Pre-1972 PP Performance

[Tortoise]

Although I try not to place too much emphasis on backtesting, I'm curious:  why doesn't anyone seem to discuss the PP's backtested performance beyond 1972--when the U.S. dollar was still partially convertible into gold?  Or, better yet, beyond 1933--when the dollar was still fully convertible into gold?

I realize the gold price was fixed in terms of dollars prior to 1972, but that can't be the entire reason for ignoring pre-1972 PP performance, can it?  After all, the PP is supposed to provide safety in the midst of not only unpredictable market movements due to investors and traders, but also government policies and regulations that can cause turbulence and distortions in the markets.  A gold price that's fixed by law in terms of U.S. dollars is just a specific example of one such distortion-inducing government policy.

And along similar lines, what would PP investors do if the U.S. government were to re-adopt the gold standard someday?  (Highly unlikely in the near term, but it's an eventual possibility if the U.S. dollar crashes and thereby puts hard money advocates on center stage.)  A gold standard would all but eliminate inflation and deflation, since the supply of money could no longer be increased or decreased by fractional-reserve banking or the printing of fiat money.  That would seem to leave prosperity and recession as the remaining two cycles to handle in the portfolio.

[Coffee]

I don't understand why there wouldn't be deflation?  Wasn't that one of the reasons given for why we went off the gold standard, in the 1930's?

[craigr]

Although I try not to place too much emphasis on backtesting, I'm curious:  why doesn't anyone seem to discuss the PP's backtested performance beyond 1972--when the U.S. dollar was still partially convertible into gold?  Or, better yet, beyond 1933--when the dollar was still fully convertible into gold?

There are data sets out there that incorporate LT bonds and stocks. You would substitute the gold for cash as they were linked and you couldn't own gold anyway. Even if you could, the official price was fixed so you couldn't harvest gains from the price float. The numbers overall will be way off because the govt. price fixing of gold understated actual inflation over this time. That's why when gold was released from the dollar in 1971 it immediately jumped from $35 an ounce to over $100 and went up from there rapidly.

I realize the gold price was fixed in terms of dollars prior to 1972, but that can't be the entire reason for ignoring pre-1972 PP performance, can it?  After all, the PP is supposed to provide safety in the midst of not only unpredictable market movements due to investors and traders, but also government policies and regulations that can cause turbulence and distortions in the markets.  A gold price that's fixed by law in terms of U.S. dollars is just a specific example of one such distortion-inducing government policy.

There is data going back to 1927 for stocks and LT bonds but I obtained it from a source who compiled it for his own needs and don't know if I can release it without permission (and I need to verify it myself). The impact of the fixed gold price lowers CAGR because that volatile piece of the portfolio is not allowed to move. The price controls distorts returns over this time. The portfolio under a gold standard would need to be modified. The gold allocation and cash would probably be merged under that situation as they are same. This would probably mean you hold more LT bonds and stocks as well. So instead of 25% in each, it would be more like 33% Cash(gold)/Stocks/Bonds.

You can have deflation/inflation under a gold standard depending on the supply of gold. But this is far less of a problem than paper currencies. IMO. Also there is always a temptation for politicians to print money or use a central bank to achieve spending they couldn't otherwise do. I see almost no possibility of the US going back to a gold standard. I'll worry about it if happens, but other than that I just don't think about it much.

[tgif]

I was attempting to come up with figures comparing performance of the PP from late 1960's to 1980's when both stocks and bonds did poorly when I read this thread.  Craig, would you be able to use the data mentioned in this posting to extend your performance data?

I am particularly interested because of our present environment and the possible vulnerability of the 3 fund BH portfolio to an extended bond and stock portfolio.

[MediumTex]

What would the purpose of this data be?

The PP theory states that the portfolio will do well during periods of inflation, deflation, economic expansion and economic contraction.  In the 1972-2012 period we have seen at least one example of each economic condition, and the portfolio did what it was supposed to do.

