Gold's Strange Behavior Shows It's No Haven

[Fremdon Ferndock]

If you’ve been paying attention to prices of gold and gold mining stocks the last few weeks, you probably noticed that they go up when the stock market goes up, and they go down when the stock market goes down, limiting the usefulness of gold as a hedge.

At the risk of stating the obvious, inflation rates are very high and the performance of gold has been terrible.

https://www.advisorperspectives.com/art ... -no-haven

I've been holding a small allocation but think it might be time to dump it.

[Whakamole]

Portfolio Visualizer shows a 0.07 correlation between VTI and GLD over the past 18 years: https://www.portfoliovisualizer.com/ass ... &months=60

The rolling correlation over the past year is 0.228.

Any recent correlation is just recency bias.

[Phyneas]

If you’ve been paying attention to prices of gold and gold mining stocks the last few weeks, you probably noticed that they go up when the stock market goes up, and they go down when the stock market goes down, limiting the usefulness of gold as a hedge.

At the risk of stating the obvious, inflation rates are very high and the performance of gold has been terrible.

https://www.advisorperspectives.com/art ... -no-haven

I've been holding a small allocation but think it might be time to dump it.

Gold is a hedge against stocks (it is positive on the year, stocks are negative), bonds (same deal), and DXY (which is become extremely strong recently, limiting gold's performance). Gold didn't crash in 2020 like stocks did, and it has actually kept pace with inflation since 2020 when all of that money printing artificially propped up the stock market. It's actually done a terrific job, and exactly what is expected of it.

[jason2459]

I'm not a gold bug by any stretch of the imagination but using just a couple weeks of data to come to any conclusion for anything is completely ridiculous and foolish.

[ivgrivchuck]

Stocks, Bonds, Gold make a nice triplet where the correlation between each of them is approximately zero.

Zero correlation is different from negative correlation. In most economic environments gold's movement is almost completely unrelated to the movement of stocks and bonds.

Holding 5%-10% of one's portfolio in gold can be justified, but it's also perfectly fine not to hold any (Disclosure: I don't hold any at the moment).

But if you do hold gold, you need to hold it for the right reasons...

[akxc]

Gold also tends to be inversely correlated both to real rates and the strength of the USD...both of which have gone up substantially recently.

https://www.portfoliovisualizer.com/ass ... &months=36
https://papers.ssrn.com/sol3/papers.cfm ... id=3934052
https://www.sciencedirect.com/science/a ... 3104000794

[KlangFool]

OP,

If you do not know why you should hold Gold, you should dump it. You cannot buy confidence from others.

KlangFool

[secondopinion]

If you’ve been paying attention to prices of gold and gold mining stocks the last few weeks, you probably noticed that they go up when the stock market goes up, and they go down when the stock market goes down, limiting the usefulness of gold as a hedge.

At the risk of stating the obvious, inflation rates are very high and the performance of gold has been terrible.

https://www.advisorperspectives.com/art ... -no-haven

I've been holding a small allocation but think it might be time to dump it.

It is not a "haven"; it is an alternative. It can go with stocks; it can go against stocks. A true alternative should be moving randomly against stocks and bonds. Otherwise, why even bother?

And let me ask a question, does it suffer when all stocks and bonds are suffering for prolonged periods (not just when it is stocks only)?

[rgs92]

I guess the Permanent Portfolio would provide a useful example of having a precious metals allocation.

[Watty]

I do not pretend to understand the price of gold but one important factor to watch is interest rates.

When interest rates were essentially zero owning $10,000 in gold cost you essentially nothing in lost interest if the price of gold did not change. If you assume a 2% interest rate then it would cost you $200 a year in lost interest.

For an individual that might not be a big factor but behind the scenes(which I understand even less) there all sorts of things going on with futures, contracts, gold loaning, etc where higher interest rate may also have an impact.

There is also a lot going on in the world with the pandemic, war, heatwave/drought in India, etc which may either cause people to sell gold to raise money or at least not buy as much gold.

[tvubpwcisla]

Bogle said that Gold was not an investment. I believe him.

[richard.h.gao]

You are confusing Gold with Gold Mining Stocks.

The former is a metal, the latter is a company.

[Trapper]

I have been watching gold & silver spot prices for several years.
My take (which is probably completely wrong) is that since the spot prices have been relatively flat (static), that inflation may be actually be ready to adjust down, bringing interest rates & prices with it, or at least leveling at current prices & rates.

[3CT_Paddler]

I don't buy the argument. Gold is more about the expected future weakness of currencies, primarily the biggest currency - the dollar. Right now the Fed is tightening and the dollar is strengthening. If the Fed relaxes monetary policy (which is more likely than a Volker like tightening) then I think you will see gold go higher. But that all depends on how long this tightening cycle lasts. I think if we saw a true Volker like response with more sound fiscal policy you would see gold crash 50% or more - which also seems very unlikely.

