EE Bond Doubling vs. Treasuries -- Question on Breakeven Rates & Tax Drags

[Noobvestor]

So I've been buying the max (10K/year) of EE bonds each year for the past decade, always with the idea in mind that if rates went up enough, I could stop buying, even sell some recent years, and buying Treasuries instead. But I'm scratching my head trying to figure out how taxes figure in.

Question: How many years back should I sell my EE bonds?

Givens:

1) I can't fit more bonds in tax-advantaged -- it is EE bonds or Treasuries in taxable.
2) My income is likely to be lower 15-20 years from now (i.e. I am likely to be in a lower tax bracket, but not sure - am currently mid-40s).
3) Subjectively, I like the simplicity of maxing out EE and I bond purchases each year, and planning to draw down once a year in the future. So if for example the 'best' option would be to only sell this year's, for simplicity, I'd just keep them. If it's a few years' worth, well, you get the idea.

Which brings up a related question: am I an idiot not to sell 0-rate I bonds in taxable in favor of TIPS with positive real yields? My thought has always been to cash them out in a future, low-tax year, but like the EE bonds, I'm not sure where the tax bracket tipping point is.

----------------------

I've been using this YTM comparison as a kind of starting point, but am not sure how to factor in potential income tax bracket changes, for example. Taxation aside, it looks like I'd be smart to cash out the last 5 years' worth, but ... would really appreciate some other thoughts!

Code: Select all

Year  YTM     Years left
1     3.53%   20    
2     3.72%   19    
3     3.93%   18    
4     4.16%   17    
5     4.43%   16    
6     4.73%   15    
7     5.08%   14   
8     5.48%   13  
9     5.95%   12  
10    6.50%   11    
11    7.18%   10    
12    8.01%    9    
13    9.05%    8    
14   10.41%    7   
15   12.25%    6   
16   14.87%    5    
17   18.92%    4
18   25.99%    3
19   42.42%    2
20  100.00%    1

[SantaClaraSurfer]

If the STRIPS YTM is significantly better than the EE Bonds current YTM, it's fairly straightforward to sell the EE Bonds and purchase the equivalent quantity of STRIPS / TIPS of the same maturity and then pay the likely insignificant taxes on any interest you've already earned on the EE Bonds. You'll have to factor in the taxes for your TIPS or STRIPS going forward to see if it's worth it to you, and only you have the information to do that.

Unlike their close cousins, U.S. Series I bonds, TIPs don’t benefit from tax deferral, The semi-annual inflation adjustments of a TIPS bond are treated as currently taxable income even though you don’t actually receive the money until you sell the bonds or they reach maturity. This is generally viewed as a downside to TIPs. Taxes on TIPS

Even though STRIPS do not pay out any interest until maturity, the IRS considers their imputed or "phantom" interest as taxable. The investor must pay taxes on this income each year, even though they do not receive the income until the bond matures.

For investors in high tax brackets, this can be a significant disadvantage unless the STRIPS are held in a tax-deferred account. Taxes on STRIPS

With I Bonds, if you cash out now, you'll have to pay taxes on all the yield you've gained over your principal, which is likely much more relevant, and subtract any interest penalties if you sell before five years. My bias as a long term buy and hold investor, in general, is to keep bonds I've already purchased if possible, and buy with today's dollar what seems like the best offering this year.

Personally, I buy a mix of bonds in taxable:

US Savings Bonds (I Bonds/EE Bonds) + Munis + STRIPS

The Munis provide tax-free income which covers the taxes on the STRIPS.
The I Bonds (varying fixed rates) look like they will cover my inflation protection needs "well enough" with the advantage of being tax deferred.
I'm inclined to let the EE Bonds I hold sit as the YTM on offer for STRIPS so far isn't enough to push me to sell given my marginal tax rate. However, I honestly don't know what I will do for EE Bonds in 2024.
I have been purchasing STRIPS, sometimes overlapping with my EE Bonds and sometimes covering shorter-term timeframes, and that's been lifting my blended YTM.

When I buy STRIPS, I try to aim for a single maturity month per year, and go for one that matches my EE Bond purchase window. (ie. prior to April 15th.)

