Bogleheads® Chapter Series – Mike Piper

Mike Piper, the Oblivious Investor, discusses Social Security and portfolio management.

Hosted by the Minnesota & Detroit Bogleheads chapters. Recorded on March 27, 2021.

The slide presentation can be accessed here.

Chapter meetings are listed in the Bogleheads blog calendar. You can add this calendar to your Google account. Notifications are also sent from an email subscription list. See the blog for more information.

Transcript

Bogleheads® Chapter Series – Mike Piper

[Music]

Host: Welcome to the Bogleheads Chapter Series. This episode was jointly hosted by the Minnesota and Detroit chapters and recorded March 27, 2021. It features Mike Piper, the Oblivious Investor, discussing Social Security claiming and portfolio management. Bogleheads are investors who follow John Bogle's investing philosophy for attaining financial independence. This recording is for informational purposes only and should not be construed as investment advice. Please note this recording began a few minutes into the presentation but nothing important was omitted.

Mike Piper: …it means that taking the money and investing it becomes more attractive. Of course right now the opposite is true, so that's a point in favor of waiting, but if you're 50 right now and you're going to be making this decision 12 years in the future, well interest rates might look very different than they do right now.

So when we move on to the more complicated situation of a married couple, the rough draft plan here is generally for the higher earner to wait all the way until age 70, and for the lower earner to file as early as possible or close to age 62. And the reason for this, it has to do with the way that survivor benefits work and specifically  what ends up happening is that once either of the two spouses dies, the surviving spouse generally is left with the larger of the two benefit amounts. It's the smaller one that disappears once one of the two spouses dies.

So in other words when the higher earner waits to take benefits it increases the household benefit for as long as either of the two people is alive. So if the higher earner waits it increases their own retirement benefit. But if that higher earner dies first, well then the lower earner will now start getting a survivor benefit that is increased as a result of the fact that the higher earner waited. So the higher earner waiting increases the household benefit for as long as either person is living. Or you could look at it as it increases the household benefit until the point at which both people have died.

Now on the other hand, when the lower earner delays benefits it only increases the household benefit for as long as both people are alive, and it's basically the opposite reason. It's if the lower earner delays that increases their retirement benefit, but if they die then that benefit's no longer relevant. And if the higher earner dies first, then the lower earner's benefit will get stepped up to what the higher earner had been receiving. So again, when the lower earner waits it only increases the amount that the household gets while both people are still living, or until the point at which one person has died. And the key point here is that until both people have died is a longer period of time than until one person has died, and that's why it's generally advantageous for the higher earner to wait.

And I think it's kind of intuitive to look at this visually. And I think if there's basically four broad scenarios that can happen when you have two people, it's a two by two matrix. Basically in one corner you have the case in which both people live beyond their life expectancy, and in the opposite corner you have the case in which both people die before reaching their life expectancy. And then in the other two corners you have the case in which one spouse dies early and the other person lives beyond their life expectancy.

Now when we're looking at these four options, or these four possible cases, for the filing date for the higher earner, we're concerned with how long it will be until both people have died. So in other words what we're asking is how likely is it that at least one person will live a long time, because if at least one person lives a long time, then it will have made sense for the higher earner to have waited to file. So if we look at these four scenarios we can see that in three of them at least one person lives past their life expectancy. It's basically just the one case in which they both die early that that doesn't happen. So in other words there's about a three out of four chance that at least one person lives beyond their life expectancy, and that's why it's advantageous for the higher earner to wait to file for benefits.

But then if we look at it again with the decision for the lower earner in mind, now what we want to know is how likely is it that at least one person will die before reaching their life expectancy, because if at least one person dies early, then it will have been advantageous for that lower earner to have filed early. So we look through our four possible outcomes here, and then three out of the four somebody dies early so there's a roughly three out of four chance that at least one person dies before reaching their life expectancy which is why it's usually advantageous for that lower earner to start their benefit early.

And for those of you who like to look at it with actual numbers this is just one basic example. It’s a married couple, husband, wife. They're both exactly the same age, 62, and they're both in exactly average health. In this case the husband's individual life expectancy is 20 years, so he would be expected, on average, to live until about age 82. And the wife's individual life expectancy is 23 years, which means she would be, on average, expected to live until about age 85.

Now with regard to that lower earners decision, what we want to know is how long will it be until somebody has died, and a lot of people look at these two pieces of information and they think the answer is 20 years. But that's not the answer because remember there's about a three out of four chance that somebody dies early, and in fact, when you actually use mortality tables to answer this question ,it turns out that in just 16 years it's more likely than not that somebody has died. And so that's again why the lower earner usually wants to file early. 

And if we look at the question for the higher earner, we ask how long will it be, on average, until both people have died, a lot of people intuitively assume the answer is 23 years. But by now you're probably catching on that that's not the answer because there's a three out of four chance that somebody lives past their life expectancy. So on average it turns out that it's about 27 years until both people have died. And that's why it's so advantageous for the higher earner to wait. That's why we have this rough draft plan where the higher earner waits as long as possible so they file at 70, and the lower earner files early, often at 62, or if not at 62, at least in the early range, roughly.

But of course there's plenty of exceptions. One important exception, and this applies to a lot of bogleheads is that if both people are just in very good health, or if you are just particularly concerned about longevity, then that's a point in favor of the lower earner waiting. And the flip side, of course, is that if both people are in very poor health then you would probably want to take the higher earners filing date and move it earlier.

Somewhat again, another exception is the minor child or adult disabled child scenario because in that case it might be a point in favor of the higher earner filing sooner because when/ if they were to do that, that would allow the child to get child benefits on that higher earners record at an earlier age. So it might make sense to do that.

Another important exception is if the lower earner is still working then we have to be thinking about the Social Security earnings test. And I'm not going to get into the math of that. The opensocialsecurity does, the calculator does account for that. By the way, if you tell it that you're still working and you put in the amount that you're earning, and when you expect to stop working, the result is basically that it usually makes sense for the lower earner to then wait, either until they stop working or until their full retirement age to file for benefits. 

Another possible exception is if one or both spouses have a government pension. Then we have to be thinking about the windfall elimination provision and government pension offset. And the calculator accounts for these too. If you tell it that you have a pension--this is a bit tricky to summarize in any concise way because it can affect the decision in either direction for either person. 

One last exception is tax planning. Usually that's a point in favor of waiting and the reason for that is that Social Security is inherently tax advantaged because at most 85 percent of your benefits are included in your taxable income, and depending on how high your income is, in many cases less than that would be included. And so when you wait to file for benefits basically what you're doing is increasing the portion of your lifetime income that is coming in the form of Social Security, so you're increasing the portion of your lifetime income that is tax advantaged. So that's usually a point in favor of waiting. But tax planning is always case by case, so there could be some other tax specific reason that would be a point favor filing early.

So that's it for the question of Social Security. Now if we move on to the second question of how much you can spend per year from your savings.If you've spent much time on the forum, or just reading about retirement planning in general, you've probably run into the 4% rule. And the idea there is that if you spend 4% of your portfolio balance in the first year of retirement, and then every year, the idea is that you would keep spending that same dollar amount except you would adjust it upwards for inflation every year, and you do that regardless of how the portfolio is performing. So even if the portfolio is going way up, or going way down, you just keep spending the same inflation adjusted dollar amount.That's the idea of this 4% rule concept.

But there's a bit of an issue and I think most people realize this as soon as they try to apply it in real life, which is, that most people in real life don't actually spend the same amount from their portfolio every year. And that's because you have different sources of income that stop and start at different ages, right. You might, for example if you're married, you might have one person retire, but the other person is still working. So maybe you're spending a little bit from your portfolio. And then the second person retires and now you're spending more from your portfolio. And then Social Security starts and now you're spending less from your portfolio. And the 4% rule just doesn't account for that in any way. 

And then there's also just the fact that for a lot of people, they just want to spend more money early on in retirement, right. Like traveling more, going out to eat more, going out to shows, more doing whatever it is that you like doing. A lot of people intentionally plan to do more of that in the early years of retirement. And so again we have a case where you just aren't spending the same dollar amount from your portfolio every year. 

