by Alasdair Macleod
The total known stock of gold is estimated to be 170,000 tonnes. This
estimate appears to be reasonable, because the bulk of it has been
extracted in the last 100 years, as exploration and mining techniques
have improved. The rate of annual supply is the green line in this
chart. The World Gold Council estimates that all of the world’s gold
in 1945 amounted to less than 60,000 tonnes, and nearly twice as much
has been mined since. In arriving at estimates of gold before
statistical records were available, estimates have been made of gold
extracted in earlier gold rushes and by earlier civilisations from
ancient Egypt onwards.
Central banking and goldThe central banks are the largest defined category of gold holders. Originally, the role of gold was to back the central banks’ issue of paper money, which was convertible on demand. They, or commercial banks in their network, obtained their gold from the public, who would deposit their gold against a receipt. This receipt was standardised, and could be used instead of gold (or silver for that matter) as money.
The first central bank was the Bank of England, which adopted this role as a result of the Banking Act of 1844. This gave it the monopoly of issuing notes in England and Wales. By law, the Bank of England had to be prepared to redeem its notes for gold, hence the statement on the face of each note, “I promise to pay the bearer on demand the sum of – pounds”, followed by the signature of the Chief Cashier. The Bank of England became the model for central banks throughout the empire when the pound sterling was the international standard currency before the First World War, a model copied by most other nations subsequently. This was how the bulk of known gold stocks, by providing the backing for government-sponsored note-issues, ended up in the vaults of central banks.
At the time of the Bretton Woods agreement in 1944, over half the world’s gold stock on WGC figures was in central bank hands, the vast bulk of it held by the US Treasury. The Bretton Woods Agreement put the world on a dollar standard with convertibility into gold reserved for central banks only. The result was that by the time this limited convertibility was suspended and stopped altogether in 1971, the US Treasury’s holdings had fallen from 72% of all central bank holdings to 25%, reflecting the redemption of dollars by non-US central banks.
This is shown in this chart, the dotted line being the point where the Bretton Woods Agreement broke down. From that point in time, the leading central banks, led by the US, embarked on a policy of removing gold entirely from the monetary system. For confidence to remain in unbacked paper currencies, it was deemed vital that gold should lose all credibility as money. As Nixon said in his speech to the nation, “The US dollar must never again be hostage to international speculators”. This is a straightforward reference to those central banks, particularly the Banque de France, who were redeeming their dollars for gold.
As can be seen in this slide, since the Nixon Shock in August 1971 there has been a small decline in the US’s official gold holdings, as the US sold gold into a rising market in a series of sales in the years following. This policy, which was designed to suppress the gold price failed completely, as gold was eagerly absorbed by the market. Since then the remaining gold has sat on the G7 central banks’ books as an embarrassing asset.
In the mid-eighties central banks started to lease their gold to private-sector bullion banks. This gold was used as the collateral for a carry-trade, which involved selling the leased bullion into the market and using the proceeds to buy government debt. The cost of leasing central bank gold was less than the yield on this debt, and at the end of the transaction, it was either rolled, or the debt sold or redeemed, the gold bought back and returned to the central bank. Everyone was happy: the central bank made some money as lessor, the bullion banks made some commission, the lessee made an interest rate turn, and governments had buyers for their debt. A win-win-win-win situation!
In theory this is fine, so long as the gold, which has been sold into the market, can be reacquired. Annual production during the 1980s averaged about 1,800 tonnes, and there is also the net balance of jewellery fabrication and commercial demand to consider. Once one takes into account hoarding and dishoarding it is impossible to know whether or not the stock was available to deliver leased gold back to the central banks.
The evidence suggests that there was not the bullion available. Firstly, the London Bullion Market has leveraged upwards its transactions on the bullion actually available, suggesting an underlying shortage of bullion, and secondly the bullion market ran into difficulties in Sept 1999, when the Bank of England and the Federal Reserve Board had to intervene.
And this was what Eddie George said about it afterwards to Nick Morrell, CEO of Lonmin after the Washington Agreement Gold Price Explosion in front of three witnesses.
This tells us that as the bear market in gold continued over two decades, the market grew progressively complacent about the dangers of running excessive short positions. This is common to established bear market trends in any market. It is a sort of anti-bubble. It also tells us that the Bank of England and the Fed sold physical into the market, yet this is not evident from the Fed’s official holdings, but there was some movement in the Bank of England’s. However, at that time the Bank was selling half Britain’s gold reserves on Gordon Brown’s instructions, which confuses the issue somewhat.
