The 20 basis point per year rule-of-thumb

Discussion of the Bond portion of the Permanent Portfolio

Moderator: Global Moderator

Post Reply
Kevin K.
Executive Member
Executive Member
Posts: 516
Joined: Mon Apr 26, 2010 2:37 pm

The 20 basis point per year rule-of-thumb

Post by Kevin K. » Tue Jun 08, 2021 9:21 am

Since there are several folks who post on these forums with a far better understanding of the bond market than I have I'm curious to hear your thoughts on the rule-of-thumb offered by Larry Swedroe that investors should generally look for an additional 20 basis points of yield for each additional year of duration in their bond holdings. I've seen that rule cited and discussed any number of times but don't know its basis (if there is one) - or whether any such notions make sense in our unprecedented negative-real-return Treasury bond world. But if they DID apply it looks like 7 years is the outer limit of the current sweet spot:

https://www.treasury.gov/resource-cente ... data=yield

EDIT: I found the post on Bogleheads where I first read about this rule of thumb. It's from 2013, which feels like about a hundred years ago given what's happened with interest rates since then:

https://www.bogleheads.org/forum/viewtopic.php?t=120617
User avatar
Kriegsspiel
Executive Member
Executive Member
Posts: 4052
Joined: Sun Sep 16, 2012 5:28 pm

Re: The 20 basis point per year rule-of-thumb

Post by Kriegsspiel » Tue Jun 08, 2021 4:11 pm

It sounds like a pretty easy way to stay away from inverted yield curves?

As to whether it's out of date, I've seen real estate investors talking about how the 1% rule (which used to be the 2% rule) has become outdated as well. It's tough to grasp though. Are all the rules still good, and we shouldn't be investing in things that don't meet them, or are the rules actually outdated. CUE KIERKEGAARD.
You there, Ephialtes. May you live forever.
whatchamacallit
Executive Member
Executive Member
Posts: 750
Joined: Mon Oct 01, 2012 7:32 pm

Re: The 20 basis point per year rule-of-thumb

Post by whatchamacallit » Mon May 16, 2022 8:33 pm

Is this something that needs to be applied to the "Permanent Portfolio"?

It would have saved a lot of grief this year over long term treasuries.

I found this other post on bogleheads

https://www.bogleheads.org/forum/viewtopic.php?t=46704

The post mentioned Larry' Swedroe's 2006 book:

The Only Guide to a Winning Bond Strategy You'll ever Need


I expect it has a further explanation.
dockinGA
Executive Member
Executive Member
Posts: 245
Joined: Tue May 04, 2021 9:29 am

Re: The 20 basis point per year rule-of-thumb

Post by dockinGA » Tue May 17, 2022 4:18 am

whatchamacallit wrote:
Mon May 16, 2022 8:33 pm
Is this something that needs to be applied to the "Permanent Portfolio"?

It would have saved a lot of grief this year over long term treasuries.

I found this other post on bogleheads

https://www.bogleheads.org/forum/viewtopic.php?t=46704

The post mentioned Larry' Swedroe's 2006 book:

The Only Guide to a Winning Bond Strategy You'll ever Need


I expect it has a further explanation.
I agree it would've missed a lot of the grief, so far at least. But even in an inverted yield curve situation, doesn't the PP need the extra juice available from extended duration treasuries? Shortening duration over the last 10 years of investing would've led to lower returns overall up until the last 6-12 months (at least that would be my guess, without doing any backtesting). And let's not forget that ITT's have been losing money over the last 6-12 months too.
Kevin K.
Executive Member
Executive Member
Posts: 516
Joined: Mon Apr 26, 2010 2:37 pm

Re: The 20 basis point per year rule-of-thumb

Post by Kevin K. » Tue May 17, 2022 9:17 am

dockinGA wrote:
Tue May 17, 2022 4:18 am
whatchamacallit wrote:
Mon May 16, 2022 8:33 pm
Is this something that needs to be applied to the "Permanent Portfolio"?

It would have saved a lot of grief this year over long term treasuries.

