Koos Jansen
BullionStar Blogs
Koos Jansen
Posted on 4 Apr 2015 by

Thoughts On The Price Of Gold

Withdrawals from the Shanghai Gold exchange (SGE), which equal Chinese wholesale gold demand, in week 12 (March 23 – 27) accounted for 46 tonnes, down 14.5 % w/w. Year to date total withdrawals have reached 607 tonnes, up 9 % from 2014, up 33 % from 2013.

Screen Shot 2015-04-03 at 11.23.15 AM
Blue (本周交割量) is weekly gold withdrawn from the vaults in Kg, green (累计交割量) is the total YTD.

Shanghai Gold Exchange SGE withdrawals delivery only 2014 - 2015 week 12 x

Shanghai Gold Exchange SGE withdrawals delivery 2015 week 12 dips x

Ever wondered why Chinese demand doesn’t move the price of gold substantially higher? A much perceived analysis in the gold space is that (central) banks suppress the price of gold. While it certainly is in their interest to control the price of gold and there are many clues they do intervene, in this post I would like to approach this subject from scratch, from what I believe is basic economics, hopefully sparking debate.

Thoughts On The Price Of Gold

In any market where goods are traded there is supply and demand. For this post we’ll look at the gold market to examine the relationship between both; there can be people offering gold for sale (supply), meeting people who are willing to buy gold (demand). If a transaction is agreed at a certain price the amount of gold sold (supply) is always equal to the amount of gold bought (demand), it’s impossible supply and demand are not equal by any measure – or one would use different metrics to measure either one.

When demand increases relative to supply (economic agents are willing to buy more gold at prevailing prices), the strength of demand will transcend the strength of supply. As a result the price of gold will rise until a new market equilibrium is found. The volume of gold bought in itself does not indicate the price will rise, for if an immense flood of supply would be unleashed that is being met by equally strong demand the price of gold will not change. No matter how much gold is sold, it won’t tell us anything about the strength of demand relative to supply, only the price can tell. The price unveils the forces of supply relative to demand.

In the graph below we can see how an increment in demand relative to supply can move the price.

Supply Demand curves

  • P – price
  • Q – quantity of good
  • S – supply
  • D – demand

In this example demand increases from Q1 to Q2, while supply remains constant; the price moves up from P1 to P2 for a new market equilibrium.

Technically, if India buys (or imports) 4,000 tonnes a year this doesn’t necessarily mean demand is strong, nor does it mean the price will go up or would have gone up in the process. If supply to India was stronger than demand from India, the price can go down while thousands of tonnes cross the globe (given India has no domestic mine production).

The gold market is quite unique and cannot be compared to other markets, like the potato market. The primary difference lays in the fact that gold can’t be consumed, as it doesn’t corrode all gold is immortal and can be recycled indefinitely. We humans can lose gold, but it can’t vanish. Therefor, all gold mined is added to the total above ground stock. In contrast, potatoes have a limited life span of itself and when eaten are digested. Yearly supply and demand of potatoes is determined by what is produced versus human trends that set our need for consumption.

Gold supply, on the other hand, is less determined by mining output, as this is effectively only a small percentage of the total above ground stock. It’s estimated yearly mining output is 1.6 % of the total above ground stock. In theory the total above ground stock is potential supply at the right price. The willingness to sell largely depends in which category the owner of the metal can be classified. Yearly mining output is likely to be sold no matter what the price is; above ground bullion can be sensitive to price movements; ancient gold artifacts are likely never to be sold; jewelry can have emotional value for owners, etc. Furthermore, no one is ever forced to sell – aside from government confiscation that have occurred in history. In short, the volume of yearly supply is elastic, but for sure it’s more than mining output.

Additionally, many other aspects determine the volume of supply and demand (the price). To name a few: technical analysis, trust in central banks, financial stability, real interest rates, stock market performance, inflation (expectations), the yield curve, disposable income, the strength of alternate currencies, industrial applications and supply and demand data (for example, if China buys 2,000 tonnes of gold per annum, but analysts worldwide state – for whatever reason – the Chinese buy 1,000 tonnes, this leads to distortion of sentiment as the market will react on false assumptions).

Next to physical supply and demand, the price is affected by gold derivatives – futures, options, forwards and unallocated gold – and the London Gold Fix. Derivatives are leveraged a multitude of physical supply and demand volumes and therefor have an equally greater impact on the price and sentiment, especially in the near term. In derivative markets the price of gold can be easily moved up or down to the likes of big traders in the short to medium term.

Terry Smeeton of the Bank Of England stated at the Australian Gold Conference in March of 1994 (from Frank Veneroso’s Gold Book 1998):

…at least 20 central banks are engaged in swaps, options and futures. This is double the number of banks who were regular players a few years ago.  

CME Group, the world’s biggest derivatives marketplace located in the US, launched a program in July 2013 to incentivize central banks outside the US to trade in a number of products, a few of which are Metals Futures Contracts traded on CME Globex, by offering them a special discount (click here to read the details from CME Group). I would be surprised if central banks don’t trade gold futures at this moment.

The London Gold Fix is set twice a day in the London gold market through an electronic, auction-based platform, at which currently seven bullion banks participate. The auction has been under scrutiny as its opaque nature is vulnerable for manipulation.

Gold Fix Chart

It should be noted that the volume of gold traded in the London OTC gold market is unknown, but estimated to be a few times the size of the futures market in New York (the COMEX).

