Asymmetric risk with nominal vs. inflation-linked bonds

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
squirrel1963
Posts: 545
Joined: Wed Jun 21, 2017 10:12 am
Location: Portland OR area

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by squirrel1963 »

Joey Jo Jo Jr wrote: Thu Aug 04, 2022 1:28 pm
vineviz wrote: Thu Aug 04, 2022 1:23 pm
Joey Jo Jo Jr wrote: Thu Aug 04, 2022 1:19 pm The original post recommends TIPS. Does not mention I bonds. Seems like relevant information to me. Honestly, a simple “hey, fair point, guy” would be fine.
The orginal post mentions "inflation-linked bonds, such as TIPS".
And mentions TIPS maybe another dozen times, and doesn’t mention I bonds as a possible better option. Again, such an extremely simple point I don’t know why there is such resistance to it.
That's a fair point to be sure, I think we immediately default to TIPS in the discussion because it's really tough to ammass sufficient I-bonds to really make a difference for a 10+ year ladder.
It also takes a lot of foresight to cumulate 30 years worth of I-bonds, whereas if your tax deferred space is large enough you can do it anytime.

I have both, but I definitely don't have enough I-bonds to make a meaningful difference for my TIPS ladder, and I suspect quite a few people are in the same situation.

Also keep in mind that right now newly issued I-bonds have a fixed rate of 0%, so in my opinion TIPS are preferable at 0.7%-0.9% real yields for long TIPS.
Yes I-bonds are better in deflation but deflation hasn't occurred enough times (as in sequence-of-return events) compared to inflation. And a deflated TIP will maintain the original purchase power even if the floor protection doesn't kick in, so it's good enough for me.
| LMP | safe portfolio: TIPS ladder + I-bonds + Treasuries | risky portfolio: US stocks / US REIT / International stocks |
Walkure
Posts: 565
Joined: Tue Apr 11, 2017 9:59 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Walkure »

bridge2benefits wrote: Thu Aug 04, 2022 10:13 am A $1,000 I Bond held over the same 5 year period would behave virtually identically. One difference is that your gains after year 4 would be realizable, because you could redeem the bond early (with a small penalty). But if you held it for a 5 year period during which net deflation occurred, you would have an identical value of $1,000.
This is incorrect. At 5 years, the interest penalty falls away and the I-bond could be redeemed for the full $1,170, which would have even higher purchasing power due to the interim deflation. There is no giveback to deflation ever (when the fixed rate is 0%) within a given 6-month period.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

Da5id wrote: Thu Aug 04, 2022 10:22 am
willthrill81 wrote: Thu Aug 04, 2022 10:16 am It seems difficult for some to understand that when the statement is made that TIPS will provide at least the real yield known at the time of purchase, regardless of what happens to inflation or deflation.
This guarantee is only for TIPs held to maturity, right? TIPs sold before maturity may not provide that real yield.

Unlike I-bonds, which other than the interest penalty will always provide the real yield indicated (which sadly is currently 0).
Yes, selling TIPS before maturity will almost certainly result in a different real yield than if they were held to maturity, just like any other marketable bond. Since I bonds are non-marketable, their real yield will be the same for years 1-30.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

Joey Jo Jo Jr wrote: Thu Aug 04, 2022 12:29 pm Well since the thesis of the thread is TIPS > nominals because of asymmetric risk...
No, it isn't. Nowhere in the OP does it say that inflation-linked bonds like TIPS are better than nominals. Rather, the contention is that inflation-linked bonds do not have the asymmetric risk that nominal bonds do.
vineviz wrote: Thu Aug 04, 2022 1:07 pmThe premise of this thread is that inflation-linked bonds are better than nominal bonds with respect to the asymmetry of inflation risk.

That premise is true regardless of WHICH TYPE of inflation-linked bonds you choose to use.
Precisely.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

Broken Man 1999 wrote: Thu Aug 04, 2022 1:08 pm Even today if a couple buys at least $20,000 each year, it isn't all that difficult to build a large inventory of I-Bonds. Honestly, outside of Bogleheads, where everyone is above average (much like Lake Wobegone), how many investors would be pinched by the purchase limits?
Considering that the median household income in the U.S. in 2020 was $67.5k, $20k for a MFJ couple is 30% of that. Few investors have a 30% saving rate, and even fewer are putting all their savings into bonds of any type.
anoop
Posts: 3193
Joined: Tue Mar 04, 2014 1:33 am

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by anoop »

The attractiveness of nominal treasuries especially at the long end is all about the potential for capital appreciation when the fed pivots. In other words, it’s a twisted world where you buy stocks for their dividend and bonds for capital appreciation.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

rossington wrote: Thu Aug 04, 2022 2:46 pm Thanks for explaining that!
:beer
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

anoop wrote: Thu Aug 04, 2022 3:31 pm The attractiveness of nominal treasuries especially at the long end is all about the potential for capital appreciation when the fed pivots.
Of course, that blade cuts both ways. Rising interest rates result in existing bond principal declining in value. It's called interest rate risk for good reason. So, unless one has a working crystal ball, there should be no expectation of bond principal appreciation over a long-term period.
Joey Jo Jo Jr
Posts: 267
Joined: Mon Jun 21, 2021 11:38 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Joey Jo Jo Jr »

willthrill81 wrote: Thu Aug 04, 2022 3:26 pm
Joey Jo Jo Jr wrote: Thu Aug 04, 2022 12:29 pm Well since the thesis of the thread is TIPS > nominals because of asymmetric risk...
No, it isn't. Nowhere in the OP does it say that inflation-linked bonds like TIPS are better than nominals. Rather, the contention is that inflation-linked bonds do not have the asymmetric risk that nominal bonds do.
vineviz wrote: Thu Aug 04, 2022 1:07 pmThe premise of this thread is that inflation-linked bonds are better than nominal bonds with respect to the asymmetry of inflation risk.

That premise is true regardless of WHICH TYPE of inflation-linked bonds you choose to use.
Precisely.
Did you not advocate for all TIPS (or inflation linked bonds) due to the asymmetric risk of nominals? If “better” is not the word you would use, how about “preferable”?

And if those are preferable might it not be relevant that I bonds have the same inflation benefit as TIPS but perform better during deflation? Why is it not possible to say “hey guy, fair enough point” and move on?

I must finally be a boglehead because my head is truly bogled! (yes I know the pronunciation and spelling are different).
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

Joey Jo Jo Jr wrote: Thu Aug 04, 2022 4:14 pm
willthrill81 wrote: Thu Aug 04, 2022 3:26 pm
Joey Jo Jo Jr wrote: Thu Aug 04, 2022 12:29 pm Well since the thesis of the thread is TIPS > nominals because of asymmetric risk...
No, it isn't. Nowhere in the OP does it say that inflation-linked bonds like TIPS are better than nominals. Rather, the contention is that inflation-linked bonds do not have the asymmetric risk that nominal bonds do.
vineviz wrote: Thu Aug 04, 2022 1:07 pmThe premise of this thread is that inflation-linked bonds are better than nominal bonds with respect to the asymmetry of inflation risk.

That premise is true regardless of WHICH TYPE of inflation-linked bonds you choose to use.
Precisely.
Did you not advocate for all TIPS (or inflation linked bonds) due to the asymmetric risk of nominals? If “better” is not the word you would use, how about “preferable”?
Risk is only part of what goes into making one asset preferable to a specific investor than another asset.

Stocks have more market price risk than bonds, for instance, but that doesn't mean that bonds are 'preferable' or 'better' in any objective way.

The purpose of the thread is not to say that inflation-linked bonds are 'better' than nominals but to acknowledge that nominal bonds have asymmetric risk than inflation-linked bonds do not.
FactualFran
Posts: 1883
Joined: Sat Feb 21, 2015 2:29 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by FactualFran »

TN_Boy wrote: Wed Aug 03, 2022 3:23 pm Should I predict the future, since T-bills had the best SWR in the worst times (say 1965) but in general, the longer duration bonds did better?

