Year of the Conflagration
feel value in gold as never before. And that value will be as the
"productive use of holding wealth thru the fire of change".
-Another (2/7/98)
Conflagration has a couple of different meanings. The first is a very intense and uncontrolled fire, which burns over a large area and destroys everything. It's the fire of change. And the second is a state of armed violent struggle between states, nations, or groups, involving a lot of people. And in this case, I'm afraid that both meanings may apply.
Solitary Monk suggested conflagration back in late 2015, for 2016. If you remember, I named 2015 the Year of the Fire, so his suggestion implied an expansion and intensification, which was a very good suggestion. But in the end, I went with 2016 - Year of the Fire Sale, writing:
Last year, I began with “The dollar is on fire right now… 90.28 at the close!” The dollar has indeed been on fire all year, and is now at 98.69. Gold was just under $1,300 last January, and today it’s on sale for only $1,062. So my title for this year, in a way, is already a fait accompli.
I picked a different name for 2024 about six months ago, but I decided to push that one to 2025 after receiving an update from my crystal ball. So, for the first time ever, I have not only already named the next year a year in advance, but I’m not even going to keep it a secret. I’m going to share it with you right now:
Year of the Denouement
In case you don't know what denouement means, here's a paragraph from my 2011 post, Deflation or Hyperinflation?:
The whole point of the deflation versus hyperinflation debate is about the denouement, the final outcome of this 100-year dollar experiment. It is about the ultimate end, and the debate has been going on ever since the 70s when the dollar was separated from gold and it became clear that there would be an end. The debate is about determining the best stance someone should take who has plenty of net worth. And I do mean PLENTY. People of modest net worth, like me, can of course participate in the debate. But then it can become confusing at times when we think about shortages or supply disruptions of necessities like food. Of course you need to look out for life's necessities first and foremost. But beyond that, there is real value to be gained by truly understanding this debate.
Normally, I can only see out into the future about a year. The difference this year that makes it so I can see out 2 years into the future is the election scheduled for November 5th, 10 months and 6 days from today. I don't know how it's going to end, just that it's going to be messy.
"Nothing is yet known as to how this will ultimately unfold; but what is known is that something will change."
"The only thing that is certain is 'uncertainty'."
"All I know for a fact is that this can’t be resolved in an orderly way, and it won’t be, period."
Conflagration and denouement are basically death and rebirth. They imply the death and rebirth of the US political and monetary systems, because the two are inseparable. And, because the US monetary system is the global monetary system, the impact will be global.
Looking back about 24 years, I can see an emergent pattern, centered around presidential election years, that began with Bush v. Gore and this guy:
You can read the rest of the post at the Speakeasy. (See the side bar to subscribe.)
Here I'm giving you a portion of Think Like a Giant 4 that pertains to GLD, from back in November. And then I'm giving you a short update regarding GLD from my New Year's post. So enjoy, and have a Happy New Year!
11/13/23:
One year ago, in November of 2022, GLD inventory hit 906 tonnes and stuck. For the preceding 6 months, from 4/20/22 until 11/4/22, GLD was drained from 1,106 tonnes down to 906 tonnes, a drain of 200 tonnes in just over 6 months. Then it hit 906 tonnes on 11/4/22, and stayed there, or at least within a relatively-tight range of 43 tonnes, between 901 and 944, for the next 9 months. That’s a range of plus or minus about 2.3% from its average during those 9 months, and that followed an 18% decline over the previous 6 months. Here’s what it looks like on a chart:
HSBC: https://www.spdrgoldshares.com/assets/dynamic/GLD/file/barlist/Barlist.pdf
JPM: https://emea-markets.jpmorgan.com/metalicsWebAppJanus/publicUnauthenticated/SPDR_GOLD_TRUST_JPM_BARLIST.pdf
One more note of interest is that this popped up on my XTL (Twitter/X timeline) just yesterday:
HSBC shifts part of US precious metal trades to UK
US subsidiary transfers all related market risk to headquarters
HSBC moved to consolidate precious metal trading under one roof in the third quarter, transferring all associated market risk, and part of its physical assets, from its US subsidiary to the non-ringfenced UK unit. HSBC USA Inc sold around $900 million worth of physical inventory to HSBC Bank Plc during the quarter, part of a decision to refer all stateside precious metal financing and trading clients to the London desk going forward. // The US subsidiary’s precious metal assets halved.
