Meet two retirees, Ann and Bea. Each retires with $1,000,000, withdraws $50,000 a year, and earns an average annual return of exactly 6.5 percent over the next twenty years. Not similar averages. The same average, from the very same list of yearly returns. The only difference is the order: Ann gets the bad years first, Bea gets them last. Twenty years later, Bea has about $1.77 million. Ann has about $468,000 and a portfolio losing altitude. Same money, same spending, same average return, and an ending gap of more than $1.2 million. That gap has a name, sequence-of-returns risk, and it is the single most important idea in retirement spending. This article explains it with real numbers, examines what the famous 4 percent rule does and does not promise in 2026, and walks through the withdrawal strategies, guardrails above all, that are built to survive markets as they actually behave.
In 1994, a financial planner named William Bengen asked a question that sounds obvious only in hindsight: looking at actual U.S. market history rather than average returns, what is the highest withdrawal rate a retiree could have taken in the worst case without running out of money over 30 years? He tested every historical starting year with a portfolio of roughly half to three-quarters stocks, withdrawing a fixed inflation-adjusted amount. The answer that survived even the cruelest start dates, like retiring in 1966 into a 16-year sideways market with raging inflation, was about 4 percent. The Trinity study later reached broadly similar conclusions. The rule's mechanics matter and are often misquoted: you withdraw 4 percent of the portfolio in year one only, then adjust that dollar amount for inflation each year regardless of what the market does.
Three things the rule never promised: it does not guarantee the future will be no worse than the past, it was not designed for retirements longer than about 30 years, and it assumed a disciplined investor who held through crashes. It also has a flip side almost nobody mentions. Because it was calibrated to the worst case, the median historical retiree following it died with more money than they started with. The 4 percent rule is simultaneously too risky for the unlucky and too stingy for the typical, which is exactly why the adaptive strategies later in this article exist.
Look again at Ann and Bea. The return list is identical, averaging 6.5 percent. Ann's version starts with three losing years, down 15, 10, and 8 percent, while Bea banks strong gains early and hits those same losses in years 18, 19, and 20. By the end of year three, Ann's million is down to about $581,000, and here is the mechanism that matters: her $50,000 withdrawals during the downturn forced her to sell shares at depressed prices. Those shares are permanently gone. When the recovery arrives in year four, it compounds on a smaller base, and twenty years of decent markets never close the wound. Bea sells nothing cheap until the final years, when losses barely matter because two decades of compounding already happened.
This is why your retirement date is the riskiest day in your financial life, and why the five years before and after it are sometimes called the fragile decade. While you are saving, a crash is a sale on shares; the order of returns barely affects the accumulator. The moment withdrawals begin, order becomes destiny. Any honest withdrawal strategy is really a machine for surviving an unlucky order.
The research community has spent thirty years stress-testing Bengen's number, and the debate has settled into a useful range rather than a single answer. Arguments for a lower starting rate: a 30-year horizon understates modern longevity for healthy early retirees, and some researchers modeling lower expected future returns have suggested starting nearer 3.7 to 4 percent. Arguments for a higher one: Bengen himself, adding more asset classes to the historical tests, has argued the worst-case rate lands meaningfully above 4 percent, and rigid inflation-adjusted spending is a behavioral fiction anyway, since real retirees cut back in bad years without being told.
The practical synthesis most planners land on: treat 4 percent as a planning anchor, not a contract. If you retire into reasonable valuations with flexibility in your budget, modestly more is defensible. If you retire at 55, or into a euphoric market, or with a budget that has no fat to trim, start lower or build in guardrails. The number matters less than the willingness to adjust, which brings us to the strategies.
The original. Withdraw 4 percent of the starting balance in year one, then give yourself an inflation raise every year, mechanically. Its virtue is a perfectly predictable income, which suits retirees whose budgets are mostly fixed costs. Its vice is that it ignores the portfolio entirely: it keeps spending through a crash as if nothing happened, which is precisely how the unlucky cohorts fail, and it never spends the surplus in the lucky scenarios. Best for: planning and benchmarking. Few people should actually fly it on full autopilot.
Withdraw a set percentage, say 4.5 or 5 percent, of whatever the portfolio is worth each year. The portfolio can never technically hit zero, since spending shrinks automatically in downturns and grows in booms. The cost is an income that can swing painfully: a 30 percent market drop means a 30 percent pay cut the following year. Best for: retirees with a strong guaranteed-income floor from Social Security or pensions who can absorb variable discretionary spending.
Guardrails strategies, the best-known designed by planners Jonathan Guyton and William Klinger, start with a higher withdrawal rate than the classic rule, often around 5 percent, and then police it with simple triggers. The version below captures the essence without the academic trimmings.