So how do we address that fact, how do we actually incorporate that fact into a plan. I think the easiest way to do it is actually to bundle in the third question as well of which assets you should be spending, because there's a very straightforward approach that can address both of these questions together and essentially the idea is to recognize that your portfolio is being asked to satisfy separate spending goals, and those spending goals last for different periods of time.

And so if you have any spending that's only supposed to last for a short period of time you can use an allocation that's a good fit for a short period of time. And that piece of the portfolio that's going to satisfy this short-term spending you wouldn't be using something like the 4% rule. You would intentionally be depleting that piece of the portfolio according to a schedule. So you would intentionally be spending from it at a high rate. For example if something's supposed to last for four years you would be spending from it at 25% per year, and then for the rest of the portfolio, the portfolio that's supposed to last through your whole lifetime, that's where you generally want to use a stock oriented allocation and a low initial spending rate. So it's three to four percent depending on who you ask. 

So I know this isn't very intuitive, so it's actually much easier to explain it visually. So what we have here, imagine that this is your desired total level of spending per year in retirement and the idea is that you are intending to spend more in the first part of retirement than in the later part of retirement. And in this particular example, it's just one decline at one moment in time, but it could be a gradual decline over several years.It doesn't really change the concepts, it's just trying to keep the example simple. and there's no numbers on the age axis, and that's because it doesn't really matter whether this decline in spending happens at age 72 or at age 83, the concepts are still basically the same. And there's no numbers on the spending axis either because it doesn't really matter whether this is fifty thousand dollars or five hundred thousand dollars of spending, the concept’s still basically the same.

So the idea is that every year you have to come up with enough dollars to get you up to that line, to get you up to your desired total amount of spending. But the key point is that it doesn't all have to come from the portfolio because at some point your Social Security benefit i's going to start and it's going to get you part of the way up to that line. And if you're married at some point the other person's Social Security benefit begins, so it gets you even further closer towards that line. And so it's just the rest of the area, the green space basically, that has to come from the portfolio. And so at the beginning of retirement it is every dollar of spending that comes out of the portfolio, but then it's a little bit less because you have some Social Security coming in, and then it's even less because you have more Social Security coming in, and then it's even less because now you've reached the point where you just aren't planning to spend as much per year. 

So the level of spending here from the portfolio, you can see that it's fluctuating quite a lot over time. And that's why the 4% rule or even similar concepts, they just kind of break down when you try to apply them here. 

But the other thing you might notice is that in this scenario this person has the potential for massive sequence of returns risk because they're spending from their portfolio at the fastest rate early in retirement, and so this initial spending rate might be way above four percent it could be six, seven, eight percent depending on how much of the total spending would be,coming from Social Security later. And so this is a risky case because if the portfolio declines significantly early in retirement and you're spending from it at a high rate you can run into a scenario basically where the portfolio gets depleted really quickly and you have a real problem. 

So the solution again is basically to separate out the portfolio into separate chunks to satisfy these different pieces of spending. So for example, the extra spending, if you look at the bottom left corner, the extra spending that comes from the portfolio until that first Social Security benefit starts, if that's three years for example, then you would use something that's a good fit for three years, like a three-year CD ladder.

And I think it's often helpful here to just forget about retirement planning for a minute and just think about basic personal finance. If you had some money that you were going to spend completely over the next three years how would you invest it? You probably wouldn't put it in the stock market. You'd probably use something like a three year CD ladder and that's basically what we're doing here. And then if the piece of the portfolio, the extra spending that's going to happen until the second Social Security benefit starts, if that's seven years, you might use a seven year CD ladder or a seven year bond ladder. And then the extra spending in early retirement just until the desired total level of spending has come down. If you imagine that's 15 years you could use a 15-year bond ladder, or maybe an intermediate term bond fund, something like that. 

And then it's really just the rest of the spending, the rest of the portfolio, where we have some questions because those CD ladder pieces and the bond ladder pieces, they pretty much solve themselves, right. If you know that you're going to spend this particular chunk of money over four years, a four-year CD ladder is an obvious solution. And the spending rate is obvious as well, you just spend one of those four CDs every year.

So for the rest of the portfolio, that's where we have more questions, and that's where you want to be using a low initial spending rate. So depending on what assumptions you're using some people say it should be three percent right now some people say four percent some people are saying even higher than four percent. But a low initial spending rate, and that's where you generally also want to be using a stock oriented allocation because this is the piece of the portfolio that basically has to last indefinitely.

Because no matter how old you get you don't know how much longer you have left and so you need this portfolio to last essentially indefinitely. So with this rest of the portfolio chunk, basically one strategy that has grown a lot in popularity over the last 10 years, and you'll see this in a lot of the research lately, is to base your spending on the RMD tables, and for those who aren't familiar RMD just stands for required minimum distribution. And this has to do with retirement accounts, basically traditional IRA, 401-k, 403-b, once you reach age 72 you have to take money out of that account every year.

And the amount that you have to take out is a percentage based on your life expectancy. So the older you are the more you have to take out. And the idea of using this as a spending strategy is that every year you would just look up the RMD percentage for somebody your age and then for the rest of the portfolio piece, that's the percentage that you would spend. So it's pretty straightforward. It's basically just a percent of portfolio strategy but it adjusts for age. Meaning that it's suggesting for the fact that the older you are the greater the percentage of your portfolio that you can spend.

So if you imagine somebody who retires early at age 50 and compare that to a 90 year old retiree, the 90 year old can spend a greater percentage of their portfolio safely every year because they don't need it to last for as many more years. And this strategy of basing spending on the RMD tables reflects that fact. This slide basically just shows what the percentages would be at various ages. And the thing I want you to notice is that it's a pretty conservative rate of spending, right.  You're spending less than three percent until you reach age 64. And you're spending less than four percent until you reach age 73.

And so what some people have decided to do is basically take this concept but just crank the numbers up a little bit. For example they might say we're going to spend the RMD percentage times 1.2 or something like that. And of course that's a trade-off. It means you get to spend more but because you're spending more you're increasing the likelihood that the portfolio ends up being depleted.

And on the topic of asset allocation for this rest of the portfolio piece, you generally want to be using an equity-oriented allocation. You wouldn't want ,for instance, to make a bond ladder like a 23-year bond ladder, because if you live 27 years then you have a problem. So the idea is that for a portfolio that has to last indefinitely, stocks are the best fit for that. But if you look at the research, you'll see that there's a really broad range in terms of what gets recommended. with some studies you might see a 40% stock allocation recommended. You might see that with target date funds for instance. Other studies will recommend all the way up to 100%  stocks for this piece of the portfolio, and the reason for that is just because if you're looking at historical data it depends a lot which periods you look at and it depends what assumptions you are making. If a particular study assumes that the stocks are all US stocks, whereas another study uses an international allocation as well, it's going to vary in terms of what it shows. So a broad range is acceptable, but generally something with a stock oriented focus is usually what works best. Now of course a high stock allocation means more volatility.

 And the other thing that it usually results in is a larger bequest. So a larger portfolio balance at death or more spending later in retirement. What surprises a lot of people is that using a higher stock allocation doesn't really let you spend more early on in retirement. A lot of people assume that it would because stocks generally have higher expected returns. But the problem with that is that stocks obviously don't always earn great returns, and so with a stock heavy portfolio you basically have to plan to get unlucky. You have to spend at a conservative rate and assume that you're going to get unlucky and spend at a rate that would be okay in that scenario.

But then what usually does happen is that you probably won't get really unlucky. You probably won't have a huge bear market right after you retire. And so what usually happens is that the portfolio grows over time if you have a significant stock allocation and so that means that you can spend more later on in retirement or leave more to your kids or grandkids or charities.

Chris: There was one question that came in that I think is a good one, and you've covered multiple topics already. I think the crux of the question is what are we talking about generally for the magnitude of difference between these multiple decisions we have to make that you just mentioned, are we talking about if you make the correct decision with your spouse on who files early, are you going to get twice as much money? Ten percent more money?  

And the same sort of question would just generally go first the sequence of returns risk that you were just talking about, and then this asset allocation decision.

