Skip to Content
Personal Finance

William Reichenstein: Avoiding Tax Headaches in Retirement

The retirement-planning expert discusses tax-efficient withdrawals, Social Security-filing decisions, and how to dodge the 'tax torpedo.'

Listen Now: Listen and subscribe to Morningstar's The Long View from your mobile device: Apple Podcasts | Spotify | Google Play | Stitcher

Our guest on the podcast today is William Reichenstein. Dr. Reichenstein is an author and emeritus professor of finance at Baylor University. He is also head of research for Social Security Solutions and Retiree, Inc., where he and William Meyer have developed software to facilitate tax-efficient Social Security and asset-drawdown strategies for retirees. His recent books include Income Strategies and Social Security Strategies with William Meyer. A fourth edition of that book will be forthcoming soon. He has also published more than 200 articles in various financial journals. He earned his B.A. in economics from St. Edward's University and his Ph.D. from the University of Notre Dame. He is also a CFA charterholder.


Income Strategies: How to Create a Tax-Efficient Withdrawal Strategy to Generate Retirement Income, by William Reichenstein

Social Security Strategies: How to Optimize Retirement Benefits, by William Reichenstein and William Meyer

Tax-Efficient Withdrawal Strategies

"Pay Attention to Marginal Tax Rates and Not Tax Brackets," by William Reichenstein,, Sept. 28, 2021.

"Tax Considerations for Relatively Wealthy Households," by William Reichenstein, Journal of Financial Planning, December 2021.

"How Social Security Coordination Can Add Value to a Tax-Efficient Withdrawal Strategy," by William Reichenstein and William Meyer,, Fall 2021.

"Saving in Roth Accounts and Making Roth Conversions Before Retirement in Today's Low Tax Rates," by William Reichenstein,, July 2020.

"Investment Implications of the Rising and Falling Pattern of Marginal Tax Rates for Retirees," by William Reichenstein and William Meyer,, Summer 2020.

"A Comparison of the Tax Efficiency of Decumulation Strategies," by Greg Geisler, Bill Harden, and David Hulse,, March 2021.

"Advice for Married Couples When One Spouse Will Die Years(s) Before the Other Spouse," by William Reichenstein and William Meyer,, January 2021.

"Using Roth Conversions to Add Value to Higher-Income Retirees' Financial Portfolios," by William Reichenstein and William Meyer,, February 2020.

"The Asset-Location Decision and Related Topics," by William Reichenstein,, June 1, 2020.

"Basic Truths About Asset Allocation: A Consensus View Among the Experts," by William Reichenstein,, July 2019.

Social Security, Medicare, and Taxes

"How to Prepare for a No-to-Low 2021 Social Security COLA," by William Reichenstein,, July 27, 2020.

"The Relationship Between Wealth and Delaying Social Security Benefits," by William Reichenstein,, June 2020.

"Social Security Claiming Strategies for Singles and Their Implications for Couples," by William Reichenstein and William Meyer,, May 2021.

"A Top Social Security Expert Blasts Biden's Reform Plan," by Jane Wollman Rusoff,, June 2, 2020.

"Understanding the Tax Torpedo and Its Implications for Various Retirees," by William Reichenstein and William Meyer,, July 2018.

"How Retirees Can Avoid the 'Tax Torpedo'," by Mark Miller,, Feb. 9, 2022.

"Minimizing the Damage of the Tax Torpedo," by William Reichenstein,, September 2021.

"Medicare and Tax Planning for Higher-Income Households," by William Reichenstein and William Meyer,, Winter 2019.

"How Retirees Can Avoid Higher Medicare Premiums," by William Reichenstein,, June 14, 2018.

Retirement Income/Other

"There Is No Perfect Way to Protect Your Investments Against Inflation," by Anne Tergesen,, Feb. 14, 2022.


Christine Benz: Hi, and welcome to The Long View. I'm Christine Benz, director of personal finance and retirement planning for Morningstar.

Jeff Ptak: And I'm Jeff Ptak, chief ratings officer for Morningstar Research Services.

Benz: Our guest on the podcast today is William Reichenstein. Dr. Reichenstein is an emeritus professor of finance at Baylor University. He is also head of research for Social Security Solutions and Retiree, Inc., where he and William Meyer have developed software to facilitate tax-efficient Social Security and asset-drawdown strategies for retirees. His recent books include Income Strategies and Social Security Strategies with William Meyer. A fourth edition of that book will be forthcoming soon. He has also published nearly 200 articles in various financial journals. He earned his B.A. in economics from St. Edward's University and his Ph.D. from the University of Notre Dame. He is also a CFA charterholder.

Dr. Reichenstein, welcome to The Long View.

Reichenstein: Thank you.

Benz: So, just to get the ball rolling, how did you get started in this topic of Social Security-claiming and tax-efficient drawdown strategies? What drew you in?

Reichenstein: I started off with the Social Security-claiming and looked at that. And then, just with the taxation of Social Security benefits and the Medicare premiums—those spikes in Medicare premiums—looking at the marginal tax rates, I tried to bring all that in and at that point, combining those two ideas. And to me, it just affects so many people, and these are topics that most people have not thought about—how the taxation of Social Security benefits, for example, and the huge spikes in Medicare premium, how they should impact tax-efficient withdrawal strategies.

Ptak: And I think we're going to get into each one of those topics a little bit later in the conversation. But to get the conversation started even further, we thought it might be helpful to discuss the key items that are in pre-retirees' toolkits as they think about their investments and how they're positioned. There's the type of investments they keep inside each of their accounts. There's also the sequence with which they draw upon those accounts. And there's withdrawal rates, Social Security filing, and filing for Medicare. Are we missing any other important variables? Is there anything that we left out from that list?

