Now that long TIPS yields are 60 bp off their highs I will…

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
billaster
Posts: 2882
Joined: Wed Apr 27, 2022 2:21 pm

Re: Now that long TIPS again yield more than 1.70% I will…

Post by billaster »

watchnerd wrote: Thu May 25, 2023 11:00 pm
billaster wrote: Thu May 25, 2023 10:41 pm
Bill Bernstein wrote: Thu May 25, 2023 10:16 pm A little financial history and a pinch of Minsky go a long way. IOW, good things almost always come to those who wait, especially with fixed income, for the simple reason that from time to time,the central banker swallows have to return to tightening Capistrano.
The thing is, you could find yourself waiting a very long time, hoping for a black swan to show up. TIPS have been above 1% for only two months out of the previous 11 years and could have easily been much longer if not for a world-wide pandemic.
I actually doubt it.

Chancellor's "Price of Time" would argue that the QE/ZIRP period was imbalanced with interest rates too far below natural interest rates, and thus the system was inherently unstable and going to be heading towards a regime change eventually.
Chancellor is a journalist, not an economist.

Economists define the "natural interest rate" as be the level of output consistent with stable price inflation, absent transitory shocks to supply. The natural rate of interest is the real fed funds rate consistent with stable inflation absent shocks to demand and supply.

Some folks seem to have a definition of "natural interest rate" as something they are used to or would aspirationally like it to be.

In the years since 2009 inflation has been very stable around 2%. It's hard to argue that Fed rates were far below the "natural rate" when there was not extraordinary inflation.
User avatar
bobcat2
Posts: 6074
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Now that long TIPS again yield more than 1.70% I will…

Post by bobcat2 »

Kevin M wrote: Thu May 25, 2023 12:10 pm
bobcat2 wrote: Wed May 24, 2023 7:31 pm
watchnerd wrote: Tue May 23, 2023 3:16 pm
Kevin M wrote: Tue May 23, 2023 3:09 pm Several highly respected forum members here, e.g., BobK, have argued that you can accomplish an LMP (non rolling ladder) with two or three funds, as you suggest. I understand the argument, but I've never been convinced that there's no risk in the rebalancing part; i.e., selling shares of the long term fund when long term yields have increased significantly to buy shares of the shorter term fund. You don't get the benefit of the return to par at maturity effect for the longer term holdings.
<snip>
Here are some reasons I prefer using duration matched TIPS funds and I-bonds instead of creating a TIPS bond ladder for safe flexible retirement income.
Correct me if I'm wrong Bob. I've seen your posts as bobcat2, but believe your username used to be BobK, and that's the one I tend to remember. Perhaps there's an interesting story behind that?

Thanks for schooling us on your views of using funds instead of a ladder.
To me the most important reason for preferring TIPS funds to TIPS bonds is during the accumulation stage of acquiring TIPS assets. At least 15 years before your target retirement year you want to be moving assets into safe TIPS. I am assuming you do this primarily in your DC plan with 26 pay periods per year. So each pay period you are buying TIPS assets. With new money you couldn't buy one TIPS bond per pay period. But you can easily add to TIPS funds per pay period and up the percent of new money going into TIPS funds over time. Buying TIPS bonds over time with old money in a 401k would be clumsy and not very cost efficient, but shifting old money into TIPS funds periodically is easy.

The problem with purchasing a TIPS ladder at a point in time is that is inherently risky. As vineviz also pointed out several times, planning on making a big one time financial move in the future is always risky. Your risky portfolio may not get to that financial target such as 25x or 100% funded ratio. Instead the goal from my perspective should be achieving a targeted level of safe income - not whatever level of safe income you can afford at some future date.

If, however, you want to purchase your liability matched assets at one time in the future - go with the TIPS ladder. But be aware that any one time future purchase is a risky strategy if you want to hit a target.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
slicendice
Posts: 563
Joined: Tue Sep 22, 2020 12:08 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by slicendice »

billaster wrote: Fri May 26, 2023 11:39 am In the years since 2009 inflation has been very stable around 2%. It's hard to argue that Fed rates were far below the "natural rate" when there was not extraordinary inflation.
Including many shorter periods where inflation failed to meet the 2% target, which absent a change in fiscal policy is a tough place for the Fed to be with the funds rate at 0. That's why I think the Fed will accept core inflation in the 3% range going forward given the relatively blunt instruments at their disposal. From their perspective 3% is vastly preferable to 1.5%. The risk of this seems like a good reason to prefer TIPS over what nominal bonds are yielding right now.
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

FWIW, I established my TIPS ladder in about two hours over two days (would have been shorter if Vanguard's systems were working better). I funded largely from TIPS with relatively short remaining maturities (some held since 1998) and Vanguard's TIPS fund and ST bond fund.

I hope this doesn't run afoul of the "The problem with purchasing a TIPS ladder at a point in time is that is inherently risky" criterion.
Bill Bernstein
Posts: 853
Joined: Sat Jun 23, 2007 12:47 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Bill Bernstein »

Code: Select all

Chancellor is a journalist, not an economist. 
For the record, he trained as a historian, and is universally recognized as one of the world's foremost financial historians.

He's worked as an investment banker and asset manager at GMO, and, more to the point, has written two very highly regarded works of primary financial historical research, The Price of Time and Devil Take the Hindmost.

I try to stay away, in any case, from credentialism. Jason Zweig, for example, took his undergraduate degree in art history, but this neurologist stands in awe of his command of neuropsychology. To give a more extreme example, Fischer Black was an applied mathematician and computer scientist, and had no formal economic training.

And, trust me, Chancellor doesn't need the Wicksellian rate of interest explained to him.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

NiceUnparticularMan wrote: Fri May 26, 2023 9:20 am
watchnerd wrote: Fri May 26, 2023 9:01 am In our case, though, it won't actually look like that 'maintaining lifestyle' because we'll be on a rising equity glide path as the LMP gets consumed.
So if you have both legacy/gift and retirement income motives, then it could make sense as you age for the legacy/gift portion of the portfolio to grow while your retirement income portion of the portfolio shrinks, and that could take the former of a rising equity glide path at an overall portfolio level.

That said, you could (and I would say should) also consider giving out of your legacy/gift portfolio while alive. In which case the overall effect could be something different.

For just the retirement income portion of your portfolio, declining duration (although not at a 1:1 rate) would make sense, but not necessarily a significant risky/non-risky shift.
We have no heirs, so any excess will be given to charity, probably.

But at age 53 we're not worrying about that just yet.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Lawrence of Suburbia
Posts: 691
Joined: Mon Aug 08, 2022 12:04 pm

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Lawrence of Suburbia »

I'm happy with the token amount of VIPSX(?) in my Target 2025 fund. See no reason to divert any more $s from my regular investments; I've been persuaded by numerous folks much smarter than I that stocks are a better long-term bet vs. inflation.
74% VTHRX/8% DODWX/12% TIAA Traditional/6% SWVXX
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

bogswenbern wrote: Fri May 26, 2023 9:25 am
Currently I copy and paste the value from the brokerage website. It's not just the value at maturity that matters, because when adding new money I need to know where to direct it to keep my desired asset allocation, stock/bond ratio etc. Also need the value for rebalancing in a boom or crash. Using GOOGLEFINANCE is how I get values automatically for funds/ETFs.
I solve this by viewing my risk portfolio (stocks, nominals bonds, alts) and LMP (TIPS) as two separate portfolios.

I don't rebalance from LMP into the risk portfolio, so I don't really care about mark-to-market of individual TIPS.

I also don't rebalance within my risk portfolio, either, as I'm not trying to achieve a certain fixed AA.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

Lawrence of Suburbia wrote: Fri May 26, 2023 12:49 pm I'm happy with the token amount of VIPSX(?) in my Target 2025 fund. See no reason to divert any more $s from my regular investments; I've been persuaded by numerous folks much smarter than I that stocks are a better long-term bet vs. inflation.
The research literature finds stocks vs inflation to often be overstated:
It is often claimed that equities are a hedge against inflation. Figure 16 showed this is incorrect. Equities performed especially well in real terms when inflation was low. High inflation impaired performance and deflation led to lower returns than on government bonds. The correlation between real equity returns and inflation was negative, i.e. equities were a poor hedge against inflation. There is an extensive literature to back this up. Fama and Schwert (2077), Fama (1981), and Boudoukh and Richardson (1993) are three classic papers, and Tatom (2011) is a useful review article.
https://www.credit-suisse.com/media/ass ... dition.pdf
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

bobcat2 wrote: Fri May 26, 2023 12:05 pm At least 15 years before your target retirement year you want to be moving assets into safe TIPS.
Why 15 years?
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Kevin M »

bogswenbern wrote: Fri May 26, 2023 8:51 am I use a Google doc spreadsheet to track asset allocation and where to direct new money and rebalance. Is there a way to update the value of individual TIPS automatically, maybe using the CUSIP? If not, isn't that a big advantage of the fund/ETF? These long TIPS are volatile...
I also track portfolios with google sheets (many of them, as I manage family member accounts as well of that of a good friend). I periodically do a download of the portfolios from Vanguard and Fidelity, and import the results into a master spreadsheet, and from there use IMPORTRANGE to get the values into the individual portfolio sheets.

If you just have one portfolio (one ownership group), then I'd just import the download into a Vanguard or Fidelity sheet, and then you can extract the values from there into your portfolio sheet. I find this much easier than copy/paste, and it only takes a few minutes to do. This way your bond values match the broker's values as of the time of download. Of course this could change a few minutes later, but the values will be close.