[RoadThunder]

I don't buy the argument. Gold is more about the expected future weakness of currencies, primarily the biggest currency - the dollar. Right now the Fed is tightening and the dollar is strengthening. If the Fed relaxes monetary policy (which is more likely than a Volker like tightening) then I think you will see gold go higher. But that all depends on how long this tightening cycle lasts. I think if we saw a true Volker like response with more sound fiscal policy you would see gold crash 50% or more - which also seems very unlikely.

Actually gold is about jewelry- approximately 78% of all gold is used in personal jewelry. Anything else is a far distance second.

[secondopinion]

Bogle said that Gold was not an investment. I believe him.

Bogle might or might not be right; did he give his reasons?

[alluringreality]

Bogle said that Gold was not an investment. I believe him.

Bogle might or might not be right; did he give his reasons?

I think that he considered it a non-productive asset, meaning that it does not provide cashflow. There are some brief comments around the 2 minute mark in the following.
https://www.google.com/url?sa=t&source= ... K0jNt95UDs

[Watty]

Actually gold is about jewelry- approximately 78% of all gold is used in personal jewelry. Anything else is a far distance second.

In some countries like India and China gold jewelry is used as an investment and store of wealth.

[secondopinion]

Bogle said that Gold was not an investment. I believe him.

Bogle might or might not be right; did he give his reasons?

I think that he considered it a non-productive asset, meaning that it does not provide a cashflow.
https://www.google.com/url?sa=t&source= ... K0jNt95UDs

Right. I just wanted hear it straight from the source and it being a "non-productive asset" is a good argument. However, I did an analysis in the past; if I remember the results right, it suggests that long-term treasuries do best with volatility when there is between 0%-25% gold in place of them (depending on the stock included and the bonds besides long-term treasuries). Not exactly a lot, but gold appeared to be the only alternative that seemed to actually produce a portfolio improvement of some kind. My guess is that those who are duration matching very long into the future might cushion the ride with some gold. Otherwise, it is hard to beat the stock/bond combo.

[halfnine]

Gold's Strange Behavior Shows It's No Haven

Owning gold is kind of like wearing a life jacket while rafting. You are still going for a swim but odds are at least you are not going to drown. And just like a life jacket most of the time you don't need it and it just sits there on you chafing you a bit. But expecting gold to be a haven right now is a bit like expecting a life jacket to do something for you at the beginning of the rapids. That's just not how a life jacket works.

[jason2459]

Bogle said that Gold was not an investment. I believe him.

Would you believe he invested in gold?

[ivgrivchuck]

Bogle said that Gold was not an investment. I believe him.

I think he is right. It's not an investment, rather it's a storehold of wealth.

If you are holding 100% stocks, don't bother.

If you are not holding 100% stocks, the question arises if it's wise to keep all your remaining wealth in dollar nominated instruments. Dollar after all is a fiat currency, its value is mostly backed by the faith that people have in it... Gold has retained its value for thousands of years. Meanwhile hundreds of fiat currencies have become worthless.

[abc132]

I'll consider a few percent gold once I retire. I think many people want to count on past relationships holding up, but my belief is simply that gold can act differently. I wouldn't need or expect a repeat of the past, but gold would be likely to be beneficial in some of the future downturns. QREARX has done well for me so far this year, even though it didn't hold up for those that held it back in 2008. Each downturn is different.

I'm not sure any one asset is really a safe haven, as you want a combination of assets that is likely to succeed.

[NiceUnparticularMan]

In late 2019, Vanguard published a white paper on unexpected inflation in which they presented data showing that free-floating gold has been a far less efficient unexpected USD inflation hedge than diversified commodities futures. This makes sense because unlike most commodities, free-floating gold has an extremely large speculative/greater-fool component to its market price, far above its intrinsic value as a productive material. Therefore, other commodities, particularly taken as a diversified pool, are going to have a far more predictable/exploitable relationship to general inflation, whereas the large greater-fool component to gold pricing is adding a lot of noise/volatility on top of that macroeconomic signal.

Anyway, Vanguard's study has gotten a great out of sample test of rather extreme unexpected inflation recently. And sure enough, diversified commodities futures have been way out-performing gold in this period.

My point is this is not recency bias. There are fundamental theoretical/model reasons to think diversified commodities futures are a better asset for this purpose than free-floating gold. There was a comprehensive empirical study which showed that was true in the entire period of free-floating gold through 2019. There is now also a powerful out-of-sample test providing additional confirmation.

What exactly more could anyone ask on this subject? If all that is not sufficient to convince someone diversified commodities futures are a better asset for this purpose than free-floating gold, is there ANY logic or evidence that could persuade them?

