Tag Archives: GFMS

The Great Physical Gold Supply & Demand Illusion

Gold supply and demand data published by all primary consultancy firms is incomplete and misleading. The data falsely presents gold to be more of a commodity than a currency, having caused deep misconceptions with respect to the metal’s trading characteristics and price formation.

Numerous consultancy firms around the world, for example Thomson Reuters GFMS, Metals Focus, the World Gold Council and CPM Group, provide physical gold supply and demand statistics, accompanied by an analysis of these statistics in relation to the price of gold. As part of their analysis the firms present supply and demand balances that show how much gold is sold and bought globally, subdivided in several categories. It’s widely assumed these balances cover total physical supply and demand, which is incorrect as the most important category is excluded. The firms though, prefer not to share the subtle truth or their business models would be severely damaged.

The supply and demand balances by the firms portray gold to be more of a commodity than a currency, as the gist of the balances reflect how much metal is produced versus consumed – put differently, the firms mainly focus on how much gold is mined versus how much is sold in newly fabricated products. However, in reality gold is everlasting and cannot be consumed (used up), all that has ever been mined is still above ground carefully preserved in the form of bars, coins, jewelry, artifacts and industrial products. Partly because of this property the free market has chosen gold to be money thousands of years ago, and as money the majority of gold trade is conducted in above ground reserves. Indisputably, total gold supply and demand is far in excess of mine production and retail demand.

As most individual investors, fund managers, journalists, academics and precious metals analysts consider the balances by the firms to be complete, the global misconception regarding gold supply and demand is one of epic proportions. Physical gold is a profound anchor in our global financial system and thus it’s of utmost importance we understand the fine details of its trading characteristics. 

Supply & Demand Metrics By The Firms

The firms can argue that the difference between what they present as supply and demand (S&D), as opposed to what I deem to be a more unadulterated approach of S&D is due to contrasting metrics. Accordingly, we’ll discuss their metrics to reveal their infirmity. In a nutshell, the firms only count the physical gold S&D flows that are easy to measure, while leaving out the most important part: institutional supply and demand. 

Although the firms all have slightly different methodologies to measure S&D, from comparisons the numbers appear to be quite similar. For our further investigation we’ll spotlight the metrics and models by GFMS. The reason being, GFMS has been the only firm that was willing to share a full description of their methodology for publication – to be viewed here. Metals Focus (MF) provided a partial methodology, the World Gold Council and CPM Group declined to comment.

Let’s have a look at GFMS its S&D categories. On the supply side is included:

  • Mine supply (newly mined gold)
  • Scrap supply (gold sourced from old fabricated products)

On the demand side is include:

  • Jewelry demand (gold content used in newly manufactured jewelry products bought locally at retail level, adjusted by jewelry exported and imported).
  • Industrial demand (the volume of gold used in industrial applications, for example bonding wire, products used in semiconductors/electronics and dental alloys).
  • Retail bar investment (the net volume of bars that are purchased by individual investors through retail channels).
  • Coin investment (a combination of published data from mints and also a proprietary survey conducted by GFMS detailing where coins are sold).

The above four demand categories summed up are often referred to as “consumer demand” by the firms.

Furthermore GFMS includes:

  • Net hedging (change in physical market impact of mining companies’ gold loans, forwards, and options positions)
  • Net official sector (total central bank selling or buying)
  • ETF inventory build (change in ETF inventory)
  • Exchange inventory build (change in exchange inventory)

The last four categories can be either supply or demand. In example, when central banks (the official sector) in total are net sellers this will be listed as a negative demand figure, as is shown in the S&D balance by GFMS below from 2006 until 2009, when central banks in total are net buyers this will be listed as a positive demand figure, as is shown in the balance from 2010 until 2015. For a clear overview of the GFMS S&D balance please have a look at all line items below.

Exhibit 1. Courtesy GFMS. Global gold S&D balance as disclosed in the Gold Survey 2016.

According to GFMS Supply consists of Mine production, Scrap and Net Hedging. In turn, Demand consists of Jewelry, Industrial Fabrication, Retail Investment, and Net Official Sector. After balancing Supply and Demand this results in a Physical Surplus/Deficit. Then, ETF Inventory Build and Exchange Inventory Build are added/subtracted from the Physical Surplus/Deficit to come to a Net Balance.