So, what does any of this mean?
Well, first of all, HSBC is a bullion bank, so "physical inventory" simply means reserves, the physical gold equivalent of physical cash in ATM machines. HSBC transferred half of its gold reserves from its US subsidiary to its LBMA headquarters in London according to that report. This fits nicely with my long-held suspicion that the LBMA has been operating without reserves for several years now.
To recap, banks need reserves for basically three reasons: 1. to meet reserve requirements, 2. for interbank clearing, and 3. for withdrawals. Bullion banks don't have physical bullion reserve requirements on the bullion side of their business, so they don't need physical reserves for #1. As for clearing, the LBMA has created the LPMCL (London Precious Metals Clearing Limited), a separate corporation consisting of HSBC, JPMorgan, UBS and ICBC Standard, for clearing.
ICBC, while it is both a clearing member of the LPMCL and a vaulting custodian for the LBMA, has recently stopped participating in LBMA gold price auctions, raising the obvious question of whether or not it's even still involved in the LPMCL:
Why are the 3 big Chinese banks no longer participating in the LBMA Gold Price auctions?
— BullionStar (@BullionStar) October 4, 2023
ICBC, Bank of Communications and Bank of China.
Did they get booted out by the LBMA or did they scarper once they realised the pricing is rigged?
Enquiring minds would like to know. pic.twitter.com/QEWI7paqKX
It's also the bank that got hacked last week:
World's Biggest Bank Has to Trade Via USB Stick After Hack
— Oumar Sissoko (@Oumar_Sissoko_1) November 10, 2023
Nov 10, 2023
Incident caused ICBC’s 🇨🇳clients to reroute some Treasury trades
‘A true shock to banks around the world,’
Trades handled by world’s largest bank in globe’s biggest market traversed Manhattan 🇺🇸on aUSB stick pic.twitter.com/RcU0WQEO3S
And UBS, while it's a clearing member of the LPMCL, is not a vaulting custodian of the LBMA. So it's really just JPMorgan and HSBC doing the physical clearing if there is such a thing, which there isn't by the way. The rest of the LBMA's reserves are unallocated liabilities of the clearing banks, and clearing is all done on ledgers. So they don't need physical reserves for #2.
That leaves us with #3, withdrawals, or allocation and delivery as it's called in bullion banking. That's the only reason the LBMA needs physical reserves, and in most cases, they simply don't offer allocation or delivery, and that's how they've been operating without reserves for at least several years.
I know I've explained it many times before, but reserves are the slack in the flow of gold flowing through the LBMA system of gold industry financing, from mining and recycling to refining, minting, fabricating and wholesaling. And GLD is where that slack now sits. It wasn't designed that way, but GLD, which began in 2004, turned out to be a convenient "coat-check room" for any slack in the flow.
Physical gold is worse than a dead asset to bullion banks, because there's a "carrying cost" (storage and security) associated with it. And GLD gave the bullion banks a place to park that slack in the flow, and let unwitting GLD shareholders pay the carrying cost of the gold. It was a win-win for the banks, because they could put it in when it wasn't needed, and pull it out when it was needed for delivery or allocation.
It's easy to think that there's something nefarious going on with the bullion banks using GLD in this way, but there's not. It's just the way things worked out, and it only applies to GLD, not any other gold ETFs, silver ETFs, SLV, or any other ETFs, period. Most ETFs are managed to some extent in order to track the price of the underlying asset. That means the inventory of underlying assets generally correlates with the price, and sometimes we see this with GLD, but not always.
No one debates me on this coat-check analogy anymore, but I know that many of you simply dismiss it out of hand. You shouldn't, though, because it's true. It's the right way to view GLD, which is different from all other ETF products simply because it's the LBMA's gold ETF, and because gold is different than anything else, even silver.
