50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Discussion of the Bond portion of the Permanent Portfolio

Moderator: Global Moderator

Post Reply
MiniB

50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Post by MiniB » Tue Sep 28, 2010 9:38 pm

Harry likes the LT Bond section to have a high maturity/duration so that it responds appropriately to deflation.

The Cash portion is in T-Bills which nearly do not respond at all to interest rates.

I've read Craig state that he likes to use some VG Short Term T-Bond fund for his cash portion to juice his returns a little bit.

I'm wondering if one can use these 2 VG Bond funds in place of actual LT Bonds and T-Bills.

Here's my theory:

The VG LT Bond fund does not have a long enough duration to be appropriate normally.  It will not respond strongly in deflation.  However if the cash portion was in the ST Bond fund, then the duration of this fund is longer than cash.  So in Deflation, your cash portion would be responding a little better, to counteract your Bond portion responding a little worse.

Anyone here that can run a historical returns using these 2 bond funds in place of cash and bonds would be great, thanks!
User avatar
KevinW
Executive Member
Executive Member
Posts: 945
Joined: Sun May 02, 2010 11:01 pm

Re: 50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Post by KevinW » Tue Sep 28, 2010 10:12 pm

Interesting.  Using that combination would certainly simplify investing with Vanguard.

If we compare average durations, we get

SHV: 0.37
TLT: 15.79
mean: 8.08

VFISX: 2.3
VUSTX: 13.5
mean: 7.90

Those two figures are very close, so on that basis it seems like the two combinations would behave very similarly.

I'm curious to hear others' views.
MiniB

Re: 50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Post by MiniB » Tue Sep 28, 2010 10:19 pm

Kevin,

I dont believe the pure duration average is appropriate.  If it were appropriate, why not just use a single intermediate treasury bond fund with an 8 years duration and go 50% Intermediate Bond, 25% Gold 25% Stocks?

Harry Browne was surely smart enough to think of that if it would be a better or equivalent alternative.

I think there's something specifically to holding 25% in high duration bonds, but I'm not sure why.  Perhaps Craig can clarify or point to a radio episode where this is discussed?

And I wonder at what point it is OK to do this?  Surely using a mix of 30 day TBills and 25 year Maturity Bonds is Browne's ideal.  What about 60 day TBills and 24 Year Maturity Bonds?  What about half in 1 year Notes and half 22 Year Bonds?  Extrapolate that down to 100% in a 10 year Treasury Bond with an 8 years duration.  At a certain deviation from the ideal it may lose a lot of steam.

My goal here is to reduce costs as much as possible and using VG Mutual Funds seems like a great solution if it's feasible.
User avatar
MediumTex
Administrator
Administrator
Posts: 9096
Joined: Sun Apr 25, 2010 11:47 pm
Contact:

Re: 50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Post by MediumTex » Wed Sep 29, 2010 8:45 am

In a scenario like the early 1980s where long bond and t-bill rates were rising, the cash portion of the PP did very well while the long bond (and stocks and gold) did very poorly.  If an investor had held 50% intermediate bonds in this scenario, the portfolio would have done very poorly.  In that situation, it was the t-bills that protected the portfolio.

With that said, a PP with 50% 10 year bonds has done well historically.  I like the 25% in cash, however, since it provides a simple pool of funds to draw from when it comes time to spend the money invested in the PP.
Q: “Do you have funny shaped balloons?”
A: “Not unless round is funny.”
MiniB

Re: 50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Post by MiniB » Wed Sep 29, 2010 9:57 am

MediumTex wrote: In a scenario like the early 1980s where long bond and t-bill rates were rising, the cash portion of the PP did very well while the long bond (and stocks and gold) did very poorly.  If an investor had held 50% intermediate bonds in this scenario, the portfolio would have done very poorly.  In that situation, it was the t-bills that protected the portfolio.
My counterargument would be that if we were 25% Semi-Long Bonds (15 year maturity RE: VUSTX) then that position would not have done as poorly as a 25% Super Long Bonds (25+ year maturity RE: TLT).