I imagine I will buy and hold TIPS someday, too, but, ideally, those will live in the TDF bond allocation in our 401(k)s. For now, I'm running with MUNIs, STRIPS, I Bonds and EE Bonds in taxable with a goal of maintaining a small tax footprint during accumulation.

[Noobvestor]

I'm not worried about cash flow -- just ultimate post-tax return (i.e. I don't need, for example, to worry about what other bonds are/n't throwing off. I have way too much in taxable as-is, and between dividends and bond payments, it's more than enough to pay off any taxes).

f the STRIPS YTM is significantly better than the EE Bonds current YTM, it's fairly straightforward to sell the EE Bonds and purchase the equivalent quantity of STRIPS / TIPS of the same maturity and then pay the likely insignificant taxes on any interest you've already earned on the EE Bonds.

Isn't that apples and oranges, though? TIPS (inflation adjusted) and EE doubling (nominal)? I hold Series I and EE as well as TIPS and Treasuries, but more specifically: I aim for a ratio of roughly 50/50 nominal/inflation protected, so I'd be swapping EE for Treasuries.

I'm inclined to let the EE Bonds I hold sit as the YTM on offer for STRIPS so far isn't enough to push me to sell given my marginal tax rate. However, I honestly don't know what I will do for EE Bonds in 2024.

I'd like to understand the logic here better -- what's your marginal rate, versus what you expect it to be? And/or: what is the tipping point for you at which you'd start selling EE bonds, or at least stop buying new ones? FWIW, I'm in the 24%, will probably be there for years yet, maybe a decade or more. I figure if I'm still there in 20 years, I'm doing great and can pay whatever taxes. But if I'm in, a lower one, well, choices I make now might help.

Right now it seems to me that all else being equal, cashing in my paltry gains on EE and using that money to buy Treasuries is a net good, except I don't know how to factor taxes into that mix. I assume in a simplified scenario like: my marginal rate stays the same for 20 years, Treasuries come out ahead, right? But even that I'm a bit fuzzy on to be honest. Add in that I may well be retired in 15 years, and I'm in deeply murky waters.

Today, Treasuries are yielding 4.8%. EE doubling is a bit over 3.5% annualized. But ... I get to defer taxes. Curious how others are weighing this all.

[tibbitts]

Today, Treasuries are yielding 4.8%. EE doubling is a bit over 3.5% annualized. But ... I get to defer taxes. Curious how others are weighing this all.

It may depend on when the 20 years will occur in someone's investing timeline. If it comes when you want to be doing Roth conversions, or post-SS-claiming, for example... those might not be the best situations. I tended not to think of all my deferred investments as being in the "deferred" category, only the obvious ones like tIRAs and employer deferred plans, when fact I also had an annuity with mandatory distributions, I/EEs, plus large capital gains in taxable. So I didn't fully appreciate that almost anywhere I wanted to pull money from (voluntarily or through RMDs) will trigger taxes. But you can't practically do something like redeem EEs before their doubling date just to avoid them coming due at an inopportune time.

[ivgrivchuck]

How I would approach this:

Let's say your retirement federal marginal tax rate is 22%.

You invest $10k today in EE-bonds.
In 20 years you have after taxes $10k + $10k * (1 - 0.22) = $17.8k.

What kind of post-tax interest do we need for a 20 year STRIPS to get the same return in 20 years:

$10k * x^20 = $17.8k
x = (17.8/10.0) ^ (1/20) = 1.0292.

So you need 2.92% annual interest after taxes to break even.

Let's say that your federal marginal tax rate is 32% for the next 20 years.
x * (1 - 0.32) = 2.92%
x = 2.92% / 0.68 = 4.3%

So you would need 4.3% yield-to-maturity before taxes to break even.

[Noobvestor]

I appreciate the math, I really do! I'm missing one dumb thing though ... where the heck do I find STRIPS YTMs? My Google-fu is failing me here. Is there a way to do a similar calculation/comparison with SEC yields? Those I feel like I understand and can find easily for basically any duration.

[Noobvestor]

Today, Treasuries are yielding 4.8%. EE doubling is a bit over 3.5% annualized. But ... I get to defer taxes. Curious how others are weighing this all.

