• 👉 Work With Me
• Blanchett: How Spending Evolves in Retirement: A Smile, a Smirk, or Something Else?
• What Is Your Funded Ratio?
• 5 Withdrawal Strategies to Boost Retirement Income (And How High You Can Really Go)
• The 3 Questions That Reveal Your Real Inflation Risk in Retirement

The longer you’re retired, the less you’re likely to spend. 

In fact, a new study tracking thousands of American retirees found that as many as 85% of households spent less than they had ten years earlier, after adjusting for inflation.

That’s almost the exact opposite of what most retirement plans assume. Most financial planning tools project that your spending will rise with inflation every year, often for 30 years or more.

And what I found especially interesting about this new research is that even retirees who could comfortably afford to maintain their lifestyle were spending less as they got older.

In other words, retirees don’t always cut back because they have to. Many do it because they choose to.

And that distinction could have a meaningful impact on your retirement income planning. For example, accounting for how retirees actually spend could support a starting withdrawal rate roughly 20% higher than popular models suggest.

So in today’s episode, I’m walking you through this new research and sharing three specific things:

1. Why spending falls even as the mostly costly retirement expense keeps climbing 
2. Why even well-funded retirees continue cutting back
3. What two very different spending patterns mean for your withdrawal rate and retirement success

Welcome to another episode of the Stay Wealthy Retirement Show. I’m your host, Taylor Schulte, and every week I cover the most important financial topics to help you stay wealthy in retirement. Ok, onto today’s episode.

NEW RESEARCH: Your Plan Is Overestimating Retirement Costs (By 20%)

To appreciate what this new research found — and why it matters — I think it helps to start with the assumption it’s challenging.

When financial planner Bill Bengen developed the famous 4% rule back in 1994, he assumed a retiree would withdraw 4% of their portfolio in year one, and then increase that dollar amount with inflation every single year for 30 years. If inflation ran at 3%, your withdrawals got a 3% raise. Automatically. Every year. No exceptions.

Researchers call this “constant real spending.” And it’s not unique to the 4% rule — it’s built into nearly every financial planning tool and retirement calculator in use today. Punch your numbers into almost any retirement software, and it will project your expenses rising in lockstep with inflation until the very end of your plan.

And honestly, it seems like a perfectly reasonable assumption. As the price of groceries, insurance, and travel rises, we would naturally expect retirees to spend more dollars each year just to maintain the same lifestyle.

Which brings me to the researcher behind today’s study, David Blanchett, who is one of the most widely cited researchers in retirement income planning.

And I’ve shared David’s work on this show before. Most recently, in an episode on retirement withdrawal strategies, I cited a study of his showing that retirees spend about 80% of the guaranteed income that arrives as a paycheck, income from sources like Social Security and pensions. However, when it comes to their investment portfolios, retirees spend less than half of what they could safely withdraw.

Well, Dr. Blanchett just published a new paper, and it asks a deceptively simple question: what actually happens to retiree spending over time?

And what I appreciate about the paper is its honesty. David openly admits that most of his own past research assumes spending increases with inflation — despite the fact that he has spent more than a decade publishing evidence to the contrary. As he puts it in the paper: “change is hard!”

Now, before we get to his findings, here’s what makes them genuinely surprising. If you simply looked at the raw ingredients, you might actually expect retiree spending to rise faster than inflation — not slower.

Let me explain what I mean.

As we age, our spending naturally shifts toward healthcare. To put some actual numbers to it, healthcare represents less than 5% of total spending for households under 35. However, by age 75, that spending triples, to about 15%.

And this shift in spending is so well documented that the government actually tracks a separate inflation rate just for Americans 62 and older. It’s universally referred to as CPI-E, with the “E” standing for Elderly. And as you might expect, medical costs carry meaningfully more weight in that index than they do in the standard inflation number we see in the headlines.

But healthcare’s growing share of the budget is only half the story — it also has a long history of inflating faster than everything else. Specifically, going all the way back to 1957, medical care inflation in the U.S. has averaged about 5.1% per year, while inflation excluding medical care has averaged closer to 3.6% during the same time period.

Put simply, retirees devote more and more of their budget to the single spending category with the fastest-rising prices. Stack those two facts together, and you would likely predict that retiree spending should outpace inflation over time.

But that’s not what the data shows. In fact, it’s not even close.

What Thousands of Real Retirees Actually Do

To see just how different the reality is, let’s start with a simple snapshot of American households.

According to the government’s most recent Consumer Expenditure Survey, households headed by someone between ages 45 and 54 spend more than any other age group — roughly $97,000 per year on average. On the other hand, households headed by someone 75 or older spend $53,000 per year, about 45% less.

