Also called: SWR.
The safe withdrawal rate is a summary of a much longer question: given an allocation, a horizon and a spending pattern, how much can come out each year before the plan is likely to fail. It is expressed as a percentage of the starting balance, not the current one — a detail that trips people up constantly, because after a 30% drop your 4% of the original balance is 5.7% of what is left.
A single number can only be "safe" relative to assumptions. Change the horizon from 30 years to 40 and the safe rate falls. Change the allocation from 60/40 to all bonds and it falls further. Include fees and it falls again, by roughly the amount of the fee.
The concept is also conditional on a rigid spending pattern. Almost nobody actually spends the same inflation-adjusted amount through a 30-year retirement regardless of what markets do, and once spending can flex, the safe rate stops being a fixed number and becomes a policy.
The honest use of a safe withdrawal rate is as a starting point and a sanity check, not as a plan. If your intended spending implies 6.5% of the portfolio in year one, no amount of modelling detail will rescue it.
Research on this
See also
- The 4% rule — A rule of thumb from William Bengen’s 1994 research: withdraw 4% of the portfolio in year one of retirement, then raise that amount with inflation each year.
- Sequence of returns risk — The risk that the order in which investment returns arrive damages a plan, even when the average return is unchanged. It matters most when money is being withdrawn.
- Guardrails (Guyton-Klinger) — A withdrawal policy that starts from fixed spending, then cuts it when the withdrawal rate breaches an upper band and raises it when it falls below a lower one.
- Ruin rate — The share of runs in which a plan exhausts its portfolio before the horizon ends. A 5% ruin rate means the money ran out in one simulated run in twenty.