It may depend on when the 20 years will occur in someone's investing timeline. If it comes when you want to be doing Roth conversions, or post-SS-claiming, for example... those might not be the best situations. I tended not to think of all my deferred investments as being in the "deferred" category, only the obvious ones like tIRAs and employer deferred plans, when fact I also had an annuity with mandatory distributions, I/EEs, plus large capital gains in taxable. So I didn't fully appreciate that almost anywhere I wanted to pull money from (voluntarily or through RMDs) will trigger taxes. But you can't practically do something like redeem EEs before their doubling date just to avoid them coming due at an inopportune time.

Fair enough! Of the EE bonds I would consider selling (more recent years, so lower YTM) I figure many or most will likely land when I'm in retirement, but before RMDs, which might weight the decision toward hanging onto them.

But you're correct that I still can't control much beyond that (unlike I bonds). My general logic is: (1) I won't be in a higher bracket, (2) I might be in the same bracket, but if that's the case, I'm fine paying more taxes later, because I'll have saved a lot, or (3) I'll be in a lower bracket.

EE bonds I buy in the next few years ... those are going to hit right around the same time as RMDs, so between current interest rates and that RMD cliff, well, that does make me wary of keeping this ladder going. Seems like the EE bond question may in some ways be more straightforward to answer than the question of I bonds versus TIPS, because the latter has a flexibility factor that's hard to quantify.

[ivgrivchuck]

I appreciate the math, I really do! I'm missing one dumb thing though ... where the heck do I find STRIPS YTMs? My Google-fu is failing me here. Is there a way to do a similar calculation/comparison with SEC yields? Those I feel like I understand and can find easily for basically any duration.

Your broker should have a list of bonds you can buy.

For example at Schwab they have right now available:
US Treasury STRIP 0% 08/15/2043 at YTM 4.985%

Disclaimer: I have never purchased individual bonds. I know some theory, but I have no practical experience managing them.

[Noobvestor]

I appreciate the math, I really do! I'm missing one dumb thing though ... where the heck do I find STRIPS YTMs? My Google-fu is failing me here. Is there a way to do a similar calculation/comparison with SEC yields? Those I feel like I understand and can find easily for basically any duration.

Your broker should have a list of bonds you can buy.

For example at Schwab they have right now available:
US Treasury STRIP 0% 08/15/2043 at YTM 4.985%

Disclaimer: I have never purchased individual bonds. I know some theory, but I have no practical experience managing them.

Since 20-year Treasuries are yielding 4.8%, might it be safe to say that Treasuries are a good (enough) proxy? And if not, there's another conversion I need to make here, right? Because ultimately my replacement purchase will be Treasuries?!

[tibbitts]

I appreciate the math, I really do! I'm missing one dumb thing though ... where the heck do I find STRIPS YTMs? My Google-fu is failing me here. Is there a way to do a similar calculation/comparison with SEC yields? Those I feel like I understand and can find easily for basically any duration.

Your broker should have a list of bonds you can buy.

For example at Schwab they have right now available:
US Treasury STRIP 0% 08/15/2043 at YTM 4.985%

Disclaimer: I have never purchased individual bonds. I know some theory, but I have no practical experience managing them.

Since 20-year Treasuries are yielding 4.8%, might it be safe to say that Treasuries are a good (enough) proxy? And if not, there's another conversion I need to make here, right? Because ultimately my replacement purchase will be Treasuries?!

A downside to Treasuries would be that if for some now-unplanned reason you need to redeem early, the Treasuries could have lost considerable value.

[tibbitts]

But you're correct that I still can't control much beyond that (unlike I bonds). My general logic is: (1) I won't be in a higher bracket, (2) I might be in the same bracket, but if that's the case, I'm fine paying more taxes later, because I'll have saved a lot, or (3) I'll be in a lower bracket.

If you're MFJ now I'd say the outcome of your betting on a lower or equal bracket is definitely in doubt. Otherwise you might be see marginally higher or lower rates, certainly not worth thinking much about, unless maybe as part of a retire-early strategy.

While I understand what you're thinking, when it comes right down to it, no Boglehead will actually be "fine paying more taxes later."

[jeffyscott]

I appreciate the math, I really do! I'm missing one dumb thing though ... where the heck do I find STRIPS YTMs? My Google-fu is failing me here. Is there a way to do a similar calculation/comparison with SEC yields? Those I feel like I understand and can find easily for basically any duration.

Your broker should have a list of bonds you can buy.

