The Rule Everyone Cites and Almost Nobody Reads
The 4% rule is everywhere in personal finance. FIRE forums treat it like scripture. Retirement calculators have it baked in as a default. It gets quoted in Reddit threads as the final answer to “how much do I need to retire?”
Here’s the thing: most people who cite it have never actually read the research it’s based on. And the research has a very specific scope that matters a lot depending on when you plan to retire.
Let’s go through what the 4% rule actually says, where it came from, and whether you should trust it with your financial independence.
Where It Came From: Bengen (1994) and the Trinity Study (1998)
The modern 4% rule traces back to two key pieces of research.
William Bengen, 1994. Bengen was a financial planner who ran an analysis of historical US market returns going back to 1926. He looked at every rolling 30-year retirement window — what if you’d retired in 1929? 1966? 1980? — and asked: what’s the highest initial withdrawal rate that never depleted the portfolio within 30 years?
His answer: 4%. Specifically, 4% of the portfolio in year one, adjusted upward for inflation each subsequent year. He used a 50/50 stock/bond allocation, and later work showed 4% held at higher stock allocations too.
The Trinity Study, 1998. This is the paper most people reference when they say “the 4% rule.” Authored by Cooley, Hubbard, and Walz (three finance professors at Trinity University), it extended Bengen’s analysis with probability framing. Rather than asking “what always worked,” they asked: what percentage of 30-year historical windows survived at various withdrawal rates?
The result that stuck: at 4% withdrawal with a 75/25 stock/bond split, the portfolio survived roughly 95% of all historical 30-year periods. “Survived” means it didn’t hit zero. Not that you’d retire wealthy — just that you’d still have money left.
The Trinity Study was updated in 2011 and continues to inform retirement planning discussions. But it shares the same core limitation as Bengen’s original work: 30 years.
What “Safe Withdrawal Rate” Actually Means
Before we go further, let’s be precise about what a “safe” withdrawal rate means in this context.
You start with a portfolio — say, $1,000,000. You withdraw 4% in year one: $40,000. Each subsequent year, you withdraw last year’s amount adjusted for inflation. If inflation runs 3%, you withdraw $41,200 in year two, and so on.
The question the research answers: over how many historical 30-year periods did this strategy not zero out the portfolio?
What it does NOT mean:
- The portfolio will always be fine
- 4% works for 40 or 50-year retirements
- It holds when starting valuations are extremely high
- You will leave a meaningful inheritance
The “safe” is statistical safety in historical US markets over a 30-year window. That’s a specific and limited claim.
Sequence of Returns: The Hidden Risk
Understanding why the 4% rule sometimes fails requires understanding sequence of returns risk. It’s the concept that trips up most people when they first encounter it.
Imagine two retirees — both with $1,000,000 portfolios, both withdrawing $40,000 per year. Over 30 years, they both experience the same average annual return of 7%. The only difference: the order in which those returns arrive.
Retiree A gets a crash in years 1–3, then strong returns for the rest of the period.
Retiree B gets strong returns for the first several years, then a crash later.
Retiree A runs out of money. Retiree B retires comfortably wealthy.
Same average return. Completely different outcome.
Here’s why: when you’re withdrawing from a portfolio that drops 40% in the first three years of retirement, you’re selling shares at the worst possible prices. Those shares never recover for you — they’re gone. Retiree B keeps their shares through the crash because they have a buffer of gains from the early strong years.
This is why retiring into a market crash is genuinely dangerous. The person who retired at the start of 2000 — right into the dot-com bust followed immediately by the financial crisis — faced a completely different math than someone who retired in 1995 and spent five years riding gains before the crash hit.
Bengen’s research captures this: the 4% rule mostly fails at exactly the worst starting points. Retire into a prolonged bear market and 4% may not be safe.
The Case Against 4% Right Now
There are three serious critiques of using 4% as your planning number today.
1. Pfau’s research on valuations and future returns.
Wade Pfau, a retirement researcher and professor, has done extensive work showing that safe withdrawal rates are sensitive to starting market valuations. When CAPE (Cyclically Adjusted Price/Earnings ratio) is high — as it has been for much of the past decade — forward returns tend to be lower. His research suggests that when you account for current valuations, the historical safe withdrawal rate might be closer to 3–3.5% for current retirees rather than 4%.
This isn’t catastrophizing. It’s using valuation-adjusted expected returns rather than the full historical average. If the next 30 years produce below-average returns from today’s starting prices, the 4% rule’s historical success rate looks different.