I use VLOOKUP based on CUSIP for Fidelity or description for Vanguard (VG does not include CUSIP in the download--boo!) to get the values from the VG or Fido sheets into the portfolio sheet.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
bobcat2
Posts: 6074
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Now that long TIPS again yield more than 1.70% I will…

Post by bobcat2 »

exodusing wrote: Fri May 26, 2023 12:21 pm FWIW, I established my TIPS ladder in about two hours over two days (would have been shorter if Vanguard's systems were working better). I funded largely from TIPS with relatively short remaining maturities (some held since 1998) and Vanguard's TIPS fund and ST bond fund.

I hope this doesn't run afoul of the "The problem with purchasing a TIPS ladder at a point in time is that is inherently risky" criterion.
Just because it's risky doesn't mean it won't work. It simply means it's risky, if you have targeted a safe level of income.

If you have targeted for a safe level of income it seams to me you would want to get there safely. :happy But to each his own. Unsafely at a point in time you will probably get there. Safely over time you are extremely likely to get there.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
User avatar
Lawrence of Suburbia
Posts: 691
Joined: Mon Aug 08, 2022 12:04 pm

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Lawrence of Suburbia »

watchnerd wrote: Fri May 26, 2023 1:00 pm
Lawrence of Suburbia wrote: Fri May 26, 2023 12:49 pm I'm happy with the token amount of VIPSX(?) in my Target 2025 fund. See no reason to divert any more $s from my regular investments; I've been persuaded by numerous folks much smarter than I that stocks are a better long-term bet vs. inflation.
The research literature finds stocks vs inflation to often be overstated:
It is often claimed that equities are a hedge against inflation. Figure 16 showed this is incorrect. Equities performed especially well in real terms when inflation was low. High inflation impaired performance and deflation led to lower returns than on government bonds. The correlation between real equity returns and inflation was negative, i.e. equities were a poor hedge against inflation. There is an extensive literature to back this up. Fama and Schwert (2077), Fama (1981), and Boudoukh and Richardson (1993) are three classic papers, and Tatom (2011) is a useful review article.
https://www.credit-suisse.com/media/ass ... dition.pdf
Impressive article, for sure.

Has anyone done a longterm study of stocks vs. TIPS, though? ... I'm still not persuaded to throw more of my limited means into those.
74% VTHRX/8% DODWX/12% TIAA Traditional/6% SWVXX
User avatar
bobcat2
Posts: 6074
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Now that long TIPS again yield more than 1.70% I will…

Post by bobcat2 »

watchnerd wrote: Fri May 26, 2023 1:05 pm
bobcat2 wrote: Fri May 26, 2023 12:05 pm At least 15 years before your target retirement year you want to be moving assets into safe TIPS.
Why 15 years?
Notice I said at least, it could be more. It's because if you are going to achieve a targeted level of income safely over time you need time to do it. The longer you hold your risky portfolio the greater the chances that a period of equity returns like 2000-2002 or 2007-2009 will show up and leave you high and dry. No one is forcing anyone to do anything safely. You can start later, take more risk, and have a better shot at higher safe income in retirement, as well as a better shot at having lower safe income in retirement. :wink:

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

bobcat2 wrote: Fri May 26, 2023 1:06 pm
exodusing wrote: Fri May 26, 2023 12:21 pm FWIW, I established my TIPS ladder in about two hours over two days (would have been shorter if Vanguard's systems were working better). I funded largely from TIPS with relatively short remaining maturities (some held since 1998) and Vanguard's TIPS fund and ST bond fund.

I hope this doesn't run afoul of the "The problem with purchasing a TIPS ladder at a point in time is that is inherently risky" criterion.
Just because it's risky doesn't mean it won't work. It simply means it's risky, if you have targeted a safe level of income.

If you have targeted for a safe level of income it seams to me you would want to get there safely. :happy But to each his own. Unsafely at a point in time you will probably get there. Safely over time you are extremely likely to get there.

BobK
I now have a government guaranteed inflation adjusted stream of income locked in until I'm 95. That seems rather safe. Most of it was funded with TIPS or a TIPS fund, albeit with a shorter average duration. What would have been the advantage of doing this over time?

I can understand the risk of using equities to fund a ladder or set of funds, as the equities could drop in price at an inopportune moment, but that is much less of a problem when using similar securities.

I just saw this in another thread and am not sure I understand it.
bobcat2 wrote: Wed Feb 07, 2018 4:25 pm You can do cash flow matching with a bond ladder. You can do duration liability matching with a bond ladder or bond funds. Most bond laddering done by professionals is done through duration matching. I would guess that most individual investors don't use ladders for duration liability matching, but rather as cash flow matching, which is less efficient (more wasteful of resources to reach a goal) than duration matching.

When you use TIPS for duration matching you immunize the assets against both changes in interest rates and inflation. That makes for safe income. :D

BobK
I estimate I will need $X/year for the next Y years and now have a guaranteed inflation adjusted income stream (i.e., TIPS ladder) to fund those expenses. Is this what you mean by cash flow matching? If so, how is duration matching different and how is it more efficient?
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

Lawrence of Suburbia wrote: Fri May 26, 2023 1:15 pm
Impressive article, for sure.

Has anyone done a longterm study of stocks vs. TIPS, though? ... I'm still not persuaded to throw more of my limited means into those.
TIPS have only been around since 1997, so if using actual historical data, it wouldn't be very long term.

I think someone synthesized inflation-linked bonds going back a century or so, but I can't find the study
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Drew31
Posts: 735
Joined: Mon Oct 22, 2012 1:18 pm

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Drew31 »

watchnerd wrote: Fri May 26, 2023 12:55 pm I also don't rebalance within my risk portfolio, either, as I'm not trying to achieve a certain fixed AA.
You just let it float and add new money based on a set % between stocks/bonds?
User avatar
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Kevin M »

exodusing wrote: Thu May 25, 2023 8:43 am I've been thinking about how to hold TIPS in taxable for part of a LMP since I'm running out of IRA space. All things equal, I prefer a ladder, since it's easy to deal with once set up.

Funds make tax reporting simpler than dealing with individual TIPS. However, rebalancing between funds is a taxable event while a maturing TIPS is not. Tax reporting is a bit of a hassle, but gains have a real economic cost.

Any thoughts on the subject?
I found tax reporting for TIPS in taxable to be quite easy this year using HRB software. The OID (inflation adjustment income) reported by Vanguard was accurate (I checked with my own calculations for one or two accounts), and IIRC, the main complication in the past has been inaccurate OID reporting, which required one to do their own calculations.

For more, see the taxation of Treasuries thread (easy to find with a forum search).

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

Drew31 wrote: Fri May 26, 2023 1:34 pm
watchnerd wrote: Fri May 26, 2023 12:55 pm I also don't rebalance within my risk portfolio, either, as I'm not trying to achieve a certain fixed AA.
You just let it float and add new money based on a set % between stocks/bonds?
The risk portfolio is global market cap weight, so it floats around accordingly.

With only 2 years planned left to work and already having exceeded 'our number', we aren't adding new money to our retirement savings anymore. We're saving for home upgrades, future vacations, etc.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

Kevin M wrote: Fri May 26, 2023 1:44 pm
exodusing wrote: Thu May 25, 2023 8:43 am I've been thinking about how to hold TIPS in taxable for part of a LMP since I'm running out of IRA space. All things equal, I prefer a ladder, since it's easy to deal with once set up.

Funds make tax reporting simpler than dealing with individual TIPS. However, rebalancing between funds is a taxable event while a maturing TIPS is not. Tax reporting is a bit of a hassle, but gains have a real economic cost.

Any thoughts on the subject?
I found tax reporting for TIPS in taxable to be quite easy this year using HRB software. The OID (inflation adjustment income) reported by Vanguard was accurate (I checked with my own calculations for one or two accounts), and IIRC, the main complication in the past has been inaccurate OID reporting, which required one to do their own calculations.

For more, see the taxation of Treasuries thread (easy to find with a forum search).

Kevin
For this specific issue, see viewtopic.php?p=7283848#p7283848
User avatar
Drew31
Posts: 735
Joined: Mon Oct 22, 2012 1:18 pm

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Drew31 »

watchnerd wrote: Fri May 26, 2023 1:44 pm
Drew31 wrote: Fri May 26, 2023 1:34 pm
watchnerd wrote: Fri May 26, 2023 12:55 pm I also don't rebalance within my risk portfolio, either, as I'm not trying to achieve a certain fixed AA.
You just let it float and add new money based on a set % between stocks/bonds?
The risk portfolio is global market cap weight, so it floats around accordingly.

With only 2 years planned left to work and already having exceeded 'our number', we aren't adding new money to our retirement savings anymore. We're saving for home upgrades, future vacations, etc.
Makes sense. Always curious to see how others handle this.
User avatar
bobcat2
Posts: 6074
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Now that long TIPS again yield more than 1.70% I will…

Post by bobcat2 »

exodusing wrote: Fri May 26, 2023 1:22 pm
bobcat2 wrote: Fri May 26, 2023 1:06 pm
exodusing wrote: Fri May 26, 2023 12:21 pm FWIW, I established my TIPS ladder in about two hours over two days (would have been shorter if Vanguard's systems were working better). I funded largely from TIPS with relatively short remaining maturities (some held since 1998) and Vanguard's TIPS fund and ST bond fund.

I hope this doesn't run afoul of the "The problem with purchasing a TIPS ladder at a point in time is that is inherently risky" criterion.
Just because it's risky doesn't mean it won't work. It simply means it's risky, if you have targeted a safe level of income.

If you have targeted for a safe level of income it seams to me you would want to get there safely. :happy But to each his own. Unsafely at a point in time you will probably get there. Safely over time you are extremely likely to get there.

