How to Safely Spend More Money in Retirement

If you’re within 5-10 years of retirement, or already drawing from your nest egg, and still using the 4-percent rule, it’s past time to upgrade your plan.

Here’s why and how.

What Is the 4-Percent Rule?

Created by financial planner William Bengen in the 1990s based on historical stock and bond returns, it works like this:

• Invest your nest egg 50/50 between large-cap US stocks and US bonds.
• For Year 1 of your retirement, withdraw 4 percent of your portfolio to fund your expenses.
• For Year 2, adjust your Year-1 withdrawal to account for inflation.
• Rinse and repeat each year thereafter.

Simple, straightforward, and from the 1920s to the 1990s, there was not a single 30-year period where this would have caused you to run out of money in retirement.

What Are the Main Problems with the 4-Percent Rule?

This “rule,” or more accurately, withdrawal strategy, has several significant drawbacks.

1. Monte Carlo simulations using more recent market assumptions show about a 13 percent chance of failure for a 30-year retirement.
2. If your retirement is longer than 30 years, e.g., if you retire very early, your risk of failure grows.
3. Unless you retire at exactly the wrong point in time, your net worth will likely grow far above where you started. This means you will not enjoy as good a retirement as you can afford to.

What’s a Better Strategy?

Introduced by financial planner Jonathan Guyton and computer scientist William Klinger in 2006, the “Guardrails Approach” is a dynamic retirement withdrawal strategy that aimed to address the shortcomings of static approaches such as the 4-Percent Rule.

The classic version of the Guyton-Klinger Guardrails Approach works like this:

• Invest your nest egg in a prudently diversified fashion.
• For Year 1 of your retirement, withdraw a reasonable fraction of your portfolio’s value, say 5.2 percent.
• For Year 2, adjust the prior year’s withdrawal amount to correct for inflation.
• Check your new withdrawal rate (new dollar amount divided by current portfolio balance). If it’s 20 percent above your starting rate, cut spending by 10 percent. If it’s 20 percent below, raise spending by 10 percent.
• Rinse and repeat each year thereafter, until you enter the last 15 years of your retirement, at which point you stop bumping your draws down, no matter what your portfolio does, because you’re beyond the “sequence of returns risk” danger zone.

For example, say you start out with a $1 million portfolio and initially draw 5 percent, or $50,000, and for simplicity, we’ll neglect inflation. At the start of the following year, if your portfolio is worth over $1.25 million, that same $50,000 would be less than 4 percent, which is 20 percent lower than the initial 5 percent draw rate, so you’d bump your draw up 10 percent to $55,000. This lets you enjoy in the present the benefits of strong returns.

On the flip side, if your portfolio dropped to under $833,333, making the $50,000 over 6 percent of the portfolio’s new value, you’d chop your withdrawal by 10 percent, to $45,000. This reduces the number of shares you need to sell at lower prices.

What Are the Benefits of the Guardrails Approach?

This dynamic approach provides important benefits relative to the static 4 percent rule.

1. It lets you start with a higher initial draw with a lower risk of failure than a static approach, allowing you to spend more over a 30-year retirement.
2. People naturally tend to reduce spending when their portfolio value drops and increase spending when it soars.

Ben Simerly, CFP, Financial Advisor and Founder of Lakehouse Family Wealth, uses guardrails frequently, “As we primarily work with near/pre-retirees and recently retired people, income strategies are a daily topic of discussion with clients. At a high level, we use a guardrail-type strategy with every client. A Guyton-Klinger guardrail-type strategy improves outcomes in almost all cases. The key is that it can improve not only the likelihood of success, ensuring money lasts throughout retirement, but also increase overall income.

“We can account for both current income needs and the need to keep up with and possibly surpass inflation. My favorite aspect of guardrails is that they amplify the success of the portfolio’s underlying components. And when you get down to it, our job as advisors is to provide the client with as many opportunities for success as possible.”

Jordan Gilberti, Founder and Financial Planner at Sage Wealth Group, also likes dynamic guardrails. “I discuss dynamic withdrawal strategies with all of my clients who are nearing retirement. It’s an incredibly effective approach to retirement distribution planning, and it works well with clients because they know their retirement spending is not static and that markets are highly unpredictable. The value here is flexibility, but it requires discipline and monitoring, which is not a fit for everyone.

