Guyton-Klinger vs Boldin Guardrails

Inspired by: Guyton-Klinger vs. Boldin: Which Retirement Strategy Wins? — The Retirement Gnome (Nov 2, 2026, 14 min).

Rewrite in site voice, not a transcript. Guyton–Klinger rules checked against the 2006 Journal of Financial Planning paper. Boldin thresholds checked against Boldin’s own help article. Not tax, legal, or personalized advice — run your own numbers (or a CFP) before you raise or cut a dollar of spending.

The Point

Spending guardrails are a thermostat, not a 4% thermostat-forever. You pick a method that says: too poor → cut a bit; too rich → spend a bit more; in the middle → leave it alone. The fight is which inputs the thermostat actually sees.

Guyton–Klinger looks at two numbers: portfolio value and how much you took from it. Boldin (and the Morningstar “probability-based” family already on this site) looks at the whole plan — Social Security, pensions, taxes, medical, the rest of the spending — and asks what Monte Carlo success you are sitting on.

Related on this site: Morningstar spend-more study · Josh’s review · Tax timeline 55–75 · Retire at 60

Same idea, different machine

	Guyton–Klinger (2006)	Boldin Spending Guardrails
What it watches	Portfolio + withdrawal rate only	Whole plan (income streams, taxes, medical, spending, balances)
Trigger	Current withdrawal % drifts ±20% from the starting rate	Monte Carlo chance of success vs 95% / 80% / 70%
Move	Cut or raise portfolio withdrawals 10%	Recalculate a gross safe-spend target
Start rate (typical)	Guyton tested ~5–6%. Morningstar’s G-K run started at 5.2%	Whatever spend keeps ~80% success
Still working?	Rules assume you are already withdrawing	Only the Safe Spending Target is useful; the $ guardrails on the page are for people already retired
What “spend” means	Dollars leaving the portfolio	All-in lifestyle: taxes, healthcare, housing, one-time costs averaged in

Guyton–Klinger — four mechanical rules

Jonathan Guyton and William Klinger (JFP, 2006): start at a fixed withdrawal rate, then apply rules every year. The Gnome video walked the two that actually move the paycheck.

Portfolio management. Keep the mix. Rebalance when you drift. Boring on purpose.
Withdrawal / inflation. Raise last year’s dollar withdrawal by last year’s inflation — unless the portfolio had a down year. The original paper’s “modified withdrawal rule” skips the inflation raise after a negative return. The video skipped this caveat. Do not skip it in real life.
Capital preservation (lower rail). Current withdrawal ÷ current portfolio. If that rate is more than 20% above the initial rate (5.0% start → 6.0%+), cut the dollar withdrawal 10%.
Prosperity (upper rail). If that rate is more than 20% below the initial rate (5.0% start → 4.0% or less), raise the dollar withdrawal 10%.

Worked example (video’s 5% start): plan is $50k off $1M. Bad year, portfolio shrinks, same $50k is now 6.5% of what’s left. 6.5% is more than 20% above 5% → 10% cut. Great year, same $50k is 3.5% of a fatter pile → 10% raise.

What it will not see: Social Security claiming, a pension, Roth vs traditional mix, ACA MAGI, IRMAA, or a big one-time truck/house hit. It is a portfolio-withdrawal machine. That is a feature if you want a spreadsheet you can run on a napkin. It is a bug if most of your retirement cash is not “sell shares.”

Morningstar’s study (on this site) still liked G-K for lifetime spending vs a static 4% rule — 5.2% start, $1.36M median lifetime spend, ~$700k median leftover — but probability-based guardrails beat it on lifetime spend and are simpler to explain.

Boldin — Monte Carlo rails on the whole plan

Boldin’s own help article (May 26, 2026): Insights → Spending Guardrails. Unlike a percentage-of-portfolio rule, it solves for spend that keeps success in a band.

Number	Chance of success	What it means
Safe Spending Target	~80%	Baseline lifestyle if things go reasonably well. Every user gets this.
Upper guardrail	~95%	Retired users only. You’re leaving money on the table — room to spend, gift, or travel.
Lower guardrail	~70%	Retired users only. Early warning. Temporary belt-tighten, not a permanent lifestyle funeral.

Still working: the only number that matters is Safe Spend vs Planned Spend. Compare them. Directional. Not a call to blow next month’s budget. Boldin even says the dollar guardrails shown before retirement are based on projected retirement-date balances — the Gnome’s complaint that they should hide those numbers until you actually retire is fair.

Already retired: Safe Spend is what you can support today through longevity, given current balances. Check it on a cadence (Gnome: every few months; Boldin: at least annually). Bull market that pushes you over the upper rail → raise. Drawdown under the lower rail → cut enough to get probability back toward the 80% zone, then reassess.

Do not spend the headline number. It is gross: user-entered lifestyle + debt + healthcare + system-computed taxes + one-time items (car, house) smeared as a monthly average across the whole retirement. A $50k truck in the plan inflates the “monthly” figure. Pay the IRS first.

Which one wins?

Wrong question if you treat them as sports teams. They answer different jobs.

Want a ruleset with no software and no Monte Carlo faith? Guyton–Klinger. Accept that it is blind to Social Security, taxes, and healthcare.
Want the thermostat to see the real household? Probability-based / Boldin. That is the same family Morningstar ranked #1 for lifetime spending and the one Josh called “even I can follow it.”
Want both? Use Boldin (or any Monte Carlo planner) as the annual decision, and keep a G-K-style 20%/10% note in the back of the notebook as a sanity check on the portfolio-withdrawal slice only.

This site’s existing pick does not change: probability-based guardrails over a static 4% rule, with RMD as the simple fallback. See the Morningstar write-up.

Pipps filter

You are not retired yet. Treat Boldin’s Safe Spend vs Planned Spend as a yearly reality check — not a spending raise. Ignore the $ upper/lower rails until the paycheck actually stops.
G-K is the wrong primary tool while wages still exist and while a taxable brokerage is supposed to be the 60–65 bridge. Those years are MAGI / ACA / Roth-conversion years, not “hit 5.2% and inflate.” See the tax timeline.
Once retired, prefer the probability thermostat (Boldin or equivalent) over mechanical G-K. It already knows Social Security timing and tax drag. G-K will cheerfully tell you to take more from the IRA in a year that also kicks IRMAA.
If moving (example: high-tax state → lower-tax state), rebuild the Boldin plan in the new tax regime before you trust the Safe Spend number. State tax is inside “gross spend.”
The video ends in a paid plan-review pitch. Skip it. The usable idea is the thermostat: check, nudge 10%-ish, don’t freeze spending for 30 years out of fear.

Do this once a year (or twice if already drawing)

Update balances by tax type, expected MAGI, and next year’s known one-time hits.
If still working: Safe Spend vs Planned Spend. Gap? Save more, spend less later, or push the date. Do not “give yourself a raise” off a pre-retirement rail.
If retired: where is Monte Carlo vs 70 / 80 / 95? Cut, hold, or raise. Temporary, not a personality change.
Separately: is the portfolio withdrawal rate more than 20% off the starting rate? If yes and the Monte Carlo still looks fine, believe the whole-plan number, not the napkin G-K cut.

Sources

Added Nov 2, 2026 · rewrite of a 14-minute guardrails comparison · confirm current Boldin UI and your own Monte Carlo before you spend a raise