BobK
I now have a government guaranteed inflation adjusted stream of income locked in until I'm 95. That seems rather safe. Most of it was funded with TIPS or a TIPS fund, albeit with a shorter average duration. What would have been the advantage of doing this over time?

I can understand the risk of using equities to fund a ladder or set of funds, as the equities could drop in price at an inopportune moment, but that is much less of a problem when using similar securities.
Then we agree. But many people believe they can hold their risk portfolio until the time they purchase the TIPS assets. But accumulating risky assets to purchase low risk assets at one future point in time is - risky.

It seems odd to me that if you want safe income in retirement you would accumulate the assets for purchasing the safe income assets using a risky investment strategy.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
User avatar
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Kevin M »

oxothuk wrote: Thu May 25, 2023 9:14 am
exodusing wrote: Thu May 25, 2023 8:43 am I've been thinking about how to hold TIPS in taxable for part of a LMP since I'm running out of IRA space. All things equal, I prefer a ladder, since it's easy to deal with once set up.
I assume you've already maxed out I-bonds?
Personally I am not doing that, as short-term TIPS have higher real yields--probably enough higher to overcome the tax deferral benefit of I Bonds. The fixed rate on I bonds issued May 1, 2023 to October 31, 2023 is 0.90%. Due to the 1-year lockup, I would compare this to a 1-year TIPS. The yield of the 4/15/2024 TIPS is 3.20% (2.79% using seasonal plus outlier adjustments), and the 7/15/2024 TIPS yield is 2.69% (2.65% SA+O adjusted). These are best ask yields pulled from Fidelity a few minutes ago.

The current I Bond combined rate of 4.30% is not at all attractive to me, with nominal Treasury yields out to one year maturity of 5.2-5.3%-- both exempt from state income tax, so TEYs are higher -- for me nominal Ts are TEY of about 6% at 22% and 9.3% marginal income tax rates.

I am also preferring short term TIPS for very busy and/or very uninterested family members, as there's no way on earth that any of them would be willing to use TD, and I certainly don't want to use it more than I have to.

I only bought I bonds for DW and me last year because of the high yields ($70K using individual and trust accounts and gifts), and plan to sell them as soon as we get 3 months of relatively low rates.

I understand that I bonds are attractive to many for various reasons, and I respect that.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

bobcat2 wrote: Fri May 26, 2023 1:56 pm
Bob, in an old thread you linked to the DFA retirement calculator (now at https://www.dimensional.com/us-en/finan ... calculator). For a retiree, it recommends/assumes 25%-20% global equities and 75%-80% inflation protected bonds. Is this in fact what they recommend (or is it just the design for their TDF)? I would have thought a better recommendation would be sufficient inflation protected bonds to generate desired income with the balance in equities or equities and nominal bonds.
Last edited by exodusing on Fri May 26, 2023 4:14 pm, edited 1 time in total.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

bobcat2 wrote: Fri May 26, 2023 1:56 pm
exodusing wrote: Fri May 26, 2023 1:22 pm
bobcat2 wrote: Fri May 26, 2023 1:06 pm
exodusing wrote: Fri May 26, 2023 12:21 pm FWIW, I established my TIPS ladder in about two hours over two days (would have been shorter if Vanguard's systems were working better). I funded largely from TIPS with relatively short remaining maturities (some held since 1998) and Vanguard's TIPS fund and ST bond fund.

I hope this doesn't run afoul of the "The problem with purchasing a TIPS ladder at a point in time is that is inherently risky" criterion.
Just because it's risky doesn't mean it won't work. It simply means it's risky, if you have targeted a safe level of income.

If you have targeted for a safe level of income it seams to me you would want to get there safely. :happy But to each his own. Unsafely at a point in time you will probably get there. Safely over time you are extremely likely to get there.

BobK
I now have a government guaranteed inflation adjusted stream of income locked in until I'm 95. That seems rather safe. Most of it was funded with TIPS or a TIPS fund, albeit with a shorter average duration. What would have been the advantage of doing this over time?

I can understand the risk of using equities to fund a ladder or set of funds, as the equities could drop in price at an inopportune moment, but that is much less of a problem when using similar securities.
Then we agree. But many people believe they can hold their risk portfolio until the time they purchase the TIPS assets. But accumulating risky assets to purchase low risk assets at one future point in time is - risky.

It seems odd to me that if you want safe income in retirement you would accumulate the assets for purchasing the safe income assets using a risky investment strategy.

BobK
This seems overly cautious to the point of potentially gimping returns during the accumulation years.

Note that DFA's table doesn't have any TIPS prior to 15 years from retirement date:

https://www.dimensional.com/us-en/finan ... calculator
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

watchnerd wrote: Fri May 26, 2023 3:52 pm
bobcat2 wrote: Fri May 26, 2023 1:56 pm
exodusing wrote: Fri May 26, 2023 1:22 pm
bobcat2 wrote: Fri May 26, 2023 1:06 pm
exodusing wrote: Fri May 26, 2023 12:21 pm FWIW, I established my TIPS ladder in about two hours over two days (would have been shorter if Vanguard's systems were working better). I funded largely from TIPS with relatively short remaining maturities (some held since 1998) and Vanguard's TIPS fund and ST bond fund.

I hope this doesn't run afoul of the "The problem with purchasing a TIPS ladder at a point in time is that is inherently risky" criterion.
Just because it's risky doesn't mean it won't work. It simply means it's risky, if you have targeted a safe level of income.

If you have targeted for a safe level of income it seams to me you would want to get there safely. :happy But to each his own. Unsafely at a point in time you will probably get there. Safely over time you are extremely likely to get there.

BobK
I now have a government guaranteed inflation adjusted stream of income locked in until I'm 95. That seems rather safe. Most of it was funded with TIPS or a TIPS fund, albeit with a shorter average duration. What would have been the advantage of doing this over time?

I can understand the risk of using equities to fund a ladder or set of funds, as the equities could drop in price at an inopportune moment, but that is much less of a problem when using similar securities.
Then we agree. But many people believe they can hold their risk portfolio until the time they purchase the TIPS assets. But accumulating risky assets to purchase low risk assets at one future point in time is - risky.

It seems odd to me that if you want safe income in retirement you would accumulate the assets for purchasing the safe income assets using a risky investment strategy.

BobK
This seems overly cautious to the point of potentially gimping returns during the accumulation years.

Note that DFA's table doesn't have any TIPS prior to 15 years from retirement date:

https://www.dimensional.com/us-en/finan ... calculator
The idea is to transition to the TIPS heavy portfolio over 15 years rather than go from the old portfolio to TIPS at once.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

exodusing wrote: Fri May 26, 2023 4:16 pm
The idea is to transition to the TIPS heavy portfolio over 15 years rather than go from the old portfolio to TIPS at once.
Bob was saying to starting acquiring TIPS prior to 15 years from retirement.

The DFA table shows 0 TIPS at -20 years.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
bobcat2
Posts: 6074
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: Now that long TIPS again yield more than 1.70% I will…

Post by bobcat2 »

watchnerd wrote: Fri May 26, 2023 3:52 pm
This seems overly cautious to the point of potentially gimping returns during the accumulation years.

Note that DFA's table doesn't have any TIPS prior to 15 years from retirement date:

https://www.dimensional.com/us-en/finan ... calculator
Yes, at 20 years out DFA shows 21% global bonds and zero percent TIPS. I hardly think if you went to 11% global bonds and 10% TIPS at 20 years out you would gimp returns by very much. :wink: At 15 years DFA shows 19% TIPS. To me that seems like a reasonable default at 15 years and would indicate that at 18-19 years DFA is moving to TIPS. I doubt very much if they plan zero percent TIPS at 16 years and 19% at 15 years.

More importantly at ten years out DFA shows 38% TIPS - not zero percent TIPS. And at five years out DFA shows 56% TIPS in the portfolio - not zero percent. I am not wedded to 38% and 56% but in general that sounds about right on average. However, few of us are exactly average. So, depending on the preferences and financial situations of individual investors, some will have somewhat higher allocations to TIPS five and ten years out and others will have lower TIPS allocations.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
slicendice
Posts: 563
Joined: Tue Sep 22, 2020 12:08 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by slicendice »

deleted
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

If I have a TIPS ladder, I know exactly how much I will receive in inflation-adjusted dollars for each year of the ladder.

If I use funds, how do I compute the annual real payout? As an example, assume I want a 30 year level income stream of $100,000/year. How would I do this with existing funds, say LTPZ and the two Vanguard TIPS funds?

For a ladder I could use #cruncher's spreadsheet or tipsadder.com. https://www.tipsladder.com/build?bondCh ... ome=100000 shows the exact bonds to buy and the $2,352,204 cost.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

exodusing wrote: Fri May 26, 2023 7:33 pm If I have a TIPS ladder, I know exactly how much I will receive in inflation-adjusted dollars for each year of the ladder.

If I use funds, how do I compute the annual real payout? As an example, assume I want a 30 year level income stream of $100,000/year. How would I do this with existing funds, say LTPZ and the two Vanguard TIPS funds?

For a ladder I could use #cruncher's spreadsheet or tipsadder.com. https://www.tipsladder.com/build?bondCh ... ome=100000 shows the exact bonds to buy and the $2,352,204 cost.
Dying to hear the answer to this because I've never been able to figure it out when it comes to funds.