In fairness to long bonds, the difference in SWR between using T-bills and long gov bonds in the worst times was relatively minor -- maybe 3.8 versus 4.0. But intermediate bonds don't look so bad to me.

I presume that assumes holding a constant duration. In this thread, we've seen advocating shifting duration down as the retirement progresses (i.e. duration matching). I don't know what effect that would have had on SWRs.
It is your choice whether to predict the future. The graph shows the SWR for historical data. The lines for T-Bills and IT Gov. around 1965 as the starting year overlap too much to see which had a higher SWR. 1963 was the only starting year for which using T-Bills resulted in a higher SWR (4.81%) than using IT Gov. (4.80%).

The lowest SWR was 3.62% when using LT Gov. with 1966 as the starting year. The SWR using IT Gov. was 4.15% for the same starting year.

The bond data in a Stocks, Bonds, Bills, and Inflation yearbook is generally for a one-bond portfolio that is periodically replaced with a new bond with the shortest term to maturity that is at least the target term to maturity (30 days for T-Bills, 5 years for Intermediate-Term Gov., 20 years for Long-Term bonds).
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

FactualFran wrote: Thu Aug 04, 2022 4:59 pm
TN_Boy wrote: Wed Aug 03, 2022 3:23 pm Should I predict the future, since T-bills had the best SWR in the worst times (say 1965) but in general, the longer duration bonds did better?

In fairness to long bonds, the difference in SWR between using T-bills and long gov bonds in the worst times was relatively minor -- maybe 3.8 versus 4.0. But intermediate bonds don't look so bad to me.

I presume that assumes holding a constant duration. In this thread, we've seen advocating shifting duration down as the retirement progresses (i.e. duration matching). I don't know what effect that would have had on SWRs.
It is your choice whether to predict the future. The graph shows the SWR for historical data. The lines for T-Bills and IT Gov. around 1965 as the starting year overlap too much to see which had a higher SWR. 1963 was the only starting year for which using T-Bills resulted in a higher SWR (4.81%) than using IT Gov. (4.80%).

The lowest SWR was 3.62% when using LT Gov. with 1966 as the starting year. The SWR using IT Gov. was 4.15% for the same starting year.

The bond data in a Stocks, Bonds, Bills, and Inflation yearbook is generally for a one-bond portfolio that is periodically replaced with a new bond with the shortest term to maturity that is at least the target term to maturity (30 days for T-Bills, 5 years for Intermediate-Term Gov., 20 years for Long-Term bonds).
It might not be obvious to many, but the unexpected inflation of the 1970s and early 1980s was a major factor driving down SWRs for those who retirees in the mid-1960s. And inflation-linked bonds would have have been very useful to them, especially if they were duration matched.
RubyTuesday
Posts: 1916
Joined: Fri Oct 19, 2012 11:24 am

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by RubyTuesday »

Joey Jo Jo Jr wrote: Thu Aug 04, 2022 4:14 pm Why is it not possible to say “hey guy, fair enough point” and move on?

I must finally be a boglehead because my head is truly bogled! (yes I know the pronunciation and spelling are different).
hey guy, fair enough point
:sharebeer
“Doing nothing is better than being busy doing nothing.” – Lao Tzu
Logan Roy
Posts: 259
Joined: Sun May 29, 2022 10:15 am

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Logan Roy »

anoop wrote: Thu Aug 04, 2022 3:31 pm The attractiveness of nominal treasuries especially at the long end is all about the potential for capital appreciation when the fed pivots. In other words, it’s a twisted world where you buy stocks for their dividend and bonds for capital appreciation.
I think this is where TIPS are even more asymmetrical. You can buy long duration TIPS when there's a guarantee of a positive real return to maturity. Yet get largely the same capital appreciation when rates fall, and the option to reduce positions – making them an inflation, recession and deflation hedge.

Image
Joey Jo Jo Jr
Posts: 267
Joined: Mon Jun 21, 2021 11:38 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Joey Jo Jo Jr »

[/quote]

The purpose of the thread is not to say that inflation-linked bonds are 'better' than nominals but to acknowledge that nominal bonds have asymmetric risk than inflation-linked bonds do not.
[/quote]

Cool. Anything in particular I should do with that information? Buy inflation linked bonds? Which ones?
Joey Jo Jo Jr
Posts: 267
Joined: Mon Jun 21, 2021 11:38 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Joey Jo Jo Jr »

RubyTuesday wrote: Thu Aug 04, 2022 5:13 pm
Joey Jo Jo Jr wrote: Thu Aug 04, 2022 4:14 pm Why is it not possible to say “hey guy, fair enough point” and move on?

I must finally be a boglehead because my head is truly bogled! (yes I know the pronunciation and spelling are different).
hey guy, fair enough point
:sharebeer
Thank you!! :sharebeer The ghost of Joey Jo Jo Jr is almost free to leave this thread and move into the afterlife.
Kenneth Almquist
Posts: 56
Joined: Mon Sep 26, 2016 1:59 am

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Kenneth Almquist »

Code: Select all

CUSIP      type  coupon maturity   date    price     YTM
912810QV3  TIPS  0.75%  2/15/42  12/1/21  128.381 -0.572%
912810QV3  TIPS  0.75%  2/15/42   8/4/22   97.930  0.865% 
912810QU5   no   3.125% 2/15/42  12/1/21  122.566  1.788%
912810QU5   no   3.125% 2/15/42   8/4/22   99.570  3.152%
Over the last 20 months, the price of the TIPS shown fell more that the price of the nominal bond with the same maturity. Unfortunately I don't have older prices so I can't do a longer time frame comparison. Also, other people have already noted that volatility doesn't matter if you plan to hold to maturity.

On 12/1/21, the difference between the interest rate on the nominal bond and the interest rate on the real bond was 2.360%. On 8/4/22, the difference had declined to 2.287%. I suppose this makes sense if market participants believe that the current bout of high inflation is a temporary phenomenon that is closer to being over now that it was on 21/1/21.
User avatar
squirrel1963
Posts: 545
Joined: Wed Jun 21, 2017 10:12 am
Location: Portland OR area

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by squirrel1963 »

Joey Jo Jo Jr wrote: Thu Aug 04, 2022 5:48 pm
Cool. Anything in particular I should do with that information? Buy inflation linked bonds? Which ones?
It depends on your situation right? If you are interested in either one you need to figure out how many of them you need and how long it takes to cumulate enough I-bonds due to the yearly purchase limit.
You can work around the purchase limit a little bit if you are married by gifting them to each other and leaving them in the gift box until it's possible to deliver them. I've done that already for 2023 and 2024 but I don't think I would pile 20 years worth of I-bonds in the gift box where they cannot be used until delivery.

TIPS on the other hand don't have this limit, although if you buy them in IRA you are of course limited by how much you have in IRA.