The clearest statistical proof that coat-check is right (because logical reasoning alone is not enough proof for some people, especially technical analysts) is in this post. If coat-check were wrong, which as I said is the case with SLV, you'd expect GLD and SLV to behave similarly during a given timeframe of any length. In that post, we look at the timeframe just preceding the big drop in April of 2013, specifically the first quarter of the year, from January 1 until April 13th, 2013, the day before the price collapsed.
During that timeframe, the gold price dropped from $1,693 to $1,535, a mild 9% decline compared to what followed. The silver price also dropped, from $30.87 to $27.40, also a 9% drop. But the GLD and SLV inventories moved in opposite directions during that time. SLV added 366 tonnes, while GLD lost 205 tonnes.
Now, you might be thinking that GLD did what you'd expect it to do during a price decline, it lost inventory, and SLV did the unexpected by gaining inventory during a price decline. But that's not all. There's an even more convincing period we looked at in that post:
As shown in Chart 2, from February 1st through September 6, 2011 (151 trading days) the price of the GLD increased from $129.92 to $184.49 or by 42%, while there was essentially no net creation of ETF shares to purchase the underlying physical gold. There were 404 million shares outstanding on February 1st and on September 6th, there were 406.8 million shares outstanding; an increase of less than 1% (virtually no net change in the amount of gold in the pot). This was the big run-up in gold, to its all-time high. You remember. It went from around $1,330 on February 1st up to $1,895 on September 6.
Over the last few years, the gold market developed with a two tier nature. Entities that had massive oil to supply a needing world would most certainly receive their gold in "allocated" form if they asked for it. Whether it be from private investors who exchanged their physical for holding paper derivatives or from the vaults of Major CBs, this gold would come from somewhere, "come hell or high water". The modern financing tool we call the "gold carry trade" is now becoming the poison that will kill this market. The demands of gold lenders to return their "at risk" positions are creating an atmosphere where no amount of physical gold exists that can supply the outstanding paper claims. Great blocks of gold are now lent into the markets at 4% or greater, where once 1% was considered a good return. As each new group of lenders enter the market they are followed close behind by former lenders demanding their gold return. Fear begins to grip those who were once bullion owners as they now became paper pawns. Each new demand for "full allocation" creates yet further demands to borrow. The supply of new lenders grows smaller and smaller as the possibility of default increases. -FOA (10/6/99) Gold Carry Trade
[Central] Banks do lend gold with a reason to control price. If gold rises above its commodity price it loses value in discount trade. They admit now to lending much where they would admit nothing before! They do this now because of the trouble ahead. Does a CB have collateral to lend its gold? Understand, they only lend their good name on paper, not the gold itself. The gold that is put on the market in these deals belongs to someone else! The world private stockpiles that could be sold have been. The CBs are heavy into their own stuff now and are over their heads if they had to make good on all the private deals. How DO they do it?
The demands of gold lenders to return their "at risk" positions are creating an atmosphere where no amount of physical gold exists that can supply the outstanding paper claims. Great blocks of gold are now lent into the markets at 4% or greater, where once 1% was considered a good return. As each new group of lenders enter the market they are followed close behind by former lenders demanding their gold return. -FOA (10/6/99) We have seen the last of cheap oil in US$ as the oil states are no longer taking paper gold! I think a large purchase of bullion was just made by them. It should have been paper. The BIS must soon take a stand! -ANOTHER (11/30/97) It was never the intent of the CBs to sell their countries gold in massive amounts. The "understanding" that was worked out years ago was good for the economies and the world. In return for the US$ remaining the "oil reserve" currency, ( oil would be not just supplied but supplied in dollars ) large amounts of gold would be supplied far into the future. The gold, while indirectly backed by the CBs would actually come from the mines of the future. With the oil money making a ready market for gold priced at a premium ( contangoed out many years ) , the mines could make a fair profit even with spot gold priced below production. All would win! And for some time, we did! I am able to know some CBs, they are not evil, their minds are for the best that can occur. But, I THINK the world ran away from them.
