There is a quiet category of harm in personal finance that's hard to name and hard to undo: optimistic withdrawal-rate assumptions adopted at 45 that compound into resource exhaustion at 80. We've watched the early-retirement community grow over the last decade with growing concern about exactly this pattern. Tens of thousands of careful, math-literate savers have built their FIRE targets around 25× annual spending without realizing that the underlying analysis was sized for a different retirement.
This piece is a synthesis, not a polemic. We pulled the primary literature — Bengen's original 1994 paper, the Trinity study, Wade Pfau's two decades of withdrawal-rate research, Michael Kitces' dynamic-spending work, and Bengen's own 2023 update — and reconciled their findings into what we call the Length-Adjusted Withdrawal Framework (LAWF): a retirement-horizon-by-confidence-level matrix that says what the actual research supports. The headline finding is that 4% is a defensible rate for traditional 30-year retirement under historically normal valuations. For 40-50 year horizons under current valuations, the safe rate sits materially lower, and the popular 25× benchmark is producing FIRE numbers that fail more often than savers realize.
What Bengen Actually Wrote
The original 1994 paper, "Determining Withdrawal Rates Using Historical Data" in the Journal of Financial Planning, was a back-test. Bengen took US stock and bond returns from 1926 onward, simulated a 30-year retirement starting in each year, and asked: what's the highest constant inflation-adjusted withdrawal rate that would have survived even the worst historical 30-year period?
His answer was approximately 4.15%, which he rounded down to 4% for the rule. The worst historical case was retiring in 1966 — high inflation, poor returns, the dollar collapsing. Even that case survived 30 years at 4%, just barely.
The framework had three specific assumptions:
- 30-year retirement horizon (traditional retirement at 65, death by 95).
- 50/50 stock-bond portfolio rebalanced annually (Bengen's preferred mix; later widely interpreted as 60/40).
- US-only data from 1926-1992. Returns specifically reflect the US 20th-century experience.
None of those three assumptions cleanly applies to a typical early retiree in 2026.
The Trinity Study: Same Findings, Different Method
In 1998, three Trinity University professors (Cooley, Hubbard, Walz) replicated Bengen's analysis with slightly different methodology and arrived at substantially the same conclusion. The "Trinity study" is what most personal finance writers cite — often without realizing it's essentially Bengen's analysis re-run with adjusted data ranges. Their findings, which are widely quoted but rarely quoted in full:
- 4% withdrawal for 30 years: 95-98% historical success rate.
- 4% withdrawal for 35 years: 85-90% success rate.
- 4% withdrawal for 40 years: 70-80% success rate.
- 4% withdrawal for 50 years: 60-70% success rate.
That last row is the one that should arrest the reader. For a 50-year retirement, the 4% rule has roughly a 1-in-3 historical chance of leaving you broke before you die. That is "the 4% rule" applied to a typical FIRE-community retirement horizon, using the same dataset that produced the original finding. The community didn't reject the data — it never noticed it, because the headline statistic ("4% works") was so much more shareable than the success-rate-by-horizon table that produced it.
Why Early Retirement Breaks the Rule
Three distinct problems compound when you extend retirement from 30 to 40-50 years:
Problem 1: Sequence of Returns Risk
The classic finding of withdrawal-rate research is that returns in the first 5-10 years of retirement matter enormously more than returns later. If your portfolio crashes 40% in year 2 and you keep withdrawing 4% inflation-adjusted, you can lock in permanent damage to your portfolio that no subsequent bull market can repair. The longer the retirement, the more years available for a bad sequence to occur near the beginning.
Wade Pfau's research (extensive papers across 2010-2024) shows that the worst 10-year cumulative return matters more than the average return across the retirement period. A 40-year retirement has roughly 30 different 10-year windows that could contain a bad sequence at the start; a 30-year retirement has only 20. The chance of catching a bad early sequence rises with retirement length.
Problem 2: Inflation Compounding
The 4% rule assumes you start with 4% of the portfolio and increase the dollar amount by inflation each year. Over 30 years at 3% inflation, your withdrawal nearly triples in nominal terms. Over 50 years, it more than quadruples. The portfolio has to generate enormous returns to keep up.
The 1966 retiree — the historical worst case — was destroyed by inflation, not by a stock crash. From 1966-1981, US inflation averaged 7.4% annually. The retiree had to roughly triple withdrawals in nominal terms over 15 years while the real value of stock and bond portfolios stagnated. It took the post-1982 bull market to save the simulation.
Problem 3: Bond Returns Have Fundamentally Changed
Bengen's 50/50 portfolio worked partly because US bonds yielded 4-8% real returns through most of the historical period. From 2010 to early 2024, real bond yields were near zero or negative — bonds were not paying their share of portfolio returns. The 2022-2024 bond market drawdown was the worst in 100+ years and broke the diversification that 60/40 had relied on.