[ivgrivchuck]

What exactly more could anyone ask on this subject? If all that is not sufficient to convince someone diversified commodities futures are a better asset for this purpose than free-floating gold, is there ANY logic or evidence that could persuade them?

What kind of long term return can one expect from holding commodity futures? Can one expect to even get to 0% real? I'm not an expert in futures, but my understanding is that constantly rolling futures has spread costs that vary depending on market environments. Gold has demonstrated that it has 0% real over long time horizons. Can the same be said about these futures?

But yes, please try to convince me that commodity futures are a better deal than gold.

[secondopinion]

In late 2019, Vanguard published a white paper on unexpected inflation in which they presented data showing that free-floating gold has been a far less efficient unexpected USD inflation hedge than diversified commodities futures. This makes sense because unlike most commodities, free-floating gold has an extremely large speculative/greater-fool component to its market price, far above its intrinsic value as a productive material. Therefore, other commodities, particularly taken as a diversified pool, are going to have a far more predictable/exploitable relationship to general inflation, whereas the large greater-fool component to gold pricing is adding a lot of noise/volatility on top of that macroeconomic signal.

Anyway, Vanguard's study has gotten a great out of sample test of rather extreme unexpected inflation recently. And sure enough, diversified commodities futures have been way out-performing gold in this period.

My point is this is not recency bias. There are fundamental theoretical/model reasons to think diversified commodities futures are a better asset for this purpose than free-floating gold. There was a comprehensive empirical study which showed that was true in the entire period of free-floating gold through 2019. There is now also a powerful out-of-sample test providing additional confirmation.

What exactly more could anyone ask on this subject? If all that is not sufficient to convince someone diversified commodities futures are a better asset for this purpose than free-floating gold, is there ANY logic or evidence that could persuade them?

Gold is not supposed to behave like commodities; it has been a fundamental store of wealth for thousands of years. Inflation was not why it was invested into; it was recognized by the large amount of the world as valuable. When commodities are needed the most, no one wants gold or currency. As a result, I think the claim it should counter inflation on the spot is false, as no one needs gold in dire situations. On far, far longer terms, it does counter inflation because it is still seen as the same thing as always.

The greater fool theory holds no claim when compared to fiat currencies; after all, the value of a currency is what is given to it. We are experiencing inflation because the dollar is not holding it value against what can be purchased. And unlike gold, many currencies are actually designed to lose value regardless of people's valuation of the currency. Gold loses value only if people stop valuing it.

In short, you are talking about two radically different prospects. Do not compare commodities to gold as the results will certainly vary.

[manuvns]

it's heaven becuse when i buy it for my wife she is happy ! and the gold i got for 500$ in 2000's is now worth 4-5x it's value

[Walkure]

Actually gold is about jewelry- approximately 78% of all gold is used in personal jewelry. Anything else is a far distance second.

In some countries like India and China gold jewelry is used as an investment and store of wealth.

More importantly, this "78%" is only true of actual above ground, refined gold in circulation. "Financialized" gold (what some gold bugs might derisively term "paper gold") is a whole different question. Let's say that Gold Miner Company A has proven reserves of x million ounces. They are valued and traded based on what the market estimates as the present value of the future stream of profit they can derive from mining that gold. Now suppose they decide to hedge their exposure to fluctuating spot prices by selling a futures contract to lock in some of their production. That futures contract is purchased by Market Maker 1, who in turn uses that asset to hypothecate the creation of a share of a gold ETN that Retail Investor Bob buys. In this example, if some enterprising soul actually added up the market cap of all gold miners, the notional exposure of the futures market, and the precious metals ETF/ETNs, one would find that each physical ounce of gold is backing 3x its actual value in financial instruments. Yet it's completely ignored in the denominator of the 78%.

[Gaston]

At the risk of stating the obvious, inflation rates are very high and the performance of gold has been terrible.

As Larry Swedroe and others have noted, gold does tend to keep pace with inflation over long periods of time. And when they say “long periods”, they are quick to point out that this means “over centuries”, not within a human life span.

[halfnine]

At the risk of stating the obvious, inflation rates are very high and the performance of gold has been terrible.

As Larry Swedroe and others have noted, gold does tend to keep pace with inflation over long periods of time. And when they say “long periods”, they are quick to point out that this means “over centuries”, not within a human life span.

I can't find anything factually accurate in your statement. Gold doesn't "tend" to keep pace with inflation over "centuries" gold "tends" to keep up with inflation over decades. Additionally, I am not aware of Larry indicating that it would take centuries. I have seen him state it could could take up to a century. And I am aware of at least one data point where this is true so he is not technically incorrect. Of course, I am also aware of more data points of stock markets not only not keeping up with inflation but completedly ceasing to exist over an even more recent time frame so there is no free lunch. As to the "others" I don't know who they are or why it should matter anyway. One could make a logical argument against gold ownership based on one's own situation and time horizon with existing data without appealing to authrority figures or hyperbole.

