How much can you actually spend in retirement?

Turning a pot of savings into a monthly paycheck is the central problem of retirement planning, and three forces make it genuinely hard. The first is longevity: a plan built around an average life expectancy fails roughly half the time by definition, so a sound plan is built to your late 80s or 90s, not to the average. The second is sequence-of-returns risk: a market decline in your first few retirement years does more damage than the same decline decades in, because you're selling shares to fund spending at exactly the wrong moment. The third is taxes and inflation: the same gross withdrawal buys a different lifestyle depending on where it comes from and what it's worth by the time you spend it.

None of this means the question is unanswerable — it means a single number rarely survives contact with any of the three. What follows is built to survive all of them.

William Bengen's 1994 study asked a specific, narrow question: what fixed, inflation-adjusted withdrawal rate would have survived the worst 30-year sequence in US market history? The answer was close to 4%, and it became one of the most cited numbers in personal finance. It was a landmark finding, and it remains a reasonable sanity check.

It was never a spending plan, though, and taking it as one exposes real gaps. It assumes you never adjust — not in a great market, not in a terrible one. It ignores taxes entirely, treating a traditional 401(k) dollar the same as a Roth dollar. It treats your first year's number as destiny for thirty years. And it was calibrated to a 30-year horizon that may not match yours at all.

Guardrails, first formalized by Jonathan Guyton and William Klinger in 2006, replace the single fixed rate with pre-agreed rules: spend more when your plan is running ahead, trim by a modest, planned amount when it falls behind. In practice that means an upper rail — spend more if your plan's success odds are comfortably high — and a lower rail — trim if they fall below a chosen threshold — recalculated as markets move and the plan updates.

The consumer version of this is a monthly range rather than a single figure: instead of "$3,300 a month, forever," the answer becomes "$3,300–$4,000 a month, and here's what would move it." That's the question most people actually wake up with — not what a historical worst case allows, but what's true for their plan today.

The same $80,000 of spending can require withdrawing very different amounts depending on where it comes from — and a spending answer that ignores this is a gross number pretending to be a net one. A few of the mechanics worth knowing, each covered in depth elsewhere:

• Withdrawal order — drawing from taxable, then traditional, then Roth accounts changes your tax bracket path across retirement, not just this year's bill.
• Social Security taxation — up to 85% of benefits can become taxable depending on your other income, which makes withdrawal timing part of the Social Security decision too.
• IRMAA — Medicare premium surcharges are based on income from two years earlier, so a large withdrawal or Roth conversion today can raise a bill you won't see until later.
• Required minimum distributions — once RMDs begin, the tax-deferred portion of your portfolio starts drawing itself, whether or not it fits your spending plan that year.

Consider a hypothetical couple, both 65, with a $1 million portfolio and Social Security covering part of their spending. A guardrail-style plan might support roughly $3,700–$4,000 a month at retirement — a range, not a fixed rate. If markets fall meaningfully in year two and their plan's success odds drop below the lower rail, the guardrail rule might suggest a modest, single-digit-percentage trim — not a crisis response, a planned one, revisited at the next check-in.

Compare that to the alternative: a fixed 4% spender who never checks back until year fifteen, discovers the plan is off track, and faces a much larger adjustment all at once. The guardrail trim is unglamorous by design. It's the plan working, not the plan failing.

Any spending answer decays. Markets move, tax law changes, and your own plans change — a spending range calculated today is a snapshot, not a permanent verdict. The educational guidance is straightforward: revisit after major market moves, at least annually, and after any real change in your life (a move, a health event, a new goal).

This is also where ongoing monitoring earns its place in a plan. A system that periodically re-checks your numbers and tells you, calmly, whether anything material changed is doing the same job as the annual review — just more often, and without you having to remember to ask. Framed correctly, that's reassurance between check-ins, not a system making decisions for you.

RetireGlide's safe-spending range comes from a deterministic, versioned planning engine that runs your plan through 1,000 simulated market futures alongside your full tax picture — Social Security taxation, RMDs, capital gains, IRMAA, and the rest. The engine, not AI, produces the range and the guardrail thresholds.

AI's role is explaining what the range means and helping you get your inputs right — proposing facts for your confirmation, never computing a projection or a tax result. Every figure the assistant states about your plan is cross-checked against those deterministic results before you see it, and only shown as verified when it matches. It explains. You decide.

See your own current range with the free calculator — no signup required.