Now, a snapshot like that has limitations. The main one is that it compares different households — and different generations — at a single point in time. As a result, it’s hard to tell whether spending truly declines with age, or whether today’s 75-year-olds simply spend less because they grew up more frugal. 

To really answer the question, you need to follow the same households through retirement and watch how their spending evolves. And that’s exactly what David did.

His analysis is built on the Health and Retirement Study — a large, ongoing research project run by the University of Michigan that has tracked the finances, health, and spending of thousands of older Americans for more than two decades now. 

For context, when David published his now-famous research on retirement spending back in 2014, he had five waves of this survey data to work with. Today, he has eleven, spanning from 2001 through 2021. So, all told, this new analysis includes nearly 8,000 observations of real retiree households between the ages of 60 and 90.

And here’s the headline finding from his new analysis: for the typical retiree, inflation-adjusted spending declines steadily throughout retirement. More specifically, over any given two-year window, a little more than half of retirees cut their real spending. But stretch the window out to ten years, and as many as 85% of retiree households experienced a decline.

Now, when you hear “spending declines,” you might assume it means that fewer dollars are leaving retirees’ bank accounts each year, but that’s not what the research is saying.

Most retirees still spend more in actual dollars every year—they just don’t increase their spending as quickly as prices rise, at least according to this research. 

For example, let’s say prices rise by 3%, but your spending only rises by 2%. More dollars are still leaving your checking account than they did last year, but because your spending didn’t fully keep pace with rising prices, you purchased slightly less overall. That’s what researchers mean when they say “real,” or inflation-adjusted spending declined. Put simply, the typical retiree gives their spending a raise most years–that raise just tends to be smaller than the rate of inflation.

And naturally, this leads us to ask “why?” Why exactly does retirement spending gradually decline in real terms year after year? Maybe it’s belt-tightening out of necessity. Or maybe it reflects a natural slowdown, with fewer big trips at 82 than at 68. We’ll dig into that question in a few minutes, because the answer changes what these findings mean for your plan. 

A Smile, a Smirk, or Something Else?

But first, we need to talk about smiles and smirks. Because, while researchers largely agree that real spending does, in fact, decline, the long-running debate is about what happens at the very end of retirement — in your mid-80s and beyond.

To understand the debate, it helps to rewind to 1988, when a financial planner named Michael Stein gave us what is still one of the most memorable ways to think about the arc of retirement. He described it in three phases: the go-go years, the slow-go years, and the no-go years. In the go-go years, retirees travel, spend on hobbies and experiences, and generally don’t consider themselves “old.” In the slow-go years — roughly ages 70 to 84 — health starts to nudge people closer to home, and spending naturally slows. And in the no-go years, lifestyle changes significantly, often driven by end-of-life health issues.

The open question is what happens to spending in those final years, and there are two competing answers.

The first is what David Blanchett famously called the “retirement spending smile.” In his original 2014 research, he found that real spending declines through the middle of retirement and then curls back up at the oldest ages — driven largely by rising end-of-life healthcare costs. When you plot this on a chart, the line looks like a subtle smile.

The second answer comes from two economists at RAND, the well-known nonprofit research institute, who documented real spending declines that persist even at advanced ages. But in their research, there is no upward curl at the end to form a smile. Plotted on a chart, their version looks more like a smirk.

So who’s right? The smile or the smirk? Well, according to this new analysis… both. And the reconciliation is genuinely useful for understanding your own retirement spending decisions.

Let me explain. When Blanchett looked at the median retiree — the household right in the middle of the pack — he found that spending continued to decline at older ages. In other words, it was closer to looking like a smirk. But when he averaged the changes across all retirees, spending ticked back up late in retirement, creating a smile.

So, why would the median and the average tell two different stories? I’ll give you a hint: it starts with an H, it gets more expensive as we age, and it has never been accused of being predictable. Unfortunately, it’s not “hobbies.” If you guessed “healthcare,” you are correct. 

Healthcare is the reason the median and the average tell two different stores. Here’s why. 

Most retirees — i.e., the median experience — most retirees never face a catastrophic healthcare bill. However, a meaningful minority do, and those large, unpredictable expenses pull the average up, even though the typical retiree never experiences them.

Let me put some numbers to this. Using the same national retirement study, David examined cumulative out-of-pocket medical expenses incurred after age 70 — these are the unexpected expenses, like hospital stays, nursing home care, and home healthcare. Among retirees who lived to age 95, the median household incurred about $50,000 in these costs over that entire 25-year stretch. But the unluckiest 5%? They incurred roughly $250,000 in unexpected out-of-pocket medical expenses.

