Safe Withdrawal Rate: What the Research Actually Says (and When to Ignore It)
The 4% rule has one job: survive the worst historical 30-year stretch the U.S. market ever produced, without ever cutting spending. That's a useful benchmark — but it's almost certainly not the right rule for you specifically.
Where the 4% rule comes from
In 1994, financial planner William Bengen asked a simple question: what is the highest withdrawal rate a retiree could have used historically, adjusted upward each year for inflation, without running out of money over any 30-year period in the historical record?1 His answer: 4% of the initial portfolio value.
The Trinity Study (1998) confirmed this finding across different asset allocations and time periods, showing that a 50/50 stock-bond portfolio at a 4% initial withdrawal rate succeeded in approximately 95% of historical 30-year periods.2
That's the whole rule. Four percent. Inflation-adjusted. 30 years. Historically safe.
What it is not: a guarantee, a universal prescription, or a number that accounts for your specific situation, current valuations, retirement length, spending flexibility, or other income sources.
How retirement length changes everything
The 4% rule was built for a 30-year retirement. A 65-year-old who lives to 95 needs it. A 55-year-old early retiree needs 40 years of coverage. A 72-year-old needs maybe 25.
The table below shows approximate historical success rates for a 50% equity / 50% bond portfolio at different withdrawal rates and planning horizons, based on U.S. historical data from 1926 onward.2
| Withdrawal rate | 20 years | 25 years | 30 years | 35 years | 40 years |
|---|---|---|---|---|---|
| 3.0% | ~100% | ~100% | ~100% | ~99% | ~98% |
| 3.5% | ~100% | ~99% | ~98% | ~97% | ~94% |
| 4.0% | ~100% | ~97% | ~95% | ~90% | ~85% |
| 4.5% | ~99% | ~94% | ~88% | ~81% | ~73% |
| 5.0% | ~96% | ~88% | ~80% | ~71% | ~62% |
Approximate historical success rates; 50/50 stock-bond portfolio, annual rebalancing, inflation-adjusted withdrawals. Based on U.S. historical data 1926–2020. Source: Cooley et al., updated analyses by Pfau and Morningstar.
The key insight: a 4% rate has an 85% historical success rate over 40 years — not 95%. If you're retiring at 55, using the same rule as someone retiring at 65 understates your risk by about 10 percentage points.
Why the 4% rule is probably both too conservative and too aggressive
Too conservative for many retirees, because:
- Most retirees naturally reduce spending in their 70s and 80s. Research shows average spending declines 1–2% per year in real terms after age 70 as travel and entertainment slow.3
- Social Security and any pension income provide inflation-adjusted guaranteed income that the model ignores — reducing the portfolio draw needed.
- The rule uses the worst historical sequence ever observed. Your expected outcome is substantially better than the worst case.
Too aggressive for some retirees, because:
- The 4% rule was built on 20th-century U.S. equity returns, which were exceptionally high by global historical standards. Lower expected returns going forward would reduce safe withdrawal rates — Morningstar's 2022 research put the starting safe rate at closer to 3.8% for a 30-year horizon given current valuations and expected bond returns.4
- A 40-year retirement has meaningfully lower historical success (85% vs 95%).
- Sequence-of-returns risk is real: a bad first five years can permanently impair a portfolio even if average returns recover. See our sequence-of-returns risk guide.
The dynamic alternative: Guyton-Klinger guardrails
In 2006, financial planners Jonathan Guyton and William Klinger published a study showing that adding simple spending rules — guardrails — allows a higher initial withdrawal rate while keeping the portfolio safe.5 The tradeoff: you accept small, rule-based spending cuts in bad markets instead of holding a fixed inflation-adjusted withdrawal.
The guardrail system has four rules. Two matter most in practice:
The withdrawal rule (skip the inflation raise in down-portfolio years when your rate already exceeds the initial rate) and portfolio management rule (skip the raise when equity drift is above initial allocation) act as additional friction, preventing overspending.
Why this matters for your withdrawal rate: Guyton and Klinger found that with these guardrails in place, initial withdrawal rates of 5.2–6.2% remained sustainable across historical periods for a 30-year retirement — substantially higher than the 4% floor. The reason: a single 10% spending cut in the worst years is enough to protect the portfolio through most historical bear markets.
