Friday, June 1, 2012
GLD Talk Continued
As many of you know from the comments here and on Twitter, Victor The Cleaner just completed a great new post called GLD – The Central Bank Of The Bullion Banks. The timing was pretty neat. Lance Lewis' GLD Puke Indicator delivered a buy signal on 5/22 and Victor and I have been emailing extensively ever since that day discussing my view of the GLD Pukes.
During our discussion Victor conducted an analysis of the Puke Indicator using a hypothetical trading strategy based on buying spot gold at the puke and selling once the gold in GLD is replenished and found that it optimized at a puke size of about .5% of the inventory. Using 250,000 ounces as the puke size, he found that the $PoG climbed four times as much (annualized) in only one-third of the time (between puke and replenishment) as it did during the remaining two-thirds of the time. This was a significant finding which, at the very least, showed that the Indicator really does work. Using Lance's 1% threshold the trading strategy was a little less optimal, but perhaps Lance's higher threshold is more immediately predictive of big moves like today. I don't know.
But that's some timing, huh? Vic finally got his post up late last night and then today we have a jumbo up-move of more than 4%! At the very least I think it demands a little bit of attention.
Costata really enjoyed Victor's post and he emailed me with a few comments about it. I felt that Costata had maybe missed some of what I thought was an important thrust in the post and so I tried to summarize in one short email a few of the things that Vic and I had discussed over nine days of long emails. Not everything we discussed made it into Victor's post. Victor is meticulous in his exposition while I often try to cover too much ground, requiring me to just touch on some things which, if it was Victor writing, would require a toy model, some characters with alphabetical names, and a fancy chart or three.
Anyway, Costata suggested that I should post my email and JR concurred, so here you go, with minimal polish:
Costata: "It also occurred to me just now that if this analysis is correct, it makes the possibility of a run on the unallocated accounts with the LBMA clearing members even more remote. The "large buyer" at the LBMA that VTC speculates about appears to be extremely disciplined - the BIS perhaps.
Thoughts?"
Me: No, VtC divided it into two possible theories. Large buyer who knows the bottoms in the markets is one theory, an unlikely one. The second theory he called “speculative interpretation” is the real message.
If the BBs are redeeming GLD shares to fulfill allocation demands, that implies that they do not have enough 400 oz. bars outside of GLD to fulfill those demands. And that also implies that the BBs are using GLD shares as reserves.
Obviously 400 oz. bars come and go. They come in from mines, scrap and hapless investors and they go out to allocation demands and deliveries. That’s the flow. So when we see a GLD puke, we infer they were essentially out of 400 oz. bars at the time. There is no other reason for a BB to redeem GLD shares, even if it is performing the arbitrage. GLD shares, for all practical purposes, are as good as 400 oz. bars not in GLD from the perspective of a BB. As shares, the BB reserves can even be lent at interest to those who want to short GLD. In fact, the BBs could potentially have only GLD shares as their reserves, which is why Randall Strauss called GLD a “central coat-check room” for the BBs. The pukes suggest to me this may be the case.
That flow from the mines to investors is the flow. Remember when there’s not enough supply in that flow is when the stock to flow ratio explodes toward infinity. With this view, we can infer this may be what is happening with each puke.
We can’t really look at the size of GLD and the size and frequency of the pukes and extrapolate a timeline. If you look at Lance’s chart, the size of GLD peaked in 2010. If we suddenly see a puke of, say, 10% of the GLD inventory, I’d say it’s game over starting there. The largest puke so far was about 4% over two consecutive days last August. That was about 50 tonnes when the PoG was around $1,800. That was HUGE. Almost $3B.
So when the pukes happen, that is BB heart attack time, but then the price rises and eventually (so far) the puke gets replenished. So as the price rises, the reserves are stretched and so is the inflow. As Victor said, the inflow of gold from the mines is relatively constant by weight but the outflow is normally constant in currency terms. So they raise the price until the puke is replenished and then they stop. That’s the message in the post.