Mike Piper: Yeah, so on the topic of Social Security it's super case by case, because the difference in earnings history matters a lot, and the difference in ages between two spouses matters a lot. The simplest case. So for a single person, like I said, filing later is slightly advantageous, not wildly advantageous. It's just that life expectancies are a little bit longer than those that are baked into the system,so you're going to get somewhat more, on average, if you wait. And how much more depends on what mortality assumptions you're using.

For a married couple generally what you'll see is that having the higher earner delay is extremely advantageous because you have this same fact that life expectancies are somewhat longer than those built into the system and there's the facts that are again, for that higher earner for looking at until both spouses die, period of time. So it's super duper advantageous.

But to put a number on it, unfortunately you can't because it depends entirely on the difference in ages, the health assumptions that you're making, and the difference in earnings history for the lower earners decision. Though what you'll find is that filing early is usually very slightly advantageous. It's pretty close to neutral because you have this one point that filing earlier is advantageous because of the way survival benefits work. But there's the countervailing point that life expectancies are longer than those assumed in the system. So it's almost neutral for the lower earners decision. It's usually you get a little bit more, on average, by filing earlier but it's not going to be a huge difference. But again to put any numbers on it. You just can't do it without knowing the exact couple that we're talking about. 

In terms of which assets to spend from, and how much to spend every year, I'm not sure exactly how to answer that. Certainly if we're talking about a scenario in which you are going to be spending well above four percent in those early years, basically until Social Security kicks in, that absolutely can be okay  if what's going to happen then is that once Social Security does start the spending rate that you're left with is pretty low. 

But in that case what you have to do is just what we're talking about, basically take a chunk of the portfolio and specifically allocate it to something very safe. So essentially, for that piece of the portfolio sequence of returns risk is eliminated because you just have a bond ladder or a CD ladder and you know that money will be there.

In the research over the last few years you'll see the term Social Security bridge or bridge payment used to describe that concept. I don't know if that answers the question or not.

Chris: Yeah I  think it did to some degree, so thank you. 

Mike Piper: Sorry, it's hard to answer. To put numbers on that particularly.

Chris: Yep, understandable. 

Mike Piper: So as far as which accounts to spend from, generally this is about tax planning and in most cases your taxable accounts are your least tax efficient accounts and the reason for that is that you have to pay tax every year on the rate of growth. You have to pay tax every year on the interest that you're earning unless you're using tax exempt bonds, but those pay less interest to begin with, so again you have a lower rate of growth. And depending on your income level you might have to pay tax on dividends and capital gains as well.So taxable accounts generally grow at a lower rate which makes them less tax efficient and that usually means that you want to spend those accounts first. 

Anything in tax planning though, there's always exceptions. One big one in this case is that if you have assets that already have large unrealized capital gains so they've gone up in value quite a bit since when you bought them and you're pretty sure that you could leave those assets to your heirs, so you know you're not going to be spending your whole portfolio during your lifetime. In that case it usually does make sense to try not to spend those assets.

And the reason for that is that when you die your heirs would get a step up in cost basis. Basically when they inherit those assets their basis would be stepped up to the market value when they inherit those assets and so the idea here is that if you never sell the asset during your  lifetime you won't have to pay tax on the capital gain and neither would your heirs. They could inherit it and sell it right away and they wouldn't have to pay tax on the gain either.

So after taxable accounts the things that most people have of course are Roth accounts and tax deferred accounts like traditional IRA and 401-k and the decision here in terms of how much to spend from each of these two types of accounts it's overwhelmingly a function of your marginal tax rate, which is just the rate of tax that you would pay on an additional amount of income. 

So for example if a thousand additional dollars of income caused your taxes for the year to go up by $300 that means you have a 30% marginal tax rate. And the general approach is that basically you would compare your current marginal tax rate to the marginal tax rate that you expect to have later on in retirement. And if your current marginal tax rate is lower than that future anticipated marginal tax rate that means you want to spend from tax deferred assets right now instead of Roth assets.

And the idea here is you're basically saying my tax rate right now is low so I’m going to take advantage of that low rate. I'm going to take money out of these tax deferred accounts right now and pay tax now at this low rate so that I don't have to pay tax later at a higher rate. And this is a very common thing in real life financial planning because what usually happens is that you retire at a given age, so your income has gone down dramatically because you've stopped working and Social Security hasn't started yet and RMDs don't start until after that.

So you've got this window of time, a period maybe of several years, or a year or two, where your income is lower than it will be, lower than it has been, and lower than it will be in the future. And that usually means a lower marginal tax rate. And so that usually means it makes sense during those years to spend primarily from tax deferred assets and maybe do Roth conversions as well, which is where you move money from a tax deferred account to a Roth account and you pay tax on it when you do that generally And the idea there is basically the same thing, you're saying I'm going to pay tax now at my low rate so that i don't have to pay tax later at a higher rate.

And then any years where the opposite is true, so any time where your current tax rate is higher than you expect it to be later that means you would want to spend from Roth assets right now and the idea there it's basically the opposite, you're saying my tax rate right now is really high so I'm going to try to keep my taxable income really low so that I don't have to pay any more tax at this high rate than i really have to.

Chris: Yeah, hey Mike a question came in. What if capital gains rates increase in the future. What would that do to the strategy you just laid out here?

Mike Piper: If you knew with a certain confidence capital gains rates are possibly going to be going up at some point, it basically just amplifies the desirability of spending down taxable accounts first. That makes your taxable accounts, they're already your least tax efficient accounts, most likely and it would make them even less tax efficient. So it would make it even, it's already probably advantageous to spend those accounts first, and it would make it even more advantageous to spend those accounts first.

Now an important thing to know about this is that it's not--some people misunderstand it and think that it's just a year by year analysis where for instance you might say this year we're going to spend entirely from Roth or entirely from tax deferred-- but in reality at least in theory it's a dollar by dollar analysis. So you would, theoretically for every dollar of spending you would be saying should this come out of tax deferred or should this come out of Roth.

And of course no one's actually going to do dollar by dollar because that would take forever. But you might do thousand dollar increments, or 5 000 ,or 10 000 or whatever you pick based on your total spending level and based on how much time you want to put into it. But the key point is that basically in a lot of years you spend from tax deferred up to a point because the more you take out of tax deferred the higher your marginal tax rate gets in most cases, and so it often makes sense to spend from tax deferred until you reach a particular threshold where your marginal tax rate would go up and then once you've hit that threshold basically you would spend from Roth for the rest of the year.

And this is something I harp on on the forum all the time, but it's really important. It's that your marginal tax rate. It's not just the same thing as your tax bracket. A lot of people just assume, for instance, I’m in the 22% tax bracket so I have a 22% marginal tax rate. But that's not the way it works in a lot of cases, especially in retirement because there's a whole bunch of things in our tax code where having some additional amount of income come in it causes the normal amount of tax based on your tax bracket, but it also causes something else bad to happen. It causes you to become ineligible for a particular credit or a particular deduction maybe, or it causes a whole other type of tax to kick in. 

And so when you account for those two effects together your marginal tax rate can be considerably higher than just your tax bracket. Some things that can cause that effect. The premium tax credit. So this is for people buying health insurance on the Affordable Care Act exchange. That's especially important for anybody retiring before 65. The way that Social Security benefits are taxed, that causes an effect. Like this also, the medicare income related monthly adjustment amount which we'll talk about in a second, causes an effect like this. The 3.8% Net Investment Income Tax causes an effect where the actual marginal tax rate is higher than just your tax bracket.

And really anything in our tax code that phases in or phases out based on your income level it's going to cause an effect where your tax rate, your actual tax rate, will be different than just your tax bracket.

Now this is another thing that I know isn't very intuitive. So let's go through an example. Let's imagine a single retiree who is age 70, and so they're already on Medicare and they're already receiving Social Security, and let's assume it's $30,000 of Social Security that they're getting per year. And we're also going to assume that there's nothing else complicated, no other complicating factors in this person's tax planning picture. So this is like the simplest retirement tax planning scenario that you would ever see.