Accounting for gold at the central banks
A number of central banks, including the Bank for International
Settlements, maintain gold accounts for other central banks. These
accounts are held to facilitate transactions of fungible gold in the
centres where gold either trades, such as London, Zurich and New York,
or in centres deemed strategically desirable. Of course, all central
banks have the option to store their own gold, but given that it has
to be shipped to a settlement centre if it is sold, gold so held is
viewed as a sterile asset.
The bullion banks
We now turn to the bullion banks. The bullion banks are defined as the
banks that are members of the London Bullion Market Association. The
LBMA is the largest physical market by far. Daily turnover is
estimated at as much as 100m ounces, with daily settlements averaging
about 20m ounces. So every day on average, more than 3,000 tonnes are
transacted, more than total annual mine production. The market is an
over-the-counter market between members and is unregulated;
unregulated for two reasons:
There are eleven market-making members who have agreed to quote two-way prices to each other during the London business day for agreed minimum quantities at least twice a day in both gold and silver. They are shown on the slide above.
This is not to be confused with the daily fix, which for gold is
between Barclays Capital, Deutsche Bank, Scotiabank, HSBC, and Societe
Generale. For silver it is between Deutsche Bank, Scotiabank and HSBC.
The fix provides an opportunity for market users through these banks
to buy and sell gold at a single quoted price. This is useful to users of the market who
wish to deal at an official price, rather than an ad hoc one between
two or three
Besides these market-makers, there are fifty-eight ordinary members,
27 of them banks from all round the world, the remainder being
commodity dealers acting mostly on behalf of producers and
fabricators. The list of banks is shown on this slide. These banks
give us a clue to the governments and citizens who are particularly
interested in gold. Thus, we see China is represented by Bank of China
and the Industrial and Commercial Bank of China. This is interesting
because China is now the world’s largest producer at 345 tonnes, none
of which she sells into the markets. So if these banks do anything,
they must be net buyers. The Swiss are represented by the two big
banks, Credit Suisse and UBS, and also by Julius Baer and Zurcher
Kantonal Bank, presumably on behalf of the whole Canton banking
About the LBMAThe LBMA’s client base is the majority of the central banks that hold gold, plus producers, refiners, fabricators and other traders throughout the world. Members of the London bullion market typically trade with each other and with their clients on a principal-to-principal basis, which means that all risks, including those of credit, are between the two counterparties to a transaction. This is known as an ‘Over the Counter’ (OTC) market, as opposed to an exchange traded environment.
The LBMA standardises both deliverable bullion and contractual agreements. Deliverable gold is the London Good Delivery gold bar. It must have a minimum fineness of 995 parts per thousand and a gold content of between 350 and 430 fine ounces with the bar weight expressed in multiples of 0.025 of an ounce - the smallest weight used in the market. Bars are generally close to 400 ounces or 12.5 kilograms. To be “good delivery” they must have been produced by a refiner on the LBMA’s good delivery list.
Deliverable silver must have a minimum fineness of 999 parts per thousand and a recommended weight between 750 and 1,100 ounces, although bars between 500 and 1,250 ounces will be accepted. Bars generally weigh around 1,000 ounces.
There are two standard LBMA template agreements used between LBMA members and their clients you should be aware of. The first is for allocated gold, and the second for unallocated gold. Allocated gold is clearly the property of the client, being segregated from the LBMA member’s own assets, whereas unallocated gold is gold that appears on the LBMA member’s balance sheet as an asset, the offsetting liability being a debt to the client.
LBMA members strongly encourage clients to hold unallocated accounts by charging little or no fees for the privilege. They discourage clients from holding allocated accounts by charging high storage and custody fees. There are very good reasons why this is so. They are unable to make use of allocated gold, whereas every ounce of unallocated gold is used to back the LBMA members’ dealings in the market. The disadvantage to the client is that he is exposed to counter-party risk. Now, if you have looked recently at the balance sheets of some of the European banks, you may not wish to take that risk. So while there has never been a bankruptcy in the market (though there have been some covert rescues – for example the one I mentioned earlier which Eddie George talked about), implying that unallocated accounts are safe, for many clients this is not an assumption they are prepared to accept any more.
This leads me onto another important risk to consider. A bank only has to hold enough physical gold to meet likely demands for delivery. This is like a bank that takes in cash deposits: it doesn’t have enough of its own money to meet a rush of depositors seeking to withdraw their deposits at the same time. We call this fractional reserve banking, or bank gearing, and it applies to bullion banks in just the same way. It is perfectly possible for a bullion bank to owe its customers 100 tonnes, but only own five or ten. Now, if the price of gold rises substantially, because the bank’s liabilities to its bullion clients exceed the bullion in its possession, it will face losses. The client is usually aware of this, so rising prices, or rather prices rising too rapidly, could destabilise the whole market. I will come back to this subject in a moment.