Derivatives can be used by hedge funds, speculators, bullion banks and central banks to influence the price, subsequently influencing technical analysis and sentiment on which the rest of the market reacts. People can be scared to sell, however, when the price in the paper markets (derivatives) moves up or down, no physical gold owner is forced to sell at the paper prices except for miners. If the paper price goes down and physical demand increases this has to be met by equal physical supply, that is, if the price for physical gold follows the paper price. If the physical price disconnects from the paper price, premiums will appear at one location.

Reality Check

In 2013 the price of gold made a spectacular nosedive, which was followed by an even more impressive flight of physical gold from Western vaults to China. The UK net exported 1,424 tonnes of bullion, China net imported 1,507 tonnes.

Screen Shot 2015-04-02 at 9.56.46 PM
BulionStar charts

According to my logic and textbooks the fall in the gold price and the physical moving east was a stronger force of supply than demand. We could quantify Chinese demand as “strong”, but apparently supply was stronger.

In the Gold Demand Trends Q2 2014, by the World Gold Council we can read:

The rapid 25% drop in the gold price during the April-June period of 2013 sparked a leap in gold demand that we have heard described as a ‘once in a generation’ event.

My point being, if central banks suppress the price of gold, this can only be done if physical gold is supplied to the market. So the question is, who is currently selling gold to China? (Or in the free market since the London Gold Pool collapsed in 1968.)

China is the largest miner of gold at 450 tonnes a year, though to satisfy domestic demand additional gold is imported; in 2013 Chinese net import exploded to 1,507 tonnes, my estimate for 2014 is at least 1,250 tonnes and year to date China has imported well over 400 tonnes. Is this sold by institutional investors in London (the LBMA system) or by central banks? Eventually time will tell. In the meantime I will continue to research how much gold is flowing to Asia and if there is any gold left in Fort Knox (read this and this post for my Fort Knox research).

Koos Jansen
E-mail Koos Jansen on:

  • Dirkk

    Thanks for the post Koos
    What about high frequency trading. This is a purely manipulated market. I can see that and I dont need to be a gold and silver guru. China and Russsia buy gold by the tonne or several thereof. It is no longer a supply and demand issue. Evey time the sales of gold and silver go up the central banks generate fake gold and silver contracts (sell orders) and the price drops. This has to stop. I purchased silver as 50% is used in manufacturing but instead of going up in price as you would normally think would happen as supply and demand would force it to, the central banks merely generate fake sell orders and the price drops. If you have a look at Kitco and change the frequency to seconds you can see the computers generating these fake contracts. We need a complete currency reset to stop the banks manipulating these commodities

  • pontiac

    I think we can learn more about supply and demand from the silver market as gold has an apparent unlimited supply. Since there are no known stockpiles of silver, increased supply must be coming from silver mines. One of the effects of crashing the price of silver has been to force the silver miners to dramatically increase the rate which they are mining. Sure enough almost all of my miners have greatly boosted production in the last couple of years, some are up 25%. The lower lead and zinc prices have also caused an increase in silver production as a by product just to keep the mines operationally profitable.

    This process will continue until the marginally profitable mines start to deplete. So far I have not seen mines closing due to lack of available ore.

    • rowingboat

      Regarding silver stockpiles, have you calculated what is held in the UK, net imports over many years? This data is available (Eurostat website) and the UK is the main supplier of silver to India, for example.
      Silver is a monetary metal and is also hoarded, much of it in London I suspect.

  • Kirill Klip

    Great Article Koos, thank you!

    Huge Miss On The Jobs Report And Gold – The Parabolic Move In US Dollar Is Over.

    After today’s huge miss on The Jobs Report the parabolic move in US Dollar is officially over. With crashing dollar commodities should finally find the bottom. Gold is already printing very strong Double Bottom Reversal, confirming what we had discussed before. Move above $1,310 will bring all the fun back now. Good Friday to everybody who is not short Gold or Euro … Monday will be interesting with Record Short position in Gold.


  • JohnSpiers


    You are doing extremely important work, but if I may suggest some parameters…

    1. Please keep definitions tight. Money is money (gold or silver usually) and everything else is not. That paper folding stuff is currency, even when it was warehouse receipts, it was currency, not money. Please say only money, and money only when referring to that medium of exchange, usually gold and silver.

    2. Yes, gold is a commodity TOO. It is always a commodity, and often money too. The commodity aspect never leaves the equation.

    3. Credit is never money, although most currency now is really credit, or IOUs. We are not in monetary deflation, we are in credit deflation, which acts the same way.

    4. Follow Spooner on the right amount of money: all gold and silver is, after mining, turned into money (coins, ingots, etc.) When there is not enough money, plate, decoration and jewelry is melted down. When there is too much money (keep the definition strict) it gets converted into decoration, plate and jewelry.

    5. Money (strictly defined) only shows up where usury-free asset-based credit, the natural trading condition in history, is no longer extended for whatever reason, usually political end times. Where politics ended civilizations, pots of gold and silver are found by archeologists. Where natural disasters ended the civilization, only tallies (credit records) are found.

    WE are shifting from 40 years of lending asset-less backed securities at interest (usury) and leveraged in fractional reserve which suggests one logic for investing and economic action. Now that has reversed, we are in credit deflation. Extremely few people know what to do in credit delation.

    And in any case, he who has the gold will be in that measure able to beck credit at the top leverl. At the bottom level, no one needs money, just credit.. and not bank credit at interest, but no-interest asset-backed vendor financing, where now, the longer you wait to get paid, the harder the currency in which you are paid, assuming the currency is backed by… yes… gold and silver.

    John Spiers

    • Tuna

      Many misconceptions.
      Gold is NOT a commodity. It has no link to the Marketplace. That is why it is the perfect wealth reserve… and Silver is not. This is why CBs hold Gold and not Silver. Having no link to the marketplace allows Gold to escalate in currency value without disturbing the marketplace.
      Gold, Silver are NOT money. Money is the Medium of Exchange – is and will continue to be paper. If an old paper hyperinflates, a new one will be introduced – like almost every HI in history.

