The ONE Thing I Don't Understand With The PP

[TripleB]

I studied the PP for a couple years before jumping in. Everything except ONE thing made sense to me. It's not a deal-breaker at all, just something I don't understand. I've asked the question a couple times and don't recall getting a suitable answer to me. Let me pose it here, now.

Why does it matter whether I have 25% Long Term Bonds and 25% "Cash" in <1year T-Bills, rather than 50% in an intermediate treasury bond fund?

My limited understanding of bonds is that mixing bonds of varying durations will result in some arithmetic mean of the durations to form a composite bond duration.

Thus, by holding 25% Long Term bonds with a duration of around 17 years, and 25% 30-day TBills with a duration close to 0 years, theoretically, you could just put 50% of your money into a bond fund that had a duration of 17/2= 8.5 years, and you would get the same result on any DAILY basis.

My best guess as to why this doesn't work in the PP has to do with momentum. The PP starts with 25% of each asset, but if Long Term bonds do really well, they may make up 33% of the portfolio and not trigger rebalancing. Thus when interest rates go down again, you now have 33% of your assets shooting up, rather than 25%. If you used 50% of an "intermediate" bond fund, then you may lose out on some momentum.

I recall asking this question a while back on Bogleheads, and a few people backtested the PP using 50% intermediate bond fund and the results were nearly identical to a regular PP. I believe CraigR chimed in, but I recall not being satisfied with the answer and it's been burning a hole in the back of my mind ever since.

I've been using the PP in pure form, holding individual long-term bonds for almost 2 years now and have no desire to change. I just like to know the answer to any possible question, in case someone asks me, I don't like to "not know" something, especially something related to an investment strategy that I have 100% of my assets in.

[MediumTex]

If you have 50% in 10 year bonds rather than the bond barbell, in a long term deflationary scenario you quickly begin to "grow out" of your deflation protection and you don't get to enjoy the longer period of higher coupon payments that the barbell approach provides.

In an inflationary scenario, a 50% 10 year bond approach doesn't provide you any dry powder for rebalancing, as you would have wanted to have in the early 1980s.

With all that said, using a 50% 10 year bond approach to the PP would have historically provided similar returns to the 4 x 25% approach (though with a bit more volatility).  This was covered somewhere in the big BH thread.

I am MUCH more comfortable with 25% in fully liquid form at all times.  You never know when you might need to spend some of that money.

[Storm]

I am MUCH more comfortable with 25% in fully liquid form at all times.  You never know when you might need to spend some of that money.

This brings up another great thing about the PP.  The 25% cash can be considered your "emergency fund," so long as it is outside of your retirement accounts.  If you keep 6-12 months of expenses in cash already you'll just have that much more to invest in the PP.

[TripleB]

I am MUCH more comfortable with 25% in fully liquid form at all times.  You never know when you might need to spend some of that money.

This brings up another great thing about the PP.  The 25% cash can be considered your "emergency fund," so long as it is outside of your retirement accounts.  If you keep 6-12 months of expenses in cash already you'll just have that much more to invest in the PP.

I'll write up a blog post on this topic in more depth in the next few weeks. The answer is that you can and should hold stocks in your taxable accounts for the emergency fund, and hold the PP Cash within the retirement account (appropriate as per the 25% x4 framework). Then if you need cash for an emergency, sell the stocks at long-term capital gain rate (or realize a loss), and then within the retirement account, shift the equivalent amount of cash into the same (or similar if realizing a loss, to avoid a wash sale) stocks. Thus you really didn't sell stocks, because your market position was completely neutral. However you were able to "access" your cash tied up within the retirement account, without having to pay annual taxes on the interest the cash spits off.

This method has 2 flaws. (1) if the stock position does really bad (i.e. 50% drop), then you might not have enough in taxable to cover your emergency. Thus I recommend having 2x in taxable stocks as what you would consider necessary for emergencies, since a drop of >50% is very unlikely, and in that event you could always liquidate some of your retirement account if really necessary. (2) if you need to access your emergency fund within the first year of buying the stock fund, and there is a huge gain, you will be hit with a huge short-term cap gain at your marginal income tax rate.

I'll cover this in more detail on my blog, and cross-link here, but this should be enough to get you started.

[HB Reader]

I am MUCH more comfortable with 25% in fully liquid form at all times.  You never know when you might need to spend some of that money.

This brings up another great thing about the PP.  The 25% cash can be considered your "emergency fund," so long as it is outside of your retirement accounts.  If you keep 6-12 months of expenses in cash already you'll just have that much more to invest in the PP.

Yes, this is often overlooked and unappreciated when the PP is compared with other approaches.  The 25% cash "emergency fund" is an intregal part of the strategy.

[Lone Wolf]

With all that said, using a 50% 10 year bond approach to the PP would have historically provided similar returns to the 4 x 25% approach (though with a bit more volatility).  This was covered somewhere in the big BH thread.

I agree with the others that the liquidity is a huge plus.  If an unexpected expense comes up, I would be able to simply take care of it without worrying about tax consequences so long as it didn't exceed 25% of my total savings.  I like the "financial agility" this gives me.

Having said that, I am very happy that the 50% 10-year bond approach also works well.  If the United States stopped offering bonds past a 10-year duration for whatever reason (high interest rates?) then I'd still have the ability to cobble together a Permanent Portfolio that works pretty well.

[Storm]

In Japan's post 1989 (deflationary period) inflation averaged 0.66% whilst a ST/LT barbell averaged 3.97% and a constant rolled 5 year (buy a 5 year and sell each year to buy another 5 year) averaged 3.66%

So LTT/ST barbell provided a (marginally) better reward during deflation, but equally likely provides a poorer reward during rising interest rate periods. Assuming you'd bought 5 year at near par/face value, then rather than rolling after a year, you might opt to hold to maturity to get your money back had prices declined significantly.

LTT's have an inverse correlation to stocks (fixed income (yield) vs variable yield effects), and as such help smooth out the ride. But generally that correlation is a relatively short lived effect. Extreme moves could trigger a PP rebalance event, which might potentially lock in some greater gains for the LT/ST than the mid-duration.

LTT's are great for locking into a fixed, potentially high (as per US 1980's) yield for a relatively long period of time.

I suspect over full cycles its a close call as to which is better overall.

Clive, I know you've done significant research into 50% 5-year treasury ladder vs. LT/ST barbel, but does your research take into account the price appreciation of LT during significant stock corrections, and the potential rebalancing this triggers?

December 2008 was a good example when 4x25 PP's were able to rebalance out of LT treasuries and make a significant profit to offset their stock losses.  With a 5 year term that opportunity would have never existed and you would be stuck with your meager few % yield.  I believe doodle experienced this recently when he decided to tweak the 4x25 PP to include 50% 5-year ladder.

Perhaps you've factored that in, and I am incorrect here.

[moda0306]

I think Clive's playing  "long game" where he assumes a reversion to a mean that has happened historically.

Japan's deflation didn't have enough of the right macroeconomic force to make their PP perform accordingly, but if we were to go through the same thing (looks like we may be), I'd be MUCH more comfortable with 25/25 instead of a 5-year ladder.

In fact, if I was going to tweak my PP to get more return out of it, I'd significantly reduce cash, as it is the "red-headed stepchild" of the PP.