The Permanent Portfolio May Be About To Break

General Discussion on the Permanent Portfolio Strategy

Moderator: Global Moderator

User avatar
tomfoolery
Associate Member
Associate Member
Posts: 35
Joined: Fri Mar 06, 2020 9:47 pm

The Permanent Portfolio May Be About To Break

Post by tomfoolery » Fri Mar 06, 2020 10:19 pm

I've been a follower of the PP for over 15 years, back when Craig had the 100-page post on Bogleheads. I've listened to all of HB's shows and have read a lot of this early forum's posts.

Today, with long-term 30 year treasuries at 1.25%, I'm sounding the alarm that the system may be able to fail. With an asset, there is an upside and a downside potential.

In the case of the stock market there's a maximum potential downside of 100%, meaning everything could drop to zero. The likelihood of that is virtually zero without a major global meltdown where everything is screwed. With stocks, the potential maximum upside is near infinity. There's no defined limit of how big the stock market could grow over time, not even limited by Earth's resources because we could colonize the galaxy or "colonize" virtual networks.

In the case of gold, the maximum potential downside is 100%, could drop to zero, but that's also virtually impossible. As with stocks, there is infinite potential upside, at least in nominal terms, not "real" (spending potential terms) because an ounce of gold can't be worth "everything" because there's a lot of ounces of gold. The theoretical maximum upside might be estimated as "value of all of the things" divided by "amount of all of the gold". However in practical terms, there's no max upside.

In the case of cash, it's not meant to have an upside, the maximum upside is limited by the interest rate and also potential deflationary effects on the economy and liquidity constraints which would raise the value of the dollar. For example, if we have massive deflation, the dollar purchases more than it did before the deflation (the opposite of inflation). Thus, the max upside might be calculated as maximum potential deflation plus the interest rate. The max downside could be zero, but in the case of the USD, that would only be secondary to a global meltdown of catastrophic proportions that no portfolio could insure against.

Now we get to the bonds portion of the PP. The maximum downside is zero, but also very unlikely because it would mean the collapse of the US Government, and end of the world as we know it. However, the maximum upside is not infinite. The maximum upside is limited to how much lower interest rates can go. At 1.3% as of Friday March 6th, they can't go much lower.

Suppose rates drop down to 0%. I don't have the exact math, but Fidelity is claiming about a 22-year duration on the longest US treasuries, which means a 22% gain for each 1% decrease in interest rates. I understand duration changes as the yield to maturity drops, so it may be an exponential increase, but I'd guess it's no more than a 40% increase to go from 1.3% down to 0%.

So if stocks tank and drop 70%, which is totally possible given a corona virus epidemic, combined with a global deleveraging and a stock market P/E reversion to the mean, then your treasuries won't be able to save you. Because of the limited downside to the treasuries.

The PP "works" when long-term interest rates are at 3%+ because there's reasonable maximum upside potential for these bonds to do what they need to do. The theoretical idea is that maybe the stock market will drop 50% and the long-term treasuries will go up 50% to compensate. But, eventually, interest rates rise and give more "powder" into the "bonds musket".

What we've seen since 2009 is that interest rates never rose. Stock market has gained, but myself and many others see these gains as tenuous. The Federal Reserve has no more power in their primary recession-fighting tool, because they need 500 basis points (5%) to be able to lower interest rates to combat a recession. Thus, if the market does drop, their options will be limited to letting the market crash happen, or printing money to buy equities.

If they print money to buy equities, I expect gold to outperform, because there is defacto inflation and stocks to do well, assuming the Fed is willing to print enough money to lift the market. But, perhaps they print just enough money to maintain the market, and prevent it from crashing, but not to bring any gains, and then stocks may be flat.

I read through an old thread on these forums from 4 years ago where CraigR said bond yields under 1% would break the PP:

viewtopic.php?f=3&t=6675&hilit=maximum+ ... 48#p141214


Yesterday, on Thursday March 5th we were at 1.5%, and today Friday March 6th we are at 1.25%, so it's certainly possible we hit 1% as early as next week.