And if we wanted to draw a chart of this person's marginal tax rate at different levels of income most people assume it would look like this, basically just stairs that reflect the different tax brackets. As this person's income goes up they go into the next tax bracket, so their marginal tax rate goes up accordingly. And that's reasonable to expect that. That's what it would look like, but it doesn't look like that. It actually looks like this. And this piece right here, it's known as the Social Security tax hump, and it's the result of the way that Social Security benefits are taxed.

And the way that works is that if your income is low enough none of your benefits, none of your Social Security benefits are included in your taxable income, so they're totally tax-free. But then as your income proceeds through a certain range, every additional dollar of income causes either 50 cents or 85 cents of Social Security to become taxable. So basically through this range of income every dollar of income is causing income tax based on your tax bracket but it's causing even more tax because it's causing some Social Security to become taxable. So that's why your actual marginal tax rate is way higher than just the tax bracket that you're in.

But then what eventually happens is that your income reaches the point where 85% of your benefits are included in your taxable income which is the most that's ever allowed to be included in your taxable income. So then this effect goes away and your marginal tax rate comes back down.

And then you'll notice the four straight up and down lines. They are the result of medicare IRMA, Income Related Monthly Adjustment amount, and that has to do with medicare premiums. And the way that that works is that your medicare premiums for a given year depend on your income level from two years prior. So a person's 2021 premiums would depend on their 2019 amount of income. And the key thing to know about medicare IRMA is that it isn't a gradual effect. There's just specific thresholds where once your income crosses that threshold your premiums two years from now go up dramatically. They'll go up by several hundred dollars, or more than a thousand dollars over the course of the year. So basically this one dollar of income that puts you right over that threshold, it's costing you hundreds or thousands of dollars in health insurance premiums. So it's a marginal tax rate that doesn't fit on the chart. It's just a straight up and down line.

And like I said a minute ago, the idea here is that basically every year in retirement you would create a chart like this,in theory at least, to show what your actual tax rate would be at different levels of income. And then you would pick one of these thresholds and try to keep your income just below that threshold so that you don't get hit with that higher extra amount of tax. And of course a CPA, CFP, tax attorney, somebody like that, can help you with that analysis.

But if obviously a lot of bogleheads are do-it-yourselfers, and so if you're going to take the DIY approach what I would encourage you to do is not to try to do all the tax calculations in a spreadsheet because almost every time when I see people do that they leave something out, they forget about a particular deduction, or a particular credit, or some phase out range and so they miss one of these thresholds. They don't recognize that it's there, and then you accidentally take too much out of tax deferred, or you do too much Roth conversions and you blow right through that threshold and you end up paying more tax than you had to.

But if obviously a lot of bogleheads are do-it-yourselfers, and so if you're going to take the DIY approach what I would encourage you to do is not to try to do all the tax calculations in a spreadsheet because almost every time when I see people do that they leave something out, they forget about a particular deduction, or a particular credit, or some phase out range and so they miss one of these thresholds. They don't recognize that it's there, and then you accidentally take too much out of tax deferred, or you do too much Roth conversions and you blow right through that threshold and you end up paying more tax than you had to.

And because it's TurboTax it is accounting for all of the deductions and credits and so on, all the other accountings, and so you can see what your actual marginal tax rate would be for that thousand dollars of income. And then you do it again, you say another thousand, and another thousand, and you can figure out your actual marginal tax rate at different levels of income.

But the one thing that you would have to manually include if you're doing it that way is the medicare IRMA  that we're looking at right here, because that's not technically a tax, it's medicare premiums. So TurboTax doesn't care about that so it's not going to tell you about that on its own. So you would have to manually include that in your analysis.

Now one more point about this topic of marginal tax rates and how that influences the decision of which accounts to spend from every year, is that for a married couple what usually happens when one spouse dies, is that the surviving spouse is left with a higher marginal tax rate going forward and the reason for that is that the standard deduction for one person or a single filer is half of what it is for a married couple filing jointly and up to the 32% bracket each tax bracket has half as much space in it for a single filer as for a married couple filing jointly. 

But when one spouse dies the usual result is that the household income falls by less than half because it's generally the smaller of the two Social Security benefits that goes away and then with regard to the portfolio income, it usually doesn't really change all that much at all just because somebody died. I mean the portfolio is still there right. It's still there doing what it does. 

And so you have half the standard deduction and half as much space in each tax bracket, but more than half as much income so the result is often that the surviving spouse has a higher marginal tax rate after that point. So the implication there is that it often makes sense during the period that both of you are retired and both of you are alive, it often makes sense to do somewhat more tax deferred spending and likely Roth conversions than you would if you weren't accounting for this fact.

Now in fairness I should note that this is not a very precise analysis because, of course, you don't know really how long you're both going to live or how long just the one person's going to live after that. So there's a lot of guesswork involved here but even when you recognize that there's guesswork and uncertainty it still usually makes sense to do somewhat more Roth conversions or tax-deferred spending when you're both still alive.

So just summarize this, which accounts to spend from, taxable accounts are usually the least tax efficient so you usually want to spend from them first. An exception would be highly appreciated assets. You often want to try to plan to leave those to your heirs or donate them to charity. And then with regard to spending from Roth versus spending from tax deferred, it's pretty much about your marginal tax rate and generally if your current marginal tax rate is lower than you expect it to be later on you want to spend from tax deferred right now, and if the opposite is true, so if your current marginal tax rate is higher than you expect it to be later on then you want to spend from Roth right now.

Chris: So just moving on to the last question,and we're going to go through this one pretty quickly, insurance during retirement.

Mike Piper: Your insurance needs basically aren't the same as they are during your working years.Life insurance, for instance, you generally don't need it in retirement because the whole point of life insurance is to protect people who are financially dependent upon you in the event of your death. But usually by the time you're retired nobody is financially dependent upon you. One big exception is for a married couple, in which one person has a pension and the survivor benefit is pretty low. Then you might want to buy life insurance on the life of the person with the pension. Another big exception would be if you do actually have dependents. So if you have a minor child or an adult disabled child ,then you likely would want to have life insurance in order to financially provide for that person in the event of your death.

Long-term care insurance, almost everybody I know hates talking about this because there's nothing really good to say if you're single. One strategy that might make sense is just to count on medicaid, by which I mean you're basically betting that either a.)  you won't need long-term care; b.) if you do need long-term care you won't need it for very long so you can just pay for it out of pocket; or c.) if you do need it and you need it for a long time then you would be okay with spending down your assets to the point at which you would qualify for medicaid. And medicaid can pay for it going forward. Of course an important point to note with that plan is that medicaid doesn't cover every type of facility out there so you would want to make sure that you are okay with the facilities that medicaid does cover.

And then of course for a married couple that strategy doesn't work nearly as well because if one spouse needs long-term care and they end up spending down the portfolio to the point at which they qualify for medicaid that leaves the other spouse without a whole lot to live on. So there's more use for insurance products for a married couple.

So you've got the traditional long-term care insurance option, although a lot of insurers aren't providing that at all anymore. They're not even offering it, and of course, you've probably heard that for people who have those policies they've generally been hit with pretty significant premium increases, which leaves you without any particularly good options. You can either just pay a higher premium, abandon the policy, or choose to have your benefits cut. Basically none of those feel very good. 

Another option is hybrid policies. So either annuities or life insurance policies, basically with the long-term care rider. So it's a rider that says that if you need long-term care then the policy will pay a certain amount of benefits. But just like with any insurance product, they're priced in such a way that's going to be profitable, on average, for the insurance company. On average it's a losing bet for you and a winning bet for them.

And so that generally means that if you have enough of a portfolio that you are very confident that you could just pay for long-term care out of pocket, that usually means that you wouldn't want to buy a policy of that nature. If you wouldn't be able to pay for it out of pocket, you have more use for insurance. But even still it's just that the options are not all that great. Unfortunately they leave something to be desired.

Disability insurance is usually a pretty easy question because the whole point of disability insurance is to replace your income from your work, right. If you're still working, you get disabled you no longer have income. So disability insurance protects you in that scenario. But if you're not working anymore you don't really need that. 