Seems like, at best, you've got some fuzzy error bars.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
JBTX
Posts: 11205
Joined: Wed Jul 26, 2017 12:46 pm

Re: Now that long TIPS again yield more than 1.70% I will…

Post by JBTX »

watchnerd wrote: Fri May 26, 2023 1:00 pm
Lawrence of Suburbia wrote: Fri May 26, 2023 12:49 pm I'm happy with the token amount of VIPSX(?) in my Target 2025 fund. See no reason to divert any more $s from my regular investments; I've been persuaded by numerous folks much smarter than I that stocks are a better long-term bet vs. inflation.
The research literature finds stocks vs inflation to often be overstated:
It is often claimed that equities are a hedge against inflation. Figure 16 showed this is incorrect. Equities performed especially well in real terms when inflation was low. High inflation impaired performance and deflation led to lower returns than on government bonds. The correlation between real equity returns and inflation was negative, i.e. equities were a poor hedge against inflation. There is an extensive literature to back this up. Fama and Schwert (2077), Fama (1981), and Boudoukh and Richardson (1993) are three classic papers, and Tatom (2011) is a useful review article.
https://www.credit-suisse.com/media/ass ... dition.pdf

I’ve never bought into the stocks inflation thing. I look at it like this

A tips bond, maturing at the time you need it, is pretty safe. You will get what you expected. As you approach the end date, the duration gets lower.

A long term TIPS fund, in itself, may not be a great inflation hedge because the change in real interest rates can overwhelm the inflation protection. LPTZ lost a third of its value in 9 months. It could take quite a while to recover that. But I don’t think anybody would plan to buy LTPZ and hold it for 30 years.

Stocks have a TIPS like component, in the their revenues and earnings tend to be inflation adjusted, all else equal. But they also have a longer duration, and they have interest rate risk in terms of multiple. So their interest rate risk adds a fair amount of non inflation related variability. Somewhat like holding LTPZ for a long time, but probably worse. Plus add in a cyclical component related to the economy, and idiosyncratic risks related to stock valuation during bubbles, etc.

If you wait long enough, they are inflation protected, in that ultimately they tend to revert to the mean, in terms of valuations (rates) and economic cycle. But that can be decades, or worse in the case of Japan (high idiosyncratic risk).
dcabler
Posts: 4482
Joined: Wed Feb 19, 2014 10:30 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by dcabler »

watchnerd wrote: Fri May 26, 2023 10:04 pm
exodusing wrote: Fri May 26, 2023 7:33 pm If I have a TIPS ladder, I know exactly how much I will receive in inflation-adjusted dollars for each year of the ladder.

If I use funds, how do I compute the annual real payout? As an example, assume I want a 30 year level income stream of $100,000/year. How would I do this with existing funds, say LTPZ and the two Vanguard TIPS funds?

For a ladder I could use #cruncher's spreadsheet or tipsadder.com. https://www.tipsladder.com/build?bondCh ... ome=100000 shows the exact bonds to buy and the $2,352,204 cost.
Dying to hear the answer to this because I've never been able to figure it out when it comes to funds.

Seems like, at best, you've got some fuzzy error bars.
In very broad, approximate strokes for duration matching using funds.
30 years of income is roughly 15 years investment horizon
Find current duration of LTPZ and your favorite intermediate bond fund. Let's just say SCHP
Calculate the weight of LTPZ and SCHP that results in a weighted duration that's 15 years. Hold that percentage of each.
Calculate the effective yield of this weighted combination of LTPZ and SCHP using the current real yields of these two funds
Withdrawal % is calculated with Excel function: PMT(Yield, 30, -1, 0, 1)
Withdraw the dollar amount.
Next year you have 29 years remaining, so a 29/2 years investment horizon
Rinse and repeat.
At some point in the future, the required duration needed is less than SCHP's duration. At that point, you will no longer own LTPZ, but instead you'll transition to owning SCHP and, say, STIP
Continue to rinse and repeat until the duration required is less than STIP's duration.
Then you have choices including:
STIP + some nominal investment (CD's, ultrashort bond fund, etc)
Ibonds - this is what BobK likes
Or even just sticking with 100% STIP to the bitter end.

Challenges
- Not every fund reports its real yield. Most report their nominal yields. SEC only recommends that such funds report real yields but they don't require it. But there are only a limited number of TIPS indexes around, so sometimes another fund reports it. For example while SCHP doesn't report real yields, iShares TIP does and they both track the same index. Sadly, there is no equivalent for LTPZ. #cruncher calculates and reports yields for all of the major indexes weekly in this thread: viewtopic.php?p=7241758#p7241758
- Same thing with duration. Most bond funds calculate duration based on nominal yields, not real yields. Nominal-based might be close enough or you can use the data from #cruncher

Yes, there are definitely fuzzy error bars. Based on some backtesting I did with #cruncher's data, it's +/- a few percent but with a relatively flat mean. But this data is very limited. There is enough data out there in the wild to use #cruncher's technique to create a history that goes further back, if anybody wished to pursue it.

There are also some other methods of varying complexity to get close as well.
There's a shortcut method that vineviz proposed here: viewtopic.php?p=6869837#p6869837
I'm not especially keen on this one as I don't think you can cover the longest timeframe you might need. It's main advantage is that you don't need to do any rebalancing between funds.

Another thing I've played with that seemingly does alright. A bit larger error bars, but less maintenance.
Start with LTPZ only. Find real yield for LTPZ. Withdrawal percentage = PMT(yield, 30, -1, 0, 1)
Same thing next year but PMT (yield, 29, -1, 0, 1)
At some point, move everything from LTPZ to SCHP and continue on.
Again at some point, move everything from SCHP to STIP and continue to the end
So instead of continually adjusting the ratios between 2 funds, you're just holding 1 fund at a time and doing the transition at pretty much the same points in time where you would transition in the original duration matching method.

Cheers.
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Phyneas »

dcabler wrote: Sat May 27, 2023 6:14 am
watchnerd wrote: Fri May 26, 2023 10:04 pm
exodusing wrote: Fri May 26, 2023 7:33 pm If I have a TIPS ladder, I know exactly how much I will receive in inflation-adjusted dollars for each year of the ladder.

If I use funds, how do I compute the annual real payout? As an example, assume I want a 30 year level income stream of $100,000/year. How would I do this with existing funds, say LTPZ and the two Vanguard TIPS funds?

For a ladder I could use #cruncher's spreadsheet or tipsadder.com. https://www.tipsladder.com/build?bondCh ... ome=100000 shows the exact bonds to buy and the $2,352,204 cost.
Dying to hear the answer to this because I've never been able to figure it out when it comes to funds.

Seems like, at best, you've got some fuzzy error bars.
In very broad, approximate strokes for duration matching using funds.
30 years of income is roughly 15 years investment horizon
Find current duration of LTPZ and your favorite intermediate bond fund. Let's just say SCHP
Calculate the weight of LTPZ and SCHP that results in a weighted duration that's 15 years. Hold that percentage of each.
Calculate the effective yield of this weighted combination of LTPZ and SCHP using the current real yields of these two funds
Withdrawal % is calculated with Excel function: PMT(Yield, 30, -1, 0, 1)
Withdraw the dollar amount.
Next year you have 29 years remaining, so a 29/2 years investment horizon
Rinse and repeat.
At some point in the future, the required duration needed is less than SCHP's duration. At that point, you will no longer own LTPZ, but instead you'll transition to owning SCHP and, say, STIP
Continue to rinse and repeat until the duration required is less than STIP's duration.
Then you have choices including:
STIP + some nominal investment (CD's, ultrashort bond fund, etc)
Ibonds - this is what BobK likes
Or even just sticking with 100% STIP to the bitter end.

Challenges
- Not every fund reports its real yield. Most report their nominal yields. SEC only recommends that such funds report real yields but they don't require it. But there are only a limited number of TIPS indexes around, so sometimes another fund reports it. For example while SCHP doesn't report real yields, iShares TIP does and they both track the same index. Sadly, there is no equivalent for LTPZ. #cruncher calculates and reports yields for all of the major indexes weekly in this thread: viewtopic.php?p=7241758#p7241758
- Same thing with duration. Most bond funds calculate duration based on nominal yields, not real yields. Nominal-based might be close enough or you can use the data from #cruncher

Yes, there are definitely fuzzy error bars. Based on some backtesting I did with #cruncher's data, it's +/- a few percent but with a relatively flat mean. But this data is very limited. There is enough data out there in the wild to use #cruncher's technique to create a history that goes further back, if anybody wished to pursue it.

There are also some other methods of varying complexity to get close as well.
There's a shortcut method that vineviz proposed here: viewtopic.php?p=6869837#p6869837
I'm not especially keen on this one as I don't think you can cover the longest timeframe you might need. It's main advantage is that you don't need to do any rebalancing between funds.

Another thing I've played with that seemingly does alright. A bit larger error bars, but less maintenance.
Start with LTPZ only. Find real yield for LTPZ. Withdrawal percentage = PMT(yield, 30, -1, 0, 1)
Same thing next year but PMT (yield, 29, -1, 0, 1)
At some point, move everything from LTPZ to SCHP and continue on.
Again at some point, move everything from SCHP to STIP and continue to the end
So instead of continually adjusting the ratios between 2 funds, you're just holding 1 fund at a time and doing the transition at pretty much the same points in time where you would transition in the original duration matching method.