[secondopinion]

At the risk of stating the obvious, inflation rates are very high and the performance of gold has been terrible.

As Larry Swedroe and others have noted, gold does tend to keep pace with inflation over long periods of time. And when they say “long periods”, they are quick to point out that this means “over centuries”, not within a human life span.

I can't find anything factually accurate in your statement. Gold doesn't "tend" to keep pace with inflation over "centuries" gold "tends" to keep up with inflation over decades. Additionally, I am not aware of Larry indicating that it would take centuries. I have seen him state it could could take up to a century. And I am aware of at least one data point where this is true so he is not technically incorrect. Of course, I am also aware of more data points of stock markets not only not keeping up with inflation but completedly ceasing to exist over an even more recent time frame so there is no free lunch. As to the "others" I don't know who they are or why it should matter anyway. One could make a logical argument against gold ownership based on one's own situation and time horizon with existing data without appealing to authrority figures or hyperbole.

Right. And one can argue for having gold. I split the difference and hold a little; since optimality has diminishing benefits occurring quite a while before the maximum, a small percentage still gives a bit of benefit. We do not ever need to be on polar ends of the discussion.

[Forester]

Gold vs US stocks & long bonds since inflation broke above 4% in April 2021:

[3CT_Paddler]

I don't buy the argument. Gold is more about the expected future weakness of currencies, primarily the biggest currency - the dollar. Right now the Fed is tightening and the dollar is strengthening. If the Fed relaxes monetary policy (which is more likely than a Volker like tightening) then I think you will see gold go higher. But that all depends on how long this tightening cycle lasts. I think if we saw a true Volker like response with more sound fiscal policy you would see gold crash 50% or more - which also seems very unlikely.

Actually gold is about jewelry- approximately 78% of all gold is used in personal jewelry. Anything else is a far distance second.

If the price of gold was driven by jewelry demand, then it should be positively correlated with good economic times and a growing economy. When the US dollar was relatively strong and inflation was low and the economy was booming during the 90s, gold did absolutely terrible. Gold is a safe haven from fiat currencies when the world is unsure if the largest governments will devalue their currencies.

[halfnine]

I don't buy the argument. Gold is more about the expected future weakness of currencies, primarily the biggest currency - the dollar. Right now the Fed is tightening and the dollar is strengthening. If the Fed relaxes monetary policy (which is more likely than a Volker like tightening) then I think you will see gold go higher. But that all depends on how long this tightening cycle lasts. I think if we saw a true Volker like response with more sound fiscal policy you would see gold crash 50% or more - which also seems very unlikely.

Actually gold is about jewelry- approximately 78% of all gold is used in personal jewelry. Anything else is a far distance second.

If the price of gold was driven by jewelry demand, then it should be positively correlated with good economic times and a growing economy. When the US dollar was relatively strong and inflation was low and the economy was booming during the 90s, gold did absolutely terrible. Gold is a safe haven from fiat currencies when the world is unsure if the largest governments will devalue their currencies.

The jewelry demand in India and China dwarf the jewelry demand in the USA so extrapolating US economic data is likely irrelevant to the overall price of gold. That doesn't necessarily mean, of course, that jewelry prices do or don't drive gold prices. It's just no conclusion can be gained from the example you've given.

[firebirdparts]

I guess the Permanent Portfolio would provide a useful example of having a precious metals allocation.

Sort of. If you had a time machine, and enter the PP in 1972, it would be awesome. One great thing about its performance is that anybody can conclude whether or not that'll happen again. It's like guessing where lightning will strike. If you guess "not here" you've got a pretty got chance of being right.

[Valuethinker]

I don't buy the argument. Gold is more about the expected future weakness of currencies, primarily the biggest currency - the dollar. Right now the Fed is tightening and the dollar is strengthening. If the Fed relaxes monetary policy (which is more likely than a Volker like tightening) then I think you will see gold go higher. But that all depends on how long this tightening cycle lasts. I think if we saw a true Volker like response with more sound fiscal policy you would see gold crash 50% or more - which also seems very unlikely.

Actually gold is about jewelry- approximately 78% of all gold is used in personal jewelry. Anything else is a far distance second.

If the price of gold was driven by jewelry demand, then it should be positively correlated with good economic times and a growing economy. When the US dollar was relatively strong and inflation was low and the economy was booming during the 90s, gold did absolutely terrible. Gold is a safe haven from fiat currencies when the world is unsure if the largest governments will devalue their currencies.

The jewelry demand in India and China dwarf the jewelry demand in the USA so extrapolating US economic data is likely irrelevant to the overall price of gold. That doesn't necessarily mean, of course, that jewelry prices do or don't drive gold prices. It's just no conclusion can be gained from the example you've given.