To me, any pre-1972 data would not be especially helpful, since gold would be performing a different function and the whole portfolio wouldn't be as well-matched to the different economic environments, especially inflation, since under a gold standard the price of gold is basically fixed and thus can't respond to rising prices the way it can in a floating currency world.

[Gumby]

Holding gold pre-1972 would have been the equivalent of holding cash. So a PP pre-1972 would have basically been 50% cash, and therefore given you a sub-par return.

Some have attempted to replace gold with other commodities pre-1972, but it's a futile analysis. The very fact that gold was illegal to own would have caused interest rates and every asset to behave differently during that time. Chaos Theory (i.e. the butterfly effect) tells us that even the slightest change in a single variable causes an enormous change over time. The fact that gold was fixed and illegal to own is a huge variable change for pre-1972 analysis. So, it's really not very useful to imagine a pre-1972 PP when the PP was clearly created for a free-floating fiat monetary system.

[WildAboutHarry]

So, it's really not very useful to imagine a pre-1972 PP when the PP was clearly created for a free-floating fiat monetary system.

To me, any pre-1972 data would not be especially helpful, since gold would be performing a different function and the whole portfolio wouldn't be as well-matched to the different economic environments, especially inflation, since under a gold standard the price of gold is basically fixed and thus can't respond to rising prices the way it can in a floating currency world.

You can have deflation/inflation under a gold standard depending on the supply of gold. But this is far less of a problem than paper currencies. IMO.

These are all good points.  The HBPP is a "modern" portfolio, and it would not have been easily investable in bygone days (i.e., you could not invest in the S&P 500 prior to the 1970s).

But there were periods of pretty wild inflation/deflation under a gold standard (world wars, etc.).  It might be a useful thought experiment to build an HBPP-type portfolio for pre-1972 conditions.  The stocks, bonds, and cash parts still work, but would would take gold's place?  Silver (I think Clive has used this)?  Farmland?  Timber?  Casino? Stamps? Diamonds? Fine Art?

And laws about what one could invest in have changed over the years.  If gold went away today (repeat the 1933 restrictions on ownership, etc.), what would you replace gold with?

[Gumby]

It might be a useful thought experiment to build an HBPP-type portfolio for pre-1972 conditions.  The stocks, bonds, and cash parts still work, but would would take gold's place?

I don't believe bonds or stocks would still work exactly the same way. A country with a reserve restraint and/or a currency peg would have bonds that behaved differently from a country with a free-floating fiat currency. And that country's stock market would be affected by those different bonds in different ways. I mean, yes, bonds would generally go up when stocks went down, but that country could also default on payments if it had reserve restraints — whereas a fiat country can always make its payments but defaults with inflation.

Anyway, there are other simple portfolios out there that would work better for pre-1972 conditions. For instance, assuming Gold is illegal to hold, I believe Clive once mentioned the "Talmud Portfolio" which was literally recommended in the ancient Talmud text.

Let every man divide his money into three parts, and invest a third in land, a third in business, and a third let him keep in reserve. - The Talmud, Circa 1200 BC – 200 AD

http://www.whatcanidoabout.com/2011/12/ ... portfolio/
http://www.joshuakennon.com/the-talmud- ... portfolio/
http://www.pinnacleadvisory.com/2012/03 ... investing/

[Pointedstick]

That Talmud portfolio is no slouch!



http://www.riskcog.com/portfolio-theme2.jsp#59h09ha9he

[craigr]

And laws about what one could invest in have changed over the years.  If gold went away today (repeat the 1933 restrictions on ownership, etc.), what would you replace gold with?

My view is all portfolios should have exposure to hard assets that you own free and clear along with stocks, bonds and cash. This same topic came up the other day when I was talking to someone about their portfolio. He said he held about 1/3rd in stocks, bonds and real estate (in industrial and residential rental properties that he did not live in). He also, I assumed, held short term reserves in some kind of safe cash. I feel that a portfolio built that way would be extremely robust and incorporates many attributes of the Permanent Portfolio holding gold.