Thus while the 25% Short Term Treasury Bond position would underperform cash, because the 2 year duration of this fund would cause drops in rising interest environment as well, you wouldn't need as much protection because the drop in the Bond portion would be slighter than otherwise.

In the case of a 25% Short Term Bond, 25% Long-ish (VG) Term Bond mix, I think the results might be the same as 25% Cash and 25% TLT because in the rising interest rate environment, the combination of how the two move should be similar.

Vanguard lets you write checks from the Short Term Bond funds so you can use it like a MMF that isn't guaranteed to hold a $1 level.

I'm not saying this is a good plan, I'm wondering why it isnt a good plan.  Something tells me that Harry would have suggested it because VG funds were around in his lifetime, so it must be bad for some reason, I just cant figure out why.
User avatar
MediumTex
Administrator
Administrator
Posts: 9096
Joined: Sun Apr 25, 2010 11:47 pm
Contact:

Re: 50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Post by MediumTex » Wed Sep 29, 2010 11:30 am

On the bogleheads thread, PP performance using 50% 10 year bonds (which is basically what you are suggesting) was backtested and it was a bit more volatile than the 25% x 4 approach with larger drawdowns in bad years. 

Depending upon what period you looked at, I think its returns may have been slightly higher than the 25% x 4 approach.

There is nothing wrong with this approach, though I prefer the 25% x 4 approach myself.
Q: “Do you have funny shaped balloons?”
A: “Not unless round is funny.”
MiniB

Re: 50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Post by MiniB » Wed Sep 29, 2010 1:14 pm

MT,

I wont refuse the data.  It speaks for itself.  Conceptually I dont understand why it wouldnt work.

Can you explain conceptually why it's different to use 50% 8 year duration bonds then 25% 0 year duration and 25% 16 year duration?

I think the answer is that duration isn't a linear scale, in that as interest rates change, the duration changes as well.  So an 8 years average duration might not be the average of the two once interest rates rise by 2%.
User avatar
MediumTex
Administrator
Administrator
Posts: 9096
Joined: Sun Apr 25, 2010 11:47 pm
Contact:

Re: 50% VG Long Term Treasury Bond Fund plus 50% VG Short Term T- Bond Fund

Post by MediumTex » Wed Sep 29, 2010 1:51 pm

MiniB wrote: MT,

I wont refuse the data.  It speaks for itself.  Conceptually I dont understand why it wouldnt work.

Can you explain conceptually why it's different to use 50% 8 year duration bonds then 25% 0 year duration and 25% 16 year duration?

I think the answer is that duration isn't a linear scale, in that as interest rates change, the duration changes as well.  So an 8 years average duration might not be the average of the two once interest rates rise by 2%.
The PP concept is based upon putting highly volatile non-correlated assets together and then benefiting from the fact that volatile asset prices have more room to move up than down (since no asset can lose more than 100%, but another asset can easily gain more than 100%).  These assets are LT treasurys, stocks and gold.

The PP concept also adds cash as a stabilizing asset so that in times where all three volatile assets are falling there is an asset class that can buoy the portfolio.

What we are discussing here is trading one volatile asset and the cash for a less volatile asset class with a larger allocation (i.e., 50% in 8-10 year treasurys).

The problem with this approach, in my view, is that in a scenario where treasury rates rise quickly (and stocks and gold are falling), there is nothing to protect the portfolio from large losses.

The difference with the 25% x 4 approach and the 50% 8-10 year bond approach is that the 25% x 4 approach has 25% of its assets in an asset that is more or less immune to interest rate fluctuations, while the other approach does not.

The real difference between the two approaches may be in the psychological reaction PP investors are likely to have in a scenario where treasury rates rise rapidly (i.e., lots of fear and a tendency to abandon the strategy if all of the assets are falling in value at the same time).
Q: “Do you have funny shaped balloons?”
A: “Not unless round is funny.”
Post Reply