GFMS likes to pretend their balance is complete and occasionally articulates any surplus or deficit arising from it is positively correlated to the price of gold, which is anything but true, as I will demonstrate step by step.  

The Firms Exclude Majority Gold Supply & Demand

Most important what’s excluded from the balance is what we’ll refer to as institutional supply and demand, which can be defined as trade in bullion among high net worth individuals and institutions. Usually the bullion in question comes in 400-ounce (12.5 Kg) London Good Delivery (GD) bars having a fineness of no less than 995, or smaller 1 Kg bars having a fineness of no less than 9999. In addition, bullion bars can weigh 100-ounce or 3 Kg, among other less popular sizes, generally having a fineness of no less than 995. Bullion can be traded without changing in weight or fineness, but it can be refined and/or recast for transactions as well, in example from GD bars into 1 Kg bars. In some cases institutional supply and demand involves cross-border trade, when bullion is sold in country A to a buyer in country B, in other cases the bullion changes ownership without moving across borders.

Provided are two exemplifications of institutional S&D:

  • An (institutional) investor orders 400 Kg of gold in its allocated account at a bullion bank in Switzerland – which would be purchased in the Swiss wholesale market most likely in GD bars. This type of S&D will not be recorded by GFMS.
  • A Chinese (institutional) investor buys 100 Kg of gold directly at the Shanghai Gold Exchange (SGE), the Chinese wholesale market, in 1 Kg 9999 bars and withdraws the metal from the vaults. Neither this transaction will be registered by GFMS – or any other firm.

These examples show the S&D balances by GFMS are incomplete.

For illustrational purposes, below is a chart based on all S&D numbers by GFMS from 2013, supplemented by my conservative estimate of institutional S&D. Including institutional transactions total S&D in 2013 must have reached well over 6,600 tonnes.

Exhibit 2. Global gold S&D 2013 by GFMS, including conservative estimate institutional S&D.

GFMS Covers The Tracks With Help From The LBMA

Although GFMS intermittently admits their number are incomplete (they have to), at the same time they’ve been battling for years to eclipse apparent institutional S&D for its audience. Dauntless tactics were needed when in 2013 institutional demand in China reached roughly 1,000 tonnes and over 500 tonnes in Hong Kong. Institutional demand in the East was predominantly sourced through GD bars from the London Bullion Market, which were refined into 1 Kg 9999 bars that are more popular in Asia. For the cover up GFMS went to great lengths to refute the volumes of gold withdrawn from SGE vaults, and accordingly have the London Bullion Market Association (LBMA) adjust statistics on total refined gold by its member refineries. Remarkably, the LBMA cooperated. Allow me to share my analysis in detail.

In 2013 something unusual happened in the global gold market as Chinese institutional demand exploded for the first time in history. Hundreds of tonnes of institutional supply from London in the form of GD bars were mainly shipped to Switzerland to be refined in 1 Kg 9999 bars, subsequently to be exported via Hong Kong to meet institutional demand in China. From customs data by the UK, Switzerland and Hong Kong the institutional S&D trail was clearly visible. From 2013 until 2015 there was even a strong correlation between the UK’s net gold export and SGE withdrawals. Demonstrated in the chart below.

Exhibit 3. Correlation between UK net gold export and SGE withdrawals.

Because of the mechanics of the gold market in China, Chinese institutional demand roughly equals the difference between the amount of gold withdrawn from SGE designated vaults (exhibit 4, red bars) and Chinese consumer demand (exhibit 4, purple bars). In the exhibit 4 below you can see this difference that brought GFMS in a quandary, especially since 2013. For more information on the workings of the Chinese gold market and the size of Chinese institutional demand please refer to my post Spectacular Chinese Gold Demand Fully Denied By GFMS And Mainstream Media.

Exhibit 4. Chinese wholesale demand (SGE withdrawals), versus GFMS consumer demand versus apparent supply.

Stunningly, since 2013 GFMS has tried to convince its readers through numerous arguments why SGE withdrawals crossed 2,000 tonnes for three years in a row, while Chinese consumer demand reached roughly half of this. Yet the arguments have failed miserably to explain the difference – they rationalize only a fraction, read this post for more information.

And GFMS did more to eclipse apparent institutional S&D. They colluded with the LBMA.