Craig suggested he might sell all of his bonds and go into cash until interest rates rose again but he wasn't sure how he'd know when to buy back in. Interestingly, this is the opposite of panic selling. In the traditional "panic selling" model, the stock market is crashing, and the average investor sells all of their stocks, with the intent of buying them back in at some unknown point in the future. In many cases, those people miss the rebound and wind up buying back into stocks after they've recovered and they rebuy them higher than they sold.

In the case of Craig suggested selling bonds at the 1% rate and buying back in, perhaps at the 2% rate, it's the opposite of that model, because you're selling bonds at the peak and re-buying them as they are losing money (due to interest rates rising).

However, what if we sell all of our bonds at 1%, wait to buy them at 2%, but in the interim they fluctuate between 1% and 2% for a few years, massively swinging 20% up or down in price (due to a 22-year duration), and otherwise providing protection against the gold and stock portion moves?

I propose a graduated model, where we reduce our bond allocation linearly with interest rates. Such that at 1%, perhaps you hold 5% bonds, down from 25%, because you might still want some, in case they do drop to 0%. Perhaps at 1.25% you hold 10% bonds, at 1.5% you hold the full 25%. And for the bonds you sold, I propose a 50-50 cash/gold split because if bond yields are that low, I expect they are low because the federal reserve is debasing the currency, printing virtual money to add to their balance sheet in order to buy the treasuries.

The problem here, is the PP no longer becomes simple. We need advanced spreadsheets and potentially daily rebalances because in this 1% to 2% range of long-term bond yields, if we subscribed to a linear adjustment model, we'd need constant rebalancing.

Perhaps 1% is the magic number and we simply sell all bonds when they hit 1%, switch to cash or cash/gold mix, and when rates go back up above 1%, we buy back in. Possibly on a daily basis. Because we're always buying low and selling high in this method. It might be like the PP but with an asterisk that says, if Long-Term Treasuries are below 1%, then the asset split is 25% stocks, 25% gold, 50% cash. Assume the bonds are held in tax shelters that do not trigger capital gains or complicated tax filings (multi-page Schedule D for US tax payers).

I haven't participated in the board so I don't know if Craig and MediumTex are gone for good, but I'd certainly like to hear their thoughts on this, if anyone has their contact info. This is the first time in my 15 years of using and studying the PP that it feels like "this time is different".

In other words, with the 30-year under 1%, we enter a state where the asymmetric risk are against us. Reduced maximum upside, but retaining the same maximum downside. It's like playing the lottery where there's a 1 in 1 Million chance to win, regardless of the pot, and the pot is $10M one day and $100M the next day, but the entry fee is still $1.
Last edited by tomfoolery on Fri Mar 06, 2020 10:55 pm, edited 1 time in total.
murphy_p_t
Executive Member
Executive Member
Posts: 995
Joined: Fri Jul 02, 2010 3:44 pm

Re: The Permanent Portfolio May Be About To Break

Post by murphy_p_t » Fri Mar 06, 2020 10:49 pm

I am also giving a little more thought to the portfolio recently. allow me to offer a devil's advocate perspective on your post.

If in fact the bonds are not going to react to the upside sufficiently as rates fall, it seems that you would require a greater percentage of US Treasury bonds, not fewer, to maintain the risk profile of the pp under a more normal interest rate environment. And roll the bonds over more frequently to stay close to the 30-year maturity.

I'm not sure that 1% is anything other than an arbitrary figure. For example, German 30-year bunds have already been trading at negative interest rates.

TLT, over the past 12 months, has returned over 40%. Not too shabby.