-ANOTHER (11/13/97) The Asians are the problem, by buying up bullion worldwide and thru South Africa they created a default situation on all the paper for the oil / gold trade! Now the CBs are selling in the open to calm nerves but it's known that they will never sell enough. It was never their intent to provide the gold, only the backing until new mining technology could increase production. Over time the forward sales, such as ABX's should have worked. But LBMA went nuts with the game and the whole mess has now accelerated. -ANOTHER (10/19/97) The commodity value of gold was forced so low in paper currency terms that all of the new mined gold, going out some 10 years is spoken for. Between the third world buying physical gold and the jewelry industry ( same people buying ) there is none left for the oil states! -ANOTHER (10/9/97) The BIS and other various governments that developed this trade ( notice I didn't use conspiracy as it was good business, as the world gained a lot ) , thought that the paper gold forward market would have allowed the gold industry to expand production some five times over! Don't ask where they got this, as they are the same people that bring us government finance and such. But, without a major increase in gold supply, the paper created by this "gold control operation" will either be paid by, 1. new supply. 2. the central banks. 3. rollover existing. 4. cash? 5. or total default! As the Asians started buying up everything last year ( 97 ) , number 5 and 5 started looking like the answer! When the CBs started selling into this black hole of demand, the discussion of #5 started in their rooms also. Trading of gold, on and off official markets dwarfs not only new mine supply but is even larger than all existing stocks. LBMA, trades a million ounces a day! And that is only what is seen. -FOA (9/29/98)
Yup, that's a 42% rise in the price of "gold" as well, so GLD did a fantastic job of tracking it. And the fact that there was virtually no change in inventory during this remarkable run should put to bed once and for all the idea that the arb is physical. A net total of 5 tonnes was added, actually less than half a percent.
There are various schools of thought on how the mythical physical arb might function, but this period alone should sufficiently debunk all of them. And the only alternative left is that creations and redemptions are a choice made by the banks, not forced upon them by investor sentiment or inflows. And if they're a choice (which they are), then "coat check" is the only view that makes sense.
[…]
For those who still don't quite get it, the argument is basically this: The consensus, the settled science, almost anyone you ask, even GLD itself, says that GLD tracks the POG through a physical arbitrage carried out by the APs, the Authorized Participants or bullion banks that act as market makers for the ETF. If net-interest in GLD is higher than in physical, the APs buy physical and put it in GLD, creating new shares. If net-interest in GLD is lower than in the underlying asset, the APs redeem baskets of shares and sell the physical, profiting the difference.
Under this view, whether gold is put into or taken out of the ETF is driven by the public's demand for GLD shares relative to gold, its underlying asset. Coat-check, on the other hand, shows that the arbitrage is all paper, unallocated gold, not physical, and therefore GLD tracks the POG not because the APs run the arb, but simply because the market keeps it tracking. The arb that keeps GLD tracking the POG is open to anybody, and that's why it tracks it so well.
The corollary is that the creation and redemption of shares (putting physical gold into or taking it out of GLD) is purely a matter of choice by the bullion banks running the show. GLD is going to track the POG no matter how much gold is in there, and no matter what the price action is doing, because the arb is between unallocated XAU/USD and GLD shares, and is therefore open to anyone with a computer.
Taking it one step further, the creation and redemption of shares (putting physical gold into or taking it out of GLD) is purely a matter of choice, not by just any LBMA bullion bank, but ultimately by only the custodian, which from 2004 until 11 months ago, was one sole bank, HSBC. Today it is two banks (more on this line of thought later).
The reason the custodian(s) has all the power, is because it's his responsibility to source the actual bar numbers that get allocated to the ETF. Say, for example, Goldman Sachs, an AP and LBMA member but not a London vault or custodian, is expecting a London delivery of some (unwanted) physical. It is unwanted because GS is going to have to pay to store it at HSBC or JPM. In any case, the physical gets delivered to HSBC's vault and credited to GS as unallocated in its interbank account with HSBC.
GS may owe some of that unallocated to someone else to clear some trades, or it can choose to exchange it for either dollar cash or GLD shares, either through another bank, with the public, or with HSBC itself. HSBC doesn't care what GS chooses, but GS's choice does not affect HSBC's choice of what to do with the physical. GS is not going to pay storage costs either way, so HSBC will place the bars wherever someone else is paying for the storage, either in GLD if they're not needed elsewhere, or allocated or delivered elsewhere if needed.