I found this other post on bogleheads

https://www.bogleheads.org/forum/viewtopic.php?t=46704

The post mentioned Larry' Swedroe's 2006 book:

The Only Guide to a Winning Bond Strategy You'll ever Need


I expect it has a further explanation.
I agree it would've missed a lot of the grief, so far at least. But even in an inverted yield curve situation, doesn't the PP need the extra juice available from extended duration treasuries? Shortening duration over the last 10 years of investing would've led to lower returns overall up until the last 6-12 months (at least that would be my guess, without doing any backtesting). And let's not forget that ITT's have been losing money over the last 6-12 months too.
Historically it just hasn't made much difference to use the barbell vs. all (or mostly, with a bit of cash/STT's) ITT's but the ride has been much smoother. Different ball game now it seems to me, given interest rates. Continuing to hold 30 year Treasuries is a massive bet on convexity still working and those bonds saving the day in a market panic. But it's easy to forget that situations like we're in now, where stocks and Treasuries tank in unison (and LTT's do worst of all) are not at all uncommon. It's kind of like folks forget that there was bond market history before 2008.

https://www.portfoliovisualizer.com/bac ... tion5_2=25
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: The 20 basis point per year rule-of-thumb

Post by Kbg » Tue May 17, 2022 2:52 pm

People should never focus on assets in a portfolio...they should focus on the portfolio...so let's do this which explains why it doesn't make much difference.

Duration (making numbers up)

LTTs 20
ITTs 10
Cash 0

25/25 = 50/50 for cash & bonds...20+0/2 = 10...voila...ITTs
whatchamacallit
Executive Member
Executive Member
Posts: 750
Joined: Mon Oct 01, 2012 7:32 pm

Re: The 20 basis point per year rule-of-thumb

Post by whatchamacallit » Tue May 17, 2022 3:26 pm

Initially I was wondering if this rule would save you from buying into long term treasuries when you shouldn't be but looking back at historical charts this doesn't look to be the case.

Some of the best times to have 30 year treasuries actually appear to be when the yield curve is flat with elevated rates just like we have now.
Kevin K.
Executive Member
Executive Member
Posts: 516
Joined: Mon Apr 26, 2010 2:37 pm

Re: The 20 basis point per year rule-of-thumb

Post by Kevin K. » Tue May 17, 2022 3:37 pm

Kbg wrote:
Tue May 17, 2022 2:52 pm
People should never focus on assets in a portfolio...they should focus on the portfolio...so let's do this which explains why it doesn't make much difference.

Duration (making numbers up)

LTTs 20
ITTs 10
Cash 0

25/25 = 50/50 for cash & bonds...20+0/2 = 10...voila...ITTs
Ideal duration for ITT’s is 5 years, not 10.
boglerdude
Executive Member
Executive Member
Posts: 1313
Joined: Wed Aug 10, 2016 1:40 am
Contact:

Re: The 20 basis point per year rule-of-thumb

Post by boglerdude » Tue May 17, 2022 4:23 pm

If you had to sell a huge chunk of your portfolio in 08-09, would you have been better off with LTTs
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: The 20 basis point per year rule-of-thumb

Post by Kbg » Tue May 17, 2022 5:44 pm

boglerdude wrote:
Tue May 17, 2022 4:23 pm
If you had to sell a huge chunk of your portfolio in 08-09, would you have been better off with LTTs
That's a loaded question and most likely would depend on how and when one sold off assets.

But here's an interesting tidbit from PVisualizer using the monthly returns display

There is not a single month from Jan 08 to Dec 09 when an IEF portfolio did not have a larger monthly balance than a 50/50 TLT/CASHX portfolio and in most months an IEF only vs. a TLT only IEF had a larger balance. The only way you would not have done better in IEF is if your timing would have been to the month perfect on when you sold. Additionally, during the time period chosen there is not a single summary metric that IEF did not perform better on.

Thanks for asking this question as it caused me to dive into something I've never looked into or thought about looking into.

In short, for this period of time...the barbell trailed the bullet.