Reichenstein: Yes, in this sense, and when you're talking about withdrawal strategies, the important thing is what's the marginal tax rate, which is the tax rate on the next dollar of income. And what happens for a typical retiree is their marginal tax rate goes from their tax bracket to 150% of their tax bracket, to 185% of their tax bracket for a wide range of income due to the taxation of Social Security benefits, and then there's a sharp drop back to the tax bracket. And then, if your income continues to rise for wealthier clients, there are five huge spikes in marginal tax rates due to those income-based Medicare premiums called IRMAAs, or Income Related Monthly Adjustment Amounts. And to figure out how to tax-efficiently withdraw money in retirement, you need to bring in that up and down, that roller coaster ride of marginal tax rates.

Benz: Do you think people often focus too much on withdrawal rates and how their investments are positioned and perhaps not enough on how taxes fit in?

Reichenstein: Yes, obviously. What you want to do is you want to maximize. You've got financial resources in retirement, whether that's Social Security or your financial portfolio, but you're wondering, how much of that can you spend during your retirement. Well spending requires aftertax dollars. And as it turns out, there's a major difference in the amount though. Well, we'll get to a simple example. Even if you live in a tax-free state, you can have a marginal tax rate for moderate-income individuals of 46.25% on a wide range of income. And that's to the degree you can avoid that and pay instead of… When I say tax-deferred account, think of a 401(k) or a traditional IRA. If you take money out of that instead of paying 46.25% marginal tax rate, let's pay, say, 22% or less, that will give you a whole lot more of your financial resources that you get to spend instead of the government.

Ptak: It's a good segue to the next question that we had, which is on a key theme of yours, which is that de-accumulation and the sequencing of withdrawals from various accounts with different tax characteristics should go hand-in-hand with Social Security filing. Can you discuss that in general terms?

Reichenstein: Yes. And I'll right now talk about a moderate income, moderate-wealth client, which we'll call $2 million or less of financial assets. Here's what typically is a good strategy for somebody like that. And we'll talk a little bit later about higher-income individuals and strategy for them, if you wish, for myself and my wife.

But for this moderate-wealth couple, we encourage to delay Social Security benefits until 70. In those years before Social Security begins, you're going to have a marginal tax rate as the same as your tax bracket. Well, the conventional wisdom is, let's take money out of your taxable account first, and then out of your tax-deferred account, the 401(k)s, those next. And if you have any Roths, and a lot of people don't, take that out last. Well, here's the trouble with that strategy. If you're taking money out of your taxable account, that's usually tax-free withdrawals of principal largely. And so, you might be in a 0% tax bracket. Your adjusted gross income isn't even equal to your standard deduction. If not, you're in a low tax bracket. Don't waste the opportunity to take some of those tax-deferred accounts and fill up those low tax brackets. The best way to do that, in that case, is a Roth conversion.

Here's what happens. So, in those early years, let's say, you're filling up the 0%, 10%, 12%, and 22% tax bracket. Now, let's go to a few years later, you've started Social Security. And by the way, by then, you're probably in 2026 or later. So, we have those higher tax rates that are already scheduled to come back. That's the current law. And so, what happens is, you might be taking your money out of a tax-deferred account, say, up to the top of the 15% bracket. Well, think about if you had to take out additional money, call it another $1,000, it would cause another $850 of Social Security benefits to be taxed. So, your taxable income goes up by $1,850. Twenty-five percent of $1,850 is $462.50. Again, your marginal tax rate is 46.25%. Every dollar of income causes taxable income to go up by $1.85, because another $0.85 of Social Security is taxed. And 25% of $1.85 is $0.4625.

So, what happened is that you're saying in those early years, we made Roth conversions to fill, say, the 10%, the 12%, and the 22% tax bracket. Those were also the marginal tax rates. And what did we do? We avoided this 46.25% marginal tax rate on additional income during most of your retirement years. Well, you can do a whole lot better and add a lot of value by taking those tax-deferred account dollars, making the Roth conversions now at much lower tax rates than those marginal tax rates. And why are the marginal tax rates higher later? Well, first of all, we're going from today's tax brackets to the higher ones. But more important, when Social Security begins, you're going to have a wide range of that income for these moderate-wealth clients, say, $2 million and less, where they're going to pay a marginal tax rate of 185% of their tax bracket. We want to avoid that. How do you do it? Get those Roth account balances in those early years, and we use those as ammunition. You can withdraw those later, you're avoiding what, taking tax-deferred account. So, you take $1,000 out of the Roth, and hey, that saves you what, from having to take out a lot of money from the tax-deferred account that would have been taxed at 46.25%. And, of course, Roth withdrawals are tax-free. So, that's a good way to try to manage your retirement accounts there.

Benz: You've talked about accelerating withdrawals from tax-deferred accounts and how that can make sense in many situations. The question is, how did we end up with this conventional wisdom that it makes sense to start with taxable accounts, move on to tax-deferred accounts, and then save Roth for last? Where did that conventional wisdom come from?

Reichenstein: Here's what happens. That's the conventional wisdom, and it's in a lot of your financial software out there is the default. But here's what happens. Let's just take this year. Somebody who newly retired, and let's say, they haven't started—so, it doesn't matter, they're going to take money out of their taxable account. And remember, withdrawals from the taxable account are usually tax-free withdrawals of principal. They may be in a 0% tax bracket, or at least a low tax bracket this year. When I say a 0%, their adjusted gross income may not even be their standard deduction. And then, what happens after the taxable account is exhausted after, say, three or four years? Now, they've got most of their money in tax-deferred accounts. And in the rest of the retirement, they're going to be in a lot higher tax bracket, because they've taken out all these pretax withdrawals from tax-deferred accounts.

Instead of that, what's happening, the marginal tax rate on a bunch of that's 46.25%. Instead, let's make the Roth conversions now. We want to fill up those low tax brackets, say the 0%, the 10%, the 12%, and perhaps the 22% with Roth conversions. That gives you that Roth account balance, it's that ammunition that later in retirement… So, let's say, 2026 and on, you take out enough from the tax-deferred account to fill the 15% tax bracket. Think about it—if you had to take out more money, what would happen? You're already at the top of the 15% tax bracket to be taxed at the 25%. But each dollar of tax-deferred account withdrawal causes another $0.85 in Social Security to be taxed. So, your marginal tax rate is 185% of the 25% tax brackets—it's 46.25%. You're going, “That strategy, it sounds good, let's save taxes early.” But no, in your early years, you're in a 0% tax bracket unless you make those Roth conversions to fill those low tax brackets. In the middle years, and all those years, you're in a much higher tax bracket.