The reverse is true as well. HSBC can pull gold out of GLD at will when it's needed elsewhere, for delivery or allocation. This is how and why the slack in the flow ebbs and flows over time. When the flow is tight, we see GLD draining. And when there's gold coming in (or not flowing out), we see GLD expanding. The last time it expanded to any significant degree was in 2020, during COVID. That was most-likely due to physical outflows being blocked by global lockdowns (it was not flowing out).
This is important to understand when looking at the larger trend which I outlined above. Remember, the bullion banks do not need (or want) physical bars for anything other than deliveries and allocation. So, when we see the big drain of 200 tonnes from 4/20/22 to 11/4/22, we can suppose that someone or multiple someones of some clout was/were requesting (I guess the clout makes it more of a demand than a request) either delivery or allocation.
Then we see it stop at 906 tonnes, and sit there for a month until they bring in a 2nd custodian, and the outflow is cut off until the two custodians have basically split the physical 50/50, and then the drain resumes. The drain not only resumes on August 4th, barely three months ago, but HSBC also transfers half of its reserves held in the US over to London. So, just in the last quarter, we see another 38 tonnes gone from GLD, and another 14 tonnes brought over from the US by HSBC, presumably for delivery or allocation, for a total of about 52 tonnes, halfway back to the drain-rate of 2022. Looks like someone, or several someones of some import, have been demanding (and taking) delivery.
-FOA (10/6/99)
The important thing to note in that paragraph is that he's talking about the two tier nature of the 1999 gold market not in terms of two prices, but in terms of the top tier being those entities who would receive physical allocation (or delivery) if and when they asked for it, and that it would be sourced one way or another, "come hell or high water."
One way the gold would be sourced would be to "borrow" it from someone who had allocated physical. FOA goes on to talk about the lending of gold via the gold carry trade, which was popular at that time:
For those who don't understand the "gold carry trade," here's a brief explanation from Nasdaq.com:
A carry trade where you borrow and pay interest in order to buy something else that has higher interest. The gold carry trade works as follows. A central bank loans a bank (sometimes called a bullion bank) some gold. The gold lease rate is usually very low. The bullion bank immediately sells the gold and invests in securities with a higher rate of return, such as government long-term bonds. The carry return is the return on the bonds minus the gold lease rate. However, this trade is risky on two dimensions. First, if the bullion bank invested in long-term bonds and the interest rate goes up, the trade could be unprofitable. More seriously, the bullion bank has effectively sold the gold short. If the loan is called by the central bank and if gold has risen in value, the bullion bank will have to go into the market and purchase higher priced gold. Indeed, if many banks are short, the unwinding of the gold carry trade could drive the gold price even higher.
An important thing to remember here is that the CBs did not lend actual physical gold. They only lent it on paper, and the physical gold, if necessary, came from someone else, either borrowed or purchased from some other entity, or from the mines.
-ANOTHER (11/16/97)
-ANOTHER (10/19/97)
It's more complicated than this but here is a close explanation. In the beginning the CBs didn't sell their own gold. They ( thru third party ) found someone else who had bullion. That "party" sold to a broker who sold forward for a mine or speculator or government ) . In the end the 3rd party had the backing from the broker that he had backing from the CB to supply physical if needed to put out a fire. The CB held a very private note from the broker as insurance and was paid a small fee. This process mobilized free standing bullion outside the government stockpiles. The world currency gold price was kept down as large existing physical stockpiles were replaced by notes of future delivery from the merchant banks ( and anyone else who wanted to play ) .
This whole game was not lost on some very large buyers WHO WANTED GOLD BUT DIDN'T WANT ITS MOVEMENT TO BE SEEN! Why not move a little closer to the action by offering cash directly to the broker/bank ( to be lent out ) in return for a future gold note that was indirectly backed by the CBs. That "paper gold" was just like gold in the bank. The CBs liked it because no one had to move gold and it took BIG buying power off the market that would have gunned the price! It also worked well as a vehicle to cycle oil wealth for gold as a complete paper deal.
Are you with me?
Well a funny thing happened right after the Gulf war ended. What looked like big money before turned out to be little money as some HK people, I'll call them "Big Trader" for short, moved in and started buying all the notes and physical the market offered. The rub was that they only bought low, and lower and cheaper. They never ran the price and they never ran out of money. Seeing this, some people ( middle east ) started to exchange their existing paper gold for the real stuff.
