How much do I need to retire? A practical way to work it out

"How much do I need to retire?" is the most searched question in personal finance and one of the most badly answered. The internet is full of single numbers — a million, two million, a multiple of your salary — and almost none of them tell you whether the figure applies to your life.

You can build a defensible answer in about an hour with a pencil, four numbers and one honest assumption. Here is the method, then the caveats that decide whether it holds up.

Step 1 — Find out what your life actually costs

Not what you earn. Not what you think you spend. What leaves your account in a normal year.

Open twelve months of bank and card statements and add up the outflows. Group them into three buckets:

• Essential: housing, food, utilities, insurance, transport, health costs, debt payments.
• Discretionary: travel, dining, hobbies, gifts. The spending you could dial down in a bad year.
• Irregular: a new roof every fifteen years, a car every eight, dental work, family help. Divide the total by the number of years it covers, then add that annual figure back in — this is the bucket people forget, and it is often 10–15% of the total.

Now adjust for retirement. Some costs fall: commuting, work clothes, the second car, retirement savings contributions themselves. Some rise: healthcare before Medicare eligibility, travel while you are still mobile, hobbies that used to be your job. A common planning assumption is 70–85% of pre-retirement spending, but if you have tracked your real numbers, use them instead of a rule of thumb.

Step 2 — Add up the income you will have anyway

You are not funding your whole spending from savings. You are funding the gap.

• Social Security. Create an account at ssa.gov and look at your actual statement. The headline figure assumes you work until full retirement age; claiming early reduces it, claiming at 70 increases it. For a household, the higher earner usually delays and the lower earner often claims earlier.
• Pensions. Any defined-benefit pension, including from a former employer. Check whether it has a survivor benefit and whether it is indexed to inflation.
• Annuities already owned. Income you have already purchased.
• Rental or business income you expect to continue, net of costs.

Be conservative here. A pension that is not inflation-indexed loses roughly a third of its purchasing power over twenty-five years at 3% inflation.

Step 3 — The gap, and the multiple that covers it

Subtract guaranteed income from spending. What remains has to come from your portfolio.

The 25× figure comes from the long-standing "4% rule": if you withdraw about 4% of the portfolio in the first year and raise that amount with inflation, historical US market data suggests a portfolio has usually lasted thirty years or more. Multiply the annual gap by 25 and you have a rough target.

Thirty-three times is the same idea at 3%, a lower and more cautious starting rate. Use 25× if you are comfortable with a fair amount of market risk and a thirty-year horizon; use 30× or more if you are retiring early, if you want a wider margin, or if you would rather not think about markets in your eighties.

Step 4 — Test the timeline, not just the total

A target without a date is a wish. Take your current savings, add what you put away each year, and see how long it takes to reach the figure at a realistic return.

Two rules for that projection:

• Use a real return, not a nominal one. A 7% nominal return with 3% inflation is about 4% real. Projecting nominal returns while spending in today's dollars flatters the result.
• Run a bad decade. Sequence matters more than average. Two portfolios can earn identical average returns over thirty years and end very differently depending on whether the bad years came first. A projection that only shows a smooth average line is hiding the part of the problem you are actually trying to solve.

If the answer is "at this rate, never", you have three levers, in order of how much they usually help: spend less, work a few years longer, or save more now. Adjusting the investment return assumption is the fourth lever and the weakest one, because you do not control it.

The five assumptions that decide the answer

Anyone who gives you a number without stating these has given you a guess.

1. How long you live. Planning to 95 for a couple is common; for a healthy 60-year-old, there is a meaningful chance of one partner reaching beyond that. The cost of over-planning is a slightly larger portfolio; the cost of under-planning is running out.
2. Inflation. Especially healthcare inflation, which has historically run above general inflation.
3. Investment returns, and the order in which they arrive.
4. Health and long-term care. One serious care event can be the largest single expense of a retirement. Insurance is expensive and unattractive; so is the alternative.
5. Taxes. Withdrawals from a traditional 401(k) or IRA are taxed as income. A million dollars in a pre-tax account is not worth a million dollars of spending. This is why the order you withdraw from matters so much.

Three things people get wrong

They use income instead of spending

A salary is not a lifestyle. Two households earning $120,000 can retire on $55,000 and $95,000 respectively. Only the spending number is useful.

They forget the withdrawal order

Which account you draw from first can change the lifetime value of a portfolio by a lot, and it is one of the few things you fully control. Taxable first, tax-deferred later, tax-free for the years that would otherwise push you into a higher bracket is a common starting point — but the right answer depends on your bracket, your state, and your estate intentions.

They stop at the number

The number is not the deliverable. The deliverable is a plan that says: here is what we spend, here is where it comes from each year, here is what we do if the market falls 35% in year three, and here is what we do if it does not. That is what makes the number safe to live on.

Frequently asked questions

Is the 4% rule still valid?

It is a reasonable starting point, not a guarantee. It came from historical US data over thirty-year periods and assumes a fairly high allocation to stocks with no investment fees. With lower expected returns, longer retirements or higher fees, 3–3.5% is a more cautious starting rate. Use the rule to set a range, then test your own plan.

Should I include my house in the number?

Only if you plan to sell it or borrow against it. For most people the home is where they live, so including it makes the target look healthier than the cash flow will be. Count the home if downsizing is a real part of your plan, and count only the difference between its value and the cost of where you would move.

What if I retire before I can claim Social Security?

Then the gap is larger in the early years and smaller after you claim. This is often the most expensive stretch of a retirement — no pension income, no Medicare, and health insurance bought at full price. A bridge plan that funds those years deliberately is usually worth the effort.

How does a pension change the calculation?

A pension reduces the gap, which reduces the portfolio you need. A $20,000 annual pension with no cost-of-living adjustment covers roughly $500,000 of portfolio at 4% in year one — less in real terms as inflation erodes it. Treat secure income as valuable, and be honest about whether it is indexed.

Is one million dollars enough to retire?

For a single person spending $50,000 a year with $20,000 of Social Security, a million is comfortable. For a couple spending $100,000 with $40,000 of Social Security, it is not. The number is meaningless without the spending, the guaranteed income and the time horizon attached to it.

If you want to run these numbers with someone who is not paid by the sale, that is what our first call is for. You can also read how to choose a financial advisor first, so you know what questions to ask.