However I believe that gold production and consumption are only a small percentage of total gold stocks?

This is different from (any?) other industrial metal - where production is quite close to consumption, over time ie stockpiles are not a large percentage of say copper consumption.

So one cannot make a lot of good predictions about gold based on "supply" and "demand" because a lot of the latter is simply to stick it in refined form in bank vaults.

What one can say is that many Central Banks have been scrambling to diversify away from USD assets. The sanctions against Russia tell us why (at least partly), but the trend has been noticeable for a lot longer than that. Gold is one of the few assets which is fungible enough & has enough volume to do that.

[Broken Man 1999]

Bogle said that Gold was not an investment. I believe him.

And yet he had 5% of an endowment for Blair Academy in gold. His thought was gold for a possible catastrophe. Plus 5% in Emerging Markets.

Gold AND International..... oh my!

Broken Man 1999

[NiceUnparticularMan]

What exactly more could anyone ask on this subject? If all that is not sufficient to convince someone diversified commodities futures are a better asset for this purpose than free-floating gold, is there ANY logic or evidence that could persuade them?

What kind of long term return can one expect from holding commodity futures? Can one expect to even get to 0% real?

So collateralized commodity futures can get returns from spot price changes, rolling futures, and the collateral. The expected real return on spot prices changes is slightly negative. Rolling futures is nominally slightly positive, but only really if you do something to prevent frontrunning and avoid contango. The collateral return depends on the market rates on your collateral.

All told, I would say 0% real, with a lot of variability, is a good forward-looking estimate.

Gold has demonstrated that it has 0% real over long time horizons.

So with gold, you have to be careful not to use data from periods where gold was not free-floating, or making the transition from fixed to free-floating.

As a free-floating commodity, its only possible source of return is spot price changes. Again, the expected return on commodity spot price changes is slightly negative. So, that's what you should expect from free-floating gold going forward. And that is consistent with the data so far from the free-floating gold era.

With an important caveat. Gold is unlike most commodities due to the large speculative/greater-fool component to its market price. That is an inherently unpredictable thing and so doesn't necessarily change the expected return calculus. But it does mean there is an unusually large risk that free-floating gold could experience large, permanent real losses versus other commodities.

But yes, please try to convince me that commodity futures are a better deal than gold.

So again, the above is standard economic modeling theory. Vanguard looked at data on this subject and it was consistent with all the above. That analysis then got a recent out of sample test, and the results of that out of sample test were also consistent with all of the above.

Of course if a person wanted to speculate on gold, and they requireed absolute proof of what will happen in the future to change their mind about that plan--well, of course no such absolute proof is possible.

But if they were looking for an unexpected inflation hedge, and making an unbiased choice between diversified CCFs and just speculating on the spot price of free-floating gold, I can't see why any unbiased person would choose the latter.

[NiceUnparticularMan]

In late 2019, Vanguard published a white paper on unexpected inflation in which they presented data showing that free-floating gold has been a far less efficient unexpected USD inflation hedge than diversified commodities futures. This makes sense because unlike most commodities, free-floating gold has an extremely large speculative/greater-fool component to its market price, far above its intrinsic value as a productive material. Therefore, other commodities, particularly taken as a diversified pool, are going to have a far more predictable/exploitable relationship to general inflation, whereas the large greater-fool component to gold pricing is adding a lot of noise/volatility on top of that macroeconomic signal.

Anyway, Vanguard's study has gotten a great out of sample test of rather extreme unexpected inflation recently. And sure enough, diversified commodities futures have been way out-performing gold in this period.

My point is this is not recency bias. There are fundamental theoretical/model reasons to think diversified commodities futures are a better asset for this purpose than free-floating gold. There was a comprehensive empirical study which showed that was true in the entire period of free-floating gold through 2019. There is now also a powerful out-of-sample test providing additional confirmation.

What exactly more could anyone ask on this subject? If all that is not sufficient to convince someone diversified commodities futures are a better asset for this purpose than free-floating gold, is there ANY logic or evidence that could persuade them?

Gold is not supposed to behave like commodities; it has been a fundamental store of wealth for thousands of years.

Gold has been a lot of different things over history, at different times and/or in different places.

Today, the type of gold you can buy is free-floating gold. You can't buy the other kinds available at different times in history.

The greater fool theory holds no claim when compared to fiat currencies . . . Gold loses value only if people stop valuing it.

I mean, the latter statement is just a different articulation of a greater fool theory.

Unlike with most commodities, most of the market price of gold is based not on its intrinsic value, but rather based on the hope that someday in the future when you sell it, someone will pay as much or more for it as you paid for it, despite that hoped-for price being well above its intrinsic value. If you are right, you do OK or better. If you are wrong, you do poorly. Possibly very poorly.