      • JohnSpiers

        Hey Tuna,
        Your definitions are going to get you into trouble. Of course gold is a commodity, as well as a money. It does trade physically on exchanges. Just as copper ends up in piping, gold ends up in jewelry. It is only a relatively good wealth reserve, hence its present supremacy, hardly a perfect wealth reserve (and your definition of wealth must be a little off too.) Paper is either warehouse receipts or IOUs, either way merely currency. Without proper definitions, into the fog you will stumble!


        • Tuna

          Of course Gold is NOT a commodity. You are again confusing paper gold with physical gold. Most of the physical gold in the world lies still. It does not move. It’s paper derivative is the one that is vastly traded and can be confused as a commodity… and if that is money then anything is… physical gold is no more money than fine art or land.
          What denominates the price of gold? The paper Gold market!! It is mutually exclusive to the physical. High physical demand does not mean a higher paper price – surely you Goldbugs see this by now.
          Your confusion will cloud your understanding of Gold’s eventual place in the monetary system. Consider paper gold collapsing – because it has no claim check on physical – it is essentially worthless. It doesn’t even need the 112:1 ratio statistic from the LBMA. Gold is the hole card for the system. It can go up 100X in currency terms without affecting the marketplace. And physical gold doesn’t need to do anything for this to happen. The paper Gold market only needs to reach its intrinsic value – zero. This is where we are headed. Of course, on that journey down – don’t expect to be able to buy physical gold anymore…

          • JohnSpiers

            You shifted your terms from gold to paper and then projected your confusion onto my argument. I said nothing of “paper gold” nor would I for there is no such thing. there maybe warehouse receipts, or some fake derivative, upon which you apparently fixate, but no, I have no opinion at all on paper gold, a term with internal contradictions. Whatever happens to it is of zero concern to me, but apparently it does to you. Hope it turns out well for you, whatever it is.

  • Gutekunst

    Koos, I have personally thought a lot about this issue and
    my conclusion is that the paper market is and will dominate the physical
    market. If the Chinese would be willing to bid up the physical market, things
    could be different, but they are as all asians price sensitive buyers, they buy
    what they could get with a little premium, but they are not paying e.g. 50 USD
    over Spot to get more physical. You could compare the “Exchange Rate”
    GOLD/USD to any other e.g. USD/EUR, they have completely disconnected from the
    underlying economic realities and are dominated by the OTC, Futures etc. The
    price is simply driven by the expectations of the market particpants, which are
    changing their views sometimes on a daily basis. Of course the exchange rate
    could fold sometimes to economic realities like the Swiss Franc, but
    unfortunately Gold does not play any important role in our current financial
    system, and currently nobody speculates alongside the Chinese in the OTC
    market. Whether we like it or not, the price is driven by sentiment, which is
    reflected in the OTC markets, the Chinese would need to buy 50-100 times more
    Gold to match the size of it, or you can state it this way, 90% of the Gold
    price is paper and 10% is physical, this 10% will never have any meaningful
    influence on the price, otherwise we would see a rise in Gold since mid 2013.
    Allmost all available physical gold goes to the east und I do not believe that
    central bank supply physical, they may lease out, swap etc., but the gold will
    never leave their vaults, only the ownership on the paper may be different,
    since they want to keep gold as the ultimate payment in case of system stress
    as Draghi said. I think there is sufficent unallocated gold at the banks which
    they can still sell to Chinese. The banks probably keep 10%% of the
    unallocated, the rest is sold on the market and there is still a plenty of
    unallocated Gold to be sold to Asia in my opinion.

  • https://shiresilver.com/ Ron Helwig

    I think the basic underlying question here is “why isn’t the price of gold as high as I (Austrian-esque economic folks/gold bugs) think it should be?”. And the answer I have is that gold is no longer money.

    Gold has many uses, and part of the demand for gold is based on those real and perceived uses. But since most gold uses are things where other materials can replace it if the price gets too high, there will never be sufficient demand to create the large increase in price desired by those interested in it. The only way that will happen is if there is a use for gold where a substitute cannot be found.

    I might be wrong, but from what I can tell there is only one use case of sufficient size to effect the desired price rise, and that is remonetization of gold. There are a couple things holding that back (besides government): bitcoin and usability. Bitcoin was designed to act like gold but in the digital world, and it does a fine job of that; so I don’t expect any attempts to introduce a digital gold currency will work. That leaves the niche of in-person anonymous physical cash; complementing bitcoin’s digital only presence.

    But gold coins, rounds, and bars are too inconvenient and will never see mass acceptance again. People would much rather hand over control of their money to acknowledged bastards to avoid handling coins. So what we need is a more usable form of physical gold. That has been provided in the form of my own Shire Silver gold and silver cards and now there is also the Valaurum product. Both embed real precious metal into user (and wallet) friendly form.

    IMHO, until we see the masses using these or similar products as money we won’t see the expected and desired rise in gold price.

    • Mark Fisher

      Nope. It is complete demonetisation which will realise gold’s true value and utility as a physical wealth reserve and final settlement vehicle. Monetisation (the US official gold price) is an attempt to unnaturally fix and limit gold’s value, or more correctly, give the Dollar value. Look at the ECB’s balance sheet, their line 1 asset is physical gold, marked to market on a quarterly basis. Gold is used to back the currency but on a floating basis; as the gold price rises the Euro’s balance sheet strengthens.
      Forward selling was introduced to enable the continuing flow of physical gold by reducing the demand from gold speculators, who are really just trading currencies. The physical market does not price gold currently.

