Safe Withdrawal Rate: Everything You Need to Know

In 1994, a financial planner named William Bengen asked a question that no one had rigorously answered before: how much can a retiree safely withdraw from a portfolio each year without running out of money? He ran the numbers against 50 years of historical stock and bond market data, and his answer — 4% — became the single most influential number in retirement planning.

Thirty years later, the 4% rule is cited in every financial blog, quoted in every FIRE forum, and built into every retirement calculator. It's also frequently misunderstood, misapplied, and occasionally wrong — particularly for the early retirees the original research wasn't designed to serve.

Here's everything the research actually says.

Where the 4% rule came from

William Bengen's 1994 paper in the Journal of Financial Planning tested withdrawal rates against every 30-year rolling period in US market history starting in 1926. For each starting year, he asked: what percentage of the initial portfolio could you withdraw in year one (adjusted for inflation each subsequent year) without depleting the portfolio over 30 years?

His answer: a 4% initial withdrawal rate had never failed over any 30-year period in US history, including the Great Depression, the 1970s stagflation, and the 1929–1932 crash. A portfolio of 50–75% stocks and 25–50% bonds was sufficient to support inflation-adjusted withdrawals for the entire period.

In 1998, three professors at Trinity University confirmed and expanded this analysis in what became known as the Trinity Study. They tested portfolio survival rates across multiple time horizons and asset allocations, cementing 4% as the industry standard.

What “safe” actually means — the research behind the number

The Trinity Study (Cooley, Hubbard & Walz, 1998) tested withdrawal rates against every 30-year rolling period in US stock and bond market history from 1926 to 1995. For a 75% stocks / 25% bonds portfolio at a 4% withdrawal rate, the study found a 98% success rate — only 2 of more than 100 overlapping 30-year periods failed.

That is where “safe” comes from. Not a theoretical model. Not an assumption. A rate tested against every real 30-year period in modern market history, including the Great Depression and the stagflation of the 1970s.

It is worth being precise about what “success” meant in the study: the portfolio value was greater than $0 at the end of the 30-year period. Nothing more. Not that you lived comfortably. Not that you had money left over for your children. Just that you did not run out.

The study also made several assumptions that matter when you apply it to your own plan: withdrawals increase with inflation every year, the portfolio rebalances annually to the target allocation, there is no flexibility in spending, there is no Social Security income, and there are no other income sources. Every one of these assumptions makes the 4% rule more conservative than a real retiree’s actual experience is likely to be — and every one of them is a lever you can use to improve on it, which is exactly what the rest of this article covers.

What the 4% rule actually guarantees — and what it doesn't

The 4% rule guarantees one specific thing: that this withdrawal rate survived every 30-year period in US market history. That's a meaningful finding. It's also a limited one.

It does not guarantee that 4% will work in all future 30-year periods. It does not guarantee survival over 40 or 50 years. It does not apply to portfolios outside US markets. And it assumes a specific asset allocation (roughly 50–75% stocks, 25–50% bonds) that many people don't maintain.

The problem for early retirees: 4% wasn't designed for 50 years

Bengen's research was based on 30-year retirements — someone retiring at 65 and planning to age 95. A person retiring at 45 needs their portfolio to last potentially 55 years. The original data simply wasn't tested at those horizons.

Subsequent research has extended the analysis, and the findings are important:

Retirement length	Retire at (approx.)	Safe withdrawal rate	FIRE multiplier
30 years	Age 65	4.0%	25×
35 years	Age 60	3.75%	27×
40 years	Age 55	3.5%	28.5×
45 years	Age 50	3.25%	31×
50 years	Age 45	3.0%	33×
55+ years	Age 40 or younger	2.75%	36×

For someone planning to retire at 50 and live to 95, the safe withdrawal rate isn't 4% — it's closer to 3.25%. On a $60,000/year budget, that means needing $1,846,000 instead of $1,500,000. The difference is $346,000 — significant, but knowable.

The asset allocation matters as much as the rate

The 4% rule's success depends heavily on holding a substantial equity allocation. An all-bond portfolio fails at withdrawal rates above about 2%. A 100% stock portfolio succeeds more often — but suffers more volatility and sequence risk in early years. The research consistently supports 50–75% stocks as the optimal range for long-term withdrawal sustainability.

Many FIRE retirees make the mistake of getting too conservative as they approach their target date — shifting to bonds and cash in anticipation of retirement. This is counterproductive: a 30–40% equity allocation won't sustain a 4% withdrawal rate. Keep equities high throughout a long retirement.

Current challenges to the 4% rule

Several current conditions make some researchers more cautious about the 4% rule going forward:

• Elevated stock valuations (high CAPE ratios) have historically correlated with lower future returns
• Lower expected bond yields reduce the total return of a balanced portfolio
• Higher life expectancy means more retirees will face 35–40 year retirements even at 65
• Geographic concentration risk — the original research used US-only data; international diversification changes the picture

Wade Pfau, one of the leading researchers on safe withdrawal rates, has suggested that in a low-return environment, 3.0–3.5% may be more appropriate for new retirees than 4%. This is not a consensus view — others, including Bengen himself, have argued that 4.5% is defensible with proper portfolio management.