Health insurance. Of course if you retire before 65 most years you are generally going to be buying insurance on the Affordable Care Act Exchange. So as a tax planner the thing I'm always thinking about there is the premium tax credit. That's the the tax credit that basically reduces your premiums. It's a subsidy against your premiums and it's based on your income level, and health insurance is expensive especially the older you get. So this can be a really, really big credit and it's worth paying a lot of attention to, that's all. It's worth really thinking about navigating your distributions from your accounts in such a way that you might want to keep your income low during those pre-65 years in order to maximize that credit.

After 65, of course, that's where medicare  kicks in and so, again, that's where you want to be thinking about the income related monthly adjustment amount, and remember that's the one where it's specific thresholds, where as soon as you cross that threshold, your premiums go up quite a bit. So if you can keep your income just below one of those thresholds rather than just above it,that's generally quite advantageous to do so.

As far as longevity insurance, so things that protect you in the event that you live a really long time, that's where annuities come into play. Obviously bogleheads don't really love most annuities, and the thing you'll see people say on the forum a lot, which is true, is that delaying Social Security is usually the best annuity deal that you can get. And again that's because most insurance products are priced in a way that's profitable to the insurance company, but that's not true with Social Security. So it's usually a better deal than you could get from an insurance company.

However if you're already delaying Social Security and you want more longevity protection, that's where an annuity might be helpful, and the type of annuity that's generally considered boglehead's approved is the basic single premium immediate annuity which is a very simple product. It's just the pension that you buy from an insurance company. So you give the insurance company a lump sum of money and it is now their money. iI's gone. It's not like it's in an account and you can pull your money back out, it's their money now. But now they promise to pay you a fixed amount every month, or quarter, or a year for as long as you are alive. Or you can get a joint life annuity, so it's as long as you and the other person are alive.

One important point about this is that since 2019 there are no longer any insurance companies that offer this type of lifetime annuity with a cost of living adjustment that's based on actual inflation. So if you're buying this type of product you will be taking on inflation risk .And that's another reason why people generally consider delaying Social Security to be the best option in this category, is because Social Security benefits have an actual inflation adjustment so they don't have that sort of inflation risk.

And that's really all I've got on the cookie cutter retirement plan concept. I hope you got something useful in terms of answering the other questions in retirement planning other than just when you can retire.

SS critic: And I'd like to go back to the idea that the low earner should  file early. That's under the assumption that the high earner is going to delay. If for example, the high earner claims at 62 and is thus getting a lower benefit, and if the low earner waits until age 70 and gets the delayed retirement credits, the low earner can have a higher benefit than the higher earner did, and that flips the thing all around. Which means when the first, when the high earner dies, the low earner keeps getting her higher benefit, or his higher benefit, and the reverse. 

So anyway, so the example Mike gave, which is correct, is if you have minor children, the higher has the minor children, you may want to claim early and thus the argument basically flips on its head then. 

Mike Piper: Yeah, that's a great point. I was assuming that the two spouses are coordinating in their plan, but of course that might not be true. The higher earner might have filed early even if that wasn't wise to do. So yeah, in the example that you gave with the minor child it could very well make sense for that higher earner to file early so the child can get that child's benefits on that higher PIA. And then the lower earner waits to increase their own retirement benefit and plausibly the survival. Of course it depends on their ages. So if the higher earner is older, the higher earner is going to hit 62 with that minor child years before the lower earner could hit 62. 

SS critic: Yeah definitely depends on ages and difference in primary insurance amounts too.

Audience question: I’m a high income guy. I'm a doctor. I actually got into your stuff from the White Coat Investor. I'm in New Jersey, so high tax bracket. I'm putting in a lot in tax deferred but then, like when is it in general, assuming it's similar tax rates, then I might have an RMD problem. Really, like maxed out tax preferred for the highest tax bracket but just, in general, when should I say like, oh man maybe I'm not going to get that tax arbitrage anymore, maybe is it five million in tax deferred. I plan on retiring at 65 and then taking Social Security at 70. 

Mike Piper: So do you mean in terms of the tax arbitrage like I'm going to have an RMD problem if my tax deferred is too high.

Audience question: Right, we're like, oh my god I'm now at the highest tax bracket in retirement based on the RMD. I didn't get the tax arbitrage of tax deferral in my peak earning years.

Mike Piper: And so are you asking whether it makes sense to be doing Roth conversions or Roth contributions instead of tax deferred contributions or because are you doing something about just doing it. 

Audience question: Yeah because that will say ,oh at this level of tax deferred when I retire at 65, I really should do this amount of tax Roth conversions between my Social Security claiming and the time I retire. So that's going to be a five year window where it's just nice to know, oh my god, I'm going to have an RMD problem, I got to do Roth conversions at age 66, 67, 68, 69.

Mike Piper: So I don't know if you have a number, just that's in general.

Audience question:  I don't have a number because, yeah it's going to depend. 

Mike Piper: I mean that's just not the way tax planning works because it's going to depend entirely on what other things you have going on in your tax planning picture. But so I would say if you have a Roth contribution option available to you, that could be definitely a point in favor of doing Roth contributions. But if what we're talking about is just either doing tax deferred contributions or just regular taxable saving, it almost always makes sense to be doing the tax deferred contributions even if there isn't the tax arbitrage concept going on, because with tax deferred accounts, again, at least they get to grow at their whole rate of growth, whereas with the taxable account you're paying tax every year.

So if you've got a Roth option, and you are really, really got quite a bit in tax deferred, then absolutely doing Roth contributions definitely can make sense, even if you are a higher earner. Another reason for that is that effectively with Roth you get to save more in a tax advantaged way because with tax deferred essentially the government owns a portion of that account. If you were planning a 30% tax rate in retirement you can think of a tax deferred account as if it's a Roth account of which you only own 70% and the government owns the other 30%. And so basically your contribution limit, every year the government's getting to use a portion of it, is kind of a way that you can think of it  And so Roth effectively lets you save more in a tax advantaged way. So even if you have a high tax rate it can definitely make sense to do Roth contributions.

But if we're talking about just not doing tax deferred contributions and instead doing regular taxable saving that can make sense, but it's it's pretty uncommon, especially the further away you are from retirement. Such that the longer the period of time that this money would be compounding then the more important it becomes to get all of the compounding rather than having a portion of it disappear to go to pay taxes.

Chris: Okay, cool. Also a related question with just looking at state taxes I know there's no income tax states, but are there, I think, I guess maybe Tennessee and Vermont might actually tax withdrawals from retirement. Are there any other states that you know of that are like, well wait, retirees should not retire here because even though there's low income tax or whatever retirement income or drawing from those accounts are actually taxed, and are actually much higher, you know any sort of state sort of tax bombs for retirees?  I know it's sort of nuanced with every state.

Mike Piper: Yeah. I'm not honestly able to speak off the top of my head to the different tax breaks, but definitely, certainly some states have no income tax. Others have income tax but it doesn't apply to Social Security or retirement account distributions.  And then some states have tax that applies to all of those things.

Chris: We have another raised hand from Duncan, do you want you to ask your question?

Duncan: In regard to the spending rules for retirement you mentioned the RMD rule and I'm wondering what you think about using an endowment spending rule where you are using a smoothing process where you take a fixed percentage of the last seven years average value of your portfolio? 

Mike Piper: Yeah, so I think of retirement spending strategies as basically existing on a spectrum, and at one end of the spectrum is the pure percentage strategy where you say we're going to spend three percent every year and then at the other end of the spectrum is the four percent rule concept, where despite having a percent in the name, it doesn't actually fluctuate, you're not letting the spending fluctuate based on the portfolio balance, you're just spending the same dollar amount every year, or inflation just a dollar amount.

And so the further you are toward that end of the spectrum where you're spending the same dollar amount and you don't adjust based on portfolio performance, the riskier it is because you're not cutting spending when the portfolio declines.So the endowment strategy is kind of in the middle of those two extremes. So it's riskier than a strategy that just says I'm going to spend three percent every year, or three and a half percent, or four percent, or whatever it is, so it's riskier than that because you're not adjusting spending as much as that strategy would. But it's not as risky as the classic four percent rule. So I think it's one of many very reasonable approaches I guess is what I would say.