For example at Schwab they have right now available:
US Treasury STRIP 0% 08/15/2043 at YTM 4.985%

Disclaimer: I have never purchased individual bonds. I know some theory, but I have no practical experience managing them.

Since 20-year Treasuries are yielding 4.8%, might it be safe to say that Treasuries are a good (enough) proxy? And if not, there's another conversion I need to make here, right? Because ultimately my replacement purchase will be Treasuries?!

The STRIPS are zero coupon, like EE bonds, so it is a fairer and easier comparison. With regular treasuries you will receive coupon payments that you would need to reinvest at unknown interest rates. That complicates the comparison.

One option could be to assume the market is efficient and so there is no reason to expect a better outcome from zero coupon STRIPS vs. regular treasuries. So if STRIPS are better than EE, then at least the expected return from treasuries would better as well. Another option could be to figure out what term of treasury has the same duration as the EE, so maybe you would need to buy a 17 year treasury to replace the 15 year EE or something like that.

Treasuries are more liquid than STRIPS (smaller bid/ask spread), so if you did need the money earlier than expected they have that advantage also you would be getting some money earlier anyway due to coupon payments.

[jeffyscott]

How I would approach this:

Let's say your retirement federal marginal tax rate is 22%.

You invest $10k today in EE-bonds.
In 20 years you have after taxes $10k + $10k * (1 - 0.22) = $17.8k.

What kind of post-tax interest do we need for a 20 year STRIPS to get the same return in 20 years:

$10k * x^20 = $17.8k
x = (17.8/10.0) ^ (1/20) = 1.0292.

So you need 2.92% annual interest after taxes to break even.

Let's say that your federal marginal tax rate is 32% for the next 20 years.
x * (1 - 0.32) = 2.92%
x = 2.92% / 0.68 = 4.3%

So you would need 4.3% yield-to-maturity before taxes to break even.

And if you are not sure of future tax bracket, you can repeat this with the likely possibilities. I'm going to use 12 and 22% for tax rates, instead of 22/32 and use the EEs with 15 years to go.

The ones that you have had for 5 years will leave you with $18,800 with a 12% tax rate in retirement, 15 years from now.

If you sold and are in the 22% bracket now, you would have about $1039 to invest for 15 years (there should be about $50 in interest before tax). To have $18,800, you need a bit more than 4% after tax and about 5.15% before tax, if I have calculated correctly. So for this scenario, keeping the EE would seem to be the better choice, but it's pretty close and probably the ones you have had for 4 years would be candidates to sell.

[Noobvestor]

The STRIPS are zero coupon, like EE bonds, so it is a fairer and easier comparison. With regular treasuries you will receive coupon payments that you would need to reinvest at unknown interest rates. That complicates the comparison.

One option could be to assume the market is efficient and so there is no reason to expect a better outcome from zero coupon STRIPS vs. regular treasuries. So if STRIPS are better than EE, then at least the expected return from treasuries would better as well. Another option could be to figure out what term of treasury has the same duration as the EE, so maybe you would need to buy a 17 year treasury to replace the 15 year EE or something like that.

Treasuries are more liquid than STRIPS (smaller bid/ask spread), so if you did need the money earlier than expected they have that advantage also you would be getting some money earlier anyway due to coupon payments.

Point taken re:fairer comparison. I think I can see what you're saying -- like even if I wouldn't be buying STRIPS, they're a more apples-to-apples 'marketable nominal government bond' thing than Treasuries, so comparing their rates to EE doubling is the closest to 'fair' yeah?

I'm comfortable with 'rounding' and 'close enough' when it comes to these kinds of optimizations. I guess what I'm trying to figure out is if I'm missing something particularly obvious. So like right now,

1) EE bonds double in 20 years, about a 3.5% annualized yield
2) 20-year Treasuries have about a 4.5% annualized yield

All else being equal (my tax bracket remaining the same, etc...) it's best to sell (1) for (2), yeah? And to scan that spreadsheet at the top that I posted, and do the same for any EEs with a YTM of less than 4.5%? Or ... am I still missing something? Let's say for the sake of argument I'm OK being off by .3%, maybe even .5%, just mainly trying to decide 'do I not sell any' or 'do I sell some' (if the latter, I can drill down to the next level). I think (per your example) if I assume my tax bracket is going down a notch, then keeping them all is probably better or at least a fine option.

