How Long Will $1 Million Last in Retirement?

How long will $1 million last in retirement? $1 million is the usual target, the number most people give when asked how much they need. Dividends plus Social Security and they feel like they can make it. The real answer is that it depends (classic attorney answer). The same $1 million can run dry in sixteen years or outlast a thirty-five-year retirement with room to spare. The spread comes from just a handful of inputs, which are how much you withdraw, what your portfolio earns, what inflation does, and how old you are when you start.

Let's run through an example. A single retiree, 65, holds exactly $1 million across a 401(k), a Roth IRA, a brokerage account, and cash, and will collect $2,500 a month from Social Security starting at 67. RetirFi's engine ran the plan to age 95 at three spending levels. Drawing $40,000 a year, the money survives every one of a thousand simulated futures, and the median run grows to $2.9 million. Drawing $60,000, it survives 78 of 100. Drawing $80,000, it survives 20 of 100, and the typical failing run goes broke at 85. Same million, same markets. The answer runs from indefinitely to about two decades.

This article walks through each of those inputs, shows the range they produce, and explains why a single number is the wrong thing to chase in the first place.

Start with the floor: divide and ignore everything else

The crudest estimate is one division. Take the balance and divide it by what you pull out each year.

Treat that as a floor, not a forecast. It assumes the money sits in cash earning nothing, which understates how long it lasts, and it assumes your spending never rises, which overstates it. Two large forces pull against each other once you leave the cash-under-the-mattress world. Investment growth stretches the money out, and inflation compresses what each withdrawal actually buys. The real answer lives between those two pulls, and where it lands depends on the levers below.

The withdrawal rate is the biggest lever you control

What you take out each year matters more than almost anything else, since it sets the pace of depletion. The table below runs the same $1 million at five withdrawal levels. The middle column is the naive floor (no growth). The right column assumes a balanced stock and bond portfolio earning about 6.25% a year, which after roughly 3% inflation is about 3.25% in real terms, with withdrawals held steady in today's dollars.

Annual withdrawal from $1M	No growth (floor)	At ~6.25% (≈3.25% real)
$30,000 (3%)	33 years	Indefinite*
$40,000 (4%)	25 years	~52 years
$50,000 (5%)	20 years	~33 years
$60,000 (6%)	~17 years	~24 years
$80,000 (8%)	~12 years	~16 years

*At a $30,000 draw you are pulling 3% while the portfolio earns about 3.25% above inflation, so growth more than replaces what you take and the balance holds in this smoothed model. That is the knife's edge the whole question turns on. Withdraw less than your portfolio earns after inflation and the principal can last as long as you do. Withdraw more and the clock starts ticking, faster the higher you go. Moving from a 4% draw to a 6% draw cuts the lifespan from about fifty years to the mid-twenties.

These are just illustrations. They assume a steady return every year, which no real portfolio delivers, and that is exactly where the next input comes in.

The smoothed table also leaves out taxes and the order returns arrive in, and the Monte Carlo simulation restores both. Below is the case-study retiree at the same three spending levels, Social Security included, a thousand randomized market sequences per plan.

Annual draw from $1M	Chance it lasts to 95	Runs that deplete	Typical depletion age	Median balance at 95
$40,000 (4%)	100%	none	never by 95	$2,900,000
$60,000 (6%)	78%	22%	91	$910,000
$80,000 (8%)	20%	80%	85	$0

Read the depletion column as the real answer to the headline question. At $40,000 the money never runs out by 95, and the median run ends nearly three times larger than it started. At $60,000 the plan holds in 78 runs of 100, and the runs that fail give out around age 91, twenty-six years in. At $80,000 the odds invert. Four in five deplete, typically at 85 and in the worst sequences at 74, nine years after the retirement party. The smoothed table called the $80,000 draw "~16 years"; the simulation says the truth is a range, and the range is what you plan around.

Returns decide the outcome, and they do not arrive on schedule

The right-hand column above quietly assumes that 6.25% shows up the same way every year. Markets do not work like that. They deliver the average as a jagged line, a few great years, a few terrible ones, long stretches of mediocre. For a retiree pulling money out, the order of those years makes a significant impact.

Same average, different order, different result

Picture two retirees, each starting with $1 million and each drawing $50,000 a year adjusted for inflation. Over thirty years they happen to earn the identical average return. The only difference is timing. Retiree A meets a sharp downturn in the first three years of retirement. Retiree B gets those same bad years near the end instead.

Retiree A is selling shares to fund spending while the market is down, locking in losses that never get the chance to recover. Retiree B spends down a portfolio that grew first, so the late downturn lands on a smaller remaining balance and does less damage. Same average return, and Retiree A can run out a decade or more before Retiree B. The effect is sequence-of-returns risk, and it is the single biggest reason a tidy projection can be wrong in practice. Our guide to sequence-of-returns risk works through the mechanics with a fuller example.