To be clear, I cannot exactly measure global institutional S&D. However, let me make an estimate of apparent institutional demand for 2013. Notable, in 2013 a flood of gold crossed the globe from West to East. Chinese institutional demand accounted for 914 tonnes and Hong Kong net imported 579 tonnes – the latter we’ll use as a proxy for additional Asian institutional demand, as Hong Kong is the predominant gold trading hub in the region. 

In total apparent institutional demand in 2013 accounted for (914 + 579) 1,493 tonnes. If we add all other demand categories by GFMS shown in exhibit 1, total demand in 2013 was at least 6,619 tonnes. Be aware, this excludes non-apparent institutional demand.

Exhibit 5. Global gold demand 2013 by GFMS, including apparent institutional demand.

Because nearly all wholesale gold demand in Hong Kong and China is for 1 Kg 9999 bars, the global refining industry was working overtime in 2013, mainly to refine institutional and ETF supply in GD bars coming from London. In December 2013 I interviewed Alex Stanczyk of the Physical Gold Fund who just before had spoken to the head of a Swiss refinery. At the time Stanczyk told me [brackets added by me]:

They put on three shifts, they’re working 24 hours a day and originally he [the head of the refinery] thought that would wind down at some point. Well, they’ve been doing it all year [2013]. Every time he thinks it’s going to slow down, he gets more orders, more orders, more orders. They have expanded the plant to where it almost doubles their capacity. 70 % of their kilobar fabrication is going to China, at apace of 10 tonnes a week. That’s from one refinery, now remember there are 4 of these big ones [refineries] in Switzerland.

As a consequence, statistics on “total refined gold production” in 2013 by “LBMA accredited gold refiners who are on the Good Delivery List”, which the four large refineries in Switzerland are part off, capture the immense flows of institutional S&D – next to annual mine output and scrap refining. On May 1, 2015, the LBMA disclosed total refined gold production by its members at 6,601 tonnes for 2013 in a document titled A guide to The London Bullion Market Association. It’s no coincidence this number is very close to my estimate on total demand (6,619 tonnes), as apparent institutional demand in Asia was all refined from GD into 1 Kg bars.

Here’s exhibit 2 from another angle.

Exhibit 6. Global gold S&D by GFMS, including apparent institutional S&D, versus total refined gold production 2013.

In the table below we can see the LBMA refining statistics for 2013 at 6,601 tonnes.

Exhibit 7. Courtesy LBMA. Screenshot from A guide to The London Bullion Market Association captured by Ronan Manly in May 2015.

After this publication GFMS was trapped; these refining statistics revealed a significant share of the institutional S&D flows they had been trying to conceal. What happened next – I assume – was that GFMS kindly asked the LBMA to adjust downward their refining statistics. First and painstakingly exposed by my colleague Ronan Manly in multiple in-depth posts, the LBMA kneeled and altered its refining statistics to keep the charade in the gold market going.

On August 5, 2015, the LBMA had edited the aforementioned document, now showing 4,600 tonnes in total refined gold production. (Click here to view the original LBMA document from the BullionStar server, and here to view the altered version from the BullionStar server.) Have a look.

Exhibit 8. Courtesy LBMA. Altered document on refining statistics by the LBMA August 2015.

In the altered version it says:

Total refined gold production by the refiners on the List was estimated to be 4,600 tonnes in 2013, owing to recycling of scrap material, above world mine production of 3,061 tonnes (source Thomson Reuters GFMS).

A few important notes:

  • In the altered version the LBMA mentions “an estimate” for “total refined gold production”, while it doesn’t need to make an estimate as all LBMA accredited gold refiners who are on the Good Delivery List are required to provide exact data to its parent body. The exact data was disclosed in the first version of A guide to The London Bullion Market Association, and it stated, “total refined gold production by the refiners on the List was 6,601 tonnes”.
  • In the altered version the LBMA states the refining statistics were sourced from Thomson Reuters GFMS, but the LBMA doesn’t need GFMS for these statistics. The fact they mention GFMS, though, suggests a coordinated cover up of institutional S&D. Not only the firms, also the LBMA publishes incomplete and misleading data.
  • The altered version stated refining production totaled 4,600 tonnes, which is a round number and obviously quickly made up. A few weeks after the numbers were adjusted, the LBMA adjusted the numbers again, this time into 4,579 tonnes (click here to view from the BullionStar server). Clearly, on several occasions there has been consultation with the LBMA to get the statistics in line with GFMS.
  • In the original document the LBMA states, “Total refined gold production by the refiners on the List was 6,601 tonnes in 2013, more than double world mine production of 3,061 tonnes”, while in the altered version they state, “Total refined gold production by the refiners on the List was estimated to be 4,600 tonnes in 2013, owing to recycling of scrap material, above world mine production of 3,061 tonnes”. Notable, GFMS prefers to have total supply focused around mine and scrap production, instead of including institutional supply.
  • The original refining statistics (6,601 tonnes) are still disclosed in the LBMA magazine The Alchemist (#78 on page 24), to be viewed from the LBMA server here.
  • The fine details about how often and when the LBMA changed its refining statistics can be read in Ronan Manly’s outstanding post Moving the goalposts….The LBMA’s shifting stance on gold refinery production statistics.

