Guardrails Withdrawal Strategy: How the Guyton-Klinger Rules Work and What They Cost

Guardrails let a retiree start at a 5% withdrawal rate instead of 4% by agreeing in advance to cut spending when the portfolio falls. Here is how the Guyton-Klinger rules work, a 25-year replay from 2000 showing every cut, why a 100% Monte Carlo success rate is not what it looks like, and how to set a floor before the first bear market.

20 min readNovember 13, 2026
Guardrails
Guyton-Klinger
Dynamic Withdrawals
Safe Withdrawal Rate
4% Rule
Sequence of Returns
Retirement Spending
Monte Carlo
Retirement Income
Risk Management

The 4% Rule Has One Job and No Reflexes

The 4% rule tells a retiree to withdraw 4% of the portfolio in the first year and then give themselves an inflation raise every year after, no matter what the market does. A retiree following it in 2009, with the portfolio down by a third, takes the same raise as in any other year.

Nobody actually behaves that way. A household that watches its savings fall by a third spends less, at least for a while. The rule's rigidity is also what makes it so conservative. Because it never reacts, the starting rate has to be low enough to survive the worst sequence in history with no help from the retiree at all.

Guardrails are the alternative. The retiree starts higher, usually around 5%, and agrees in advance to a written rule for when spending gets cut and when it gets raised. This article explains the rules, replays them through 25 real years, and puts a number on the part most explanations skip, which is what the cuts add up to.

The Short Version

Guardrails trade a steady paycheck for a portfolio that almost cannot run dry. A retiree who started in 2000 at $50,000 instead of $40,000 took five pay cuts and no raises in 25 years, ended with $1,183,523, and spent the last fourteen of those years on about 59% of the purchasing power they started with. The retiree who began in 2008 took two cuts and still out-spent the 4% rule every single year. The rule works. What it costs depends entirely on the decade you retire into.

Where Guardrails Come From

The idea was formalized by financial planner Jonathan Guyton in "Decision Rules and Portfolio Management for Retirees: Is the 'Safe' Initial Withdrawal Rate Too Safe?", published in the Journal of Financial Planning in October 2004. He argued that retirees could begin at a higher withdrawal rate if they committed to a set of rules for adjusting along the way.

In March 2006, Guyton and computer scientist William Klinger published a follow-up in the same journal, "Decision Rules and Maximum Initial Withdrawal Rates," which tested the rules with Monte Carlo simulation and added the two that gave the strategy its name. The paper described them as acting like "financial guardrails" when markets push the withdrawal rate far from where it started.

The full Guyton-Klinger framework has four parts:

  • A portfolio management rule that decides which assets each year's withdrawal comes from.
  • A withdrawal rule that governs the annual inflation raise and withholds it after certain losing years.
  • The capital preservation rule. If the current withdrawal rate has risen more than 20% above the starting rate, cut this year's withdrawal by 10%. As published, this rule switches off in the last 15 years of the planning horizon.
  • The prosperity rule. If the current withdrawal rate has fallen more than 20% below the starting rate, raise this year's withdrawal by 10%.

Their conclusion was that initial withdrawal rates of 5.2% to 5.6% were sustainable at a 99% confidence standard over at least 40 years for portfolios holding at least 65% stocks. With 50% stocks, the maximum fell to as low as 4.6%. Either way, that is a meaningfully bigger first paycheck than 4%.

The last two rules are the guardrails, and they are what most planning software means by "Guyton-Klinger." They are often summarized as 20/20/10: a 20% band above, a 20% band below, and 10% adjustments.

How the Two Guardrails Work

The rule watches one number, the current withdrawal rate: this year's planned withdrawal divided by today's portfolio balance.

Take a retiree with $1,000,000 who starts by withdrawing $50,000. The starting rate is 5%.

  • The upper guardrail is 20% above that, or 6%.
  • The lower guardrail is 20% below, or 4%.

Each year the retiree increases the withdrawal for inflation and then checks the rate. Inside the 4% to 6% band, nothing happens. Above 6%, spending is cut 10%. Below 4%, spending is raised 10%.