      • Tuna

        Yes, Mark – you have it. Exactly. Much in like with Freegold. Physical demand is mutually exclusive to paper price. Gold is not supply/demand – it is ‘stock/flow’.

        • Mark Fisher

          I may have read a little Another and his derivatives!

      • https://shiresilver.com/ Ron Helwig

        At first I was confused by your response, but then I saw your definition of monetisation. That’s one I would NEVER use, as anything government touches turns to crap and I’d never want the dollar tied to gold. What I want, and am suggesting, is that gold become money again; not that it be the theoretical backing supposedly stored away in some warehouse but more likely hypothecated. When I call for remonetization I mean for people to be physically using gold in their transactions instead of fiat.

        IMHO if your monetary system requires managing and uses highly technical terms that most people don’t understand then it is guaranteed to be manipulated and corrupted.

        • Mark Fisher

          Fiat is for making and spending money, gold is for storing wealth. Why do you need the ultimate SoV to also act as your MoE or UoA?

          • https://shiresilver.com/ Ron Helwig

            Technically you don’t need that, but allowing them to be different leads to problems. Most importantly, none of them should be politically managed. By having them be the same and also be something where it cannot be inflated at a whim you avoid a lot of problems.

            At this point someone usually brings up the fact that there were problems with the gold standard that led to what we have now. I have two responses to that:

            1) it is indicative of the typical political strategy of taking a small problem, implementing a good sounding but flawed solution, which leads to more problems that “need” more “fixing”

            and more importantly

            2) the problem with the gold standard was not the gold but the standard. By trying to fix the price the government was fighting the market and the market always wins eventually.

  • sb

    From what I see gold price is mainly driven by futures market. US hedge funds use it to bet on US rates. Every time there is a bad number in US and FED become more dovish gold goes up (hopefully on Monday price will jump by $10-$20) . Hedge funds/banks has access to billions of funding and they can easily trade 10000-20000 contracts in a few minutes. This mechanism is visible when you look at COT managed money position, it is highly correlated with gold price. FED gets more hawkish, money managers read Wall Street research and sells gold futures, gold price goes down. Money managers are heavy short, there is no one left to short, price stabilizes, US bad numbers is released, shorts cover, price goes up.

    Spot gold price is linked to future price through gold market makers, big banks which set price based how easily they can borrow gold. Spot market is mainly paper market where you trade both physical gold and gold claims. When you buy gold through LBMA you don’t need to take physical delivery, you can just have claim on seller.

    Physical gold demand is quite small compared to paper market and any demand short term mismatch between paper and physical gold market can be smoothed with the physical gold inventories in London vaults. In the long term that is no possible if there is constant deficit of physical gold.

    Physical demand slowly drain available gold stocks in London vaults. This relentless draining (500mt last year) is slowing making gold borrowing more difficult. Every time there is a surge in gold demand GOFO rates are more negative, in Nov/14 -60bp, Mar/15 -90bp (implied from basis). At some point in the futures (probably within 1-2 years, 500-1000mt of gold easy to borrow) the shortage of gold to borrow become so acute that physical price will either decouple from paper price or physical gold market will drive paper market. Now spot price can be $1 higher than future price, but if this difference become bigger e.g. $10 then you can’t ignore it, it is 10% annual return on lending gold (or 10% loss for borrower).

    Contrary to cash squeeze there is no way out by printing money, you can’t create gold from the thin air. Central banks would have to start selling/lending their gold reserves but I am not sure they will do this this time.

    • KoosJansen

    • rowingboat

      There won’t be any need to create gold from thin air. Look at what happened to Indian imports in 2009 and the start of the bull market. They declined as holdings in London increased with a rising price (Western investment demand). Also Swiss imports rose, particularly during the GFC years.

      When we sum what London + Switzerland were importing annually, well China has simply replaced that. Except the inventory build-up in London is being depleted so the selling pressure is greater than the buying pressure (West driving the price not the East).

      By my calculation the UK net exported 1,838mt in 2013-14. That’s one-third of what was net imported from 2001-12 so potentially the current state of play could continue for years until London holdings increase again and Chinese/Indian imports fall to accommodate (new bull market).

      What’s interesting sb are the ETP holdings. At the end of 2014 there were about 1700mt in total…. 700mt GLD (in London). Are the other ETP holdings in London as well, or scattered around the world? Overall there are still 3600mt in the UK, net imported since 2001.

      • sb

        Where did you get the UK net imports between 2001-2004? My estimation from 2015-1014 is around 1800mt, which implies that you have 1800mt for 2001-2004. I think this is quite big number, in 2005 the UK net exported (not imported) around 200mt.

        700mt in GLD can’t be leased that’s why I assume that of my 1800mt, no more than 1000mt can be leased. Probably less as some of this gold is in strong hands e.g. another ETP.

        2009 year is a good example when gold demand/supply drove gold price, new Western demand could only be met by rising price, higher prices decreased Indian demand leaving gold for Western investors.

        These days gold demand/supply imbalances show not in the gold price but in gold leasing rates. When there is a surge in demand GLR jumps (GOFO goes negative). It happened in Nov-14 and Mar-15. The gold now is still in backwardation despite the fact that India demand dropped almost to zero last two weeks. This suggests that the gold market is getting tighter and tighter. This is probably the longest period of negative GOFO in history. I think if price of gold don’t rise we will likely to have permanent backwardation soon.

        • rowingboat

          From the Eurostat link you posted sb, then deriving fine tonnage from gross weight and average annual gold price (WGC).