This has in fact happened to free-floating gold already. More than once. Sometimes it has gone up a lot in real price. Sometimes it has gone down a lot in real price. This what you would expect from a speculative/greater-fool asset.

I might note that I think the only reason to actually hold paper cash or such is for the purpose of using it to actually directly buy goods or services. I don't actually hold much paper cash as a result.

In short, you are talking about two radically different prospects. Do not compare commodities to gold as the results will certainly vary.

Well, I think it is true that a reasonable discussion of free-floating gold would focus on the fact it is a type of greater-fool asset. Calling it a commodity, although technically true, would be peripheral to such a discussion.

But people make a lot of arguments for holding gold that are not consistent with its nature today. Like, you can't actually use it to directly buy things. It is not in fact a reliable store of real value. It is not in fact a particularly good unexpected inflation hedge. And so on.

It is a greater-fool asset. It is not any of those things.

But people will promote it as all of those things. And so comparing it to things which actually do serve those other purposes can be useful. As in, if you want something you can actually use to directly buy things, cash is a good option. If you want to reliably store real value, individual TIPS (if you are a USD spender, at least) are a good option. If you want something to hedge against unexpected inflation, diversified CCFs are a good option. And so on.

But again, if you want to speculate on gold prices, then obviously speculating on gold prices is what you should do. But it isn't any of those other things.

[NiceUnparticularMan]

Gold vs US stocks & long bonds since inflation broke above 4% in April 2021

Now include a diversified CCF ETF like CMDY.

[ivgrivchuck]

So again, the above is standard economic modeling theory. Vanguard looked at data on this subject and it was consistent with all the above. That analysis then got a recent out of sample test, and the results of that out of sample test were also consistent with all of the above.

Of course if a person wanted to speculate on gold, and they requireed absolute proof of what will happen in the future to change their mind about that plan--well, of course no such absolute proof is possible.

But if they were looking for an unexpected inflation hedge, and making an unbiased choice between diversified CCFs and just speculating on the spot price of free-floating gold, I can't see why any unbiased person would choose the latter.

You raised some interesting points, thanks for that. I have no bias, I don't currently hold any gold, but I'm considering to hold a small amount in the future.

Looking at the correlations though, it seems that CCFs are at least somewhat correlated with the stocks. While gold seems to have practically no correlation to stocks:
https://www.portfoliovisualizer.com/ass ... &months=36

When holding a low-yielding high volatility asset, it's quite crucial that it's not correlated with the rest of your portfolio (otherwise you get hit in two ways: lower returns, maximum drawdowns are magnified).

Do you have opinions or views why the correlations seem to look as they are (is it just too small of a sample size)?

[000]

Gold is not much of a productive asset which is probably explains the difference in correlation.

This fundamental property is exactly why it its behavior will likely remain distinct from stocks, bonds, and industrial commodities.

Some will say this a reason not to hold gold, but usually these are the same kind of people who tell us debt instruments backed by nothing but more debt issued in a debt currency backed by nothing but more debt are a risk free asset, so one might take it with a grain of salt.

[NiceUnparticularMan]

Looking at the correlations though, it seems that CCFs are at least somewhat correlated with the stocks. While gold seems to have practically no correlation to stocks:
https://www.portfoliovisualizer.com/ass ... &months=36

So correlations between different assets tend not to be fixed, and instead tend to depend on the macroeconomic conditions of the relevant time. And, in the case of free-floating gold, whatever else might be involved in determining its speculative value.

Indeed, CCFs were much more popular here at one point when backtesting showed they apparently had negative correlations with both stocks and bonds AND stock-like returns. Here is a representative sort of analysis from that era:

https://www.nber.org/papers/w10595

Abstract:

We construct an equally-weighted index of commodity futures monthly returns over the period between July of 1959 and March of 2004 in order to study simple properties of commodity futures as an asset class. Fully-collateralized commodity futures have historically offered the same return and Sharpe ratio as equities. While the risk premium on commodity futures is essentially the same as equities, commodity futures returns are negatively correlated with equity returns and bond returns. The negative correlation between commodity futures and the other asset classes is due, in significant part, to different behavior over the business cycle. In addition, commodity futures are positively correlated with inflation, unexpected inflation, and changes in expected inflation.

Your backtest picks up right after that period, and it is quite true during that period CCFs did not at all behave like some people were hoping based on studies like that. CCF returns were much lower than stocks, and not negatively correlated.

What some people back then failed to understand, among other things, is that this synthetic CCF index was in part just reflecting that their study period included a period when inflation was unexpectedly high, and US stock and bond real returns were really bad. We haven't had a significant period like that until just recently, and both returns and correlations depend on such things.