Cheers.
With this scenario, is there a risk of selling LTPZ at an incredibly inopportune time (like 2022-2023) and never making up the price difference as it isn't a rolling ladder, just one used in decumulation. I.e. if the number of years it takes you to move from LTPZ + SCHP to just SCHP is a period in which LTPZ never recovers its price, even with the increased coupons, could you still lose vs a regular TIPS ladder where the bond is maturing nearest to your funding year and you are guaranteed (absent a default) to get both the principal and coupon?
Last edited by Phyneas on Sat May 27, 2023 11:29 am, edited 1 time in total.
60% AVGE | 20 Year TIPS LMP | 5% Cash
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

dcabler wrote: Sat May 27, 2023 6:14 am <snipped>
Broad approximate strokes with fuzzy error bars and limited backtesting does not exactly inspire confidence, especially when the ladder alternative allows for exact calculation and predictability (subject to the 2033-2039 gap, but funds have the same issue). The question was how much to invest today to produce a specified guaranteed real cash flow for a specified number of years.

Given that the real yield and duration of the funds change over time, it would seem that you can not calculate today an amount to invest that would guarantee you a specified annual real cash flow for a specified number of years.

We do have the self-correcting nature of bonds and bond funds, that is, if yield goes down principle value goes up (and vice versa) which keeps interest payments relatively constant for some time, but if there's a long enough period of low yield, then interest payments will suffer. A ladder does not have this problem.

Without the real yield information for the funds you're using, I don't see how to proceed. For longer horizons, LTPZ seems necessary if you're to use funds and it doesn't report real yields. Are its actual holdings in close to real time reported anywhere? Without this information, our fuzziness suffers a massive increase.

If the goal is to generate a specified cash flow each year for a specified number of years, individual bonds seem to have a great advantage - in that you can figure out how to achieve the goal. Buying them is easy. There's a touch more hassle if holding them in a taxable account, but it seems tax reporting is easy (see posts above). You just need the cash flow to pay the taxes on the inflation adjustment.

One should be wary of being overly precise, since expenses are not as predictable as one may wish, but the fuzziness of funds may go beyond this.

I hope I'm being too negative on funds.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

dcabler wrote: Sat May 27, 2023 6:14 am
watchnerd wrote: Fri May 26, 2023 10:04 pm
exodusing wrote: Fri May 26, 2023 7:33 pm If I have a TIPS ladder, I know exactly how much I will receive in inflation-adjusted dollars for each year of the ladder.

If I use funds, how do I compute the annual real payout? As an example, assume I want a 30 year level income stream of $100,000/year. How would I do this with existing funds, say LTPZ and the two Vanguard TIPS funds?

For a ladder I could use #cruncher's spreadsheet or tipsadder.com. https://www.tipsladder.com/build?bondCh ... ome=100000 shows the exact bonds to buy and the $2,352,204 cost.
Dying to hear the answer to this because I've never been able to figure it out when it comes to funds.

Seems like, at best, you've got some fuzzy error bars.
In very broad, approximate strokes for duration matching using funds.
30 years of income is roughly 15 years investment horizon
Find current duration of LTPZ and your favorite intermediate bond fund. Let's just say SCHP
Calculate the weight of LTPZ and SCHP that results in a weighted duration that's 15 years. Hold that percentage of each.
Calculate the effective yield of this weighted combination of LTPZ and SCHP using the current real yields of these two funds
Withdrawal % is calculated with Excel function: PMT(Yield, 30, -1, 0, 1)
Withdraw the dollar amount.
Next year you have 29 years remaining, so a 29/2 years investment horizon
Rinse and repeat.
At some point in the future, the required duration needed is less than SCHP's duration. At that point, you will no longer own LTPZ, but instead you'll transition to owning SCHP and, say, STIP
Continue to rinse and repeat until the duration required is less than STIP's duration.
Then you have choices including:
STIP + some nominal investment (CD's, ultrashort bond fund, etc)
Ibonds - this is what BobK likes
Or even just sticking with 100% STIP to the bitter end.

Challenges
- Not every fund reports its real yield. Most report their nominal yields. SEC only recommends that such funds report real yields but they don't require it. But there are only a limited number of TIPS indexes around, so sometimes another fund reports it. For example while SCHP doesn't report real yields, iShares TIP does and they both track the same index. Sadly, there is no equivalent for LTPZ. #cruncher calculates and reports yields for all of the major indexes weekly in this thread: viewtopic.php?p=7241758#p7241758
- Same thing with duration. Most bond funds calculate duration based on nominal yields, not real yields. Nominal-based might be close enough or you can use the data from #cruncher

Yes, there are definitely fuzzy error bars. Based on some backtesting I did with #cruncher's data, it's +/- a few percent but with a relatively flat mean. But this data is very limited. There is enough data out there in the wild to use #cruncher's technique to create a history that goes further back, if anybody wished to pursue it.

There are also some other methods of varying complexity to get close as well.
There's a shortcut method that vineviz proposed here: viewtopic.php?p=6869837#p6869837
I'm not especially keen on this one as I don't think you can cover the longest timeframe you might need. It's main advantage is that you don't need to do any rebalancing between funds.

Another thing I've played with that seemingly does alright. A bit larger error bars, but less maintenance.
Start with LTPZ only. Find real yield for LTPZ. Withdrawal percentage = PMT(yield, 30, -1, 0, 1)
Same thing next year but PMT (yield, 29, -1, 0, 1)
At some point, move everything from LTPZ to SCHP and continue on.
Again at some point, move everything from SCHP to STIP and continue to the end
So instead of continually adjusting the ratios between 2 funds, you're just holding 1 fund at a time and doing the transition at pretty much the same points in time where you would transition in the original duration matching method.

Cheers.
Jeeze.....

Yeah, those are some error bars.

I'll stick with individual bonds where I know much more precisely what my future income stream will be.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Now that long TIPS again yield more than 1.70% I will…

Post by watchnerd »

exodusing wrote: Sat May 27, 2023 7:06 am If the goal is to generate a specified cash flow each year for a specified number of years, individual bonds seem to have a great advantage - in that you can figure out how to achieve the goal.
KISS principle :sharebeer
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Now that long TIPS again yield more than 1.70% I will…

Post by Kevin M »

Phyneas wrote: Sat May 27, 2023 6:37 am
dcabler wrote: Sat May 27, 2023 6:14 am
watchnerd wrote: Fri May 26, 2023 10:04 pm
exodusing wrote: Fri May 26, 2023 7:33 pm If I have a TIPS ladder, I know exactly how much I will receive in inflation-adjusted dollars for each year of the ladder.

If I use funds, how do I compute the annual real payout? As an example, assume I want a 30 year level income stream of $100,000/year. How would I do this with existing funds, say LTPZ and the two Vanguard TIPS funds?

For a ladder I could use #cruncher's spreadsheet or tipsadder.com. https://www.tipsladder.com/build?bondCh ... ome=100000 shows the exact bonds to buy and the $2,352,204 cost.
Dying to hear the answer to this because I've never been able to figure it out when it comes to funds.

Seems like, at best, you've got some fuzzy error bars.
In very broad, approximate strokes for duration matching using funds.
30 years of income is roughly 15 years investment horizon
Find current duration of LTPZ and your favorite intermediate bond fund. Let's just say SCHP
Calculate the weight of LTPZ and SCHP that results in a weighted duration that's 15 years. Hold that percentage of each.
Calculate the effective yield of this weighted combination of LTPZ and SCHP using the current real yields of these two funds
Withdrawal % is calculated with Excel function: PMT(Yield, 30, -1, 0, 1)
Withdraw the dollar amount.
Next year you have 29 years remaining, so a 29/2 years investment horizon
Rinse and repeat.
At some point in the future, the required duration needed is less than SCHP's duration. At that point, you will no longer own LTPZ, but instead you'll transition to owning SCHP and, say, STIP
Continue to rinse and repeat until the duration required is less than STIP's duration.
Then you have choices including:
STIP + some nominal investment (CD's, ultrashort bond fund, etc)
Ibonds - this is what BobK likes
Or even just sticking with 100% STIP to the bitter end.

Challenges
- Not every fund reports its real yield. Most report their nominal yields. SEC only recommends that such funds report real yields but they don't require it. But there are only a limited number of TIPS indexes around, so sometimes another fund reports it. For example while SCHP doesn't report real yields, iShares TIP does and they both track the same index. Sadly, there is no equivalent for LTPZ. #cruncher calculates and reports yields for all of the major indexes weekly in this thread: viewtopic.php?p=7241758#p7241758
- Same thing with duration. Most bond funds calculate duration based on nominal yields, not real yields. Nominal-based might be close enough or you can use the data from #cruncher

Yes, there are definitely fuzzy error bars. Based on some backtesting I did with #cruncher's data, it's +/- a few percent but with a relatively flat mean. But this data is very limited. There is enough data out there in the wild to use #cruncher's technique to create a history that goes further back, if anybody wished to pursue it.

There are also some other methods of varying complexity to get close as well.
There's a shortcut method that vineviz proposed here: viewtopic.php?p=6869837#p6869837
I'm not especially keen on this one as I don't think you can cover the longest timeframe you might need. It's main advantage is that you don't need to do any rebalancing between funds.

Another thing I've played with that seemingly does alright. A bit larger error bars, but less maintenance.
Start with LTPZ only. Find real yield for LTPZ. Withdrawal percentage = PMT(yield, 30, -1, 0, 1)
Same thing next year but PMT (yield, 29, -1, 0, 1)
At some point, move everything from LTPZ to SCHP and continue on.
Again at some point, move everything from SCHP to STIP and continue to the end
So instead of continually adjusting the ratios between 2 funds, you're just holding 1 fund at a time and doing the transition at pretty much the same points in time where you would transition in the original duration matching method.

