When the 4% Rule Works, and When It Fails

Take $1 million. Spend $40,000 in your first year of retirement, then raise that amount each year for price increases. Will the money last 30 years?

In our test it did about 85 times in 100. The other 15 ran out of money, and most of those ran out late. That is a good result, but it is not a promise. This guide shows what we ran, what pushed the odds up or down, and how to test your own plan.

What is the 4% rule?

The 4% rule is a simple way to plan spending. You take 4% of your savings in year one. That is $40,000 on $1 million. After that, you raise the dollar amount each year to keep up with prices. You do not change it when the market moves.

The rule comes from adviser William Bengen's 1994 study and a 1998 follow-up called the Trinity Study. They looked at past US stocks and bonds. They found that about 4% was the most you could spend and still have money 30 years later, even in the worst past starts.

That was a study of the past. Your future may not look like it. So we asked a different question: how often does 4% work if the market could go many different ways?

How did we test it?

We built a free calculator and ran the same plan through 10,000 random retirements. Each one has its own string of good and bad years. The setup follows a recent Wall Street Journal column by finance professor Derek Horstmeyer. The numbers below come from our own run.

• Start: $1 million at age 65, spending $40,000 in year one.
• Prices: rise 3% a year, and your spending rises with them.
• Stocks: earn 10% a year on average, with ups and downs of 15% a year.
• Bonds: earn 4% a year on average, with ups and downs of 5% a year.
• Mix: 50% stocks at age 65, sliding in a straight line to 0% stocks at age 95.
• Timing: you spend first, then what is left earns that year's return.
• Left out: taxes, health shocks, and any change in how much you spend.

Your answer will change if you change these. The 4% rule stress test lets you do that.

How often does the money last?

In 8,507 of 10,000 retirements, there was still money at 95. That is about 85 in 100. In 1,493 the money ran out first.

The blue line is the typical retirement. It slides down, and that is normal. You are spending your savings. The shaded bands show how far apart results can land.

Here is the part most calculators leave out. The dashed line is what you get if you assume the same return every year. It never runs out of money. Real markets do not move in a straight line, and the bad years matter most.

Among the retirements that lasted, the typical one ended with about $399,000 in today's dollars. That is about $968,000 in the prices of that future year.

When the money ran out, it was almost always late. About 99.9 in 100 retirements still had money at age 85. About 97 in 100 did at age 90. Among the ones that failed, the typical retirement ran out in year 28, around age 93.

Look at how far apart the outcomes are. Some retirements end with more than $1.2 million in today's dollars. Some end with nothing. The same rule gives very different results.

Why does the order of good and bad years matter?

Two retirements can earn the same average return and end up in very different places. The reason is simple. When you are taking money out, a loss early costs more than a loss late.

Say stocks fall in your first year. You still need your $40,000. So you sell shares at a low price to get it. Those shares are gone, so they cannot recover when prices bounce back. The experts call this the order of returns. It is the biggest reason the 4% rule fails.

Both retirements above earn 9% a year for 20 years and 0% for 10 years. The only difference is the order. With flat years first, the money runs out around age 90. With good years first, there is about $831,000 left at 95, in today's dollars.

What breaks the 4% rule?

We changed one thing at a time and kept everything else the same. Every row uses the same 10,000 random markets, so the gaps come from the change, not luck.

Here is what each big change does, in plain terms.

• Fees. A 1% yearly fee cut the odds from about 85 in 100 to about 61. That is a drop of about 24 points. A fee comes out every year, even when your savings are falling. Check what you pay for advice and funds.
• Higher prices. If prices rise 6% a year instead of 3%, your spending doubles faster. Only about 14 in 100 lasted. The 4% rule was built for a world of moderate price rises.
• A weak first decade. If stocks earn only 3% a year for the first 10 years, the odds fall to about 42 in 100. That is the order-of-returns problem in action.
• Lower stock returns. If stocks earn 7% a year instead of 10%, about 61 in 100 lasted. Our 10% is roughly the long-run past average for US stocks, but nobody knows the future.
• All bonds. This sounds safe, but only about 33 in 100 lasted. Bonds at 4% barely beat 3% price increases, so your money shrinks in real terms.
• Bigger ups and downs. Raising stock swings by 5 points a year cut the odds to about 77 in 100.

One result surprised us. Keeping 50% in stocks for the whole 30 years lasted in about 91 in 100, better than sliding to zero stocks. The tradeoff is bigger swings along the way. If a 30% drop would make you sell, the safer mix may be the one you can hold.

How long does the money need to last?

The 4% rule was built for 30 years. If you retire earlier, or live longer, the odds change a lot.

At 4% spending, the money lasted in about 100 of 100 retirements over 20 years. It was 97 over 25 years, 85 over 30, 68 over 35 and only 52 over 40. If you plan to retire in your 50s, 4% is too high for most people. Try 3.5% or less, or plan to earn some income.

What spending rate gives you better odds?

We tried several spending rates on the same plan.

The most you could spend in year one and still get 90 in 100 was 3.8%, or $38,000. To get 95 in 100, it was 3.5%, or $35,000. That is close to the 3.9% that Morningstar published in 2026 for a 30-year retirement. Different methods landing in the same area is a good sign.

The lesson is that a small cut early buys a lot of safety. Going from 4% to 3.5% takes $5,000 off your year-one spending and adds about 11 points of safety.

What can you do about it?

You cannot control the market. You can control four things.

1. Spend a little less at the start. It is the strongest lever you have, and it costs the least when you do it early.
2. Keep fees low. One percent a year is a big drag. Ask for the full cost of every account and fund.
3. Be ready to cut spending after a bad year. This works, but it has a price. See below.
4. Count other income. If Social Security or a pension pays part of your bills, your savings only fill the gap.

In this test, spending 10% less when things got tight kept the money alive in all 100 retirements. But in 74 of them, spending fell below 75% of the plan at some point. The typical retiree's leanest year was about 59% of the plan. That is about $23,600 in today's dollars, not $40,000. The rule does not make the risk disappear. You pay for it with a smaller lifestyle.

Other income helps a lot. If $24,000 a year of Social Security starts in year 3, the savings only cover the rest. In our test, the money lasted in all 100 retirements. That is a large share of spending, so enter your own number.

Some people also keep a year or two of spending in cash, so they never have to sell stocks at a low. See where to keep that cash and our retirement withdrawal playbook.

What does this test leave out?

• Taxes. Withdrawals from pre-tax accounts are taxed. You may need to take out more than you spend.
• Health shocks. A big medical or care bill early could hurt more than any market.
• Changing spending. Many people spend less as they age. See how spending changes in retirement.
• The future. Returns here are assumptions, not a forecast. Try lower returns in the calculator.

Try it with your numbers

Change the savings, spending, years and stock share below. For the full version with the what-if table and the heat map, open the 4% rule stress test.