How they raise it, which he didn’t go into, is a little more interesting. From that LBMA survey, we can see that the LBMA had net sales in one quarter of 7,575 tonnes of paper gold. That’s a gross increase in the amount of paper gold in existence over only three months. 100:1 actually seems conservative in this light. That’s most likely FOREX use of gold as a hedge or a currency play. But even still, the BBs have to hedge their price exposure when selling that much paper gold. Without a hedge, that would be a 7,575 tonne naked short position for the BBs.
So that net increase in paper gold is also a net inflow of cash for the BBs, cash which they use to hedge that net exposure. In fact, we can see from the LBMA survey exactly how much cash it was. It was $338B. That’s over 3 months, so it’s more like $5.4B per day inflow. That’s a small percent considering the daily turnover in paper gold used as a FOREX currency is $240B and the daily turnover of all currencies is $4T. So in a $4T/day FOREX market, that’s a $5.4B/day net flow from other currencies into gold. That was $5.4B per day in Q1 2011 that needed to be hedged by the BBs.
There’s no way they hedged all of that in the “gold” market (Comex/mining forwards/GLD). It’s simply not big enough to absorb that rate of flow without rising a lot faster than we saw it rise. So the BBs must be hedging this exposure the way they hedge net positions against other currencies in the FOREX market, simply using complex formulas and derivatives that look at correlations between different things. Correlations change slower than raw price changes which (they think) gives them time to adjust their hedges if the correlations start to exceed the model parameters.
Anyway, that is a plausible way they are hedging their exposure to the price of gold without doing so in the “gold” market per se. But they can also hedge some of that exposure in the gold market as well, by going long gold on the Comex or some other way. So if they want the price to rise in order to stretch the physical side and (hopefully) replenish the GLD puke, they would simply shift some hedges from complex derivatives into Comex.
So even though they have some control over the price of gold, they are still relying on other market players from the physical side to respond as expected. And from the view of GLD as their reserve pool, we can see that reserves are not only quite finite, but they also peaked almost two years ago.
A/FOA said the ECB/BIS strategy was to “expand and support” the dollar paper gold market so the dollar would eventually “bankrupt itself” just to keep the gold market going and stay in the game with the euro.
FOA (08/13/01; 07:24:30MT - usagold.com msg#96)
A very large part of that war strategy, employed by the ECB/BIS, was to let the dollar / IMF faction hang themselves by expanding and supporting the whole arena of this dollar paper gold market [the ECB/BIS is supporting and expanding paper gold as a strategy]. Inflating the gold market place with so much "paper gold" that we would eventually have to bankrupt ourselves just to keep the dollar in the war game against the Euro.
[…]
So, don't count on this destruction of our paper gold market to mark the real value and availability of physical gold; that ratio will split somewhere down the goldtrail. This action will scare most harden gold investors to death; especially the ones in leveraged gold stocks and lesser white metals!
The war between gold and the dollar has been over for a while now. The action, today, is between the dollar and the euro arena and this is what will break the price lock on gold. Leaving gold bugs with a lot of questions that ask why this: both systems will strive for a higher currency price for gold; one doing it because they have to; the other doing it because they want to! The casualty on this battlefield will be the world gold market as we know it. A market caught between how Western perception thinks gold's price should be "discovered" and at what price level trading in physical gold craters the entire paper structure. A structure of American based "paper gold".
We have been saying for some time that this will be "the" show to watch unfold; but only if your holdings allow you to stay still in your seat as it happens (smile).
They shifted their war on gold to become a war on the Euro,,,, only too late. Now, knowing that the Euro is a fact, we must have a super gold price if the dollar is to stay in the game! The question becomes one of supporting a cheap paper price for the sole function of keeping the market and all its bullion players alive. With the war on gold over, they need to turn their tanks around to face the real enemy but cannot.
So it seems that as the war switched from dollar v. gold to dollar v. euro, the euro side helped make the dollar gold market TBTF. But with a rising physical gold price/demand, the dollar paper gold market has to keep up because it’s TBTF now. Too many of those “gold” FDIC stickers out there! If those stickers fail, the dollar loses. So the “gold” market is TBTF. Remember this from FOA?
FOA (10/9/01; 10:05:48MT - usagold.com msg#117)
What doesn't seem to be obvious is the "why for" the paper market grew so large. It grew to dominate because worldwide dollar expansion reached its "non-hedged" peak. In other words, the dollar's timeline was ending as its ability to produce non price inflationary economic gains came into sight.
