Is $1 Million Enough to Retire? The 4% Rule, Run Honestly
The Million Dollar Question: According to Northwestern Mutual’s 2026 Planning & Progress Study, how much do Americans now say they need to retire “comfortably”?
A) $1 million B) $1.26 million C) $1.46 million D) $2.67 millionRead on for the answer.
Everyone knows the shorthand: $1 million at 4% a year is $40,000 a year, forever. What almost nobody mentions is that the man who invented that rule no longer believes the 4% part, and the people who study it hardest can’t agree on a replacement number either. Here’s what $1 million actually pays out in retirement, who disagrees about how much, and why the public idea of “enough” keeps drifting upward anyway.
What it is
The 4% rule is a single instruction: withdraw 4% of a portfolio’s starting balance in year one of retirement, then adjust that same dollar amount for inflation every year after, regardless of how markets perform. It comes from a specific piece of research — financial planner William Bengen’s October 1994 paper, “Determining Withdrawal Rates Using Historical Data,” which tested every rolling 30-year retirement period in U.S. market history back to 1926 and found the highest withdrawal rate that survived the worst stretch. That was a deliberately conservative exercise: at the time, as the Journal of Financial Planning’s 20th-anniversary retrospective on the paper recounts, the prevailing wisdom — Peter Lynch included, as late as 1995 — assumed a 7% withdrawal rate from an all-stock portfolio was perfectly safe. Bengen’s 4% was the sober correction to that.
What made Bengen’s paper different from the shorthand that followed it is that it was one of the first attempts to test a withdrawal rate against actual sequences of historical market returns — year by year, through real crashes and real recoveries — rather than simply dividing a portfolio by an assumed average annual return. That distinction matters enormously, because an “average” 7% return smooths over the bad decades a real retiree actually has to live through in order. Bengen’s contribution was finding the one number that survived every bad decade in the data, not just the typical one.
What rarely makes it into the shorthand is that Bengen’s own number has never stayed still. His original research pointed to roughly 4.15%; a 2006 book pushed it to 4.5%; a more recent refresh of the same method put his “SAFEMAX” at 4.7%. In his 2025 book A Richer Retirement, he told The Motley Fool’s Robert Brokamp that a typical new retiree today can safely start closer to 5.5% — “it’s a lot better than 4.7%, about 15% to 20% higher.” Meanwhile, Morningstar’s Christine Benz, whose research now anchors much of the opposing camp, currently puts the safe starting rate closer to 3.9% for a 30-year retirement — though she notes a flexible, income-adjusting approach can support something nearer 5% in good years. Same underlying math, same historical data, two serious researchers landing more than a percentage point and a half apart. “4%” was never a law of physics; it was one historical worst case, and it’s been re-litigated ever since.
Who uses it
This is specifically a question for the household approaching a traditional retirement age — roughly 62 to 67 — with household income in the neighborhood of $75,000–$200,000, a house that’s paid off or close to it, and Social Security arriving soon. It isn’t the Fat FIRE household already targeting $4 million–$10 million to retire at 45 (that’s its own post), and it isn’t the family-office tier either, where the question stops being “can I stop working” and becomes something structurally different.
It’s also a smaller club than the public conversation around it suggests. Only about 4.6% of U.S. households — roughly 6 million — hold $1 million or more in retirement accounts, according to the Federal Reserve’s 2022 Survey of Consumer Finances, as reported by The Motley Fool. Fidelity, which administers retirement accounts for 24.8 million plan participants, counted roughly 654,000 of them with $1 million-plus 401(k) balances alone. Reaching $1 million at all puts a household in a thin single-digit slice of the country — even as the public bar for “enough” has moved well past it.
That’s worth placing on the bracket shorthand this site uses elsewhere. A household at $1 million in retirement assets, with a paid-off or nearly paid-off home, is in a different position than the sub-$500,000 household still years from a comfortable number, and in a very different position from the $4 million–$10 million range this site’s Fat FIRE households target for an early exit, let alone the $30 million-plus tier where a family office, not a withdrawal-rate spreadsheet, does the planning. “Is $1 million enough” is specifically the question at the first rung of that ladder — the one a majority of people who ever reach financial independence actually have to answer, rather than the more extreme brackets that get disproportionate attention.
Why they use it
The honest appeal of a single percentage is that it converts an unanswerable 30-year forecasting problem — what will markets return, what will inflation do, how long will I live — into one number a person can actually act on without a financial model of their own. Nobody can know their own lifespan or the next three decades of market returns in advance, so a rule of thumb trades precision for something usable: open the account, apply the percentage, done.