So it's up to you. You can also buy TIPS in taxable but phantom income is a big issue IMHO.
| LMP | safe portfolio: TIPS ladder + I-bonds + Treasuries | risky portfolio: US stocks / US REIT / International stocks |
BigJohn
Posts: 2461
Joined: Wed Apr 02, 2014 11:27 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by BigJohn »

squirrel1963 wrote: Thu Aug 04, 2022 6:23 pm You can also buy TIPS in taxable but phantom income is a big issue IMHO.
Just to take the other side, I’d say phantom income in taxable needs to be considered but shouldn’t stop you if you decide you want/need TIPS. Use TIPS mutual funds or ETFs and there is no phantom income since they are required to pay out the inflation adjustment. As discussed several places in this discussion, you can achieve similar duration matching performance by picking the right ratio of fund durations.
"The greatest enemy of a good plan is the dream of a perfect plan" - Carl Von Clausewitz
skeptical
Posts: 265
Joined: Fri Jul 18, 2014 12:24 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by skeptical »

BigJohn wrote: Thu Aug 04, 2022 6:52 pm
squirrel1963 wrote: Thu Aug 04, 2022 6:23 pm You can also buy TIPS in taxable but phantom income is a big issue IMHO.
Just to take the other side, I’d say phantom income in taxable needs to be considered but shouldn’t stop you if you decide you want/need TIPS. Use TIPS mutual funds or ETFs and there is no phantom income since they are required to pay out the inflation adjustment. As discussed several places in this discussion, you can achieve similar duration matching performance by picking the right ratio of fund durations.
I am contemplating TIPS, but they would mostly be in taxable, and right now, that probably eliminates individual bonds due to the phantom income, so if I go this way it would be funds as whatever you need in taxes is already distributed. The other advantage to funds is that since income is taken from NAV, there is possible TLH, but this benefit seems cloudy.

The decision to go with TIPS, for me, is highly dependent on the size of the potential differential of returns between TIPS and nominals, and taxes have a big impact on that calculation, not only because a large part of the returns will be taxed, but also because the havoc it wreaks on the rest of the portfolio in terms of being bumped up tax brackets - Roth conversions, rebalancing, and qualified dividends.

So, when I look to compare nominals and TIP funds, the CAGR over time is not a simple comparison as the tax drag is highly sequence of returns dependent, as well as having unpredictable impact on the return of other assets and ability to do Roth conversions.
User avatar
vineviz
Posts: 12280
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by vineviz »

skeptical wrote: Thu Aug 04, 2022 7:09 pm
So, when I look to compare nominals and TIP funds, the CAGR over time is not a simple comparison as the tax drag is highly sequence of returns dependent, as well as having unpredictable impact on the return of other assets and ability to do Roth conversions.
You’re almost certainly overestimating the difference here.

Even a portfolio that is 100% TIPS can be set up so this “phantom income” is a non-issue.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
BigJohn
Posts: 2461
Joined: Wed Apr 02, 2014 11:27 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by BigJohn »

skeptical wrote: Thu Aug 04, 2022 7:09 pm
BigJohn wrote: Thu Aug 04, 2022 6:52 pm
squirrel1963 wrote: Thu Aug 04, 2022 6:23 pm You can also buy TIPS in taxable but phantom income is a big issue IMHO.
Just to take the other side, I’d say phantom income in taxable needs to be considered but shouldn’t stop you if you decide you want/need TIPS. Use TIPS mutual funds or ETFs and there is no phantom income since they are required to pay out the inflation adjustment. As discussed several places in this discussion, you can achieve similar duration matching performance by picking the right ratio of fund durations.
I am contemplating TIPS, but they would mostly be in taxable, and right now, that probably eliminates individual bonds due to the phantom income, so if I go this way it would be funds as whatever you need in taxes is already distributed. The other advantage to funds is that since income is taken from NAV, there is possible TLH, but this benefit seems cloudy.

The decision to go with TIPS, for me, is highly dependent on the size of the potential differential of returns between TIPS and nominals, and taxes have a big impact on that calculation, not only because a large part of the returns will be taxed, but also because the havoc it wreaks on the rest of the portfolio in terms of being bumped up tax brackets - Roth conversions, rebalancing, and qualified dividends.

So, when I look to compare nominals and TIP funds, the CAGR over time is not a simple comparison as the tax drag is highly sequence of returns dependent, as well as having unpredictable impact on the return of other assets and ability to do Roth conversions.
Not sure I understand the worry. If TIPS earn more than nominals, isn't that a good thing? Since you're paying taxes on both you're still net ahead. Surely you're not worried that the TIPS returns will be too high? Not sure how rebalancing or qualified dividends are impacted by tax bracket. Unless you're getting into NIIT territory for the dividends but even then I suspect you'd be net better off in TIPS with a higher return. As far as Roth conversions, just go part way and top up in December once you have a better handle on income.

In any case, I'd sure take all the issues that come up with being bumped up a tax bracket rather than lose value in nominals to high inflation. Be careful that your not letting the tax tail wag the portfolio dog. :beer
"The greatest enemy of a good plan is the dream of a perfect plan" - Carl Von Clausewitz
User avatar
squirrel1963
Posts: 545
Joined: Wed Jun 21, 2017 10:12 am
Location: Portland OR area

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by squirrel1963 »

BigJohn wrote: Thu Aug 04, 2022 6:52 pm
squirrel1963 wrote: Thu Aug 04, 2022 6:23 pm You can also buy TIPS in taxable but phantom income is a big issue IMHO.
Just to take the other side, I’d say phantom income in taxable needs to be considered but shouldn’t stop you if you decide you want/need TIPS. Use TIPS mutual funds or ETFs and there is no phantom income since they are required to pay out the inflation adjustment. As discussed several places in this discussion, you can achieve similar duration matching performance by picking the right ratio of fund durations.
Sure that's what I would do if I didn't have available space in tax deferred.
| LMP | safe portfolio: TIPS ladder + I-bonds + Treasuries | risky portfolio: US stocks / US REIT / International stocks |
User avatar
Kevin M
Posts: 12934
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Kevin M »

I don't like the term phantom income, although I understand it's a common term, and I'm not going to change that. Unless one is cash flow constrained, I don't see the difference between reinvesting nominal dividends, which are taxed, and being taxed on the inflation adjusted principal of TIPS. One way to view it is as forced dividend reinvestment.

Regarding a fund which distributes the inflation adjusted principal, unless you reinvest it you are not getting the full benefit of the TIPS fund, at least the way I view it. So if you reinvest TIPS fund dividends, what's the difference? If you are cash flow constrained, and must receive the dividends to pay the taxes, then use a fund.

So far I am buying my individual TIPS in an IRA, but I am buying TIPS for family members in taxable, since they have little or no tax-advantaged space. I am not worried at all about the so called phantom income (and they don't even know what it is).

Bringing it back on topic, I don't see "phantom income" as a big deal with respect to the asymmetric risk being discussed.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
Joey Jo Jo Jr
Posts: 267
Joined: Mon Jun 21, 2021 11:38 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Joey Jo Jo Jr »

bridge2benefits wrote: Thu Aug 04, 2022 10:13 am
Joey Jo Jo Jr wrote: Thu Aug 04, 2022 8:00 am
bridge2benefits wrote: Thu Aug 04, 2022 6:58 am TIPS (when bought at original auction and held to maturity) and I Bonds are both guaranteed to maintain a face value that is at least equal to their original purchase price. This means they provide equal protection from deflation, and they are both guaranteed to have positive real returns if there is deflation.

TIPS Reference: https://www.treasurydirect.gov/indiv/re ... ps_faq.htm
What happens to TIPS if deflation occurs?
The principal is adjusted downward, and your interest payments are less than they would be if inflation occurred or if the Consumer Price Index remained the same. You have this safeguard: at maturity, if the adjusted principal is less than the security's original principal, you are paid the original principal.
I Bonds Reference: https://treasurydirect.gov/forms/savpdp0039.pdf
Will the value of a Series I bond decrease during periods of deflation, when the CPI-U declines?
No. In periods of deflation, the bond’s redemption value won’t decline.
But TIPS give back value with deflation whereas I bonds don’t. This guy explains it better than I can: https://tipswatch.com/2015/02/06/tips-v ... ary-times/ Thus, I bonds have better protection from the asymmetric risk apparently being discussed here (waiting on OP to confirm my understanding of the risk).
To reduce that chances of others misunderstanding, it's worth pointing out some important nuances relevant to that discussion on Tipswatch. TIPS "give back" value with deflation only if you are tracking unrealized gains. If you buy your TIPS at auction, in a tax-deferred account, and hold them to maturity, the intermediate value fluctuations before maturity are irrelevant.

