The Rule Everyone Knows, and the Part Most Explanations Skip
If you want the full breakdown of what the 4% rule actually says and how to turn it into your own savings target, How Much Do You Need to Retire already covers that ground. Quick recap since it matters here: withdraw 4% of your portfolio in year one of retirement, adjust that dollar amount for inflation every year after, and historically that held up for at least 30 years in nearly every period researchers tested.
What usually gets left out is what happens after you retire and actually start pulling the money out. The rule isn’t a guarantee, it’s a probability based on specific historical assumptions, and those assumptions just shifted in two different directions in 2026. Worth understanding both the update and the mechanics behind it before you build a plan around any single number.
Short version: In 2026, the 4% rule’s own author revised his safe withdrawal number up to 4.7%, while independent research from Morningstar moved the other direction, down to roughly 3.9%. Both are defensible, they’re measuring different risks. What matters more than picking a number is understanding sequence of returns risk, the mechanism that makes the first few years of retirement disproportionately important, and being honest about whether 30 years is even the right planning horizon for your situation.
The 2026 Update Nobody’s Retirement Calculator Has Caught Up To
Bill Bengen, the researcher who created the 4% rule back in 1994, revisited his own data using a more diversified portfolio than his original study, spreading stock exposure across large-cap, mid-cap, small-cap, micro-cap, and international equities instead of just large-cap US stocks. Testing that mix against every 30-year retirement period since 1926, including the worst one on record (a retiree starting in October 1968, who faced both a market collapse and runaway inflation at the same time), his updated worst-case safe withdrawal rate came out to 4.7%, up from the 4.15% his original research actually found (that number got rounded down to a clean 4% and stuck for three decades). Bengen has said that under more typical, non-worst-case conditions, something closer to 5% to 5.5% is often supportable.
Morningstar’s independent research moved the opposite direction. Running forward-looking simulations based on current market valuations and bond yields rather than backward-looking historical data, their 2026 baseline sits at roughly 3.9%, up slightly from 3.7% the year before, but still meaningfully more conservative than either version of Bengen’s number.
Neither side is wrong, they’re answering different questions. Bengen is asking: what’s the worst 30-year stretch that has ever actually happened, and what would have survived it? Morningstar is asking: given where valuations and yields sit right now, what’s likely to hold up going forward, even if the future looks worse than anything in the historical record? A retiree with a genuinely diversified portfolio and some flexibility to adjust spending in a bad year has more room to lean toward Bengen’s end. A retiree who wants to lock in one number and never revisit it is better served planning around something closer to Morningstar’s.
Why the Number Matters Less Than the Mechanism: Sequence of Returns Risk
Here’s the part that actually explains why this whole debate exists. Two retirees can have the exact same average return over 30 years and end up in completely different financial positions, depending on when the bad years happen to land.
Picture two retirees, both starting with $1,000,000, both withdrawing $40,000 in year one. Retiree A hits a rough market right out of the gate, a 15% drop in year one. That portfolio drops to $850,000 before the withdrawal even comes out, and now $40,000 is being pulled from a smaller base, leaving less principal to ride the eventual recovery. Retiree B has the exact same 15% drop, but it lands in year 29 instead of year 1. By then, decades of growth mean that same percentage loss is hitting a portfolio that’s grown substantially larger, and there’s far less time remaining for it to matter.
Same market event, same average return over the full 30 years, wildly different outcome depending purely on timing. This is why the first five to ten years of retirement carry outsized weight in every withdrawal-rate study, and it’s also the real argument for holding more in bonds or cash specifically in the years right around retirement (sometimes called a “bond tent”), then gradually shifting back toward stocks once you’re a decade or so in and the highest-risk window has passed.
When 30 Years Isn’t the Right Number
Every safe withdrawal rate discussion above, Bengen’s, Morningstar’s, the original Trinity Study, assumes a roughly 30-year retirement starting somewhere around age 65. That assumption doesn’t hold for a lot of people doing physical trade work.
A desk job and 30 years of knees, backs, and shoulders taking a beating on job sites are different retirement timelines. A lot of guys in the trades aren’t choosing to retire at 55 or 58, their body is making that decision for them, whether or not the savings are actually there yet. That changes the math in a way none of the withdrawal-rate research above accounts for directly: a 55-year-old retiree isn’t funding 30 years, they’re potentially funding 35 to 40, and every one of the safe withdrawal rates discussed here gets less safe the longer the money has to last. A lower starting withdrawal rate, closer to the Morningstar end of the range or below it, is the more honest planning assumption if an early, physically-forced exit is a real possibility for your trade.
Two things help hedge against this specifically. First, disability coverage that would actually replace income if the job ends before the plan does, worth a real look at what your policy or Social Security Disability Insurance would actually pay if the trade work has to stop early. Second, treating the retirement age in your own plan as a range rather than a fixed number, and stress-testing your savings target against the earlier end of that range, not just the optimistic one.