Now if you take your test and set the beginning to March 1, 2021, which is approximately when inflation really started taking off, you will see a couple things of note:

https://www.portfoliovisualizer.com/ass ... &months=36

First, now IAU has the higher correlation with VTI, 0.25, and PCRIX is -0.07. Second, PCRIX now has a 46.79% (!) annualized return, versus 4.35% for VTI and 8.21% for IAU.

So CCFs did do what was expected of them--when the right macroeconomic conditions came around.

By the way, USERX is one of the oldest gold (mostly) funds I know of. VFINX is also one of the oldest funds approximating the US total stock market I know of. Using those instead in your test, you get back to September 1976, and the average correlation between USERX and VFINX is 0.20:

https://www.portfoliovisualizer.com/ass ... &months=36

But if you click on the rolling correlation, you get a better picture of the real story. That rolling correlation has been as low as -0.26 (in 1994), and as high as 0.65 (unfortunately, now).

Again, whether you are talking about CCFs or free-floating gold, there is no one particular answer as to how they will correlate with stocks. It all depends on macroeconomic conditions, plus in gold's case this speculative factor which is inherently unpredictable.

For the record, though, USERX has spent more time positively correlated with US stocks than negatively correlated.

When holding a low-yielding high volatility asset, it's quite crucial that it's not correlated with the rest of your portfolio (otherwise you get hit in two ways: lower returns, maximum drawdowns are magnified).

As an aside, a sub-zero return, high-volatility, uncorrelated asset generally is not expected to improve portfolio efficiency.

Indeed, if it were that easy, you could take a portion of your portfolio each month to Vegas, and put it on black on a roulette wheel.

Mathematically, that is a slight sub-zero return and high-volatility "asset", and it is totally uncorrelated to anything else!

And yet, it is a bad idea.

[NiceUnparticularMan]

Some will say this a reason not to hold gold, but usually these are the same kind of people who tell us debt instruments backed by nothing but more debt issued in a debt currency backed by nothing but more debt are a risk free asset, so one might take it with a grain of salt.

Well, debt instruments by nature are promises of payment. There might also be some security involved as well, but usually with higher-quality debt, the main source of value is just the expected reliability of the borrower in terms of keeping their promises.

There are a few reasons why most people treat the U.S. government in particular as a very reliable borrower. Obviously, one is it has never defaulted, and as a consequence has low borrowing rates, which would one think it would like to keep going if possible. Only a little less obviously, since it can simply make USD, and it promises payment in USD, under most circumstances it is fairly hard to foresee a reason why the U.S. government would need to break such a promise. Finally, it also has the power to tax in USD.

Now, if someone wants to say all that does not quite equal "risk free", then sure. Anything is theoretically possible, and you can imagine scenarios in which something actually leads to a US government default on its bonds.

But practically speaking, I think it is a reasonable conclusion that U.S. government bonds are at very low risk of default.

Of course I am not sure what this really has to do with free-floating gold. With free-floating gold at today's prices, you are just hoping that a buyer after N units of time will value it as high or higher as you are valuing it now. No one has made such a promise, though. There is nothing fundamentally useful about free-floating gold that would serve to explain that future price you are hoping to get. You are just hoping it works out that way nonetheless.

So in one case, you have an actual promise of payment, which is at least at very low risk of being broken.

And in the other case, you have . . . nothing like that at all.

[000]

Obviously, one is it has never defaulted

No, that is incorrect. That list does not include liberty bonds and a 1979 technical default.

Only a little less obviously, since it can simply make USD

Also incorrect. The banking system makes USD, not the Treasury (the Treasury can mint coins with fixed values or perhaps a platinum coin with an unspecified value). The simple proof of this fact is that they need to borrow USD in the first place. If they could simply "make" it, why wouldn't they just do that when new funds are needed?

under most circumstances it is fairly hard to foresee a reason why the U.S. government would need to break such a promise.

The reason such promises have usually been voluntarily broken is when continued inflation is perceived as being more destructive than default.

Finally, it also has the power to tax in USD.

Which would be unnecessary if they could simply make it, right?

[halfnine]

Some will say this a reason not to hold gold, but usually these are the same kind of people who tell us debt instruments backed by nothing but more debt issued in a debt currency backed by nothing but more debt are a risk free asset, so one might take it with a grain of salt.

Well, debt instruments by nature are promises of payment. There might also be some security involved as well, but usually with higher-quality debt, the main source of value is just the expected reliability of the borrower in terms of keeping their promises.

There are a few reasons why most people treat the U.S. government in particular as a very reliable borrower. Obviously, one is it has never defaulted, and as a consequence has low borrowing rates, which would one think it would like to keep going if possible. Only a little less obviously, since it can simply make USD, and it promises payment in USD, under most circumstances it is fairly hard to foresee a reason why the U.S. government would need to break such a promise. Finally, it also has the power to tax in USD.

Now, if someone wants to say all that does not quite equal "risk free", then sure. Anything is theoretically possible, and you can imagine scenarios in which something actually leads to a US government default on its bonds.

