Gyroscopic Investing

Question: Since 1928, is there a way to have a portfolio get at least the nominal CAGR of the vanilla PP, but with less risk? 

Answer: Yes there is...

Stocks 22.64%
T-Bills 25%
T-Bonds .52%
AAA C-Bonds 39.68%
Gold 12.15%

Reub will be happy with this.

What does this mean?

Fun topic.  Where do you get your C-bond data?

Know of any funds that satisfy your bond allocation? VWESX is made up of bonds in the "A" range, but not all "AAA". I'd be nervous about buying individual corporate bonds. Corporations don't last nearly as long as the U.S. government has.

Tyler wrote: Fun topic.  Where do you get your C-bond data?

FRED

Pointedstick wrote: Know of any funds that satisfy your bond allocation? VWESX is made up of bonds in the "A" range, but not all "AAA". I'd be nervous about buying individual corporate bonds. Corporations don't last nearly as long as the U.S. government has.

There's only three companies with AAA left: Exxon Mobil, Microsoft and Johnson & Johnson.  I doubt think just three companies could handle an inflow of capital when Treasuries go wonky.  Default rates 1920-2014:

[align=center][img width=800]http://i60.tinypic.com/33nx6o6.png[/img][/align]

I believe recoveries average 50%.

Not sure about this allocation. As recently as 1980, there were 60+ companies with AAA ratings.

iwealth wrote: Not sure about this allocation. As recently as 1980, there were 60+ companies with AAA ratings.

Seems to me its like the survivorship bias effect going on!

MachineGhost wrote:

iwealth wrote: Not sure about this allocation. As recently as 1980, there were 60+ companies with AAA ratings.

Seems to me its like the survivorship bias effect going on!

Or the ratings agencies have gotten smarter.

***

MG, what's with the .52% in T-bonds?  How could an allocation of that size make a real difference in performance?

MediumTex wrote: MG, what's with the .52% in T-bonds?  How could an allocation of that size make a real difference in performance?

I started with 25% so the optimizer was drawing it down.  Excel is real picky about starting weights.

Moody's tries to include bonds with remaining maturities as close as possible to 30 years. Moody's drops bonds if the remaining life falls below 20 years, if the bond is susceptible to redemption, or if the rating changes.

Perhaps that explains it versus fixed 20yr maturity in T-Bonds pre-1978?

Here's 1928-1977:

Stocks 18.74%
T-Bills 25%
C-Bonds 13.83%
Gold 42.43%

And 1978-2014:

Stocks 28.85%
T-Bills 25%
T-Bonds 24.48%
C-Bond 1.50%
Real Estate 9.19%
Gold 10.89%

I don't think 87 years is enough data....  bull and bear markets easily last decades.

(drumroll)

So what are the returns for comparison?

Tyler wrote: (drumroll)

So what are the returns for comparison?

1978-2014: 8.69% CAGR
1928-1977: 5.13% CAGR
1928-2014: 6.63% CAGR

MachineGhost wrote: Question: Since 1928, is there a way to have a portfolio get at least the nominal CAGR of the vanilla PP, but with less risk? 

Answer: Yes there is...

Stocks 22.64%
T-Bills 25%
T-Bonds .52%
AAA C-Bonds 39.68%
Gold 12.15%

Reub will be happy with this.

What does this mean?

It means you can get tremendous results by data mining.
Of course those results are available only to investors who have access to a working time machine, in which case there are easier ways to make lots of money!

MG,

Great chart on C-Bonds.

Thanks.

My current favorite non-PP portfolio still looks a lot like the PP when you break down the underlying assets.  75% VWINX, 25% Gold.  Maybe consider it a gateway drug for Bogleheads.

[img width=300]http://i61.tinypic.com/oi7l1v.jpg[/img]

Note that having the separate assets in the PP instead of so much in one fund also presents tax optimization options that are very real but harder to quantify.