11 April 2010

London unallocated: Fractional Fubar or Benevolent Banking?

A number of readers of my blog have asked me to comment on the 100:1 comment by Mr Christian of CPM Group in his CFTC testimony and whether London unallocated metal accounts are fractional. Well short answer is no, they are only 10:1 fractional. Do you feel much better now?

The 10:1 statement was made by Mr Christian in a April 10 interview with Jim Puplava of Financial Sense. Mr Christian is interviewed at the 26 minute mark and explains his 100:1 statement at the 36 minute mark. However, it is the comments at the 44 minute mark that are most illuminating, which I have transcribed below:

If you are a bank in the United States and you take in a deposit the office of the controller of the currency says you have a reserve requirement of 12.5% or something which means that for every dollar you take in deposits you can lend out 8 and that's how the money in banking 101 works.

Now if you're a bank in the United States and you take in gold and silver deposits not on an allocated basis but on an unallocated basis the same way you take in dollars when you put them in – when people put money into a savings account or a chequing account that's an unallocated account and the bank is allowed to lend it out. If they put the money in their safety deposit box that money belongs to the investor and the bank can't lend it, the bank can't hypothecate it, it stays there, and it means nothing to the money in circulation.

In the gold market if you put your gold and silver in a safety deposit box or an allocated account the bank can't touch it legally but if you put it in an unallocated account that is now an asset on the bank's book, they have a liability to give it to you if you ever want it back but in the meantime they can lend it out. Now if you give the bank in the United states money the law, the office of the controller of the currency says the bank can lend it out 8 times. If you give it gold and silver the office of the controller of the currency says the bank can lend it out in a “prudent fashion” and the bank has the discretion to decide what's a prudent multiple for its credit lending. Most of the banks I know, commercial banks, 8, 10, maybe 12 as a leverage factor.

AIG was not a bank, was not a commercial bank, and under the US laws non-commercial banks don't come under the law, the guidance of the office of the controller of the currency. AIG used a leverage factor of 40, so if people gave them a million ounces of gold to hold for them, they could lend out 40. I mean, I have friends who are metals traders who were looking for job years ago and, you know, they went to AIG and AIG said “we use a leverage factor of 40” and the trader is a seasoned guy and he's worked at major banks and investment banks, he said “I can't operate at that level of leverage its just too risky more me” and AIG trading said “well this is what we do”, right, so there is a loophole in our regulatory system, its doesn't really have anything to do with gold and silver per se but it allows non-banks to participate in banking activities in a way that skirts banking regulations that are designed to promote stability in the banking system.

In the interview, Mr Christian recommended that listeners go to the CPM Group website where there was a free download Bullion Banking Explained. I took him up on the offer. Below are are some extracts that fill out his statements above.

This article may help to clarify the complex world of commodity banking, in which gold, silver, and other commodities are treated as assets, collateralized and traded against. When we explain these processes to clients, we often refer to the same mechanics as they are applied to deposits, loans, and assets by commercial banks in U.S. dollars and other currencies. Banks treat their metal deposits in much the same way as they do deposits denominated in money, as the reserve asset against which they lend additional money to borrowers. ...

Many banks use factor loadings of 5 to 10 for their gold and silver, meaning that they will loan or sell 5 to 10 times as much metal as they have either purchased or committed to buy. One dealer we know uses a leverage factor of 40. (Long Term Capital Management had a leverage factor of 100 when it nearly collapsed in 1998.)

A bank does not even have to be buying gold at a particular time to be able to use it as collateral against which it can trade, sell forward, and lend gold. If a bank has gold held in an unallocated account, or a forward purchase on its books committing a producer to sell it gold later, it can use these gold assets as collateral for additional gold trades.

Is London unallocated fractional fubar or just benevolent banking? Maybe this statement by Mr Christian in a presentation to the International Cotton Advisory Council in October 2002 will help you decide:

A producer should use an advisor such as CPM Group, which is not trading against the producer. Banks and dealers have a conflict of interest between their own trading positions and the hedges they advise their clients to take.

