Gold - How Do Investors Model Price?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Tue Jan 14, 2020 1:40 am

Pu239 wrote:
Tue Jan 14, 2020 12:33 am
danielc wrote:
Tue Jan 14, 2020 12:26 am

Let's see if we can convince watchnerd's wife to hide more gold coins in the backyard and we can all go to watchnerd's home and have a metal detector party. :wink:
I'm in if she starts burying antique gold watches though I presume watchnerd would have a greater interest in finding those.
Well, yeah. Those aren't commodities. ;)
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Tue Jan 14, 2020 1:49 am

Pu239 wrote:
Tue Jan 14, 2020 12:54 am
It's much easier to buy some or none and hope, regardless of outcome, that your decision was correct.
This is the "buy and pray" method from the first post.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Tue Jan 14, 2020 1:51 am

anoop wrote:
Tue Jan 14, 2020 12:47 am
The problem with the price of gold is that given that we have ETFs trading in the exchanges the price can easily be manipulated.

You can tell something is not right when every coin dealer in my local area listed on the treasury website as an authorized dealer say that they have none to sell. In some sense, there is no liquidity and we need the fed to dump a few gold coins at each such dealer. :)

I'm not a gold bug, but everything going on in the financial world now is beyond my comprehension.
I don't understand why the ETFs are a problem.

Doesn't that make price discovery easier and the price more transparent?

Maybe the dealers don't like it because they have to compete with it?
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Tue Jan 14, 2020 2:05 am

JP Morgan forecast for gold over next 10 years:

3% per year
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
grayfox
Posts: 5253
Joined: Sat Sep 15, 2007 4:30 am

Re: Gold - How Do Investors Model Price?

Post by grayfox » Tue Jan 14, 2020 9:12 am

RE: Gold as Money.

Gold and Silver were once the exclusive form of Money in the U.S.A. and just about every other country. The United States Constitution gives Congress the power to coin money. See The Constitution of the United States, Article I, Section 8.

"Congress shall have the power...To coin Money, regulate the Value thereof, and of foreign Coin, and fix the Standard of Weights and Measures;"

Now this does not mean that Congress has to stamp out coins on the floor of the Senate. Congress exercises their power to coin money by writing laws which says such and such shall be done, and then the POTUS and his executive branch executes the law.

So in 1792, they exercised that power when they wrote legislature that established the Money system in the U.S. In the Coinage Act of April 2, 1792. The Eagle, Dollar, Dime, Cent, and Mille were defined in terms of a certain weight of Gold, Silver or Copper.

So all the U.S. coins were made of Copper, Silver, or Gold. There was no paper money. Private banks did issue paper banknotes. But these were promises to pay a certain amount of Silver or Gold on Demand, that the bank kept safe in their vaults. On the East Coast people would just trade the banknotes because it was more convenient, and the Gold coins just sat in the bank vault. On the West Coast, people didn't trust the banks as much, so they tended to prefer "hard money", money in specie.

Then Congress established the Federal Reserve System with the Federal Reserve Act of 1913. I don't know all the details, but eventually all the banknotes were from the Federal Reserve, but they were still exchangeable for Gold or Silver in the vault at the Treasury or at any Federal Reserve Bank.

Image
Gold Certificate

When I was a kid, the bills were all Silver Certificates.

At various points in time, they changed the laws. In 1934 they outlawed owning Gold, so a private citizen could no longer hold Gold coins. In 1965, they stopped making coins out of Silver. Up until 1968, you could still exchange a Silver Certificate for Silver coins or Silver bullion. Now the money is just a Federal Reserve Note or coins made from Copper and Nickel. No Silver or Gold in circulating coins.

:idea: The U.S. Constitution gives Congress the power to coin money. I interpret that more broadly to mean that Congress can define what the monetary system is. They have the authority to make it whatever they want. They could make it Gold, Silver and Copper. They could make it Euros or Swiss Franks. Cowry Shells. Bitcoin. Or Federal Reserves Notes that are backed by nothing.

:arrow: So I would say at this point in time, Gold is not used as money. Neither is Bitcoin. Money is only what the laws written by Congress say, according to the U.S. Constitution, which is the supreme law of the land. You may not like the Federal Reserve or Congress, but only the U.S. Congress has the legal power in the U.S. to make the rules regarding what money is.

I guess that settles it... or does it?

There is one interesting clause in the U.S. Constitution. Article I, Section 10.

"No State shall...make any Thing but gold and silver Coin a Tender in Payment of Debts;"

:?: What does that mean? It seems to imply that individual States also have the power to coin money, as long as it's Gold or Silver. E.g. New York State could mint NY Dollars, made out of Silver or a $10 NY Eagle made from Gold.

User avatar
grayfox
Posts: 5253
Joined: Sat Sep 15, 2007 4:30 am

Re: Gold - How Do Investors Model Price?

Post by grayfox » Tue Jan 14, 2020 9:13 am

To summarize:

In the past, Gold and Silver were money. Pretty much everywhere in the world.

In the present time, Gold and Silver are not used as money. Money is banknotes from the Central Bank, created out of thin air and backed by nothing. So-called fiat money. This is the monetary system almost everywhere in the World.
Some people may not like this, but the only way to change it is by an Act of Congress.

The Future? Gold may be used as money again some day.
I can see that happening, at some point.
I think Central Banks are preparing for it. Why do you think they are stockpiling so much Gold?
Maybe we should all be preparing for it.

Here is what you want:
Image
1 oz Gold American Eagles

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Tue Jan 14, 2020 9:29 am

grayfox wrote:
Tue Jan 14, 2020 9:13 am

The Future? Gold may be used as money again some day.
I can see that happening, at some point.
I think Central Banks are preparing for it. Why do you think they are stockpiling so much Gold?
Maybe we should all be preparing for it.

Here is what you want:
Image
1 oz Gold American Eagles
For the sake of entertainment, let's hypothesize fiat money goes away and there is a return to the gold standard:

--Does having gold coins help if the government fixes the price of gold by law?
--What if the price of gold fixed by law is below the current gold market price? Is holding gold good in that situation?
--What if part of the return to the gold standard also makes private ownership of gold illegal again?

As you said, Congress can define money as whatever they want, and that definition needn't be in favor current gold owners, whom I believe are not a majority of the population.

It's a fun drinking game to play, but I don't think it's a useful framing for creating AA as you can't even do any rational probabilistic analysis of likely outcomes.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

beehivehave
Posts: 208
Joined: Thu Aug 08, 2019 1:21 pm

Re: Gold - How Do Investors Model Price?

Post by beehivehave » Tue Jan 14, 2020 12:04 pm

grayfox wrote:
Tue Jan 14, 2020 9:13 am
To summarize:

In the past, Gold and Silver were money. Pretty much everywhere in the world.

In the present time, Gold and Silver are not used as money. Money is banknotes from the Central Bank, created out of thin air and backed by nothing. So-called fiat money. This is the monetary system almost everywhere in the World.
Some people may not like this, but the only way to change it is by an Act of Congress.

The Future? Gold may be used as money again some day.
I can see that happening, at some point.
I think Central Banks are preparing for it. Why do you think they are stockpiling so much Gold?
Maybe we should all be preparing for it.

Here is what you want:
Image
1 oz Gold American Eagles
Return to the gold standard is not at all likely, since all governments and mainstream economists, who control policy, are against it. There is not a single country in the entire world on it.
But if you like the euro, which is controversial, you should love gold.
It is interesting that gold bugs refer to money as "fiat currency". As if calling it a derogatory name somehow denigrates it. What could be more "fiat" (i.e., arbitrary) than accepting a gold metal established as a standard thousands of years ago? Why gold, chosen for its physical portability, in an age when most transactions are electronic?
Gold will also be fine come the apocalypse for which the gold bugs are preparing. Who wouldn't want a metal whose primary use is making jewelry when people are fighting over food and water?
Americans should be particularly resistant to the idea of a new standard to replace the current ad hoc standard - the US dollar..
And if you choose to invest in commodities, the rule is the same as for all investments - diversify.