-ANOTHER (10/12/97)
I know this one is a little confusing. But don't worry, I'll simplify it for you.
When he says brokers, he's talking about the bullion banks of the LBMA. Remember, there's no shortage of gold. All the gold ever mined is still out there, owned by somebody. You just don't know who those owners are. So, the idea here in the mid-90s was to "mobilize" some of those large, non-government, non-CB stockpiles of physical to supply physical demand to those top-tier entities who would receive physical allocation (or delivery) if and when they asked for it, which would be sourced one way or another, "come hell or high water."
When he says third party, or just party, he's talking about the person or entity that has a large stockpile of physical. And when he says that party "sold" the gold to the broker (the bullion bank), he really means lent, in a repurchase kind of way. It was "sold" with, essentially, a repurchase agreement, because he then says that the 3rd party (the original owner of the gold) had assurances from the broker (the bullion bank) that it had assurances from the CB that CB physical would be made available in extremis (ie., to put out a fire).
In essence, the CB was lending its gold to the bullion bank on paper (the "very private note" he mentions was that promissory note from the bullion bank to the CB, for which the bullion bank paid a small rate of interest, 1% or lower, for the guarantee that CB gold would be made available if necessary). A sale with a repurchase agreement is essentially a loan. The interest is worked into the repurchase price, so the seller of the gold is assured that he will be able to buy it back, whenever he wants, at the market price minus a discount representing the interest. Either that, or he's paid a premium up front representing the interest.
In any case, this was 1997, and the price of gold had been declining for more than a decade. So, if you had a very large private stockpile of gold, and you had a guarantee from a CB that CB gold would be used to keep the price from rising (ie., to put out a fire), then it made sense to sell your gold for cash to buy Treasuries that paid interest, and then buy the gold back later when the price started rising. So, there may or may not have been a formal repurchase agreement, but these are the people that FOA was talking about in 1999 when he wrote:
The "parties" whose "free standing bullion outside the government stockpiles" was mobilized and "replaced by notes of future delivery from the merchant banks (ie., the LBMA bullion banks)" that Another was talking about in October of 1997, were the "former gold lenders" involved in the "gold carry trade," "demanding their gold return," that FOA was talking about two years later, in October of 1999.
The rest of the quote from Another above is about how it then evolved into a purely paper market, where gold traders would buy those same CB-guaranteed "notes of future delivery from the merchant banks (ie., the LBMA bullion banks)" with cash rather than "free standing bullion stockpiles," and the CBs were fine with it because "it took BIG buying power off the [physical] market," and it "also worked well as a vehicle to cycle oil wealth for gold as a complete paper deal."
The problems started in the mid-90s, when "Big Trader" from Hong Kong "started buying all the notes and physical the market offered." "He" was buying the CB-guaranteed notes AND the physical. Seeing this, others who had been sitting on CB-guaranteed notes (ie., the Saudis), started demanding allocation and delivery, and they were the "entities who would receive physical allocation (or delivery) if and when they asked for it, which would be sourced one way or another, 'come hell or high water.'"
That was the basic situation when Another showed up on the scene, warning of the possible collapse of the gold market. The CBs that came up with this deal thought that the "notes of future delivery" they were guaranteeing with their own gold would be fulfilled by new gold coming from the mines:
-ANOTHER (2/16/98)
The problem back then, the amount of physical needed at the time, was about 14,000 tonnes according to Another. Since then, some 70,000 tonnes have been added to the above-ground supply, Barrick (ABX) retired its forward sales hedge book, and the pressing issues of the time (1997-1999) were overcome.
My point of walking you through all of this is because I think that what we see happening today with GLD draining, and HSBC moving reserves from New York to London, is that some top tier entities, who are not necessarily the same ones as 25 years ago, are demanding (and receiving) allocation and/or delivery.
It is my suspicion that the LBMA does not have any slack in the flow other than what's in GLD, and it uses GLD as its reserve pool. I have also deduced that most if not all of what's in GLD must already be spoken for, or at least its ultimate distribution is under the control of some group of central bankers (I'm not using the term BIS here because it triggers some of you, and all you can think of is the woke fat guy 😉).