And the ultimate hail mary question: there isn't by chance a spreadsheet floating around people use for this stuff, is there?!?!

[Noobvestor]

A downside to Treasuries would be that if for some now-unplanned reason you need to redeem early, the Treasuries could have lost considerable value.

I'll make an exception for Series I and EE bonds, but beyond that I refuse to add the complexity of DIY ladders or individual bonds in general.

If you're MFJ now I'd say the outcome of your betting on a lower or equal bracket is definitely in doubt. Otherwise you might be see marginally higher or lower rates, certainly not worth thinking much about, unless maybe as part of a retire-early strategy.

While I understand what you're thinking, when it comes right down to it, no Boglehead will actually be "fine paying more taxes later."

I'm single. Not really sure if I'll get married or not (my last long-term relationship we stayed unmarried). "Fine" might be a stretch, but I've had a target number in mind for a while now, and I'm pretty sure if I worked for 15 more years I'd more than meet it (and I've already met my 'minimum to survive'). It's one of the few things that hasn't really changed -- I set that number about 15 years ago, and I still feel like it's a solid high-end target!

[tibbitts]

A downside to Treasuries would be that if for some now-unplanned reason you need to redeem early, the Treasuries could have lost considerable value.

I'll make an exception for Series I and EE bonds, but beyond that I refuse to add the complexity of DIY ladders or individual bonds in general.

If you're MFJ now I'd say the outcome of your betting on a lower or equal bracket is definitely in doubt. Otherwise you might be see marginally higher or lower rates, certainly not worth thinking much about, unless maybe as part of a retire-early strategy.

While I understand what you're thinking, when it comes right down to it, no Boglehead will actually be "fine paying more taxes later."

I'm single. Not really sure if I'll get married or not (my last long-term relationship we stayed unmarried). "Fine" might be a stretch, but I've had a target number in mind for a while now, and I'm pretty sure if I worked for 15 more years I'd more than meet it (and I've already met my 'minimum to survive'). It's one of the few things that hasn't really changed -- I set that number about 15 years ago, and I still feel like it's a solid high-end target!

Just to clarify I didn't mean "fine" in that I'm sure you'll do well financially; I mean that even if a Boglehead had billions, he or she would definitely not be "fine" with paying $1 more in taxes than absolutely necessary... and every year would still be on the lookout for that last-minute deal on tax prep software. Well, except for the pencil-and-paper crowd, who'd probably write a "Boxes on 1040 Too Small" post.

[jeffyscott]

A downside to Treasuries would be that if for some now-unplanned reason you need to redeem early, the Treasuries could have lost considerable value.

I'll make an exception for Series I and EE bonds, but beyond that I refuse to add the complexity of DIY ladders or individual bonds in general.

But that's just what you're asking about, buying individual Treasury bonds to replace EE

With regard to your question about 4.5% EE vs. 4.5% Treasuries, there is still a difference because taxes on the EEs are deferred until you redeem.

Of course, there's the possibility to put the EE proceeds in stocks and then offset that by trading stocks for bonds in an IRA. That would reduce, but not eliminate, the added tax drag.

[Noobvestor]

A downside to Treasuries would be that if for some now-unplanned reason you need to redeem early, the Treasuries could have lost considerable value.

I'll make an exception for Series I and EE bonds, but beyond that I refuse to add the complexity of DIY ladders or individual bonds in general.

But that's just what you're asking about, buying individual Treasury bonds to replace EE ..... Of course, there's the possibility to put the EE proceeds in stocks and then offset that by trading stocks for bonds in an IRA. That would reduce, but not eliminate, the added tax drag.

OH, sorry, I should have been more specific (guilty as usual of using 'bonds' as a casual shorthand for 'bond funds'). I'd be swapping EE bonds for Treasuries held in a Treasury bond fund. I also hold most of my investments in taxable (business windfalls years back made it inevitable). From the start, part of the reason for going for EE and I bonds was expanding effective tax-advantaged space (any sold EE bonds, for the sake of simplicity, would be traded for Treasuries held in a fund in a taxable account -- so selling EEs reduces effective tax-advantaged space, in a sense). But I don't know if that 'effective expansion' is worth much, or really: how to value that effectively.

