ReadoutKit

Retirement Savings Calculator

How long the money lasts once you stop earning — with withdrawals that rise with inflation, as they actually do.

Runs entirely in your browser. Nothing you enter is uploaded.

Written and maintained by ReadoutKit / Takashi Yamazaki

How this works

The question is not how much you will have

Most retirement calculators answer the easy half of the problem: compound a pot forwards and print a large number. That number is not the decision. The decision is whether the pot outlives you, and that depends on what happens after the saving stops — a phase with entirely different arithmetic, because money is now leaving while it grows.

This tool runs both phases. It accumulates to your retirement age, then draws the pot down against your spending until it either runs out or reaches age 100, and it tells you which. The chart shows the shape: a rise, a peak, and a decline whose steepness is the whole story.

The price of starting later

Same $500 a month, same $100,000 already saved, same retirement at 65. Only the age you start changes.

Start atPot at 65In today's moneyFirst-year withdrawalMoney runs out at
25$1.96M$886,3724.5%94
30$1.44M$718,6485.6%87
35$1.05M$578,9566.9%82
40$758,374$462,2528.7%78
45$541,427$364,36511.0%75
50$379,312$281,83414.2%73

Each five years of delay costs between a quarter and a third of the final pot. Note the second column against the third: the nominal figures look impressive and the inflation-adjusted ones are the honest version — $1.96M saved from age 25 has the purchasing power of about $886,000 today.

How long a pot actually lasts

Retiring at 65 and spending $40,000 a year in today's money, with 4% returns against 2% inflation.

Pot at retirementFirst-year withdrawalLastsRuns out at age
$500,0008.0%15 years80 years
$750,0005.3%24 years89 years
$1M4.0%34 years99 years
$1.25M3.2%35 yearsNever
$1.5M2.7%35 yearsNever

The relationship is not linear. Doubling the pot from $500,000 to $1,000,000 does not double the years — it takes them from 15 to 34, because the withdrawal rate falls below the point where growth roughly keeps pace with spending. Somewhere around a 4% first-year withdrawal, the money stops being consumed and starts being sustained.

Withdrawals that grow, because costs do

You enter your spending in today's money, and the model raises it every year by your inflation assumption before removing it. This is not a refinement. It is the difference between a projection you can use and one that quietly flatters you.

Consider a pot drawn down by a fixed cash amount. Year 30 removes the same number of pounds as year one, but that money buys roughly half as much at 2% inflation. A projection built that way is not modelling a constant standard of living; it is modelling a retiree whose spending power halves and who does not mind. Holding the real withdrawal constant instead — which is what people actually need — shortens the pot's life substantially, often by five to ten years on a typical set of assumptions.

Real return is the number that matters

The headline return is the one people compare. The real return is the one that determines outcomes. It is what remains after inflation, and it is calculated by division rather than subtraction: a 6% return with 2% inflation leaves 3.92%, not 4%.

The intuition is worth having. When the real return is zero, a pot lasts almost exactly as many years as the pot divided by the annual withdrawal — a million at forty thousand a year lasts twenty-five years, no more. Positive real returns extend that; the extension is dramatic near the point where growth matches withdrawals, which is why small changes in assumed return produce alarming swings in the depletion age. When the real return is negative, there is no sustainable withdrawal rate at all: the pot is shrinking even untouched.

The tool shows the real return next to the withdrawal rate for exactly this reason. If the two are close, your plan is finely balanced and deserves a stress test.

Nominal returns are noise; real returns are the signal

A $1,000,000 pot, $40,000 a year, under different combinations of return and inflation.

ReturnInflationReal returnLastsRuns out at age
2.0%2.0%0.0%25 years90 years
3.0%2.0%1.0%29 years94 years
4.0%2.0%2.0%34 years99 years
5.0%2.0%2.9%35 yearsNever
6.0%2.0%3.9%35 yearsNever
4.0%4.0%0.0%26 years91 years

Look at the first row and the last. A 2% return with 2% inflation and a 4% return with 4% inflation are different-looking plans with the same real return of zero, and they last within a year of each other. Meanwhile 4% against 2% inflation — the same nominal return as that last row — lasts eight years longer. Whenever someone quotes you a return, the number is incomplete until you know what inflation they assumed alongside it.