“The Guyton-Klinger Guardrails approach can feel intuitive since it ties spending to changes to the performance of your investment accounts, while risk-based guardrails focus more on probability and sustainability of one’s portfolio. The tradeoff is simplicity vs. precision, and neither approach eliminates the need for ongoing monitoring and review.”

Dr. Steven Crane, Founder of Financial Legacy Builders, agrees, “I’m a fan of dynamic withdrawal strategies, but not because the math is perfect. I like them because they respect human behavior. Most middle-class retirees don’t fail because their spreadsheet was wrong; they fail because fear or guilt causes them to underspend early or panic later. A flexible approach gives people permission to spend when life is good and pull back when it’s not, which is far more realistic than telling someone to blindly take the same dollar amount every year, no matter what’s happening in their life or the market.”

Brennan Decima, Owner, Decima Wealth Consulting, is also a big fan, “I absolutely recommend dynamic withdrawal strategies with my clients. Many of my clients are accustomed to receiving bonuses during their working years. With dynamic spending plans, we help our clients understand the bonus that the market has given them and what their safe amount of additional spending can be. If clients want to save their ‘bonus’ for a rainy day, we really want to determine what they actually consider a rainy day.”

But there’s a catch that most people, even some pros (present company excluded 😊), don’t see coming.

Criticism of the Guyton-Klinger Guardrails Framework

The Guyton-Klinger approach is a huge improvement over the static 4-percent rule. But it’s far from perfect, as detailed by Derek Tharp, PhD, and Justin Fitzpatrick on Kitces.com, “… this strategy … can result in sharp reductions in retirement income that would be unfeasible for some retirees. Additionally, these income reductions tend to overcorrect for market losses, meaning that far more capital is often preserved than necessary at the cost of severe reductions in the retiree’s standard of living.”

The authors back-tested the Guyton-Klinger Guardrails over the past century or so and found that even if retirees start with a relatively tame 4.3 percent (14 percent smaller than a more typical 5 percent initial draw), four specific periods result in dramatic income cuts (in increasing order of severity).

• Retiring in 2007, through the Global Financial Crisis, would have resulted in a 28 percent drop in income a few years into retirement, staying at that low for several years, and spending most of the first 15 years of retirement with income below the initial draw.
• Retiring in 1999, through the Dot-Com Bubble, would have resulted in an almost immediate set of cuts, totaling a 36 percent drop in income, staying at that low for several years, and spending most of retirement with income lower than the initial draw.
• Retiring in 1936, during the Great Depression, would have resulted in an almost immediate set of cuts, totaling a 45 percent drop in income (!), staying at that low for years, and spending most of retirement with income lower than the initial draw.
• Retiring in 1965, during the Stagflation Era, would have resulted in an almost immediate set of cuts, totaling a 54 percent drop in income (!), staying for years under half the initial draw, and never recovering to the initial draw.

They also note that another author found that a retirement starting in 1966 would have resulted in a maximum 59 percent cut in income, while a retirement starting in 2000 would have suffered a 50 percent cut. 

Admittedly, these periods were outliers, but the authors note that retirement researcher Wade Pfau published results of a Monte Carlo simulation with a Guardrails approach starting at 4.8 percent draw that showed the median scenario (i.e., one that’s better than half and worse than the other half of scenarios) led to cuts ending at 36 percent below the initial draw by the end of a 30-year retirement.

Most retirees would find such cuts unacceptable. Ironically, rather than resulting from aggressive draws, they stem from overly conservative rules.

A Better Guardrails Approach

To address the above shortcomings, the authors suggest a different method of calculating the necessary draw corrections, using risk scores rather than current portfolio values.

Specifically, they suggest starting with a draw rate that would lead to an 80 percent likelihood of “success,” where that’s defined as the probability of ending retirement with a positive balance (even $0.01 qualifies).

Each year, the probability of success is recalculated, and if the likelihood of success reaches 100 percent (because the portfolio has grown so much), one would increase spending to whatever level would return that likelihood to the initial 80 percent.

On the other hand, if the likelihood of success drops to 25 percent (!), one would reduce the draw to a level with a 45 percent likelihood of success. 