Just to clarify I didn't mean "fine" in that I'm sure you'll do well financially; I mean that even if a Boglehead had billions, he or she would definitely not be "fine" with paying $1 more in taxes than absolutely necessary... and every year would still be on the lookout for that last-minute deal on tax prep software. Well, except for the pencil-and-paper crowd, who'd probably write a "Boxes on 1040 Too Small" post.

I was just looking at a chart about how everyone wants just a bit more to be 'happy' -- like someone making 50K wants 70K, someone making 70K wants 100K and so on. IDK ... I used to be an optimizer, but as I get older, I'm more and more relaxed about finances. Obviously, I'm here asking about something that isn't going to make or break me, so I see the irony, but broadly, I'm less frugal/obsessive than I used to be

[Noobvestor]

I suppose it wouldn't hurt to mention that about half of my bond holdings are in Series I and EE bonds (a bit less). The rest are in TIPS and Treasuries, but to be specific: TIPS and Treasury bond funds (and Money Markets) averaging out to a target duration, more or less.

I also would find the answer easier if I were, say, 3 or 4 years into an EE ladder. As it is, I don't like the idea of leaving a 'gap' just because rates have changed a bit (i.e. selling a few years' worth, but still having older ones, and maybe buying more down the line). It's probably just an OCD thing, but I'd rather carry on with EE bond purchases, or bail and not go back to that strategy, and just let the ones I keep now double.

[jeffyscott]

OH, sorry, I should have been more specific (guilty as usual of using 'bonds' as a casual shorthand for 'bond funds'). I'd be swapping EE bonds for Treasuries held in a Treasury bond fund. I also hold most of my investments in taxable (business windfalls years back made it inevitable). From the start, part of the reason for going for EE and I bonds was expanding effective tax-advantaged space (any sold EE bonds, for the sake of simplicity, would be traded for Treasuries held in a fund in a taxable account -- so selling EEs reduces effective tax-advantaged space, in a sense). But I don't know if that 'effective expansion' is worth much, or really: how to value that effectively.

Okay, so the question is really more something like: "Are EE & I bonds a bad investment choice or does tax deferral make them an okay choice?".

The newest EE bonds at least have a better short term return, so may not be the worst of them. The worst would probably be those bought in 2021 or 2022 with a 0.1% short term rate. Those have 18-19 years to maturity.

You might compare those by assuming you could get something like 4.75-4.95% nominal over the next 18 years (based on Treasury yield curve and STRIPS quotes) or alternatively get about 3.9% tax deferred over that same time.

In the 12% bracket, using a taxable account and 4.75% return, the after tax return would be (0.88)(4.75%)=4.18%, so you can see that even at a 0% tax rate at maturity you would beat EE bonds.

In the 22% bracket, it would be a 3.7% return, so there is some chance the EEs could come out ahead. A 3.7% return on $10K over 18 years would give you $19,232. From that you can see what tax bracket you would need to be in for the EE to be approximately equivalent.

$10K at 3.9% for 18 years for the EE, would give you $19,910 before tax. So if the tax was $678 ($19,910-$19,232) the EE would equal the taxable investment. Your gain is $9910, so the break-even tax rate is about 6.85% (678/9910). Probably an unlikely tax bracket to anticipate, but you'd not lose that much at 12%. In that bracket, tax would be $1189, so your after tax total would be $18,721, $500 less than the taxable alternative.

You could repeat this sort of analysis for other terms and using your actual tax bracket.

A factor that could tilt things a little more toward the EEs would be if you anticipate the tax brackets reverting to 15% and 25%. That's make 4.75% equivalent to 3.56% after-tax and leave you with $18,778 and a break-even tax rate of 11.4%. There could certainly be an agreement to allow higher tax rates to revert, while keeping 12%, making this about as close to a toss-up as it could be.

I'd say that for one in the 22% bracket, who might be in a lower bracket when the EEs are redeemed and with only a taxable account to hold the EE's replacement, keeping or redeeming one with 18 years looks to be pretty close to a toss-up.

Another thing that could tilt one more toward the EEs would be if you thought you might move into a higher tax bracket (due to income, rather than rates reverting) during the term of the EEs, even though you might be in a lower bracket when actually redeeming. Maybe it would even be all that interest from the taxable treasuries that would push you into a higher tax bracket .