South Africa is now part of the BRICS, who have essentially splintered off from the half of the world that contains the LBMA. China, also part of the BRICS, is buying gold directly from Africa, which was part of the problem in the 90s as it diverted LBMA inflow. So, there's not only no slack in the flow of gold passing through the LBMA's gold industry financing network, but the outflow is significantly higher than the inflow, as can be seen in the above-mentioned draining and movements.
The initial 200 tonne drain began in April of 2022, less than two months after the entire West sanctioned Russia and anyone who did business with Russia (essentially sanctioning the BRICS), and froze Russian assets. The timing could be related.
There was absolutely no obvious need for a 2nd custodian. It wasn't like they needed more vault space; GLD was getting smaller, not bigger. And HSBC had apparently done a fine job as the sole custodian for 18 years.
I later speculated that perhaps JPMorgan was added to location-swap GLD inventory with its vaults in Switzerland and New York, but that hasn't happened. There is no GLD gold in Switzerland, and the amount in JPMorgan's New York vault is the same as it was back in May, 22.8 tonnes, which is only 2.6% of GLD.
What happened was that the drain resumed right around the time that JPMorgan and HSBC hit 50/50 parity in terms of GLD holdings, which happened around July 12, 2023, about a month before the BRICS summit in South Africa. Perhaps the timing is related.
All interbank GLD share creation and redemption transactions between HSBC and all of the other APs, including JPMorgan before it became a custodian, were transacted in unallocated gold, meaning gold credits, or "gold of the mind." If GS, for example, wanted to redeem some GLD shares, whether on behalf of itself or a client doesn't matter, then it would receive unallocated credits from HSBC. There is nothing that says that HSBC has to deallocate bars. HSBC could simply take the shares and issue credits to Goldman Sachs. This probably happened many times, when one bank would be redeeming while another was creating, HSBC would simply manage the flow.
Goldman Sachs, if it wanted physical gold, whether for itself or its client doesn't matter, could then request allocation or delivery. But again, HSBC would be under no GLD-related obligation to grant the request. There's nothing wrong with this, it's just the way GLD and the LBMA operate, and understanding the way they operate in extremis, helps us understand the actions they might take when facing extremis.
During normal times, I'm sure this all worked just fine. It was HSBC's job as custodian of GLD to source physical when needed, and to provide physical when requested. As GLD gradually became the entirety of the LBMA's physical gold reserves, as everyone eventually learned that parking any slack in the flow in GLD was the most efficient way to store reserves, HSBC ended up with a tremendous amount of power with regard to controlling the slack in the flow.
In extremis, having a single entity in charge of managing the LBMA's reserves, that entity can unilaterally decide where those bars go. But once you add a second entity, neither one can make unilateral decisions. So, I now think that JPMorgan was brought in to eliminate the risk of a single entity unilaterally giving away the rest of the gold.
I don't think anyone at either bank even knows this stuff. They're just doing their jobs. As I've said many times, the banks don't even own this gold, and banks aren't individuals. And most people who work at banks don't even like gold, haven't even heard of Freegold, and, if I'm being honest, aren't very creative or curious. Perhaps they're smart, greedy, and willing to break a rule here or there for a buck, but they aren't thinking like us about how unallocated differs from physical, and its inevitable revaluation.
Now, as I've said, not anyone can just go buy physical gold from a bullion bank. It's just not what they do. They'll tell you to go to a gold broker or a coin shop, depending on how much you want to buy. But if you happen to have $65M, you could probably buy a pallet of gold bars, and you might even be able to take delivery from HSBC's vault, although it's pretty unlikely.
That said, there are individuals and entities that have much more than $65M which they want to protect with the timeless wealth asset known for revaluing whenever monetary and financial systems collapse, and new ones rise from the ashes. And I have no problem imagining that, following the seizure of state and private Russian assets in March of 2022, that the LBMA was hit up for whatever gold it could cough up to individuals and entities with that kind of money, and I'm talking billions, not millions.
