This recent post by me on the Mint's corporate blog is the simplest way I can think of to explain the Mint's version of unallocated. Note that there are only so many "pipes", which will become clear in a few month's time.
28 February 2011
06 February 2011
Gold lenders of last resort
Another post by FOFOA that will get you thinking. My response below.
FOFOA,
There would be many within the Mint who would be amused at you categorising me as a “mainstream” view. I understand you are using my explanations as representative of the mainstream view but I would like readers to be clear that my personal view is different. To clarify this, some comments on your piece.
While I have no direct evidence of the bullion bank's (BB) activities, I am not as sure as others that the BBs are massively financially short gold (this is not to say they don't run short term speculative positions). My reasoning for this is that gains and losses on such positions impact their reported profit and loss. If they were as massive short for as long as some claim, their losses would have been visible.
I would also suggest readers ask why a BB would take on hundreds and hundreds of tonnes of short gold positions over time in some attempt to suppress the gold price. You only do this if you have a philosophical hatred of gold. I understand that gold ownership is a political action, a rejection of fiat currency and banking, but would (at this time) a bank with a BB division really be threatened by the pathetic fraction of a percent of those investors who hold gold? Threatened to the extent that they would of their own accord take on a massive short position?
Now the above does not mean I think everything is OK. On my fractional fubar post you mentioned, I commented “ It troubles me as well. The Mint has been under no illusions about London unallocated as the legals say we are an unsecured creditor and the bullion banks would never make any statement one way or another about what they did with it. We have operated accordingly.”
Bankers make money by intermediating – buying from one, selling to another, borrowing from one, lending to another – and taking their cut along the way. I would suggest readers consider the theory that BBs would be willing to intermediate for someone else with that philosophical hatred of gold and take their riskless cut along the way. Why risk your own money when someone else is willing to do so, with the bonus that their activity protects your banking “franchise”?
This then leads on to your statement that “there is no clearly defined lender of last resort to cover the risks”. Is this really the case? You mention two risks the BBs have.
1. Default – Borrowing gold doesn't solve this problem, as the act of borrowing gives you an gold asset but also a gold liability. The only way to solve this problem is to buy the gold, which results in a loss because you have to give up dollars to acquire the gold asset.
2. Liquidity – Buying gold doesn't solve this problem as while it gives you gold to give to your creditor but also gives you price exposure as you have technically bought your gold asset which is due in the future. The solution is to borrow gold directly, repaying it when your gold asset comes due. Alternative, you can borrow synthetically by buy spot gold and then selling forward (using the gold from your gold asset to deliver into this forward sale).
I would therefore agree that BBs have "exchange rate risk" for the default situation but not for the liquidity situation. This assumes that holders of gold (which in cases of large volume really just means central banks) are willing to sell (in case of default) or lend ( in case of liquidity) to BBs. I therefore suggest the question is not whether central banks are lenders of last resort, but whether they have the capacity, or willingness, to fulfil that role now or in the future.
In my previous post I stated that central bankers are the gold market's lenders of last resort. The fact that central banks hold gold as a physical asset (and only gold) in addition to fiat currencies is clear indication to me that gold is not just another commodity. However, the other side of this is that central banks can be lenders of last resort of this “money”, just as they are of dollars.
Central banks have been more than willing to lend dollars to banks to help them out with their liquidity problems, eg taking on their crappy mortgages etc, rather then have them fail and to avoid a systematic collapse of the banking system.
Consider the situation where a bank comes to its central bank and say “Hey, I've got all these pesky unallocated gold holders wanting physical but all I have is these long term loans. Can you lend me some of your physical gold and I'll replace it later when those borrowers repay their leases? If you don't I'll have to declare bankruptcy, the gold price will rocket up, this will cascade through the gold market and we will have a systematic collapse of the banking system.”
Why would a central bank not be willing to support a bank's BB division in such a situation, especially when they would do the same for dollars? For me this is not the issue, I think they will do (are doing?) it.
You mention the CBGA as proof that (some) central banks are “are no longer going to be the lender of last resort to this system”. I think it is therefore very interesting that the 2009 statement makes no reference to leasing as the previous two statements did. Why the change?