It helps to translate the guardrails into portfolio balances, because a balance is easier to watch than a ratio. In the first year, a $50,000 withdrawal crosses the 6% line when the portfolio falls below $833,333 and crosses the 4% line when it rises above $1,250,000. Those two balances move up each year with the inflation raise.

Notice what the rate is really measuring. The withdrawal rate rises for two reasons: the portfolio fell, or spending grew faster than the portfolio did. A portfolio that merely goes sideways for a decade while spending inflates at 3% a year will hit the upper guardrail without a single dramatic crash.

A 25-Year Test: Retiring in January 2000

The replay below uses the same retiree, the same returns, and the same conventions as the one in the bond tent article: $1,000,000, a 60/40 portfolio rebalanced every year, published calendar-year total returns for the S&P 500 and the Bloomberg U.S. Aggregate Bond Index, withdrawals at the start of each year, and 3% inflation. It applies the two guardrail rules only, with the full inflation raise every year and no 15-year cutoff, which is how most software implements them.

Three retirees, identical portfolios:

  • Fixed 4%. Starts at $40,000 and takes a 3% raise every year.
  • Fixed 5%. Starts at $50,000 and takes a 3% raise every year.
  • Guardrails from 5%. Starts at $50,000, takes a 3% raise every year, and applies the 20/20/10 rule.

What happened to each

The fixed 5% retiree ran out of money. The balance was down to $576,525 after 2008 and never recovered, because the withdrawals kept growing while the portfolio did not. By the start of 2022 there was $73,796 left to cover a $95,805 withdrawal. The account was empty in the plan's 23rd year.

The fixed 4% retiree survived, as the rule promises. They spent $40,000 of purchasing power every year and finished 2024 with $777,158. But that final year's $81,312 withdrawal was 10.8% of what was left. The plan made it through 25 years. It is not on course to make it through 35.

The guardrails retiree never came close to running out, and paid for it in spending.

YearFixed 4% withdrawalGuardrails withdrawalGuardrails purchasing power vs. year oneFixed 4% year-end balanceGuardrails year-end balance
2000$40,000$50,000100%$952,243$942,324
2001$41,200$51,500100%$876,806$857,347
2002$42,436$47,74190%$757,975$735,479
2003$43,709$44,25581%$848,891$821,505
2008$50,671$51,30481%$679,810$648,015
2009$52,191$47,55973%$742,147$710,028
2010$53,757$44,08766%$768,601$743,536
2011$55,369$40,86959%$744,628$733,598
2019$70,140$51,77159%$838,487$1,019,830
2024$81,312$60,01759%$777,158$1,183,523

The first cut came in 2002. After two losing years the planned withdrawal was 6.19% of the portfolio, over the 6% guardrail, so spending dropped 10%. The market fell again and a second cut followed in 2003. Then 2008 arrived, and the rule cut three years in a row: 2009, 2010, and 2011. The 2009 check found the withdrawal rate at 8.15%.

Five cuts took the retiree's purchasing power from $50,000 to $45,000, $40,500, $36,450, $32,805, and finally about $29,500. From 2011 through 2024 they lived on 59% of the lifestyle they retired with.

The part that should bother you

Compare that to the fixed 4% retiree, who started $10,000 lower.

The guardrails retiree out-spent the 4% retiree for the first nine years. From 2009 on, they spent less, and by 2011 they were living on about $29,500 of purchasing power while the 4% retiree was still spending $40,000. Over the full 25 years, measured in constant dollars, the guardrails retiree spent $870,598 and the 4% retiree spent $1,000,000. The strategy that was sold as a way to spend more delivered $129,402 less.

What the guardrails retiree got in exchange is the balance. They ended 2024 with $1,183,523, which is $406,365 more than the 4% retiree, and their withdrawal rate was back inside the band rather than sitting near 11%. One retiree has a plan that is still healthy at 90. The other had the better 25 years and has a problem coming.

And the prosperity rule never fired. Even in 2021, with the portfolio at $1,221,276, the withdrawal rate was still above 4%, because 21 years of inflation raises had moved the target. This is the least understood feature of the rule: cuts are permanent until the portfolio grows enough to trigger a raise, and after a bad decade that can take a very long time.