Example: $1,000 in a 5 year TIPS

After year 4, due to cumulative inflation of 17%, the theoretical value of the TIPS is $1,170. However this is a theoretical value, and it's only relevant for tax purposes if you hold the TIPS in a taxable account. You won't be able to sell the TIPS for $1,170. (Its market value will be higher or lower, depending on investor demand and interest rate changes since it was issued).

Let's say that there is dramatic deflation in year 5, such that cumulative deflation is 5%. Your TIPS matures, and $1,000 is deposited into your account. But the TIPS did not really "give back" $170 to deflation over the period that you held it, because that $170 was an unrealizable, paper gain.

A $1,000 I Bond held over the same 5 year period would behave virtually identically. One difference is that your gains after year 4 would be realizable, because you could redeem the bond early (with a small penalty). But if you held it for a 5 year period during which net deflation occurred, you would have an identical value of $1,000.
I don’t think you are correct. The composite I bond rate (both the fixed rate and the inflation rate) can never be below 0. https://www.treasurydirect.gov/indiv/re ... dterms.htm
So in your example the I bond maintains its gains and can’t go in reverse like the TIPS can. Deflationary asymmetry, you might say. :)
bridge2benefits
Posts: 55
Joined: Sat May 07, 2022 10:53 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by bridge2benefits »

Joey Jo Jo Jr wrote: Thu Aug 04, 2022 9:22 pm
bridge2benefits wrote: Thu Aug 04, 2022 10:13 am To reduce that chances of others misunderstanding, it's worth pointing out some important nuances relevant to that discussion on Tipswatch. TIPS "give back" value with deflation only if you are tracking unrealized gains. If you buy your TIPS at auction, in a tax-deferred account, and hold them to maturity, the intermediate value fluctuations before maturity are irrelevant.

Example: $1,000 in a 5 year TIPS

After year 4, due to cumulative inflation of 17%, the theoretical value of the TIPS is $1,170. However this is a theoretical value, and it's only relevant for tax purposes if you hold the TIPS in a taxable account. You won't be able to sell the TIPS for $1,170. (Its market value will be higher or lower, depending on investor demand and interest rate changes since it was issued).

Let's say that there is dramatic deflation in year 5, such that cumulative deflation is 5%. Your TIPS matures, and $1,000 is deposited into your account. But the TIPS did not really "give back" $170 to deflation over the period that you held it, because that $170 was an unrealizable, paper gain.

A $1,000 I Bond held over the same 5 year period would behave virtually identically. One difference is that your gains after year 4 would be realizable, because you could redeem the bond early (with a small penalty). But if you held it for a 5 year period during which net deflation occurred, you would have an identical value of $1,000.
I don’t think you are correct. The composite I bond rate (both the fixed rate and the inflation rate) can never be below 0. https://www.treasurydirect.gov/indiv/re ... dterms.htm
So in your example the I bond maintains its gains and can’t go in reverse like the TIPS can. Deflationary asymmetry, you might say. :)
You're right, thank you (and Walkure) for this correction. I'm sorry I didn't follow your point the first time. As Walkure explained it:
Walkure wrote: Thu Aug 04, 2022 2:54 pm At 5 years, the interest penalty falls away and the I-bond could be redeemed for the full $1,170, which would have even higher purchasing power due to the interim deflation. There is no giveback to deflation ever (when the fixed rate is 0%) within a given 6-month period.
Joey Jo Jo Jr
Posts: 267
Joined: Mon Jun 21, 2021 11:38 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Joey Jo Jo Jr »

Sorry I missed that Walcure had mentioned it. :sharebeer
bluesky50
Posts: 69
Joined: Tue Jan 09, 2018 11:05 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by bluesky50 »

willthrill81 wrote: Thu Aug 04, 2022 10:16 am
bridge2benefits wrote: Thu Aug 04, 2022 10:13 am
To reduce that chances of others misunderstanding, it's worth pointing out some important nuances relevant to that discussion on Tipswatch. TIPS "give back" value with deflation only if you are tracking unrealized gains. If you buy your TIPS at auction, in a tax-deferred account, and hold them to maturity, the intermediate value fluctuations before maturity are irrelevant.

Example: $1,000 in a 5 year TIPS

After year 4, due to cumulative inflation of 17%, the theoretical value of the TIPS is $1,170. However this is a theoretical value, and it's only relevant for tax purposes if you hold the TIPS in a taxable account. You won't be able to sell the TIPS for $1,170. (Its market value will be higher or lower, depending on investor demand and interest rate changes since it was issued).

Let's say that there is dramatic deflation in year 5, such that cumulative deflation is 5%. Your TIPS matures, and $1,000 is deposited into your account. But the TIPS did not really "give back" $170 to deflation over the period that you held it, because that $170 was an unrealizable, paper gain.

A $1,000 I Bond held over the same 5 year period would behave virtually identically. One difference is that your gains after year 4 would be realizable, because you could redeem the bond early (with a small penalty). But if you held it for a 5 year period during which net deflation occurred, you would have an identical value of $1,000.
Excellent example! :thumbsup

It seems difficult for some to understand that when the statement is made that TIPS will provide at least the real yield known at the time of purchase, regardless of what happens to inflation or deflation.
I don't think I understand how the deflection case works for TIPS and Ibonds. Hopefully someone help me here.

In the above 5-year TIPS example, lets assume that the TIPS has a coupon that pays out semi-annually and in year 5 there is 5 % deflation. When it matures, I am getting back the principle of $1000 but that does not account for the interest that has already been paid for the last 1-4 years, is that right?

For I bond, since it does not have any coupon, the only payout is the inflation adjustment that is paid semi-annually. This is not phantom money, this is real money that is deposited into my TD account semi-annually. So, in year 5 when there is 5% deflation, how much money do I get back from this Ibond?
User avatar
squirrel1963
Posts: 545
Joined: Wed Jun 21, 2017 10:12 am
Location: Portland OR area

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by squirrel1963 »

bluesky50 wrote: Thu Aug 04, 2022 10:34 pm
willthrill81 wrote: Thu Aug 04, 2022 10:16 am
bridge2benefits wrote: Thu Aug 04, 2022 10:13 am
To reduce that chances of others misunderstanding, it's worth pointing out some important nuances relevant to that discussion on Tipswatch. TIPS "give back" value with deflation only if you are tracking unrealized gains. If you buy your TIPS at auction, in a tax-deferred account, and hold them to maturity, the intermediate value fluctuations before maturity are irrelevant.

Example: $1,000 in a 5 year TIPS

After year 4, due to cumulative inflation of 17%, the theoretical value of the TIPS is $1,170. However this is a theoretical value, and it's only relevant for tax purposes if you hold the TIPS in a taxable account. You won't be able to sell the TIPS for $1,170. (Its market value will be higher or lower, depending on investor demand and interest rate changes since it was issued).

Let's say that there is dramatic deflation in year 5, such that cumulative deflation is 5%. Your TIPS matures, and $1,000 is deposited into your account. But the TIPS did not really "give back" $170 to deflation over the period that you held it, because that $170 was an unrealizable, paper gain.

A $1,000 I Bond held over the same 5 year period would behave virtually identically. One difference is that your gains after year 4 would be realizable, because you could redeem the bond early (with a small penalty). But if you held it for a 5 year period during which net deflation occurred, you would have an identical value of $1,000.
Excellent example! :thumbsup

It seems difficult for some to understand that when the statement is made that TIPS will provide at least the real yield known at the time of purchase, regardless of what happens to inflation or deflation.
I don't think I understand how the deflection case works for TIPS and Ibonds. Hopefully someone help me here.