Now with that said, he understood real estate was not liquid, etc. as gold was and he was OK with that. So as long as you understand the limits of real estate as a hard asset I think that is fine.

In this way I differ from Harry Browne in the strict sense. I really want to see a portfolio with stocks, bonds, cash and a hard asset. I think gold works best as a hard asset because it's a monetary asset the central banks themselves own and it's incredibly liquid no matter what is going on in the world. BUT, if someone just couldn't bring themselves to own gold and wanted to hold real estate instead I wouldn't think it was too far off the mark. They just need to understand that real estate may not be as reliable during a monetary crisis, liquidity issues, etc.

So to answer the question, if gold and silver were banned tomorrow I would probably:

1) Hold Stocks
2) Hold Bonds and Cash
3) Try to work around government restrictions locally and own gold overseas anyway.
4) Assuming (3) is not possible:
4a) Own liquid hard assets of another form locally.
4b) Own real estate (not the house you live in, but extra to that purpose).
4c) Even consider having that real estate be overseas sending back rental income to me in a foreign currency.

Reason for 4c is you still get the geographic diversification and the income stream in a foreign currency in case your local currency is having problems. During the conversion you will obtain a lot of value. Plus the fact that the property is overseas still gives you protection against government problems at home.

[Gumby]

Silver has long been considered a poor-man's-gold' and IMO is a reasonable proxy to backtest the PP prior to the 1970's. That's JMHO however.

Even if you used peanuts as a reasonable proxy for gold, it still wouldn't make sense to hold a 4x25 PP-style portfolio since the entire currency — and everything that was denominated in that currency — was already fixed to a hard asset. A PP was clearly designed to be used with a free-floating fiat currency.

[Pointedstick]

If you use LTT in place of TSM as was suggested in one of Gumby's links, the return drops only a little to 9.42%, but the max DD drops A LOT to 3.99%!  

It gets even better if you add gold to the mix:



http://www.riskcog.com/portfolio-theme2 ... c74a74e748

If you substitute stocks for the REIT component you've basically got a PP.

[craigr]

The problem with many REITs is they are really speculating on mortgage bonds and don't own the actual properties. REITs like those from TIAA-CREF actually own the properties in their portfolios and are a better choice (but many can't buy into it). So the behavior of many REITs may not reflect property value as much as mortgage bonds rates at times.

[MediumTex]

Isn't that mix close to PRPFX? Cuggino has only half of the 'equity' portion of the portfolio [15%] in stocks, with the other half [15%] of the 'equities' in REIT and natural resources.

It's REITs and natural resource company stocks, such as oil and gas, gold miners, coal miners and timber companies.

[craigr]

Isn't that mix close to PRPFX? Cuggino has only half of the 'equity' portion of the portfolio [15%] in stocks, with the other half [15%] of the 'equities' in REIT and natural resources.

It's REITs and natural resource company stocks, such as oil and gas, gold miners, coal miners and timber companies.

I think it would be great to have a fund that held stocks, bonds, cash, gold and direct ownership of things like timber tracts, office buildings, oil fields, etc. But the overhead would most likely be onerous.

[Gumby]

Just gotta say that it's pretty incredible that the Talmudic author(s) could have devised a "portfolio" circa 1000 B.C without charts and backtesting software. Kinda makes you wonder how they settled on those recommendations. Maybe the Talmud was the Biblical version of a weekly political/investment podcast scrollcast.

[clacy]

Just gotta say that it's pretty incredible that the Talmudic author(s) could have devised a "portfolio" circa 1000 B.C without charts and backtesting software. Kinda makes you wonder how they settled on those recommendations. Maybe the Talmud was the Biblical version of a weekly political/investment podcast scrollcast.