At first glance, a 75 percent failure probability sounds terrifying. However, as the authors state, the whole success/failure terminology is misleading, since the strategy will, by definition, modify draw levels to ensure success. 

Thus, it’s acceptable to wait until the probability of success falls that much before cutting spending. In plain English, you don’t cut spending just because the market had a bad year. You cut only when your long-term plan is genuinely at risk.

An analysis of income levels for retirements starting at the same four periods mentioned above, using the same 4.3 percent initial draw, shows:

• Income drops by at most 3 percent vs. 28 percent for the Global Financial Crisis, with most of the initial 15 years of retirement spent with higher income.
• The Dot-Com Bubble start would have resulted in no cut at all, with income rising far above the initial level from about 15 years into retirement.
• The Great Depression, the worst period for US investments in well over a century, would have led to cuts of at most 8 percent, rather than 45 percent.
• The Stagflation Era would have resulted in cuts of up to 32 percent, but that’s in place of 54 percent, and would have only lasted a few years instead of through the entire retirement period. Comparatively speaking, while not great, this is a far better worst-case scenario.

The goal isn’t to slash your lifestyle every time markets wobble. It’s to make small, intentional budget adjustments, but only when you need them to avoid catastrophic cuts later.

Finally, the authors evaluated the portfolio size for retirees starting retirement in 2000. They found that by 2023, the original Guardrails Approach with a 4.3 percent initial draw would have led to a portfolio value that’s 59 percent higher than at the start of retirement, while their revised approach would have dropped by 29 percent relative to the starting balance.

The remaining balance, throughout the first 23 years analyzed, would have been lower for the new approach compared to the initial Guardrails strategy. This is a feature, not a bug. Avoiding excessive cuts naturally leaves a smaller remaining balance.

This is ok if leaving a large bequest is not a priority. Obviously, if it is a priority, you can set the upper guardrail target at, say, 90 percent instead of 80 percent, and the lower guardrail at a success likelihood level of, say, 50 percent instead of 25 percent. Such changes would almost certainly result in worse income cuts, but would also leave more for your heirs.

But does all this work for most people? 

Crane isn’t sure. “From a psychological standpoint, guardrails work because they create boundaries, not because they optimize returns. The Guyton-Klinger approach feels more intuitive to real people because it ties spending adjustments to portfolio reality, not abstract probabilities. Risk-based guardrails make sense on paper, but many everyday retirees don’t emotionally connect with percentages and Monte Carlo outcomes. If someone doesn’t understand the rule, they won’t follow it when emotions are high, and that’s when plans usually break.”

Decima also points out a potential psychological problem with guardrails: “The single biggest challenge with guardrails is that the majority of people we work with love the option to spend more in good times, but really don’t want to consider reducing their spending in down times. Wealth is meant to be a tool to make our quality of life better. If every headline gives retirees anxiety that their spending will have to be cut, it becomes very difficult to stick with a plan.”

Would Retirement Income Jump Up and, More importantly, Down a Lot?

With the original Guardrails approach, the answer depends heavily on the market era. When markets behave themselves, you wouldn’t expect many cuts.

However, as pointed out by Tharp and Fitzpatrick, there were multiple periods when this approach would have hit the upper guardrails repeatedly, leading to massive income cuts. That’s why they proposed their risk-based modification.

The table below compares the depth of maximum income cuts for the two guardrail approaches in four problem periods.

While no cut is pleasant, all but the worst case here are manageable, and even the worst is temporary and bearable.

Discretionary Budget Size: Your Secret Lever to Higher Safe Initial Draws

The thing that determines how aggressive you can be with your initial draw level is the discretionary fraction of your retirement budget (think travel, entertainment, eating out, gifts, charity, etc.). 

Retirement research shows a significantly higher safe withdrawal rate if a large fraction of your retirement budget is discretionary. This is intuitively clear – If your discretionary spending comprises 50 percent of your budget, you can survive a much deeper income cut than if that fraction is just 5 percent.

Two things can make this even better. 

First, assuming you’re drawing from a tax-deferred retirement account (e.g., a traditional IRA), every dollar you reduce from your spending will result in more than a dollar lower draw since you don’t have to pay taxes on that dollar of avoided draw. Assuming your overall marginal tax rate is, say, 20 percent, a dollar lower spend means $1.25 less needed to be drawn. 