Pu239
Posts: 186
Joined: Mon Dec 17, 2018 6:24 pm

Re: Gold - How Do Investors Model Price?

Post by Pu239 » Tue Jan 14, 2020 12:18 pm

watchnerd wrote:
Tue Jan 14, 2020 2:05 am
JP Morgan forecast for gold over next 10 years:

3% per year
Well, there's your modeling result. Time to buy some gold!
Between the idea And the reality...Between the motion And the act...Falls the Shadow - T. S. Eliot

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Tue Jan 14, 2020 12:20 pm

Pu239 wrote:
Tue Jan 14, 2020 12:18 pm

Well, there's your modeling result. Time to buy some gold!
Yeah, all the volatility of a stock without the return!

(based on that projection)

Sounds awesome.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Tue Jan 14, 2020 12:41 pm

beehivehave wrote:
Tue Jan 14, 2020 12:04 pm
[
Return to the gold standard is not at all likely, since all governments and mainstream economists, who control policy, are against it. There is not a single country in the entire world on it.
But if you like the euro, which is controversial, you should love gold.
It is interesting that gold bugs refer to money as "fiat currency". As if calling it a derogatory name somehow denigrates it. What could be more "fiat" (i.e., arbitrary) than accepting a gold metal established as a standard thousands of years ago? Why gold, chosen for its physical portability, in an age when most transactions are electronic?
Gold will also be fine come the apocalypse for which the gold bugs are preparing. Who wouldn't want a metal whose primary use is making jewelry when people are fighting over food and water?
Americans should be particularly resistant to the idea of a new standard to replace the current ad hoc standard - the US dollar..
And if you choose to invest in commodities, the rule is the same as for all investments - diversify.
I find it amusing that the gold bugs often make cases for gold ownership based upon scenarios involving massive upheavals of economic systems or civilizational collapse. There is something going on with that thinking beyond pure economics.

Whereas the World Gold Council, the guys funded by the gold miners to market gold, really don't. They emphasize a nearly purely speculative momentum argument:

"There are millions of people in India and China who love to buy gold and are buying at an unprecedented rate as they continue to get richer. Don't you want to ride that train up?"

The World Gold Council must have concluded that doom & gloom doesn't sell.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

sean.mcgrath
Posts: 165
Joined: Thu Dec 29, 2016 6:15 am
Location: US in NL

Re: Gold - How Do Investors Model Price?

Post by sean.mcgrath » Tue Jan 14, 2020 4:11 pm

watchnerd wrote:
Sat Jan 11, 2020 10:18 am
grayfox wrote:
Sat Jan 11, 2020 10:01 am
Consider this: Back about 1915, Henry Ford raised the wages of auto workers to $5 per day.
$5 is a half Eagle, which has 0.2419 troy ounce of Gold.
At today's price $1562, that is $377.85 worth of Gold.

50 weeks per year x 5 days per week = 250 work days per year
250 work days x $377.85 = $94,462.50 per year wages for 1915 auto worker in Jan 2020 dollars.
You'd be on the losing end of the bargain:

According to data on autoworkers union, "in the reduction of the average hourly labor cost from about $73 to about $69 per hour"....in 2007.

https://work.chron.com/average-pay-auto ... 24071.html

That's $138,000.00 annually, that's $40,000 more than the 1915 gold wage adjusted for time.
Late reply, but labor costs are not the same as salary. As your link states, it includes benefits (including SS, I would expect). The salaries are "Roughly speaking, the average hourly pay for a member of the United Auto Workers currently ranges from $28 to $38 or so for those hired before September 2007, and between $16 and $20 for workers hired afterward." (from your link)

We can discuss whether benefits should be included as wages in the comparison, but they certainly would not be paid as much as the half Eagle (even pre-tax).

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Tue Jan 14, 2020 4:19 pm

sean.mcgrath wrote:
Tue Jan 14, 2020 4:11 pm
watchnerd wrote:
Sat Jan 11, 2020 10:18 am
grayfox wrote:
Sat Jan 11, 2020 10:01 am
Consider this: Back about 1915, Henry Ford raised the wages of auto workers to $5 per day.
$5 is a half Eagle, which has 0.2419 troy ounce of Gold.
At today's price $1562, that is $377.85 worth of Gold.

50 weeks per year x 5 days per week = 250 work days per year
250 work days x $377.85 = $94,462.50 per year wages for 1915 auto worker in Jan 2020 dollars.
You'd be on the losing end of the bargain:

According to data on autoworkers union, "in the reduction of the average hourly labor cost from about $73 to about $69 per hour"....in 2007.

https://work.chron.com/average-pay-auto ... 24071.html

That's $138,000.00 annually, that's $40,000 more than the 1915 gold wage adjusted for time.
Late reply, but labor costs are not the same as salary. As your link states, it includes benefits (including SS, I would expect). The salaries are "Roughly speaking, the average hourly pay for a member of the United Auto Workers currently ranges from $28 to $38 or so for those hired before September 2007, and between $16 and $20 for workers hired afterward." (from your link)

We can discuss whether benefits should be included as wages in the comparison, but they certainly would not be paid as much as the half Eagle (even pre-tax).
Fair.

But to flip it around, I'm pretty darn sure Ford workers in 1914 had little to zero benefits beyond wages. If someone wants to bother to look up the total benefits package of a 1914 Ford worker it might be interesting to know, but I'm sure it's "less".

So the total compensation comparison is still valid.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

MarginalUtility
Posts: 6
Joined: Wed Aug 21, 2019 3:39 pm

Re: Gold - How Do Investors Model Price?

Post by MarginalUtility » Tue Jan 14, 2020 5:10 pm

"So all the U.S. coins were made of Copper, Silver, or Gold. There was no paper money. Private banks did issue paper banknotes. But these were promises to pay a certain amount of Silver or Gold on Demand, that the bank kept safe in their vaults. On the East Coast people would just trade the banknotes because it was more convenient, and the Gold coins just sat in the bank vault. On the West Coast, people didn't trust the banks as much, so they tended to prefer "hard money", money in specie."

There was paper money. When private banks issued notes before the Civil War, the notes could function as money because they were a medium of exchange and a store of value, and were denominated in the standard unit of account (dollars). But these banknotes were not equivalent to gold. Payment in gold depended on the banks' willingness and ability to pay, and the noteholders' willingness and ability to travel to the banks to present the banknotes for payment. Private banks frequently failed before the Civil War.

During the Civil War, the federal government issued United States Notes (greenbacks) that were not redeemable for gold or silver. The government eventually authorized redemption, but not until 1878.

User avatar
grayfox
Posts: 5253
Joined: Sat Sep 15, 2007 4:30 am

Re: Gold - How Do Investors Model Price?

Post by grayfox » Wed Jan 15, 2020 8:07 am

watchnerd wrote:
Tue Jan 14, 2020 9:29 am

For the sake of entertainment, let's hypothesize fiat money goes away and there is a return to the gold standard:

--Does having gold coins help if the government fixes the price of gold by law?
--What if the price of gold fixed by law is below the current gold market price? Is holding gold good in that situation?
--What if part of the return to the gold standard also makes private ownership of gold illegal again?

As you said, Congress can define money as whatever they want, and that definition needn't be in favor current gold owners, whom I believe are not a majority of the population.
As long as you have the gold coins in your possession and you can sell them at a fair market price, you will get the fair value.
If someone robbed you or the government confiscated your Gold, then you won't have it to sell.
If it was illegal to buy and sell Gold, but you still had possession, you would have to sell it on the black market. A Goldfinger character would probably be buying up all the Gold on the black market.