By the way, for somebody who does have Roth balances, if you take those out last, what are you doing? And again, your adjusted gross income is going to be zero in those years. You want to fill in the lower tax brackets and avoid those really high tax brackets in those middle years when you're taking out tax-deferred accounts. And for most retirees, that's most of their retirement years, and most of their retirement balances are in those. Almost never does the conventional wisdom come anywhere close to being an optimal strategy. Just think about—in those early years, you're in a 0% or maybe a 10% tax bracket. In the middle years, you're going to be in this 25% with a marginal tax rate of 46.25%. That doesn't make sense. Just go fill those lower tax brackets with, in this case, whether it's conventional tax-deferred account withdrawals or Roth conversions. We recommend the Roth conversions.

Ptak: Given that retirees aren't making decisions on these items in a vacuum, and a decision in one area could have repercussions for another, is there any way for retirees and pre-retirees to triangulate this problem without the aid of software? You've created some software to bring a lot of these variables together, but not everybody will have access to that. So, what should a retiree or a pre-retiree do?

Reichenstein: First of all, I don't mean it wrong. But yes, you could go to our software, and even for individual investors, we'll give you a free report, which will help you look at—you got to put in some of your financial information—but we can help you there. But here's why it's tough to do this without financial software. The decision—do you want to make a Roth conversion this year, 2022? Well, there are four questions you got to ask: What would be the additional taxes on that Roth conversion? How, if at all, would it affect your 2024 level of Medicare premiums? Because remember, you have those income-based Medicare premiums, otherwise known as IRMAAs, those spikes in marginal tax rates there as your income goes up. And since you're taking more out of the tax-deferred accounts this year, with a Roth conversion, you're going to have lower income in 2023 and later years, so you potentially could have lower Medicare premiums in 2025 and later years. That's just based on the decision one year. You got a question of should you make a Roth conversion in other years too, and then you also got the question of, if you're taking out just money from the tax-deferred account, what's the marginal tax rate on that across the various years? And there are so many moving parts, it would be very difficult to handle this without software.

Benz: We want to switch over to discuss Social Security specifically. Many people have heard the admonition to delay Social Security filing. Can you discuss the benefits of delaying?

Reichenstein: Well, first of all, if somebody has a very short life expectancy—a single has a very short life expectancy—then no, let's go ahead, we're talking about 77 years or less, start it as soon as possible. But let's say that your life expectancy is at least 85 years and you're single right now. Your real lifetime Social Security benefits pretax, it will end up being maximized by waiting till 70. And most people are worried about longevity risk, the risk of outliving their financial resources. Well, that's obviously the major risk there is what if you live a lot longer than expected? Well, in that case there, you can see where delaying till 70 ends up being the right decision.

So, for singles—and again, that roughly starting at full retirement age versus 70—the breakeven age is about 82.5. And I can give a simple example there. But between starting at 69 and 70, the breakeven age is about 85. So, let's just keep it simple. Somebody with a $2,000 primary insurance amount and the person's full retirement age of 67 today. Now, I know that would be a few years from now. But if you start at $2,000 and real benefits remain constant. Each year, you have that COLA increase to reflect inflation. So, think about 2,000 per month from age 67 to 82.5, for that 15.5-year period, and you calculate that, and that's the same amount as if you instead waited to 70. So, you would get 24% more, $2,480 a month in real terms each year from age 70 to 82.5. You work those two out and they're equal. So, the breakeven age between starting at 67 versus 70 is 82.5%. And again, if you changed it to 69 versus 70, it's 85 years. But it doesn't take long before it really pays to wait for Social Security. And again, if you're worried about, as most clients are, the risk of running out of money in their lifetime, well, that occurs primarily if they live longer. We've looked at there, if you wish, that's the decision for a single individual. Unfortunately, the decision for a married couple gets a lot more difficult. And I can go through a couple of examples just to explain that if you wish.

Ptak: Sure. Why don't you go ahead and explain that if you don't mind?

Reichenstein: Well, take my wife and I as an example. There are two sets of rules. If you were born Jan. 1, 1954, or earlier, as I was, then you can follow restricted application for spousal benefits but only if your spouse has already applied. In this case, my wife has already applied for her retirement benefits. If you were born after Jan. 1, 1954, you can't follow a restricted application. So, there's a different set of rules potentially—and for my wife and I—for each set of spouses. So, what did we do? In our case here, as soon as she was eligible for benefits, she filed, so I could file that restricted application, get half of her primary insurance amount at 70, which will be next month. I will start Social Security benefits and I get that for me 32% more, and that will last till the second spouse dies. So, notice what was the best decision for us? It turns out to be her actually filing early so I could get the restricted application for spousal benefits.

Now, let's change that. Suppose both married couples are born on Jan. 2, 1954, or later. So, they can't file the restricted application. And just to keep things relatively simple here, let's suppose they're already a same-age couple, just turned 66.5 today, which is their full retirement age. And for simplicity, I'm going to assume the husband has the higher PIA, primary insurance amount. Suppose his is $2,600 and hers is $2,000. What happens? Benefits on her record, at the death of the first spouse—and it doesn't matter who dies first—the survivor keeps the $2,600, of course before COLA increases. And the benefits based on the record of the $2,000, her primary insurance amount, that goes away. So, what happens? When should he begin benefits, the higher earner? It's based on the age he would be when the second spouse is expected to die. When should she begin? Benefits based on her earnings where it could go away at the death of the first spouse, and it doesn't matter which one dies first. So, suppose he's got a short life expectancy, he comes up with cancer, and they say you've got a few years to live. What should happen? She should start her Social Security benefits. He shouldn't start his. This is assuming that she lives past 85, you would maximize benefits based on his record by him not filing.