Actually...I did a TON of work analyzing ITTs vs. LTTs late last year and for me this issue is settled. ITTs are a better way to go.
Last edited by Kbg on Thu May 19, 2022 12:09 pm, edited 1 time in total.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: The 20 basis point per year rule-of-thumb

Post by Kbg » Tue May 17, 2022 5:50 pm

Kevin K. wrote:
Tue May 17, 2022 3:37 pm
Kbg wrote:
Tue May 17, 2022 2:52 pm
People should never focus on assets in a portfolio...they should focus on the portfolio...so let's do this which explains why it doesn't make much difference.

Duration (making numbers up)

LTTs 20
ITTs 10
Cash 0

25/25 = 50/50 for cash & bonds...20+0/2 = 10...voila...ITTs
Ideal duration for ITT’s is 5 years, not 10.
Fair enough, let's do some real numbers using VGLT and VGIT duration metrics.

18.6+0/2 = 9.3

VGIT 5.7

IEF 8.5

ITT's are a bit nebulous as you know running anywhere from 3-10 years.
Kevin K.
Executive Member
Executive Member
Posts: 516
Joined: Mon Apr 26, 2010 2:37 pm

Re: The 20 basis point per year rule-of-thumb

Post by Kevin K. » Tue May 17, 2022 5:59 pm

boglerdude wrote:
Tue May 17, 2022 4:23 pm
If you had to sell a huge chunk of your portfolio in 08-09, would you have been better off with LTTs
Kbg’s new post in response to this is excellent. I also suggest reading the Morningstar article I posted the link to in my “Bills, Bonds & Cash” post in this forum. LTT’s did slightly better than ITT’s in ‘08 and worse in every other market crash since 1926.

Regarding LTT’s as reliable portfolio saviors is right up there with thinking that gold’s an inflation hedge in the list of common PP articles of faith unsupported by facts.
User avatar
joypog
Executive Member
Executive Member
Posts: 561
Joined: Wed Apr 27, 2022 7:42 pm

Re: The 20 basis point per year rule-of-thumb

Post by joypog » Thu May 19, 2022 10:22 am

Slightly related to this question. How do you compare yields for bonds of different durations? The bloomberg site seems pretty clear, but it doesn't include all the available durations (such as 3 or 7 year). So just treasury direct for most recent auction results? Or the treasury.gov daily par yield curve?

https://www.bloomberg.com/markets/rates ... t-bonds/us

https://www.treasurydirect.gov/instit/a ... result.htm

https://home.treasury.gov/resource-cent ... nth=202205

(and based on the 20bp rule, it seems that either 2 year or 3 year is the sweet spot?)
1/n weirdo. US-TSM, US-SCV, Intl-SCV, LTT, STT, GLD (+ a little in MF)
dockinGA
Executive Member
Executive Member
Posts: 245
Joined: Tue May 04, 2021 9:29 am

Re: The 20 basis point per year rule-of-thumb

Post by dockinGA » Sat May 21, 2022 6:00 am

Kbg wrote:
Tue May 17, 2022 5:44 pm
boglerdude wrote:
Tue May 17, 2022 4:23 pm
If you had to sell a huge chunk of your portfolio in 08-09, would you have been better off with LTTs
That's a loaded question and most likely would depend on how and when one sold off assets.

But here's an interesting tidbit from PVisualizer using the monthly returns display

There is not a single month from Jan 08 to Dec 09 when an IEF portfolio did not have a larger monthly balance than a 50/50 TLT/CASHX portfolio and in most months an IEF only vs. a TLT only IEF had a larger balance. The only way you would not have done better in IEF is if your timing would have been to the month perfect on when you sold. Additionally, during the time period chosen there is not a single summary metric that IEF did not perform better on.

Thanks for asking this question as it caused me to dive into something I've never looked into or thought about looking into.

In short, for this period of time...the barbell trailed the bullet.