Would it matter how many of the new gold-backed dollars are set to how much Gold? No matter what price they set, the price of everything else will adjust.
E.g Let's say a Toyota Yaris costs $15,000 today. Recent price of Gold has been about $1,500 per ounce. That's about 10 ounces of Gold for one Yaris.
If they set $2,000 = 1 oz Gold, a Yaris would be about $20,000.
If they set $20 = 1 oz Gold, a Yaris would be about $200.

:?: The United States has 8,133.5 tonnes of Gold. Why does the U.S. need that much Gold when we are not on a Gold Standard?
It seem like an insurance policy to me. The U.S. only went off the Gold Standard in 1971. That's less than 50 years ago. It was in my lifetime. When it happened it was a shock to the world.

:idea: Money backed by nothing is an experiment. In many countries, it has led to hyperinflation. In the U.S. it's already led to two asset bubbles and a financial crisis. No one knows how the experiment will turn out. But right now with interest rates so low, the Fed is just about out of ammunition to fight inflation. [ :oops: Correction: not inflation, recession. No room to lower rates if recession. To fight inflation, they would raise interest rates.] The people in charge are hedging their bet by holding 8,000 tons of Gold. If it becomes necessary, they can quickly replace the current money system with a Gold-backed system.

When Bernanle was asked at a Congressional hearing why the U.S. held so much Gold, he channeled Tevye and answered "Tradition!"
If anyone has a better explanation why the U.S. holds 8,000 tons of Gold, let's hear it.

watchnerd wrote:
Tue Jan 14, 2020 9:29 am

It's a fun drinking game to play, but I don't think it's a useful framing for creating AA as you can't even do any rational probabilistic analysis of likely outcomes.
As far as probabilities go, what's the probability your house will burn down? If you knew something was going to happen for certain, you would not be able to buy insurance for it. No one would sell it to you. Insurance is for low probability events that have high damage. The best time to buy insurance is when nobody expects the calamity too happen.

Top 10 countries with the largest gold reserves
Last edited by grayfox on Wed Jan 15, 2020 9:46 am, edited 1 time in total.

MarginalUtility
Posts: 6
Joined: Wed Aug 21, 2019 3:39 pm

Re: Gold - How Do Investors Model Price?

Post by MarginalUtility » Wed Jan 15, 2020 9:08 am

grayfox wrote:
Wed Jan 15, 2020 8:07 am

:idea: Money backed by nothing is an experiment. In many countries, it has led to hyperinflation. In the U.S. it's already led to two asset bubbles and a financial crisis. No one knows how the experiment will turn out. But right now with interest rates so low, the Fed is just about out of ammunition to fight inflation. The people in charge are hedging their bet by holding 8,000 tons of Gold. If it becomes necessary, they can quickly replace the current money system with a Gold-backed system.
The worst financial crisis in U.S. history, the Great Depression, began while the U.S. adhered to the gold standard. In Lords of Finance, Liaquat Ahamed attributes the 1920s asset bubble in part to efforts to maintain the gold standard in the U.S. and Great Britain. The Great Depression was preceded by severe financial crises beginning in 1837, 1873, 1893, and 1907.

Today's low interest rates do not suggest the Fed is just about out of ammunition to fight inflation. The Fed can fight inflation by raising interest rates, as it did in the early 1980s. Instead, low interest rates arguably indicate the Fed may have difficulty fighting deflation.

DB2
Posts: 496
Joined: Thu Jan 17, 2019 10:07 pm

Re: Gold - How Do Investors Model Price?

Post by DB2 » Wed Jan 15, 2020 9:23 am

MarginalUtility wrote:
Wed Jan 15, 2020 9:08 am
The Fed can fight inflation by raising interest rates, as it did in the early 1980s.
That is true. Considering how we have far more debt today compared to the early 1980s it's a scary thought although I assume they would try to essentially inflate it away.

User avatar
grayfox
Posts: 5253
Joined: Sat Sep 15, 2007 4:30 am

Re: Gold - How Do Investors Model Price?

Post by grayfox » Wed Jan 15, 2020 9:40 am

MarginalUtility wrote:
Wed Jan 15, 2020 9:08 am

Today's low interest rates do not suggest the Fed is just about out of ammunition to fight inflation. The Fed can fight inflation by raising interest rates, as it did in the early 1980s. Instead, low interest rates arguably indicate the Fed may have difficulty fighting deflation.
Right. I got that backwards. :oops: They would want to lower rates in case of recession.

User avatar
TheTimeLord
Posts: 6372
Joined: Fri Jul 26, 2013 2:05 pm

Re: Gold - How Do Investors Model Price?

Post by TheTimeLord » Wed Jan 15, 2020 9:52 am

watchnerd wrote:
Sun Jan 05, 2020 11:58 am
How do people even create SWAG models for gold when it's almost completely outside the normal financial system, and often responds strongly to emotional events? Or without a model, how do people know how much gold is buy?

Or is it all just 'buy and pray you're lucky'?
I would say Gold is a luxury portfolio allocation for those who already have more than enough and want to protect their finances against certain unlikely events rather than look for additional growth. I would assume you would decide to purchase it monthly in a set amount (dollars or ounces) in your preferred form (ETF or physical) then set it aside hoping it never becomes truly valuable since that would likely mean the devastation of the remainder of your holdings.

But what do I know, I know nothing, nothing.
IMHO, Investing should be about living the life you want, not avoiding the life you fear. | Run, You Clever Boy! [9085]

beehivehave
Posts: 208
Joined: Thu Aug 08, 2019 1:21 pm

Re: Gold - How Do Investors Model Price?

Post by beehivehave » Wed Jan 15, 2020 10:33 am

grayfox wrote:
Wed Jan 15, 2020 8:07 am
watchnerd wrote:
Tue Jan 14, 2020 9:29 am

For the sake of entertainment, let's hypothesize fiat money goes away and there is a return to the gold standard:

--Does having gold coins help if the government fixes the price of gold by law?
--What if the price of gold fixed by law is below the current gold market price? Is holding gold good in that situation?
--What if part of the return to the gold standard also makes private ownership of gold illegal again?

As you said, Congress can define money as whatever they want, and that definition needn't be in favor current gold owners, whom I believe are not a majority of the population.
As long as you have the gold coins in your possession and you can sell them at a fair market price, you will get the fair value.
If someone robbed you or the government confiscated your Gold, then you won't have it to sell.
If it was illegal to buy and sell Gold, but you still had possession, you would have to sell it on the black market. A Goldfinger character would probably be buying up all the Gold on the black market.

Would it matter how many of the new gold-backed dollars are set to how much Gold? No matter what price they set, the price of everything else will adjust.
E.g Let's say a Toyota Yaris costs $15,000 today. Recent price of Gold has been about $1,500 per ounce. That's about 10 ounces of Gold for one Yaris.
If they set $2,000 = 1 oz Gold, a Yaris would be about $20,000.
If they set $20 = 1 oz Gold, a Yaris would be about $200.

:?: The United States has 8,133.5 tonnes of Gold. Why does the U.S. need that much Gold when we are not on a Gold Standard?
It seem like an insurance policy to me. The U.S. only went off the Gold Standard in 1971. That's less than 50 years ago. It was in my lifetime. When it happened it was a shock to the world.

:idea: Money backed by nothing is an experiment. In many countries, it has led to hyperinflation. In the U.S. it's already led to two asset bubbles and a financial crisis. No one knows how the experiment will turn out. But right now with interest rates so low, the Fed is just about out of ammunition to fight inflation. [ :oops: Correction: not inflation, recession. No room to lower rates if recession. To fight inflation, they would raise interest rates.] The people in charge are hedging their bet by holding 8,000 tons of Gold. If it becomes necessary, they can quickly replace the current money system with a Gold-backed system.

