Safe Withdrawal Rate: What 150 Years of Data Shows

Where the 4 Percent Rule Came From — and What It Actually Promises

The 4 percent rule originates from William Bengen's 1994 study in the Journal of Financial Planning. Bengen analyzed rolling 30-year retirement periods from 1926 to 1992 using a 50/50 stock-bond portfolio and found that a 4% initial withdrawal, adjusted annually for inflation, never depleted a portfolio in any historical 30-year window.

The formula is simple: withdraw 4% of your portfolio in year one, then adjust that dollar amount by CPI each subsequent year. On a $1 million portfolio, that's $40,000 in year one. If inflation runs 3%, you take $41,200 in year two — regardless of what the market does.

The critical word in Bengen's finding is "never" — but it applied only to the 1926–1992 dataset. The Trinity Study (Cooley, Hubbard, and Walz, 1998) extended this analysis and found 4% succeeded in approximately 95% of 30-year periods within that dataset. Any failure rate represents real historical scenarios — specifically, retiring at the wrong time with an unlucky sequence of early losses. Neither study was a guarantee; both were historical backtests.

What 150 Years of Market Data Changes About Safe Withdrawal Rates

Extending the dataset back to 1871 — using the Shiller CAPE data — tells a more sobering story. Researcher Wade Pfau published work in 2011 showing that when international markets and pre-1926 U.S. data are included, the 4% rule fails more often than the original studies suggested. In some developed-market datasets (Japan, Germany, the UK post-WWII), a 4% withdrawal rate resulted in portfolio depletion within 20–25 years with distressing regularity.

The survival rates by withdrawal rate across all 30-year historical windows in the extended U.S. dataset (1871–2023) break down roughly as follows:

| Withdrawal Rate | 30-Year Success Rate | 40-Year Success Rate | |----------------|---------------------|---------------------| | 3.0% | 99% | 97% | | 3.3% | 95% | 90% | | 4.0% | 86% | 74% | | 4.5% | 77% | 62% | | 5.0% | 68% | 51% |

The 40-year column matters enormously. If you retire at 55 or even 60, a 30-year plan may leave you broke at 90. A 4% rate survives only 74% of 40-year historical windows — roughly the odds of a coin flip plus a little luck. For early retirees, 3.3% is the number that holds at 95% confidence across 30-year horizons and approximately 90% confidence across 40-year horizons; for 95% confidence over a 40-year horizon, the data in this article points to the 3.0% rate.

The culprit isn't average returns — it's sequence-of-returns risk. A portfolio that loses 30% in years one and two of retirement, even if it fully recovers by year ten, may never recover enough to sustain withdrawals. The math of withdrawing from a shrinking base is brutally asymmetric.

How Sequence-of-Returns Risk Destroys the Average-Return Assumption

Sequence-of-returns risk is the single most underappreciated threat to retirement income. Two retirees can experience the exact same average annual return over 30 years and end up with vastly different outcomes, simply because of when the bad years hit.

Consider two portfolios, both averaging 7% annually over 20 years. Portfolio A earns strong returns early (15%, 12%, 10%...) and weak returns late. Portfolio B earns weak returns early (-15%, -10%, -5%...) and strong returns late. After 20 years with no withdrawals, both portfolios end at the same value. But add $40,000 annual withdrawals from a $1 million starting balance, and Portfolio B can be depleted well before year 20 while Portfolio A may still have substantial assets remaining.

The formula that captures this: your portfolio's real terminal value is a function not just of CAGR, but of the specific ordering of returns multiplied against a declining asset base. This is why Monte Carlo simulations — which randomize return sequences — consistently produce lower safe withdrawal rates than simple average-return calculations. Michael Kitces and Wade Pfau's research on "retirement income scenarios" shows that the worst historical 10-year sequences (1966–1976, 1929–1939, 2000–2009) all share the same feature: large negative real returns in the first several years of retirement.

The practical implication: your first five years of retirement are disproportionately important. A 20% portfolio loss in year three of retirement is roughly three times as damaging as the same loss in year twenty.

The CAPE Ratio Adjustment: A Smarter Way to Set Your Rate

The best forward-looking predictor of safe withdrawal rates is the Shiller CAPE ratio (Cyclically Adjusted Price-to-Earnings ratio) at the time of retirement. Pfau's research demonstrates a strong negative correlation: high CAPE at retirement predicts lower safe withdrawal rates over the subsequent 30 years.

The historical relationship, roughly:

| CAPE at Retirement | Estimated Safe Withdrawal Rate | |-------------------|-------------------------------| | Below 12 (cheap market) | 5.0–6.0% | | 12–20 (fair value) | 4.0–5.0% | | 20–28 (elevated) | 3.5–4.0% | | Above 28 (expensive) | 3.0–3.5% |

As of early 2026, the Shiller CAPE sits above 35 — well into historically expensive territory. That places the historically-grounded safe withdrawal rate for someone retiring today in the 3.0–3.5% range, not 4%. On a $1 million portfolio, the difference between 4% and 3% is $10,000 per year in spending — significant, but far less catastrophic than running out of money at 82.