In order to push dollar holdings further, international players needed and purchased "paper financial hedges" to balance their risk. Within their total mix of derivative hedges were found "paper gold price hedges"; modern gold derivatives. The important thing to remember is that these positions are not and never will be used to demand physical gold. They are held to buffer financial and currency risk associated with holding any form of dollar based asset. To work these items don't need to really perform "dollar price movements" in the holders favor as much as they are present in the portfolio to act as insurance stickers.
In that truth, these paper gold positions act like FDIC insurance at our banks. It can and will manage only a small determined portion of bank runs,,,,, not a full scale failure of the banking system. In a real full banking failure we would all get, perhaps, 80% of our covered $100,000 and 10% of the rest.
The same is true for these gold position's performance; real gold delivery along with true price performance, matching real bullion trading, would be only for the very few. For that matter, an actual functioning paper gold marketplace would be for the very few, too! But, in the same way a bank account owner understands the credibility of FDIC insurance when times are good; the international dollar asset owner will not grasp that modern paper gold hedges cannot be allowed to work until after a real serious price inflationary run begins.
For the first time in this portion of the dollar's timeline and our lifetimes, such an inflation is about to show its face!
So the paper gold of the bullion banks is now TBTF. Of course that doesn’t mean it can’t fail. It either fails, or the USG hyperinflates the dollar as prices rise. They are related, and each will likely cause the other almost immediately, but either one could end up being the initial cause IMO. If price inflation forces the USG to hyperinflate then the paper gold insurance stickers will have to fail to perform. And if these price rises in the gold market fail to manage the flow (demand) of physical as they have so far, we’ll likely see a 10% or larger GLD puke at some point. That would signify more than a 120 tonne allocation demand, a system-busting size. They might think they can rocket the price at that point and get it back, but more likely we’ll see more allocation requests coincident with a falling (paper) "gold" price as the longs dump their worthless “insurance” while wishing they had the real thing.
FOA (06/12/00; 19:48:25MT - usagold.com msg#26)
Put your cards on the table!
The current paper gold world will die (burn) as its value to users erodes, not increases!
…Again, most everyone in the Western Gold bug game is running with the ball in the wrong direction.
…So who is in danger of being hurt as this unfolds?
That's right, the Western paper gold long! I'm not talking about just the US market! This is about the entire world gold market as we know it today. The real play will be for the ones that get out in front of the move by owning physical…
It seems every Gold bug sees only half the trade and has great faith that contract law will favor a short squeeze. Yet, none of them see where it is the long that will be dumping and forcing the discount!
As I have said in the past, gold is so oversubscribed through the BB's paper gold it's more of a wonder when the $PoG rises than when it falls. Perhaps now we have a plausible explanation for why and how it has been rising over a decade, and also how it will end.
Sincerely,
FOFOA
PS. I realize there's a ton of stuff I only touched on here. I hope that Victor will grace us with his presence and his meticulous Thoughts. ;)
"Effect And Cause"
I guess you have to have a problem
If you want to invent a contraption
First you cause a train wreck
Then they put me in traction
Well first came an action
And then a reaction
But you can't switch around
For your own satisfaction
Well you burnt my house down
Then got mad at my reaction
Well in every complicated situation
There's a human relation
Making sense of it all
Takes a whole lot of concentration
Well you can blame the baby
For her pregnant ma
And if there's one of these unavoidable laws
It's just that you can't just take the effect and make it the cause
[Chorus:]
Well you can't take the effect
And make it the cause
I didn't rob a bank
Because you made up the law
Blame me for robbing Peter
But don't you blame Paul
Can't take the effect
And make it the cause
I ain't the reason that you gave me
No reason to return your call
You built a house of cards
And got shocked when you saw them fall
Well I ain't saying I'm innocent
In fact the reverse
But if your heading to the grave
You don't blame the hearse
You're like a little girl yelling at her brother
Cause you lost his ball
You keep blaming me for what you did
But that ain't all
The way you clean up a wreck
Is enough to give one pause
You seem to forget just how this song started
I'm reacting to you
Because you left me broken hearted
See you just can't just take the effect and make it the cause
[Chorus]
Subscribe to:
Post Comments (Atom)
404 comments:
«Oldest ‹Older 401 – 404 of 404holdinmyown,
I can't understand how any modern day "master of the universe" could be so stupid as to believe that a synthetic derivative construct could be used to hedge fat tail risks.