And so nothing is spared in trying to uphold the illusion of the GFMS S&D balance to be complete. In another example GFMS excluded gold purchases by the central bank of China from its S&D balance. In June 2015 the People’s Bank Of China (PBOC) increased its official gold reserves by 604 tonnes, from 1,054 tonnes to 1,658 tonnes. During that quarter (Q2 2015) all other central banks worldwide were net buyers at 45 tonnes. Thus, in total the Official Sector was a net buyer at 649 tonnes. Now, let’s have a look at GFMS’ S&D balance for Q2 2015:

Exhibit 9. Courtesy GFMS. Global gold S&D balance as disclosed in the Gold Survey 2015 Q2. 

Net Official Sector purchases are disclosed ay 45 tonnes. GFMS decided not to include the 604 tonnes increment by the PBOC simply because it didn’t fit their balance model. A 604 tonnes increment in would have set the “net balance” at -480 tonnes. Readers would have questioned the balance from this outlier, and so GFMS decided not to include the tonnage.

According to my sources PBOC purchases were sourced from institutional supply (from abroad and not through the SGE), which is a supply category not disclosed by GFMS and therefore the tonnage was a problem. (Note, GFMS disclosed the PBOC increment in text, but not in their balance.) For more information read my post PBOC Gold Purchases: Separating Facts from Speculation.

Gold Is More A Currency Than A Commodity

The biggest flaw of the balance model by GFMS is that it depicts gold to be more of a commodity than a currency. It’s focused on mine output and gold recovered from old fabricated products on the supply side, versus retail sales of newly fabricated products on the demand side. In parlance of the firms, how much is produced (supply) versus consumed (demand). Official sector, ETF and exchange inventory changes are then added to the balance. This commodity S&D balance approach by GFMS has caused deeply rooted misconceptions about the essence of gold and its price formation.

The price of a perishable commodity is mainly determined by how much is annually produced versus how much is consumed (used up). However, gold is everlasting, it cannot be used up and its exchange value is mainly based on its monetary applications, from being a currency, or money if you will. Logically the best part of its trading is conducted in above ground reserves. From my perspective the impact of global mine supply, which increases above ground stocks by roughly 1.5 % annually, and retail sales have less to do with gold’s price formation than is widely assumed.

Back to GFMS. Have a look at the picture below that shows their S&D flows for 2015. 

Exhibit 10. Courtesy GFMS. The global S&D flows for 2015.

GFMS pretends total supply is mine production plus some scrap, which is then met by jewelry demand in addition to retail investment, industrial fabrication and official sector purchases. The way they present it is misleading. These S&D flows are incomplete; they suggest gold is traded like any other commodity. But what about institutional S&D in above ground bullion? Trades that define gold as an international currency.

Let’s do another comparison; this time between what GFMS calls Identifiable Investment demand, consisting of…

  • Retail bar & coin
  • ETF demand

…versus my what I deem to be a more unadulterated approach of investment demand, consisting of…

  • Retail bar & coin
  • ETF demand
  • Institutional demand

According to my estimates, in 2015 apparent Chinese institutional demand accounted for roughly 1,400 tonnes (exhibit 4). In the Gold Survey 2016 GFMS states on page 15 [brackets added by me]:

Total [global] Identifiable Investment, … posted a modest 5 % increase in 2015, to reach 990 tonnes.

That’s quite a tonnage between global Identifiable Investment by GFMS at 990 tonnes and apparent Chinese institutional demand at 1,400 tonnes. We should also take into account non-apparent institutional demand, gold that changes hands in trading hubs like Switzerland. Unfortunately we can’t always measure institutional S&D, but that doesn’t justify denying its subsistence.

