Retirement Calculator

The 25× rule is one division, and it hides three decisions

Every “how much do I need to retire” answer comes out of the same division: the annual spending your portfolio has to cover, divided by the withdrawal rate you are willing to take. At a 4% rate, dividing by 0.04 is the same as multiplying by 25, which is why the 4% rule and the 25× rule are not two guidelines but one statement said twice. If your savings need to produce $50,000 a year, that division points at roughly $1.25 million. The arithmetic is trivial. The three decisions buried inside it are not.

The spending figure is a convention, not a measurement. The calculator starts from your current income and a replacement percentage, because almost everyone can state their salary far more reliably than their retirement budget. That is a stand-in, and it is the input to argue with first, because it scales the entire corpus. If the mortgage ends before you retire, if you plan to move somewhere cheaper, or if you intend to travel hard for the first decade and then stop, the replacement percentage you accepted by default is doing more damage or more flattery than any other number on the page.

Outside income is subtracted before the division, not after. A pension or a Social Security payment does not reduce your target by its own size; it reduces the spending the portfolio has to fund, and that reduction is then multiplied by the same 25. Every $1,000 of reliable annual income from outside the portfolio removes $25,000 from the corpus you have to build at a 4% rate. This is the largest single lever on the number, and it is why two households with identical spending can face targets that differ by a million dollars.

The withdrawal rate is the most contested input here. The 4% figure is a historical guideline drawn from US market data over rolling 30-year periods, not a guarantee: it assumes a particular portfolio mix and it ignores fees and taxes. Some analysts argue a lower starting rate of around 3% to 3.5% is safer when returns or interest rates are low; others note that a retiree willing to cut spending in weak years can sustain more. The multiplier moves sharply with that choice — 4% gives 25×, while 3.5% gives roughly 28.6×, so the same $50,000 of spending asks for about $1.43 million instead of $1.25 million. Treat the rate as something to test across the conservative and optimistic scenarios, not a setting to accept.

Behind those four formulas the model commits to a handful of things. All of them are visible, and all of them are editable:

• The projection runs month by month with a single, constant pre- and post-retirement return; contribution step-ups, inflation, and the drawdown all apply across the schedule.
• The required corpus divides your annual income need by the withdrawal rate — at 4% this is the same as the 25× rule.
• Future values are shown in both nominal terms and today’s money, discounted by the inflation rate you enter.
• The readiness score blends corpus adequacy, longevity, saving pace, and withdrawal sustainability into one 0–100 figure — it is descriptive, never a hard-coded label.
• Everything is currency-neutral: switching currency changes formatting only, never the math.

Why 3% inflation halves what your balance buys in about 24 years

At 3% inflation prices roughly double in about 24 years, which is the same sentence as: the purchasing power of a fixed sum halves over that period. A plan that runs from 35 to 90 spans more than two of those doublings, and one of them falls entirely inside retirement, after the contributions have stopped. That is why every future figure on this page is shown twice — once in nominal money and once discounted back into today’s money at the inflation rate you entered — and why the second figure is the one worth reacting to. The nominal number is the one that feels like the answer; the real number is the one you can actually spend.

The same erosion runs through the drawdown. A plan that withdraws a flat amount every year is quietly cutting your standard of living each year, so the schedule raises the withdrawal with inflation instead of holding it level. That is what makes the “how long will it last” figure so sensitive: the balance is being drained by a payment that grows while the returns funding it do not.

It also sets the bar your return assumption has to clear. A 7% expected return against 3% inflation leaves roughly four percentage points of real growth — enough to compound meaningfully across thirty years of saving. The lower post-retirement return most plans use, 5% in the worked figures below, leaves only about two points above inflation, which is why the money stops growing much once withdrawals begin. If you ever find your return assumption sitting close to your inflation assumption, the plan has stopped depending on investment growth at all and is running purely on how much you save.

Thirty years of saving, twenty-five years of spending: one plan end to end

A 35-year-old has $50,000 saved and adds $750/month, expecting a 7% return before retirement and 5% after, with 3% inflation. They plan to retire at 65, expect to live to 90, earn $90,000 a year, and want to replace 80% of that income, withdrawing at 4%.

