FIRE glossary
Safe vs Perpetual Withdrawal
A safe withdrawal rate is sized to survive your own retirement horizon; a perpetual ("never touch principal") rate is sized to hold the pot steady forever, which needs a smaller draw — or a bigger pot — than merely lasting your lifetime does.
Two different design targets
"Safe" and "perpetual" describe two different jobs a withdrawal rate can be asked to do, and everyday usage blurs them together. A safe rate — the sense the original 4% Trinity Study aimed at — was tested against real historical market sequences over a fixed horizon (30 years, in that study) and picked because it survived nearly all of them: it's sized to outlast a specific, finite retirement, not to run forever. A perpetual rate is the "never touch principal" ideal instead: a draw small enough that the pot's real value never shrinks at all, so it could in principle support the withdrawal indefinitely — the target an endowment, a family preserving wealth across generations, or anyone who deliberately doesn't want to spend the pot down would actually need. The two targets sound similar but solve different problems, and a rate that's genuinely safe for a 40-year retirement isn't automatically one that holds forever.
Why the arithmetic makes them easy to conflate
Ember's standard safe-withdrawal-rate basis prices your required pot as spend ÷ rate. Run through a single constant real-return assumption, that formula leaves the pot roughly self-sustaining whenever the chosen rate sits at or below the assumed return — in that closed-form model, the pot mathematically never fully depletes, which is exactly why a "safe" 4% rate can read as if it were perpetual. But the 4% figure wasn't derived from a perpetuity requirement; it came from testing a finite historical horizon and finding it survived almost every rolling window. Ember's drawdown-to-zero basis makes the finite-horizon intent explicit instead, pricing the pot to run out at (or near) a specific life expectancy on the same return assumption — and it usually comes out smaller for the same spend, because it isn't carrying the extra margin true perpetuity needs.
Which one should guide your number
A true perpetual rate has to sit at or below the real return you expect to sustain, since drawing more than the pot grows depletes it eventually even without any run of bad luck. A rate that's merely safe for your own horizon can afford to run a bit higher, because it's allowed to spend down to a defined endpoint rather than hold steady forever. Choosing between them is really a choice about intent — preserve capital indefinitely versus spend deliberately down to a horizon you've estimated. Ember doesn't pick one for you: switch your plan's withdrawal basis between safe-withdrawal-rate and drawdown-to-zero to see how the required pot moves for the same spending target, or turn on the optional draw-to-zero secondary target to compare both on your dashboard at once.
Across borders
How long "perpetual" and "your lifespan" actually diverge depends partly on where you live and retire — life expectancy varies by country, and a horizon-limited plan built around one country's figure understates the pot needed if a later move adds years of retirement in a place where people typically live longer. Inheritance and estate rules also shape which framing even matters: in a jurisdiction with heavy inheritance tax, preserving principal "for heirs" nets a family less than the gross figure implies, which can tilt the choice back toward spending down deliberately. Whichever basis you pick, moving tax residency still changes the required pot through the tax gross-up on withdrawals — that effect is separate from, and stacks with, the safe-versus-perpetual choice itself.
Common questions
Is the 4% rule actually a perpetual withdrawal rate?
No — it was derived from testing one finite historical horizon (30 years, in the original Trinity Study), not built to guarantee an indefinite pot. In a simple constant-return model it can behave close to perpetual because 4% sits near many long-run real-return assumptions, but that's a feature of the maths, not the rule's original design intent.
Do I need a perpetual rate if I'm not leaving money to anyone?
Not necessarily. If the goal is to fund your own retirement rather than preserve capital forever, a rate sized to your own life expectancy — Ember's drawdown-to-zero basis — usually supports a smaller required pot for the same spend than a genuinely perpetual rate would.
How do I compare the two in Ember?
Set the withdrawal basis on your plan to safe-withdrawal-rate or drawdown-to-zero and Ember re-solves the required pot on that basis, with the full maths trace behind the number — switch between them to see how much of the gap is genuinely about horizon length versus a margin for never touching principal.
Related terms
See safe vs perpetual withdrawal in your numbers
The free calculator gives a rough estimate; the full planner models your actual accounts, pensions, residency moves and taxes — with the maths behind every figure shown.