A 10% Cut Is Rarely Just 10%

The rule is usually described as "a 10% spending cut in bad years." In a bad decade it is a 10% cut, then another, then another, each one applied to an already reduced paycheck and none of them reversed. Five cuts compound to a 41% reduction in purchasing power. Any guardrails plan should be presented with the cumulative number, not the single-step one.

The Same Rule in Two Other Decades

The 2000 retiree is the hard case. Here is the same $50,000 guardrails plan started in 2008 and in 2010, with the 4% rule alongside. Spending totals are in constant dollars.

Retired inYearsCutsRaisesLowest purchasing powerGuardrails total spendingFixed 4% total spendingGuardrails ending balanceFixed 4% ending balance
2000255059%$870,598$1,000,000$1,183,523$777,158
2008172081%$761,000$680,000$1,417,622$1,666,063
20101501100%$765,000$600,000$1,892,762$2,308,571

The 2008 retiree walked straight into the worst year for stocks since the 1930s and took one cut, in 2009. The recovery came quickly enough that no second cut followed until 2023, after stocks and bonds both fell in 2022. Their purchasing power never dropped below $40,500, which is more than the 4% retiree spent in any year. Over 17 years they spent $81,000 more in constant dollars.

The 2010 retiree never took a cut. The portfolio grew until the withdrawal rate fell to 3.73% at the start of 2022, the prosperity rule fired, and spending rose 10% to $55,000 of purchasing power. The raise arrived just in time for the 2022 decline, which is a fair illustration of what a rule based on last year's balance can and cannot know.

One crash followed by a recovery is what guardrails handle best: a single cut, a faster recovery, and more lifetime spending than a fixed rule would have dared. Two crashes in a decade, with a flat market between them, is what turns the cuts into a ratchet.

Why a 100% Success Rate Is Not What It Looks Like

Guardrails look spectacular in a Monte Carlo simulation. The table below runs the same $1,000,000, 60/40 portfolio over 30 years, from 65 to 95, with no other income, across 10,000 simulated paths. Each row compares the fixed rule and the guardrails on identical return paths. The assumptions are 10% average stock returns with an 18% standard deviation, 5% bond returns with a 6% standard deviation, and 3% inflation.

First-year withdrawalFixed: success rateGuardrails: success rateFixed: median ending balanceGuardrails: median ending balance
$40,000 (4.0%)92.3%100.0%$2,852,000$2,602,000
$45,000 (4.5%)85.0%100.0%$2,225,000$2,214,000
$50,000 (5.0%)75.1%100.0%$1,500,000$1,815,000
$55,000 (5.5%)63.8%100.0%$827,000$1,501,000
$60,000 (6.0%)52.1%100.0%$119,000$1,241,000

A fixed 6% withdrawal is a coin flip. The same 6% start with guardrails did not fail once in 10,000 tries.

That should make a careful reader suspicious rather than relieved. If a rule scores 100.0% at 4% and 100.0% at 6%, the success rate has stopped measuring anything. The guardrails cannot fail in this model because nothing limits how many times they can cut. A path heading for zero is met with a 10% cut, then another, for as long as it takes. The account survives because the spending did not.

The success rate counts whether the balance stayed above zero. It does not count whether the retiree could still pay the bills. For a fixed withdrawal those are the same question. For guardrails they are different questions, and the 2000 replay shows the gap between them: a plan that "succeeded" while paying 59% of what it promised.

So the right way to judge a guardrails plan is not the success rate. It is three questions about the spending path. How many cuts in a poor sequence? How low does spending go? How long before the first raise?

Putting a Floor Under the Cuts

A retiree whose withdrawals pay for groceries and property taxes cannot take five cuts. The standard fix is a floor: a spending level the rule is not allowed to cut below.

A floor has a price, and the replay shows it. Give the 2000 retiree a floor at $40,000 of purchasing power, the level the 4% rule would have paid. The first two cuts happen as before. After 2008, the rule tries to cut and the floor stops it. From 2009 on, this retiree is simply a 4% retiree who spent more in the early years, and they end 2024 with $631,877, drawing 12.9% of what is left.

