What is the 4% rule, and is it still right in 2026?

The 4% rule is the most famous number in retirement planning, and one of the most widely misunderstood. Most articles either:

1. Repeat it as gospel ("just withdraw 4% and you're fine"), or
2. Dismiss it as broken ("4% is dead — try 3%"), or
3. Use it as a marketing hook for a paid product.

The honest version is more interesting and more useful: the 4% rule is a specific, well-documented historical finding with specific assumptions, and it's mostly held up — but the assumptions matter.

What Bengen actually said

In 1994, financial planner William Bengen published a paper testing every 30-year retirement window in US history from 1926 to 1976. He asked: what's the highest initial withdrawal rate, indexed annually for inflation, that survived every single window without depleting the portfolio?

The answer was 4%. With a 50/50 stocks/bonds portfolio, a retiree could have withdrawn 4% of their starting balance in year 1, then increased that dollar amount by inflation each year, and not run out of money in any 30-year window in the historical record.

The Trinity Study (Cooley, Hubbard, Walz, 1998) replicated the analysis with stocks/bonds blends from 75/25 to 25/75 and confirmed the finding. The 4% rule was born.

What the rule does NOT say

Five things the 4% rule is regularly mischaracterized as saying:

1. It does NOT guarantee 4% works in your specific case. It says 4% has worked historically in every 30-year window. The next 30 years could include a regime worse than anything in the historical record — there's no theorem proving that won't happen.
2. It does NOT mean "withdraw 4% of your balance each year." It means "4% of your year-1 balance, then increase by inflation." If your portfolio drops 30% in year 5, you keep withdrawing the inflation-adjusted year-1 amount, not 4% of the new (smaller) balance.
3. It does NOT account for taxes — or IRMAA. The 4% is gross withdrawal. Federal and state taxes come out of that; the 2026 federal brackets and standard deduction are broken down here. Withdrawals that push MAGI past a Medicare IRMAA threshold also add $1,148 to $1,736 per person per year in extra Medicare premiums two years later — see the IRMAA cliff for the 2026 bracket table and the two-year-lookback mechanic.
4. It does NOT model Social Security. The rule is portfolio-only. If Social Security covers $30K of your annual spending, you only need the portfolio to cover the rest.
5. It does NOT extend to 40+ year retirements. The 30-year window was the test horizon. For longer retirements, the safe rate drops.

The 2026 critiques

Three serious critiques have emerged in the 30 years since Bengen:

Critique 1: The historical sample includes some uniquely good periods

Bengen's data ran 1926-1976. That period included a unique 50-year stretch where US stocks delivered ~10% nominal returns with a strong dollar, growing labor productivity, and demographic tailwinds. Whether the next 50 years will be similar is an open question.

Researchers like Wade Pfau and Big ERN have stress-tested the rule against international data and against expected lower returns. The conclusion most academic researchers land on: 4% is probably still safe for a 30-year horizon, but the safe range for cautious planners has shifted to roughly 3.3-3.7%.

Critique 2: Sequence-of-returns risk dominates the answer

The reason the rule needed to be 4% (and not 5%) was the worst historical sequence: a retiree who started in 1966 lived through stagflation that nearly broke the plan. Bad sequence-of-returns in the first decade of retirement is what kills most plans, regardless of long-term average return.

If you happen to retire into a strong first decade (like 2010-2020), you can sustain a much higher withdrawal rate. If you retire into a 1966-style decade, even 3.5% can fail. You can't know which you'll get.

The implication: the 4% number is the historical worst-case answer, not the typical answer. In most historical 30-year windows, retirees could have withdrawn 5-6%+ and been fine. They just didn't know which window they were in.

Critique 3: It assumes you don't adjust spending

The rule is mechanical: withdraw the inflation-adjusted year-1 amount every year, regardless of portfolio performance. Real retirees don't do this. If the portfolio drops 40%, most retirees spontaneously cut discretionary spending — which materially improves the success rate.

Variable withdrawal strategies (taking 4% of current balance, or using the Guyton-Klinger guardrails) typically allow higher initial withdrawal rates because they self-correct. The trade-off is income volatility: you spend less in down years.

Is 4% still right in 2026?

The pragmatic answer:

• For a 30-year retirement (retire at 65, plan to 95): Yes, 4% remains defensible for a balanced portfolio. The historical data support it; the forward-looking critiques argue for a slightly more conservative range (3.5-4%) but don't break the rule.
• For a 40-year retirement (retire at 55): drop to 3.0-3.5% initial withdrawal. Bengen himself (2020 update) recommends about 3.3% for early retirees.
• For a 50-year retirement (FIRE retirement at 45): drop to 2.8-3.0%. The math compounds: more years to fund, more years exposed to a bad sequence.

These are starting points, not laws. The Yearfold calculator runs your specific spending against 10,000 simulated futures and reports the actual success probability — which is the number you should plan against, not a rule of thumb.

What about variable strategies?

Two variable strategies that empirically beat the static 4% rule on sustainability:

1. Variable percentage withdrawal (VPW). Withdraw a fixed percentage of your current balance each year (typically 4-5%). Income tracks the market — higher when stocks are up, lower when they're down. The trade-off is income smoothness.
2. Guyton-Klinger guardrails. Start at a higher initial withdrawal rate (say 5%), then ratchet up by inflation in normal years and ratchet down if the portfolio drops below a threshold. Adds 0.5-1.0 percentage points of safe initial withdrawal vs static 4%.

Both are available as withdrawal strategies in the Yearfold calculator's Withdrawal tab. Try them against your numbers.

The bottom line

The 4% rule is approximately right for a 30-year retirement with a balanced portfolio. It's an academic floor, not a target. Real retirees should plan for 3.5-4% as their default and budget the discretionary buffer that lets them adapt to whichever sequence of returns history hands them.

And remember the rule only sizes the target — which is often smaller than the headlines suggest once Social Security and taxes are counted properly. Before you assume you're behind, find out: do you already have enough to retire?

The Monte Carlo answer is always more useful than the rule of thumb — because "my plan succeeds in 92% of historical futures" tells you something the rule doesn't: how robust your plan actually is.

---

Related reading:

• How much do I really need to retire?
• Monte Carlo simulation for retirement, explained in 600 words
• Methodology — the data sources, paired sampling, and known limitations