Cheers.
With this scenario, is there a risk of selling LTPZ at an incredibly inopportune time (like 2022-2023) and never making up the price difference as it isn't a rolling ladder, just one used in decumulation. I.e. if the number of years it takes you to move from LTPZ + SCHP to just LTPZ is a period in which LTPZ never recovers its price, even with the increased coupons, could you still lose vs a regular TIPS ladder where the bond is maturing nearest to your funding year and you are guaranteed (absent a default) to get both the principal and coupon?
This is a point I've made several times over the last few days (and a number of times over the years here). I believe the answer to your question is "yes".

The only rebuttal I've heard is that there will be opportune times to offset the inopportune times to sell, and it will balance out over the long term. My rebuttal to that is that you may not be so lucky if there is a sustained period of increasing yields, or even if the yields don't return to the lower levels you bought at before you've sold a lot of your longer term fund.

This assumes a positive yield curve, so that you are earning less on the shorter term TIPS fund shares you buy than you would if you continued to hold the longer term fund shares instead of selling them to buy the shorter term shares.
If I make a calculation error, #Cruncher probably will let me know.
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

Since taxes are a consideration, here's a post by #cruncher on why it's better to hold TIPS in an IRA than in taxable. viewtopic.php?p=2100674#p2100674

However, it can be better to hold TIPS than other taxable bonds in taxable.
dcabler
Posts: 4482
Joined: Wed Feb 19, 2014 10:30 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by dcabler »

watchnerd wrote: Sat May 27, 2023 10:00 am
dcabler wrote: Sat May 27, 2023 6:14 am
watchnerd wrote: Fri May 26, 2023 10:04 pm
exodusing wrote: Fri May 26, 2023 7:33 pm If I have a TIPS ladder, I know exactly how much I will receive in inflation-adjusted dollars for each year of the ladder.

If I use funds, how do I compute the annual real payout? As an example, assume I want a 30 year level income stream of $100,000/year. How would I do this with existing funds, say LTPZ and the two Vanguard TIPS funds?

For a ladder I could use #cruncher's spreadsheet or tipsadder.com. https://www.tipsladder.com/build?bondCh ... ome=100000 shows the exact bonds to buy and the $2,352,204 cost.
Dying to hear the answer to this because I've never been able to figure it out when it comes to funds.

Seems like, at best, you've got some fuzzy error bars.
In very broad, approximate strokes for duration matching using funds.
30 years of income is roughly 15 years investment horizon
Find current duration of LTPZ and your favorite intermediate bond fund. Let's just say SCHP
Calculate the weight of LTPZ and SCHP that results in a weighted duration that's 15 years. Hold that percentage of each.
Calculate the effective yield of this weighted combination of LTPZ and SCHP using the current real yields of these two funds
Withdrawal % is calculated with Excel function: PMT(Yield, 30, -1, 0, 1)
Withdraw the dollar amount.
Next year you have 29 years remaining, so a 29/2 years investment horizon
Rinse and repeat.
At some point in the future, the required duration needed is less than SCHP's duration. At that point, you will no longer own LTPZ, but instead you'll transition to owning SCHP and, say, STIP
Continue to rinse and repeat until the duration required is less than STIP's duration.
Then you have choices including:
STIP + some nominal investment (CD's, ultrashort bond fund, etc)
Ibonds - this is what BobK likes
Or even just sticking with 100% STIP to the bitter end.

Challenges
- Not every fund reports its real yield. Most report their nominal yields. SEC only recommends that such funds report real yields but they don't require it. But there are only a limited number of TIPS indexes around, so sometimes another fund reports it. For example while SCHP doesn't report real yields, iShares TIP does and they both track the same index. Sadly, there is no equivalent for LTPZ. #cruncher calculates and reports yields for all of the major indexes weekly in this thread: viewtopic.php?p=7241758#p7241758
- Same thing with duration. Most bond funds calculate duration based on nominal yields, not real yields. Nominal-based might be close enough or you can use the data from #cruncher

Yes, there are definitely fuzzy error bars. Based on some backtesting I did with #cruncher's data, it's +/- a few percent but with a relatively flat mean. But this data is very limited. There is enough data out there in the wild to use #cruncher's technique to create a history that goes further back, if anybody wished to pursue it.

There are also some other methods of varying complexity to get close as well.
There's a shortcut method that vineviz proposed here: viewtopic.php?p=6869837#p6869837
I'm not especially keen on this one as I don't think you can cover the longest timeframe you might need. It's main advantage is that you don't need to do any rebalancing between funds.

Another thing I've played with that seemingly does alright. A bit larger error bars, but less maintenance.
Start with LTPZ only. Find real yield for LTPZ. Withdrawal percentage = PMT(yield, 30, -1, 0, 1)
Same thing next year but PMT (yield, 29, -1, 0, 1)
At some point, move everything from LTPZ to SCHP and continue on.
Again at some point, move everything from SCHP to STIP and continue to the end
So instead of continually adjusting the ratios between 2 funds, you're just holding 1 fund at a time and doing the transition at pretty much the same points in time where you would transition in the original duration matching method.

Cheers.
Jeeze.....

Yeah, those are some error bars.

I'll stick with individual bonds where I know much more precisely what my future income stream will be.
What I saw was +/- a few 1's of % when I did this. I don't see it as being that large. Turns out there's more data if you backtest the same thing with nominal treasury funds and make nominal withdrawals. Saw the same magnitude of error bars over a longer time period with them. You can also make withdrawals/rebalance more often, which also helps - I'm doing both the withdrawal and rebalancing quarterly. I know at least one person doing rebalancing monthly but withdrawing on a longer schedule. I like what I'm doing because it means the withdrawals are on my schedule, not on the coupon/maturity schedule of the underlying bonds.

Anyway like most things in investing, it comes down to personal preference/taste. And it doesn't need to be once-and-forget decision to do duration matching and never switch to actual bonds. You can also take some hybrid approaches. For example, you can build a bond ladder, but duration match for the black hole, selling from the duration matched portion to buy a new rung to fill each of the empty years as they become available.

Cheers.
Last edited by dcabler on Sun May 28, 2023 5:51 am, edited 3 times in total.
dcabler
Posts: 4482
Joined: Wed Feb 19, 2014 10:30 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by dcabler »

exodusing wrote: Sat May 27, 2023 11:29 am Since taxes are a consideration, here's a post by #cruncher on why it's better to hold TIPS in an IRA than in taxable. viewtopic.php?p=2100674#p2100674

However, it can be better to hold TIPS than other taxable bonds in taxable.
Yep - things would need to get pretty hosed in my portfolio for me to hold any otherwise taxable bonds in anything but my IRA.

cheers.
oxothuk
Posts: 889
Joined: Thu Nov 10, 2011 7:35 pm

Re: Now that long TIPS again yield more than 1.70% I will…

Post by oxothuk »

dcabler wrote: Sat May 27, 2023 5:58 pm
exodusing wrote: Sat May 27, 2023 11:29 am Since taxes are a consideration, here's a post by #cruncher on why it's better to hold TIPS in an IRA than in taxable. viewtopic.php?p=2100674#p2100674

However, it can be better to hold TIPS than other taxable bonds in taxable.
Yep - things would need to get pretty hosed in my portfolio for me to hold any otherwise taxable bonds in anything but my IRA.

cheers.
That's the beauty of I-bonds. They are a great way to extend your tax-deferred space.
User avatar
#Cruncher
Posts: 3975
Joined: Fri May 14, 2010 2:33 am
Location: New York City
Contact:

Re: Now that long TIPS again yield more than 1.70% I will…

Post by #Cruncher »

Edited 2:30 PM to correct "d Tax at End" for I Bond. Thanks to erp in this post below.
oxothuk wrote: Sat May 27, 2023 7:06 pm
dcabler wrote: Sat May 27, 2023 5:58 pm
exodusing wrote: Sat May 27, 2023 11:29 am Since taxes are a consideration, here's a post by #cruncher on why it's better to hold TIPS in an IRA than in taxable. viewtopic.php?p=2100674#p2100674 ...
Yep - things would need to get pretty hosed in my portfolio for me to hold any otherwise taxable bonds in anything but my IRA. ...
That's the beauty of I-bonds. They are a great way to extend your tax-deferred space.
I often hear on the forum that I Bonds (and EE Bonds) "extend your tax-deferred space". This is only partially true. I Bonds get better tax treatment than a regular taxable account; but not as good as a Roth or Traditional IRA. To show this, I've extended ( :wink: ) the table from the 2014 post exodusing references to include an I Bond having a 0% fixed rate to match the assumed 0% real yield of the TIPS in the Roth, TIRA, and Taxable cases.

Note that the I Bond's performance for both 2% and 5% inflation is better than Taxable but worse than either IRA. And like the Taxable case, its real after tax return deteriorates as inflation increases -- unlike the case with the Roth or traditional IRA.

