What Bengen actually proved
In 1994, financial planner William Bengen published 'Determining Withdrawal Rates Using Historical Data' in the Journal of Financial Planning. He tested every rolling 30-year retirement window from 1926 to 1976 using a 50/50 stocks/intermediate-term Treasuries portfolio. His finding: a starting withdrawal rate of 4% of portfolio value, adjusted each year for inflation, had a 100% historical success rate of lasting 30 years.
The Trinity Study (Cooley, Hubbard, Walz, 1998) replicated and extended the work with different asset mixes and time horizons. With a 75/25 stocks/bonds allocation, 4% withdrawals succeeded 95–98% of the time over 30 years, depending on the period studied. The 4% rule entered mainstream financial planning and has anchored retirement math ever since.
What's held up, and what hasn't
- Held up: the core insight that stocks-heavy portfolios sustain higher long-term withdrawals than bonds-heavy portfolios.
- Held up: inflation adjustment matters more than nominal-return optimisation over 30+ years.
- Partially held: the 'safe' rate. Bengen himself published an update arguing 4.5% is safer with a wider asset allocation including small-cap and international equities.
- Aged poorly: the assumption that 30 years is the right horizon. Many 60-year-olds today are planning for 35–40 years.
- Aged poorly: the assumption that historical U.S. equity returns will repeat. Recent valuation-adjusted research (Kitces, Pfau) suggests 3.3–3.8% for retirees starting at very high valuations.
Sequence-of-returns risk, the rule's real enemy
The number-one threat to a 4% withdrawal is not lower average returns, it is bad returns in the first 5–10 years of retirement. The combination of withdrawals and a falling market depletes the portfolio so quickly that no subsequent recovery can save it. A retiree who started withdrawing 4% in 2000 (just before the dot-com crash) had a meaningfully different outcome than one who started in 2003.
The standard defence is a 'bond tent' or 'rising equity glide path': enter retirement with 30–40% bonds, then increase equity allocation back to 60–70% over the first decade. This reduces early-retirement volatility while keeping long-run growth on track. Wade Pfau's research suggests this approach raises the safe withdrawal rate by 0.3–0.5 percentage points.
Modern variants worth knowing
The Guyton-Klinger guardrails approach allows withdrawals to flex within a band: increase 10% in good years, decrease 10% in bad years. This dynamic approach raises sustainable starting withdrawals to 5–5.5%, at the cost of variable income.
The Kitces ratchet method (a 10% raise after a 50% portfolio increase) gives early-retirement spending guardrails for upside without the early downside flexibility. The bond-tent / rising-equity-glide-path method, as mentioned above, hardens the portfolio against sequence risk.
How to use the 4% rule today
- Use it as a planning starting point, multiply projected annual spending by 25 to size the portfolio target.
- Pair with at least a Monte-Carlo run from a planner or free tool (Boldin, ProjectionLab) for your specific situation.
- If you're retiring at age 60 with a 35+ year horizon, plan closer to 3.5–3.8% for the first decade.
- If you're retiring at 67+ with a 25-year horizon, 4.5–5% is supportable.
- Build in flexibility: aim to live on 90% of the safe rate in normal years so you have headroom for surprises.
What Bill Bengen actually found
William Bengen's 1994 paper used 1926–1992 U.S. market data to test withdrawal rates against rolling 30-year retirement periods. He found that a 50/50 stocks-bonds portfolio could sustain an inflation-adjusted 4% initial withdrawal across every rolling period, including retirees who began in 1929 and 1966, the two worst starts in modern history. Bengen later updated his work and now suggests 4.7% may be sustainable with a slightly higher equity allocation.
The Trinity Study (Cooley, Hubbard, Walz 1998) confirmed the result with portfolio mixes from 50/50 to 75/25, finding 4% had a 95–98% historical success rate over 30 years. Both studies define 'success' as the portfolio not running to zero, they say nothing about ending balance, which can range from near-zero to many multiples of the starting amount.
Sequence-of-returns risk, in numbers
Two retirees with identical 30-year average returns can finish at radically different balances. A retiree starting in a bad first decade (large drawdowns early) burns through capital faster than one starting in a good decade, even if the long-run average is the same. Pfau and Kitces' work shows the first 10 years of retirement returns explain ~75% of portfolio survival.
The hedge is a cash-and-bond bucket of 2–5 years of expenses, drawn down preferentially during equity bear markets so you never sell stocks low. This is the foundation of the bucket strategy and why a 100% equity retirement portfolio almost always underperforms a 60/40 even though 100% equity has higher expected return.
Variable withdrawal strategies that beat fixed 4%
Pure 4% is rigid: you withdraw the same inflation-adjusted dollar amount whether the portfolio is up 30% or down 30%. Variable rules adjust to portfolio performance and historically support starting withdrawals 5–6% higher with similar success rates.
Guyton-Klinger guardrails: increase withdrawals 10% if portfolio is up substantially, decrease 10% if down. Vanguard's Dynamic Spending: floor at 1.5% below baseline, ceiling at 5% above. Kitces' ratcheting: increase 10% only when portfolio exceeds 1.5x its starting value. All three deliver more spending in the average retirement and slightly less in the worst, which is the trade most retirees would make.
Practical implementation rules
- Use 4.0% only for a 30-year horizon retiring at 65; drop to 3.5% for early retirees with 35–45 year horizons.
- Keep at least 50% in equities, counterintuitively, more bonds reduces success rate over long horizons because of inflation drag.
- Adopt a Guyton-Klinger or Vanguard Dynamic Spending rule rather than rigid CPI adjustments, historical safe rates rise meaningfully with these guardrails.
- Hold 2–3 years of cash and bonds outside the equity portfolio to avoid selling during bear markets.
- Re-evaluate the withdrawal rate every 3–5 years based on portfolio performance, life expectancy and any unexpected expenses.