          Yes, I calculate UK net exports of 184mt in 2005. Net imports for 2001-04 are 230, 532, 594 & 191mt, respectively. Note the large imports in 2002 and 2003, when gold price made one of its biggest gains in percentage terms during the bull market.

          Now turn to the historic Swiss data, which is now available. UK net exports to Switzerland that had been averaging +300mt during the 1990s, collapsed in 2002 and 2003 to 12mt and -1mt. In turn look what happened to net exports from Switzerland to India:

          1999: 287mt
          2000: 298mt
          2001: 378mt
          2002: 189mt
          2003: 204mt
          2004: 308mt
          2005: 435mt

          The big boys choked supply, storing gold in London and forcing sufficiently higher prices to curtail ROW demand in 2002/03 via Switzerland. It’s going to be fun to watch if/when the UK becomes a net importer of 500-1000mt again, and Switzerland becomes a net supplier to the UK like it did in 2009.

          • sb

            Eurostat must have uploaded some data, a year ago when I checked last time there were no reliable data. I see now data from 2002. Which database did you use to extract 2001? Do you see any earlier data?

            If the numbers for 2000-2004 are correct then that implies that was only 3500mt in London at the begining of 2000 (in 2005 there was 5200mt in total, 4000mt in BOE custody). Not a lot if you take into account that only German and the UK had 2000mt. It suggests that a big chunk of gold reserved stored at BOE was leased and exported. UK gold sell-off could have been a way to safe LBMA.

            German moved 1000mt from London in 2000-2001. This should show up in the UK import/export numbers if they are complete. It doesn’t for 2001. It could be the case that this gold was moved in 2002-2003 but it is not included in the numbers as it counts as monetary gold.

          • rowingboat

            I’m pretty sure it was the HS2-HS4 database, but I’ll check tomorrow (late where I am). My interpretation viewing the data is that it excluded monetary gold (e.g. net imports of 122mt in 1997). The data I extracted goes back to 1988 with net imports from 1988-2000 of 2697mt. From 1988-2014 net imports of 6355mt, which is less than what the LBMA states is now stored in London on their website (7500mt) so that leaves 1200mt imported before 1988; makes sense. However, does the LBMA estimate include central bank holdings???? I suspect it may not.

          • sb

            I am sure LBMA number includes CBs gold. They state that 75% is in BOE vaults, which matches BOE annual report gold in custody 5500mt. In 2013 LBMA claimed that there was 9000mt in London, of which 66% in BOE vault. This also matches BOE 6200mt in custody. 1200mt in London vaults in 1988 is very low number. It doesn’t seem to be correct. One possibility is that Eurostat numbers don’t include CBs moving gold from London.

            Did you crosscheck UK exports to Switzerland with Swiss imports from UK? The Swiss data shows 4200mt imported from the UK between 1988-2000.

          • rowingboat

            Yes, it was the HS2-HS4 database that I accessed under “International Trade Detailed Data” on the Eurostat website. That was months ago but I’ve checked again today and the numbers are the same. I’ve also gone into the CN8 database for different components under the 7108 umbrella:

            1) Eurostat hasn’t reported any UK-Swiss “monetary”
            gold flows from 1988-2014 (inputs are left blank). For that matter the Swiss database hasn’t either, 7108.1200 (non-monetary gold).

            2) Yes, the LBMA number on their website (7500mt London holdings) must include central bank gold for reasons you outline.

            3) Therefore the UK could NOT have net-imported 6355mt (non-monetary gold) from 1988-2014 as reported by

            4) Sure enough, the Eurostat UK export numbers to Switzerland are much less than what the Swiss database reveals Switzerland actually received from the UK, particularly in years 1988-2000. The years 2013-14 are close but there’s still a wide 500mt discrepancy from 2001-12 as well:

            2013-14: 1992mt (Eurostat) versus 2012mt (Swiss database)

            2001-12: 1858mt (Eurostat) versus 2359mt (Swiss database)

            1988-2000: 264mt (Eurostat) versus 4369mt (Swiss database)

            5) Theoretically gold exports from the UK could have passed through another country and this may explain the difference (unlikely). However, Swiss-reported imports from the UK are
            greater than the TOTAL exported gold from the UK to all countries as reported by Eurostat, each year 1988-2000. And this is 7108.1200 non-monetary gold. There would’ve been significant monetary flows in the 1990s too.

            PS: just seen your new message above.

          • sb

            I’ve found some data before 2002 but they are not available for all countries.

            I believe that Eurostat UK data before 2005 are incomplete and not useful. UK net exported 6000 mt 1991-2005. 2000-2005 net exports were probably flat.
            2005 saw low of around 5000 mt gold in London when leasing mess was finally sorted out.

            Some arguments
            a) Eurostat data don’t agree with Swiss imports from the UK

            b) there is a reason why historical HMRC data for gold trade were only uploaded starting from 2005. BOE has also started publishing its gold in custody from 2005.

            c) Turk’s data http://www.fgmr.com/more-proof.html makes more sense and agree more with Swiss data ( net exports in 1991-2002 3500mt agrees with UK exports to Switzerland)

            d) Turk’s data and Swiss data fit the story of CBs in London selling/leasing gold for 20-30 years around 3000mt per decade. This almost caused LBMA to default in 1998-1999. The system was saved by intervention from CBs which sold leased gold. See this guy blogs where he puts together different pieces of the puzzle https://victorthecleaner.wordpress.com/2012/11/26/central-bank-gold-leasing/

          • rowingboat

            Isn’t the Turk data referring to monetary flows? The Swiss import data from UK is non-monetary gold. What the folks crying foul in the 1990s totally missed is that people who bought gold in 1980 fearing hyperinflation and nuclear Armageddon had every reason to sell in 1990. Isn’t this what the Swiss data reflects? 4200mt divestment from West (London) to East (India) via Switzerland until the West decided to buy again early 2000s causing exports from UK to Switzerland to dry up and exports from Switzerland to India to fall given the higher gold prices.