Second, having non-portfolio income means that a 10 percent cut in what you draw from your portfolio would result in a smaller cut in your overall retirement income that year.

Adding Buckets to Mitigate Market Loss Risk

But even with smarter guardrails, there’s still one problem left – what do you actually sell in a bad year?

This can be addressed by the so-called “Bucket System.” 

Here’s how I apply that system.

1. Hold enough cash to cover 2 years’ worth of draw needs. This isn’t necessarily the same as 2-3 years’ worth of spending, because most retirees have at least some non-portfolio income (e.g., Social Security, annuities, rental income, part-time work, etc.). Assuming a 5 percent initial draw, this equals about a 10 percent cash allocation.
2. Hold enough bonds to cover another 3 years’ worth of draws (possibly including international bonds to reduce the risk of rising domestic interest rates, but accepting the risk of those foreign markets experiencing increasing rates). With the same 5 percent draw level, this is another 15 percent of your portfolio.
3. Hold the remainder in diversified stock funds (US and international) and any other growth assets I understand and would be comfortable holding through a downturn. This risk or growth bucket would be 75 percent of the portfolio.

Spending comes out of the cash bucket, which partially depletes it, so I need to refill it. This refill comes out of whichever bucket is highest relative to its initial allocation. 

For example, say the growth bucket balance went up 15 percent, the bond bucket balance increased by 5 percent, and the cash bucket balance dropped by 50 percent (neglecting interest income, having spent half of the initial 2 years’ worth of draw). If we started from a $1 million portfolio, the initial amounts were $750k growth, $150k bonds, and $100k cash. At the end of this hypothetical year, the new balances would be $862.5k growth, $157.5k bonds, and $50k cash, for a total portfolio value of $1.07 million, and allocations of 80.6 percent, 14.7 percent, and 4.7 percent, respectively.

Assuming we don’t need to change the $50k overall draw (i.e., the risk-based guardrails didn’t activate), we need to bring the cash bucket back up to 2 years’ draw, or $100k, so we need to add $50k. The bond bucket, to return to 3 years’ worth of draws, has to return to $150k, so it can shed $7.5k. The growth bucket supplies the remaining $42.5k, dropping to $820k, a 76.6 percent allocation.

Let’s look at a less rosy hypothetical year next. Say the growth bucket crashes 25 percent, the bond bucket increases by 3 percent, and the cash bucket drops by the 50 percent we spend. The resulting balances would be $562.5k growth, $154.5k bonds, and $50k cash. The total balance is $767k, for allocations of 73.3 percent (growth), 20.1 percent (bonds), and 6.5 percent (cash).

We need $50k to top off the cash bucket, but selling $50k of the depressed growth bucket would deplete it by nearly 9 percent, rather than the 6.7 percent it would have taken had the growth bucket stayed flat. Thankfully, the bond bucket is up, so we take $50k from there. This drops the bond bucket to just over 2 years’ worth of draws, but that’s better than either allowing the cash bucket to fully deplete in the coming year or selling a much larger fraction of the remaining growth assets. Once the growth bucket recovers, and before the bond bucket fully depletes, we’ll refill the latter from the former.

But what if both the growth bucket and the bond bucket drop? In that scenario, we may need to trim discretionary expenses (the risk-based guardrails may well require this). We can also refill the cash bucket minimally, so it doesn’t fully deplete, and refill it more the following year from the growth or bond bucket, depending on which one performed better by then.

Since we’re spending from the cash bucket, that should not be impacted by market developments. However, if the growth and bond buckets crash massively, it’s plausible to trim spending mid-year.

What do the pros say about buckets?

Decima says, “Most people have four components to their financial picture when they are working. They have a salary that pays the bills, a bonus they can enjoy or save for a rainy day, an emergency fund for unexpected expenses, and a retirement account for the future. When one pot of money is expected to perform all four tasks, it can be nerve-racking. In my experience, buckets are the best opportunity to align portions of the pot for each task. This allows retirees to know exactly what purpose each part of the plan serves.”