Your idea about trading in and out of the bond sure sounds a lot like market timing. I know my experience is less than favorable with timing.
User avatar
tomfoolery
Associate Member
Associate Member
Posts: 35
Joined: Fri Mar 06, 2020 9:47 pm

Re: The Permanent Portfolio May Be About To Break

Post by tomfoolery » Fri Mar 06, 2020 11:10 pm

murphy_p_t wrote:
Fri Mar 06, 2020 10:49 pm
If in fact the bonds are not going to react to the upside sufficiently as rates fall, it seems that you would require a greater percentage of US Treasury bonds, not fewer, to maintain the risk profile of the pp under a more normal interest rate environment. And roll the bonds over more frequently to stay close to the 30-year maturity.

I'm not sure that 1% is anything other than an arbitrary figure. For example, German 30-year bunds have already been trading at negative interest rates.
Thank you for the reply and thoughts. I agree that to protect against a stock market decline, one would need a greater percentage of bonds as the rates drop. However, this injects significantly more risk if interest rates rise.

Further, if interest rates ever rise again, I believe that would put downward pressure on the stock market. In general, rising interest rates are bad for stocks because the companies have to finance expenses at higher rates. I'm concerned that in an increasing interest-rate environment that stocks and bonds will move down in tandem.

If interest rates rise, will gold save us? Gold seems to be a hedge against unexpected inflation. If the inflation is expected, then it's priced into the market price. For example, there's a high probability the Fed prints more money to get out of the next recession. More money printing = inflation, but that inflation is expected because it's expected the Fed will perform QE and thus the probabilistic boost to gold is already priced in. Since it's not 100% guaranteed the Fed will print money, the boost to gold isn't fully priced in, it's probabilistically priced in, such that when the Fed prints money, gold will get a little price jump, but only a relatively small one because the majority of the boost was already priced in before the money printing.

Think about Apple's annual October new iPhone event. The stock doesn't jump immediately after the iPhones are announced because investors already know with 95%+ certainty that new iPhones will be announced and have that priced into the Apple share price. However, if Apple announces some spectacular unexpected new iPhone, then the stock price will jump a bit after the reveal, but not too much because the majority of the "new iPhone economic gain" was already priced in using a discounting formula, with an expected new iPhone release every year forever.

Here's what I suspect will happen. I suspect the Fed will QE infinity until the collapse of the US, which may be in 10 years or 1000 years. Because politicians are under constant pressure for the stock market to perform, and we have set a precedent that they'll bail out the system as needed. Further, pension funds are a huge problem in America and they rely on stock market returns to deliver.

Thus, my crystal ball says: QE infinity, using the newly printed money to buy both stocks and long-term treasuries, which keep long-term rates low forever, which punish savers, artificially prop up pension fund returns, while causing inflation that is underreported by federal CPI numbers. We've already seem Trump do a switcheroo on inflation by changing the measure by which inflation adjusts tax brackets from CPI-U to C-CPI-U which is significantly lower than CPI-U, because in C-CPI-U methodology, if the price of meat gets too expensive, people will just become vegetarians and there's no inflation because food costs are the same in spite of substitution.

Pensions and SS will be adjusted to different inflation metrics which significantly underreport inflation, thus reducing the buying power for recipients. This, I suspect, will boost gold. I'm calling $3k to $5k gold by 2030.

The problem, is that with macro economics, we can predict what will happen, but not when it will happen. So maybe it doesn't happen until 2070 and I'm long dead.

Perhaps staying the course is best, even with LTTs under 1%, but it caused me alarm, and then reading CraigR's post from 2016 about this being a fundamental breakage to the PP is further alarming.
murphy_p_t
Executive Member
Executive Member
Posts: 995
Joined: Fri Jul 02, 2010 3:44 pm

Re: The Permanent Portfolio May Be About To Break

Post by murphy_p_t » Sat Mar 07, 2020 2:19 pm

I suspect QE infinity is correct. I have heard at least one financial commentator predict the same. It makes perfect sense, considering our form of government. Politicians prefer to keep their jobs.