2. The 30-year horizon problem.
The Trinity Study’s 95% success rate is for 30-year periods. If you retire at 65, that covers you to 95 — solid.
If you retire at 40, you need the portfolio to last 50 years. Nobody has done a rigorous analysis of 50-year periods with 4% because there isn’t enough historical data to get statistically meaningful results. What research does exist on longer horizons suggests 4% is substantially riskier over 50 years than over 30 years.
Early retirees using 4% as their target number are extrapolating beyond the research’s actual scope.
3. US-centric data.
Bengen and Trinity studied US markets. The US had an exceptional 20th century — no domestic wars, dominant reserve currency, massive productivity gains. Running the same analysis on international markets produces lower safe withdrawal rates. If you plan to live globally or just want a less US-optimistic model, 4% looks less comfortable.
The Case For 4% (or Something Close)
None of the above means 4% is broken. Here’s the other side.
Real retirees aren’t robots. The Trinity Study assumes you withdraw exactly 4% × CPI every single year regardless of what the market does. Real people don’t do that. If the market drops 40% in year two of your retirement, you’re going to spend less. You’ll cut a vacation, delay a car purchase, trim discretionary spending. That flexibility — which is entirely realistic — dramatically improves real-world outcomes versus the model’s mechanical withdrawal assumption.
Research on “guardrails” strategies (reduce spending by 10% if the portfolio drops 20%; resume normal withdrawals when it recovers) shows meaningfully higher success rates than rigid fixed withdrawals. You don’t need to be a robot.
Social Security shows up eventually. If you retire at 55 and plan on a 40-year retirement, Social Security starts arriving at some point in that window. Even a modest SS benefit significantly reduces required portfolio withdrawals for the back half of retirement. The 4% rule calculation usually ignores this, making it more conservative than necessary.
“Failure” in the models means running out at year 30. But if a 65-year-old retiree exhausts their portfolio at age 93, have they failed? They may have SS and Medicare and family resources. The terminal zero in the model is treated as total disaster; in practice it’s rarely that clean.
The portfolio usually survives with money left. The median outcome in historical simulations isn’t “barely survived.” It’s leaving behind a substantial portfolio. In most historical periods, 4% withdrawals leave more money than you started with. The failure cases are the tail — real, but a tail.
What to Actually Use for Planning
Here’s the practical breakdown by situation:
Traditional retiree, age 60–65, 30-year horizon. 4% is a reasonable starting point. It’s historically supported, you likely have Social Security reducing your portfolio dependency later, and you have the flexibility to adjust spending. Use 4% for your initial number, build in the expectation that you’ll flex down 10% in down years.
Early retiree, age 40–55, 40–50 year horizon. Use 3.5% as your planning rate. You’re beyond the scope of what the Trinity Study actually validated. The extra cushion isn’t pessimism — it’s acknowledging you’re in different territory. You also have more years of potential part-time income, which reduces sequence risk more than most people appreciate. Even modest income in the first five years of retirement dramatically reduces the risk of a bad sequence wrecking you.
Very early retiree, age 35–40, 55+ year horizon. Think hard about whether to count on pure portfolio withdrawals at all. Barista FIRE, consulting income, rental properties — blending income sources reduces your dependence on portfolio survival over 55 years. 3.5% might still be the target but treat income flexibility as a core part of the plan, not a nice-to-have.
The guardrails method as a practical improvement. Rather than a single fixed number, build a decision rule into your plan: if your portfolio value divided by annual spending drops below 20 (a 5% implied withdrawal rate), cut spending by 10%. If it rises above 33 (a 3% implied rate), you can safely spend more. This converts the static rule into an adaptive one that handles real-world market volatility much better.
The Bottom Line
The 4% rule is the best single-number framework we have for thinking about retirement spending. It’s historically validated, easy to understand, and gives you a concrete target: multiply your expected annual spending by 25.
But it’s a starting point, not a guarantee. It was calibrated to 30-year retirements using US historical data, and you’re reading this in a different market environment than 1994. The smart play is to treat 4% as your planning anchor while building in the things that actually determine whether you’re okay: spending flexibility, the ability to earn modest income early in retirement, and an honest look at your horizon.
The 2 AM FIRE calculator session that produces a number and declares victory is fine. Just remember that number was derived from a rule with assumptions, and your job is to understand those assumptions — not just cite the rule.
Your real buffer isn’t 4% vs 3.5%. It’s how willing you are to buy fewer vacation packages in a bad market year. That’s the variable nobody puts in the spreadsheet, and it matters more than the second decimal.