I think it speaks to the common sense that a well diversified approach like the PP, Ivy Portfolio, PRPFX, etc are.  They (and others) are basically cut from the same cloth.  Also, I think this is different than the stock-heavy, "diversified" portfolios that are common among bogleheads and investment advisers, etc.  I do not consider a 70/30 portfolio that invests 30% into a total bond fund, and 70% into 6 different stock funds is well diversified.

[craigr]

This is an interesting topic that I've thought about in the past. I just put some of my thoughts down in the blog:

https://web.archive.org/web/20160324133 ... re-banned/

Basically, if you couldn't own gold I'd strongly consider real estate in some forms. Particularly if you could own it overseas and have a stream of foreign income coming from it back to you. This is just one option to consider.

[Arturo]

This is an interesting topic that I've thought about in the past. I just put some of my thoughts down in the blog:

https://web.archive.org/web/20160324133 ... re-banned/

Basically, if you couldn't own gold I'd strongly consider real estate in some forms. Particularly if you could own it overseas and have a stream of foreign income coming from it back to you. This is just one option to consider.

Hi Craig,

but to have a real statistic fact on that case (return to gold pattern like before 70s), a person should wait at least some decades, to backtest if the new portfolio (no matter who design it) has a good performance on the four possible states of the economy, right? otherwise, it would be based on hypothesis, and wouldn't follow the principles of security, right?

By the way, the scenario of returning to gold pattern is no that imposible. We just have to look back at the history and what happend in EEUU the last 200 hundred years:


    1785-1861: Gold pattern – 76 years.
    1862-1879: Fiat money – 7 years. Civil War.
    1880-1914: Gold pattern – 34 years.
    1915-1925: Fiat money: 10 years. I World War.
    1926-1931: gold pattern: 5 years.
    1931-1945: Fiat money: 14 years. Great depression.
    1945-1968: Gold pattern: 26 years.
    1971-1971: Fiat money: 5 months.
    1971-1973: dollar pattern: 2 years.
    1973-?: Fiat money: 37 years.

is PP designed to work against political changes? because PP is for the very long run (30 years?) ...

[Gumby]

but to have a real statistic fact on that case (return to gold pattern like before 70s), a person should wait at least some decades, to backtest if the new portfolio (no matter who design it) has a good performance on the four possible states of the economy, right? otherwise, it would be based on hypothesis, and wouldn't follow the principles of security, right?

For what it's worth, Terry Coxon, John Chandler and Harry Browne did do some wonkish backtesting in different scenarios (hyperinflations, depressions, etc) on the Permanent Portfolio concept before they started the Permanent Portfolio fund. They never really went into detail on what kind of backtesting they did, but I know they specifically bought one of the first (expensive) personal computers to crunch the numbers when computers were first coming out to the public. So, it's not like Harry Browne woke up one day and just convinced people it was a good idea without trying to test it. And the concept is somewhat based on a Talmudic portfolio, but with a fiat focus, so I don't think we can say that the general concept hasn't been tested over long periods.

Besides, as craig already pointed out, backtesting doesn't prove anything. backtesting can only disprove strategies.

[craigr]

but to have a real statistic fact on that case (return to gold pattern like before 70s), a person should wait at least some decades, to backtest if the new portfolio (no matter who design it) has a good performance on the four possible states of the economy, right? otherwise, it would be based on hypothesis, and wouldn't follow the principles of security, right?

I would want to see some time to see a backtest yes, but again I mainly use it to disprove ideas.

Here's the bottom line: This isn't a science! Investing is an inherently risky and unknowable business. Backtests are part of a portfolio design, yes. But also I believe there needs to be some analysis and intuition applied for those large areas which you simply won't be able to know about. For instance my intuitive sense based on historical lessons is owning gold in a portfolio is a really good idea. I can't *prove* that will be the case going forward. But looking at human history it tells me that not owning gold is probably the riskier approach ironically enough.

Same thing for other assets like stocks. If someone says they don't want to own stocks, I would simply say that the history of the markets indicate that having ownership in the productive capacity of a growing economy is a good idea. Again, I can't *prove* we won't hit a 40 year depression and make that invalid. I can only use backtests to support he hypothesis and then analysis and intuition to determine it's the right approach.