Where the 4% rule fits

The withdrawal rate shown here is your first year's withdrawal as a share of the pot at retirement, which is the same quantity the well-known 4% rule refers to. That rule came from studies of historical US market data finding that 4%, raised annually for inflation, survived 30 years across the periods tested.

It is a reference point with real limits. It assumes a particular mix of equities and bonds, a 30-year horizon, no fees, and the return history of one country in one century. A retirement beginning at 60 may need to fund 35 or 40 years. Fees of 1% come directly off the real return. Use the 4% figure as a sanity check: a withdrawal rate well above it is a signal to revisit assumptions, not a verdict either way.

What a fixed return cannot show you

This model applies the same return every year. Markets do not. During accumulation that simplification is mostly harmless, because the order of good and bad years barely changes the final total. During withdrawal it is a significant omission.

Once money is leaving the pot, the sequence of returns matters as much as the average. A poor first decade, combined with withdrawals, permanently removes capital that would otherwise have compounded through the recovery. Two retirees with identical average returns can finish decades apart depending only on which years were bad. No fixed-rate projection can display this risk, which is why the honest use of a tool like this one is to explore the shape of the problem, then re-run it with a return one or two points lower and check whether the plan still holds.

The projection also excludes pensions, taxes, fees and one-off costs. Every one of those moves the answer by years rather than months. Treat the output as a frame for a conversation with a qualified adviser, not a substitute for one.

Common questions

How long will $1,000,000 last in retirement?
Spending $40,000 a year in today’s money, with 4% returns and 2% inflation, about 34 years — it runs out at 99. Drop the pot to $750,000 and it lasts 24 years; at $500,000 it is gone in 15. The cliff between those rows is steep because the withdrawal rate crosses from sustainable to not: $40,000 out of $1,000,000 is 4%, out of $500,000 it is 8%.
Does starting ten years earlier really matter that much?
Yes, and the table below puts a number on it. Saving $500 a month from 30 instead of 40 nearly doubles the pot at 65 — $1.44M against $758,000 — and moves the year the money runs out from 78 to 87. The extra contributions account for only $60,000 of that difference; the rest is the compounding those early payments were allowed to do.
Why does the calculator raise my spending every year?
Because your costs will rise whether or not the model says so. You enter what you want to spend in today's money, and each year that figure is increased by your inflation assumption before it is taken out of the pot. Holding the withdrawal flat in cash terms is the single most common way retirement projections flatter themselves: at 2% inflation, a pot that appears to last 35 years on fixed withdrawals may only last 27 once the withdrawals grow.
What is the real return, and why does the tool emphasise it?
It is the return left after inflation, and it is what actually governs how long money lasts. It is not the simple difference between the two rates — a 6% return with 2% inflation gives 3.92%, not 4%, because the calculation divides rather than subtracts. When the real return is zero, the pot depletes in almost exactly the number of years given by dividing it by the annual withdrawal. When the real return is negative, no withdrawal rate is sustainable indefinitely.
Is the 4% rule reliable?
It is a useful reference point rather than a guarantee. The original studies found that a portfolio withdrawn at 4% of its starting value, raised annually for inflation, survived 30 years across historical US market data. That result is specific to a mix of stocks and bonds, a 30-year horizon, and one country's history. It says nothing about a 40-year retirement, a bond-heavy portfolio, or fee drag. Treat 4% as a sanity check on the withdrawal rate this calculator shows, not as a target.
Why does a fixed annual return not reflect reality?
Real returns arrive unevenly, and the order matters enormously once you are withdrawing. Two retirements with identical average returns can end very differently depending on whether the bad years come first — poor returns early, combined with withdrawals, permanently shrink the capital that would have compounded later. This is called sequence-of-returns risk, and no fixed-rate model can show it. Use this tool to understand the shape of the problem, then stress-test with a lower return before relying on the answer.
What is left out of this projection?
A great deal, deliberately. There is no state or workplace pension income, no tax on withdrawals, no investment fees, no lump-sum expenses such as care costs or a new roof, and no partner with a separate pot and a different lifespan. Each of those can move the answer by years. The projection is a starting frame for a conversation with someone qualified, not a plan.

Sources

The formulas and rates used by this tool come from the documents below. Rules change; when a figure here disagrees with the source, the source is right.

Figures checked against the sources below on August 28, 2026.

Other tools