Let's change that example. Suppose it's the same married couple, both born Jan. 2, 1954, or later, but his PIA is $2,600, hers is $1,000. What happens? Now, he has got to consider this: What is her benefit? Well, she gets $1,000 just because she is full retirement age today. But if he files today, she gets $1,300, half of his primary insurance amount. So, he has got to consider not only how his starting date would affect the survivor's benefits, he has got to consider how it would also affect his wife's benefits today and until that first spouse dies. So, you can see there's a lot of moving parts there. And so, it makes it a much more difficult decision.

Benz: There used to be all kinds of more complicated strategies for Social Security filing, but the government put the brakes on those several years back. Why do you think they went by the wayside and are no longer allowable?

Reichenstein: Well, they called them loopholes, but they were certainly part of the law for all that time. And I think they tried to make things a little simpler. And like my case, I can file a restricted application for spousal benefits, but my wife has to file. And they just felt like there were, I guess, a lot of complications. If I was born Jan. 2, 1954, or later, I wouldn't be able to do that. So, it would not be our best strategy based on projected life expectancy. So, I think they tried to simplify that. They call them loopholes. I don't call them loopholes. But that was their term, and that's why they chose to get rid of those.

Ptak: We often hear from pre-retirees and retirees who say they plan to take their Social Security benefits at 62 or their full retirement age and then invest what they don't need. They point out that they can out-earn, so to speak, with their investments the enhanced benefit that they receive through delaying. Do you think that's a sound way to look at it?

Reichenstein: No, I don't. Let me explain why. First of all, we're talking about pre-retirees and people that are at or near retirement age. They need some fixed income in their portfolio. Well, think of Social Security and their extended portfolio of Social Security benefits are like a bond. What happens? Today, with the really low interest rates, what's the best returning fixed-income asset? It's delaying Social Security benefits and getting the 8% delayed retirement credits. And then, again, assuming they don't have a really short life expectancy, an unusually short, but that's their best source of, the best return on the fixed-income portion of their portfolio where I'm thinking of Social Security as part of that extended portfolio. And yeah, you could say bond returns are higher, but that's like saying, hey, why don't you put everything in stocks? That doesn't make sense. People at that age they need a good chunk of their money in fixed income. And the best-returning fixed-income asset right now, especially in the low-return environment, is delaying Social Security benefits.

Benz: So, you're not saying that people shouldn't have fixed income, right?

Reichenstein: Correct. But my point is, well, right now, congratulations, we're finally up to where we get 2% on 10-year Treasury. So, actually interest rates have been rising, as I think you're aware. But even, say, short-term, two-year Treasuries are now close to 2%, right at 2%. In fact, I think today, they were actually slightly over. But call it 2%. That's before the expenses of the mutual fund, and that 2%, that's awfully low. If you can get 8% delayed retirement credits on the Social Security portion, that's actually a better return than the returns available on fixed-income securities today. So, somebody who is throwing away the best way to increase what's, in essence, a portion in their extended portfolio—I'm including Social Security benefits, and that's, in essence, a bond. It's an inflation-linked bond, real lifetime annuity, and that's offering a better return than you can get in the rest of your fixed income. But yes, at that age—I turn 70 next month—yes, I do have a fair amount of fixed income in my portfolio, but I certainly don't want all stock and why throw away the best returning portion, which is the Social Security of that fixed-income portion.

Benz: You've made the case that pulling from tax-deferred retirement accounts often make sense if it can help someone delay Social Security. But what about market environment? How does that fit into this? If someone retires and the market tumbles right out of the box, might accelerating Social Security make sense if it can help forestall withdrawals from a depressed portfolio?

Reichenstein: So, somebody who says, "I just retired, and all of a sudden, we had this huge decrease in the stock market," and bond markets may have been hurt too. Part of your portfolio, extended portfolio, includes the fixed-income portion you have in your financial assets and then also the Social Security. And my gosh, again, the better-returning part of that is the Social Security. So, yes, the markets went down, take money out, you can take it out of the fixed-income portion of your financial portfolio, but I would delay that Social Security. That's better than taking the Social Security out. You got more money now earning 2% in the bond fund, gross return. I'll take the delaying Social Security and get the 8% delayed retirement credits. That's a better deal. And I know it can be trouble when you retire, the markets go down. At that point, you may take money again out of the fixed-income portion of your financial portfolio and try to give time for the stocks to rebound. But the idea of your taking your Social Security now on forgoing the higher returns on that compared with what you're getting the 2%, you're going to get in the in the bond market? No, I'd rather keep that higher-returning, the 8% delayed retirement credits, keeping the Social Security.

Ptak: Another reason that people sometimes cite for taking Social Security early is they're concerned about the program's solvency. Is that legitimate, in your opinion? Are there any age bands or income profiles who should give their projected benefits from Social Security a haircut if they want to be conservative in their retirement planning?

Reichenstein: In our software, we built everything based on the current promises of the Social Security system. Let's go back to the early 1980s, when the trust fund was virtually depleted, and they made significant changes. But none of those changes affected benefits to current retirees, or retirees roughly 55 or older. And you read those talking about here are the changes that we may consider. And often the comment is, "What's the thing we have to do?" We can't take away benefits to current recipients or those in or near retirement. Because they've already planned on that. And usually, that age is roughly 55 or older; we're not going to bother the benefit. So, that's been the history. Now, there's no guarantee that something doesn't change in the future. But that's what they've done in the past. And I fully expect them to continue to do that.

So, I would not do that. I think of a good friend of mine who was concerned about that. And I said, well, that's not what our software is built on. We're based on the current promises. But if you feel they're going to start doing that, OK, take your benefits early. And by the way, that was years ago, and he did, and I'm sure he wishes now we hadn't. He's well into his 70s, and it ended up hurting him. So, no, I do not think that's a good reason. And I read that stuff. And all the proposals like tends to be, we got to protect those people in or near retirement. They've already depended on this. The adjustments have to be for people younger than that.