Crane also sees benefits. “Buckets aren’t about chasing returns; they’re about buying peace of mind. When retirees can clearly see which money is for ‘now,’ ‘soon,’ and ‘later,’ they make better decisions and sleep better at night. For middle-income retirees especially, buckets reduce the urge to overreact during market downturns because they know next month’s groceries aren’t tied to today’s headlines. The best retirement plans don’t just work financially; they work emotionally, and buckets are one of the simplest ways to make that happen.”

Gilberti sees a lot of value too. “Buckets work best as a behavioral tool that helps retirees visually separate short-term spending from long-term growth buckets. When designed with intention, buckets can reduce stress during periods of market downturns. If the markets are down 30 percent, but as an example, you have 2 years in cash that isn’t tied to the volatility of the market at any given time, that can be a massive relief when undergoing stress from the markets and news headlines.”

Simerly especially likes guardrails when combined with other systems, such as buckets. “The real power of guardrails and many other retirement strategies comes into play when you combine approaches. For example, combining a guardrails approach with buckets for different spending timelines. If we can help smooth out portfolio dips and use guardrails and buckets together, clients see far less volatile income. If guardrails are the brain, then buckets become our backbone. If a client communicates that they are seriously worried about a recession lasting for, say, 3 years, so they don’t want more aggressive investments, we can address that concern using a short-term bucket holding 3 years’ worth of income. 

“In good times, we can pull from the more aggressive investments that are doing well, in line with the income numbers calculated by the guardrails. In bad times, we can pull from cash or the short-term bucket, so the more aggressive bucket can recover.”

What Does This Look Like in Practice? (Your Annual Checklist)

To turn all the above into concrete action steps to take once a year:

1. Update your growth, bond, and cash balances.
2. Use a planning tool (ideally Monte Carlo) to recalculate your plan’s success probability. If you don’t have one, ask your advisor or use a reputable retirement calculator. However, keep in mind that these will be estimated probabilities, not guarantees.
3. If needed, adjust your draw (and thus your spending) per the above-described risk-based guardrails approach. However, run a sanity check to ensure the new draw still covers at least all your “needs” spending.
4. Update your non-portfolio income so you can more accurately determine how much you need in your cash and bond buckets.
5. Refill your cash bucket by rebalancing from your growth and/or bond buckets if those performed well, and defer full replenishment of the cash bucket if both growth and bond crashed. Note that you don’t need to slavishly rebalance to an exact percentage, just close to where you started.

Several additional steps may require some help, including:

• Calculating Required Minimum Distributions (RMDs), once they apply to you.
• Strategize (legal) tax minimization steps and estimate your resulting taxes.
• Decide on the optimal time to claim your Social Security retirement benefits.
• Identifying large unbudgeted expenses, if any (e.g., healthcare, long-term care, housing changes, etc.), and modifying your overall retirement plan accordingly. Large health or long-term care expenses aren’t handled by withdrawal strategies. They require insurance, reserves, or separate planning.
• Estate planning.

The Bottom Line

The venerable 4 percent rule is far riskier than most realize. You can reduce this risk and still start with higher withdrawals by using a dynamic approach such as the Guyton-Klinger Guardrails method.

However, redefining guardrails around plan risk instead of portfolio swings avoids potential unacceptably large cuts in retirement spending without materially increasing the risk of failure.

If you’re concerned about the potential impact of periods of high inflation, that’s why many retirement portfolios keep a large growth allocation, which historically returned about 7 percent above inflation over long periods.

What if you retire straight into a bear market (also known as “sequence of returns risk”)? While clearly not an optimal situation, that’s exactly what the risk-based guardrails and Buckets System are designed to address.

For obvious reasons, I can’t give you “the number” for your initial safe withdrawal rate. However, for most households using guardrails, somewhere in the range of 4-6 percent is common.

Keep in mind, though, this framework (as would be the case for any system) only works if you actually follow it. Continuing to spend more than your plan allows will likely sink your finances.

Finally, using the bucket method, in concert with the above dynamic withdrawal strategy, lets you avoid selling too much in depressed assets when your growth and/or bond holdings crash, so your portfolio recovers more easily from bear markets.

Disclaimer: This article is intended for informational purposes only, and should not be considered financial advice. You should consult a financial professional before making any major financial decisions.