That is $551,646 less than the guardrails retiree with no floor. The floor protected the lifestyle and gave back most of what the cuts were protecting. There is no version of this where both the spending and the portfolio are safe. A floor is the retiree deciding, in advance, how the pain is split between the two.

Smaller adjustments are not gentler

It is tempting to soften the rule by cutting 5% at a time instead of 10%. In the 2000 replay that produced 12 cuts instead of 5 and left purchasing power at 54% of where it started, lower than the standard rule. Smaller steps leave the withdrawal rate above the guardrail for longer, so the rule keeps firing. The size of the step matters less than the width of the band and where the floor sits.

Guaranteed Income Changes the Arithmetic

The cuts apply to portfolio withdrawals, and most retirees do not live on withdrawals alone.

Consider a household with $36,000 a year of Social Security and a $50,000 withdrawal, for a total budget of $86,000. A 10% guardrail cut takes $5,000 from the withdrawal, which is 5.8% of what the household actually spends. Even the full five-cut sequence from the 2000 replay, which removed 41% of the withdrawal, would have reduced this household's total budget by 23.8%.

That is still a serious reduction. But it explains why guardrails suit some retirees and not others. The more of the essential budget that is covered by Social Security, a pension, or annuity income, the more the portfolio withdrawal is discretionary money, and discretionary money is exactly what a guardrail should be cutting.

This is one more reason delaying Social Security and building a guaranteed income floor pair naturally with dynamic spending. The larger the guaranteed share, the smaller the income gap the guardrails are adjusting, and the easier each cut is to live with.

Other Ways to Build Flexible Spending

Guyton-Klinger is the best known dynamic rule. It is not the only one.

  • Ceiling and floor. Vanguard's dynamic spending approach recalculates spending as a percentage of the portfolio each year, then limits how much it can move in real terms, for example by no more than 5% up or 2.5% down. Adjustments are smaller and happen far more often.
  • Percentage of portfolio. Withdraw the same percentage of the current balance every year. The portfolio can never be exhausted, and income swings as much as the market does.
  • The RMD method. Divide the balance by a life-expectancy factor each year, the way required minimum distributions are calculated. Spending rises as a share of the portfolio with age.
  • Risk-based guardrails. A newer approach replaces the withdrawal-rate trigger with a full re-run of the plan. Instead of asking whether the withdrawal rate has crossed 6%, it asks whether the whole plan, including future Social Security, pensions, and the shorter remaining horizon, has become too risky. Its advocates point out that a fixed withdrawal-rate trigger treats a 62-year-old bridging to Social Security and an 85-year-old the same way, and cuts harder than the plan needs.

That last criticism is fair. A 6% withdrawal rate means something very different at 65 than it does at 88, when the money only has to last a few more years. Guyton and Klinger's own 15-year cutoff on the capital preservation rule was an acknowledgment of this, and it is the piece most simplified versions leave out.

Guardrails, Bucket Strategy, or Bond Tent?

These three defenses against a bad early sequence work on different levers and are often combined.

  • Guardrails protect the portfolio by adjusting the spending.
  • The bucket strategy protects the withdrawals by making sure near-term spending never comes from a depressed asset.
  • The bond tent protects the allocation by holding the least stock exposure in the years a crash hurts most.

In the same January 2000 start, a rising equity glide path with a fixed $40,000 withdrawal ended 2024 at $1,284,052 without a single spending cut. That is not an argument against guardrails. It is a reminder that sequence-of-returns risk can be managed from the investment side as well as the spending side, and that a plan relying on spending cuts alone is asking the retiree to carry the whole load.

How to Set Guardrails Before You Need Them

A guardrails plan is a written policy. It needs five decisions.

1. The starting withdrawal. Higher starts mean more early spending and more cuts later. The 2000 retiree started at 5% and took five. Guyton and Klinger's 5.2% to 5.6% range assumed at least 65% in stocks and all four of their rules.