Benz: We want to get back to the interplay between Social Security-claiming and taxes, which is a big specialty of yours. You've talked extensively about that so far. But let's talk about this thing called the tax torpedo, which you alluded to earlier in your comments. Can you describe what that is, and perhaps more generally talk about how Social Security benefits are taxed?

Reichenstein: Let's talk about how they're taxed. It depends upon the level of provisional income, sometimes called combined income. Provisional income is equal to the following: modified adjusted gross income, plus half of Social Security benefits, plus tax-exempt interest. One of the problems is modified adjusted gross income is used in a lot of places in the tax code. And its definition changes with each use. For almost all taxpayers, that measure of modified adjusted gross income, as used for professional income, it's everything in adjusted gross income except the taxable portion of Social Security benefits. If you think about it, we're using this measure of income to figure out how much of Social Security benefits should be taxable or included in adjusted gross income. So, that measure cannot include the taxable portion of Social Security. And here's what happens.

For single individuals, there's two income thresholds: $25,000 and $34,000. For married couples filing jointly: $32,000 and $44,000. Let's just take singles right now. In general, by the way, for financial advisor clients, and even moderately wealthy couples, they are going to probably have the amount of Social Security benefits that are included in taxable income is lesser of the following two. It'd be $4,500, which is $0.50 for each dollar of provisional income between the $25,000 and the $34,000, plus $0.85 for each dollar above $34,000 or 85% of Social Security benefits. And that's what happens once you get above $34,000, and that doesn't take much if you get above that. What happens is that each dollar above that causes another $0.85 of Social Security to be taxed until you finally get to the point that 85% is taxed. And for married couples filing jointly, the two PI-income threshold levels are $32,000 and $44,000. And again, unless somebody has an unusually low level of Social Security benefits—in this case, less than $6,000— you will pay taxes on the lesser of between $32,000 and $44,000, and for each dollar of income in there an extra $0.50 of Social Security would be taxed.

So, assuming you're above $44,000, you're going to pay taxes on $6,000 there, plus $0.85 for each dollar of PIA above $44,000 until you finally get to the point that 85% of Social Security benefits are taxed. And that's a good size income range where every dollar of additional income, call that a tax-deferred account withdrawal or whatever it is, it could be earned income, causes another $0.85 of Social Security to be taxed. So, your marginal tax rate is 185% of the tax bracket. That's a big jump. That needs to be considered in your withdrawal strategies.

And the tax torpedo, it just refers to the point between that first PI-income threshold level will go back for singles, $25,000, until you hit the point where 85% of Social Security benefits are taxed. So, actually, for most of that range, it's for every dollar of additional income, an extra $0.85 of Social Security benefits are taxed and your marginal tax rate is 185% of the tax bracket. Once you hit the point that 85% of Social Security benefits are taxable, your marginal tax rate drops down from—and let's go back to—most people for most of their retirement, they're going to be in the 25% or higher tax bracket—the point where 85% of Social Security benefits are taxed are in the 25% starting in 2026. So, 185% of 25%, it's 46.25%. After that, the marginal tax rate drops back down to the tax bracket, which is 25% in that case. Until what? Until you hit one of those Medicare income thresholds where all of a sudden you get five huge spikes in your marginal tax rate because of those income-based Medicare premiums, sometimes called IRMAAs or Income-Related Monthly Adjustment Amounts. So, you can see and notice what happens to marginal tax rates, go from the tax bracket to 150% of the tax bracket, for a typical retiree to 185%, then back to the tax bracket. And then, if the income continues to go up for higher-wealth individuals, you get these five huge spikes in marginal tax rates due to those income-related monthly adjustment amounts. That's the marginal tax rates you need to navigate in order to tax-efficiently withdraw money. As you can see, that's pretty complex.

Ptak: And I think we want to talk some more about IRMAAs in a moment. But before we did that, it does sound like more retirees are hit by this tax torpedo every year. Why is that happening?

Reichenstein: Well, take this for singles, but it's the same thing that two PI-income threshold levels of $25,000 and $34,000, those have never been changed since they started taxing Social Security back in the '80s. So, obviously, in essence, what happens, it's going to hit a larger and larger portion of the population. So, that's why it was designed originally that it wasn't going to affect but just the relatively wealthy. Well, now, unless you're at the really low-income level, and you just haven't saved for retirement, it's going to affect you. And then, admittedly, that's half of the people, as I understand now, don't pay any taxes on their Social Security benefits. But for financial advisor clients, that's not the case for the overwhelming majority of them. They're going to pay taxes on a good portion of their Social Security benefits, sometimes 85%.

Benz: Well, I wanted to discuss that other extreme, because it seems like there's a potential for some financial advisors and higher-income, higher-net-worth clients to just say there's no use and even bother trying to figure this out; that they will probably pay taxes at the highest level, and there is no use trying to worry about it. Should they bother with trying to figure out where they pull the funds from on a year-to-year basis?

Reichenstein: Two things: First of all, a lot of them, based on their current withdrawal strategy, they can hit the point that 85% of Social Security benefits are taxable with financial accounts of $1 million or less. But if you do that strategy, we make those Roth conversions in the early years, delaying Social Security and then withdrawing money afterward. And you go back to The Journal of Retirement, September 2021 article, and also articles by Geisler, Hulse, and Harden in the Journal of Financial Planning, and you could have $2 million of financial assets and still, if you're doing that, you're going to pay taxes on less than 85% of Social Security benefits.