The critics and the debate

The Trinity Study is not settled science. It has been revised, challenged, and extended by researchers taking the FIRE community’s longer time horizons seriously — and their conclusions do not all agree.

William Bengen: the originator revised upward

Bengen himself did not stop at 4%. In a 2006 update, he found that adding a small-cap stock allocation to the original large-cap/bond mix supported a 4.5% withdrawal rate with similar historical success rates. His original 4% figure was, by his own account, a conservative floor built around the worst historical starting year (1966) — not the average one.

Michael Kitces: the case for confidence

Kitces has pushed back on the pessimism, pointing out that in the large majority of historical 30-year scenarios, a 4% withdrawal rate did not just survive — the portfolio grew substantially, often ending with more real purchasing power than it started with. His framing: 4% is the floor for bad-luck scenarios, not the expected outcome for most retirees.

Karsten Jeske (Big ERN): the most rigorous FIRE-specific analysis

Jeske’s Safe Withdrawal Rate series, published across more than 100 parts at earlyretirementnow.com, is the most detailed publicly available analysis of withdrawal rates specifically for retirements longer than 30 years. His conclusions are broadly consistent with the table above: roughly 3.5% for 40-year retirements, 3.25% for 50-year retirements — and unlike the original Trinity Study, his work explicitly addresses the sequence-of-returns risk that matters most for people retiring in their 40s and 50s.

The honest summary: for a 30-year retirement, 4% is well-supported by decades of research. For a 40–50 year retirement, 3.25–3.5% is the working consensus among the researchers who have actually studied that horizon. For maximum flexibility, a dynamic strategy that adjusts to market performance — covered next — beats any fixed rate.

Alternatives to a fixed safe withdrawal rate

The 4% rule is a fixed, rigid rule: you withdraw the same inflation-adjusted dollar amount every year regardless of portfolio performance. This rigidity is both its strength (simple, predictable) and its weakness (no adjustment when markets decline).

Several dynamic alternatives have emerged that improve portfolio survival by adjusting spending based on market conditions:

• The Guardrails Strategy: Spend more when your portfolio grows, spend less when it drops — within defined upper and lower limits.
• Dynamic Withdrawal: Adjust spending each year based on portfolio performance, rather than a fixed inflation-adjusted amount.
• Variable Percentage Withdrawal (VPW): Withdraw a percentage of your remaining portfolio rather than a fixed dollar amount. Spending naturally decreases when the portfolio shrinks.
• The Bucket Strategy: Separate assets by time horizon. Use cash for near-term expenses, bonds for medium-term, stocks for long-term. This behaviorally prevents panic-selling during downturns.

Dynamic withdrawal strategies, in depth

Every rate discussed so far assumes a fixed, inflation-adjusted dollar withdrawal every year regardless of what the market does. Real retirees do not have to accept that constraint — and the research shows that spending flexibility is worth more than almost any other lever available.

The guardrails approach (Guyton & Klinger, 2006)

Set an initial withdrawal rate higher than the fixed-rate baseline — often 5–5.5%. Define an upper and lower guardrail around your target withdrawal percentage. If the portfolio drops enough that your current withdrawal rate breaches the lower guardrail, cut spending by a fixed percentage (commonly 10%). If the portfolio grows enough that your withdrawal rate falls below the upper guardrail, you are allowed a spending increase. The mechanism works because you cut back exactly when it matters most — during the early bad years that do the most damage to long-term portfolio survival — which is what allows a materially higher starting rate with a similar overall success rate.

The floor-and-upside approach

Split your spending into two categories: essential (housing, food, healthcare, insurance) and discretionary (travel, entertainment, upgrades). Cover the essential floor with guaranteed income — Social Security, a pension, or an annuity — and let the portfolio fund only the discretionary upside. Because your survival does not depend on the portfolio at all under this structure, sequence-of-returns risk on your essential spending effectively disappears — you simply spend less on travel in a bad year.

The cash-buffer approach

Hold two years of spending in cash, separate from the invested portfolio. In a down market, draw from the cash buffer instead of selling depressed assets; in an up market, draw from the portfolio and refill the buffer. This is the simplest dynamic approach to implement and one of the most effective ways to blunt sequence-of-returns risk in the specific early years when it does the most damage.

Variable percentage withdrawal (VPW)

Instead of withdrawing a fixed dollar amount, withdraw a fixed percentage of the portfolio’s current value every year. In a down year, the dollar amount withdrawn falls along with the portfolio; in an up year, it rises. This is mathematically the most robust approach against depletion — a portfolio withdrawing a fixed percentage of itself can never technically reach zero — but it requires genuine comfort with a fluctuating income, which is a real behavioral constraint for many retirees, not just a math problem.

Which approach is right for you depends on your flexibility and risk tolerance, not a universal answer. MyFIRE uses the fixed, inflation-adjusted approach as the baseline in its Monte Carlo simulation — the most conservative and best-tested method — specifically so your plan’s success rate reflects worst-case discipline, not an optimistic assumption that you will always cut spending exactly when needed.