I hope others on the forum will weigh in.
pmward
Executive Member
Executive Member
Posts: 1177
Joined: Thu Jan 24, 2019 4:39 pm

Re: The Permanent Portfolio May Be About To Break

Post by pmward » Sat Mar 07, 2020 2:25 pm

Might I introduce you to Tyler's wonderful article on bond convexity: https://portfoliocharts.com/2019/05/27/ ... convexity/

Combine that with the fact that gold is highly correlated to negative real interest rates, and the fact that lower risk free rates correlate to higher stock multiples, and I don't think you have anything to worry about at all. This could actually be one of the BEST times to hold the PP.
User avatar
LittleDinghy
Associate Member
Associate Member
Posts: 41
Joined: Sun Mar 17, 2019 11:44 am

Re: The Permanent Portfolio May Be About To Break

Post by LittleDinghy » Sat Mar 07, 2020 5:39 pm

pmward wrote:
Sat Mar 07, 2020 2:25 pm
Might I introduce you to Tyler's wonderful article on bond convexity: https://portfoliocharts.com/2019/05/27/ ... convexity/

Combine that with the fact that gold is highly correlated to negative real interest rates, and the fact that lower risk free rates correlate to higher stock multiples, and I don't think you have anything to worry about at all. This could actually be one of the BEST times to hold the PP.
Thank you pmward for linking Tyler's article, and thank you Tyler for writing it! About 10% of my Golden Butterfly cash allocation to Tbills matured this past week (in my Schwab rollover account in picture below). I then held my nose and bought a new tranch of Tbills at 0.3% interest (maturing late Jan 2021). Tyler, your reminder from your article "Portfolios like the Golden Butterfly use a “barbell” approach of two very different bond maturities, in part, to capture the declining rate upside of long term bonds with the rising rate safety of short term bonds" combined with pmward's statement "This could actually be one of the BEST times to hold the PP" give me some assurance that maybe I did the right thing. I really hope my GB is not "about to break" (the title of this thread) because my wife and I will be really relying on it pretty soon. Here's my asset allocation as of the end of Feb (any advice much appreciated!):
Image
Attachments
20200301.PNG
20200301.PNG (51.2 KiB) Viewed 1999 times
User avatar
craigr
Administrator
Administrator
Posts: 2540
Joined: Sun Apr 25, 2010 9:26 pm

Re: The Permanent Portfolio May Be About To Break

Post by craigr » Sun Mar 08, 2020 9:30 pm

pmward wrote:
Sat Mar 07, 2020 2:25 pm
Might I introduce you to Tyler's wonderful article on bond convexity: https://portfoliocharts.com/2019/05/27/ ... convexity/

Combine that with the fact that gold is highly correlated to negative real interest rates, and the fact that lower risk free rates correlate to higher stock multiples, and I don't think you have anything to worry about at all. This could actually be one of the BEST times to hold the PP.
Bond convexity is a topic that eludes even professional financial advisors. A lot of people simply don't understand how powerful falling interest rates on non-callable long bonds can be.

I think Tex had a great way to explain it. He said something to the effect that short-term bond interest rates were like playing tennis. But long term bond interest rates were like playing ping pong. The movements in LT bonds is very fast and powerful and shouldn't be underestimated. Once I understood how it worked, it made total sense why the bond barbell (cash and LT bonds) in the Permanent Portfolio was chosen and I became comfortable holding them.

It also explains why you shouldn't replace LT bonds with callable bonds like munis or mortgages. You don't want the issuer being able to take the bonds away from you at the exact moment you are capturing the gains from falling interest rates. This is another thing I saw with people that were recommending portfolios with munis and mortgages. They didn't understand convexity either and what it would mean at the extremes in a falling interest rate environment like we have today.
User avatar
dualstow
Executive Member
Executive Member
Posts: 10444
Joined: Wed Oct 27, 2010 10:18 am
Contact:

Re: The Permanent Portfolio May Be About To Break

Post by dualstow » Sun Mar 08, 2020 9:38 pm

viewtopic.php?f=1&t=8131&p=140483&hilit ... 9C#p140483
MediumTex wrote:
Tue Feb 02, 2016 5:40 pm
...