I'll just take my wife and I as an example. We were above moderate wealth at that point. Yeah, we're going to pay taxes on 85% of Social Security. But remember, at the end of that tax torpedo, the marginal tax rate drops from 185% of the tax bracket, down to the tax bracket, and it stays there until you hit that first IRMAA Medicare premium spike. Well, let's take advantage of that. In the case of my wife and I, she is six years younger than me, and again, I'm about ready to turn 70, she just turned 64. The last several years, I've been making Roth conversions where I'm saying, OK, I'm willing, since I'm the only one paying Medicare right now—she is not old enough—we're going to pay the first two spikes in Medicare premiums, but I'm going to try to avoid the third one. Now that my income for this year, even last year, my income, the 2021, will determine our joint level of Medicare premiums two years since, when we'll both be on Medicare. So, at that point, now, we're trying to keep our level of income below that second mid-income threshold level. But that does give us an ability to make reasonable amount of Roth conversions since we're already beyond the end of the tax torpedo, where we're going to pay 85% of Social Security benefits. But that's where, remember, your marginal tax rate is now back down to your tax bracket until you hit those IRMAA, those Medicare premium spikes. And in this case, we can't avoid the first one, but we're trying to avoid the second one. But to the degree we can do that, let's say, it's maybe only $20,000, but if we make a $20,000 Roth conversion and pay only 22%, that's a good deal. That may avoid several years of Medicare spike down the road. So, you can see that these ideas also affect higher-wealth clients, that their incomes are beyond the end of the tax torpedo.

Ptak: You've mentioned IRMAAs several times during the conversation. I don't know if there's anything to add to what you've already mentioned. But these are the surcharges that high-income Medicare beneficiaries can run into. From a planning standpoint, any sort of principles or lessons that you would impart to those that want to make sure that they're not getting dinged by those, should they encounter them?

Reichenstein: I'll just take my wife and I as an example. Let's take our income in 2020. It would not affect my wife's Medicare premiums in 2022, because she is not on Medicare this year, and she won't be on Medicare until early next year. So, what happened is, our income was already going to take us beyond the end of the tax torpedo. Well, we now have the lower marginal tax rates equal to the tax brackets until we hit—that first IRMAA, by the way, is much smaller than the second, third, and fourth, but here's what happens.

I went ahead and took mine to the top of the second one. Now, here's what happens in 2020. I had to estimate what would be those Medicare income threshold levels in 2022. And that's a problem for planning because at the end of '20, I'll make those Roth conversions at the end of the year where I can have the best estimate of my income, but we know what the level of those Medicare-income thresholds are for 2021. They've been announced, but not for 2022. But to be careful, I just use the 2021, because I want to leave a margin for error. Because if you go one penny above, here's what can happen. For us, being both on Medicare, if our 2020 income caused our 2022—married couple, both on Part B and D for all 12 months—it could cause, if you hit past that second, third, or fourth PI or Medicare-income threshold level, your level of Medicare premiums two years since can cause you to pay more than $2,900 more to the federal government in Medicare premiums. Well, think about it. A $1 increase that caused your income two years early to exceed one of those second, third, or fourth income-threshold levels by a penny caused you to owe another more than $2,900 more to the government two years later in Medicare premiums.

Think about it—what's the marginal tax rate. And that's the tax rate on the next dollar of income. If that next dollar of income causes your taxes to go up by $1, that's 100% marginal tax rate. In this case, here, they exceed going up by more than $2,900. That's the marginal tax rate—I hope you're sitting—exceeding 290,000%. Let's take it a little different. Somebody takes that last $1,000 out and says, “Oh, I'm only in a 22% tax bracket.” Well, as it turns out, that causes them to exceed the second, third, or fourth income threshold level. They owe more than $2,900 more in Medicare premiums two years hence based on that, not to mention the ordinary income tax. So, my point is, hey, when your income goes up, you owe more to the federal government. That's an income tax. But what's different here on the IRMAAs is there are huge spikes are going a penny over above a threshold level, and so, the marginal tax rates for a married couple exceed 290,000%.

Benz: You've referenced the virtue of IRA conversions in those early retirement years. How would charitable giving fit in here, perhaps as a means of reducing future tax bills and future required minimum distributions?

Reichenstein: Well, that's good. Yes, for your clients that are charitably inclined, or for individuals out there charitably inclined—and by the way, in my book, Income Strategies, 2019 version, go to Apple and Google "income strategies, William Reichenstein," and you'll see—I talk about qualified charitable distributions and donor-advised funds there. If you're going to give, those are really tax-efficient ways to give money. On the qualified charitable distribution, you have to be at least 70.5. But you're giving pretax dollars, and by doing that, that money, first of all, it would count toward your required minimum distribution. It's made with pretax dollars, it's not taxed; it doesn't show up in your measure of income used to calculate the taxable portion of Social Security benefits, and they don't show up in your measure of income to calculate the Medicare premiums. So, they're a very tax-efficient way to give money.

Ptak: Many 401(k) plans now include a Roth option, and younger investors often opt for a Roth IRA rather than a traditional one. Do you think required minimum distributions and the related tax issues will play less of a role for future retirees than they do today?

Reichenstein: First of all, yes. As you know, I think it was 1997 that Roths finally became available. So, they haven't been there for most of my retirement. But yes, for younger retirees today, when you get to the problems of taxation of Social Security benefits and these income-based Medicare premiums, and they affect retirees, well, they aren't thinking about that, most of them. But because of that, because their marginal tax rate in retirement is going to be highly elevated, they ought to think about making contributions to Roth accounts today instead of traditional IRAs or a 401(k), so if you raise your Roth 401(k) instead of a 401(k).

And why do I think that's not going to go away? You look at it and all the tax laws that have occurred in the last few years, and most of them are saying, “If somebody is retired, and they have more than a modest amount of income, we really want to go after them. That's where we ought to get our taxes.” So, think about the SECURE Act. You die, well, instead of my sons, my wife and I, our sons inheriting the money and taking it out slowly over their lifetime, now they have 10 years to take it out. And they may have required minimum distributions even in those first 10 years. And then, you take a look at the IRMAAs, these income-related monthly adjustment amounts, that started in 2010, and what have they done? They keep increasing those last year, that Part B, first IRMAA increase was, if I remember, was like 15.5%. They keep raising those dramatically. That's where we're going to get our money. And I think for people who have more than just this really minimum amount of income in retirement, they look at that as a source of income, and I don't think that's going to change.