What do these banks actually do on the LBMA? The answer is they deal in bullion in a fully flexible manner, by agreeing settlement dates in the future. The deals are called forwards, the difference between the price for settlement today and a future date being the sum of the bullion price and the gold, or silver, forward rate. These are termed GOFO and SIFO rates, and equate to LIBOR, usually with a small positive margin. The difference between LIBOR, the money-market borrowing rate, and GOFO or SIFO is effectively the lease rate. It works like this. A bank sells gold for spot delivery, and buys it back for delivery in 3 months’ time, the difference in price being the three-month GOFO rate. It buys 3-month LIBOR with the proceeds and at the end of the three months closes the position and takes delivery of the bullion. The profit is the difference between LIBOR and GOFO (or SIFO in the case of silver), and is effectively the cost of leasing gold or silver for three months.
Now let’s look at two more examples.
Forward transactions are the basis for a range of other activities, such as loans, swaps, repurchase agreements, leases, etc. First we will look at a simple transaction. In this deal, a producer delivers gold to the Bullion Bank against payment, and the bullion bank sells it on to a fabricator. The BB makes a simple turn on the deal.
Derivative marketsNow we move on to futures and options, derivatives, and the largest futures market by far is Comex. For those unfamiliar with futures I’ll give you a brief definition: a future is a standardised contract on a listed exchange that on expiry gives the owner the right to have the underlying asset delivered. Its standardisation is the principal difference from a forward contract, which is more flexible. Being listed on a regulated exchange they are more accessible to members of the public, both those who wish to use futures to hedge risk, such as miners or farmers, and those that wish to speculate.
Unlike over-the-counter markets such as the LBMA, information is made available to non-participants, in the interest of ensuring a fair market.
By far the largest precious metals futures market is the CME, or Comex. Besides noting that China is developing futures markets in Hong Kong and Shanghai, these are as yet small compared with Comex. So this analysis will concentrate on Comex.
Comex provides large amounts of price, volume, settlement and warehouse data all of which is available through their website. Unless you are a dedicated dealer, monitoring it is very time consuming. Instead, there are a number of blogs you can follow. Perhaps the most detailed free-bee site is Harvey Organ’s daily Gold and Silver Report. Ed Steer’s daily email also produces a good market summary together with links to stories that have caught his eye. There are also subscription-based services, and in Silver I should mention Ted Butler. If you troll around the GATA website, you will come across other names where a daily analysis of Comex trading is available.
One of the most useful tools is the weekly Commitment of Traders Report, which can be accessed through the CFTC’s website: the CFTC is the Commodities Futures Trading Commission. For those not familiar with actual dealing on an exchange floor, understanding the weekly COT Report can be very difficult to comprehend, but if you follow the basics, you will be able to understand what others are saying about the figures.
There are two sets of information break-down given to us. The first, with which most of us are familiar, is shown on this slide. This is termed a Legacy Report, because it is now supplemented by a separate report on what is termed disaggregated data, which I will come on to in a moment.
This legacy report, adapted here by Goldseek.com, is for gold contracts of 100 ounces for the week ending Tuesday April 3rd (they always run Wednesday to Tuesday and are published on the following Friday), and as you can see traders are broken down into three categories: Large speculators, Commercials and others (small speculators). Commercials are at the heart of the system, and consist of producers, fabricators, dealers in physical and banks dealing in bullion and swaps. A swap by the way is the hedging of another position off-market.
The speculators or non-commercials include professional money managers, such as hedge funds, as well as a wide array of non-commercial, or speculative traders. The term speculator here is something of a misnomer. It includes investors buying futures with the intention of taking delivery. It includes hedge funds who take a position to hedge another unrelated asset. The categories are catch-all to a degree, because the COT reports are produced for all futures markets. There is also a further legacy report which includes options on futures which I will omit for simplicity, but is also available from the CTFC and published in this form by Goldseek.
To understand COT legacy reports, you have to appreciate that the commercials are in control of the market most of the time. Producers in this category use the market to transfer entrepreneurial risk to speculators. They will use the market to lock in today’s known prices to fund future production costs, removing the possibility that they might get a lower price when physical product is ready to be sold. So they sell futures with the intention of delivering product to the market. Fabricators and processors, who are also Commercials obviously might want to take delivery in the future, and so might want to lock in today’s prices for future delivery, removing the risk of a price rise in the intervening time. The bullion banks, the most influential Commercials, might deal for their trading book, they might hedge a position they have taken with clients on other markets (this is swap activity), or they may be dealing for a client. The large bullion banks use futures as only part of their overall activities.