The projected $1,320,803 falls short of the estimated need of $4,369,072 by $3,048,269, giving a readiness score of 33/100 — At risk. Switching to "Find monthly saving" shows how much extra to put away to close it.

Two rows in that table carry most of the meaning. The first is the distance between the projected balance and the same balance in today’s money: that gap is thirty years of 3% inflation and nothing else, and it is the reason a seven-figure projection is not automatically a comfortable retirement. The second is the estimated monthly income, which is only the corpus at the chosen withdrawal rate spread across twelve months — the plan is never better than the spending that one figure has to cover. Move the retirement age by two years, or the return by half a point, and every row here shifts; that sensitivity is precisely why it is worth running the projection instead of trusting a rule of thumb.

A readiness score of 75 means your assumptions agree, not that your plan is safe

The score out of 100 is the fastest read on the page and the easiest to over-trust. It blends four things: how the projected savings compare with the estimated need, whether the money lasts the full retirement you specified, how your current saving compares with the saving the goal actually requires, and how sustainable the withdrawal rate is against the post-retirement return. A score of 75 or above generally means those four measures point the same way. Below it, the calculator shows the size of the gap and what would close it rather than leaving you to guess at the shortfall.

What the score cannot do is check your inputs. It is a summary of the assumptions you entered, so it inherits every one of them, including the optimistic ones. Raise the expected pre-retirement return by two points and the score climbs without a single thing changing in your finances. Shorten the life expectancy by five years and it climbs again. The honest way to use it is to move one input at a time and watch which one the score reacts to hardest — that input is what your plan is really resting on, and it is usually either the return or the retirement age rather than the contribution.

The score is descriptive rather than prescriptive: it reports what your own numbers imply, and it is deliberately not a verdict, a rating, or advice. Two people holding the same score can be in very different positions, because identical arithmetic can be built on a cautious set of assumptions or a hopeful one, and the score has no way to tell those apart.

Delaying retirement by three years pulls two levers at once

A shortfall has five possible answers: save more each month, retire later, spend less in retirement, accept a higher and therefore riskier expected return, or bring in more income from outside the portfolio. They are not equally powerful, and exactly one of them does two jobs at the same time.

Retiring later is the compound lever. Moving retirement from 65 to 68 adds three more years of contributions and three more years of growth on everything already saved — and because the plan still runs to the same life expectancy, it shortens the drawdown from twenty-five years to twenty-two. The projected corpus rises while the number of years it has to cover falls, so both sides of the comparison move in your favour at once. Nothing else on the list does that, which is why even a two- or three-year delay closes gaps that look hopeless when you attack them with contributions alone.

Saving more is the most reliable lever, because it is the only one wholly under your control. The “Find monthly saving” mode solves it directly: it returns the level monthly amount that reaches your target from where you stand today, which is a far more useful answer than being told you are behind. Spending less in retirement works through the same multiplier as everything else, so trimming $2,000 from annual spending removes $50,000 from the required corpus at a 4% withdrawal rate — a reminder that decisions about how you will live are also decisions about how much you must accumulate.

Raising the expected return deserves suspicion. It is the one lever that improves the projection without changing anything you actually do, and it buys that improvement with risk the model does not price. If the only route to the target is a higher assumed return, the plan has not been fixed; the spreadsheet has. Run the conservative and optimistic scenarios before you let a return assumption carry the outcome, and treat any gap that only closes under the optimistic case as still open.

Sequence of returns: the risk a constant return rate cannot show

This projection applies one constant rate before retirement and one constant rate after it. Real markets do not deliver an average; they deliver a sequence, and once you are withdrawing money the order of that sequence matters more than its average. A withdrawal taken during a downturn sells a larger share of the portfolio than the same withdrawal taken after a recovery, and that share is simply gone — it is not there to participate when prices come back. Two retirements with identical average returns and identical spending can therefore end in completely different places, decided by nothing but which years happened to be the bad ones.

During accumulation the order barely matters, because you are buying rather than selling and a weak decade in your thirties is partly a discount. In drawdown the same volatility becomes a permanent leak. That asymmetry is why a crash in the first few years of retirement does damage that later good years never fully undo, and it is why the “years your money lasts” figure here should be read as a central estimate rather than a floor. A plan that only just survives on a constant return has no margin left for a bad opening decade.