Even in the current 2026 rate environment with 10-year Treasury yields around 4-5%, the bond contribution is weaker than the historical 1926-1992 average. A bond-heavy retirement portfolio in 2026 is meaningfully different from the same allocation in 1990 or 1995.
What the Research Actually Recommends for Longer Retirements
Wade Pfau's Updated Research
Pfau has published extensively on safe withdrawal rates accounting for current valuations and lower expected returns. His core findings as of recent papers:
- For a 30-year retirement starting today (high valuations, moderate yields): safe rate roughly 3.0-3.5%.
- For a 40-year retirement: safe rate roughly 2.8-3.2%.
- For a 50-year retirement: safe rate roughly 2.5-3.0%.
These are the rates that historically had 90-95% success in similar valuation/yield environments to today.
Michael Kitces's "Ratcheting" and Dynamic Approaches
Kitces has argued that the fixed-withdrawal framework misses an important behavioral truth: retirees actually adjust spending in response to portfolio performance. His "ratcheting" approach starts at a higher initial rate but reduces withdrawals if portfolio performance is poor. In simulations, this allows initial rates of 4.5-5% to succeed at similar safety levels — but only if you actually cut spending when needed.
The hidden requirement: you have to be willing and able to cut. A retiree whose lifestyle is tightly bound to their initial withdrawal rate (mortgage payment, children's tuition, etc.) cannot ratchet down meaningfully. Kitces's framework works best for retirees with significant discretionary spending that can flex.
Bengen Himself: 4.7% as the Modern Update
In a 2023 update, Bengen revised his own analysis using data through 2020 and concluded that 4.7% might be supportable as the new "safe rate" given updated market history including the bull markets of 1982-1999 and 2009-2019. This number gets cited by FIRE optimists. Three caveats most citations miss:
- Bengen's 4.7% is for traditional 30-year retirement, not 50-year FIRE retirements.
- The analysis still uses historical data — it can't account for whether future market behavior will resemble the historical distribution.
- Bengen himself notes that current valuations (CAPE ratio above historical norms) suggest using a lower-than-historical starting rate.
The Length-Adjusted Withdrawal Framework (LAWF)
Synthesizing across Pfau, Kitces, Trinity, and Bengen's update produces the following matrix. Read the rows as retirement horizons and the columns as confidence levels — the cell is the starting withdrawal rate that historical data supports for that combination. These rates assume a 60/40 stock-bond portfolio with annual rebalancing, US data, and the assumption that retirees do not dynamically adjust withdrawals during downturns:
| Retirement length | Aggressive (90% success) | Standard (95% success) | Conservative (98% success) |
|---|---|---|---|
| 30 years (traditional) | 4.0% | 3.5% | 3.2% |
| 35 years | 3.6% | 3.3% | 3.0% |
| 40 years (early FIRE) | 3.3% | 3.0% | 2.8% |
| 50 years (very early FIRE) | 3.0% | 2.8% | 2.5% |
| 60+ years | 2.8% | 2.5% | 2.3% |
Compare this to the popular "use 4% rule, target 25x" advice. A 40-year-old planning to retire at 40 (50-year retirement horizon) should target 30-40x annual spending if they want a high-confidence retirement, not 25x. That's a 20-60% larger nest egg requirement.
Translating to FIRE Numbers
For an early retiree planning $60,000/year spending:
| Approach | Required portfolio |
|---|---|
| Popular 4% rule | $1,500,000 |
| Standard 50-year framework (2.8%) | $2,143,000 |
| Conservative 50-year (2.5%) | $2,400,000 |
The conservative target is 60% higher than the popular FIRE math. That's an extra 5-10 years of working for most savers — meaningful real-world implications.
Counter-Arguments and Where 4% Might Still Work
The 4% rule isn't wrong in all contexts. Three scenarios where it remains reasonable:
1. Hard Spending Floors With Lots of Flex
If $40,000 of your $60,000 spending is truly essential (housing, food, healthcare, insurance) and $20,000 is discretionary (travel, dining out, hobbies), you can sustainably start at 4% because you'll cut the discretionary $20K in market downturns. The 4% with optional 33% spending cut in bad years has historically succeeded at near-100% rates for 50-year retirements.
2. Additional Income Sources
FIRE retirees with side income, part-time work, rental properties, or other cash flow that covers a portion of spending have an entirely different math. If you only need the portfolio to cover $30K of $60K spending, your withdrawal rate against the portfolio is half the headline rate. 4% withdrawal rate on a portfolio funding only half of expenses is effectively a 2% safe withdrawal rate against the full spending need.
3. Social Security and Pension Backstops
If Social Security at full retirement age will cover $30K of your $60K spending, your portfolio needs to bridge the gap until Social Security starts and then supplement at a lower rate. This essentially shortens the high-risk portfolio dependency from 50 years to 25-30 years, restoring 4% as a reasonable rate.