So this is Open Social Security. For those of you who haven't used it-- I gather many of you have--so Open Social Security is hopefully pretty straightforward to use. tThe base case here, it just takes your very basic inputs: marital status, gender, date of birth, primary insurance amount, and you can see, as we were talking about, the very basic scenario, an unmarried woman should file at age 70 because it's slightly actuarially advantageous. But we can start to come up with all sorts of other scenarios. So,for example, we were one where you were talking about this box at the top gives you additional options. So if you say this person now has a child and this child is disabled so they could get child benefits on this person's work record. Now she should file as early as possible.  Basically it's just a very case-by-case analysis and one change in the inputs can completely flip the output on its head. 

And I know most of you have already been using this so I don't want to just show you a million different scenarios, but one thing that I don't see used a lot on the forum which I think is pretty useful is this feature right here. You can put in all of your inputs, so let's say you've got one person born in 1958 with a higher primary insurance amount and then if you want to, and obviously this isn't all that many inputs, but if you do the advanced options and then one person is still working and maybe there's a government pension and there's a whole bunch of input you've put in, you can right click right here and save that link. And then you could just paste that in. 

I mean either to another browser window later, or you could post it on the forum. And once the page loads it'll be pre-loaded with all of your inputs. And that can be useful, for if you want input on a particular, like if you want to talk on the forum about a particular scenario, you can put in all of your inputs, grab that link, and then paste it into a forum post. And then other people can look at it and see the exact scenario that you're talking about. That's also useful if you work with a financial advisor. For instance you could copy that link and then include it in an email. So that's useful there.

I think that's mostly it. That I wanted to show people  that you've got all the different options here. So disability is obviously relevant, still working is where we've got the earnings tests coming into play, pension is for applying the windfall elimination provision and government pension offset rules, you can use different mortality assumptions. So the base case is we use the Social Security’s period life table, but you can use other ones based on your health status, or you can pick a particular, you know, let's assume that I die at age 84. You can put that in if you want to. Children are relevant because child benefits are a big deal.

The discount rate, this is basically what's the inflation-adjusted return. It's the take the money and invest that option, or alternatively if you delay Social Security, you're basically spending down your portfolio at a faster rate, so that means that some piece of your portfolio is no longer invested, so you're giving up the returns on that piece of the portfolio. So the discount rate is essentially reflecting that by default. It pulls in the yield on 20-year TIPs but you can adjust it to be whatever you want. You can also have this option to assume that benefits will be cut at some point in the future. You can choose the year and the percentage, so you can see how that changes the analysis. 

And I think that's mostly it. I know that most of you, frankly, have been using this for a while now and I see it all the time on the forum anytime somebody asks about Social Security. People are coming in with what, the link to this calculator.

So if you guys have specific questions about it I’m happy to answer them but I mostly just wanted to show you what the additional options are because I think some people miss them and just this little link down here at the bottom of the page it's really useful. It saves you having to put in your inputs over and over and over.

Eric: Mike this is Eric. I did miss those options up top and I have a question about the still working option. I assume that's for somebody who is old enough to retire, old enough to collect Social Security at this point in time, or is there some way you can assume that in the future, if I'm trying to plan that, that I would still be working. Or how does that work in the context of the calculation.

Mike Piper: So with the still working, this is specifically with regard to the earnings test. So it, the earnings test,  applies for people who are--they've already filed for benefits and are younger than full retirement age. The earnings test can result in your benefit being either partially or completely withheld.

Eric: Yeah, okay. And so this would account for that. But if what you're wanting is how would my primary insurance amount change based on additional years of earnings.

Mike Piper: This calculator won't account for that just because by definition it takes your primary insurance amount as an input. So you have to tell it what your primary insurance amount is; ssa.tools is another calculator made by another boglehead that could help you with that calculation though.

Eric: Yeah, and I've actually- the PIA amount for those of us that aren't quite to retirement age yet that also assumes future years of earnings equal to or about the same as what you're earning today, I think, right. So that if you were to retire, if you were to quit working early, that number actually might be inflated a bit, what you get from the Social Security statement itself.

Mike Piper: Yes exactly. The number on your Social Security statement is not actually your PIA, that's the key point to know. It's an estimate of what your PIA would be if you continue to earn the same amount that you've been earning until the age of which you file.

Eric: Thank you, sir.

Chris: So at this point I think we can have time for a lot of questions and answers from Mike. We have a couple that have come in through the chat too.

Steve: Yeah, thank you. I have a question on delaying Social Security. If I'm the high wage earner and I’m planning to delay benefits until age 70, and my spouse is the lower wage earner and she plans to collect hers as soon as she can, let's say 62. When I begin to collect mine at age 70 does her Social Security payment change? I think I was under the impression that hers changes to half of mine at that point, but I might be wrong.

Mike Piper: So her benefit, as your spouse, if she's already receiving her retirement benefit at that point in time then when you file for your retirement benefit, her benefit as your spouse would automatically kick in that's called a deemed filing. And that benefit amount, it's only, it's not half of your benefit, it's half of your primary insurance amount, so it's half of what you would have gotten at your full retirement age. So if you filed at 70, you're getting more than your primary insurance amount and she would not be getting half of that. She'd be getting half of your primary insurance amount, so less than half of your benefit.

The other key point here is that if she's 62 at this time, or if she's younger than full retirement age then her benefit as your spouse is going to be reduced because it will be beginning prior to full retirement age. If you want to email me, I  could, for example, give you some actual numbers using certain assumptions. If you wanted to see how that would actually work out. But she would be getting less than your actual, less than half of your benefit is the answer. 

Steve: Okay thank you very much.

MIke Piper: Sure.

Chris: I've got a quick one from the chat here. Is there a difference between a divorced person and just a single person in terms of how you strategize.

Mike Piper: There can be. So if you were married for at least 10 years prior to the divorce, then you basically would qualify for spousal benefits on your ex-spouse's record, in the same way that you would if you were still married,roughly speaking. So if you are the higher earner of you and your ex-spouse, then it doesn't really matter for your own planning, unless you, for example, if you delayed taking your benefit, it would increase your ex-spouse's survivor benefit if they were to outlive you. So if you're still on good terms and that's something you would want to do, then go for it.

But if you're the lower earner then it really, really can change the analysis. So for instance I was just talking to someone the other day who is, I think she's 10 years, she's several years younger than her ex-spouse. So the ex-spouse is the higher earner and quite a bit older. So for this person, it makes sense to file early, start her own benefit as early as possible, because at some point, most likely she's going to start getting that survivor benefit on the ex-spouse's work record because the ex-spouse is a male and much older. So it's likely that she's going to outlive him. So she could start getting her own retirement benefit as early as possible and then start getting a survivor benefit on the ex-spouse's record if or when he dies.

There are also spousal benefits for ex’s so that could come into play. One thing a lot of people aren't familiar with. You can file for a wife or husband's benefit as an ex if your spouse has filed, your ex has filed, However if the divorce is more than two years then you don't have to wait for your ex to file. You can still claim as a spouse against your ex just like that. That's called independently entitled. 

Chris: Good answer.

Mike Piper: I'll just add that that's especially relevant for anybody who's still able to do the restricted application option. So anyone born before January 2, 1954, you could get a spousal benefit on your ex-spouse's work record while waiting for your own retirement benefit to max out at age 70. 

Chris: All right, here's another question.You know Roth seemed like a great idea in a lot of ways but is there any discussion going on regulatory wise that they may eventually be taxed. Or is that something we should not try to concern ourselves with. 

Mike Piper: Is that a thing I've heard proposed? Yes. But there's a million things proposed that never happen. And then I mean among tax professionals who I know, I don't know anyone frankly who was anticipating the Tax Cut and Jobs Act. We were expecting, maybe, the standard deduction will be dramatically increased and exemptions will go away. And the itemized deduction for state income taxes will be limited to ten thousand dollars. I don't-- people aren't very good at predicting legislative changes.