When Bernanle was asked at a Congressional hearing why the U.S. held so much Gold, he channeled Tevye and answered "Tradition!"
If anyone has a better explanation why the U.S. holds 8,000 tons of Gold, let's hear it.

watchnerd wrote:
Tue Jan 14, 2020 9:29 am

It's a fun drinking game to play, but I don't think it's a useful framing for creating AA as you can't even do any rational probabilistic analysis of likely outcomes.
As far as probabilities go, what's the probability your house will burn down? If you knew something was going to happen for certain, you would not be able to buy insurance for it. No one would sell it to you. Insurance is for low probability events that have high damage. The best time to buy insurance is when nobody expects the calamity too happen.

Top 10 countries with the largest gold reserves
"Money backed by nothing is an experiment."
The US dollar is not backed by nothing. It is backed by the full faith and credit of the US government. Gold, on the other hand, is backed by faith, in an arbitrarily chosen metal (aka as superstition).

"Nobody knows how the experiment will turn out."
We know how gold turned out. Contributed to the Great Depression and unprecedented turmoil in the economy in the 19th century.

"But right now with interest rates so low, the Fed is just about out of ammunition to fight inflation."
If we were on the gold standard, the Fed would have absolutely zero tools to fight inflation or deflation. That is the main problem with the euro - individual governments have no control over their own currency. And why should the US give up having control of the only real international monetary standard - the dollar?

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 10:35 am

grayfox wrote:
Wed Jan 15, 2020 8:07 am

As long as you have the gold coins in your possession and you can sell them at a fair market price, you will get the fair value.
This isn't true if a country is using the gold standard.

Prior to the abandonment of the gold standard, there was no fair market value of gold. The price was fixed by the government.

If the government says that 1 oz gold = $500, and you can trade dollars for specie, why would anyone give you more?
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 10:45 am

grayfox wrote:
Wed Jan 15, 2020 8:07 am


As far as probabilities go, what's the probability your house will burn down? If you knew something was going to happen for certain, you would not be able to buy insurance for it. No one would sell it to you. Insurance is for low probability events that have high damage.
I don't know this, but insurance actuaries make a living calculating this stuff and pricing insurance accordingly.

I know of no actuaries calculating odds on the dissolution of fiat money and selling policies around it.

As for insurance for low probability events -- there comes a point where the probability is so low that buying insurance is just a waste of money.

But this is a key pivot:

Are we now saying gold is not an investment, but insurance, i.e it's a cost?

In other words, buying gold is like buying a short on fiat money?
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

gougou
Posts: 245
Joined: Thu Sep 28, 2017 7:42 pm
Location: San Francisco Bay Area

Re: Gold - How Do Investors Model Price?

Post by gougou » Wed Jan 15, 2020 2:55 pm

danielc wrote:
Mon Jan 13, 2020 8:06 pm
gougou wrote:
Mon Jan 13, 2020 12:47 pm
You obviously don’t know how the stdev of a portfolio is calculated. Let me give you a simple example. If a portfolio returns exactly 10% every year, then its annual returns are 10% with stdev of 0. Those are the numbers you’ll get on the portfoliovisualizer website.
That would be the standard deviation of the return series. That is what I did in my post, after subtracting inflation. And if you want to know whether an asset has a stable value relative to a basket of commodities, you need to compute the standard deviation of the values of that asset in terms of the value of that basket of commodities.
gougou wrote:
Mon Jan 13, 2020 12:47 pm
You suggested calculating the stdev of the total portfolio value series ($1, $1.1, $1.21 ...).
That is not what I said. I said that if you want to know whether 10-year treasuries or gold has a more stable value relative to a basket of commodities, you need to take the standard deviation of the ratio of the two prices. What you wrote is very very different from what I said. I don't know how you could have misunderstood my post, since I even told you which column of data from your own post I was referring to.

gougou wrote:
Mon Jan 13, 2020 12:47 pm
The stdev of this series is positive and increasing with every new entry.
Read that sentence again. I don't think you meant to say what you just wrote. A standard deviation is always, by definition, a positive quantity, and it is not "increasing" since it is a single value for a data series.
gougou wrote:
Mon Jan 13, 2020 12:47 pm
Now let’s go back to my analysis. I calculated how many Commodities each portfolio can buy year over year. Then I calculated how many % extra Commodities each portfolio can buy year over year (the returns of each portfolio in terms of Commodities). You should clearly see which returns are less volatile here. I calculated stdev on these two series and compared them.
1) The size of the basket of commodities that you can buy is more variable for gold.

2) The rate of return relative to the basket of commodities is more variable for 10-year treasuries.

You computed (2) and I computed (1).
gougou wrote:
Mon Jan 13, 2020 12:47 pm
CPI is the return of holding a bunch of stuff vs USD as defined by the US govt. CPI has a strong US bias and its underlying assets can change year over year.
Are you now going to argue that the basket of commodities from Portfolio Visualizer is a better measure of stuff people buy than the Consumer Price Index? Seriously? Also, CPI is not "defined" by the US gov't in any meaningful way, it is calculated by the US gov't. You make it sound like the US govt just picked whatever the heck it wanted, presumably for some nefarious purpose. CPI is calculated by the BLS; it is a survey of the price of goods that people actually buy.
gougou wrote:
Mon Jan 13, 2020 12:47 pm
I also don’t see where you incorporated CPI in your stdev analysis.
I told you. I subtracted the rate of inflation reported in Portfolio Visualizer. That's the easiest way to convert the nominal returns of 10 year treasuries and gold into "real" returns (i.e. relative to the basket of goods people buy).

gougou wrote:
Mon Jan 13, 2020 12:47 pm
Finally, I said gold had good long term returns and probably beaten bonds, and I was mostly just taking about the returns. You wanted to calculate the whole stdev and Sharpe ratio stuff. Those are difficult to do right.
I also was the one that looked up the returns. You said that gold had probably beaten bonds, but I'm the one who looked it up. The result was that gold had lower return than 10 year treasuries despite having a much higher stdev and much lower Sharpe ratio. You have alternated between arguing that volatility doesn't matter when looking at returns, and arguing that the volatility is wrong and somehow it must be the US dollar that's jumping up and down and making gold look bad. I used inflation data to show the volatility of USD (3.06%), and the volatility real returns for treasuries (IT = 7.82%; LT = 10.82%) and gold (29.20%).

gougou wrote:
Mon Jan 13, 2020 12:47 pm
Returns are indisputable while the stdev can change a lot depending on what base currency/asset you are using. And I just used Commodities to prove that you should not use USD as base to compare Gold vs US Treasury.
You used commodities to prove that gold returns are less volatile relative to a basket that contains gold, at least if you also chop off 75% of the history of gold to pick an atypical sample of 12 years. I used inflation data to prove that gold returns are more volatile relative to a basket of stuff that people actually buy, using the entire history of gold since the end of the gold standard. I cannot fathom why you keep insisting on pinning the blame on USD for the crappy behaviour of gold.
Gold returned more money since 1972 to 2019 with 7.53% CAGR vs 10-year Treasury's 7.09% CAGR. Gold is the world's ultimate safe haven asset but you just repeatedly tried to prove gold is risky using your flawed methods.

You only proved gold is volatile against USD, or the prices of consumer goods and services within the US. All of those are still highly correlated with USD itself. When I removed the USD correlation and looked at some truly international assets such as a diversified basket of commodities suddenly the USD/US Treasury is more volatile.

Gold being independent from all countries in the world and being used as THE safe haven asset for thousands of years is a good enough proof that gold is extremely safe. The 10 year Treasury that couldn't even beat holding a piece of metal and do nothing should tell you how lousy that investment is.

With US being the most powerful country winning the cold war and having the largest economy in the past 50 years or so, its paper currencies still did terribly against gold. There's no telling how US will fair in the future. I'll sleep much better owning gold for the risk-averse part of my portfolio than holding some papers that promise to pay more paper.

