Safe Withdrawal Rate in Retirement: A Plain-English Guide

3.9%. That's Morningstar's current estimate of a safe withdrawal rate in retirement. It's the largest share of your savings you can take out in year one, then raise each year for inflation, and still have money left in 90% of simulated futures over 30 years. On $1 million, that works out to $39,000 in the first year. The figure comes from Morningstar's The State of Retirement Income 2025 report, published in December 2025 (FA Mag coverage).

That's the short answer. Everything below is about why it might not be your answer.

The short answer: what is a safe withdrawal rate in retirement?

A safe withdrawal rate is a starting point for spending. Here's how it works:

1. Year one: take a fixed percentage of your portfolio. At 3.9%, $800,000 gives you $31,200.
2. Every year after: raise that dollar amount by inflation, whatever the market did.
3. The test: does the money last for the full retirement in nearly all of the scenarios?

"Safe" doesn't mean guaranteed. It means the rate held up in a high share of tested outcomes. Morningstar sets that bar at 90%.

One more detail matters. Morningstar's number assumes a portfolio with 30% to 50% in stocks, fixed spending, and a 30-year horizon (Boldin). Change any of those and the number changes too.

Where the 4% rule came from

The famous "4% rule" goes back to financial planner Bill Bengen's research in the 1990s. He tested withdrawal rates against historical market returns to find the highest rate that would have survived the worst periods on record. The Trinity Study in 1998 reached similar conclusions, and "4%" became shorthand for safe spending (SmartAsset).

Bengen's worst historical starting year was 1968. Someone who retired then ran into poor returns and high inflation early on, and early losses do the most damage. This is called sequence risk: the same average return can lead to very different outcomes depending on when the bad years come.

Why the numbers disagree: 3.9%, 4.7%, or 5.7%?

If you've searched this before, you've probably seen conflicting figures. They don't contradict each other. They answer different questions.

Morningstar vs. Bengen. Morningstar builds its estimate from forecasts of future returns and inflation. Bengen relies on what actually happened in the past (Kiplinger). Bengen raised his own figure from about 4.15% to 4.7% and calls it a "once-in-a-century worst-case scenario" number (Morningstar AU).

The flexible figure. In Morningstar's testing, two flexible methods tied at 5.7% as a starting rate for a 30-year retirement. "Guardrails" is one well-known approach: you cut withdrawals when your portfolio falls below a set level and raise them when it rises above one. In one guardrails example, a 5.2% starting rate pays $52,000 in year one on $1 million (Boldin).

The catch is that a higher starting rate comes with spending cuts in bad years. Before you choose a flexible strategy, ask how big those cuts could get and whether you could live on the lowest income it might produce.

The number moves every year. Morningstar's past estimates were 3.3% (2021), 3.8% (2022), 4.0% (2023), and 3.7% (2024). Morningstar suggests locking in the rate for your retirement year instead of resetting it every January (Boldin).

What "90% success" actually means

Nearly every article quotes the 90% figure. Few explain it.

Monte Carlo simulation (named after the casino) is a way of testing a plan against many possible futures instead of one. Each simulated future gets its own random mix of good and bad years for markets and inflation. Run 1,000 of them and count how many keep the money going until the end.

So a 90% success rate means:

• About 900 out of 1,000 futures still have money at the end of 30 years.
• About 100 out of 1,000 run out first.

Keep two things in mind:

• "Success" can be very thin. Ending with $1 counts as success. Many of the successful futures end with a lot more, and some end with very little.
• "Failure" usually doesn't mean $0 overnight. In most failed futures, the money runs short late in retirement, which gives you years to notice and adjust.

That's why a success rate is a probability, not a promise. It also explains why a plan built on a single average return can be misleading. Averages hide the bad sequences that cause the damage.

What moves your safe withdrawal rate

The headline figure is for a typical portfolio. Your plan has its own details. These are the factors that matter most.

How long your retirement lasts

The 3.9% figure assumes 30 years. Retiring at 62 and living to 95 takes 33. According to the Social Security Administration's Retirement Information for Medicare Beneficiaries (2026), about one out of every three 65-year-olds may live to at least 90. A plan that stops at 85 quietly bets against a third of retirees. Longer horizons call for lower starting rates, which is why early retirees planning for 40 or 50 years usually aim below the headline number.

Your stock and bond mix

Morningstar's estimate assumes 30% to 50% in stocks. Bengen's higher figure depends on broader diversification and a stock share that rises over time. A different allocation produces a different safe rate.