Overall, there is likely to be a bias in gold and silver such that the commercials have a net short position. This is because bullion banks in the Commercial category providing liquidity will respond to demand, which as in any market will tend to be on the buy side, because non-commercials, members of the public and money-managers, tend to be buyers on balance if they are at all interested. So the commercials will always be net short, and the bigger that position, the more overbought the market will be.
So when the commercials are adding to their longs and reducing their shorts, as you can see in this report, that is giving us a possible signal. It would be reasonable for us to think that at the margin the commercials think price falls are less likely than they did the previous week. We see that the large speculators are cutting their positions long and short. The market appears to be less overbought, which is at least less bearish, and probably bullish. Their spreading position, which is when they are long of one maturity matched by a short on another (in other words they are taking a financial arbitrage between maturities) we can basically ignore here.
The small speculators are also reducing positions, and since they are always the weakest holders, this has to be bullish. And we can see that the level of outstanding contracts has fallen by 9,500. This is also positive, but not in addition to the factors already covered.
The money in the market is always unbalanced in favour of the Commercial bullion banks. They have lots of money, even your money as a taxpayer if they are too big to fail, and can always bluff anyone not prepared to put up funds for delivery. This is because the non-commercials and speculators have geared positions, which will magnify their losses. We see this happening time and time again. The big commercial bullies wait for the punters to build up their long positions, and then they whack it hard. They know that by doing so they will trigger all those stop-losses. In an afternoon they can make $100 an ounce this way. They make lots more money trading this way than they lose from being continually short in a bull market. You have been warned!
In the chart above I have derived from disaggregated data the net positions of the swap dealers. Two years ago, they were short a net 120,000 contracts, which is the equivalent of 373 tonnes of gold. Since March 2011 they have reduced their net shorts from -110 thousand to zero give or take at the peak of the gold market last September. At the same time Open Interest fell from a peak 650,000 in November 2010 to the 420,000 level.
Why were the swaps short? The only logical reason has to be the central banks were supplying the market with physical. It is that extra physical that was being hedged two years ago. And what is interesting is that at that time Portugal was rumoured to have given its gold up as collateral to the Bank for International Settlements. The amount that actually showed up in the BIS accounts was 349 tonnes, and the date was late 2009. Fits perfectly! Put another way, Portugal’s entire gold stock appears to have been sold and absorbed into the market.
I want to draw your attention more specifically to the sudden reduction in swaps from net short in August 2011 to a positive 8,700 in October. That reduction drove the gold price up to its record highs. Now these are swaps, so they indicate that shorts were running for cover on other markets, and this must be mostly on the London Bullion Market. This is important information. It tells us the reason gold got to $1900 was a short squeeze on physical bullion, which is what markets fear most.
However, it is not a requirement that swaps have to take a large short position for a bear squeeze to happen again. While systemic risks are high, unless there is a dramatic change in sentiment, the swap dealers will more likely maintain a more balanced position. Also, their positions link back to the London Bullion Market where the bullion banks naturally have short positions, the result of their unallocated accounts.
Since 28 February, when gold was $1785, the commercials have reduced their short position by 101,500 contracts, the equivalent of 10,150,000 ounces. This is up to last Tuesday. Overall, I would say that these two charts show a bullish outlook for gold.
Now let’s look at silver.
Silver contracts are for 5,000 ounces. This is a metal which has significant industrial use, unlike gold, and that use is rising all the time. Annual mine production is about 750m ounces, and scrap silver adds about 215m oz. So supply in round terms is a billion ounces. Industrial use is significant and growing, which with jewellery, silverware, coins and medals absorbs nearly 900 million ounces. The balance is implied net investment. So the market is very tight when people are investing in silver, but potentially becomes weak if they turn into net sellers. But in the longer term, industrial consumption looks like exceeding mine supply. Also, silver is a by-product of other metal mining for many of the producers, which does affect their view of the metal. 22% of global production is from China, Russia and Khazakstan who do not offer their own silver on western markets, and many producers send their metal to China for refining, making China the biggest supplier of refined silver by far.
That is the background to bear in mind. Here we can see the commercials have added to their shorts and longs, as have the large and small speculators. On the face of it this is moderately bearish, particularly when we concluded the opposite trend for gold was bullish. But let’s look at it in more detail and go to the disaggregated data.
We know that the commercials have had a large illiquid short position, and that the swaps are hedging shorts elsewhere. This is shown on this chart.