Smith1776
Posts: 62
Joined: Fri Apr 21, 2017 5:37 pm

Re: Gold - How Do Investors Model Price?

Post by Smith1776 » Wed Jan 15, 2020 3:01 pm

gougou wrote:
Wed Jan 15, 2020 2:55 pm

Gold returned more money since 1972 to 2019 with 7.53% CAGR vs 10-year Treasury's 7.09% CAGR. Gold is the world's ultimate safe haven asset but you just repeatedly tried to prove gold is risky using your flawed methods.

You only proved gold is volatile against USD, or the prices of consumer goods and services within the US. All of those are still highly correlated with USD itself. When I removed the USD correlation and looked at some truly international assets such as a diversified basket of commodities suddenly the USD/US Treasury is more volatile.

Gold being independent from all countries in the world and being used as THE safe haven asset for thousands of years is a good enough proof that gold is extremely safe. The 10 year Treasury that couldn't even beat holding a piece of metal and do nothing should tell you how lousy that investment is.

With US being the most powerful country winning the cold war and having the largest economy in the past 50 years or so, its paper currencies still did terribly against gold. There's no telling how US will fair in the future. I'll sleep much better owning gold for the risk-averse part of my portfolio than holding some papers that promise to pay more paper.
+1 on this one, mate.

Gold is an asset that is a must have in a portfolio as far as I'm concerned (say, a 10% allocation). It's your hedge and disaster protection.

The problem is that traditional frequentist analysis doesn't really work with gold, because it's typically an asset that gives you protection on left tail events. You need to use a kind of Bayesian approach.

In the worst case, gold tends to improve risk adjusted returns on a traditional portfolio. In the best (worst?) case it can really save your bacon.

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 3:07 pm

gougou wrote:
Wed Jan 15, 2020 2:55 pm

You only proved gold is volatile against USD, or the prices of consumer goods and services within the US. All of those are still highly correlated with USD itself. When I removed the USD correlation and looked at some truly international assets such as a diversified basket of commodities suddenly the USD/US Treasury is more volatile.
Take gold as the 0 reference point framing, and thus all currencies fluctuate around it.

Hypothetically, I can buy into that . It's equally valid as taking the euro, yen, or renminbi as the 0 reference point.

Everybody lives in a country that uses fiat currencies as the medium of exchange, and it's logical to use your local currency as your 0 reference point.

But I can also make my 0 reference point oil, silver, copper, uranium, etc.

So any commodity or currency can be a 0 reference point and claim everything else fluctuates around it.

What good does any of this do me?


It's like when my wife, who grew up in Japan, but has lived in American for >20 years and is a US citizen, translates dollars to yen to get a sense of if something is expensive or not.

It's pointless, as she doesn't get paid in yen, she doesn't pay taxes in yen, and doesn't buy groceries or goods in yen.
Last edited by watchnerd on Wed Jan 15, 2020 3:34 pm, edited 4 times in total.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 3:09 pm

Smith1776 wrote:
Wed Jan 15, 2020 3:01 pm


Gold is an asset that is a must have in a portfolio as far as I'm concerned (say, a 10% allocation). It's your hedge and disaster protection.

How did you arrive at 10% as the right number?

Ray Dalio says 7.5%.

Harry Browne says 25%.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

beehivehave
Posts: 208
Joined: Thu Aug 08, 2019 1:21 pm

Re: Gold - How Do Investors Model Price?

Post by beehivehave » Wed Jan 15, 2020 3:12 pm

gougou wrote:
Wed Jan 15, 2020 2:55 pm
danielc wrote:
Mon Jan 13, 2020 8:06 pm
gougou wrote:
Mon Jan 13, 2020 12:47 pm
You obviously don’t know how the stdev of a portfolio is calculated. Let me give you a simple example. If a portfolio returns exactly 10% every year, then its annual returns are 10% with stdev of 0. Those are the numbers you’ll get on the portfoliovisualizer website.
That would be the standard deviation of the return series. That is what I did in my post, after subtracting inflation. And if you want to know whether an asset has a stable value relative to a basket of commodities, you need to compute the standard deviation of the values of that asset in terms of the value of that basket of commodities.
gougou wrote:
Mon Jan 13, 2020 12:47 pm
You suggested calculating the stdev of the total portfolio value series ($1, $1.1, $1.21 ...).
That is not what I said. I said that if you want to know whether 10-year treasuries or gold has a more stable value relative to a basket of commodities, you need to take the standard deviation of the ratio of the two prices. What you wrote is very very different from what I said. I don't know how you could have misunderstood my post, since I even told you which column of data from your own post I was referring to.

gougou wrote:
Mon Jan 13, 2020 12:47 pm
The stdev of this series is positive and increasing with every new entry.
Read that sentence again. I don't think you meant to say what you just wrote. A standard deviation is always, by definition, a positive quantity, and it is not "increasing" since it is a single value for a data series.
gougou wrote:
Mon Jan 13, 2020 12:47 pm
Now let’s go back to my analysis. I calculated how many Commodities each portfolio can buy year over year. Then I calculated how many % extra Commodities each portfolio can buy year over year (the returns of each portfolio in terms of Commodities). You should clearly see which returns are less volatile here. I calculated stdev on these two series and compared them.
1) The size of the basket of commodities that you can buy is more variable for gold.

2) The rate of return relative to the basket of commodities is more variable for 10-year treasuries.

You computed (2) and I computed (1).
gougou wrote:
Mon Jan 13, 2020 12:47 pm
CPI is the return of holding a bunch of stuff vs USD as defined by the US govt. CPI has a strong US bias and its underlying assets can change year over year.
Are you now going to argue that the basket of commodities from Portfolio Visualizer is a better measure of stuff people buy than the Consumer Price Index? Seriously? Also, CPI is not "defined" by the US gov't in any meaningful way, it is calculated by the US gov't. You make it sound like the US govt just picked whatever the heck it wanted, presumably for some nefarious purpose. CPI is calculated by the BLS; it is a survey of the price of goods that people actually buy.
gougou wrote:
Mon Jan 13, 2020 12:47 pm
I also don’t see where you incorporated CPI in your stdev analysis.
I told you. I subtracted the rate of inflation reported in Portfolio Visualizer. That's the easiest way to convert the nominal returns of 10 year treasuries and gold into "real" returns (i.e. relative to the basket of goods people buy).

gougou wrote:
Mon Jan 13, 2020 12:47 pm
Finally, I said gold had good long term returns and probably beaten bonds, and I was mostly just taking about the returns. You wanted to calculate the whole stdev and Sharpe ratio stuff. Those are difficult to do right.
I also was the one that looked up the returns. You said that gold had probably beaten bonds, but I'm the one who looked it up. The result was that gold had lower return than 10 year treasuries despite having a much higher stdev and much lower Sharpe ratio. You have alternated between arguing that volatility doesn't matter when looking at returns, and arguing that the volatility is wrong and somehow it must be the US dollar that's jumping up and down and making gold look bad. I used inflation data to show the volatility of USD (3.06%), and the volatility real returns for treasuries (IT = 7.82%; LT = 10.82%) and gold (29.20%).

gougou wrote:
Mon Jan 13, 2020 12:47 pm
Returns are indisputable while the stdev can change a lot depending on what base currency/asset you are using. And I just used Commodities to prove that you should not use USD as base to compare Gold vs US Treasury.
You used commodities to prove that gold returns are less volatile relative to a basket that contains gold, at least if you also chop off 75% of the history of gold to pick an atypical sample of 12 years. I used inflation data to prove that gold returns are more volatile relative to a basket of stuff that people actually buy, using the entire history of gold since the end of the gold standard. I cannot fathom why you keep insisting on pinning the blame on USD for the crappy behaviour of gold.
Gold returned more money since 1972 to 2019 with 7.53% CAGR vs 10-year Treasury's 7.09% CAGR. Gold is the world's ultimate safe haven asset but you just repeatedly tried to prove gold is risky using your flawed methods.

