Safe Withdrawal Rate Calculator

It is the share of a portfolio a retiree can withdraw in the first year, and then sustain in real terms, with a high probability of the money outliving them. The best-known estimate is the 4% rule from William Bengen's 1994 study and the Trinity study, both based on historical US stock and bond returns over 30-year retirements. More recent research by Anarkulova, Cederburg, O'Doherty and Sias (2025), drawing on a broad sample of developed markets rather than the exceptional US record, puts the comparable figure near 2.7%. No single number is safe in every market and at every horizon, which is why this calculator shows how long your own spending lasts instead of promising a rate.

The steady projection grows the balance at your expected real return and withdraws the annual spending at each year's end. Solving that recurrence for the year the balance reaches zero gives the annuity duration formula, so the answer is exact rather than stepped in whole years. Two special cases matter. If the annual spending is no more than the portfolio times the real return, growth replaces every withdrawal and the money never runs out. And at a 0% real return the formula degrades to a straight division, the portfolio over the spending.

The calculator simulates 5,000 possible market histories. In each one, every year's real return is drawn from a distribution whose average matches your expected return and whose spread is the volatility you set, and the spending is withdrawn at the year's end. The success probability is simply the share of those histories in which the portfolio still holds money at your horizon. An 80% success probability therefore means the plan failed in 1,000 of the 5,000 simulated histories. It is a way of measuring how much room the plan has for bad luck, not a forecast of any particular market.

Two retirements can earn the same average return and end completely differently depending on the order in which the returns arrive. When poor years come first, withdrawals are taken from an already shrunken portfolio, and the strong years that follow work on too small a base to repair the damage. That is sequence-of-returns risk, and it only exists while money is being withdrawn; a saver with no withdrawals ends in the same place whatever the order. It is the reason the steady projection in this calculator is systematically more optimistic than the simulation, and why the gap between the two is worth taking seriously.

A real return, meaning after inflation, entered as an arithmetic mean. The calculator keeps every figure in today's money, so a withdrawal that keeps its purchasing power is a constant number only if growth is measured net of inflation too. Diversified equity portfolios have historically delivered roughly 4 to 5% real over long horizons, which is why the presets are 2, 4 and 5%. For volatility, a broad equity-heavy portfolio has seen annual swings near 15%, and mixes with more bonds sit lower. Do not enter a nominal 7 or 8% return here, and do not subtract inflation from a return that is already real.

Taxes on investment income and withdrawals, fund and platform fees, and any state or employer pension, which for most people replaces part of the spending; check how your own country taxes retirement withdrawals. The simulation draws each year's return independently from a lognormal distribution, which understates crashes that are deeper or longer than that model allows, and it keeps spending fixed in real terms, while real retirees usually cut back after bad years. It also assumes one portfolio with a single return, not separate accounts with different rules. Treat the output as a stress test of your plan, not a guarantee.