          • sb

            Turk’s data contain financial gold (both monetary and non-monetary) see page 35 http://www.financialsensearchive.com/fsn/presentations/files/IMF_Sept1993.pdf
            (btw notice that Swiss data in this old report matches recently published)

            I think the selling in 1990s is coming from CBs, if you look at total global reserves they started dropping at beginning of 1990s, in total CBs lost 5000mt.

            Not only India was buying. In 1980s and 1990s around 4000mt was exported to UAE and Saudi.

          • rowingboat

            Thanks for that very serious and excellent 1993 paper, sb.
            Turk’s data would have included 940mt German repatriation from UK in 2000/01 plus 2000-04 Swiss sales in part. Flat overall net imports therefore implies significant investment flows to the UK during that time, supported by UK–>Switzerland non-monetary flow drying up (Swiss import data).

            Given the proven unreliability of the earlier UK data, this is how I best view it at the moment. Feb 2006 BOE holdings of 3530mt are central bank gold (there are 72 central banks storing gold at the BOE and I’ve accounted for 2900mt following a Google search of the main holdings below – if you know of others please let me know). UK net inflow from 2006 is investment demand taking the total to 4000mt in Feb 2014 (7500-3500mt) reduced by 400mt in the next 12 months and probably +100mt in March 2015 given Swiss imports from the UK were 97mt in March.

            Proving there’s nothing new under the sun, I’ve plotted Switzerland –> HK/China flow showing gold flowing East during bear markets, slowing and reversing during the bull. The next deceleration / reversal will see fireworks:

            1982-2000: +497mt
            2001-2009: -334mt
            2010-2014: +2386mt

            A Google search of central bank holdings at BOE:

            IMF: 985mt (35%)
            Germany: 445mt
            United Kingdom: 310mt
            India: 265mt
            Austria: 224mt
            Switzerland: 208mt
            Belgium 200mt (?)
            Netherlands: 110mt
            Australia: 80mt
            Sweden: 61mt
            Finland: 25mt
            Italy: 12mt (?)

          • sb

            “4000mt in Feb 2014 (7500-3500mt)” I think 4000mt is too high, you take only 3500mt in the BOE vault as gold in London in 2006. Probably 1000mt was in other vaults, so it is more likely that there was around 4500mt of gold in London in 2006 (in 2014 around 2000mt was in other vaults including GLD gold in the HSBC vault).

            My current understanding of UK data is

            1) before 2005 HMRC published industrial and financial (they call it monetary) gold separately, Turk and 1993 doc took only financial gold. I don’t know what industrial means exactly, it looks rather strange that the UK net imported 200-300mt gold a year for internal consumption.
            2) since 2005 HMRC has started publishing non-monetary gold (industrial + financial) but also keeps publishing old way (industrial and financial). However the definition of industrial gold seems to be changed because data shows a big drop in imported gold.
            You can find financial and industrial gold data
            in United Kingdom Mineral Yearbooks

            3) Eurostat data shows only industrial gold before 2005, then non-monetary gold after

            Some interesting documents showing scale of gold lending in 1980s and 1990s if you haven’t seen them alreadyhttp://www.gata.org/…llStudy1999.pdf

          • KoosJansen

            Great guys!

            *work in progress*, HMRC wrote me once

            For trade prior to 2014, the UK had a long standing exemption from the International reporting requirements for gold. This means that gold held as a ‘store of wealth’ had not been recorded as trade in goods. It had previously been classified as monetary gold, which is excluded from trade in goods as per international guidelines.

            In order to bring the UK recording of non-monetary gold in line with international standards and legislation, the majority of the UK trade originally classified in the monetary gold commodity code (71082000) was reclassified as non-monetary gold. All that Monetary gold re-classified as Non-Monetary gold was put into the comcode 71081310. This was for the period 2005 to 2013 and allows better comparability with the statistics produced by other countries.

            This could explain large differences between tonnage ad value.

          • rowingboat

            I wasn’t very clear, sb. I only meant that the gold in BOE vaults on Feb 28, 2006 (3520mt) was purely central bank holdings (or very close to it) and assumed that central bank holdings at BOE have not changed much since. Therefore the LBMA statement near the time BOE released its annual report in 2014, implies 4000mt held as a financial asset by banks in the London Bullion Market (7500-3520mt) at the audited date (Feb 2014). This 4000mt of financial gold includes net imports, pre- and post- 2006, with much of it stored in BOE vaults after 2006 (overflow?).

            Thanks for the new links, will read them soon. When did “Industrial” imports to the UK start to fall significantly, was it 2007? From the Eurostat data there are relatively minor differences between gross and refined weights of imports, 2007-2014. Yet from 1988-2006 the differences are often substantial, implying a lot of scrap or impure bullion entering the UK during this time. Two observations from the Swiss data, which may be related:

            a) Italian imports from Switzerland declined dramatically during the 2000s with Italy becoming a net exporter to Switzerland in 2009. Could UK’s former “Industrial” gold now be flowing to Italy instead to supply its jewellery industry?

            b) Swiss imports from 1995-2002 were nearly pure, 2-8% difference between gross and refined weight. This difference has increased significantly (15-26%, 2004-2014) so could this “Industrial” gold also be flowing to Switzerland for refining?

          • sb

            Yes I misunderstood you. Thank you for the clarification.