But practically speaking, I think it is a reasonable conclusion that U.S. government bonds are at very low risk of default.

Of course I am not sure what this really has to do with free-floating gold. With free-floating gold at today's prices, you are just hoping that a buyer after N units of time will value it as high or higher as you are valuing it now. No one has made such a promise, though. There is nothing fundamentally useful about free-floating gold that would serve to explain that future price you are hoping to get. You are just hoping it works out that way nonetheless.

So in one case, you have an actual promise of payment, which is at least at very low risk of being broken.

And in the other case, you have . . . nothing like that at all.

Even if one was to concede all that to be true that only addresses the needs of 5% of the world's population. Please continue.

[strummer6969]

Regardless of whether it's a haven, my problem with gold is the tax. It's considered a 'collectible', so gains are taxed at 28%. And you pay sales tax when you buy it. You could hold bullions through a self-directed IRA but it's just too much of a hassle.

[NiceUnparticularMan]

No, that is incorrect. That list does not include liberty bonds and a 1979 technical default.

I don't think either of those count as a default as normally defined, but if you want to say those are cases in which a promise from the US government was not perfectly reliable, OK.

Since I already agreed US government bonds are not risk free, just very low risk, I am not sure of the point.

Again, in contrast free-floating gold simply has no promise of future value at all.

The banking system makes USD, not the Treasury

Well, it is true that it is the Federal Reserve which determines how many USD are created.

But the Federal Reserve is part of the US government.

Finally, it also has the power to tax in USD.

Which would be unnecessary if they could simply make it, right?

No. If the US government simply created new USD in order to buy necessary real resources, that would act as financially equivalent to a tax on marginal consumers, which would be regressive.

Taxation is therefore a means by which the US government can exert control over which private entities have to give up real resources in order the the US government to have real resources to use.

[NiceUnparticularMan]

Even if one was to concede all that to be true that only addresses the needs of 5% of the world's population. Please continue.

So some people have access to government bonds in their own currency that essentially have as little default risk as US government bonds.

Other people do not, but it then becomes an interesting question whether bonds in their own currency would actually be all that great for them anyway.

I don't think there is a simple answer to that question that fits every single case.

Again, though, this all seems irrelevant to the topic at hand. I guess if you live in a place where you can actually walk into a grocery store and buy food with gold, you can view gold in those places as a form of money. If you don't live in one of those places, and you would have to convert your gold into some currency before buying food, gold is just a commodity, and you are speculating on its price.

[ivgrivchuck]

Again, whether you are talking about CCFs or free-floating gold, there is no one particular answer as to how they will correlate with stocks. It all depends on macroeconomic conditions, plus in gold's case this speculative factor which is inherently unpredictable.

For the record, though, USERX has spent more time positively correlated with US stocks than negatively correlated.

I agree that correlations are changing all the time. Macro-economic conditions absolutely play a role. Still it is kind of scary that commodities and stocks have stayed clearly correlated for the last ~13 years.

https://www.portfoliovisualizer.com/ass ... &months=36

Perhaps it's as you say and the shift in inflationary environment will change the correlation in more permanent way, we'll see... if that happens, I'll need to change my perception about commodities.

If we look at gold and stocks correlation that you posted:
https://www.portfoliovisualizer.com/ass ... &months=36

It has gone up and down like crazy, but those two assets haven't ever stayed strongly correlated for an extended period of time. Maybe this time is different, we'll see... (I would expect the correlation to fall soon, but it's hard to predict the future...)

As an aside, a sub-zero return, high-volatility, uncorrelated asset generally is not expected to improve portfolio efficiency.

Indeed, if it were that easy, you could take a portion of your portfolio each month to Vegas, and put it on black on a roulette wheel.

Mathematically, that is a slight sub-zero return and high-volatility "asset", and it is totally uncorrelated to anything else!

And yet, it is a bad idea.

In principle, I agree. However we need to make a clear distinction between 0% nominal return and 0% real return assets.

In the era of very low (even negative) real interest interest rates, the picture is less clear.

For example, let's run an example assuming that inflation is 3% and real interest rates are zero:
Stocks: 7% nominal expected return (vol: 18%)
Bonds: 3% nominal expected return (vol: 6%)
Gold: 3% nominal expected return (vol: 24%)

and assume zero correlation between all asset classes.

In this case gold gets picked up on the efficient frontier:
https://www.portfoliovisualizer.com/eff ... eight1=100

Now, I readily admit that the chosen assumptions were somewhat favorable to gold (e.g. correlation is not exactly zero), but the takeaway is that adding gold in the era of low real interest rates is unlikely to take you very far from the frontier (as long as you stay within 5%-10% of the portfolio), and there is no long term change in correlations... Is it worth it? That is hard to say.