This doesn't mean 4% is reckless. It means the margin of safety is thinner than it was when Bengen published his research in 1994, when the CAPE was approximately 20. Investors retiring into an expensive market simply need either more assets, more flexibility in spending, or a dynamic withdrawal strategy.

Dynamic Withdrawal Strategies That Beat the Static 4 Percent Rule

The static 4% rule — same inflation-adjusted dollar amount every year regardless of portfolio performance — is the worst-case version of retirement spending. It's also the one most commonly cited. Better approaches exist.

Guardrails Method (Guyton-Klinger): Set a target withdrawal rate (say 5%). If the portfolio performs well and the effective rate drops below 4%, you give yourself a 10% raise. If the portfolio struggles and the effective rate rises above 6%, you cut spending by 10%. Research by Jonathan Guyton and William Klinger (2006) shows this method can allow initial rates as high as 5.2–5.6% while maintaining 95%+ portfolio survival rates — because you're not locked into inflation-adjusted increases when the portfolio can't support them.

Floor-and-Upside Method: Fund non-negotiable expenses (housing, healthcare, food) with guaranteed income — Social Security, annuities, bond ladders — and use your portfolio only for discretionary spending. Because the floor can't be cut, you're spending the portfolio on wants, not needs. This structurally reduces the risk of running short on essentials.

RMD-Based Method: Withdraw based on IRS Required Minimum Distribution tables, dividing your portfolio balance by your remaining life expectancy factor each year. In early retirement this produces a rate around 3–3.5%; it rises gradually as you age. It is inherently self-adjusting and can never technically deplete the portfolio to zero.

For many retirees, a hybrid approach can outperform any single rule. Run your actual numbers — not a hypothetical average — with our 4 percent rule calculator to see how each strategy performs against your specific balance, spending, and timeline.

How Much Can I Withdraw in Retirement: The Real Answer by Scenario

The right withdrawal amount depends on four variables: portfolio size, retirement age (which determines horizon length), asset allocation, and spending flexibility. Here's what the historical data suggests for specific situations:

Scenario 1 — Standard retirement at 65, $1M portfolio, 60/40 allocation, 30-year horizon: Historically safe rate: 3.8–4.0%. Annual withdrawal: $38,000–$40,000. This is close to the Bengen scenario. It works in approximately 86–95% of historical periods depending on the dataset used.

Scenario 2 — Early retirement at 50, $1.5M portfolio, 70/30 allocation, 40-year horizon: Historically safe rate: 3.0–3.3%. Annual withdrawal: $45,000 at 3.0% to $49,500 at 3.3%. The longer horizon compresses the safe rate even on a larger portfolio. A $1.5M portfolio at 3.3% generates meaningfully more dollar income than a $1M portfolio at 3.0% ($49,500 vs. $30,000).

Scenario 3 — Late retirement at 70, $800k portfolio, 50/50 allocation, 20-year horizon: Historically safe rate: 5.0–5.5%. Annual withdrawal: $40,000–$44,000. A shorter horizon can expand flexibility. RMD tables at age 70 imply roughly a 3.8% rate, rising to approximately 5.5% by age 80.

Scenario 4 — Flexible spender, $1M portfolio, willing to cut 10% in down markets: Guyton-Klinger guardrails can allow a 5.0–5.2% initial rate. Annual withdrawal: $50,000–$52,000. The flexibility premium over a rigid 4% rule is roughly $10,000–$12,000 per year in potentially sustainable income.

None of these scenarios account for Social Security, pensions, or part-time income — all of which reduce portfolio dependency and can allow higher effective withdrawal rates. A retiree with $2,000/month in Social Security and a $1M portfolio is in an entirely different position than one relying solely on the portfolio.

What to Actually Do With This Data Before You Retire

The 4 percent rule is a starting point, not a retirement plan. The data from 150 years of market history gives us a clear framework, but your situation requires specific numbers.

First, calculate your portfolio-to-spending ratio. Divide your investable assets by your annual spending (not income — spending). If that ratio is 25 or higher, you're at the 4% threshold. If your ratio is 30 or above, you're at the 3.3% level, which the historical data suggests holds at 95% confidence across 30-year horizons and approximately 90% confidence across 40-year horizons. For 95% confidence over a 40-year horizon, the data in this article points to the 3.0% rate.

Second, stress-test your plan against a bad sequence. Model what happens if your portfolio drops 35% in years one and two. If that scenario depletes your portfolio before age 85, you need either more assets, less spending, or guaranteed income to cover your floor.

Third, know your CAPE context. Retiring into a market with a CAPE above 30 historically suggests trimming your initial rate by 0.3–0.5 percentage points relative to Bengen's base findings.

Finally, build in a review trigger. Commit to reassessing your withdrawal rate if your portfolio drops 15% from its starting value in real terms. Dynamic adjustment in year two beats discovering a problem in year fifteen.

Building a full retirement income plan that accounts for your asset mix, Social Security timing, and spending profile takes about ten minutes with the right tool. Start with our free retirement plan builder to get your personalized safe withdrawal rate and a projection that accounts for your actual numbers — not someone else's average.

Frequently Asked Questions

Related Calculators

Related Scenarios

Related Articles