To me then the key issue is whether the central banks have the capacity”.The interesting thing about capacity in respect of gold of course is that you can't print it! Easy to do if your bank has a dollars problem, not so if they have a gold problem. Questions to consider:
a. How much gold would central banks be willing to lend to prop up banks? All of it? Or would they balk in the case of gold? Who cares about dollars, just print more – but risk the country's only real asset?
b. Out of the total they are willing to lend, how much has already been lent?
A speculation: maybe the reference to no more leasing in the 1999 and 2004 CBGA statements was a message to the BB to clean up their books. However, around 2008-09 the banks said they will fail without the backstop, need more time to unwind, have increasing physical redemptions, so CBGA drops the leasing reference to enable them to continue the "extend and pretend that there is not a run on the bullion bank reserves."
In conclusion, I would like to suggest the following in respect of the two risks
1. Default – This is most likely to happen if the BB lend to a short seller who is now bust. In this case we should see buying and thus an increase in the gold price.
2. Liquidity – As agreed, this will happen if unallocated holders are calling for physical. In this case we should see an increase in the lease rate.
Note that in the past the lease rate was around 1% to 2% with low gold prices, during gold's bear market. This was because of the large amount of miner forward selling that was going on. What we have seen in recent times with miners closing down their hedging is low lease rates and high gold prices (see this post for a chart). So there is a very general relationship between short selling and lease rates over the long term.
What would be interesting would be sustained increasing gold prices AND higher lease rates, as it may indicate buying to cover defaults and borrowing to cover liquidity.
FOFOA,
There would be many within the Mint who would be amused at you categorising me as a “mainstream” view. I understand you are using my explanations as representative of the mainstream view but I would like readers to be clear that my personal view is different. To clarify this, some comments on your piece.
While I have no direct evidence of the bullion bank's (BB) activities, I am not as sure as others that the BBs are massively financially short gold (this is not to say they don't run short term speculative positions). My reasoning for this is that gains and losses on such positions impact their reported profit and loss. If they were as massive short for as long as some claim, their losses would have been visible.
I would also suggest readers ask why a BB would take on hundreds and hundreds of tonnes of short gold positions over time in some attempt to suppress the gold price. You only do this if you have a philosophical hatred of gold. I understand that gold ownership is a political action, a rejection of fiat currency and banking, but would (at this time) a bank with a BB division really be threatened by the pathetic fraction of a percent of those investors who hold gold? Threatened to the extent that they would of their own accord take on a massive short position?
Now the above does not mean I think everything is OK. On my fractional fubar post you mentioned, I commented “ It troubles me as well. The Mint has been under no illusions about London unallocated as the legals say we are an unsecured creditor and the bullion banks would never make any statement one way or another about what they did with it. We have operated accordingly.”
Bankers make money by intermediating – buying from one, selling to another, borrowing from one, lending to another – and taking their cut along the way. I would suggest readers consider the theory that BBs would be willing to intermediate for someone else with that philosophical hatred of gold and take their riskless cut along the way. Why risk your own money when someone else is willing to do so, with the bonus that their activity protects your banking “franchise”?
This then leads on to your statement that “there is no clearly defined lender of last resort to cover the risks”. Is this really the case? You mention two risks the BBs have.
1. Default – Borrowing gold doesn't solve this problem, as the act of borrowing gives you an gold asset but also a gold liability. The only way to solve this problem is to buy the gold, which results in a loss because you have to give up dollars to acquire the gold asset.
2. Liquidity – Buying gold doesn't solve this problem as while it gives you gold to give to your creditor but also gives you price exposure as you have technically bought your gold asset which is due in the future. The solution is to borrow gold directly, repaying it when your gold asset comes due. Alternative, you can borrow synthetically by buy spot gold and then selling forward (using the gold from your gold asset to deliver into this forward sale).
I would therefore agree that BBs have "exchange rate risk" for the default situation but not for the liquidity situation. This assumes that holders of gold (which in cases of large volume really just means central banks) are willing to sell (in case of default) or lend ( in case of liquidity) to BBs. I therefore suggest the question is not whether central banks are lenders of last resort, but whether they have the capacity, or willingness, to fulfil that role now or in the future.
In my previous post I stated that central bankers are the gold market's lenders of last resort. The fact that central banks hold gold as a physical asset (and only gold) in addition to fiat currencies is clear indication to me that gold is not just another commodity. However, the other side of this is that central banks can be lenders of last resort of this “money”, just as they are of dollars.