Or maybe Recent Lessons Learned About Hedging (January 2000):

Hedgers should not rely on their trading counterparts for hedging strategies. These entities take the opposite side of the hedge transactions, have inherent conflicts of interest, and always keep their own best interests in mind, even if these are the short-term best interests and arguably not in the banks’ own long term best interests.


  1. Bron,

    Didn't we establish in "The King of Currencies"that London is not fraction? But now it is? But only 10:1 not 100:1?
    Of have I missed your point?

  2. Jayson,
    Can you point me to the "King of Currencies" article? I can't find it. Thanks.

  3. Gordon, link is http://goldchat.blogspot.com/2009/10/king-of-currencies.html

    Jayson, in that post I was disputing the reliance on turnover as proof of fractional, not necesarily that I believed it was or wasn't fractional - at that time I had no direct proof one way or the other.

    For example, you may have 1oz physical with 10oz of unallocated balances, but if every one of those 10oz paper gold holders are "buy and hold" then the turnover will be very low. You could also have a 100% backed system but all composed of speculators, day trading like crazy and thus really high turnover figures.

  4. "the same mechanics as they are applied to deposits, loans, and assets by commercial banks in U.S. dollars and other currencies. Banks treat their metal deposits in much the same way as they do deposits denominated in money"

    In other words; "the deposits may well not be there. The banks might be holding the 'hedges' of various gold miners that produce little or no gold".

    I think the question is pertinent; are these 'bullion' banks liquid?

  5. That is certainly an important question but the other one is whether the holders of unallocated are going to test that liquidity by taking delivery?

  6. Indeed.

    Another pertinent question would be; how much of the RBA's 'including gold on loan' (as described by the IMF), is actually the 'hedges' of various gold miners that produce little or no gold?, which by the way seems to be a fair portion of the ASX.

    Somebody is backstopping the gold banking market. What's the chances of central banks ask for delivery?

  7. Bron,

    I have been reading your blog for a while. I really appreciated the great series you did on the manufacturing and business processes of Mints and the chain of supply. It really helped to cut through the hysteria about retail shortages that was prevalent at the time. Thanks.

    This latest post of yours troubles me. The notion that the banks of Western developed nations such as the US, UK and Australia hold fractional reserves of their deposits is just plain wrong. If anything, it is an accounting fiction. In Australia, for example, there are no formal reserve requirements. The system is based on a metric termed "capital adequacy".

    Happy to provide as many links to supporting evidence as you wish. Links to papers from Deutsche Bank, BIS and other "reputable" sources only if you so stipulate. Christian's understanding of the bank "reserve" system seems to be frozen in the era (long ago) when he worked for Goldman Sachs.

  8. Thanks costata. 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.

    To have it confirmed however at 10:1 somehow crystalises the exposure in your mind.

    Does "capital adequacy" work on a currency by currency basis, or can you have some AUD reserves backing a USD exposure? If you have a link explaining the rules that would be helpful on understanding what the potential risks are.

  9. Justin,

    RBA I think is at least open that it has leased out "our" gold, compared to some central banks who, it seems, just combine physical and receivables into one asset line.

    Because central banks would be doing everything with bullion banks, I doubt they would have any idea where/what their lease was used for.

    Doubtful they will ever ask for physical delivery, especially if it would cause a problem because then they would just have to support the bank with cash. Of course if you have lent it out to banks in other countries, then it is that country's central bank who has to clean up the mess, so maybe some will!

  10. Bron,

    Thank you once again for helping us understand what is going on. Also, thank you for the link at Kitco to the Perth historical data in Excel files.

    This is off-topic, but I am wondering if you can direct me to some information regarding the gold content of the $200AUS gold coins (Diana/Charles comm., Koalas) from 1980 onward.

    I believe I read somewhere that from 1980-1986, both the UNC and Proofs were .2948oz total gold content. But I think it changed around 1987 in that the UNC started having much less. Is there a link to such information regarding the total content? Thank you in advance!