The behavioral cost is real. A retiree who started at 5% and hits a capital preservation cut must temporarily drop to 4.5%. For a retiree spending $80,000/year, that means $76,000 for a year or two. Most people can manage this by cutting discretionary spending (travel, gifts, dining). The guardrail rarely triggers, but when it does, it works.
Guardrail status checker
Use this tool to see which Guyton-Klinger zone you're currently in — prosperity (can raise spending), neutral (hold steady), or guardrail hit (time to cut).
Variable Percentage Withdrawal (VPW)
A third approach, developed by the Bogleheads community, takes a different angle entirely: instead of a fixed dollar amount, you withdraw a percentage of your current portfolio each year, where that percentage increases as you age to reflect your shorter remaining time horizon.6
The math ensures you never run out of money — you always withdraw a percentage, never a fixed amount. The tradeoff is spending volatility: in a down year, both your portfolio and your VPW withdrawal shrink. This approach works best when you have a substantial guaranteed income floor (Social Security + pension + annuity) that covers basic needs, and VPW governs only the discretionary overlay.
How most fee-only advisors actually do this
The cleanest framework in practice: split spending into floor and discretionary.
- Floor: Fixed expenses — housing, healthcare, food, utilities, insurance. Fund from guaranteed income: Social Security, pension, income annuity. Aim to cover 70-80% of baseline spending from these sources.
- Discretionary: Travel, dining, gifts, home improvements. Fund from portfolio using a dynamic rule (Guyton-Klinger guardrails or VPW).
This structure matters because a 10% guardrail cut on discretionary spending — say, $15,000/year out of $80,000 total — is a 16% cut to discretionary. That's a reduced travel year, not an existential crisis. The same cut applied to total spending would feel much more severe.
The floor-and-upside model also lets you choose a higher withdrawal rate on the discretionary portion because any shortfalls there are manageable. It's the guaranteed floor that makes the flexibility psychologically tolerable.
Practical starting points by situation
| Situation | Starting withdrawal rate range | Key constraint |
|---|---|---|
| Age 55-60, early retirement (40-year horizon) | 3.0–3.5% fixed, or 4.5–5.0% with guardrails | Long horizon, no SS yet |
| Age 62-67, typical retirement (30-year horizon) | 3.5–4.0% fixed, or 5.0–5.5% with guardrails | SS timing choice still open |
| Age 68-72, late retirement (25-year horizon) | 4.0–4.5% fixed, or 5.5–6.0% with guardrails | RMDs starting or imminent |
| Age 73+, RMDs in progress (20-year horizon) | RMD amount often dictates the floor | IRMAA exposure, Roth opportunity |
These are starting points for a conversation, not final answers. The right number depends on your specific account balances (taxable vs. IRA vs. Roth), Social Security income, health, longevity estimate, whether you have a pension or annuity, and how much flexibility you genuinely have in spending.
Sources
- Bengen, W.P. (1994). "Determining Withdrawal Rates Using Historical Data." Journal of Financial Planning. Original 4% rule paper.
- Cooley, Hubbard, Walz (1998). "Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable." AAII Journal. The Trinity Study — historical success rates by WR and time horizon.
- Kitces, M. (2012). "Understanding Sequence of Return Risk." Kitces.com. Sequence of returns, spending trajectories, and dynamic rules.
- Morningstar (2022). "The State of Retirement Income: Safe Withdrawal Rates." Updated research suggesting ~3.8% starting rate for 30-year horizon given current valuations.
- Guyton, J.T. and Klinger, W.J. (2006), summarized by Kitces. "Decision Rules and Maximum Initial Withdrawal Rates." Original guardrail methodology; 5.2–6.2% initial rates with guardrails for 30-year retirements.
- Bogleheads Wiki — Variable Percentage Withdrawal. VPW methodology and spreadsheet tools.
Safe withdrawal rate research is based on historical U.S. market data and evolves as new periods are added. The 4% rule was published in 1994; updated research (Morningstar 2022, Pfau 2021) adjusts for current expected returns. The Guyton-Klinger guardrail values cited (20% upper band, 10% cut) are from the original 2006 paper. All figures verified as of April 2026.
Related reading
Get your withdrawal rate modeled properly
The right withdrawal rate depends on your account mix, Social Security timing, tax situation, and how much spending flexibility you actually have. A specialist models all of it. Free match.