Almost every retirement calculation starts from the same shortcut, usually without saying so. It is called the 4% rule, and it is genuinely useful as long as you understand what it claims and what it does not.

The rule, stated properly #

In the first year of retirement, withdraw 4% of your portfolio. Every year after that, withdraw the same dollar amount adjusted for inflation.

Notice what that is not. It is not “withdraw 4% of the current balance each year,” which is a different and safer strategy. The rule fixes a real dollar amount in year one and then keeps paying it out regardless of what the market does.

Flipped around, the version everyone quotes: you need about 25 times your annual spending.

The input is spending, not income. What you earn is irrelevant to this calculation. Two people earning $200,000 where one spends $60,000 and the other spends $180,000 need completely different numbers, and the frugal one needs a third as much and gets there roughly a decade sooner.

That is the most important idea in this entire article. Your savings rate does double duty: it raises the amount you accumulate and it lowers the target you have to reach.

Where the number came from #

The 4% figure originates with a 1994 paper by financial advisor William Bengen, later reinforced by what became known as the Trinity study. Both looked at historical US market returns and asked what withdrawal rate would have survived every rolling 30-year period on record, including retirements starting right before the 1929 crash and the 1970s stagflation.

Four percent survived essentially all of them. That is the claim. It is an empirical statement about US history, not a law of nature.

Worth knowing: Bengen himself later revised the number upward, to something in the neighborhood of 4.5% to 4.7%, when he included small-cap stocks and additional asset classes in the portfolio. The person who invented the 4% rule does not think the answer is exactly 4%.

The three assumptions that break it #

A 30-year retirement. The studies tested 30 years. Someone retiring at 65 has a reasonable chance of needing that money for 30 years. Someone retiring at 40 needs it for 50, and the failure rate at 4% rises meaningfully over longer horizons. For early retirement, most careful analysis lands closer to 3.25% to 3.5%, which means 29 to 31 times spending rather than 25.

US historical returns. The data is one country over one century, and it happens to be the century where that country did unusually well. Applying US historical returns to the future, or to any other country, is an assumption, not a finding. Studies using international data produce lower safe withdrawal rates.

That you behave mechanically. The rule assumes you keep withdrawing the same inflation-adjusted amount even as your portfolio drops 40%. No actual person does this. Which is fine, because flexible spending is precisely what makes the rule safer in practice than on paper.

Sequence of returns risk, which nobody warns you about #

This is the thing that actually sinks retirements, and it is not obvious.

Two retirees, same average return over 30 years. One gets bad years at the start, one gets them at the end. The one with bad years at the start can run out of money while the other one dies rich.

The reason: when you sell shares to fund spending during a downturn, you sell more shares to get the same dollars. Those shares are permanently gone and never participate in the recovery. Early losses compound into a permanently smaller portfolio in a way late losses do not.

The practical defenses are straightforward. Hold one to three years of expenses in cash or short-term bonds so you are not forced to sell equities into a crash. Be willing to cut spending in a bad year, since trimming ten percent for two years after a downturn dramatically improves survival rates in most simulations, and that one flexibility is worth more than almost any portfolio tweak. Keep some earning capacity, because part-time work in the first few years of retirement, if things go badly, is enormously protective, and it happens to be the thing most retirees find they want anyway. And consider a small bond allocation, since a 100% stock portfolio has higher expected returns and meaningfully worse behavior in exactly the scenario that hurts you most.

Building a number from your own spending #

Skip the generic targets and do this instead.

Step 1: find your real annual spending. Twelve months of bank and card statements. The actual total, not the budget you wish you followed. Most people are off by fifteen to thirty percent.

Step 2: adjust for retirement. Some costs vanish and some appear. Commuting, work clothes, payroll taxes, retirement contributions themselves, and often the mortgage all go away. Healthcare goes up, and that is the big one, easy to underestimate badly if you retire before Medicare eligibility at 65. So do travel and hobbies, at least early on.

Step 3: subtract other income. Social Security is the main one. The Social Security Administration publishes an estimate of your benefit based on your actual earnings record, and it is worth pulling. For a lot of people it covers a third to a half of retirement spending, which reduces the portfolio target substantially. Pensions and rental income go here too.

Step 4: multiply the remainder. Take annual spending minus other income, and multiply by 25 for a traditional retirement age, or by 30 if you are retiring early or want more margin.

Worked example, a single person in San Francisco. Current spending $58,000. In retirement, adjusted, $52,000. Estimated Social Security at 67, $28,000. Gap to cover from the portfolio, $24,000. Target: $24,000 × 25 = $600,000.

Six hundred thousand, not the two million the internet quotes at you. The difference is entirely that this person has low housing costs and counted their Social Security instead of pretending it does not exist.

The savings rate table #

What determines how long this takes is your savings rate, meaning the percentage of take-home pay you save. Assuming a 5% real return, roughly:

Read that table for a second, because it says something the individual numbers do not. Going from 10% to 20% saves you fifteen years. Going from 50% to 60% saves you four. The early gains are enormous and they flatten out, which means the most valuable move for most people is getting from barely saving to saving a fifth, not from very frugal to extremely frugal.

This is also why the two levers are income and fixed costs, and why the third lever, cutting small discretionary spending, is mostly noise. Moving from a $3,800 apartment to a rent-controlled $1,680 one changes your savings rate by more than every coffee you will ever skip.

Just track the actual number #

A target calculated once and written on a napkin is worth very little. Your spending changes. Your income changes. Returns are not 7% every year, they are plus twenty two, minus eighteen, plus nine. The projection has to be revisited or it becomes fiction.

That was the reason I built Retire Goals. You set a target amount and a date, log contributions as you make them, and it projects forward from your actual contribution rate and expected return rather than the one you assumed three years ago. Adjust the return assumption and watch the date move, which is a fast and slightly sobering way to understand how sensitive all of this is to a number nobody can predict. Free, no account, data stays on the device. I wrote up how it was built separately.

The honest summary #

25 times spending is a reasonable starting target for a normal retirement age, and 30 times is the safer number for early retirement or for anyone who wants margin. Your spending sets the target, not your income. Flexibility in bad years is worth more than portfolio optimization. Count Social Security, because ignoring it inflates your target by hundreds of thousands of dollars. And recalculate every year or two, because everything in the input changes.

None of this works without the part that comes first, which is starting. The arithmetic on why the first decade matters so disproportionately is in why saving in your twenties beats saving twice as much in your thirties, and the account-type decision is in Roth versus traditional.