Why this works: the adjustments are small, rare, and happen exactly when they are most powerful. A 10 percent spending cut during a deep bear market dramatically reduces the shares you liquidate at the bottom, which is the precise mechanism that destroyed Ann in our example. In historical testing, guardrails retirees both spent more over their lifetimes and failed less often than fixed-rule retirees, paying for it with the occasional lean year. The strategy formalizes what sensible people do anyway, and the formality is the feature: deciding the rules at 65, while calm, beats improvising them at 73, while terrified.
Divide the portfolio by time horizon. Bucket one holds one to three years of withdrawals in cash and equivalents in a high-yield savings account or money market fund. Bucket two holds roughly years three through ten in bonds. Bucket three holds the long-term money in stocks. Spend from cash, refill it from bonds, refill bonds from stocks, but only when markets are up. In a downturn, you stop refilling and live off the safe buckets, giving stocks years to recover before you touch them. Mathematically, bucketing is mostly an asset allocation wearing a comforting costume, but the costume matters: retirees who can point at five years of safe money demonstrably panic-sell less, and not panic-selling is worth more than most optimizations.
Cover your essential expenses with income that cannot fail: Social Security, any pension, possibly a plain immediate annuity purchased with a slice of the portfolio, possibly a ladder of Treasury bonds or TIPS. Then invest the rest for growth and spend it flexibly on the life you actually want. Delaying Social Security to 70 is the cheapest floor-builder available, since the benefit grows about 8 percent for each year of delay past full retirement age and arrives inflation-adjusted for life. A retiree whose groceries, housing, and insurance are covered by guaranteed income can watch a bear market with something approaching indifference, which is the entire point.
Every withdrawal rate is secretly a bet about longevity, and the bet deserves daylight. The classic research assumed 30 years, which covers a 65-year-old living to 95. That is conservative for the average retiree and optimistic for planning purposes, because you are not funding the average outcome; you are funding your own tail risk. For a healthy 65-year-old couple, the odds that at least one spouse reaches their mid-90s are substantial, and that surviving spouse is precisely the person a depleted portfolio would hurt most.
The horizon moves the math in both directions. A 55-year-old early retiree may need the money to last 40 years, which argues for a starting rate meaningfully below 4 percent, more flexibility, or a plan to earn income in the early years. A 75-year-old beginning withdrawals, by contrast, is funding perhaps 20 years and can responsibly start near 5 percent or higher. Single retirees with serious health conditions sit differently than healthy couples with longevity in the family tree. None of this requires morbid precision. It requires noticing that "safe" is always relative to a horizon, and choosing yours deliberately rather than inheriting the 30-year default.
Theory becomes useful when it survives contact with an actual household, so here is one. Carol retires at 67 with $800,000, a paid-off house, and a Social Security benefit of $2,600 a month, about $31,200 a year. Her essential expenses, housing costs, food, insurance, utilities, and health premiums, total $3,400 a month, or $40,800 a year. Her comfortable life, with travel and grandkids and a reliable car, costs about $66,000 a year.
Step one, the floor: Social Security covers $31,200 of the $40,800 in essentials, leaving a $9,600 essential gap. That gap is small enough that Carol skips the annuity and simply earmarks her bond allocation to cover it for a decade in any market. Step two, the starting rate: she needs $34,800 from the portfolio for the full comfortable life, which on $800,000 is a 4.35 percent withdrawal rate, comfortably inside guardrails territory. Step three, the rails: her upper guardrail is 5.2 percent and her lower is 3.5 percent. If a bear market drags her balance to $670,000, her $34,800 withdrawal hits 5.2 percent, and she trims spending 10 percent to about $31,300, postponing one trip. If markets boom and her balance reaches $1,000,000, her rate falls under 3.5 percent and she gives herself a raise. Step four, the buffer: she keeps $70,000, two years of withdrawals, in cash and short-term Treasuries, refilled in up years. Carol's plan fits on an index card, and every number on it came from the strategies above.
Two technical notes keep real-world plans on the rails. First, withdrawal rules describe gross withdrawals, and taxes come out of that gross. If Carol pulls $34,800 mostly from a traditional IRA, some slice goes to income tax, so her spendable amount is smaller than her withdrawal. Mixing account types changes the picture: Roth withdrawals are tax-free, taxable-account withdrawals incur only capital gains on the growth portion, and the standard sequencing of taxable first, pre-tax second, Roth last is a sensible default that low-income early-retirement years can improve on with Roth conversions.