2. The band. The standard is 20% above and below the starting rate. A narrower band reacts sooner and more often.

3. The adjustment. The standard is 10%. Decide whether it applies to all spending or only to the discretionary part.

4. The floor. Name the spending level the plan will not cut below, and accept in writing that the floor brings back depletion risk. The essential budget minus guaranteed income is the natural place to set it.

5. The review date. Check once a year, on the same date, using the balance on that date. A rule checked whenever the market is frightening becomes market timing.

A Guardrails Policy in One Paragraph

"We start at $50,000 a year and raise it with inflation every January. If that withdrawal is more than 6% of the portfolio on January 1, we cut it by 10%. If it is less than 4%, we raise it by 10%. We will not cut below $40,000 in today's dollars, and if we reach that floor we will meet to discuss other options." Written before retirement, this is a plan. Improvised in a bear market, it is a guess.

Common Mistakes With Guardrails

Quoting the success rate as if it measured safety. A guardrails plan scores 100% because it cuts without limit. Report the spending path alongside it.

Describing the cut as temporary. Nothing in the rule restores a cut. Spending recovers only when the prosperity rule fires, and in the 2000 replay it never did.

Starting high because the rule will fix it. A 6% start with guardrails survives every simulated path and commits the retiree to cuts in most of them.

Applying portfolio guardrails to the whole budget. A 10% cut to withdrawals is a smaller cut to total spending when guaranteed income is in the picture. Present the household number.

Having no floor. If the withdrawals fund essentials, an uncapped rule will eventually ask for a cut the retiree cannot make.

Checking too often. Annual is the design. Monthly checks in a volatile market trigger cuts that a year-end balance would never have required.

Why This Matters for Advisors

Guardrails are easy to recommend and hard to live with, and the difference shows up years after the recommendation is made. A client who heard "you can start at 5% and we will trim a little in bad years" is not prepared for a third consecutive cut.

The conversation that holds up has the cost in it from the start. Show the starting paycheck guardrails allow. Show the balances at which a cut and a raise would happen, in dollars. Then show a poor sequence: how many cuts, how low spending went, and how long it stayed there. A client who has seen the 2000 replay and still chooses guardrails has made an informed decision. A client who has only seen "100% probability of success" has not.

It also matters for the file. A guardrails recommendation is a commitment about future behavior, the client's and the advisor's. The band, the adjustment, the floor, and the review date should be documented when the plan is adopted, so that a cut made in year nine is the execution of an agreed policy and not a judgment call made under pressure.

How RetirementForge Helps

The Monte Carlo Simulation tool runs a plan through 10,000 return paths under either a fixed withdrawal or Guyton-Klinger 20/20/10 guardrails, and when guardrails are selected it reports the fixed-withdrawal result on the identical paths beside it. That is the comparison in the table above. It reports success rates and ending balances, not the year-by-year spending path, so pair the result with a replay like the one in this article when you present it. The Retirement Income Gap Analyzer separates guaranteed income from the spending the portfolio has to fund, which is the number a floor should be built on, and the Social Security Break-Even tool shows what a delayed claim adds to the guaranteed share. The Drawdown Planner lays out each account's withdrawals year by year once the spending level is set.

Every assumption is documented and every client session is captured in an immutable audit trail, so the guardrails a client agreed to before retirement are on the record when the first cut comes due. Get started free and put the policy in front of a client with the numbers attached.


This article is for educational purposes only and does not constitute financial, tax, or investment advice. The historical replay uses published calendar-year total returns for the S&P 500 and the Bloomberg U.S. Aggregate Bond Index, a fixed 3% inflation adjustment, start-of-year withdrawals, and annual rebalancing, applies only the capital preservation and prosperity rules, and ignores fees and taxes. The Monte Carlo figures depend on the stated return and inflation assumptions. Both are illustrative, not predictive, and past performance does not guarantee future results. Consult a qualified financial advisor for guidance specific to your situation.