Have a look at the chart below that shows the large discrepancy. In the next chapter we’ll specifically discuss the significance of investment demand in relation to the price of gold.

Exhibit 11. Global Gold Investment Demand 2015.

My point being: what many gold market participants and observers think is total supply and demand is just the tip of the iceberg. This truly is a staggering misconception created by the firms.

The global gold market. H/t Dan Popescu.

When observing the GFMS balance in exhibit 1 its incompleteness is self-evident. At the bottom we can see the line item “net balance”, which reflects the difference between total supply and total demand. According to GFMS, if the “net balance” is a positive figure there was a surplus in the global gold market, and if “net balance” is a negative figure the market has been in deficit. In the real world this figure is irrelevant. Gold supply and demand are by definition always equal. One cannot sell gold without a buyer, and one cannot buy gold without a seller. Furthermore the gold market is deep and liquid. So how come there is a difference between total supply and total demand in the GFMS balance? As I’ve demonstrated before, because GFMS doesn’t include institutional S&D that in reality makes up for the difference and far beyond. In all its simplicity the “net balance” item reveals their data is incomplete.

Let’s have another stab at this. How can “net balance” exist in the real world, for example in 2009? According to GFMS the gold market had a 394 tonnes surplus in 2009. But how? Were miners left with 394 tonnes they couldn’t sell? Or some supranational entity decided to soak up the surplus to balance the market? Naturally, this is not what happens. Total supply and total demand are always equal, but GFMS doesn’t record all trades.

Moreover, in my opinion the words “surplus” and “deficit” do not apply to gold. There can be no deficit in gold; there will always be supply. At the right price that is. Sometimes Keynesian economists claim there is not enough gold in the world for it to serve as the global reserve currency. Austrian economists then respond by saying that there will always be enough gold at the right price. I agree with the Austrians and their argument also validates why there can be no deficit in gold.  

There is more proof the “net balance” item presented by GFMS is meaningless. Although according to GFMS the market had a 394 tonnes “surplus” in 2009 the price went up by 25 % during that year. This makes no economic sense. A surplus suggests a declining price, not the other way around. Tellingly, S&D forces presented in GFMS balances are often negatively correlated to the gold price, as was the case in 2005, 2006, 2009, 2010 and 2014 (exhibit 1). In conclusion, GFMS S&D balances are not only incomplete, the resulting “net balance” items are misleading with respect to the price. Below are a few charts that demonstrate this conclusion.

If we plot “net balance” versus the end of year price of gold we can see the correlation is often negative. Have a look below. Green “net balance” chart bars show a positive correlation to the gold price, red chart bars show a negative correlation (note, the left axis is inverted for a more clear overview between any “deficit/surplus” and the price of gold). As you can see nearly half of the “net balance” chart bars are negatively correlated to the price of gold.

Exhibit 12. GFMS’ gold market “net balance” versus the gold price. We can quarrel if the “net balance” in 2014 was positively or negatively correlated to the price. I say the correlation was negative as the gold price in 2014 remained flat in US dollars but was up in all other major currencies, in contrast to the “surplus” presented by GFMS.

Mind you, although the “net balance” item is often negatively correlated to the gold price, in the Gold Survey 2016 GFMS states on page 9:

In terms of the Net Balance, 2015 marked the third year in which the gold market remained in surplus, and therefore it is not surprising that the bear market continued.     

And on page 14:

The forecast reduction in global mine output and a gradual recovery in demand will see the physical surplus narrow in 2016, providing support to the gold price and laying the foundation for better prospects.  

GFMS likes to pretend any “surplus” or “deficit” arising from their balance is correlated to the price, but the facts reveal this is not true.

Let us plot the “physical surplus/deficit” line item by GFMS (exhibit 1) versus the gold price. This results in even more negative correlations.

Exhibit 13. GFMS gold market “physical surplus/deficit” versus gold price.

This exercise reveals that a positive correlation between either a “surplus” or “deficit” arising from a GFMS balance and the price of gold is just a coincidence. No surprise when one is aware their S&D data is incomplete.

Remarkably, the last chart was also published in the Gold Survey 2016, but GFMS chose not to invert the left axis and doesn’t disclose what we see is a surplus or deficit. As a result the largest surpluses (2006, 2007, 2009, 2010) seem to correlate with a rising price, though in reality they did the opposite. Compare the chart below with the one above.