Practical Implications for FIRE Planners
If you've been using 25x annual spending as your FIRE target, three actions worth considering:
- Stress-test your number against 30x and 33x. What if you needed to work 3-5 more years to hit a more conservative target? Is that acceptable, or does retirement-at-X feel locked-in?
- Be honest about spending flex. What percent of your annual spending is truly mandatory? If you're 80%+ mandatory, you need a higher target. If you're 50%+ discretionary, 4% with flex works.
- Factor Social Security explicitly. If you'll claim SS at 62 or 67, your portfolio dependency timeline is significantly shorter than your retirement horizon. Don't treat SS as bonus; treat it as a known income stream that shortens portfolio dependence.
The "Flexible 4%" Hybrid
For most early retirees, the practical synthesis is:
- Plan around 3.0-3.5% withdrawal at retirement start.
- Allow yourself to spend more if early returns are strong (don't lock in low spending if the market gives you room).
- Have a written spending cut plan for portfolio drops >25% — what specifically gets cut, in what order.
- Re-baseline every 5 years using the remaining horizon and updated portfolio value (Guyton-Klinger guardrails are one published approach for this).
The Bottom Line, Without Hedging
The 4% rule is not exactly wrong. It was built honestly for 30-year traditional retirement under specific historical conditions, and within that scope its conclusions hold up. The early retirement community then extended the framework to 40-50+ year horizons with current high valuations, and the math doesn't follow — the same data that produces "95% success at 30 years" produces uncomfortable failure rates in the longer-horizon, present-valuation scenarios that actually describe modern FIRE.
The honest answer for a 2026 early retiree planning 40-50 years of post-employment life is that a 2.8-3.3% starting withdrawal rate is what the research actually defends. The operational implication: your FIRE number should sit closer to 30-40× annual spending than 25×. That's a meaningful difference — typically 3-7 more years of working, depending on savings rate.
The objection most people raise at this point is fair: "Five more years of work to avoid a 10% probability of failure at age 85 seems like a bad trade." It would be — except the failure mode in question is not abstract. Running out of money at 85 after using optimistic withdrawal math at 45 is among the worst outcomes in personal finance. It is unrecoverable, it lands on a vulnerable person with no labor market access, and it is functionally invisible until it isn't. The extra working years are insurance against that specific scenario. Whether you want to buy that insurance is an individual decision; pretending the risk doesn't exist isn't.
Plan Around the Numbers, Not the Slogan
Two practical next steps if you've been using 25× as your FIRE target:
- Stress-test against 30-33×. What does the additional savings horizon do to your life plan? Is the gap acceptable, or does it require revisiting retirement age? Be honest before you talk yourself into the lower number.
- Quantify your spending flex. The 4% rule with dynamic spending cuts (Guyton-Klinger guardrails, Kitces ratcheting) historically succeeds at higher rates. But the strategy requires that you can actually cut. If your spending is 80%+ committed to fixed costs, you can't ratchet; if it's 50%+ discretionary, you have flexibility to deploy.
Tools below if you want to run the numbers yourself:
- FIRE Number Calculator — adjusts the SWR input directly so you can compare your number at 4% vs 3.3% vs 2.8%.
- Retirement Calculator — model years to FIRE under various savings rates and return assumptions.
- Compound Interest Calculator — what consistent monthly savings produce over decades, the underlying mechanism that makes FIRE possible in the first place.
- FIRE Planning Concierge — guided 5-step planning sequence covering spending, SWR selection, healthcare gap, and time-to-retirement.
Sources
- Bengen, W. (1994). "Determining Withdrawal Rates Using Historical Data." Journal of Financial Planning, 7(4), 171-180.
- Cooley, P., Hubbard, C., Walz, D. (1998). "Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable." AAII Journal. The Trinity Study.
- Pfau, W. (2011-2024). Multiple papers in the Journal of Financial Planning and Retirement Researcher. Body of work on safe withdrawal rates accounting for valuations and current rate environment.
- Kitces, M. "The Ratcheting Safe Withdrawal Rate" series, Nerd's Eye View. Dynamic withdrawal approaches.
- Guyton, J., Klinger, W. (2006). "Decision Rules and Maximum Initial Withdrawal Rates." Journal of Financial Planning. Guardrail framework for dynamic withdrawals.
- Bengen, W. (2023). Updated analysis discussions, various interviews and articles. The "4.7%" update.
- Shiller, R. Long-term US market data (CAPE ratio, real returns).
shillerdata.com - Trinity Portfolio (2020). Updated Trinity-style analysis with data through 2020.
The withdrawal rate framework presented in this article synthesizes findings across these sources. Specific percentages are intended as practical guidance for individual planning, not as fixed forecasts.
Disagree with the framework? Found a research source we missed? Email hello.goledigitalstudio@gmail.com. Substantive critiques are credited and incorporated.
This article is educational and not financial advice. FIRE planning is highly individual; work with a fee-only fiduciary financial planner for personalized analysis.