You only proved gold is volatile against USD, or the prices of consumer goods and services within the US. All of those are still highly correlated with USD itself. When I removed the USD correlation and looked at some truly international assets such as a diversified basket of commodities suddenly the USD/US Treasury is more volatile.

Gold being independent from all countries in the world and being used as THE safe haven asset for thousands of years is a good enough proof that gold is extremely safe. The 10 year Treasury that couldn't even beat holding a piece of metal and do nothing should tell you how lousy that investment is.

With US being the most powerful country winning the cold war and having the largest economy in the past 50 years or so, its paper currencies still did terribly against gold. There's no telling how US will fair in the future. I'll sleep much better owning gold for the risk-averse part of my portfolio than holding some papers that promise to pay more paper.
Define "risk". Compare the volatility of the price of gold to Treasury bonds. And it is not "some papers" promising, it's the full faith and credit of the US. Who is promising anything for gold? Speculators who supplement their returns by selling it at high fees to small suckers, er, investors.
Last edited by beehivehave on Wed Jan 15, 2020 3:13 pm, edited 1 time in total.

Smith1776
Posts: 62
Joined: Fri Apr 21, 2017 5:37 pm

Re: Gold - How Do Investors Model Price?

Post by Smith1776 » Wed Jan 15, 2020 3:13 pm

watchnerd wrote:
Wed Jan 15, 2020 3:09 pm
Smith1776 wrote:
Wed Jan 15, 2020 3:01 pm


Gold is an asset that is a must have in a portfolio as far as I'm concerned (say, a 10% allocation). It's your hedge and disaster protection.

How did you arrive at 10% as the right number?

Ray Dalio says 7.5%.

Harry Browne says 25%.
Yeah, I don't think there's any real way to arrive at a "precise" number.

Part of the rationale for the 10% number on my part is just recognizing investor behaviour. Many people can't even stomach the mere idea of holding more than 10% gold. It seems to be a number that's just enough to make a difference, but not so much that it's too bitter a pill to swallow.

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 3:16 pm

Smith1776 wrote:
Wed Jan 15, 2020 3:13 pm


Yeah, I don't think there's any real way to arrive at a "precise" number.

Part of the rationale for the 10% number on my part is just recognizing investor behaviour. Many people can't even stomach the mere idea of holding more than 10% gold. It seems to be a number that's just enough to make a difference, but not so much that it's too bitter a pill to swallow.
Which means we're back to the original thread point:

Nobody knows how do valuation of gold, or project future growth.

Without that, any portfolio allocation is arbitrary and has to be based on other factors, just as behavioral risk tolerance or just pure speculation.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 3:26 pm

beehivehave wrote:
Wed Jan 15, 2020 3:12 pm

Define "risk". Compare the volatility of the price of gold to Treasury bonds. And it is not "some papers" promising, it's the full faith and credit of the US. Who is promising anything for gold? Speculators who supplement their returns by selling it at high fees to small suckers, er, investors.
Okay, so.....

1. If I'm go into cryogenic sleep for 1,000 years
2. If I travel at relativistic velocities to a distant star system and then come back to earth 1,000 years from now
3. Build myself a giant pyramid tomb and load it with treasure for people to dig up and find 1,000 years in the future

....gold is probably superior to any fiat money backed by the taxation powers of a current regime that may not longer exist 1,000 years from now.

I can buy into that argument. That over 1,000 year timeframes, gold stands the test of time.

Errr, okay, great.....

None of those things are part of my life plan at the moment.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

Smith1776
Posts: 62
Joined: Fri Apr 21, 2017 5:37 pm

Re: Gold - How Do Investors Model Price?

Post by Smith1776 » Wed Jan 15, 2020 3:35 pm

watchnerd wrote:
Wed Jan 15, 2020 3:16 pm
Smith1776 wrote:
Wed Jan 15, 2020 3:13 pm


Yeah, I don't think there's any real way to arrive at a "precise" number.

Part of the rationale for the 10% number on my part is just recognizing investor behaviour. Many people can't even stomach the mere idea of holding more than 10% gold. It seems to be a number that's just enough to make a difference, but not so much that it's too bitter a pill to swallow.
Which means we're back to the original thread point:

Nobody knows how do valuation of gold, or project future growth.

Without that, any portfolio allocation is arbitrary and has to be based on other factors, just as behavioral risk tolerance or just pure speculation.
Yeah, I'd agree with that.

Fundamentally, traditional models just don't work with gold because it's a self referencing asset. There are no cash flows or counterparty risk to analyze. It's valuable because it's valuable. It's as good as gold because it's gold. It's old timey money because it's old timey money.

It seems it just "is". I do believe in holding a modicum of gold, but I don't blame anyone saying "to hell with it" and just buying stocks and bonds.

I've seen some analyses looking at all the materials and elements that exist around us and juxtaposing that with all the qualities that money needs to have. Gold (and partially silver) ended up having those intrinsic properties, while others fell by the wayside.

This could explain some of the value -- in the sense that gold is valuable because it has the intrinsic properties necessary for a valuable medium of exchange that other entities do not: it's rare but not too rare; it's durable but malleable; totally fungible; can't be printed at a whim; physically stable; minimal industrial uses to influence the price. There were others I can't remember.

By contrast, others put under the same scrutiny didn't stand up. Aluminum is too common. Arsenic is poisonous. Others are gaseous. Platinum is too rare. Paper can be printed ad infinitum.

I'm not necessarily advocating for those arguments, but they are interesting water cooler talk.

On a more practical level, a small gold allocation seems to improve traditional portfolio performance by a smidge. Good enough for me. And who knows, maybe the gold bugs will have their day.

sean.mcgrath
Posts: 165
Joined: Thu Dec 29, 2016 6:15 am
Location: US in NL

Re: Gold - How Do Investors Model Price?

Post by sean.mcgrath » Wed Jan 15, 2020 3:59 pm

sean.mcgrath wrote:
Tue Jan 14, 2020 4:11 pm
watchnerd wrote:
Sat Jan 11, 2020 10:18 am


Late reply, but labor costs are not the same as salary. As your link states, it includes benefits (including SS, I would expect). The salaries are "Roughly speaking, the average hourly pay for a member of the United Auto Workers currently ranges from $28 to $38 or so for those hired before September 2007, and between $16 and $20 for workers hired afterward." (from your link)

We can discuss whether benefits should be included as wages in the comparison, but they certainly would not be paid as much as the half Eagle (even pre-tax).
Fair.

But to flip it around, I'm pretty darn sure Ford workers in 1914 had little to zero benefits beyond wages. If someone wants to bother to look up the total benefits package of a 1914 Ford worker it might be interesting to know, but I'm sure it's "less".

So the total compensation comparison is still valid.
Yep, that is why I mentioned that it is debatable whether the benefits should be included. Some for sure, but there is an agenda and room for definition behind the benefit numbers, while the salary numbers are what they are. In general, I would prefer salary to benefits as it maximizes my choice (especially in a virtually no income tax world).

But yes, I believe you could argue it either way (and I am not even sure which side of the argument I would take).

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 4:24 pm

Smith1776 wrote:
Wed Jan 15, 2020 3:35 pm

On a more practical level, a small gold allocation seems to improve traditional portfolio performance by a smidge. Good enough for me. And who knows, maybe the gold bugs will have their day.
It does. But it doesn't seem to be much better than long treasuries at this.

Here is a historical efficient frontier for a port with:

US Stocks: 60% max
Long Treasuries: any allocation
Gold: any allocation

https://www.portfoliovisualizer.com/eff ... sset3=Gold

As equities increase, gold gets increasingly smaller relative to LTT.