First, let’s deal with the swaps, which are the yellow line and the right hand scale. Unlike gold, the swaps have been generally net long, with the exception of the beginning of 2011, when in February the swap position spiked down to minus 8,932 contracts. Otherwise, over the period covered by the chart, the average position works out at just over 2,500 contracts long. In February 2011 silver crossed the $30 level on its way up to $50, so the swaps scrambled for cover, indicating that silver bullion was in short supply. Interestingly, despite the dramatic slide from the highs at $48 on 28th April 2011 to the lows last September and late December, the swaps continued to build long positions. This indicates that despite the collapse in the silver price, there was still a shortage of physical metal. Since then, the swaps have reduced their long positions, indicating that the balance of bullion relative to physical demand has begun to ease market shortages.
Now let’s look at the Producers, bullion banks and fabricators, the green line and the left-hand scale. Back in October 2009 their net short position was 640,000 contracts (320m ounces), while the gross short position with the longs stripped out was 730,000 contracts (365m ounces). This is half global mine production. Since then they have been reducing their short position, during the the run-up in silver to its high at $48 and in the subsequent shake-out, and only began to increase their net shorts from December. Of course, we do not know the actual identity behind these shorts, though it is well-known in the market that JPMorgan, or a client of JPMorgan is the big short.
There is continual debate about this position, which was apparently inherited from Bear Sterns, when JPMorgan took it over. Those of you who have followed this debate will know that JPMorgan are routinely accused of market manipulation and even illegal dealing.
However, the head of JP Morgan’s, commodities division (Blythe Masters) publicly stated on TV recently that they did not run uncovered positions. I cannot believe a senior executive of JPMorgan goes on TV to tell a straightforward lie. So, taking her statement at face value, JPMorgan’s silver position, which they took on from Bear Sterns, has to be for a client, and that is most probably China, who besides being a major producer is the largest refiner of silver concentrate and doré (which is semi-refined ingots). Others have suggested the Fed, who certainly have a history of market manipulation. But I can’t see any logical reason why they should manipulate silver: Gold yes, silver no. This also explains why JPMorgan has been unable to comment and enquiries by the CTFC have yielded no comment, until Ms Masters narrowed the field for us.
So the variable we need to know is actually how many months forward are the Chinese government looking to sell their refined output. Given that management of commodities dealing is a centralised government function, this is probably both a strategic, political decision and a commercial decision. Bear in mind that from a commercial point of view, the Chinese want low silver prices, but from a strategic point of view they probably want to build a strategic reserve. However, they have been shortening their time horizon, which we can deduce from the green line on the chart above has gone from many months (like six plus when you allow for forward purchases by fabricators) to perhaps less than three months. Looked at another way, on the price decline from Feb 28 when silver was at $37, the commercials have bought back 27,800 net contracts of silver (139 million oz). That is up to last Tuesday (2nd May).
So, a legacy COT report which looks bearish on first sight is not bearish at all. All we can conclude is that someone, most probably China, is controlling the price, and has been reducing their position. Commercial considerations mean that ideally for China the price of silver should not rise too fast. But it is already under-priced relative to gold at over 50 to 1, and the next rise in the gold price only needs to take investment demand above 100m ounces annually for the price itself to take off again.
For the answer we have to look at China’s global political strategy and her gold position, which is at the heart of my next subject.
Politics and precious metals
In my economic analysis earlier I gave you the history and reasons why
gold became demonetised. The advanced economies in the West set
themselves on a course of increasing government control, encouraged by
their neo-classical economic approach. This was a course that the
communist countries backed down from in the face of crisis twenty
years ago.In my economic analysis earlier I gave you the history and reasons why
gold became demonetised. The advanced economies in the West set
themselves on a course of increasing government control, encouraged by
their neo-classical economic approach. This was a course that the
communist countries backed down from in the face of crisis twenty
This is the reason they set up the Shanghai Cooperation Organisation
eleven years ago, with these full members. It has four principal
These are India, Iran, Pakistan, Mongolia; and Afghanistan has
announced her intention to become an observer state. Then there are
dialog partners, who share the same values but have no current plans
This slide shows some interesting information on our subject. The
combined output of the SCO group is 26% of the world’s total, and
there is little doubt that the full gold mineral potential of Central
Asia still lies untapped. The gold reserves shown here are those that
have been officially declared. It is quite likely that Russia has
under-declared her position, having been actively mining gold for many
decades. China is certainly under-declaring her position. Her holdings
here are the central bank holdings, and are not regularly updated.
Furthermore my contacts tell me that the Communist Party itself is a
major holder, and it is quite possible her total holdings exceed
10,000 tonnes – perhaps even more. Furthermore, she is trying to buy
gold mines in Australia and elsewhere.
In conclusion, today we have covered these topics that relate to gold
and silver, and they will determine their future prices.