And again for other aspects of portfolio planning like eliminating credit risk in cash/bonds, etc. I can't prove it will be needed going forward. And indeed a backtest may not even prove it was required. But a look at history outside of the spreadsheet shows it's a really good idea.

I call these things "Extra Spreadsheet Risks". They are the risks/rewards that a spreadsheet will never show you. There needs to be some thought and analysis applied that moves away from that kind of model. It's not perfect, but it's what we've got. Again, this is not a science.

[Arturo]

but to have a real statistic fact on that case (return to gold pattern like before 70s), a person should wait at least some decades, to backtest if the new portfolio (no matter who design it) has a good performance on the four possible states of the economy, right? otherwise, it would be based on hypothesis, and wouldn't follow the principles of security, right?

I call these things "Extra Spreadsheet Risks". They are the risks/rewards that a spreadsheet will never show you. There needs to be some thought and analysis applied that moves away from that kind of model. It's not perfect, but it's what we've got. Again, this is not a science.

This is the point i wanted to get in. For my point of view (personal, no scientific, and from the perspective of a guy with no expertise in economics science), could be catalogued as a high risk investment, if a person trust all his savings on PP during the next 30 years, when there si no 100% scientifc evidence that PP will behave 100% as the system backtested in the past.

By the way, still reading your book (excelent) and trying to find a trustful brokerage in Europe that operates with american short and long bonds :-)

[craigr]

if a person trust all his savings on PP during the next 30 years, when there si no 100% scientifc evidence that PP will behave 100% as the system backtested in the past.

Yes. But then again nobody else can claim these things either. And if they do claim them, they are lying.

Again the best you can do is use backtests to discard models that have failed in the past and apply the portfolio theory in a way that diversifies your risks as best you can. There are no guarantees in investing (or life). The best you can do is just widely diversify in a way that lowers any single risk of something very bad happening to your life savings. For me, that Permanent Portfolio does that. It could fail and I have no way of knowing if, when or how it may happen. But I do think it's still better than other approaches I've used and if something very bad is happening in the world to cause huge losses to the Permanent Portfolio allocation, then I suspect other approaches are doing much worse!

[Pointedstick]

Arturo, there is no and can never be any scientific evidence that a certain investment will perform 100% as backtested in the past. If that's the standard you're looking for, you'll never find it, and anyone who claims they can provide you with it is probably a liar trying to take your money.

Besides, being all-in the PP is the equivalent to being 25% in stocks, 25% in bonds, 25% in cash, and 25% in gold. That's better diversification than nearly every other portfolio out there. A PP is actually very low-risk compared to pretty much anything else you could invest in. That's the whole point. Even a "safe" investment like T-bills or cash under the mattress gives you a 100% guarantee of losing your purchasing power to inflation right now.

[AgAuMoney]

[The PP] could be catalogued as a high risk investment, if a person trust all his savings on PP during the next 30 years, when there si no 100% scientifc evidence that PP will behave 100% as the system backtested in the past.

That's right.

Can you name one thing that isn't?  There isn't anything that cannot lose because everything has been a loser at some point in the past.

Substitute "US currency" instead of PP.  The statement is equally true.  (According to the BLS, U.S. currency has lost more than 50% of its purchasing power in the past 30 years.)

Since the PP has 25% in the U.S. dollar, you can expect that portion will be worth 50% less in 30 years if the next 30 are like the last 30.

Stocks?  See 1929.

Bonds?  The last 30 years bonds have been on an uptrend because of the general trend of declining interest rates.  Rates will go up (assuming the system doesn't collapse) and that will be as bad as declining rates were good.

Gold?  See 1980.

The key is having all four.  And I recommend to monitor it at least annually to rebalance back to equal allocations, but also to see if the concept has been broken.  I prefer you monitor it more often to check and rebalance whenever a +/-5% or +/-10% threshold on any component is breached.