Code: Select all

Pretax wages              1,000
Years                        30
Federal tax                 25%
State tax                    5%
                            Roth IRA         Trad IRA          Taxable          I Bonds
                          ------------     ------------     ------------     ------------
CPI Annual Change         2.00%  5.00%     2.00%  5.00%     2.00%  5.00%     2.00%  5.00%
a  Growth Rate            2.00%  5.00%     2.00%  5.00%     1.50%  3.75%     2.00%  5.00%
b  After Tax Investment     700    700     1,000  1,000       700    700       700    700
c  Grows To               1,268  3,025     1,811  4,322     1,094  2,112     1,268  3,025
d  Tax at End                -      -        543  1,297        -      -       [142    581]  I Bond corrected
e  After Tax              1,268  3,025     1,268  3,025     1,094  2,112     1,126  2,444
f  Real After Tax           700    700       700    700       604    489       622    565
If you wish to duplicate the calculations with other assumptions: Select All, Copy, and Paste [*] the following at cell A1 of an empty Excel sheet:

Code: Select all

Pretax wages	1000
Years	30
Federal tax	0.25
State tax	0.05
	Roth	IRA	Trad	IRA	Tax	able	I	Bonds
CPI Annual Change	0.02	0.05	=B6	=C6	=D6	=E6	=F6	=G6
a  Growth Rate	=B6	=C6	=D6	=E6	=F6*(1-$B3)	=G6*(1-$B3)	=H6	=I6
b  After Tax Investment	=$B1*(1-$B3-$B4)	=$B1*(1-$B3-$B4)	=$B1	=$B1	=$B1*(1-$B3-$B4)	=$B1*(1-$B3-$B4)	=$B1*(1-$B3-$B4)	=$B1*(1-$B3-$B4)
c  Grows To	=B8*(1+B7)^$B2	=C8*(1+C7)^$B2	=D8*(1+D7)^$B2	=E8*(1+E7)^$B2	=F8*(1+F7)^$B2	=G8*(1+G7)^$B2	=H8*(1+H7)^$B2	=I8*(1+I7)^$B2
d  Tax at End	0	0	=D9*($B3+$B4)	=E9*($B3+$B4)	0	0	=$B3*(H9-H8)	=$B3*(I9-I8)
e  After Tax	=B9-B10	=C9-C10	=D9-D10	=E9-E10	=F9-F10	=G9-G10	=H9-H10	=I9-I10
f  Real After Tax	=B11/(1+B6)^$B2	=C11/(1+C6)^$B2	=D11/(1+D6)^$B2	=E11/(1+E6)^$B2	=F11/(1+F6)^$B2	=G11/(1+G6)^$B2	=H11/(1+H6)^$B2	=I11/(1+I6)^$B2
* If you have trouble pasting, try "Paste Special" and "Text".
Last edited by #Cruncher on Sun May 28, 2023 1:32 pm, edited 1 time in total.
dcabler
Posts: 4482
Joined: Wed Feb 19, 2014 10:30 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by dcabler »

oxothuk wrote: Sat May 27, 2023 7:06 pm
dcabler wrote: Sat May 27, 2023 5:58 pm
exodusing wrote: Sat May 27, 2023 11:29 am Since taxes are a consideration, here's a post by #cruncher on why it's better to hold TIPS in an IRA than in taxable. viewtopic.php?p=2100674#p2100674

However, it can be better to hold TIPS than other taxable bonds in taxable.
Yep - things would need to get pretty hosed in my portfolio for me to hold any otherwise taxable bonds in anything but my IRA.

cheers.
That's the beauty of I-bonds. They are a great way to extend your tax-deferred space.
Yes they help. I own them. But no practical way to purchase enough to fund the amount of safe income I'd like at this point. So, for me, they're a supplement to TIPS.

Cheers.
dcabler
Posts: 4482
Joined: Wed Feb 19, 2014 10:30 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by dcabler »

exodusing wrote: Sat May 27, 2023 7:06 am
dcabler wrote: Sat May 27, 2023 6:14 am <snipped>
Broad approximate strokes with fuzzy error bars and limited backtesting does not exactly inspire confidence, especially when the ladder alternative allows for exact calculation and predictability (subject to the 2033-2039 gap, but funds have the same issue). The question was how much to invest today to produce a specified guaranteed real cash flow for a specified number of years.

Given that the real yield and duration of the funds change over time, it would seem that you can not calculate today an amount to invest that would guarantee you a specified annual real cash flow for a specified number of years.

We do have the self-correcting nature of bonds and bond funds, that is, if yield goes down principle value goes up (and vice versa) which keeps interest payments relatively constant for some time, but if there's a long enough period of low yield, then interest payments will suffer. A ladder does not have this problem.

Without the real yield information for the funds you're using, I don't see how to proceed. For longer horizons, LTPZ seems necessary if you're to use funds and it doesn't report real yields. Are its actual holdings in close to real time reported anywhere? Without this information, our fuzziness suffers a massive increase.

If the goal is to generate a specified cash flow each year for a specified number of years, individual bonds seem to have a great advantage - in that you can figure out how to achieve the goal. Buying them is easy. There's a touch more hassle if holding them in a taxable account, but it seems tax reporting is easy (see posts above). You just need the cash flow to pay the taxes on the inflation adjustment.

One should be wary of being overly precise, since expenses are not as predictable as one may wish, but the fuzziness of funds may go beyond this.

I hope I'm being too negative on funds.
I noted in my post that #cruncher posts the real yield of the index that LTPZ follows on a weekly basis, as well as pretty much all of the other TIPS indexes any funds use, using actual data from the underlying bonds that make up these indexes. Because I don't want to count on #cruncher continuing to make these posts over the course of my retirement, I've taken the time to learn what he's done and can maintain this for my own usage. So while I'd certainly like PIMCO to post LTPZ's real yield and duration based on real yield, it's not necessary for me that they don't.

Your last point is correct for me. My expenses are not going to be that predictable. That's why I avoid using the term LMP. For me, it's just a matter of how much inflation protected income I want relative to income from risk sources. Having income from both of those sources is, together, going to be extremely fuzzy in the end because of the nature of the equity portion. So I can afford a little bit more fuzziness from the one source which, eventually be 2 sources: TIPS & SS

There is no question in my mind that purchasing an actual ladder gives one absolutely deterministic inflation adjusted income after you purchase it, where that income is known for the days that coupons pay out and on the days that bonds mature. Doing this with bond funds does have some noise around it, no question. There are ways to improve that, but it will never be as deterministic as an actual bond ladder. BobK gave some great reasons why, for him, using bond funds is good enough for his purposes. Some of the same reasons apply for me as well. I also like that I can make the withdrawals on my schedule (quarterly) and not on the schedule of coupon payments/bonds maturing.

Still, I have to be mindful of my DW in case I pass before her. Though she can run rings around me in excel because she was a CS major and I was only an EE, she has little interest in any of this stuff. I have a Plan B and Plan C for her depending on a number of factors. That can include a conversion to an actual TIPS ladder. We have only one kiddo and she was a business major and is more than capable of understanding as well in my absence and would be available to help my DW transition to something else.

By the way, there is one more variant of this that I've seen mentioned on the forum. That is, only calculate the first withdrawal directly using PMT, yields and number of years. Make the same withdrawal every year after that, inflation adjusted. You still need to rebalance between the 2 funds you're holding at any given time, though. Removes all of the income fuziness, but trades that off for uncertainty as to when you actually run out of money. Starts to seem a lot like SWR. I wouldn't do this. If that first withdrawal ends up being above the long term mean of doing the calculations for each withdrawal period, then you're likely to run out of money before your desired end date.

Cheers!
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

dcabler wrote: Sat May 27, 2023 5:57 pm What I saw was +/- a few 1's of % when I did this. I don't see it as being that large.
Would you please post your backtest?

Here's a good thread on implementation viewtopic.php?p=3771655#p3771655, including BobK writing "The effort here, however, is not to almost perfectly immunize the portfolio, but rather to provide relatively safe retirement income." It also includes that one should rebalance between funds frequently to limit interest rate risk, which seems a lot of effort, especially compared to the one-time setup of a ladder.

#cruncher's thread on Consistent Yield & Duration to Help Choose TIPS Fund is viewtopic.php?t=104845

There seems to be two uses of "duration" in this threads. One use is a period of time, the other relates to interest rate sensitivity. Some posters echo Bob's notion that anything close is good enough.
dcabler
Posts: 4482
Joined: Wed Feb 19, 2014 10:30 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by dcabler »

exodusing wrote: Sun May 28, 2023 6:46 am
dcabler wrote: Sat May 27, 2023 5:57 pm What I saw was +/- a few 1's of % when I did this. I don't see it as being that large.
Would you please post your backtest?

Here's a good thread on implementation viewtopic.php?p=3771655#p3771655, including BobK writing "The effort here, however, is not to almost perfectly immunize the portfolio, but rather to provide relatively safe retirement income." It also includes that one should rebalance between funds frequently to limit interest rate risk, which seems a lot of effort, especially compared to the one-time setup of a ladder.

#cruncher's thread on Consistent Yield & Duration to Help Choose TIPS Fund is viewtopic.php?t=104845

There seems to be two uses of "duration" in this threads. One use is a period of time, the other relates to interest rate sensitivity. Some posters echo Bob's notion that anything close is good enough.
I'll dig it up and post it. Might be a while before I can get to it/find it. In the meantime, #cruncher posted an example a while back, but it wasn't a backtest using historical data - more of a hypothetical but it illustrates the point well.
viewtopic.php?p=3768998#p3768998
(link fixed - cut and paste error the first time I posted)

Rebalancing is very little effort if you're set everything up in a spreadsheet beforehand.
- Takes me less than 2 minutes, once a quarter to update my version of #cruncher's spreadsheet to get the latest durations and yields to plug into my spreadsheet which calculates the withdrawals/rebalance simultaneously. I plan to add more automation such that upon opening the spreadsheet it gives me directly what the transactions need to be by scraping data sources. Till then, I'll sacrifice those couple of minutes. :D
- Then the withdrawal and rebalance also just takes a couple of minutes.

I'm with BobK on the "good enough" comment.

cheers.
Last edited by dcabler on Sun May 28, 2023 8:45 am, edited 2 times in total.
exodusing
Posts: 2196
Joined: Thu Oct 13, 2022 7:32 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by exodusing »

dcabler wrote: Sun May 28, 2023 7:40 am
exodusing wrote: Sun May 28, 2023 6:46 am
dcabler wrote: Sat May 27, 2023 5:57 pm What I saw was +/- a few 1's of % when I did this. I don't see it as being that large.
Would you please post your backtest?