Central banks have been more than willing to lend dollars to banks to help them out with their liquidity problems, eg taking on their crappy mortgages etc, rather then have them fail and to avoid a systematic collapse of the banking system.
Consider the situation where a bank comes to its central bank and say “Hey, I've got all these pesky unallocated gold holders wanting physical but all I have is these long term loans. Can you lend me some of your physical gold and I'll replace it later when those borrowers repay their leases? If you don't I'll have to declare bankruptcy, the gold price will rocket up, this will cascade through the gold market and we will have a systematic collapse of the banking system.”
Why would a central bank not be willing to support a bank's BB division in such a situation, especially when they would do the same for dollars? For me this is not the issue, I think they will do (are doing?) it.
You mention the CBGA as proof that (some) central banks are “are no longer going to be the lender of last resort to this system”. I think it is therefore very interesting that the 2009 statement makes no reference to leasing as the previous two statements did. Why the change?
To me then the key issue is whether the central banks have the capacity”.The interesting thing about capacity in respect of gold of course is that you can't print it! Easy to do if your bank has a dollars problem, not so if they have a gold problem. Questions to consider:
a. How much gold would central banks be willing to lend to prop up banks? All of it? Or would they balk in the case of gold? Who cares about dollars, just print more – but risk the country's only real asset?
b. Out of the total they are willing to lend, how much has already been lent?
A speculation: maybe the reference to no more leasing in the 1999 and 2004 CBGA statements was a message to the BB to clean up their books. However, around 2008-09 the banks said they will fail without the backstop, need more time to unwind, have increasing physical redemptions, so CBGA drops the leasing reference to enable them to continue the "extend and pretend that there is not a run on the bullion bank reserves."
In conclusion, I would like to suggest the following in respect of the two risks
1. Default – This is most likely to happen if the BB lend to a short seller who is now bust. In this case we should see buying and thus an increase in the gold price.
2. Liquidity – As agreed, this will happen if unallocated holders are calling for physical. In this case we should see an increase in the lease rate.
Note that in the past the lease rate was around 1% to 2% with low gold prices, during gold's bear market. This was because of the large amount of miner forward selling that was going on. What we have seen in recent times with miners closing down their hedging is low lease rates and high gold prices (see this post for a chart). So there is a very general relationship between short selling and lease rates over the long term.
What would be interesting would be sustained increasing gold prices AND higher lease rates, as it may indicate buying to cover defaults and borrowing to cover liquidity.
01 February 2011
Further discussions with FOFOA on GLD
FOFOA,
Firstly, I'll have to be more careful in how I write. My post was really mixing up responding to specific quotes of yours but then veering off on to related concepts/positions that are not yours. My exploration of the idea that APs would fraudulently take GLD gold or “GLD is bad because bullion banks involved” was directed at the simplistic anti-GLD ranters not looking to the subtleties and not at yourself. One of the problems with writing rather than speaking face to face I think. Anyway, on to the discussion.
“Market-price reversible swap” makes more sense, I read “essentially lent” as implying some obligation to return the physical. With regards to the “naked short” I was talking from a financial point of view, whereas you are using the term in the sense of physical.
To clarify the distinction for our readers, let us consider a bullion bank with a physical ounce asset backing an unallocated ounce liability to its clients. If that bullion bank then lends that physical to a jewellery company who use it in their operations, then the bullion bank now has an ounce claim asset backing it unallocated ounce liability. From your point they are short “physical” but I would also note that the bullion bank is not short “financially”, that is they are not exposed to any movement in the price of gold.
Yes they are exposed to the risk the jeweller does not return the physical at the end of the lease. Probably more importantly, they are exposed to liquidity risk. I think this is the sense that you use “short” and is reflecting the issue of “maturity transformation” (see Unqualified Reservations blog for an excellent explanation of why this is a big problem).
My use of the word “short” is for situations where the bullion bank exchanges (or sells) the physical backing its unallocated ounce liabilities for cash. This creates a financial risk as there is a mismatch between the denominations of the liability (ounces) and the asset (dollars). When you used the phrase “sell them for dollars that can then chase an ROI” this implied to me a financial short and that was what I was addressing.
I now understand what you meant by "special right" when you say “once the price of physical gold starts running away from the paper price”. I will have to disagree with you on this to some extent. Now by that I'm not saying GLD does not have its risks or that any not-in-your-hands gold is better than in-your-hands gold, but I have, maybe naively, a stronger belief in arbitrage and greed.