A Fixed Percentage Isn’t Your Only Option
The classic 4% rule is rigid by design, the same inflation-adjusted dollar amount comes out every year, whether the market had a great year or a terrible one. That rigidity is exactly what creates sequence of returns risk in the first place, a bad market plus a fixed withdrawal is the worst possible combination.
Dynamic withdrawal strategies address this directly. The general idea behind approaches like the Guyton-Klinger guardrails method: give yourself a raise in good years, but skip the inflation adjustment (or actively cut spending) in years where the market drops enough to push your withdrawal rate meaningfully above your starting target. It’s a more involved approach to manage than “withdraw the same number every year,” but it directly targets the mechanism that does the most damage, pulling a fixed amount out of a shrinking portfolio during exactly the years it can least absorb it.
You don’t need a formal guardrail system to get the benefit. Even an informal version, being willing to trim discretionary spending (travel, big purchases) in a year the market’s down 15%, and only taking the full inflation raise in years it isn’t, captures most of the protection without needing a spreadsheet formula to manage it.
The Order You Draw From Actually Matters
Which account the money comes out of first is its own decision, separate from how much comes out. The general default sequence: taxable brokerage accounts first (capital gains rates are typically lower than ordinary income rates, and it lets tax-deferred and Roth balances keep compounding longest), then tax-deferred accounts like a traditional 401(k) or IRA, then Roth accounts last, since Roth money grows tax-free for as long as you leave it alone.
That default order isn’t always the smartest one, though. A more tax-efficient approach often involves pulling some money from tax-deferred accounts earlier than strictly necessary, specifically to “fill up” the lower tax brackets each year, rather than saving all of it for later and taking a bigger, more expensive hit once required minimum distributions kick in.
Speaking of which: the IRS eventually forces the issue. Required minimum distributions on traditional 401(k)s and IRAs currently start at age 73 for anyone born 1951 through 1959, and 75 for anyone born 1960 or later. Miss one and the penalty is steep, 25% of the amount that should have come out, though that drops to 10% if corrected within two years. Roth 401(k) and Roth 403(b) accounts no longer have RMDs at all as of 2024, one more reason Roth money is often the account to preserve the longest. This is exactly why the order matters as much as the amount, a plan that ignores RMDs can end up forcing a much larger, more heavily taxed withdrawal later than a little planning now would have required.
Putting It Together
There’s no single correct percentage here, and anyone claiming otherwise is selling something. A reasonable starting range for 2026 sits somewhere between Morningstar’s roughly 3.9% and Bengen’s 4.7%, with your specific number depending on how diversified your portfolio actually is, how much flexibility you’re willing to build into your spending, and honestly, how long you’re actually likely to need this money to last. That last one is where the math genuinely differs for a lot of readers here: if a physical trade career means retirement might arrive well before 65, planning around the more conservative end of that range, or lower, is the honest move, not the pessimistic one.
Frequently Asked Questions
It’s still a reasonable starting reference point, but the number itself moved in two directions in 2026. Bengen, the rule’s original author, revised his own worst-case safe rate up to 4.7% using a more diversified portfolio. Morningstar’s independent forward-looking research puts it closer to 3.9%. Both are defensible depending on your portfolio and risk tolerance.
It’s the risk that a market downturn in your first few retirement years does far more damage than the same downturn happening later, because you’re withdrawing money from a portfolio while it’s already down, leaving less principal to recover. Two retirees with identical average returns over 30 years can end up in very different positions depending purely on when the bad years landed.
Every major withdrawal-rate study assumes roughly a 30-year retirement starting around age 65. Physical trade work often forces earlier retirement due to wear on the body, which means the money may need to last 35 to 40 years instead. The safe withdrawal rate for a longer horizon is meaningfully lower than the rate for a standard 30-year one.
Instead of taking the same inflation-adjusted dollar amount every year regardless of market performance, a dynamic strategy adjusts spending up in good years and holds or cuts it in bad ones. It takes more active management than a fixed percentage, but it directly reduces sequence of returns risk, which is the biggest single threat to a retirement portfolio lasting.
The general default is taxable accounts first, then tax-deferred (401(k), traditional IRA), then Roth accounts last. But pulling some tax-deferred money earlier to “fill up” lower tax brackets each year, rather than saving it all for later, can reduce the total tax bill once required minimum distributions force larger withdrawals starting at age 73 or 75.
Roth IRAs never have RMDs for the original owner. Roth 401(k) and Roth 403(b) accounts were exempted from RMDs starting in 2024 as well. Traditional 401(k)s and IRAs still require withdrawals starting at age 73 (for those born 1951-1959) or 75 (born 1960 or later).
Sources