In the above 5-year TIPS example, lets assume that the TIPS has a coupon that pays out semi-annually and in year 5 there is 5 % deflation. When it matures, I am getting back the principle of $1000 but that does not account for the interest that has already been paid for the last 1-4 years, is that right?

For I bond, since it does not have any coupon, the only payout is the inflation adjustment that is paid semi-annually. This is not phantom money, this is real money that is deposited into my TD account semi-annually. So, in year 5 when there is 5% deflation, how much money do I get back from this Ibond?
Yes correct. The TIP will always pay the coupon interest adjusted for inflation every year. The concept of "phantom income" only applies to TIPS held in taxable. You can keep the I-bond in TD until maturity, and you'll only pay taxes when you withdraw.
| LMP | safe portfolio: TIPS ladder + I-bonds + Treasuries | risky portfolio: US stocks / US REIT / International stocks |
User avatar
squirrel1963
Posts: 545
Joined: Wed Jun 21, 2017 10:12 am
Location: Portland OR area

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by squirrel1963 »

Kevin M wrote: Thu Aug 04, 2022 8:34 pm I don't like the term phantom income, although I understand it's a common term, and I'm not going to change that. Unless one is cash flow constrained, I don't see the difference between reinvesting nominal dividends, which are taxed, and being taxed on the inflation adjusted principal of TIPS. One way to view it is as forced dividend reinvestment.

Regarding a fund which distributes the inflation adjusted principal, unless you reinvest it you are not getting the full benefit of the TIPS fund, at least the way I view it. So if you reinvest TIPS fund dividends, what's the difference? If you are cash flow constrained, and must receive the dividends to pay the taxes, then use a fund.

So far I am buying my individual TIPS in an IRA, but I am buying TIPS for family members in taxable, since they have little or no tax-advantaged space. I am not worried at all about the so called phantom income (and they don't even know what it is).

Bringing it back on topic, I don't see "phantom income" as a big deal with respect to the asymmetric risk being discussed.

Kevin
My TIPS ladder is all in tax-deferred IRA, so I confess that I haven't spent much time to the thought of keeping TIPS in taxable, so perhaps I could be completely wrong, and if so, apologies. I fear that phantom income can be an issue in taxable even for a large nest egg under some circumstances and actually be victim of asymmetric risk.

During accumulation years you can pay taxes out of your pay, so presumably TIPS in taxable will not become an issue until retirement.

But consider a new retiree at 60 who has a nest egg of $3 M, who now wants to build a 30 year ladder paying 50K a year. Let's assume 0% TIPS just to keep things simple, let's ignore SSA benefits and let's pretend an emergency fund is not needed.

So this retiree builds a $50K * 30 = $1,500K TIPS ladder and puts $1.5M in VTI -- let's ignore the need for emergency funds in these calculations. If the TIPS ladder is in a pre-tax IRA the retiree will have $50K real dollars income a year, and because presumably tax bracket thresholds will be adjusted with inflation, (s)he will pay tax on such income at a very low bracket and use excess returns from stocks (if any) for discretionary expenses.

Now think of the taxable case and suppose this retiree did the ladder on 2022. Due to phantom income inflation adjustments there will be a taxable income on 9% of $1,500K = 135K in 2022, and because the retiree wanted to inflation protect the TIPS ladder the choice is now to pay taxes selling a portion of the stocks which has lost about 14.5% since the beginning of the year. So not only the retiree has a large tax bill at higher brackets, (s)he also needs to sell stocks during the worst possible time.

Now just imagine having sequence-of-return for 3 or more of these years, and all of a sudden you are exposed to sequence-of-returns asymmetric risk, having to pay a lot in taxes at higher tax brackets with funds that have lost a lot of value. This looks like a pretty bleak scenario to me and makes me think having TIPS in taxable may not be feasible for many folks who want a TIPS ladder for their LMP approach, while looking quite doable in tax-deferred space.

Again maybe I am wrong, so hopefully someone else will do a better math.
| LMP | safe portfolio: TIPS ladder + I-bonds + Treasuries | risky portfolio: US stocks / US REIT / International stocks |
czaj
Posts: 253
Joined: Sun Oct 04, 2015 4:01 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by czaj »

There has been some discussion on the deflation protection when buying TIPs at auction. TIPs funds do not “pass along” this benefit? Is this because they do not hold the bonds to maturity?
rossington
Posts: 1137
Joined: Fri Jun 07, 2019 2:00 am
Location: Florida

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by rossington »

Kevin M wrote: Thu Aug 04, 2022 8:34 pm I don't like the term phantom income, although I understand it's a common term, and I'm not going to change that. Unless one is cash flow constrained, I don't see the difference between reinvesting nominal dividends, which are taxed, and being taxed on the inflation adjusted principal of TIPS. One way to view it is as forced dividend reinvestment.

Regarding a fund which distributes the inflation adjusted principal, unless you reinvest it you are not getting the full benefit of the TIPS fund, at least the way I view it. So if you reinvest TIPS fund dividends, what's the difference? If you are cash flow constrained, and must receive the dividends to pay the taxes, then use a fund.

So far I am buying my individual TIPS in an IRA, but I am buying TIPS for family members in taxable, since they have little or no tax-advantaged space. I am not worried at all about the so called phantom income (and they don't even know what it is).

Bringing it back on topic, I don't see "phantom income" as a big deal with respect to the asymmetric risk being discussed.

Kevin
Kevin what is your take on:

1) duration matching bond funds for a retiree (as has been discussed in this current thread by Will).

..as well as:

2) duration matching bond funds for a retiree with a stock/bond portfolio as discussed in this thread: viewtopic.php?p=6809002#p6809002 by TN_Boy ?

Any advice not already discussed on these topics?

Thanks.
"Success is going from failure to failure without loss of enthusiasm." Winston Churchill.
Dude2
Posts: 1670
Joined: Fri Jun 08, 2007 3:40 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Dude2 »

The point about TIPS in taxable and phantom income is that taxes are assessed and paid in nominal dollars. People that are thinking in a real reference frame do not necessarily recognize that there have been any taxable gains on an inflation adjustment. This blows away the simplicity that an LMP can provide to pay yourself future income.
Then ’tis like the breath of an unfee’d lawyer.
skeptical
Posts: 265
Joined: Fri Jul 18, 2014 12:24 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by skeptical »

squirrel1963 wrote: Thu Aug 04, 2022 11:06 pm
Kevin M wrote: Thu Aug 04, 2022 8:34 pm I don't like the term phantom income, although I understand it's a common term, and I'm not going to change that. Unless one is cash flow constrained, I don't see the difference between reinvesting nominal dividends, which are taxed, and being taxed on the inflation adjusted principal of TIPS. One way to view it is as forced dividend reinvestment.

Regarding a fund which distributes the inflation adjusted principal, unless you reinvest it you are not getting the full benefit of the TIPS fund, at least the way I view it. So if you reinvest TIPS fund dividends, what's the difference? If you are cash flow constrained, and must receive the dividends to pay the taxes, then use a fund.

So far I am buying my individual TIPS in an IRA, but I am buying TIPS for family members in taxable, since they have little or no tax-advantaged space. I am not worried at all about the so called phantom income (and they don't even know what it is).

Bringing it back on topic, I don't see "phantom income" as a big deal with respect to the asymmetric risk being discussed.

Kevin
My TIPS ladder is all in tax-deferred IRA, so I confess that I haven't spent much time to the thought of keeping TIPS in taxable, so perhaps I could be completely wrong, and if so, apologies. I fear that phantom income can be an issue in taxable even for a large nest egg under some circumstances and actually be victim of asymmetric risk.

During accumulation years you can pay taxes out of your pay, so presumably TIPS in taxable will not become an issue until retirement.