The thing we talk about on bogleheads all the time is how hard it is to time the market, time the stock market, and with the stock market there's really only three things that can happen right, it could go up, it could go down, or it could stay flat, and even in that really simple case it's hard to do. 

With taxes there's a million things that could happen just with Roth IRAs. It's might they become taxable, might the rules for inherited Roth IRAs change, might there be RMDs later, might the contribution limit be increased to the 401-k contribution limit. There's a million things that could change.

And then there's still the timing question. So if we were assuming, so just the case we're talking about Roth IRA distributions becoming taxable, well when does it happen. Does it happen five years from now, twenty-five years from now? If you get the idea right but the timing wrong it still doesn't help. Just like if you predict a bear market but you predict it five years in advance and so you miss five years of stock market run up, it doesn't really help to have predicted the bear market. It's similar here. You have to get the timing right. But you also have to get right what's even going to happen. 

And then taking it a step further there's for a lot of things, there's the question of  the numbers. If you're going, for example, one common thing that people talk about is the income thresholds for Social Security benefits being taxable. Those were set in 1983, and they're not adjusted for inflation. So they're still what they were in 1983. So sometimes people talk, maybe those benefits or those thresholds will be increased. But what are we going to increase them to? And so unless you get the numbers right your tax planning based on that prediction isn't going to be very useful.

So tax planning based on predictions about legislative changes, it's extremely hard. It's much harder than timing the stock market, which is already as we say on bogleheads,not worth doing. And also just to comment. Because it's kind of fruitless to talk about it, it's a prohibited subject on the forum. So as far as upcoming legislation, I think that's probably one of the good reasons why. 

Chris: So I agree. Brock, do you have a question?

Brock: Yes, thanks Mike, great info. If you can drill down into the long-term care insurance option, or more broadly,  what you said was look if you have enough in your portfolio to handle your health care needs during retirement you may not need long-term care insurance. And I'm wondering how to intelligently look at this topic. You know, I have not looked at the bogleheads forum or anything like that yet, so is there anything you could say to someone like me that's just beginning to explore this topic, to understand how to approach healthcare in retirement and how much to save for it.

Mike Piper: So are we talking healthcare or long-term care?

Brock: Yeah, like I  think healthcare costs in general and of which you know long-term care insurance is one strategy, right, to handle that well.

Mike Piper: So long-term care insurance is specifically covering long-term care, so care in a nursing home for instance, or an assisted living facility. Whereas healthcare we're talking about medical procedures, drugs, things like that. So long-term care insurance is covering something different than what your health insurance covers.

Brock: Okay, okay yeah. So I think in that case y let's just stick with the long-term care piece okay and how to handle those costs. 

Mike Piper: Sure. During retirement I i think step one, a useful step, is to look because it varies a lot geographically, you know if you live in San Diego long-term care is going to cost more than if you live in rural Missouri. So look at the facilities around you and see how much they actually do cost. And that's going to be step one, actual just research where you live.

And then you can look at statistics on how long long term care needs typically last. But of course you don't necessarily just want to plan on the typical length of stay, just like when you're retirement planning you don't want to assume that you're going to die at your life expectancy because you might live well beyond that, and then you need your savings to last longer than that.

So same thing with long-term care, if you can afford the average length of a long-term care need but no more than that, well then that still could be quite a problem. But you can look at statistics about how long long-term care needs typically last and you can look at the actual prices where you live. That's what I would do in terms of how to start to gauge whether it's something you could pay for out of pocket or whether you might be wanting to look into insurance to cover that sort of cost.

Brock: Okay thank you.

Mike Piper: Sure.

Diane: A very early question that came up which had to do with what used to be called the stretch IRA and the question comes from Ben. Do heirs have to cash out both inherited traditional IRAs and Roth IRA within 10 years of receiving them as of 2020. 

Mike Piper: It's for both.  So in certain cases, depending on who inherits the IRA, it might have to be distributed over a ten year period now that applies for traditional IRAs as well as Roth IRAs.

Stearns: I live down in Worthington not too far from South Dakota which has no personal income tax and I’m wondering if it might be a wise strategy to move to South Dakota for a year or two and do some conversions there, and then move up to Minneapolis where the grandkids are and spend money like mad. But you know given this low interest rate environment we're in. 

And I'm probably two years away from retirement and I claim Social at 70. But yeah the bulk of the assets are in the traditional IRA.. So I mean I'm looking at a pretty big tax bill from Minnesota and from the Feds in the future. So I've got maybe two or three years before starting Social Security and there'd be no income. We'd live off the brokerage account basically, which is already tax paid. So I would use those funds to live on and maybe pay income tax if we moved to South Dakota and started doing Roth conversions. So you may have heard this story before but what do you think?

Mike Piper: Yeah. I mean obviously it depends on how willing you are to move, but it's not out of the question. I would specifically look into the actual dollar amount of savings because, of course, moving itself involves a fair bit of financial cost and a fair bit of work. But it's not out of the question, people do certainly move for tax reasons.

Stearns: Okay all right yeah. We're only 50 miles away so it's not a big deal. We could rent over there, and then you know buy a house up in the cities if need be, but thank you.

Mike Piper: Now I would note though that you would want to check to make sure that you're not still-- so if you still own your home-- you know if you're just renting and if  you don't just make sure that you're not accidentally still going to be paying state income tax.

Stearns: Sure, right, sure. And then as a risk averse person, you know, up in the late 60s, with the Federal Reserve throwing us savers under the bus, there's probably not very many opportunities to get a decent rate of return that's pretty low risk at this point. Yeah wait maybe 10 years for interest rates to go up again. 

Mike Piper: So Alan Roth spends much time on bogleheads, you're probably familiar with him. Something he often suggests is looking at CD rates because in many cases banks will offer CD rates that are quite a bit better than you could get on bonds, especially bonds of a similar credit quality, which would be treasury bonds, assuming you're staying under the FDIC limit. 

Another thing that a lot of people don't realize is that if you have a mortgage you've got right there a risk-free rate of return available to you paying down that mortgage. So those are a couple of options, but yes absolutely, when interest rates are low there aren't very high return options available in any safe way.

Stearns: All right, thank you Mike.

Audience question: Referring to using the RMD's distribution option as instead of four percent withdrawals, when I try this out on portfolio visualizer, the monte carlo simulation, all the RMD option always comes up with over 90 percent maximum withdrawal or something. I assume it has to do with  exhausting the portfolio. Is there a way to modify, kind of play with this to see what else is there that can avoid looking at the 90 percent withdrawal. Thank you.

Mike Piper: Sure. So I’m not especially familiar with Portfolio Visualizer. I've used it a little bit but I haven't done it enough to speak to all of the various options that it has. But what a lot of people do is make adjustments to the RMD strategy. So they keep that same age-based percentage concept, but you could spend a greater or smaller percentage every year but still have something that's based on the idea that the older you are the greater the percentage that you could spend. You could take the RMD percentage, multiply it by something greater than one or something less than one, depending on what change you want to make,and of course again, the trade-off is exactly what you'd expect it to be. The more you spend in early retirement the more likely it is that you end up depleting the portfolio.

Chris: Thank you, Mike, great. There's another question that came in on chat. How would you factor in a pension in the timing of when to best claim Social Security benefits, and which  in retirement, Roth or Roth versus traditional, when factoring a pension.

Mike Piper: Sure  with the pension it's important to know whether it's a pension from non-covered employment. So a government pension. So then we need to be thinking about the when fall elimination provision and government pension offset.  Also important question is whether you have options, and to what extent you have options in terms of when the pension begins, because you might have an option to delay the pension and then you can have basically this case where, like we were talking about, where you have some window where your income is relatively low because you've retired and the pension hasn't kicked in yet.

And maybe you're also delaying Social Security, and then you can be doing some Roth conversions during those years. But the analysis is still basically the same in terms of which accounts to spend from. It's still about what's your marginal tax rate right now, and how does that compare to the marginal tax rate you expect to have later. The point is just that the pension affects both of those questions because it's providing income and likely changing your tax rate. But the analysis is still exactly the same concept. It's how does my current marginal tax rate compare to my future marginal tax rate.