            Based on BOE annula reports, LBMA website and UK export/import data I have for 2005
            3960mt BOE vault
            1200mt other vaults

            Industrial gold UK imports dropped in 2005 to refined 28mt from refined 324mt in 2004. For me this looks like as it was caused by a change in the methodology of reporting data by HMRC .

            a) Based on this article https://www.bullionvault.com/gold-news/italy-gold-101820136 Italian gold industry has declined significantly since 1998.

            b) That is interesting. I haven’t found any useful information on UK gold refining industry in 1990s. It seems unlikely to die so quickly in 2005.

          • rowingboat

            In actual fact the increase between gross and refined weight is mainly attributable to increases of South American imports in the 2000s (dore production). In the year of the largest difference (2010) South America accounts for 410mt of the total 599mt difference. African production probably accounts for the rest but I haven’t quantified that yet. Interestingly imports from South & Central America peaked in 2009 (215mt) and have been falling since (159mt annualised, 2015 Q1).

          • sb

            Central & South America is a playground of USA and Canada miners.

            US imports of gold dore increased by 40 mt of refined gold between 2009 and 2014



          • rowingboat

            Thanks for that. I’ve been doing American research and by my calculation using USGS data, gold holdings in USA have increased by 2,232mt from 1991-2014 (net imports + mine production). A whopping 40% of that occurred in three QE years, 2009/10/11 (881mt) and just like the UK is now being unwound, 260mt since 2012. Given US consumer demand is significant (150mt per yr) then the unwinding process is more advanced than that and I’d expect holdings in USA to increase again soon to meet consumer demand (net exports to fall further squeezing UK and Swiss supplies).

          • sb

            2232mt of net imports + mine production over 23 years that is not a lot if consumption is 150mt/year. I think only the spike of imports after 2008 represents any meaningful investment hoarding. 260mt decline from 2012 is slightly more than the amount of gold removed from Comex vaults.
            I don’t think USA can supply more than a few hundred tonnes of gold from dishoarding in the worst case.

            I am watching now Turkey data. They have switched from being net importers to being net exporters this year. They net exported from January to March around 90mt, which is roughly what they net imported in one year before.

          • rowingboat

            Could this be capital flight you reckon? In Jan-Feb Turkey also exported 14mt to the UK after exporting just 29mt in the last 26 years. At the same time my wife cancelled our vacation plans to Turkey & Jordan (isis fears) and we’re now visiting America instead.

            Have you viewed the Swiss-UAE data yet? Much of those central bank exports from UK flowed to Dubai in 1997 (451mt). The dynamic has changed dramatically after 2001 however, with UAE a slight net-exporter to Switzerland since 2002, including 257mt in 2012. UAE imported 3000mt from Switzerland 1982-2001 (net).

          • sb

            It doesn’t look like as capital flight based on official Turkish data. Turkey has been running trade deficit for many years. Gold exports help a little bit to lower it. TRY has dropped 20% this year which could trigger some CB intervention but there is no big change in Turkish official gold reserves.

            Turkey could potentially supply a lot of gold, it has 500mt of official reserves and a few thousand tonnes among its population.

            There is also a possibility that Turkey its selling somebody else gold e.g. Iranian or Isis.

            I did look at UAE data. I think raising gold prices lowered demand as it happened in Europe. It is also possible that India direct imports reduced buying of gold by Indians in UAE.

            Swiss data shows also how Asian crisis forced South Korea to sell huge quantities of gold. People buy gold in good times than sell it in bad times. Something similar has happened in Europe recently when gold from south of Europe flowed to Germany.

          • rowingboat

            Hi sb, I’ve just posted this on Bron Suchecki’s blog after plotting lots of data this week. I’d be be interested in any comments you may have about it, data you see that I’ve missed etc:

            In recent days I’ve been plotting the Swiss export data. Dominant importers of bullion from Switzerland in the 1980s and 90s were Italy, the Middle East (mainly UAE & Saudi Arabia), Japan and Turkey. From 1982-2005 these countries collectively imported on a net basis almost 15,000mt (600mt pa) but after 2005 their imports have hit a wall, averaging just 20mt per year in the last 10 years.

            In the last several years India + HK/China have made up for falls from these former “Eastern” sources of demand, but there were years during the bull market when large lulls or gaps occurred…. 2002/03/06/07/08/12 & especially 2009 when the above mentioned countries switched (collectively) from importer to net exporter to Switzerland. So rhetorically asking, who was buying if not the East?

            These years correspond with significant falls of imports into Switzerland from USA + Canada + UK, through which much of the world’s mine production flows. We also observe large net imports into the UK during this time, particularly 2009 when 1,100mt were imported. Interestingly 2009 saw the start of large exports from Japan to Switzerland for the first time since 1982 (300mt, 2009/10/11).

            I wish I could post a chart but this demonstrates a clear dynamic in my view of Western supply being choked, gold stocks rising in London, higher prices, decline and eventual reversal of West-Eastern flow. Followed then by large-scale selling by Western hoarders. This selling will likely continue until a) the Western narrative/sentiment changes, or b) the 4000-5000mt imported into the UK, 2001-12 is drained, and we may be half-way through that process by now.

  • Andreas Agorander

    “So the question is, who is currently selling gold to China?”.

    I think you touched on the answer in your post : “Yearly mining output is likely to be sold no matter what the price is”.

    Think about it, more than 3000 tonnes a year, roughly “one german gold reserve” a year is made available, at todays prices, from the weakest hands imaginable. Is this a maintainable price level? That depends for how long the mines can continue to dig up “one germany” each year at current price.