But consider a new retiree at 60 who has a nest egg of $3 M, who now wants to build a 30 year ladder paying 50K a year. Let's assume 0% TIPS just to keep things simple, let's ignore SSA benefits and let's pretend an emergency fund is not needed.

So this retiree builds a $50K * 30 = $1,500K TIPS ladder and puts $1.5M in VTI -- let's ignore the need for emergency funds in these calculations. If the TIPS ladder is in a pre-tax IRA the retiree will have $50K real dollars income a year, and because presumably tax bracket thresholds will be adjusted with inflation, (s)he will pay tax on such income at a very low bracket and use excess returns from stocks (if any) for discretionary expenses.

Now think of the taxable case and suppose this retiree did the ladder on 2022. Due to phantom income inflation adjustments there will be a taxable income on 9% of $1,500K = 135K in 2022, and because the retiree wanted to inflation protect the TIPS ladder the choice is now to pay taxes selling a portion of the stocks which has lost about 14.5% since the beginning of the year. So not only the retiree has a large tax bill at higher brackets, (s)he also needs to sell stocks during the worst possible time.

Now just imagine having sequence-of-return for 3 or more of these years, and all of a sudden you are exposed to sequence-of-returns asymmetric risk, having to pay a lot in taxes at higher tax brackets with funds that have lost a lot of value. This looks like a pretty bleak scenario to me and makes me think having TIPS in taxable may not be feasible for many folks who want a TIPS ladder for their LMP approach, while looking quite doable in tax-deferred space.

Again maybe I am wrong, so hopefully someone else will do a better math.
Unless you are incorrect, this is exactly why I believe that a TIPs ladder cannot work in taxable. In addition, the qualified dividends on the equity side are also now taxable (Using munis means no federal taxes for the bonds and also the equity dividends), and doing any Roth conversions is more expensive. And, you also need to count in the extra tax on selling the equities, because even though they are "down", they still have a low cost basis.

So, in your example, the extra taxable income on the portfolio (which, if using munis, currently has a taxable income of $75K of qualified dividends, so no fed tax due) is almost $167K ($135K from the TIPS inflation adjustment and $32K from the selling of $63K of VTI to cover the taxes), and extra taxes would be $63K on what was supposed to be a $100K withdrawal rate with no taxes.

Extra taxes would be (assuming 40% fed+state and 20% cap gains (fed+state) would be $52K (tips) + 5K (VTI dividends)+ 6K (LTCG on the selling of stock at 50% basis), or $63K. Plus, Roth conversions will be far more expensive.

So, 1/2 of the inflation adjustment is paid out, significantly narrowing the potential upside of TIPS over nominal munis. It also creates what feels like a very unpredictable and unstable portfolio, which could decimate the equity holdings at the exact time you do not want to be selling. And, it is not clear what you "get back" when inflation cools down.

And, yes, under these scenarios, nominals will take a beating due to rising interest rates, however, if your bond duration is appropriate, you will (should) come back ok, and as a retiree, you will actually get more and more income which can be put back into the portfolio, instead of taking out of the portfolio to pay taxes.

And, here is the kicker: Data over the past 75 years indicate that interest rates correlate highly with inflation, and are typically higher than inflation. If this continues to be true (maybe this time is different ?), then the differences between TIPS and nominals once the unexpected inflation becomes expected becomes very problematic for TIPS in taxable unless the real rate goes significantly higher than 0%.

https://www.spglobal.com/spdji/en/docum ... ersect.pdf
https://www.gzeromedia.com/the-graphic- ... rest-rates

So, TIPS look pretty good in tax sheltered while there is unexpected inflation. A lot of that goodness is tempered if they are in taxable. But what happens once the "unexpected" goes away and interest rates are the same or higher than inflation ?

Are TIPS a good hedge against inflation ? Or are they a good hedge against a prolonged increase in inflation ? There is a big difference between these two scenarios.

I am sure I have a number of things incorrect here, but I do not think this is simple and clear, especially in taxable.
skeptical
Posts: 265
Joined: Fri Jul 18, 2014 12:24 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by skeptical »

Dude2 wrote: Fri Aug 05, 2022 4:25 am The point about TIPS in taxable and phantom income is that taxes are assessed and paid in nominal dollars. People that are thinking in a real reference frame do not necessarily recognize that there have been any taxable gains on an inflation adjustment. This blows away the simplicity that an LMP can provide to pay yourself future income.
There is a Nobel laureate (in economics) who believes that inflation is not a problem for people as long as their wages increase at the same rate as inflation. Funny but true.
User avatar
vineviz
Posts: 12280
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by vineviz »

squirrel1963 wrote: Thu Aug 04, 2022 11:06 pm Again maybe I am wrong, so hopefully someone else will do a better math.
The math is okay, at least at quick glance, but it is incomplete.

That's because the (taxable) inflation adjustment isn't exactly "phantom income": it represents the fact that, even if inflation is 0% for the remaining 29 years of the ladder, the annual income from the TIPS ladder is permanently higher by 9%.

In other words, a TIPS ladder will have 30 years of nominal income that is higher than a comparable nominal Treasury ladder would have provided. As a result, the stress on the remaining TIPS and/or remaining stocks is less due to the presences of the TIPS ladder even accounting for the tax bill in the first, 9% inflation, year.

People who obsess over this "TIPS in taxable" account are often succumbing to a combination of biases (e.g. money illusion and myopic loss aversion) that lead to them seeing only part of the picture.

I'm not saying there aren't nuances you could explore relative to tax planning and asset location, especially if you the retiree is near a breakpoint in tax brackets, but this issue is almost universally overblown.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

skeptical wrote: Fri Aug 05, 2022 5:43 am
Dude2 wrote: Fri Aug 05, 2022 4:25 am The point about TIPS in taxable and phantom income is that taxes are assessed and paid in nominal dollars. People that are thinking in a real reference frame do not necessarily recognize that there have been any taxable gains on an inflation adjustment. This blows away the simplicity that an LMP can provide to pay yourself future income.
There is a Nobel laureate (in economics) who believes that inflation is not a problem for people as long as their wages increase at the same rate as inflation. Funny but true.
Sadly, many investors let the tax tail wag the investment dog. Having lower income so one can pay fewer taxes is not a good strategy.

And beyond that, many forget that the tax brackets are all adjusted annually for inflation.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

czaj wrote: Thu Aug 04, 2022 11:08 pm There has been some discussion on the deflation protection when buying TIPs at auction. TIPs funds do not “pass along” this benefit? Is this because they do not hold the bonds to maturity?
TIPS funds do 'pass along' the deflation protection of TIPS, but the fact that they are constantly buying new TIPS means that the market price for new TIPS has an impact on the fund. That's where duration matching with multiple funds is particularly useful.
FactualFran
Posts: 1883
Joined: Sat Feb 21, 2015 2:29 pm

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by FactualFran »

willthrill81 wrote: Thu Aug 04, 2022 5:10 pm It might not be obvious to many, but the unexpected inflation of the 1970s and early 1980s was a major factor driving down SWRs for those who retirees in the mid-1960s. And inflation-linked bonds would have have been very useful to them, especially if they were duration matched.
In a post to the TIPS Confession topic there is a SWR graph for 50:50 stock:bond portfolios where the bonds are 10-Year nominal or 10-year inflation-indexed Treasury Notes, using published extrapolated yields back to 1971 for the inflation-indexed notes.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