So, what I would encourage you to do—and I've encouraged my kids—is they put money in Roth accounts. Between you and I, if we have to kind of help them finance that a little bit for the taxes or whatever, well, we can give now at $16,000 a year to each of them so we can help them with that retirement decision. Yes, and they're better off, if you wish, if they're in a 22% tax bracket now saving $7,800 in a Roth of aftertax dollars instead of $10,000 in a 401(k) in a pretax, I'd rather say the $7,800 after taxes. And for most, I'd encourage them to do that. By the way, if you turn to your clients and tell them to help their kids on that or give them that advice—most of these kids, they have no idea about these elevated marginal tax rates in retirement. Now, these tax-deferred accounts are going to affect them then. So, if they did, they would have a lot more emphasis in savings in Roth accounts. And no, I do not think that's going to go away. That's been the source and that's where they're looking at: “This is where we want to get our money. Hey, these people don't really need it that bad anyway, so they're a great target.”

Benz: You've referenced tax-efficient withdrawal sequencing several times. A related question is how to position each of those accounts? What types of assets should go into traditional tax-deferred Roth and taxable accounts to best address the drawdown sequence as well as to reduce taxes? Do you have any thoughts on that?

Reichenstein: We are talking about the asset location—where do we want to hold stocks? Where do we want to hold bonds? And I want to point out that while meeting your target asset allocation, let's say, for a retiree, they're shooting for a 40% stock, 60% fixed-income portfolio. Let's say, you've got tax-deferred accounts, you've got taxable accounts, and you've got Roth. In the traditional argument on asset location—and I don't mean this wrong, but I'm certainly one of the leading authors in that area—the general argument is, you want to hold stocks in the taxable accounts and bonds in those retirement accounts—tax-deferred accounts, the Roth—to the degree possible while hitting your target asset allocation. And the key argument there is those stocks in the taxable accounts are generally taxed long-term capital gains at 15%, a preferential capital gains tax rate, and those are better.

I did write something in Advisor Perspectives a couple years ago that still applies today going, for new money you're putting into the taxable account—I can think of a widow that I'm dealing with now with most of her money in her taxable account, has huge capital gains in it. We don't want to sell those. But if she had new money to go in, what would we do? I would say, that's OK, let's put some of that in bonds. Bonds for the last two years have had basically a 0% return. When you're trying to save, what do you want to hold in the taxable account that's going to have the lowest tax liability? Well, if they're earning a 0% return, or two tenths of 1%, well, it's not bad to hold bonds there. But we're quickly getting back to a more normal interest-rate environment. So, I think, we go back to the let's hold stocks in the taxable accounts and passively held stocks, ideally, so that you can get that long-term capital gains rate of 15%, which is preferential than if it's held in a tax-deferred account, when you withdraw, that's ordinary income and may be taxed at 25% or 28% pretty soon. But remember, that's while meeting your target asset allocation. So, for sake of argument, some retiree who wants a 40% stock allocation, it doesn't mean, well, you want to hold all stocks in your taxable account even if that means it's 60% stocks. No, you want to get the right stock allocation. It may mean for a lot of people that they would be holding virtually all stocks in the taxable account. Then in the retirement accounts, tax-deferred accounts, and the Roth, hold a mixture of stocks and bonds but hit that target asset allocation of yours.

Ptak: Wanted to switch and talk retirement income. Safe withdrawal rates haven't been central to your work in recent years, but you probably follow the research. Is the 4% guideline too aggressive given current market conditions? Yes, fixed-income yields have risen off the lows, but they're still quite low by historical standards, and you could probably argue that valuations are elevated by a number of measures. So, what's one to do?

Reichenstein: Interesting that you should mention that. I have just finished an article, well, a column, that I think will come out in the Journal of Financial Planning. I haven't had something rejected by them, and for that matter, any other journal and several here. So, I have a hunch it will come out in a few months there. But first of all, the point I'll make there is that, yes, real returns today on bonds—well, you just look at the real return on TIPS, they're negative still on 10 years and well, well below historic levels. And the 10-year Treasury, fine, it's finally hit 2%, or recently, now, it's actually more like 2.3% in the last few days. But it's real returns. And real returns with inflation the way it is, it's still well, well below historic averages. And even if the equity risk premium is as high now as it usually is, its real returns on stocks are going to be below average.

So, what I did? I said, OK, if returns in today's low-return environment, what does that tell us instead of that 4% rule? Well, first of all, that's Bengen, and he has done great work, but he assumes a 50% stock exposure. And when does that 4%, when can you only withdraw 4%? Well, if you have really bad returns in those early years. Well, in the stuff I did here, I was assuming the same real return every year, but it's much lower than the historic averages. First of all, I recommend, if you look at target retirement date funds, 40% or less is the typical target retirement asset allocation in retirement and not the 50% Bengen uses. And in my case, I was using the same real return. It was low every year and what changes, but what if you happen to have really bad returns in the first few years. But bottom line, instead of 4% today, it'd probably be more like 2.8% to 3% would be the safe rate because of the much lower return prospects today.

And it's hard to argue those much lower return prospects. You look at the bond yields. Again, the Treasury inflation-protected securities 10-year, you're talking about negative real return. We aren't anywhere near the 2.2%. Historic average going back to 1926, intermediate-term Treasuries earned 2.2% real returns. Well, we aren't anywhere near that. And even if stocks are in the 5.2% more than intermediate-term Treasuries, then they're still well below historic averages. So, returns looking forward are likely to be well below average. And with that, instead of 4%, it's probably more like 3%, maybe 2.8%. And part of that difference is, whether he assumes 50% stocks, but I think that's a little high. Based on the target-retirement date funds, I almost always say, take Vanguard, they're down the 30% once you hit age 70. And I could go through the other ones, but 50% is above what target-retirement date funds recommend.