The case study's $80,000 draw shows the spread in one plan. Across a thousand runs with identical inputs, the worst sequences go broke at 74 while the best tenth still holds about $900,000 at 95. Nothing differs between those futures except the order the returns arrived in.

Inflation shortens every number above

The table holds withdrawals constant in today's dollars on purpose, because that is the only way the years mean anything. A fixed $50,000 withdrawal looks like it lasts longer on paper, but at 3% inflation it buys about half as much after twenty-four years. To keep your standard of living level, the dollar amount you withdraw has to climb every year, and a rising withdrawal drains a portfolio faster than a flat one.

So when a back-of-the-envelope figure says "$1 million lasts 25 years at $40,000 a year," ask whether that $40,000 is frozen or growing with prices. If it is frozen, the real lifespan is shorter than it looks. Keep every figure in today's dollars so the numbers stay honest.

The $1 million rarely funds the whole bill

Every figure so far assumes the portfolio pays for all of your spending. For most retirees it does not. Social Security, and a pension if you have one, covers part of the budget, and that part needs nothing behind it from the portfolio. What drains the $1 million is only the gap between what you spend and what your guaranteed income pays.

Take a retiree who spends $70,000 a year and collects $30,000 from Social Security. The portfolio is not on the hook for $70,000; it covers the $40,000 gap. On the table above, that moves them from a roughly sixteen-to-twenty-three-year window down to a forty-plus-year one, on the same $1 million. The benefit check did not change the spending. It changed how much of the spending the portfolio has to carry. Sizing the draw against the gap rather than the whole budget is the step most rough estimates skip, and it is covered in how much you need to retire.

The simulation puts a hard number on the carry. Re-run the case study's $60,000 draw with Social Security deleted and the success rate collapses from 74% to 12%, and the typical failure arrives at 84 instead of 91. The $30,000 benefit is worth 62 points of success probability on an identical budget. On this plan the benefit check is not a supplement; it is the load-bearing wall.

A traditional $1 million is not a Roth $1 million

Where the money sits changes how long it lasts, because the tax has not all been paid. A $1 million balance in a traditional 401(k) or IRA is pre-tax. Every dollar you withdraw is taxable income, so a portion of each draw goes to the IRS rather than to you. A $1 million Roth balance is yours after tax, and qualified withdrawals come out free. In spendable terms, the traditional account is worth less than its statement says, which means it funds fewer years of real spending than an equal Roth balance. The drawdown order across your accounts changes the bill too. Our comparison of Roth versus traditional accounts covers how the tax treatment plays out in retirement. The case-study figures above already carry the tax drag; the retiree's $600,000 traditional balance is grossed up for federal tax on every withdrawal the engine takes from it.

Your starting age sets the bar it has to clear

"Long enough" is not a fixed target. Retire at 65 and a thirty-year horizon is a reasonable plan. Retire at 55 and the same $1 million may need to cover forty years or more, which pushes you toward the lower, safer end of the withdrawal table. Retire at 70 and the money has fewer years to span, so a higher draw can be reasonable. The question is never just how long $1 million lasts; it is whether it lasts longer than you need it to, and that depends on when you start the clock.

The bottom line

$1 million does not come with an expiration date. Depending on what you withdraw, what your portfolio earns and in what order, what inflation does, how much Social Security carries, and how old you are when you begin, the honest range runs from the mid-teens of years to indefinitely. Our case-study retiree landed on every part of that range without changing anything but the draw: 100% at $40,000 a year, 74% at $60,000, 22% at $80,000, with the failing runs giving out anywhere from 75 to well past 90. A single number cannot capture that, and any source that gives you one is trading accuracy for a clean headline.

The way to pin down your own figure is to stop solving for a fixed lifespan and start measuring odds. Enter your spending, your Social Security estimate, your accounts, and your retirement age, then run a Monte Carlo simulation. It replays your plan against many randomized return sequences, including the bad-timing ones, and reports how often the money lasts to the age you set. The output is a probability of success rather than a guess, and for a question with this many moving parts, a probability is the only answer worth trusting. You can run it in a few minutes in the calculator.

Sources

• Philip L. Cooley, Carl M. Hubbard, and Daniel T. Walz, "Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable," AAII Journal, 1998 (the Trinity study).
• Bureau of Labor Statistics, Consumer Price Index
• Social Security Administration, Actuarial Life Table
• RetirFi simulation methodology, described on our about page; scenario figures come from the same engine that runs the calculator.