  11. I'm not sure, this http://www.silverstackers.com/calculators/index.php?page=6 has it at .2948oz but maybe best to post your question at that forum as it is Aussie based

  12. Bron,

    "Does "capital adequacy" work on a currency by currency basis, or can you have some AUD reserves backing a USD exposure?"

    Re-reading my earlier comment I feel I expressed myself badly. It might help to approach the concepts of reserves and capital adequacy separately under a single currency scenario. Otherwise the conditionality explodes.

    Under a traditional fractional reserve system a percentage of the bank's deposits are held in "cash" in "reserve" against its loan book. At least some "cash" is available on demand. This is a relatively simple system.

    Provided the bank's loan book is sound, the main day-to-day risk is a temporary liquidity problem. A Central Bank can solve that type of problem with a short term loan against the bank's assets.

    Under the capital adequacy regime loans are "risk weighted" and an "adequate" amount of capital is required to be held against each class of loan. The quality of the capital in each "Tier" is rated 1, 2 etc. If you have enough capital you can run with virtually zero cash reserves.

    A few issues arise; What can you treat as capital? Is the "quality" assessed accurately? Do the risk weightings reflect reality?

    Under the fractional reserve system the cash may not be sufficient but it is either there or it is not. Under the capital adequacy regime the inherent complexity is vastly greater. With loose regulations and a pliant regulator there are also ample opportunities for some "innovative" accounting.

    If debt is accepted as "capital" counter-party risk becomes the threat du jour. "Capital adequacy" can vanish in an instant.

    If that basic, over-simplified description makes sense I can compile a list of links that discuss various aspects of the system. Unfortunately I haven't found a single source that lays it all out.


  13. I think the main point is not being addressed. Fractional reserves works for fiat because the bank, ANY bank can just create more fiat with the click of a few keys.
    To extend that concept to comodities is fraud. No one can eat paper oats, you can't make a car out of paper steel, you can't run your car on paper gas. gold is fabricated and gold is saved. A person decides to save gold to get out of the fiat ponzi approach to money. Paper gold is a fraud.

  14. Great final comment there costata.

    And scruffy makes a similar blunt point, that leads me this question:

    What if there is a 'run' on gold, or a significant request (globally) for physical delivery of gold?


  15. Pete,

    A run on gold (and silver?) may already be in progress at the Comex. I don't know if this could have broader implications.

    For a rapidfire analysis of the Comex PM action I go to Harvey Organ's blog. (He was one of the guys who gave evidence at the CFTC hearings.)





    BTW our next door neighbour has a Scottish terrier named "Scruffy" so this exchange of comments is kind of surreal for me.

  16. Bron,

    Totally off topic and really outside your Blog. But as someone in a management position at the PM I am using this source to try to get some action.

    This year I have spent over $60,000 in direct orders to the Perth Mint and - including orders through my local coin dealer - another $10,000.

    My last order, for a 5oz bar and 2 1oz coins was made on March 30 this year. It was supposedly sent to me by AAE on April 6 and delivered to Sydney on April 7. Since then the tracker keeps showing re-delivery to Sydney but no delivery and no progress.

    I emailed Perth Mint who told me to phone AAE. This I did and after about 30 min getting through I was promised a return call or email which never arrived. After more correspondence by email with the Perth Mint, and another worthless call to AAE still nothing has happened.

    In my last communication with the Perth Mint I was told that this had been escalated to the Security Team and that investigations would need to be conducted before a replacement was sent. I was absolutely re-assured that this would be dealt with by Monday (today) and that I would be informed.

    It is now 9.35PM EAT and I have sent 2 emails to the Perth Mint with absolutely no response.

    This is no way to do business.

    I guess my broader point is that if you cannot even trust the Perth Mint to conduct a normal business of delivery, how would anyone trust them with an unallocated account. After this, I certainly would not.