By the time you get to 58% equities, gold is down to 0% on the tangency portfolio.
Last edited by watchnerd on Wed Jan 15, 2020 4:30 pm, edited 1 time in total.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

gougou
Posts: 245
Joined: Thu Sep 28, 2017 7:42 pm
Location: San Francisco Bay Area

Re: Gold - How Do Investors Model Price?

Post by gougou » Wed Jan 15, 2020 4:29 pm

watchnerd wrote:
Wed Jan 15, 2020 3:07 pm
gougou wrote:
Wed Jan 15, 2020 2:55 pm

You only proved gold is volatile against USD, or the prices of consumer goods and services within the US. All of those are still highly correlated with USD itself. When I removed the USD correlation and looked at some truly international assets such as a diversified basket of commodities suddenly the USD/US Treasury is more volatile.
Take gold as the 0 reference point framing, and thus all currencies fluctuate around it.

Hypothetically, I can buy into that . It's equally valid as taking the euro, yen, or renminbi as the 0 reference point.

Everybody lives in a country that uses fiat currencies as the medium of exchange, and it's logical to use your local currency as your 0 reference point.

But I can also make my 0 reference point oil, silver, copper, uranium, etc.

So any commodity or currency can be a 0 reference point and claim everything else fluctuates around it.

What good does any of this do me?


It's like when my wife, who grew up in Japan, but has lived in American for >20 years and is a US citizen, translates dollars to yen to get a sense of if something is expensive or not.

It's pointless, as she doesn't get paid in yen, she doesn't pay taxes in yen, and doesn't buy groceries or goods in yen.
Let’s think about why gold was “volatile” in the most recent boom and bust. Gold ran up strongly 2009 to 2012 when US economy was in bad shape, and ran down 2012 to 2016 when US economy was strong.

Now you can calculate some volatility metrics for gold using USD as the reference asset. But that doesn’t mean gold is volatile or risky when US economy was the roller coaster ride. People just like using USD as the 0 volatility asset when it is not. Gold is actually insulated from the volatile US economy.

over45
Posts: 49
Joined: Sat Nov 18, 2017 8:35 am

Re: Gold - How Do Investors Model Price?

Post by over45 » Wed Jan 15, 2020 4:29 pm

You may not like the Federal Reserve or Congress, but only the U.S. Congress has the legal power in the U.S. to make the rules regarding what money is.
I think it might be more accurate to say that they have the power to make the rules regarding what "legal tender" is. Money is a concept and "money" differs from "currency".

beehivehave
Posts: 208
Joined: Thu Aug 08, 2019 1:21 pm

Re: Gold - How Do Investors Model Price?

Post by beehivehave » Wed Jan 15, 2020 4:33 pm

watchnerd wrote:
Wed Jan 15, 2020 3:26 pm
beehivehave wrote:
Wed Jan 15, 2020 3:12 pm

Define "risk". Compare the volatility of the price of gold to Treasury bonds. And it is not "some papers" promising, it's the full faith and credit of the US. Who is promising anything for gold? Speculators who supplement their returns by selling it at high fees to small suckers, er, investors.
Okay, so.....

1. If I'm go into cryogenic sleep for 1,000 years
2. If I travel at relativistic velocities to a distant star system and then come back to earth 1,000 years from now
3. Build myself a giant pyramid tomb and load it with treasure for people to dig up and find 1,000 years in the future

....gold is probably superior to any fiat money backed by the taxation powers of a current regime that may not longer exist 1,000 years from now.

I can buy into that argument. That over 1,000 year timeframes, gold stands the test of time.

Errr, okay, great.....

None of those things are part of my life plan at the moment.
In the next 1,000 years and at the present pace of technological progress, somebody will figure out how to create gold from thin air, so there goes that argument.

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 4:33 pm

gougou wrote:
Wed Jan 15, 2020 4:29 pm

Let’s think about why gold was “volatile” in the most recent boom and bust. Gold ran up strongly 2009 to 2012 when US economy was in bad shape, and ran down 2012 to 2016 when US economy was strong.

Now you can calculate some volatility metrics for gold using USD as the reference asset. But that doesn’t mean gold is volatile or risky when US economy was the roller coaster ride. People just like using USD as the 0 volatility asset when it is not. Gold is actually insulated from the volatile US economy.
It doesn't matter to me. It's an academic point of zero practical importance to me.

Just like the volatility of USD relative to GBP, yen or euro doesn't matter much to me, either.

I manage my financial matters, pay my taxes, earn my salary, and pay my bills in USD, so that's my 0 reference point.

If people were paying me in gold, things would be different.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 4:37 pm

beehivehave wrote:
Wed Jan 15, 2020 4:33 pm

In the next 1,000 years and at the present pace of technological progress, somebody will figure out how to create gold from thin air, so there goes that argument.
Well, crap, I forgot about that.

Yeah, with enough energy, you can use a particle accelerator or nuclear reactor to mash other elements together to make gold.

It's already been done before, it's just not cheap enough yet for gold:

https://en.wikipedia.org/wiki/Synthesis ... ous_metals
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

Smith1776
Posts: 62
Joined: Fri Apr 21, 2017 5:37 pm

Re: Gold - How Do Investors Model Price?

Post by Smith1776 » Wed Jan 15, 2020 4:38 pm

watchnerd wrote:
Wed Jan 15, 2020 4:24 pm
Smith1776 wrote:
Wed Jan 15, 2020 3:35 pm

On a more practical level, a small gold allocation seems to improve traditional portfolio performance by a smidge. Good enough for me. And who knows, maybe the gold bugs will have their day.
It does. But it doesn't seem to be much better than long treasuries at this.

Here is a historical efficient frontier for a port with:

US Stocks: 60% max
Long Treasuries: any allocation
Gold: any allocation

https://www.portfoliovisualizer.com/eff ... sset3=Gold

As equities increase, gold gets increasingly smaller relative to LTT.

By the time you get to 58% equities, gold is down to 0% on the tangency portfolio.
That's absolutely true. I believe vineviz came to a similar conclusion somewhere way back.

My personal take is that I'd still like to have that small slice of gold for the sake of having an asset that is entirely divorced from the traditional stocks and bonds I have.

To each their own I guess.

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 4:47 pm

Smith1776 wrote:
Wed Jan 15, 2020 4:38 pm

That's absolutely true. I believe vineviz came to a similar conclusion somewhere way back.

My personal take is that I'd still like to have that small slice of gold for the sake of having an asset that is entirely divorced from the traditional stocks and bonds I have.

To each their own I guess.
From an efficient frontier / risk parity / MVO POV, the answer might go something like this:

1. Model the portfolio and see how high LTT goes for the bond portion on the tangency portfolio.

2. Figure out how comfortable you are with current duration risk and yields of LTT. If that, when combined with the allocation from #1, freaks you out, reduce LTT and add gold.

3. If adding gold gets large enough to also freak you out, add the remainder in cash (0 correlation to anything).
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

gougou
Posts: 245
Joined: Thu Sep 28, 2017 7:42 pm
Location: San Francisco Bay Area

Re: Gold - How Do Investors Model Price?

Post by gougou » Wed Jan 15, 2020 5:09 pm

watchnerd wrote:
Wed Jan 15, 2020 4:33 pm
gougou wrote:
Wed Jan 15, 2020 4:29 pm

Let’s think about why gold was “volatile” in the most recent boom and bust. Gold ran up strongly 2009 to 2012 when US economy was in bad shape, and ran down 2012 to 2016 when US economy was strong.

Now you can calculate some volatility metrics for gold using USD as the reference asset. But that doesn’t mean gold is volatile or risky when US economy was the roller coaster ride. People just like using USD as the 0 volatility asset when it is not. Gold is actually insulated from the volatile US economy.
It doesn't matter to me. It's an academic point of zero practical importance to me.