Second, required minimum distributions currently begin at 73 for pre-tax accounts and rise as a percentage of the balance with age. RMDs can exceed what your strategy says to withdraw, and retirees sometimes read that as the government overriding their plan. It is not. An RMD forces money out of the tax shelter, not out of your savings. Withdraw it, pay the tax, and reinvest whatever your spending plan does not need in a plain taxable brokerage account. Your withdrawal strategy governs spending; the RMD merely governs location.
Market crashes are loud. Inflation is silent, and over a 30-year retirement it does comparable damage. At just 3 percent inflation, $50,000 of annual spending becomes about $104,700 of required spending in 25 years. A withdrawal plan that ignores inflation does not fail dramatically; it fails by inches, as the retiree quietly stops replacing the car, then trims the travel, then the heating. Drag the sliders below and watch what your own spending number becomes.
The defenses are specific. Social Security is inflation-indexed, which is yet another argument for maximizing it. TIPS, Treasury securities whose principal adjusts with CPI, can immunize a slice of the bond allocation. And stocks themselves, for all their short-term violence, have been the most reliable long-run inflation outpacer, which is why even conservative retirement portfolios usually keep a meaningful stock allocation deep into the 80s.
One last defense deserves its own paragraph because it outperforms all the clever ones: the willingness to bend. Researchers who study actual retiree behavior keep finding that real households do not spend like spreadsheets. They naturally ease off in scary markets and loosen up in good ones, and their portfolios are far more durable than rigid simulations predict. The strategies in this article do not ask you to become a different person. They take the flexing you would do anyway in a panic and move it forward in time, into calm, written rules with specific triggers. That transfer, from improvisation under stress to policy under calm, is most of what separates the retirements that survive bad luck from the ones that do not.
The 4 percent rule is not broken, and it was never magic. It is a measurement of how bad U.S. market history has been to its unluckiest retirees, and a reminder that averages tell you almost nothing about your own sequence. You cannot control which decade the bear market arrives in. You can control everything else: a starting rate chosen with honest margins, guardrails decided in advance, a few years of cash standing between your spending and a crash, an inflation-protected floor built as high as Social Security timing allows, and the humility to take a small cut in a bad year so the portfolio lives to fund a long, good life. Ann and Bea got identical markets. Strategy is the only difference you get to choose.
Contribution rates matter, but the salary they multiply against matters more. Whether you are mid-career or planning a second act, RealWorldCareers shows which work fits your brain so your strongest earning years are actually your strongest.
Find the career your brain was built forIt remains a reasonable planning anchor rather than a law of nature. The rule survived every historical 30-year U.S. period it was tested on, including retirements that began just before the brutal markets of 1966 and 1973. Researchers continue to debate the right number, with some recent work suggesting modestly lower starting rates and the rule's own creator arguing history supports somewhat higher ones. The honest answer: 4 percent with flexibility is durable; 4 percent on rigid autopilot with no willingness to adjust is the fragile version.
It is the risk that the order of your investment returns, not just their average, determines whether your money lasts. When you are withdrawing, early losses force you to sell more shares at low prices, and those shares are gone when the recovery arrives. The same returns in a luckier order leave you selling fewer shares early and compounding more later. It is why two retirees with identical average returns can have wildly different outcomes.
A common approach is one to three years of planned withdrawals in cash or short-term instruments. The purpose is not return; it is to avoid selling stocks during a deep downturn. After a bad market year, you spend from the cash bucket and let the portfolio recover, then refill the bucket in good years. More than a few years of cash starts to drag on long-term growth, so it is a buffer, not a strategy by itself.
No, and this surprises people. The 4 percent framework describes gross withdrawals; taxes come out of that money, not in addition to it. Required minimum distributions, which currently begin at age 73 for pre-tax accounts, can also force you to withdraw more than your rule suggests. You do not have to spend an RMD, though. You can reinvest the after-tax remainder in a taxable account, keeping your actual spending on plan.
Guardrails set your starting withdrawal, often between 4 and 5.5 percent, then watch your current withdrawal rate as markets move. If a falling portfolio pushes the rate above an upper guardrail, you trim spending, commonly by about 10 percent. If a rising portfolio drops the rate below a lower guardrail, you give yourself a raise. Small, rule-based adjustments at the right moments historically allowed higher lifetime spending than rigid rules with the same failure risk.
An annuity is not a rival to a withdrawal strategy; it can be a component of one. Using part of a portfolio to buy a plain single-premium immediate annuity creates a guaranteed income floor that, combined with Social Security, covers essential expenses no matter what markets do. That floor lets the remaining portfolio take growth risk with money you will not need in a panic. The products to avoid are the complex, high-fee variety sold on fear.
One smart money idea each week, charts included. Join free and get the printable 2026 Money Calendar in your welcome email.