That’s a five-fold difference between the typical household and the unluckiest few. Researchers refer to that small, expensive end of the range the “tail” — a term worth remembering, because we’ll come back to it before we wrap up. And that wide range is also why spending gets harder to predict with age: the year-to-year swings among retirees in their late 80s are about 50% larger than they are for retirees in their early 60s.

So, to sum this all up, when you hear retirement “smile versus smirk,” here’s the practical translation: the data shows that a typical retiree’s real spending does decline throughout retirement and ends up looking more like a smirk. However, a minority of retirees get hit with large healthcare shocks late in life creating a smile — and your plan needs to acknowledge that possibility.

Is It Because They Have To — or Because They Want To?

But before I share how to plan for that possibility properly, I want to first address an important question I flagged earlier — and honestly, the part of the paper I found most compelling. 

If real spending declines throughout retirement, is that because retirees are forced to cut back? Or is it because they choose to?

Now, at first glance, the answer may seem obvious. After all, according to a survey from the National Institute on Retirement Security, 79% of Americans agree that the country is currently facing a “retirement crisis.” 

But if retirees are spending less because they are genuinely running out of money, then building those declines into a retirement plan would be a serious mistake. You would essentially be planning to fall short.

The picture changes, however, if retirees with more than enough money are also reducing their spending over time. That would suggest the decline is not driven entirely by financial necessity. At least part of it may simply reflect how people choose to live as they age.

To test that, David used something called a “funded ratio” — a concept borrowed from the world of pension plans. In plain English, a funded ratio compares two numbers to determine how healthy a retirement plan is.

1.) The first number is everything you have — your total assets, plus future income like Social Security.

2.) The second number is everything your current lifestyle is expected to cost over the rest of your life.

If those two numbers match, your funded ratio is 1.0 — meaning you have just enough to fund your current lifestyle through retirement. If it’s above 1.0, you have a surplus. And if it’s below 1.0, you would be facing a shortfall.

He then sorted thousands of retiree households into five groups, from “very underfunded” to “very overfunded,” and watched what happened to their spending.

Really quick, one sobering note before the main finding: only about 65% of the retirees in the study had a funded ratio of 1.0 or higher. In other words, roughly one in three were spending at a level their resources couldn’t fully sustain. For perspective, David notes that if a 4% withdrawal rate is roughly optimal, the most underfunded households were spending at a pace closer to 8% — while the most overfunded households were spending closer to 2%.

Ok, so with that context in mind, let’s get to the main finding. Remember, the question we’re trying to answer is whether retirees cut back because they have to — or because they choose to. And the way to answer it is to watch what each of the five funded groups did with their spending. Let’s start at the bottom of the ladder.

The most underfunded retirees at the bottom of the ladder cut their real spending by about 7.5% per year on average — deep, repeated spending cuts that look a lot like necessity. No surprise there really.

But here’s what’s interesting: retirees who were adequately funded still reduced their real spending — by about 3% per year on average. Even the overfunded group, people with a clear surplus, still trimmed their spending. 

The only group to actually increase their inflation-adjusted spending was at the top of the ladder, the VERY overfunded group. And the increase wasn’t much, only about 1% per year — nowhere close to what their resources would have allowed.

There’s also one more pattern worth highlighting, especially because I suspect it applies to many of our listeners. When David grouped households by how much they spent, every group spending $80,000 or more per year reduced its inflation-adjusted spending, regardless of how well-funded it was. In fact, the higher a household’s spending, the more likely they were to reduce their spending as they aged. In other words, the retirees who look like many of the listeners of this show — diligent savers, comfortable spenders, well-funded plans — are precisely the people whose spending declines appear to be driven most by choice.

And if that finding sounds familiar, it should, because it’s consistent with everything we’ve covered on this show about the underspending problem in retirement. The Federal Reserve study I shared in a prior episode found that retirees, on average, die with nearly twice as much savings as they had when they retired. 

And I get it. After 30 or 40 years of diligent saving, flipping the switch from accumulating to spending is genuinely hard. There’s the fear of a long-term care event. The fear of a market crash early in retirement. The fear of simply not knowing how long we’ll live and how long the money needs to last. Those fears are valid — and they’re exactly why the final piece of this research matters so much, which is what all of this means for your retirement income plan.

What This Means for Your Withdrawal Rate

To start, if retirees don’t actually increase spending with inflation every year, then plans that assume they do are systematically overestimating the cost of retirement. And if you’re overestimating the cost, you’re likely underestimating what you can safely spend — especially in the early go-go years when the money arguably delivers the most joy.