How Social Security changes the calculation

Every withdrawal rate discussed above assumes your portfolio is your only source of income in retirement. For most people, it will not be — Social Security exists, and modeling it correctly can be the difference between a plan that looks fragile and one that is actually solid.

Consider a retiree spending $72,000 a year who will receive $24,000 a year in Social Security starting at 67. Once benefits begin, the portfolio only needs to cover the remaining $48,000. On a $1,800,000 portfolio, that is the difference between a 4% effective withdrawal rate before Social Security and a 2.67% effective withdrawal rate after it begins — and Monte Carlo success rates improve accordingly, often dramatically.

For early retirees, the math is genuinely two-phase: from your retirement age until your Social Security claiming age, the full spending amount comes from the portfolio; after that, the portfolio only needs to cover the gap between total spending and the benefit. A plan that shows a fragile 65% Monte Carlo success rate under a naive, portfolio-only calculation can look meaningfully stronger — 85% or better — once Social Security is modeled correctly as a second income phase rather than ignored.

This two-phase structure also changes what a defensible initial withdrawal rate looks like. If you have a substantial Social Security benefit arriving at 67 or 70, you may be able to sustain a higher initial withdrawal rate in the years before it starts — 4%, or even a bit more — specifically because your effective rate in the second phase of retirement drops well below whatever you used in the first. MyFIRE models this explicitly: enter your expected benefit and claiming age in the Inputs tab, and the withdrawal engine reduces the required portfolio draw starting at that age, rather than treating Social Security as an afterthought.

The claiming-age decision itself is worth getting right independent of withdrawal-rate math. Delaying Social Security from 62 to 70 increases the eventual benefit by roughly 8% per year of delay. For most early retirees with a portfolio large enough to bridge the gap, delaying to 70 maximizes both lifetime Social Security income and overall portfolio survival probability — though for someone with health concerns or a strong preference for certainty over optimization, claiming earlier is a reasonable, defensible choice too.

The right safe withdrawal rate for you

There is no single "right" safe withdrawal rate that applies to everyone. The answer depends on your retirement age, your portfolio allocation, your flexibility to adjust spending, whether you have income sources outside your portfolio (Social Security, part-time work, rental income), and your personal risk tolerance.

A practical framework:

• Retiring at 65+: 4.0% is well-supported by historical research
• Retiring at 55–64: 3.5% is more appropriate for the longer horizon
• Retiring at 45–54: 3.0–3.25%; plan for Medicare healthcare and consider a flexible income backup
• Retiring before 45: 2.75–3.0%; sequence risk management is critical

Whatever rate you use, build in a buffer. The research defines "success" as not running out of money by day one of year N. Most people want more than technical solvency — they want to leave something, maintain lifestyle, and handle surprises. Adding 10–15% to your target number above the calculated minimum is reasonable risk management.

Calculating your personal safe withdrawal rate

There is no single correct safe withdrawal rate. There is only your rate — a function of your specific retirement length, income sources, spending flexibility, and risk tolerance.

Several factors push your safe rate higher than the generic table above would suggest:

• Genuine spending flexibility — a willingness to actually cut discretionary spending in a down market, not just a theoretical willingness
• Social Security, a pension, rental income, or other income arriving later in retirement
• The ability or willingness to pick up part-time or consulting work if a plan comes under stress
• A shorter retirement horizon than you might assume — a semi-retirement or coast plan changes the math
• A higher equity allocation, which historically has provided more upside in the majority of scenarios

Other factors push it lower:

• A genuinely long horizon — 40, 45, or 50+ years
• No income source other than the portfolio itself
• Fixed, largely non-discretionary essential spending
• A spending level that is already a high percentage of the portfolio
• Uninsured healthcare costs before Medicare eligibility at 65
• Little genuine ability or willingness to reduce spending if markets turn against you

Rather than picking a rate off a generic table and hoping it fits, the more useful approach is to test your specific numbers against real historical data. MyFIRE runs 1,000 Monte Carlo simulations using actual historical S&P 500 return sequences — not a smoothed average — against your age, savings, spending, and Social Security estimate, and shows the percentage of those 1,000 scenarios in which your plan actually survives. You can adjust the safe withdrawal rate directly in the Inputs tab, under Return assumptions, and watch your success rate move in response — 3.5% versus 4% versus 4.5%, on your numbers, not someone else’s.

That is more useful than any rule of thumb, generic table, or research citation in this article — because it is your plan, stress-tested against real history, rather than a guideline built for someone else’s retirement.

Frequently asked questions

References

• Bengen, W. (1994) — original safe withdrawal rate research, Journal of Financial Planning
• Cooley, Hubbard & Walz (1998) — the Trinity Study
• Pfau, W. (2013) — critique and update on withdrawal rates in low-yield environments
• Guyton & Klinger (2006) — the guardrails withdrawal strategy
• Jeske, K. — Safe Withdrawal Rate series, earlyretirementnow.com
• Damodaran / NYU Stern School of Business — historical market return data