... I thought it was called tail risk precisely because it wasn't hedged exactly ...
No, seriously, there are plenty of precedents for fooling around with correlations. An expensive one I remember was Morgan Stanley in early 2008. They had figured out as early as 2007 that there would be defaults in the market for subprime mortgages, and they wanted to hedge their existing exposure.
So what they did was to 'hedge' it by betting on an increase of the spread between the lower and the higher tranches of some CDOs, i.e. they bought CDSs on the lower tranches and sold CDSs on the upper tranches in such a way that their aggregate position was flat (at that time). Very clever, isn't it?
So how did it end? Well, in order to be hedged at that time, they had sold a lot more notional CDSs on the high tranches than theu had bought on a lower tranches. Eventually, not only the lower tranches of the CDOs got wiped out, but rather the entire CDOs including the AAA tranches.
There was a week in spring 2008 during which Morgan Stanley, first thing in the morning, received a phone call from their counterparty (Deutsche) requesting additional collateral to the tune of one *billion*. Every morning. I think they lost some 12-13 billion on that trade, most of it to Deutsche. Lippman, the person in charge on the other side at Deutsche later commented that he had the impression that MS had never understood what had happened to them.
Victor
I'd finally like to draw your attention to some comments on my GLD article:
Baltic Mist: I chased down the typical costs to borrow GLD – seem around 0.10% p.a. …much cheaper than GOFO. I.e., a bullion bank neding physical would prefer to borrow GLD and redeem than borrow gold on LBMA
VtC: I think you should compare the borrowing costs of GLD shares with the Gold Lease Rate (GLR), i.e. LIBOR-GOFO rather than with GOFO. This is because you only borrow GLD shares, but you do not lend the other party dollars at the same time. GOFO is lending gold and borrowing dollars at the same time.
In any case, this figure is interesting. If you can borrow more than 2.5 million shares (threshold of the trading strategy) at this rate of 0.1% p.a., this is probably the cheapest way of borrowing gold. For a comparison, Kitco quotes a GLR of 0.5% for a 1-year term. (There is a difference though: your borrowing costs of GLD shares are probably subject to a recall by the lender at any time whereas the rate quoted by Kitco is fixed-term for the full year).
Why is it so much cheaper to borrow GLD rather than physical gold itself? Perhaps many investment funds have GLD, but view it only as a paper position, do not care about counterparty risk and are therefore happy to lend it out?
Baltic Mist: Sorry..I meant the implied gold interest rate from GOFO of course.
I think you are exactly right…GLD is considered by many pension fund type holders as just another security to be lent out to get fee income in. The loan will be collateralised (with some other of paper or cash) so is perceived to have no risk. Holders of GLD have no intention (or no capacity – was talking to a fund manager today who simply isn’t allowed to hold physical gold) to take delivery of physical. Only the bullion banks are playing across the two markets (LBMA and GLD).
Risk of recall is generally trivial… at least until the level of borrow gets to extreme levels (say 25% of all securities) and I doubt the first to have the securities recalled would be the bullion banks themselves (their clients come first I would think).
So cheap and easy borrow in very large amounts.
Tightness in GLD borrow (interest rate > GLR; GLR rising) would therefore be a very important signal that the game is close to over (and may well come before the final massive gold puke)
VtC: I agree, that’s all very plausible. Funny that in this scenario there are some who borrow stock not in order to sell it short, but rather in order to get to the gold.
Victor
@Milamber,
You got it. Your synopsis should be in the next thread, right up at the top, where it will get the reads it deserves. Consider reposting it there.
FWIW I can't see the USG mobilising their gold unprompted... hard to do much with your head between your knees and tears in your eyes.
byiamBYoung
Thanks
and VtC
Thanks for that tale about MS and Deutch. :)
Post a Comment
Comments are set on moderate, so they may or may not get through.