Guyton-Klinger Calculator

Guardrail withdrawals: cut the flexible part of your spending in a year that is bad on both counts, raise it in a year that is clearly good, and see what that flexibility is worth across 2,000 market lifetimes.

A guardrail strategy is a promise you make to your future self: if the markets go badly enough, I will spend less for a while. The reason it is worth modeling rather than merely intending is that the promise is worth a measurable amount — the same plan, run twice, is the only way to see how much. This page sets out precisely which rule this engine applies, on which dollars, under which conditions, and where you can watch it act.

The rule, exactly as implemented

Once anyone in the household is retired, each simulated year the engine compares the household portfolio against two things: its own running peak, and that year’s market return. With the default trigger, both conditions have to be true before spending is cut — the portfolio has fallen below 85% of the highest value it has ever reached in this run, and the year returned worse than -10%. A bad year in an otherwise healthy portfolio does not trigger it, and neither does a drifting portfolio in a flat market.

When both are true, the cut is not graduated: the reducible spending charged that year is reduced by the maximum-reduction setting, 50% of it, for that year only. The next year is judged fresh on the same two tests, so a household that recovers goes straight back to full spending rather than climbing back gradually.

The prosperity rule, the other direction

In a year where no cut is triggered, the engine asks the opposite question. If the portfolio is above 110% of its starting value — not its peak, which is what the cut rule watches — and the year returned more than 5%, the reducible spending goes up by 10% for that year. It is deliberately the smaller lever: the downside rule protects the plan, the upside rule stops it from quietly assuming you would never spend a windfall.

Which of your dollars can flex

Only the ones you say can. The flexible pool is the spending you have marked as reducible — travel, discretionary categories, whatever you would genuinely trim — measured as the amount actually charged in that year, not as a share of the whole budget. The mortgage, the health cover and the groceries are charged in full whatever the market does, because you did not mark them.

The consequence matters more than it sounds: if nothing is marked reducible, the guardrail never runs at all. There is no toggle to switch it on, and no hidden default flexibility injected on your behalf. Marking one expense stream as reducible is the switch, and the size of what you marked is the size of the lever.

Where you can watch it act

In the simulation, not in the single projection. The deterministic year-by-year projection has no market-return series to test — every year is your assumed return — so with the default both-conditions trigger nothing fires there. Set the trigger to portfolio-only in Settings and it can fire in the projection too; left on the default, guardrails are a Monte Carlo phenomenon.

Which is why the simulator below runs your plan twice. Alongside the main result it runs a Fixed Spending comparison — the same household, same market sequences, with every flexible mark stripped off — and the gap between the two success rates is what your willingness to trim is worth, in percentage points, for your own plan. That number is usually more persuasive than any argument about guardrails in the abstract.

The four thresholds are yours

Under Settings → Spending the four thresholds are yours to set: the portfolio trigger, the market trigger, the largest cut a trigger may apply, and the prosperity trigger. The trigger condition itself is a choice of three — portfolio only, market only, or both. Tightening the portfolio trigger makes the rule cautious and cuts more often; widening the maximum cut makes each cut deeper. The simulation reprices the plan against whatever you choose.

How this differs from the published rules

Worth being straight about, since the strategy carries two people’s names. Guyton and Klinger’s decision rules are a set of four, and their guardrails are drawn around the withdrawal rate: spending is adjusted when the current withdrawal rate drifts a set distance above or below its initial value, with separate rules for the annual inflation raise and for which account the money comes from.

What this engine implements is a guardrail-shaped rule with different instruments — a portfolio drawdown against its own peak and the year’s market return — applied to the reducible share of spending. The family resemblance is real and the arithmetic is not theirs. If you came here for a literal implementation of the 2006 paper, this is not one, and you should know that before you read a success rate off it.

Related: the retirement withdrawal calculator sets the same idea against the 4% rule, the Monte Carlo simulator explains what a success rate is, and the guide’s Chance of Success section documents how the trials are built. Or go straight to the full simulator.