FactualFran wrote: Fri Aug 05, 2022 2:44 pm
willthrill81 wrote: Thu Aug 04, 2022 5:10 pm It might not be obvious to many, but the unexpected inflation of the 1970s and early 1980s was a major factor driving down SWRs for those who retirees in the mid-1960s. And inflation-linked bonds would have have been very useful to them, especially if they were duration matched.
In a post to the TIPS Confession topic there is a SWR graph for 50:50 stock:bond portfolios where the bonds are 10-Year nominal or 10-year inflation-indexed Treasury Notes, using published extrapolated yields back to 1971 for the inflation-indexed notes.
Thanks. Sadly, we don't know precisely how TIPS would have worked during that period since they didn't exist back then. And the SWR in the U.S. was determined by the 1966 cohort, which wasn't included in that post.
User avatar
vineviz
Posts: 12280
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by vineviz »

willthrill81 wrote: Fri Aug 05, 2022 3:55 pm
FactualFran wrote: Fri Aug 05, 2022 2:44 pm
willthrill81 wrote: Thu Aug 04, 2022 5:10 pm It might not be obvious to many, but the unexpected inflation of the 1970s and early 1980s was a major factor driving down SWRs for those who retirees in the mid-1960s. And inflation-linked bonds would have have been very useful to them, especially if they were duration matched.
In a post to the TIPS Confession topic there is a SWR graph for 50:50 stock:bond portfolios where the bonds are 10-Year nominal or 10-year inflation-indexed Treasury Notes, using published extrapolated yields back to 1971 for the inflation-indexed notes.
Thanks. Sadly, we don't know precisely how TIPS would have worked during that period since they didn't exist back then. And the SWR in the U.S. was determined by the 1966 cohort, which wasn't included in that post.
I've made some crude attempts to estimate the way that TIPS might have performed for the 1966 cohort, but any such estimate is going to be highly dependent on the assumption you make about the breakeven rate between TIPS and nominal bonds. That's obviously unknowable, and even in the not-too-distant past we've seen some temporary fluctuations during recessions and/or liquidity crises that likely would have been impossible to predict.

But across a reasonable (IMHO) range of assumed breakeven inflation rates, the improvement in SWR from using TIPS instead of nominal bonds for a 1966 retiree would likely have been somewhere between 90bps and 125bps.

In other words, Bengen would likely have been talking about the "5% rule" instead of the "4% rule".
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

vineviz wrote: Fri Aug 05, 2022 4:37 pm
willthrill81 wrote: Fri Aug 05, 2022 3:55 pm
FactualFran wrote: Fri Aug 05, 2022 2:44 pm
willthrill81 wrote: Thu Aug 04, 2022 5:10 pm It might not be obvious to many, but the unexpected inflation of the 1970s and early 1980s was a major factor driving down SWRs for those who retirees in the mid-1960s. And inflation-linked bonds would have have been very useful to them, especially if they were duration matched.
In a post to the TIPS Confession topic there is a SWR graph for 50:50 stock:bond portfolios where the bonds are 10-Year nominal or 10-year inflation-indexed Treasury Notes, using published extrapolated yields back to 1971 for the inflation-indexed notes.
Thanks. Sadly, we don't know precisely how TIPS would have worked during that period since they didn't exist back then. And the SWR in the U.S. was determined by the 1966 cohort, which wasn't included in that post.
I've made some crude attempts to estimate the way that TIPS might have performed for the 1966 cohort, but any such estimate is going to be highly dependent on the assumption you make about the breakeven rate between TIPS and nominal bonds. That's obviously unknowable, and even in the not-too-distant past we've seen some temporary fluctuations during recessions and/or liquidity crises that likely would have been impossible to predict.

But across a reasonable (IMHO) range of assumed breakeven inflation rates, the improvement in SWR from using TIPS instead of nominal bonds for a 1966 retiree would likely have been somewhere between 90bps and 125bps.

In other words, Bengen would likely have been talking about the "5% rule" instead of the "4% rule".
A 25% improvement in the 30 year SWR would clearly have been of tremendous value to such retirees.
User avatar
Kevin M
Posts: 12934
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Kevin M »

squirrel1963 wrote: Thu Aug 04, 2022 11:06 pm
Kevin M wrote: Thu Aug 04, 2022 8:34 pm I don't like the term phantom income, although I understand it's a common term, and I'm not going to change that. Unless one is cash flow constrained, I don't see the difference between reinvesting nominal dividends, which are taxed, and being taxed on the inflation adjusted principal of TIPS. One way to view it is as forced dividend reinvestment.

Regarding a fund which distributes the inflation adjusted principal, unless you reinvest it you are not getting the full benefit of the TIPS fund, at least the way I view it. So if you reinvest TIPS fund dividends, what's the difference? If you are cash flow constrained, and must receive the dividends to pay the taxes, then use a fund.

So far I am buying my individual TIPS in an IRA, but I am buying TIPS for family members in taxable, since they have little or no tax-advantaged space. I am not worried at all about the so called phantom income (and they don't even know what it is).

Bringing it back on topic, I don't see "phantom income" as a big deal with respect to the asymmetric risk being discussed.

Kevin
<snip>

Now think of the taxable case and suppose this retiree did the ladder on 2022. Due to phantom income inflation adjustments there will be a taxable income on 9% of $1,500K = 135K in 2022, and because the retiree wanted to inflation protect the TIPS ladder the choice is now to pay taxes selling a portion of the stocks which has lost about 14.5% since the beginning of the year. So not only the retiree has a large tax bill at higher brackets, (s)he also needs to sell stocks during the worst possible time.

Now just imagine having sequence-of-return for 3 or more of these years, and all of a sudden you are exposed to sequence-of-returns asymmetric risk, having to pay a lot in taxes at higher tax brackets with funds that have lost a lot of value. This looks like a pretty bleak scenario to me and makes me think having TIPS in taxable may not be feasible for many folks who want a TIPS ladder for their LMP approach, while looking quite doable in tax-deferred space.

Again maybe I am wrong, so hopefully someone else will do a better math.
I see what you're talking about for someone with stocks and 100% of fixed income in TIPS.

My first thought would be that taxes are part of your expenses, so the LMP should provide enough to pay the taxes on the inflation adjustments. The problem is that taxes are on nominal income, and we don't know what the nominal income will be, since we don't know what inflation will be. The difficulty in planning for this is a point in favor of nominals--i.e., the asymmetry is in nominal's favor in terms of planning for taxes.

However, it's also quite difficult to know your other liabilities in the future, so there may be as much uncertainty here as for taxes. This might be an argument to over fund the TIPS ladder relative to your estimate of necessary residual expenses, which would include taxes as well as other expenses.

My next thought would be to hold some cash or nominal bonds to help with this contingency. The taxable accounts in which I'm buying TIPS have plenty of nominal fixed income as well, so this will not be a problem.

I believe that the asymmetry argument applies to funds as well, so owning some or all of your taxable TIPS in a fund might be a good solution for those who want to hold 100% TIPS in taxable and don't have other income sources to pay the taxes.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

Kevin M wrote: Fri Aug 05, 2022 6:07 pm My first thought would be that taxes are part of your expenses, so the LMP should provide enough to pay the taxes on the inflation adjustments. The problem is that taxes are on nominal income, and we don't know what the nominal income will be, since we don't know what inflation will be. The difficulty in planning for this is a point in favor of nominals--i.e., the asymmetry is in nominal's favor in terms of planning for taxes.
If the TIPS are held in tax-advantaged accounts, isn't this a moot point since the tax brackets are adjusted for inflation? The tax brackets aren't 'taking' anything away that the TIPS aren't 'giving' you?
User avatar
squirrel1963
Posts: 545
Joined: Wed Jun 21, 2017 10:12 am
Location: Portland OR area

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by squirrel1963 »

Kevin M wrote: Fri Aug 05, 2022 6:07 pm
squirrel1963 wrote: Thu Aug 04, 2022 11:06 pm
Kevin M wrote: Thu Aug 04, 2022 8:34 pm I don't like the term phantom income, although I understand it's a common term, and I'm not going to change that. Unless one is cash flow constrained, I don't see the difference between reinvesting nominal dividends, which are taxed, and being taxed on the inflation adjusted principal of TIPS. One way to view it is as forced dividend reinvestment.