That tradeoff is also exactly why the rule keeps getting revisited rather than settled. A number built to survive history’s single worst 30-year stretch is, by design, wrong for every other stretch — usually too conservative, occasionally not conservative enough. The appeal of simplicity and the cost of simplicity are the same fact, viewed from two directions.
The alternative to a fixed rule is a dynamic one — adjusting spending year to year based on how the portfolio is actually performing, the way Benz’s more flexible guidance allows. Dynamic approaches tend to produce better average outcomes and let a retiree safely start somewhat higher, but they ask for something a flat percentage doesn’t: an annual check-in, a willingness to spend less after a bad market year, and either the temperament or the advisor relationship to actually follow through on an uncomfortable adjustment in a down year rather than spending as if nothing changed. A fixed 4%-style rule trades some of that potential upside for the comfort of never having to make that call.
How it works
Run the actual arithmetic: $1,000,000 at a 4% starting withdrawal is $40,000 in year one, then that same $40,000 is adjusted upward for inflation every year after — a flat, predictable, “constant real dollar” income stream that doesn’t change with the market (a materially different, lower-risk approach is to withdraw a fixed percentage of the current balance each year instead, which flexes income up and down with performance but never guarantees a steady number).
The model behind the number has barely changed since 1994; what keeps changing is which inputs it’s run against. Sequence-of-returns risk — a market downturn arriving in the first five to ten years of retirement — does outsized, sometimes permanent damage, because withdrawals taken during a decline lock in losses a portfolio never gets the chance to recover from. Portfolio allocation matters too: Bengen’s original base case assumed roughly 55% stocks, 40% intermediate government bonds, and 5% T-bills, and part of his newer, higher number comes from modeling a more stock-heavy mix, closer to 65% equities, which produces higher sustainable withdrawal rates over most historical periods. Time horizon moves the number more than almost anything else: Bengen’s own figures run 4.7% for a 30-year retirement, 4.1% for a 60-year one, and roughly 8% for someone who only needs the money to last 10 years. Averaged across roughly a century of hypothetical retirees rather than the single worst case, he says the sustainable rate comes out “a little bit over 7%” — which is most of the gap between his number and Benz’s right there: his 5.5%–4.7% is built to beat most historical outcomes with room to spare; her 3.9% is built to survive the worst one with less margin for error.
What it costs
Translate the competing percentages into real income, and the honest range for an identical $1 million is wide. At Benz’s roughly 3.9% starting rate, that’s about $39,000 in year one. At the classic 4%, $40,000. At Bengen’s 4.7% SAFEMAX, $47,000. At his newer 5.5% recommendation, $55,000. Same million dollars, same retirement date — a $39,000-to-$55,000 range depending entirely on which camp’s research a household trusts.
Layer in Social Security, which most “is $1 million enough” math leaves out entirely. The average Social Security retirement benefit was about $2,075 a month as of January 2026 — roughly $24,900 a year — and specifically about $2,017 a month (about $24,200 a year) for someone claiming at the full retirement age of 67, a figure designed to replace only around 40% of pre-retirement income on its own. Add that to the portfolio withdrawal and a household drawing both lands somewhere around $64,000–$80,000 a year before taxes — meaningfully more livable than $40,000 alone, and still a different order of magnitude from the $4 million–$10 million range this site’s Fat FIRE households are targeting, or the family-office tier several brackets above that. At $1 million, “enough” means a modest income, not a lavish one.
It’s worth running the same math against the numbers people now say they actually want, for comparison. Northwestern Mutual’s 2026 survey put the “comfortable retirement” figure at $1.46 million — which, at a 4% starting rate, is about $58,400 in year one, or roughly $83,000–$88,000 combined with Social Security. The $2.67 million that people who already hold $1 million-plus say they’d need produces about $106,800 at 4% alone, before Social Security — a figure that starts to look less like “comfortable” and more like the lower end of Fat FIRE math. The pattern across all three numbers is the same: every added few hundred thousand dollars buys real, calculable extra income, which is exactly why the “enough” conversation never fully resolves — there’s always a true, defensible reason more would help.