In terms of bond pricing, this is known as yield curve "convexity." 

If you use zeroes, all of this same stuff happens, it just happens with more volatility in the value of the underlying bond.  Go compare a chart of TLT and EDV and you will see what I am talking about.
...
I always think of a very low interest environment in the bond market as sort of like a game of tennis turning into a game of ping pong.  Everything gets faster and tighter the smaller the court becomes.
RIP Charlie Daniels
User avatar
craigr
Administrator
Administrator
Posts: 2540
Joined: Sun Apr 25, 2010 9:26 pm

Re: The Permanent Portfolio May Be About To Break

Post by craigr » Sun Mar 08, 2020 9:49 pm

dualstow wrote:
Sun Mar 08, 2020 9:38 pm
viewtopic.php?f=1&t=8131&p=140483&hilit ... 9C#p140483
MediumTex wrote:
Tue Feb 02, 2016 5:40 pm
...

In terms of bond pricing, this is known as yield curve "convexity." 

If you use zeroes, all of this same stuff happens, it just happens with more volatility in the value of the underlying bond.  Go compare a chart of TLT and EDV and you will see what I am talking about.
...
I always think of a very low interest environment in the bond market as sort of like a game of tennis turning into a game of ping pong.  Everything gets faster and tighter the smaller the court becomes.
Yep that's it. That is a great way to explain the concept to people. Falling interest rates leverage the gains greatly for LT bonds, as long as they're not callable.
User avatar
tomfoolery
Associate Member
Associate Member
Posts: 35
Joined: Fri Mar 06, 2020 9:47 pm

Re: The Permanent Portfolio May Be About To Break

Post by tomfoolery » Sun Mar 08, 2020 9:50 pm

craigr wrote:
Sun Mar 08, 2020 9:30 pm
pmward wrote:
Sat Mar 07, 2020 2:25 pm
Might I introduce you to Tyler's wonderful article on bond convexity: https://portfoliocharts.com/2019/05/27/ ... convexity/

Combine that with the fact that gold is highly correlated to negative real interest rates, and the fact that lower risk free rates correlate to higher stock multiples, and I don't think you have anything to worry about at all. This could actually be one of the BEST times to hold the PP.
Hi Craig,

Welcome back! Do you have any new thoughts to your 4 year old statement regarding selling long term treasuries at 1% yield? Have you changed your mind since?
User avatar
craigr
Administrator
Administrator
Posts: 2540
Joined: Sun Apr 25, 2010 9:26 pm

Re: The Permanent Portfolio May Be About To Break

Post by craigr » Sun Mar 08, 2020 9:53 pm

tomfoolery wrote:
Sun Mar 08, 2020 9:50 pm

Hi Craig,

Welcome back! Do you have any new thoughts to your 4 year old statement regarding selling long term treasuries at 1% yield? Have you changed your mind since?
I'm likely to just rebalance according to plan. I don't have any better options or ideas right now. Each person is different and I can't offer specific advice due to varying circumstances. Ask me again when they are 0%. ;)
User avatar
tomfoolery
Associate Member
Associate Member
Posts: 35
Joined: Fri Mar 06, 2020 9:47 pm

Re: The Permanent Portfolio May Be About To Break

Post by tomfoolery » Sun Mar 08, 2020 9:58 pm

craigr wrote:
Sun Mar 08, 2020 9:53 pm
tomfoolery wrote:
Sun Mar 08, 2020 9:50 pm

Hi Craig,

Welcome back! Do you have any new thoughts to your 4 year old statement regarding selling long term treasuries at 1% yield? Have you changed your mind since?
I'm likely to just rebalance according to plan. I don't have any better options or ideas right now. Each person is different and I can't offer specific advice due to varying circumstances. Ask me again when they are 0%. ;)
Futures are showing the yield has broken 1% in overnight trading. Let's schedule us for a chat on Thursday this week when they push through zero ;)
Post Reply