Those are the limitations that matter most, and they are not the only ones. This calculator does not predict markets, and it does not include the following unless you enter them yourself:

• Sequence-of-returns risk and real market volatility (the model uses a constant return)
• A market crash early in retirement, which can do lasting damage
• The full detail of income, capital-gains, and estate taxes
• Healthcare, long-term-care, and other lumpy late-life costs
• Exact Social Security, State Pension, or national pension calculations

Tax deserves a specific warning. The optional tax estimate applies one flat rate to withdrawals; it does not distinguish an account taxed on the way out from one taxed on the way in, it does not model capital-gains treatment, and it does not know about withdrawals a plan may force you to take at a certain age. Healthcare and long-term care are similarly lumpy and back-loaded in a way that a smooth, inflation-adjusted withdrawal cannot represent. Use the projection to compare decisions against each other, which it does well, rather than to predict a balance on a date, which nothing can.

Pensions, Social Security and superannuation enter as one number you supply

The calculator takes guaranteed retirement income as an input and never estimates it. Whatever you expect from Social Security, a company pension, an annuity, or rental property goes into “Other annual retirement income” as a single yearly figure, and it is subtracted from your income need before the corpus division runs. That placement is what gives it leverage: at a 4% withdrawal rate, every $1,000 of dependable annual income from outside the portfolio removes $25,000 from the corpus you have to build, and it lowers the monthly saving required to get there in the same proportion.

It does not compute your Social Security benefit, because it cannot. The benefit depends on your own earnings record and the age at which you claim, and the Social Security Administration is the only body that can produce your figure — get it from your SSA account and bring the number here. The same holds, more strongly, for defined-benefit plans. FERS and military retirement pensions are calculated from years of service and salary history under their own statutory formulas, which this tool does not implement; if you know the expected annual amount, entering it as other income makes the rest of the projection correct.

Account type matters less than people expect. Australian superannuation, a 401(k), an IRA, and their equivalents elsewhere are all the same object to this model: a balance that compounds until retirement and is drawn down afterwards. Enter the balance as current retirement savings and the projection behaves identically. What the model deliberately does not do is enforce the annual contribution caps those accounts carry — it projects the saving rate you type, so if you enter more than your plan allows, it will cheerfully build a corpus you are not permitted to fund. The IRS pages cited below are the authority on the current caps.

Currency is presentation only. Switching between USD, INR, GBP, EUR, CAD and AUD reformats every figure, the year-by-year schedule and the Excel workbook, and changes nothing in the mathematics — the model holds no view on local tax rules, pension systems, or market returns in any particular country. That is a convenience for comparison and a limitation for anyone who needs jurisdiction-specific rules, which is the point at which a local adviser beats any calculator.

Sources

Social Security income: this calculator takes your expected Social Security (or pension or annuity) income as an input and never estimates the benefit itself — benefit amounts depend on your earnings record and claiming age, and the Social Security Administration is the only authority that can compute yours. Get the figure from your SSA account or the SSA retirement pages below, then enter it here.

Account contribution limits: the accounts this page references (401(k), IRA, and their equivalents) carry annual contribution caps this calculator deliberately does not model or enforce — it projects the saving rate you enter. The IRS pages below are the authority for the current caps; verify your plan year there.

The 25× / 4% rule of thumb: discussed above as a historical guideline from US rolling-period studies, not a standard set by any authority — treated here as a reference point to test, not a rule to follow.

• Social Security Administration — retirement benefits (how benefits are determined; where to get your own estimate)
• IRS — COLA dollar limitations on benefits and contributions (401(k) and IRA caps by plan year)
• IRS Publication 590-A — contributions to IRAs (the rules behind the IRA references on this page)
• How we calculate — site-wide methodology

Related calculators

More in Finance, or browse all calculators.

Read the guide

For how a withdrawal projection is actually built, and why the "years your money lasts" figure shifts so much when you nudge one input, see How to Estimate Retirement Withdrawals Without Overtrusting One Number.