Frequently Asked Questions

What is a guardrails withdrawal strategy?
A guardrails strategy sets a starting withdrawal and two trigger points around it. Spending rises with inflation each year, but if the withdrawal rate drifts above the upper guardrail the retiree takes a pay cut, and if it drifts below the lower guardrail the retiree takes a raise. It replaces the 4% rule's fixed, inflation-adjusted withdrawal with one that responds to what the portfolio has actually done.
What are the Guyton-Klinger rules?
They are the decision rules published by Jonathan Guyton in 2004 and refined with William Klinger in 2006. The two best known are the capital preservation rule, which cuts spending 10% when the current withdrawal rate rises more than 20% above the starting rate, and the prosperity rule, which raises spending 10% when the rate falls more than 20% below it. Their 2006 paper called these two rules financial guardrails.
How much can you withdraw with guardrails?
Guyton and Klinger concluded that initial withdrawal rates of 5.2% to 5.6% were sustainable at a 99% confidence standard over 40 years for portfolios holding at least 65% stocks, and as low as 4.6% at 50% stocks. That compares with roughly 4% for a fixed, inflation-adjusted withdrawal. The higher starting rate is paid for with spending cuts in poor markets.
How big are the spending cuts with guardrails?
Each cut is 10%, but cuts can stack. In a replay of a retiree who started in 2000 with $1 million and a $50,000 withdrawal, the guardrails triggered five cuts and no raises in 25 years. Spending fell to about 59% of its original purchasing power by 2011 and stayed there, which was less than a retiree who simply followed the 4% rule was spending.
Do guardrails guarantee you will not run out of money?
In a simulation, nearly. Guardrails scored a 100.0% success rate at every starting rate from 4% to 6% in a 10,000-path Monte Carlo run. That is because the rule keeps cutting spending as the portfolio shrinks, so the account almost never reaches zero. The risk does not disappear. It moves from running out of money to living on less, and the success rate cannot see that.
Should guardrails have a spending floor?
Most retirees need one, because essential bills do not shrink 10% at a time. A floor limits how far the cuts can go, and it also brings back some of the depletion risk the cuts were removing. In the 2000 replay, a floor at $40,000 of purchasing power left the retiree with $631,877 after 25 years instead of $1,183,523.
Are guardrails better than the 4% rule?
They are a different trade. The 4% rule holds spending steady and lets the portfolio absorb every shock. Guardrails hold the portfolio steady and let spending absorb the shock. A retiree with flexible spending and solid guaranteed income is well suited to guardrails. A retiree whose withdrawals pay for essentials is not, unless the plan has a floor.

Run the numbers on your own situation

Free, no signup required.

For advisors

Monte Carlo Simulation

Run 10,000 randomized retirement scenarios to estimate the probability of your portfolio lasting through retirement. Visualize outcomes with percentile bands.

Part of RetirementForge, the advisor platform. Requires an advisor account.

Investment & Risk20 min read

Monte Carlo Simulation in Retirement: What a 90% Success Rate Actually Means

Almost every retirement plan is now graded with a probability of success. Almost nobody explains what the number counts, how precise it is, or why a 100% score is usually a warning. Here's how Monte Carlo simulation really works — and how to read the output without being misled by it.

Investment & Risk13 min read

The 4% Rule and Safe Withdrawal Rates: What It Really Means for Your Retirement

The 4% rule says you can withdraw 4% of your portfolio in year one and adjust for inflation each year after, with a high chance the money lasts 30 years. Here's where the rule came from, what it actually guarantees, the assumptions that quietly break it, and how modern planners use dynamic withdrawal rates to spend more safely than a rigid 4% ever allowed.

Investment & Risk16 min read

Sequence-of-Returns Risk: Why the Order of Returns Matters More Than the Average

Two retirees with identical average returns can end up worlds apart, just because of when the bad years arrived. Here's how sequence-of-returns risk works, why it's the most dangerous force in retirement, and the proven strategies advisors use to defuse it.

Investment & Risk21 min read

The Bond Tent and the Rising Equity Glide Path: Why Some Retirees Should Own More Stocks at 75 Than at 65

A bond tent starts retirement conservative and buys back into stocks over the first decade, the opposite of the classic age-in-bonds rule. Here is how it works, a 25-year replay from 2000 showing what it bought, the 2010 retiree who paid for protection they never needed, why 2022 broke the tent's assumption, and how delaying Social Security builds one automatically.