    I’m not convinced we need any further explanation for price and “supply for the chinese” than mining output, and the general sentiment in the west where gold isn’t the default savings vehicle for your typical saver.

    • rowingboat

      This is where the World Gold Council’s data is useful, that being global retail demand for gold (jewelry, bars, coins) not held within the banking system. Excluding India & China (the world’s largest consumers) this demand is about 1500mt annually, last time I checked. India + China together import 2000mt. Global mine production is 2500mt (excluding China, which produces 500mt) so where is the other 1000mt coming from to supply India & China? Answer: from the West’s accumulated holdings, particularly gold held within the bullion banking system (institutional investors) in London. The UK net imported about 5500mt during the bull market (2001-12) and has exported a third of that since (net). Western institutional investors appear to be the swingers, setting price at the margin.

  • Dress

    They don’t need to increase demand physically. They have a trading account with the COMEX and do what they do best – create a lot of paper IOUs.

    • KoosJansen

      For every short on the COMEX there is a long. The COMEX can be used a as steering tool; it doesn’t create infinite paper gold.

      • Dress

        Thanks for your reply, Koos. Your posts are always great.

        There is a long for every short over time but not at any point in time. Two questions for you:
        1.) How much physical gold is delivered on COMEX?
        2.) Is there a limit on how much amount of gold can be shorted?

        • KoosJansen

          If you go short who is on the other side of that contract? A long. Both are always equal.

          1) Not much
          2) Not really

          • Dress

            1.) None. For several years now. The COMEX is a pure paper market.

            2.) None. Gold is the only commodity that can be shorted without any limits on the COMEX.

            The other side of a short contract is a long but there is no long contract.

            “…who is on the other side of that contract?” Exactly. This is the question. And what do you think is on the other side?

          • Dress

            Just for clarification. If I buy a short (contract) from myself I don’t lose any money because it is a zero sum game. No physical gold will change hands. But in the process the price of gold has been suppressed. Q.E.D.!

  • c1ue

    Interesting article.
    Your comment on supply vs. demand is a little too ivory tower though.
    What are the circumstances which drive the West to sell to the East? It isn’t a mystery: the economies of the West are declining – certainly for most of the West’s population if not necessarily nominally. In contrast, the economies of the East are still improving even if the rate has slowed down.
    Thus while I do believe you are correct in that supply is meeting demand – the factors which are driving this supply onto the market do matter.
    And if this thesis is correct, then what is important isn’t the supply or demand of gold per se – it is how much more supply can be driven out of the West vs. how much more demand can China generate.
    I would posit that the demand that China can generate is immense and will continue to grow so long as that economy grows – which it still seems to.
    How much supply available in the West is then available would then seem to be the future limiting factor. Total world gold supply is around 165000 metric tons. CBs hold around 30K metric tons. India supposedly has 18K tons in private hands. China has accumulated thousands of tons in net sales – but how does that compare vs. gold actually in the West?

    • rowingboat

      Very good question. People forget that China only started importing gold significantly since 2011. So who was buying all of the gold before then? It’s an inconvenient question which gold bugs don’t want to address/answer… but much of this historical information is available via national trade statistics.

      There is a precedent to China and that is India who started importing hundreds of tonnes annually in the 1990s. Chinese demand may or may not continue to grow and is a function of the gold price. India is a much richer country than it was two decades ago but gold demand isn’t multiples higher… higher gold prices have limited import growth in terms of tonnage (although value in terms of rupee is much higher of course).

      • Mountain Man

        Because of tariffs fool! The Indian Central bank was threatened by the FED and they cowered and did what they wanted them to do which is raise tariffs on gold!

        • rowingboat

          And what did raising tariffs do? They increased premiums above the spot price and the gold entered anyway via Dubai and other countries instead. There is only one way to stop the flow East and that is higher prices, and this is precisely what happened from 2001-12. Indian imports have been relatively flat for so long because the gold price rose so much in the last 15 years. Nothing to do with tariffs fool.

  • Baruch Obamawitz

    A lot of concern about definitions here. We know gold prices are manipulated downwards because US (and US lackey)corporate-controlled media are constantly propagandizing against gold, and because all asset prices and financial markets are being manipulated to try to prevent an equity price collapse, which a rapid rise in gold prices would trigger. The strategy of the financial parasites is faute de mieux, which gold threatens.

  • Phil Scott

    The use of Samuelson’s imaginary smooth curves of demand and supply is very misleading. I suggest that you try and draw real curves from actual data and you will very quickly discover that the much vaunted supply/demand curves do not relate to reality no matter how much you try to juggle them.

    This is just another fantasy of economists to explain things that don’t exist in reality.

    • KoosJansen

      Do the curves of the lines matter, as long as they cross?

      • Phil Scott

        These are imaginary lines of an imaginary economic world. Thye have no relevance to the real world. They do not add to any discussion or provide proof of anything that is true – that is they do not accord with reality.

        They are a paper money red herring to confuse and confound the unwary.

        Your arguments stand fine and clear without Samuelson’s approval.

        • KoosJansen

          The visual cross helps people get the relationship

          • Phil Scott

            I don’t think it does. BUT it continues the myth that an economy is in a state of equilibrium and that clever central bankers and economists can keep it in equilibrium.

            Please see my new book – Real Economics – Creating wealth and keeping it.

Copyright Information: BullionStar permits you to copy and publicize blog posts or quotes and charts from blog posts provided that a link to the blog post's URL or to https://www.bullionstar.com is included in your introduction of the blog post together with the name BullionStar. The link must be target="_blank" without rel="nofollow". All other rights are reserved. BullionStar reserves the right to withdraw the permission to copy content for any or all websites at any time.