Hidden costs and tradeoffs
The number people set years ago doesn’t stay put. $1,000,000 saved in 1990 has the purchasing power of roughly $2,577,965 today — cumulative inflation of about 158% since then — so a “$1 million” target isn’t one fixed thing; it’s a moving line that means something different every decade, and a household that locked in a nominal target years ago is already working from a stale number without realizing it.
Taxes complicate the clean $40,000 figure further: withdrawals from a traditional 401(k) or IRA are taxed as ordinary income, so the “safe withdrawal” is a pre-tax figure, not spendable cash — and required minimum distributions eventually force withdrawals on a schedule set by the IRS, whether or not the retiree actually needs that money in a given year. Geography adds another layer that resists false precision: the same $1 million and the same Social Security check stretch very differently depending on a state’s cost of living and how it taxes retirement income, which is part of why relocation is such a common move around retirement — directionally real, even without pretending a single state-by-state dollar figure applies to every household’s actual spending. And for anyone retiring before 65, there’s a healthcare bridge to Medicare that has gotten sharply more expensive in the last two years as enhanced ACA subsidies lapsed — a cost this site covers in more depth in the piece on early retirement at higher wealth levels, and one that applies with equal force at the $1 million mark.
Longevity itself is a hidden cost in the math, not just a personal hope. Bengen’s own figures show the sustainable rate dropping from 4.7% over a 30-year retirement to 4.1% over 60 years — meaning a household that retires earlier, or simply lives longer than the historical 30-year assumption baked into the classic rule, needs to plan around a lower number than the one most calculators default to. A healthy 62-year-old household today has a real chance of needing the money to last 30-plus years, which is closer to Bengen’s more conservative end of the range than to his newer, higher one.
What people get wrong
The most common mistake is treating “4%” as settled science rather than a single historical worst case that its own inventor has revised upward at least three times since 1994 — most recently to roughly 5.5% for a typical new retiree. Running the opposite error is just as common: some retirees anchor to the most conservative number they’ve seen (Benz’s ~3.9%) without noting that even she allows for a more flexible approach closer to 5% in good years.
The second mistake is leaving Social Security out of the math entirely, which understates realistic retirement income by roughly $24,000–$30,000 a year for a typical household — the difference between “$1 million produces $40,000” and “$1 million plus Social Security produces closer to $65,000–$80,000.”
The third is assuming $1 million still signals “rich” the way it did a generation ago. Fewer than 5% of U.S. households actually hold $1 million or more in retirement savings, yet survey respondents now say they need 46% more than that — $1.46 million — to feel merely “comfortable.” That gap says less about the $1 million figure itself than about how far expectations and cost of living have moved around it.
The fourth is conflating this question with two different, related ones covered elsewhere on this site: whether $1 million buys what it used to (the purchasing-power question, in The Moving Target) and where $1 million sits on the overall ladder of wealth (in Wealth Levels). This post is narrower and more mechanical than either: not “does $1 million feel rich,” but “how much retirement income does $1 million, drawn down, actually produce.”
The fifth is treating “is $1 million enough” as a yes-or-no question at all, when it’s really a function of four separate variables that each household controls or experiences differently: how long the money needs to last, how the portfolio is allocated, how much Social Security is coming, and where the household actually lives and spends. Two households with an identical $1 million can have genuinely different honest answers — one comfortably retired, one not — without either one doing the math wrong.
Bottom line
The Million Dollar Question answer is C: $1.46 million — the amount Northwestern Mutual’s 2026 Planning & Progress Study found Americans now say they need to retire comfortably, up more than 15% from $1.26 million the year before. (Americans who already have $1 million or more invested set the bar even higher, at an average of $2.67 million — a reminder that the number keeps climbing precisely for the people closest to it.)
Run honestly, $1 million on its own produces somewhere between $39,000 and $55,000 a year in year one, depending on which camp’s safe-withdrawal research a household trusts — and closer to $64,000–$80,000 once an average Social Security benefit is added in. That’s a real, livable income in much of the country, especially with a paid-off house, and it’s a figure millions of American households never reach at all. It is also, unmistakably, a meaningfully tighter number than the $1.46 million people now say they want — which is the real answer underneath the Million Dollar Question: $1 million is enough to retire on. It just isn’t enough, anymore, to feel like the number used to feel.
Related reading:
– Wealth Levels: Life at $1M, $10M, $100M, and $1B
– The Moving Target
– HENRY: $500K and Still Paycheck-to-Paycheck
– FIRE for the Affluent: Coast FIRE, Fat FIRE, and What Early Retirement Really Costs