  17. basle 3 determines capital requirements

  18. Gordon, can you email me with your problem and I will get it addressed, email is bron.suchecki at perthmint.com.au Sometimes parcels get mislaid and don't want to send out a replacement and the original also gets delivered, but even so your emails should be answered.

    In terms of a "run" on gold, it does matter if this is a retail run or wholesale (ie 400oz/1000oz bars). If the latter then there is more capability of the market to withstand it but if it is retail, then the minting industry as a whole (and even small bar manufacture at refineries) is not able to meet mass market demand. Premiums will rise and rationing will occur, even if there is plenty of raw wholesale gold out there.

    Interestingly, the best time for a squeeze may be right now while there are no flights in/out of the UK, as this is a key centre of the gold market - they can't get gold out and into COMEX so best time to take delivery?

  19. Hi Bron,

    I appreciate your well thought out and rational take on all matters gold-related. There seems to a debate between the CEO of BullionVault Paul Tustain and GATA Board Member Adrian Douglas. I'm trying to decipher some of the nuanced arguments and was wondering if you could provide your take. Here is the link:


    In particular, Paul suggests the large value/number of LMBA gold forwards are sold by miners/producers who have the physical gold in their posession albeit in the ground. Adrian seems to be using Paul's comments to suggest that it is not the miners selling the forwards but the bullion banks, i.e. the folks who don't have the physical to actually sell forward. Something to do with miners not hedging and forwards being illiquid. It was kind of hard for me to follow so I was hoping you could help clarify. Thanks!


  20. I've been in Singapore, hence no response. I've left another comment on those articles to the post following this one.

    Mr Tustain's article is hard to follow unless you are right up to speed on the basis as discussed at the Gold Standard Institute sessions, or in other words the arbitrage between spot and futures.

    In fact what Mr Tustain is talking about is arbitraging between spot and futures, spot and forwards and futures and forwards. Needless to say it can get complex but my short answer is:

    1. Mr Tustain does not discuss the counterparty risk (futures and forwards are promises to do something) involved in the transactions.
    2. Mr Douglas does not discuss the legitimate role of arbitraging and paints the whole market as bogus.

    Both are I think emphasising certain aspects of the market to push their point, which is what can make it confusing. In some way they are both right and both wrong.

    Trying to untangle it is a major article I'm not sure I have time for at the moment. I'll have a look at both articles and see if I can add anything useful.

  21. Thanks Bron,

    I think I need to brush up on the concept of "hedging" and "arbitraging".

    In the meantime I believe I understand the concept of counterparty risk with respect to when miners sell gold forward but it seems to me that this risk is considerably lower than the risk a bullion bank makes when it sells gold forward because, according to Adrian, the bullion banks do not have the gold to sell. At least the miners have something in the ground or do they????

    Do you see a dinstinction in comparable counterparty risk when a miner sells forward vs when a bullion bank does?

  22. I don't think one can make a simple rule that bullion bank (BB) forwards are more risky than those done by miners (or jewellers, for that matter).

    If you look at Sons of Gwalia example you can see a situation where they did too many forwards and other funny stuff beyond what they could deliver.

    I could argue that BB forward sales are actually less risky because a BB can perfectly hedge that forward sale by borrowing cash and buying physical gold, then sell that gold forward at a price greater than their borrowing cost and storage costs. They then just on the physical gold and deliver it against the forward sale when it fall due.

    By contrast, even if a miner is conservative about its production and thus its forward sales, it is still at risk of unforseen production problems - eg labour strikes, mine collapse etc and thus not being able to supply gold against its forward sales.

    In the end the assessment of which is riskier depends on one's assessment of the prudence of the management of either the miner or BB. Your answer may well be neither!

    I don't think Adrian sufficiently notes that a fair amount of the forward selling or short futures that BBs hold could be perfectly hedged - he seems to paint the entire short position as dodgy. I think it is a fair question as to whether the whole lot could be hedged, but that is different to implying that the whole game is a ponzi scheme.