Market prices when you retire

Bengen has warned about high stock valuations, measured by the CAPE ratio (a price-to-earnings measure averaged over ten years): "I think if you get much above the low 20s, the risk starts to build" (Morningstar AU). High prices at the start of retirement raise the odds of a weak first decade.

Health care costs

Fidelity's 2025 Retiree Health Care Cost Estimate (July 2025) puts health care for a 65-year-old retiring this year at $172,500 over retirement. That covers Medicare premiums, deductibles, and prescriptions, and leaves out long-term care entirely. It's a real line item, not a rounding error, and it doesn't arrive in neat, even yearly amounts.

Taxes

Published withdrawal rates are almost always pre-tax. A $40,000 withdrawal from a traditional IRA isn't $40,000 you can spend. Which accounts you draw from first, and required minimum distributions (the amounts the IRS makes you withdraw from many retirement accounts starting at a certain age), both affect how far your money goes.

Other income

Social Security, a pension, or part-time work all reduce how much your portfolio has to cover. The more of your spending that comes from guaranteed income, the less any single withdrawal rate matters.

The other risk: spending too little

Running out of money is the risk everyone talks about. Underspending gets far less attention. If you treat 3.9% as a hard ceiling while your portfolio grows, you could end up giving up years of travel, help for family, or simple comfort that you could have afforded. Some coverage argues that retirees who stick rigidly to the 4% rule are "leaving thousands on the table" (AOL).

The goal isn't the lowest possible rate. It's a rate you've tested and are comfortable with, plus a plan for what you'll change if things go badly.

How to find your own number

A national average is a reasonable place to start. It can't tell you how your own plan holds up. EBRI's 2025 Retirement Confidence Survey found that only 52% of workers have even estimated the monthly income they'll need in retirement. The other 48% are guessing.

Here's a simple way to start:

1. Write down three numbers: your age, what you've saved for retirement, and what you spend each month.
2. Run them through many futures, not one. Look for a success rate, not a single projected balance.
3. Stretch the horizon. Test to 95 or beyond, not 85.
4. Break the plan on purpose. Try a bad first few years, higher health costs, or a later Social Security start, and see where the success rate falls.
5. Adjust and rerun. Small changes to monthly spending can move your odds more than you'd expect.

That's what Monte Climb's retirement calculator is built for. Enter your three numbers and it runs your savings through 1,000 simulated futures, then tells you how many stay funded. Plans run to age 95 by default, and the engine charges Medicare premiums and income-based surcharges year by year instead of using a flat guess. Everything runs in your browser, and nothing is saved until you say so. If you have a follow-up question, like "Am I on track?" or "Can I retire at 62?", you can ask William, our AI planner, who shows his work.

It's planning software, not advice. A probability, not a promise. But it's your probability, and that tells you more than a headline number.

Want to see your number? Show my answer →

Frequently asked questions

What is a safe withdrawal rate in retirement for 2026?

Morningstar's December 2025 research puts it at 3.9%. That's the highest starting withdrawal rate, raised each year for inflation, that left money remaining in 90% of simulated outcomes over a 30-year retirement, assuming a portfolio with 30% to 50% in stocks.

Is the 4% rule still valid?

It's still a reasonable rule of thumb, but estimates vary. Morningstar's forecast-based figure is 3.9%. Bill Bengen, who created the rule, now suggests 4.7% using historical data and a more diversified portfolio. Flexible spending strategies tested by Morningstar reached 5.7%. The right figure depends on your timeline, investments, and how willing you are to cut spending in bad years.

What does a 90% success rate mean in retirement planning?

It means that in about 900 out of 1,000 simulated futures, the money lasted through the whole retirement, and in about 100 it ran out first. 'Success' can include ending with very little left, so it's worth looking at the range of outcomes, not just pass or fail.

Why do Morningstar and Bill Bengen give different safe withdrawal rates?

They use different methods. Morningstar builds its estimate from forecasts of future returns and inflation, while Bengen tests against historical market data. Bengen's 4.7% also assumes a broader mix of investments, including small-cap and international stocks, with a stock share that rises over time.

Should I use a lower withdrawal rate if I retire early?

Generally, yes. Published safe withdrawal rates usually assume a 30-year retirement. A longer retirement means your money has to last more years, which typically requires a lower starting rate. Running your own numbers over a longer horizon shows how much the length of retirement changes your odds.