Let us consider your scenario where the markets have been closed for a week, during which no doubt the price for physical gold has risen. On market opening I agree we are likely to see much selling by retail investors who no longer have any trust in the markets. They are wanting cash so they can buy physical gold. Their selling pushes the price of GLD down.
Now you state that “the APs can just scoop up those shares at a panic discount”. This I'm not so sure about. The prospectus lists 16 APs, only some of which are actually bullion banks (with their angry) unallocated creditors):
BMO Capital Markets, CIBC World Markets, Citigroup Global Markets, Credit Suisse Securities, Deutsche Bank Securities, EWT, Goldman, Sachs, Goldman Sachs Execution & Clearing, HSBC Securities, J.P. Morgan Securities, Merrill Lynch Professional Clearing, Morgan Stanley, Newedge, RBC Capital Markets, Scotia Capital and UBS Securities.
I find it hard to believe that all of these will conspire to agree to hold off on buying GLD until a significant discount appears. Arbitrage traders in each firm will be watching GLD drop relative to the physical price of gold. As it goes to $1 to $2 discount per ounce etc, the traders will be thinking “if I don't take that discount and lock in easy profit now then one of the other APs will and I'll lose the profit”. With 16 traders I find it hard to believe that one will not jump first, providing offers to buy and thus arresting the decline in GLD's price. What does Newedge care about JP Morgan's angry unallocated customers and why let them get GLD gold at a big discount to save them and deny yourself a profit?
Now I will concede that for my scenario to work all of the 16 APs have to have access to physical in the OTC market, which may not be the case for the smaller players. But then when you say the physical price is diverging from GLD's price, this implies that there is market for physical at a price and thus would it not be more easier for the 16 APs to acquire physical compared to retail investors, given their connections in the OTC market?
As you say we are talking about systemic failure. I suppose I'm nit picking, but is not systemic failure a situation where gold goes into Feketian “hiding” in which case there ceases to be a gold price? What I'm saying is that up until that parabolic breaking point, while gold is still being sold for cash, the backstabbing greedy profit motive of the 16 APs will ensure GLD's price stays in touch with the physical gold price. That is my answer to your question “Will anything other than physical gold itself track the price of physical gold in a physical-only market?”
For readers who don't find this particularly helpful or are not comfortable assessing these risks, I would suggest taking FOFOA's advice:“I don't know the answers but I do know one way to avoid the risks.” - that is, buy physical!
The ability of non-APs to borrow GLD shares and then sell them short I think we are in agreement on and is another problem with GLD, or to be fair with stock exchanges in general it seems.
Finally, I take some issue with your statement that “or some other coin the ETF shareholders would have bought had there not been an ETF. The ETF diverted demand in many ways”.I partly disagree with this, but I also partly disagree with those who think GLD's tonnage is “additional” demand. The truth is in between both in my opinion.
There is no doubt that a fair portion of investment in GLD would have occurred anyway, either into other funds (eg Central Trust), Allocated account or cash and carry coins and bars. In this sense all GLD does is make this investment more visible than it would have been. Unfortunately commentators obsess about GLD simply because it is visible and ignore the other 28,000t or so of privately held gold (not to mention Asian demand in “jewellery”, which is really investment in nature).
However, I do believe that the creation of stock exchange listed gold products has increased demand for gold by making it easier to get exposure to gold. Buying it through a stock broker eliminated the perceived inconvenience to some investors of having to go down to a coin shop and then worry about where to store gold.
As to the WGC, in my dealings with them I don't agree with your view that “they are focused on all aspects of the gold market, including the structural integrity of the Bullion Banks' fractional reserves given that the CBs have removed their physical backstop.” They are a miner trade group. Their focus was on physical offtake and thus obsessed about the metal behind GLD being Allocated gold. Funny when you consider that the legal structure introduced, in my opinion, some holes that negated the “security” of the Allocated gold backing.
I can't say anymore except that I'd guess I'm one step closer to them than yourself (note: the first exchange traded gold product in the world was the Australian Gold Bullion Securities, the second was the Perth Mint's ASX code:PMGOLD, the WGC naturally took some interest in these Aussie upstarts). Unless of course you are close to the WGC, but then that would be revealing a bit too much? :)
Subscribe to:
Posts (Atom)