Chris: I have another question Mike from the chat. So we had one question, says, I understand that Mike uses the Vanguard LifeStrategy Growth Fund. If he were to use a DIY portfolio would he have changed the asset allocation stocks/bonds ratio over the last two years.

Mike Piper: I would not have changed that. I would have eliminated the international bonds, definitely, no question about that, but I wouldn't really have changed the stock/bond allocation at all.

Chris: What percentage of a portfolio should be invested in real estate?

Mike Piper: If we're talking about Reits, so real estate stocks, I think a total market allocation makes perfect sense. So that would be, I think, somewhere between three and four percent basically. There's no particular need to have an extra Reit holding to overweight that industry any more than you necessarily need a fund to overweight the financial services sector or any other particular sector.

If we're talking about how much of your net worth should be allocated to investment real estate, so you know a condo that you're going to rent out or something, that's a tricky question to answer.Investment real estate is an undiversified asset as compared to an index fund which is extremely diversified. When you buy a rental property you have a lot of risk involved in this one property. And then there's additional risk because generally with investment real estate it's a leveraged investment because you're borrowing a good chunk of the money used to make the investment. 

So I think for a lot of people the answer is zero. Basically you can own your own home and then there isn't really any need to do any individual investment real estate. But for people who want to take on that risk, it certainly has the potential for very high returns but you should just recognize that it's an undiversified leveraged investment so there's quite a bit of risk involved.

Chris: Just to just to review, I think, and just make sure I understand. The Total Stock Market  Index Fund includes some like three to four percent of real estate, right, so any anything you buy you're actually overweighting, right. 

Mike Piper: Yes, that's precisely it and some people like to do that. But you could overweight, you could make a case for overweighting other industries as well if you wanted to.

Chris: One of the other questions that came in advance asks the following. What are your overall recommendations for an early retirement strategy? If details are helpful assume a two-worker household in their mid to late 40s with two children graduated from high school. In about 10 years mortgage paid by children's graduation, retire at children's graduation, at 100,000 a year. College funded via 529s or loans or other. So a very early retirement scenario.

Mike Piper: I think the things to primarily recommend are make sure you're using a really low spending rate because this portfolio might have to last for a very long time, and for the exact same reason you probably want a pretty high stock allocation because that's what's most likely to sustain a given level of spending over a very long period of time.

Chris: As far as further details beyond that, like the tax planning details, you know we just need more information about the individual person.  

Mike Piper: Yeah, that's mostly all I can say is high stock allocation very low initial spending rate.

Chris: One of the other questions that came in is where at this time is the best place to put your money for a five to ten years investment?

Mike Piper: Just like any other time we're talking about investing it depends on risk tolerance, right. So if you say five to ten years, well is it money that you are going to be spending definitely on year five, and on year six, and in year seven and eight, and just like that. Or is it maybe you're going to be spending it in year five, but if things don't go well you're fine spending it in year ten. So it depends on that. It also depends on if this money that absolutely needs to be there at the time, or is it something where you'd be okay spending somewhat less because you took a risk and it didn't pan out. So it depends entirely on risk tolerance.

I think stocks are clearly very risky over periods of less than 10 years so you could use a modest stock allocation, so 20% stocks/ 80% bonds, that's not insane. But I think it also makes perfect sense just to buy a seven year CD for money that you're spending in seven years. So it depends entirely on what the actual goal in question is.

Chris: There's a question that came in on the pros and cons of international bonds.

Mike Piper: Sure the pros. In theory it provides a diversification benefit because the things that affect interest rates in other countries are not the same things that affect them here, specifically right now. The cons are if you compare Vanguard Total International Bond Index Fund to the domestic version you'll see that it has a significantly lower yield and significantly more risk, so if you look at it as a standalone investment it doesn't exactly look like a winner. But in theory it can provide some diversification benefit.

One of the reasons I don't find that argument to be especially compelling is that you don't really need diversification in the fixed income part of your portfolio, because there's always CDs and treasury bonds. You can put it all in one type of thing,whereas with the stock portion of your portfolio you need to diversify. You don't want it all in one stock, that'd be insane.

But with the bond portion of the portfolio you can use just CDs and that's not diversified, but it's not as if it's risky because you've got FDIC insurance. So in theory there's a diversification benefit but right now international bonds have lower yield for higher risk which doesn't seem very appealing to me.

Chris: What percentage of international stocks do you recommend in a balanced portfolio?

Mike Piper: I think the starting point for analysis should just be the market capitalization, the market weighting. So what you would see from the Vanguard Total World ETF, for Vanguard Total World Index Fund. But that weighting of course varies over time because sometimes the US market does better, which means that after that happens a greater portion of that allocation would be domestic. 

But then I think you want to make adjustments from there, and for most people what you'll find is that you generally should end up skewing towards domestic because the primary reason is that living here in the US we generally spend in dollars and so that means that international funds have an additional source of risk, which is currency risk. It's basically the risk that the performance is affected because the currency in which those other investments are denominated decreases in value relative to the dollar. So once you account for that you generally want to skew somewhat towards the US allocation and that's especially true for retirees. So basically once you're actually spending from your portfolio that becomes more relevant. 

So you'll often recommend it, but you know it varies depending on what the best is, it varies depending on what historical period you look at and so on.

Chris: I've got a question. You've talked about CDs and treasury bonds. One of the things I learned from the boglehead forum about initially, were I-bonds and then also EE-bonds. Could you comment on those two options to fill that portion of the portfolio that you showed to pay for spending early on in retirement. In fact, if you would show that graph again, I don't know if you could do that, but that was one of my favorite slides.

Mike Piper: Okay. so just on the topic of other bond options. I-bonds, yeah they do often get left out, and I'm as guilty as anyone. I didn't even mention them today but they're very useful because like TIPs they're providing an inflation-adjusted return but they don't have market risk like TIPs do.  It's either you have to buy them through treasury direct, which some people don't like, or you can, up to a limit--Harry Sit, the Finance Buff wrote about this recently, you can intentionally overpay your estimated taxes or have a little bit too much withheld on purpose, and then basically one of the ways you can get your refund for the year is have them automatically purchase I-bonds for you.  Also another advantage of any type of treasury bond is that it's exempt from state income tax. So if you're looking at savings outside of retirement accounts that's an additional advantage there as well. 

Just to comment on EE-bonds and  I-bonds and if they could fill one of those slots. Sure, absolutely. So this space here you know we're going to spend this money over this predetermined period of time, absolutely other types of treasury bonds can be a great fit for that. I would say especially for the longer windows, because with I-bonds, one of the advantages is the inflation protection. Of course inflation protection over two years is less of a big deal than inflation protection over an extended period. So I-bonds can be an even better fit for that dedicated spending over an extended period. That's a great way to reduce the inflation risk that you're facing there.

Because that's basically the only risk you have if you're using FDIC insured CDs. You know that the money is going to be there. The only risk that you have is pretty much inflation, and I-bonds are a great way to alleviate even that risk.

Chris: So the question came in on this slide. What are strategies to address planning for possible late retirement balloon of financial need regarding assisted living, increased health care needs, etc. not covered by insurance?

Mike Piper: Yeah, one thing that kind of coincidentally works out well is when we were talking about spending rates, we're saying that you need to use a low initial spending rate because you don't know whether stocks are going to provide good returns, but what usually happens is stocks do provide good returns. And so if you're using a very conservative initial spending rate what you actually see is most, a lot of people, see their portfolios grow throughout retirement even though they're spending from it and so that can actually all by itself really significantly alleviate that problem of health care costs going up significantly later in retirement, or long-term care needs or so on. 

But then of course, you do obviously want to have health insurance, no doubt about that. Long-term care insurance you might want to have it, again, it's more useful if you're married than for a single person. And there's again the hybrid options to consider, annuities or life insurance policies with long-term care riders.

Chris: That's great so there are questions that are still coming in so I want to just kind of pause here.  First of all Mike, if somebody does want to reach you through email, what is the best email that they should use for that, please.

Mike Piper: mike at oblivious investor dot com

[Music]

Slides

The slides used in this presentation are available at Slides.