Just like the volatility of USD relative to GBP, yen or euro doesn't matter much to me, either.

I manage my financial matters, pay my taxes, earn my salary, and pay my bills in USD, so that's my 0 reference point.

If people were paying me in gold, things would be different.
Sure, you can put your full faith in the value of USD and assume it is the perfectly stable asset. Then you’ll reach the conclusion that gold is volatile and that its returns cannot make up for its volatility.

But when people say gold is volatile they usually don’t put a disclaimer that they are assuming USD is perfectly stable. That’s my whole point. I’ve also talked about why I prefer using gold as the stable and risk free asset and why I am bothered when my portfolio can’t even beat the return of gold.

over45
Posts: 49
Joined: Sat Nov 18, 2017 8:35 am

Re: Gold - How Do Investors Model Price?

Post by over45 » Wed Jan 15, 2020 5:27 pm

This is a very informative thread and I have been debating whether to buy a little physical gold to hold vs. investing in GLD which is substantially more liquid and easier to convert to dollars if the need arises. It's pretty expensive now (IMHO), but in the back of my mind I have a feeling that there will come a day sometime soon where the US will have to devalue the dollar (further).

I don't think we can ever go back to a fully gold backed currency or a sound money system for that matter because politicians wouldn't be able to make promises and give away the farm without the ability to print, print, print. Some say the system will keep expanding until it can't - globally - and then at some point there will be a good old fashioned debt jubilee.

The main question I keep asking myself if buying gold is: if I want to sell it - who is going to buy it? When it was $500 oz. that wasn't so hard to answer - but now it is over $1550/oz. and I know most of that run up is from investors fleeing other countries looking for safe haven assets.

They are people talking about going to an all digital currency and possibly negative interest rates - what could possibly go wrong?
Last edited by over45 on Wed Jan 15, 2020 5:49 pm, edited 1 time in total.

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 5:30 pm

gougou wrote:
Wed Jan 15, 2020 5:09 pm


Sure, you can put your full faith in the value of USD and assume it is the perfectly stable asset. Then you’ll reach the conclusion that gold is volatile and that its returns cannot make up for its volatility.
Yep, that's what I do. If I was European, I'd do the same thing with the euro. Japanese, I'd use yen.

Because if I don't do that, then CAPM, CAPE, Fama-French, MPT, Black Scholes, etc, all go out the window as they're based on the concept of the risk free asset being ultra short government bills in the currency of your choice.

All of these models require the ability to model a risk free rate of return which, at the crudest level, is a T-bill yield (or the euro/GBP/yen etc equivalent).

Because gold doesn't have a risk free rate of return, none of those analytical tools will work with it.

If you make gold your 0 reference you basically have no way to model stock valuations, equity risk premiums, bond duration, or much of anything.

You're basically intentionally choosing to abandon all of modern portfolio theory and monetary economics.

At that point, I'd rather just gamble -- because at least I can calculate the probabilities involved in a game of craps better than a financial world that uses gold as the definition of a stable asset.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 5:34 pm

over45 wrote:
Wed Jan 15, 2020 5:27 pm
I know most of that run up is from investors fleeing other countries looking for safe haven assets.
Gold is definitely driven by a lot of buyers in India or China.

But they're definitely not all fleeing (you can't flee 3 billion people to anywhere between the two of them). And I'm not so sure it's safe haven, either, as much as it is increasing status / wealth -- old cultures that have a long history of displaying gold as a social signal of status.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

over45
Posts: 49
Joined: Sat Nov 18, 2017 8:35 am

Re: Gold - How Do Investors Model Price?

Post by over45 » Wed Jan 15, 2020 5:42 pm

And I'm not so sure it's safe haven, either, as much as it is increasing status / wealth -- old cultures that have a long history of displaying gold as a social signal of status.
It's probably both. Goldbugs will tell you that is the beauty of gold.

sean.mcgrath
Posts: 165
Joined: Thu Dec 29, 2016 6:15 am
Location: US in NL

Re: Gold - How Do Investors Model Price?

Post by sean.mcgrath » Wed Jan 15, 2020 5:45 pm

watchnerd wrote:
Wed Jan 15, 2020 4:37 pm
beehivehave wrote:
Wed Jan 15, 2020 4:33 pm

In the next 1,000 years and at the present pace of technological progress, somebody will figure out how to create gold from thin air, so there goes that argument.
Well, crap, I forgot about that.

Yeah, with enough energy, you can use a particle accelerator or nuclear reactor to mash other elements together to make gold.

It's already been done before, it's just not cheap enough yet for gold:

https://en.wikipedia.org/wiki/Synthesis ... ous_metals
Calculate the energy costs -- pretty prohibitive and not related to state of technological advancement unless energy becomes virtually free. In that case, life is easy anyway (says the geophysicist ;-)

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 5:55 pm

over45 wrote:
Wed Jan 15, 2020 5:27 pm

I don't think we can ever go back to a fully gold backed currency or a sound money system for that matter because politicians wouldn't be able to make promises and give away the farm without the ability to print, print, print.
The problem of politicians and regimes mucking with the money supply predates fiat money by millennia.

In the old days, it was done by increasing the amount of base metals (e.g. lead) in the coinage. "Debasement" is the term.

https://en.wikipedia.org/wiki/Roman_currency

https://en.wikipedia.org/wiki/Qing_dynasty_coinage

"Sound money" is a bit of a utopian dream that hasn't really existed in history for any extended duration. Even the vaunted Swiss Franc is considered to be overvalued by the Swiss National Bank and they've grumbled about wanting to devalue it.

Politicians will almost certainly keep finding new ways to manipulate any currency in the future, as well.

IMHO, some strong gold advocates conflate "money" with "wealth".

I strongly disagree.

Money is a medium of exchange which is used to accumulate wealth.

Money is not, in itself, a durable form of wealth, and probably shouldn't ever be expected to be. It needs to be converted into wealth.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 5:57 pm

sean.mcgrath wrote:
Wed Jan 15, 2020 5:45 pm


Calculate the energy costs -- pretty prohibitive and not related to state of technological advancement unless energy becomes virtually free. In that case, life is easy anyway (says the geophysicist ;-)
We were assuming 1,000 years from now, when we'll have fusion reactors or miniature black hole particle accretion disk x-ray power sources.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

over45
Posts: 49
Joined: Sat Nov 18, 2017 8:35 am

Re: Gold - How Do Investors Model Price?

Post by over45 » Wed Jan 15, 2020 6:00 pm

The other question I keep asking myself about gold (or silver) is: what would it be worth to anyone if there was no "spot price" system in place? What would the value be if someone had it and needed to sell it and there was no reference point to work off of ? We may have all seen videos where an interviewer hits the street and offers passers by a choice: a one ounce bar of silver -- or a chocolate bar. Virtually everyone takes the chocolate bar. If I have 10 ounces of gold and you have a bushel of apples and I am starving - those apples are more valuable to me than the gold. If I offered people a one ounce bar of silver or a $10 bill - 99.9% of them would take the $10 bill.

User avatar
Topic Author
watchnerd
Posts: 2527
Joined: Sat Mar 03, 2007 11:18 am
Location: Seattle, WA, USA

Re: Gold - How Do Investors Model Price?

Post by watchnerd » Wed Jan 15, 2020 6:01 pm

over45 wrote:
Wed Jan 15, 2020 5:42 pm

It's probably both. Goldbugs will tell you that is the beauty of gold.
Ultimately, I don't think it matters.

If I was going to buy gold, I'd much rather bank of the demand (for whatever reason) of billions of Indians and Chinese, than some zombie apocalypse rationale.
70% Global Market Weight Equities | 15% Long Treasuries 15% short TIPS & cash || RSU + ESPP

Post Reply