Benz: One question we've put to a number of our guests over the years has been whether we're in the midst of a retirement crisis, or about to have a retirement crisis? And the really interesting thing is just the diversity of opinions that we've heard on that question. What's your take?

Reichenstein: First of all, there's maybe half of or more than half of people entering retirement now that have saved almost nothing. So, they've got, you may call it, a crisis for them. They're in trouble because they haven't saved enough. But let's take the other, call it, half of the people that have saved enough, I don't think they're in retirement crisis. Yes, returns are going to be lower going forward, or certainly look that way. So, the withdrawal rate, again, might be 3% instead of 4% for a safe withdrawal rate. But if you have saved up enough for retirement, you should not be in a crisis situation. If you haven't saved enough for retirement, there's always that half of the population, yes, they're going to be in trouble. But that's been the case, and I don't think that will ever change. But as far as for those that have saved reasonably for retirement, there should not be a crisis situation. Again, maybe a little bit lower the safe withdrawal rate, maybe 3% instead of 4%, but it should not be a countrywide crisis.

Ptak: Has spending so much time focusing on retirement planning affected your thinking about your own retirement, and if so, how?

Reichenstein: Well, yes, and I've adopted the strategies—again, without getting too detailed, we have enough that our income is going to be to the point that 85% of Social Security benefits are going to be taxed. So, we're beyond that tax torpedo. But we have those five huge increases in marginal tax rates due to the IRMAAs. And yes, to try to minimize that, I have actually to wait till December when we have the best estimate of our income and in 2021, at the end of 2021, we see what the level of those Medicare income threshold levels for 2022. Well, the 2021 actually affects the level of Medicare premiums in 2023. We don't know what those levels of Medicare income threshold levels are. So, it's a little hard. But to be conservative, I just go up to the '22 level and say, I want to be conservative. Because again, if you go a penny over, you owe that huge spike. So, yes, I have done that. And my wife and I, now that we will both be on Medicare, I will try to limit it to the second-income threshold level. So, I've used all this stuff in how we manage our money here. And in this case here, it's the IRMAAs that we are concerned with. And I have made a reasonable amount of Roth conversions the last three, four years, which if anything, I wish years ago, I would have saved a lot less in tax-deferred accounts and a lot more in Roth accounts. And again, you can tell your clients that.

Let me also mention one other group because it's important. We'll take my wife who is six years younger than me. Well, based on at least longevity of Reichenstein men, let's say, she is probably going to live a lot longer than me. Suppose for sake of argument, I die in 2030. In 2031, she is going to file for tax as a single. Well, her tax bracket and therefore marginal tax rate is going to be a lot higher in 2031, one year after my death, and the rest of her life. And three years after, in 2031, her income, filing as single, will affect her level of Medicare premiums in 2033. Those Medicare income threshold levels for singles, currently $91,000, $114,000, and so on, they're half the level of where they are for couples. It could very well be the case that three years after I die, if I die in 2030, in 2031, she could be paying more in Medicare premiums herself than she and I both paid while we're both alive.

Well, what's the solution? While we're both alive, let's consider Roth conversions, where those Roth conversions will be taxed at the tax brackets for married couples filing jointly, which at least for the first few are twice the level of that for single individuals, and those Medicare income threshold levels are for the first four of the five, they're half the level. So, that's another strong incentive. And by the way, who does that apply to? As long as one spouse lives at least one calendar year past the other one, it's going to affect them. That marginal tax rate is single once you file as a single, and that's going to affect most couples. So, that was another reason we were pretty strong in making Roth conversions until recently. Now, all of a sudden, we got to worry about both of us paying those spikes in Medicare premiums. So, our opportunities are going to be much less going forward.

By the way, at the end of the year, I will still go back. OK it's December. Well, let's look now and see where those Medicare… If I can make a Roth conversion of—last year it wasn’t as much, say it was $18,000 up to the safely below that next modified adjusted gross income threshold level. OK, that's taxed at, in this case here, let's say, I think it was 22% or 24%. That's a lot better than the potential adverse effects of paying a spike in Medicare premium down the road, for many years down the road, potentially.

Benz: Well, Dr. Reichenstein, you've given us lots of good food for thought. Thank you so much for being with us here today.

Reichenstein: Well, I appreciate it. Thank you. And I left Baylor years ago and actually took an income that's quite a bit lower because the last few years where I want to help a lot of people, and I feel good about being able to provide advice that may help people. So, thank you for the opportunity to hear some of my thoughts.

Ptak: Thanks for being with us today. We really appreciate it.

Reichenstein: Thank you.

Benz: Thank you for joining us on The Long View. If you could, please take a moment to subscribe to and rate the podcast on Apple, Spotify, or wherever you get your podcasts.

You can follow us on Twitter @Christine_Benz.

Ptak: And @Syouth1, which is, S-Y-O-U-T-H and the number 1.

Benz: George Castady is our engineer for the podcast, and Kari Greczek produces the show notes each week.

Finally, we'd love to get your feedback. If you have a comment or a guest idea, please email us at Until next time, thanks for joining us.

(Disclaimer: This recording is for informational purposes only and should not be considered investment advice. Opinions expressed are as of the date of recording. Such opinions are subject to change. The views and opinions of guests on this program are not necessarily those of Morningstar, Inc. and its affiliates. Morningstar and its affiliates are not affiliated with this guest or his or her business affiliates unless otherwise stated. Morningstar does not guarantee the accuracy, or the completeness of the data presented herein. Jeff Ptak is an employee of Morningstar Research Services LLC. Morningstar Research Services is a subsidiary of Morningstar, Inc. and is registered with and governed by the U.S. Securities and Exchange Commission. Morningstar Research Services shall not be responsible for any trading decisions, damages or other losses resulting from or related to the information, data analysis, or opinions, or their use. Past performance is not a guarantee of future results. All investments are subject to investment risk, including possible loss of principal. Individuals should seriously consider if an investment is suitable for them by referencing their own financial position, investment objectives and risk profile before making any investment decision.)

More on this Topic