To quantify this, David modeled a hypothetical retiree using a framework from his earlier research. This retiree collects $30,000 per year in Social Security and has a $60,000 total annual spending goal — $40,000 of it essential, and $20,000 more flexible. Their portfolio is a conservative mix of 40% stocks and 60% bonds, and retirement is projected to last 30 years.

With the retiree profile documented, he then estimated the highest sustainable starting withdrawal rate under three different spending assumptions: constant inflation-adjusted spending, the spending smile, and the spending smirk.

Under the traditional constant spending assumption — with a moderate level of risk — the model supported a starting withdrawal rate of about 5.2%. Not bad compared to the outdated 4% rule that many still follow.

However, under the spending smile and smirk assumptions, the starting withdrawal rate jumped to 6.2% and 6.4%, respectively.

That’s roughly 20% more starting income, simply from modeling spending the way retirees actually behave rather than the way our software assumes they behave.

On a $2 million portfolio, that’s the difference between $104,000 and as much as $128,000 of portfolio income in the first year of retirement. Same nest egg. Same market and planning assumptions. The only change is replacing an assumption the data doesn’t support with one that it does.

Now, two quick notes before anyone runs off and cranks their withdrawal rate up to 6.4% after listening to this.

1.) First, you may have noticed that even the “traditional” scenario came in at 5.2% — well above the 4% rule. That’s because David’s model isn’t the rigid, set-it-and-forget-it approach from the original 4% rule research. It dynamically adjusts spending over time and distinguishes between essential and flexible expenses. In other words, flexibility is doing real work in all three of those numbers. If you want or need perfectly stable, never-adjusting income, your safe starting rate needs to be lower.

2.) Second, notice how close the smile and smirk numbers are to each other: 6.2% versus 6.4%. After years of debate about the precise shape of retirement spending, the two shapes support nearly identical withdrawal rates. In other words, moving beyond the constant spending assumption matters far more than which specific declining-spending model you choose.

If you’ve been listening to this show for a while, this conclusion should feel familiar. It’s the same principle behind the guardrails and dynamic withdrawal strategies I covered in my episode on boosting retirement income. When your plan is allowed to adapt — to markets, and, in this case, to how your spending naturally evolves — the evidence consistently supports a higher starting withdrawal rate and more lifetime income.

Before We Wrap Up: Two Important Caveats

Now, before we wrap up, I want to briefly address two caveats — because I can already hear a thoughtful listener pushing back and saying: “But Taylor, if I build declining spending into my plan and I’m one of the unlucky ones who needs long-term care at 88, haven’t I just planned my way into a problem?”

It’s the right question, and it’s why the healthcare tail we discussed earlier deserves its own line item in your plan. Remember the numbers: the median retiree who lived to 95 incurred about $50,000 in unexpected out-of-pocket medical costs after age 70, but the unluckiest 5% incurred roughly $250,000. 

A well-built plan lets you spend confidently through the go-go years precisely because it has set money aside for that tail. For some households, that means long-term care insurance — if you can qualify and stomach the premiums. For others, it means earmarking a dedicated pool of assets for late-in-life care. 

What you shouldn’t do is let an unquantified fear of the worst case push you to underspend by hundreds of thousands of dollars across your healthiest years. Instead, quantify the risk, fund it, and then give yourself permission to actually use the rest.

The second caveat I wanted to touch on is about inflation. Nothing in this research says inflation doesn’t matter — it absolutely does, and a few episodes ago I walked through the three questions that reveal your personal inflation risk in retirement, which I’ll link to in the show notes. But this research does soften the problem in one important way: if your real spending is likely to decline with age, your portfolio doesn’t need to be a perfect, dollar-for-dollar inflation hedge. And thankfully, most retirees already own the best inflation hedge available — Social Security, which is explicitly adjusted for cost of living increases every year. To sum it up, the evidence suggests your portfolio needs to be inflation-aware, not inflation-obsessed.

The Bottom Line

As we all know, retirement planning is full of assumptions — about markets, lifespan, taxes, and more. But one assumption almost nobody questions is the idea that your spending will rise with inflation for 30 straight years. A growing body of research — now including one of the most robust studies to date — suggests that for the typical retiree, it simply doesn’t.

But the goal of all this research — and of good retirement planning in general — isn’t to motivate people to spend recklessly. It’s to replace vague fears with measured risks, and rigid assumptions with evidence, so that every dollar you spend — and every dollar you hold back — is a decision you actually made, not a default your planning software made for you.

Thank you, as always, for listening. And to view David Blanchett’s full paper, along with the other research and resources referenced in today’s episode, just head over to youstaywealthy.com/295.

Disclaimer

This podcast is for informational and entertainment purposes only, and should not be relied upon as a basis for investment decisions. This podcast is not engaged in rendering legal, financial, or other professional services.