Here's a good thread on implementation viewtopic.php?p=3771655#p3771655, including BobK writing "The effort here, however, is not to almost perfectly immunize the portfolio, but rather to provide relatively safe retirement income." It also includes that one should rebalance between funds frequently to limit interest rate risk, which seems a lot of effort, especially compared to the one-time setup of a ladder.

#cruncher's thread on Consistent Yield & Duration to Help Choose TIPS Fund is viewtopic.php?t=104845

There seems to be two uses of "duration" in this threads. One use is a period of time, the other relates to interest rate sensitivity. Some posters echo Bob's notion that anything close is good enough.
I'll dig it up and post it. Might be a while before I can get to it/find it. In the meantime, #cruncher posted an example a while back, but it wasn't a backtest using historical data - more of a hypothetical but it illustrates the point well.
viewtopic.php?p=7286996#p7286996
I look forward to your post.

Is that the correct link? There are no posts by #cruncher in that thread.
dcabler wrote: Sun May 28, 2023 7:40 amRebalancing is very little effort if you're set everything up in a spreadsheet beforehand.
- Takes me less than 2 minutes, once a quarter to update my version of #cruncher's spreadsheet to get the latest durations and yields to plug into my spreadsheet which calculates the withdrawals/rebalance simultaneously. I plan to add more automation such that upon opening the spreadsheet it gives me directly what the transactions need to be by scraping data sources. Till then, I'll sacrifice those couple of minutes. :D
- Then the withdrawal and rebalance also just takes a couple of minutes.
I like python for that sort of thing. It's very good for scraping data sources and manipulating spreadsheets.
dcabler wrote: Sun May 28, 2023 7:40 amI'm with BobK on the "good enough" comment.

cheers.
So long as what you're doing really is good enough. :sharebeer
dcabler
Posts: 4482
Joined: Wed Feb 19, 2014 10:30 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by dcabler »

exodusing wrote: Sun May 28, 2023 8:38 am
dcabler wrote: Sun May 28, 2023 7:40 am
exodusing wrote: Sun May 28, 2023 6:46 am
dcabler wrote: Sat May 27, 2023 5:57 pm What I saw was +/- a few 1's of % when I did this. I don't see it as being that large.
Would you please post your backtest?

Here's a good thread on implementation viewtopic.php?p=3771655#p3771655, including BobK writing "The effort here, however, is not to almost perfectly immunize the portfolio, but rather to provide relatively safe retirement income." It also includes that one should rebalance between funds frequently to limit interest rate risk, which seems a lot of effort, especially compared to the one-time setup of a ladder.

#cruncher's thread on Consistent Yield & Duration to Help Choose TIPS Fund is viewtopic.php?t=104845

There seems to be two uses of "duration" in this threads. One use is a period of time, the other relates to interest rate sensitivity. Some posters echo Bob's notion that anything close is good enough.
I'll dig it up and post it. Might be a while before I can get to it/find it. In the meantime, #cruncher posted an example a while back, but it wasn't a backtest using historical data - more of a hypothetical but it illustrates the point well.
viewtopic.php?p=7286996#p7286996
I look forward to your post.

Is that the correct link? There are no posts by #cruncher in that thread.
dcabler wrote: Sun May 28, 2023 7:40 amRebalancing is very little effort if you're set everything up in a spreadsheet beforehand.
- Takes me less than 2 minutes, once a quarter to update my version of #cruncher's spreadsheet to get the latest durations and yields to plug into my spreadsheet which calculates the withdrawals/rebalance simultaneously. I plan to add more automation such that upon opening the spreadsheet it gives me directly what the transactions need to be by scraping data sources. Till then, I'll sacrifice those couple of minutes. :D
- Then the withdrawal and rebalance also just takes a couple of minutes.
I like python for that sort of thing. It's very good for scraping data sources and manipulating spreadsheets.
dcabler wrote: Sun May 28, 2023 7:40 amI'm with BobK on the "good enough" comment.

cheers.
So long as what you're doing really is good enough. :sharebeer
Link fixed! (I don't think it will necessarily automatically update in your quote of it, though, or in my quote of your quote)

Yep Python is great for that. Another thing my wife is good at that I'm not. I'm using some of the built in capability in Excel for such things..

BTW, from time to time a little bit of maintenance is needed on #cruncher's spreadsheet as bonds mature and new ones are added. That generally also takes a few minutes to clean up.

Some day I also want to be able to grab data directly from the funds themselves. Today both LTPZ and SCHP have detailed daily holdings. But what each reports is a bit different.

Cheers again.
dcabler
Posts: 4482
Joined: Wed Feb 19, 2014 10:30 am

Re: Now that long TIPS again yield more than 1.70% I will…

Post by dcabler »

Kevin M wrote: Sat May 27, 2023 11:09 am
Phyneas wrote: Sat May 27, 2023 6:37 am
dcabler wrote: Sat May 27, 2023 6:14 am
watchnerd wrote: Fri May 26, 2023 10:04 pm
exodusing wrote: Fri May 26, 2023 7:33 pm If I have a TIPS ladder, I know exactly how much I will receive in inflation-adjusted dollars for each year of the ladder.

If I use funds, how do I compute the annual real payout? As an example, assume I want a 30 year level income stream of $100,000/year. How would I do this with existing funds, say LTPZ and the two Vanguard TIPS funds?

For a ladder I could use #cruncher's spreadsheet or tipsadder.com. https://www.tipsladder.com/build?bondCh ... ome=100000 shows the exact bonds to buy and the $2,352,204 cost.
Dying to hear the answer to this because I've never been able to figure it out when it comes to funds.

Seems like, at best, you've got some fuzzy error bars.
In very broad, approximate strokes for duration matching using funds.
30 years of income is roughly 15 years investment horizon
Find current duration of LTPZ and your favorite intermediate bond fund. Let's just say SCHP
Calculate the weight of LTPZ and SCHP that results in a weighted duration that's 15 years. Hold that percentage of each.
Calculate the effective yield of this weighted combination of LTPZ and SCHP using the current real yields of these two funds
Withdrawal % is calculated with Excel function: PMT(Yield, 30, -1, 0, 1)
Withdraw the dollar amount.
Next year you have 29 years remaining, so a 29/2 years investment horizon
Rinse and repeat.
At some point in the future, the required duration needed is less than SCHP's duration. At that point, you will no longer own LTPZ, but instead you'll transition to owning SCHP and, say, STIP
Continue to rinse and repeat until the duration required is less than STIP's duration.
Then you have choices including:
STIP + some nominal investment (CD's, ultrashort bond fund, etc)
Ibonds - this is what BobK likes
Or even just sticking with 100% STIP to the bitter end.

Challenges
- Not every fund reports its real yield. Most report their nominal yields. SEC only recommends that such funds report real yields but they don't require it. But there are only a limited number of TIPS indexes around, so sometimes another fund reports it. For example while SCHP doesn't report real yields, iShares TIP does and they both track the same index. Sadly, there is no equivalent for LTPZ. #cruncher calculates and reports yields for all of the major indexes weekly in this thread: viewtopic.php?p=7241758#p7241758
- Same thing with duration. Most bond funds calculate duration based on nominal yields, not real yields. Nominal-based might be close enough or you can use the data from #cruncher

Yes, there are definitely fuzzy error bars. Based on some backtesting I did with #cruncher's data, it's +/- a few percent but with a relatively flat mean. But this data is very limited. There is enough data out there in the wild to use #cruncher's technique to create a history that goes further back, if anybody wished to pursue it.

There are also some other methods of varying complexity to get close as well.
There's a shortcut method that vineviz proposed here: viewtopic.php?p=6869837#p6869837
I'm not especially keen on this one as I don't think you can cover the longest timeframe you might need. It's main advantage is that you don't need to do any rebalancing between funds.

Another thing I've played with that seemingly does alright. A bit larger error bars, but less maintenance.
Start with LTPZ only. Find real yield for LTPZ. Withdrawal percentage = PMT(yield, 30, -1, 0, 1)
Same thing next year but PMT (yield, 29, -1, 0, 1)
At some point, move everything from LTPZ to SCHP and continue on.
Again at some point, move everything from SCHP to STIP and continue to the end
So instead of continually adjusting the ratios between 2 funds, you're just holding 1 fund at a time and doing the transition at pretty much the same points in time where you would transition in the original duration matching method.

Cheers.
With this scenario, is there a risk of selling LTPZ at an incredibly inopportune time (like 2022-2023) and never making up the price difference as it isn't a rolling ladder, just one used in decumulation. I.e. if the number of years it takes you to move from LTPZ + SCHP to just LTPZ is a period in which LTPZ never recovers its price, even with the increased coupons, could you still lose vs a regular TIPS ladder where the bond is maturing nearest to your funding year and you are guaranteed (absent a default) to get both the principal and coupon?
This is a point I've made several times over the last few days (and a number of times over the years here). I believe the answer to your question is "yes".

The only rebuttal I've heard is that there will be opportune times to offset the inopportune times to sell, and it will balance out over the long term. My rebuttal to that is that you may not be so lucky if there is a sustained period of increasing yields, or even if the yields don't return to the lower levels you bought at before you've sold a lot of your longer term fund.

This assumes a positive yield curve, so that you are earning less on the shorter term TIPS fund shares you buy than you would if you continued to hold the longer term fund shares instead of selling them to buy the shorter term shares.
I think that #cruncher at least touched on this with this post back in 2019: viewtopic.php?p=4770710#p4770710

Cheers
Post Reply