Regarding a fund which distributes the inflation adjusted principal, unless you reinvest it you are not getting the full benefit of the TIPS fund, at least the way I view it. So if you reinvest TIPS fund dividends, what's the difference? If you are cash flow constrained, and must receive the dividends to pay the taxes, then use a fund.

So far I am buying my individual TIPS in an IRA, but I am buying TIPS for family members in taxable, since they have little or no tax-advantaged space. I am not worried at all about the so called phantom income (and they don't even know what it is).

Bringing it back on topic, I don't see "phantom income" as a big deal with respect to the asymmetric risk being discussed.

Kevin
<snip>

Now think of the taxable case and suppose this retiree did the ladder on 2022. Due to phantom income inflation adjustments there will be a taxable income on 9% of $1,500K = 135K in 2022, and because the retiree wanted to inflation protect the TIPS ladder the choice is now to pay taxes selling a portion of the stocks which has lost about 14.5% since the beginning of the year. So not only the retiree has a large tax bill at higher brackets, (s)he also needs to sell stocks during the worst possible time.

Now just imagine having sequence-of-return for 3 or more of these years, and all of a sudden you are exposed to sequence-of-returns asymmetric risk, having to pay a lot in taxes at higher tax brackets with funds that have lost a lot of value. This looks like a pretty bleak scenario to me and makes me think having TIPS in taxable may not be feasible for many folks who want a TIPS ladder for their LMP approach, while looking quite doable in tax-deferred space.

Again maybe I am wrong, so hopefully someone else will do a better math.
I see what you're talking about for someone with stocks and 100% of fixed income in TIPS.

My first thought would be that taxes are part of your expenses, so the LMP should provide enough to pay the taxes on the inflation adjustments. The problem is that taxes are on nominal income, and we don't know what the nominal income will be, since we don't know what inflation will be. The difficulty in planning for this is a point in favor of nominals--i.e., the asymmetry is in nominal's favor in terms of planning for taxes.

However, it's also quite difficult to know your other liabilities in the future, so there may be as much uncertainty here as for taxes. This might be an argument to over fund the TIPS ladder relative to your estimate of necessary residual expenses, which would include taxes as well as other expenses.

My next thought would be to hold some cash or nominal bonds to help with this contingency. The taxable accounts in which I'm buying TIPS have plenty of nominal fixed income as well, so this will not be a problem.

I believe that the asymmetry argument applies to funds as well, so owning some or all of your taxable TIPS in a fund might be a good solution for those who want to hold 100% TIPS in taxable and don't have other income sources to pay the taxes.

Kevin
Yes you most definitely need to account for taxes either way of course.
If I had no choice (I.e. No space or not enough space in tax deferred) I would still pick TIPS in taxable but probably prefer a combination of short/long TIPS fund, which seems more manageable tax wise, or at least easier to deal with.
Either way, it's not a slam dunk I think, I would run a lot of numbers to make sure the math works out.
| LMP | safe portfolio: TIPS ladder + I-bonds + Treasuries | risky portfolio: US stocks / US REIT / International stocks |
User avatar
Kevin M
Posts: 12934
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Kevin M »

willthrill81 wrote: Fri Aug 05, 2022 6:16 pm
Kevin M wrote: Fri Aug 05, 2022 6:07 pm My first thought would be that taxes are part of your expenses, so the LMP should provide enough to pay the taxes on the inflation adjustments. The problem is that taxes are on nominal income, and we don't know what the nominal income will be, since we don't know what inflation will be. The difficulty in planning for this is a point in favor of nominals--i.e., the asymmetry is in nominal's favor in terms of planning for taxes.
If the TIPS are held in tax-advantaged accounts, isn't this a moot point since the tax brackets are adjusted for inflation? The tax brackets aren't 'taking' anything away that the TIPS aren't 'giving' you?
First, the context of the reply was regarding paying taxes on the inflation adjustments in a taxable account, so tax-advantaged is not relevant to this particular point.

Second, even if tax brackets are adjusted for inflation, you still end up paying more taxes on higher inflation adjustments in the same tax bracket. If inflation is such that your inflation adjusted income is $100,000, at 22% fed marginal you pay $22,000 in tax. If inflation adjustment income is $150,000, you pay $33,000 in tax at 22% fed marginal. More after-tax money is better, but you have to have some plan for the high inflation adjustment scenarios.

In a tax-deferred account, higher inflation -> higher nominal value -> larger RMDs -> higher taxes, but the larger tax can be paid out of the larger RMDs.

Of course in a Roth there are no taxes, but most people here seem to prefer holding higher risk, higher expected return assets in their Roths (even though this is not necessarily sound reasoning if the higher risk of a Roth is factored in).

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

Kevin M wrote: Fri Aug 05, 2022 6:28 pm
willthrill81 wrote: Fri Aug 05, 2022 6:16 pm
Kevin M wrote: Fri Aug 05, 2022 6:07 pm My first thought would be that taxes are part of your expenses, so the LMP should provide enough to pay the taxes on the inflation adjustments. The problem is that taxes are on nominal income, and we don't know what the nominal income will be, since we don't know what inflation will be. The difficulty in planning for this is a point in favor of nominals--i.e., the asymmetry is in nominal's favor in terms of planning for taxes.
If the TIPS are held in tax-advantaged accounts, isn't this a moot point since the tax brackets are adjusted for inflation? The tax brackets aren't 'taking' anything away that the TIPS aren't 'giving' you?
First, the context of the reply was regarding paying taxes on the inflation adjustments in a taxable account, so tax-advantaged is not relevant to this particular point.

Second, even if tax brackets are adjusted for inflation, you still end up paying more taxes on higher inflation adjustments in the same tax bracket. If inflation is such that your inflation adjusted income is $100,000, at 22% fed marginal you pay $22,000 in tax. If inflation adjustment income is $150,000, you pay $33,000 in tax at 22% fed marginal. More after-tax money is better, but you have to have some plan for the high inflation adjustment scenarios.

In a tax-deferred account, higher inflation -> higher nominal value -> larger RMDs -> higher taxes, but the larger tax can be paid out of the larger RMDs.

Of course in a Roth there are no taxes, but most people here seem to prefer holding higher risk, higher expected return assets in their Roths (even though this is not necessarily sound reasoning if the higher risk of a Roth is factored in).

Kevin
That makes sense. Thanks.
JayB
Posts: 100
Joined: Sat May 28, 2022 9:57 am

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by JayB »

willthrill81 wrote: Fri Aug 05, 2022 10:10 am And beyond that, many forget that the tax brackets are all adjusted annually for inflation.
Under current tax law, tax brackets are adjusted in accordance with the chained CPI, not the CPI-U measure used for TIPS inflation adjustments. The chained measure is typically lower than the CPI-U by up to about 0.25% per year. So we have the potential for more income being pushed into higher marginal brackets over a number of years. Also, the NIIT threshold amounts are not adjusted for inflation.
User avatar
Lee_WSP
Posts: 7686
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by Lee_WSP »

How is the recommendation to prefer TIPS to nominals any different from the common and older saying that TIPS are for inflation insurance? I suppose taking the two sentences apart, we arrive at an actual recommendation rather than “just” a reason to buy TIPs. Nevertheless, the thinking and reasoning behind both statements appears to be the same to me.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Asymmetric risk with nominal vs. inflation-linked bonds

Post by willthrill81 »

Lee_WSP wrote: Sat Aug 06, 2022 5:20 pm How is the recommendation to prefer TIPS to nominals any different from the common and older saying that TIPS are for inflation insurance?
There was no recommendation in the OP to 'prefer TIPS to nominals'. Rather, the point was to make clear the asymmetric risk of nominals vs. TIPS.
Locked