All such requests, if not outright turned down due to Russian-sounding surnames, would eventually make their way to HSBC, the sole LBMA member capable of unilaterally deciding whether or not to honor a physical delivery request. And having never had to actually turn one down before (even though they could have), because that's just the way it worked, I can easily see them unwittingly draining 200 tonnes from GLD before someone stepped in and stopped them.
Again, the drain lasted from April 20, 2022 until November 4, 2022, a little more than 6 months, with GLD draining at an average rate of about 30 tonnes a month, or 1 tonne per day. Then it stopped, held steady for a month and a weekend, and voila, GLD suddenly had a 2nd custodian.
Then, no more drain until the two custodians held equal amounts, and now the drain is on again, only this time a little bit slower, as if it's being managed by a third party.
And that's why I now think that JPMorgan was added as a GLD custodian. There were other theories, but after 11 months, none of them panned out. And with 11 months of hindsight, now, I think this explanation makes perfect sense.
Whether the gold was already spoken for, previously purchased, or is being sold now to top tier parties at 2nd tier prices is beyond the purview of this speculation, at this time. But it appears that distribution is ongoing, rather than just waiting for the inevitable termination event.
[There is much more to this post at the Speakeasy, as well as 525 comments under it. But this was the part about GLD.]
And here's the New Year's Eve update:
12/30/23:
GLD is currently at 879 tonnes. There's not much to be gleaned from that number alone, but it is still consistent with what I wrote back in November, in Think Like a Giant 4, and there's an interesting new trend emerging. It's up about 11 tonnes overall since 11/11/23. But if I look at HSBC and JPM individually, HSBC is up 28 tonnes, while JPM is down 14 tonnes since 11/11, for a net increase in GLD of 14 tonnes. So, why don't the HSBC and JPM bar lists match the GLD spreadsheet (14t vs. 11t)?
Remember, the bar lists that the custodians put out are two or three days behind the daily spreadsheet that GLD puts out. When shares are created or redeemed, it happens on paper, using unallocated or paper gold, and GLD reports it that day. Then, the custodians have three days to either source the new bars, or decide which bars to deliver. So what they report each day actually matches what GLD reported two or three days earlier.
So, if we go back two days, we see that GLD had 882 tonnes on Wednesday, which matches the custodian bar lists, with an increase of 14 tonnes since 11/11/23. This is kind of interesting since we know that HSBC transferred about 14 tonnes from its US subsidiary to its London vault.
So, 14 tonnes (net) were added to GLD, and it appears that it was HSBC's turn to source that gold. Remember, back on 11/11, even though they were pretty balanced, HSBC had about 53 tonnes less than JPMorgan.
So, for convenience, here are the HSBC bar list and the JPMorgan bar list from 12/29/23. And here are the live links that get updated every business day:
HSBC: https://www.spdrgoldshares.com/assets/dynamic/GLD/file/barlist/Barlist.pdf
JPM: https://emea-markets.jpmorgan.com/metalicsWebAppJanus/publicUnauthenticated/SPDR_GOLD_TRUST_JPM_BARLIST.pdf
We'll check again on Tuesday and Wednesday, and if both withdrawals came out of JPMorgan's vault, then it looks like someone is trying to make them even-steven.
I'll also note that the amount of GLD inventory in JPMorgan's New York vault remains the same as it has been for the past 7 months, 22.83 tonnes. This raises the question, in my mind at least, whether HSBC actually physically moved 14 tonnes from New York to London, or if it was only on paper. And if it was only on paper, the question remains, are those 14 tonnes in New York or London? Because even though HSBC USA Inc. "owned" that gold, and sold it to HSBC Bank Plc. London last fall, it could have been in London all along. Or it could still be in New York.
We don't know what any of this means, yet, but it's interesting to watch, since it doesn't quite seem random that HSBC now holds 49.4% of GLD's inventory, and JPM has 50.6%. And if the next 2.6 tonnes come out of JPM, it'll be 49.6% and 50.4%. So if there's a third party managing GLD now (and it would be the BIS acting through the BOE), which I speculated in Think Like a Giant 4, then it seems to want equal amounts in the two banks for some unknown reason.
Happy New Year, everyone!!!
Sincerely,
FOFOA
