include international data, fund costs, taxes, or behaviour during drawdowns.
What it doesn't say
- Nothing about taxes — assume after-tax withdrawals.
- Nothing about fund fees — gross-of-fee data.
- No international diversification.
- No spending flexibility.
- No flagged risk of "ending with nothing" — a portfolio that dies at year 31 is "success" under the original methodology even though that's no comfort to the retiree.
Modern updates
The Trinity authors updated their paper in 2011 with data through 2009, and the headline numbers barely moved. Wade Pfau and others have re-run the analysis on international data and found lower safe rates abroad — 3.5% is closer to the global safe rate.
The single biggest improvement over Trinity is allowing withdrawal flexibility. Variable withdrawal strategies (Guyton-Klinger guardrails, VPW, etc.) lift the safe rate by 0.5–1% with modest spending volatility. See our guardrails article.
How to use it today
The Trinity Study is a good first-cut benchmark, not a planning tool. For your actual FIRE plan, you want:
- A horizon that matches your expected lifespan
- A withdrawal model that allows flexibility
- Global data, not just US
- Fund costs baked in
Our withdrawal survival tool does all of this against 155 years of data.
Frequently asked questions
- Did the Trinity Study include the 2008 crisis?
- Not in the original 1998 paper. The 2011 update extended through 2009 and the success rates barely changed — 2008 was a bad year but the recovery was fast enough that retirees who didn't panic-sell were fine.
- Why does an all-bond portfolio do so badly?
- Inflation. 4% withdrawals adjusted for inflation outrun bond yields over long horizons. Bonds preserve capital but not purchasing power, and a 30-year retirement needs purchasing power.
- Is the Trinity Study still relevant in 2026?
- Yes as a foundational reference, no as a planning tool on its own. Use it as the anchor for understanding why 4% is approximately right at 30 years, then move to modern tools for your actual horizon.
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.