Can I Spend More in Good Years and Less in Bad?
The 4% rule spends the exact same inflation-adjusted dollar amount every year for thirty years, market be damned. Nobody actually lives that way. Here's what happens when spending is allowed to flex instead.
- The 4% rule assumes you spend the same real dollar amount every single year of retirement, no matter what the market does. That's a modeling convenience, not how anyone actually lives — and it means the rule has to plan as if every retirement might be the worst historical case, because it has no way to adapt if things go badly.
- Guyton-Klinger guardrails let spending move with the portfolio: a raise when you're comfortably ahead of plan, a cut (commonly 10%) when your withdrawal rate climbs too high after a bad stretch. It typically supports a higher starting withdrawal rate with a comparable or better success rate — the cost is that you may need to actually take the cut when it's called for, not just in theory.
- You can run both strategies side by side on your own numbers below — free, no signup, nothing leaves your browser.
The assumption nobody notices they're making
Ask most people how they picture their retirement spending and they'll describe something that flexes — a little more when things are good, tighter when they're not, exactly like they've spent money their entire working life. Then ask them to picture how the 4% rule actually works, and it's the opposite: pick a dollar amount in year one, adjust it for inflation every year after, and hold it there for three decades regardless of what the market is doing underneath it.
That's not a criticism of the 4% rule's math — the math is sound, and it's been stress-tested against real historical market history since Bengen's original study. The issue is the assumption baked into it. A withdrawal strategy that can never adapt has to be defensive enough to survive the worst 30-year stretch in the historical record, every single time, even in the far more common years where the market does fine. The insurance premium for that rigidity is real: in good markets, a retiree following the constant rule is often sitting on a portfolio far larger than they'll ever spend, having quietly under-lived their own plan to protect against a bad outcome that never showed up.
What Guyton-Klinger guardrails actually do
Guyton-Klinger guardrails replace the single fixed number with a dial. The mechanics are simple to state, even though the underlying math is doing real work:
- A capital-preservation rule. If the portfolio has fallen enough that the current withdrawal rate has drifted too high relative to where it started, spending gets pulled back — commonly a 10% cut — to bring the rate back down before the drawdown compounds.
- A prosperity rule. If the portfolio has grown enough that the withdrawal rate has drifted comfortably low, spending is allowed to rise — the plan isn't just protecting against bad outcomes, it's responding to good ones too.
- Bands around the target rate. The guardrails only trigger when the withdrawal rate has moved meaningfully off its starting point (a common setup uses a 20% band in either direction) — small, ordinary market noise doesn't trigger a change. Only a real shift does.
The result isn't spending chaos. It's a strategy that responds to what's actually happening to the money instead of pretending every year of a thirty-year retirement will look identical.
The honest trade-off
Here's the part that matters most, stated plainly: guardrails are not a free lunch. Across most simulated futures, they let a retiree support a higher starting withdrawal rate than the static 4% rule with a comparable or better long-run success rate — but the mechanism that makes that possible is the same one that asks you to actually take a spending cut when the market has had a genuinely bad stretch. That's not a hypothetical footnote. It's the entire trade.
A static 4% plan gives you more certainty about next year's number and less overall spending power across the plan. A guardrails plan generally gives you more overall spending power and a real chance — not a certainty, but a real chance — that you'll need to tighten your belt in a bad year. Which one is "better" isn't a math question. It's a question about how you'd rather absorb risk: locked-in predictability, or flexibility that comes with the obligation to actually flex when it's called for.
The surprising part isn't that guardrails are safer. It's that across most simulated futures, they also let a retiree spend more, on average, than the constant rule — because the constant rule is quietly over-insuring against the worst case every single year, not just the years that turn out bad.
Where this matters most
For a plan that's comfortably above the numbers that matter, the choice between constant and flexible spending is closer to a lifestyle preference than a survival question. For a plan sitting near the edge of "safe" — a success rate in the gray zone, a starting withdrawal rate that's a little aggressive for a fixed strategy — flexibility can be the difference between a plan that doesn't work and one that does, at the honest cost of accepting that some years won't look like the plan on paper. That's worth knowing before you commit to either approach, not after a bad market year forces the question.
A worked example
Take a household retiring at 65 with a $1,000,000 portfolio, planning to a life expectancy of 90, starting at a 4.5% initial withdrawal rate — a little more aggressive than the traditional 4%, which is exactly the situation where the two strategies tend to diverge most. Run both through a 1,000-scenario simulation at that same starting spend level and the comparison isn't close to a wash:
| Strategy | Success rate | What it means |
|---|---|---|
| Constant (4.5% rule) | Fixed spending, no adjustment | Same real dollar amount every year, regardless of market performance |
| Guyton-Klinger guardrails | Dynamic, band-based | Spending rises in strong markets, cuts (~10%) in weak ones — reported alongside median cut/raise counts across the simulated paths |
The calculator above reports the real numbers for whatever inputs you enter — success rate, median and percentile ending balances, and average real spending for each strategy, plus how often the guardrails strategy actually triggered a cut or a raise across the simulated paths. That last figure is the honest part most comparisons skip: it's not enough to know guardrails can support more spending, you should also know roughly how often you'd actually be asked to adjust.
What this doesn't capture
This comparison is deliberately apples-to-apples with the original Trinity Study and Bengen 4%-rule framing: portfolio-only, no Social Security, no pension income layered in. That's the right way to isolate the withdrawal-strategy question on its own terms, but it means the numbers here aren't your full retirement picture — most real retirees have Social Security covering a meaningful share of spending, which changes how much the portfolio itself needs to carry. It also doesn't model one-time expenses, staggered retirement dates, or real estate transactions. For the full plan with all of that layered in, the complete app is the right tool; this page isolates one specific, genuinely important decision.
How this is calculated
Both strategies run through the same 1,000-scenario Monte Carlo simulation using randomized historical market returns, with a 3% inflation assumption and 7% nominal pre- and post-retirement returns. The Guyton-Klinger strategy uses a 20% band with a 10% spending adjustment when triggered — the standard setup referenced across FIRE-community and retirement-research writing on the topic. It's the same engine behind the full app — free to run, no account required. The complete math, including the specific assumptions and data sources, is documented in the methodology.
Common questions
This article runs one calculator on a portfolio-only, apples-to-apples comparison. The full app models your complete plan — Social Security, taxes, one-time expenses, and stress-testing against real market history.
Run your full plan free →