Actually...I did a TON of work analyzing ITTs vs. LTTs late last year and for me this issue is settled. ITTs are a better way to go.
But how big is the difference? And does it overcome the fact that with a barbell approach, you have the added flexibility of using your emergency fund as part of the cash portion of the portfolio, as well as the possibility of adding I Bonds to juice portfolio returns during times like these? Just having a glance again at results in portfoliovisualizer, and on a portfolio level the differences look pretty small
User avatar
dualstow
Executive Member
Executive Member
Posts: 14227
Joined: Wed Oct 27, 2010 10:18 am
Location: synagogue of Satan
Contact:

Re: The 20 basis point per year rule-of-thumb

Post by dualstow » Sat May 21, 2022 9:20 am

dockinGA wrote:
Tue May 17, 2022 4:18 am
whatchamacallit wrote:
Mon May 16, 2022 8:33 pm
Is this something that needs to be applied to the "Permanent Portfolio"?
..https://www.bogleheads.org/forum/viewtopic.php?t=46704
I agree it would've missed a lot of the grief, so far at least. But even in an inverted yield curve situation, doesn't the PP need the extra juice available from extended duration treasuries?
Yeah, it seems like it could be a case of too many strategies.
RIP Marcello Gandini
User avatar
joypog
Executive Member
Executive Member
Posts: 561
Joined: Wed Apr 27, 2022 7:42 pm

Re: The 20 basis point per year rule-of-thumb

Post by joypog » Sat May 21, 2022 9:39 am

dockinGA wrote:
Sat May 21, 2022 6:00 am
on a portfolio level the differences look pretty small
Could this be said about 95% of the discussions on all these forums after someone has set their Equities percentage of their portfolio?

(assuming the remining percentage isn't going to crypto)
1/n weirdo. US-TSM, US-SCV, Intl-SCV, LTT, STT, GLD (+ a little in MF)
whatchamacallit
Executive Member
Executive Member
Posts: 750
Joined: Mon Oct 01, 2012 7:32 pm

Re: The 20 basis point per year rule-of-thumb

Post by whatchamacallit » Sat May 21, 2022 9:51 am

dockinGA wrote:
Sat May 21, 2022 6:00 am

But how big is the difference? And does it overcome the fact that with a barbell approach, you have the added flexibility of using your emergency fund as part of the cash portion of the portfolio, as well as the possibility of adding I Bonds to juice portfolio returns during times like these? Just having a glance again at results in portfoliovisualizer, and on a portfolio level the differences look pretty small
I think their end point was to still do barbell but use intermediate instead of long.
User avatar
joypog
Executive Member
Executive Member
Posts: 561
Joined: Wed Apr 27, 2022 7:42 pm

Re: The 20 basis point per year rule-of-thumb

Post by joypog » Tue May 31, 2022 1:18 am

whatchamacallit wrote:
Mon May 16, 2022 8:33 pm
The post mentioned Larry' Swedroe's 2006 book:

The Only Guide to a Winning Bond Strategy You'll ever Need

I expect it has a further explanation.
I skimmed the book on the Internet Archive Open Library and I could only find a brief single paragraph explanation on page 201....I'd try to summarize it but I'm only learning about bonds and yield curves so I'm hesitant to say much since the concepts are new to me. Best to just go check it out.

Roughly, he suggests 20bps for taxable bonds per year, (15 bps for muni). He also suggests a min weighted-average maturity of 3 years and max of 5 years. But these are very tentative suggestions, much closer to rules of thumb than hard recommendations.

https://archive.org/details/onlyguideto ... ew=theater
1/n weirdo. US-TSM, US-SCV, Intl-SCV, LTT, STT, GLD (+ a little in MF)
User avatar
joypog
Executive Member
Executive Member
Posts: 561
Joined: Wed Apr 27, 2022 7:42 pm

Re: The 20 basis point per year rule-of-thumb

Post by joypog » Tue May 31, 2022 6:58 am

For giggles, I took the US Treasury Yield info and imported the data into google sheets. I then added columns for the deltas between bonds and color coded it. (green for 20bp+, yellow for 10bp, red 0bp or less)

https://docs.google.com/spreadsheets/d/ ... sp=sharing
1/n weirdo. US-TSM, US-SCV, Intl-SCV, LTT, STT, GLD (+ a little in MF)
Post Reply