Exhibit 14. Courtesy GFMS.

GFMS also publishes S&D balances for silver (a monetary metal that is comparable to gold). For silver the presented correlations by GFMS between a “surplus” or “deficit” in relation to the price are even weaker.

Exhibit 15. GFMS silver market “net balance” versus silver price, as disclosed in the Silver Survey 2016.
Exhibit 16. GFMS silver market “physical surplus/deficit” versus silver price, as disclosed in the Silver Survey 2016.

According to GFMS the silver market is always in deficit, but the price goes up and down. Obviously GFMS neglects to measure institutional S&D for silver. 


In my opinion, when Gold Fields Mineral Services (GFMS) was erected many decades ago they made a mistake to adopt a commodity S&D balance approach. Surely with the best intentions they gather intelligence and retrieve data from the market. But we must be aware this is not the full picture. The most significant data is not disclosed by GFMS.

When it comes to what drives the price of gold GFMS and I agree it’s determined by gold’s role as a currency in the global economy. When reading the chapter PRICE AND MARKET OUTLOOK in the Gold Survey 2016, GFMS shares its insights with respect to the gold price. Factors mentioned are:

  • Turmoil in global stock markets
  • A Chinese hard landing
  • Geopolitical tensions in the Middle-East
  • Central bank stimulus (QE)
  • Global economic weakness
  • Interest rates policy by central banks
  • Low risk asset / safe haven demand

So if these factors drive the gold price, in what S&D category would this materialize? Would (large) investors buy and sell jewelry? Or bullion bars? I think the latter. According to my analysis the price of gold is largely determined by institutional demand, and to a lesser extent ETF and retail bar & coin demand.

Let’s do an exercise to see what physical gold S&D trends correlate to the price. The majority of supply on the GFMS balance consists of mine output and the majority of demand on the GFMS balance consists of jewelry consumption. But if we plot these volumes versus the price of gold in a chart, there is no push and pull correlation. For example, when the gold price surged from 2002 until 2011 jewelry consumption was not rising. Neither was it outpacing mine supply. The opposite happened, to be seen in the graph below. This is because jewelry demand is price sensitive – when the price goes up jewelry demand goes down, and vice versa. Jewelry demand is not driving the price of gold.

Exhibit 17. GFMS retail demand, versus mine and scrap supply versus the gold price.

I also added retail bar & coin demand. Interesting to see is that retail bar & coin demand is on one hand a price driver, moving up and down in sync with the gold price, on the other hand it can be price sensitive having brief spikes when the price of gold declines.

The best correlation between physical S&D in relation to the gold price can be seen in institutional and ETF S&D. One of the largest gold trading hubs in the West is the UK, home of the London Bullion Market that also vaults the largest ETF named GLD. The UK has no domestic mine production, no refineries and national gold demand is neglectable in the greater scheme of things. Therefore, by measuring the net flow of the UK (import minus export) we can get a sense of Western institutional and ETF demand and supply. For example, if the UK is a net importer – import demand being greater than export supply – that signals a net pull on above ground stocks. Approximately one third of the UK’s net flow corresponds to ETF inventory changes, the other two thirds reflect pure institutional S&D.

Exhibit 18. UK net flow versus the gold price.
Exhibit 19. UK net flow, GLD inventory change, gross import and gross export versus the gold price.

In the charts above we can observe a remarkable solid correlation between the UK’s net flow and the gold price. The UK is a net importer on a rising price and net exporters on declining price. The shown correlation can’t be a coincidence, though there’s no guarantee it will prevail in the future.

The two charts above show the gold price is mostly determined by institutional supply and demand in above ground reserves. Effectively, GFMS is hiding the most important part of global physical gold flows.

When I asked an analyst at one of the leading firms why his company doesn’t measure institutional S&D he told me candidly, “because it’s extremely difficult to accurately estimate it”. And it is. As I wrote previously, I can’t exactly measure global institutional S&D either. However, very often publicly available information gives us a valuable peek at it, and it shows to be more relevant to the gold price than what the firms keep staring at. Not knowing exactly what institutional S&D accounts for doesn’t mean GFMS shouldn’t pay attention to it.

But the firms keep trying to uphold the illusion the data they’ve been selling for decades is complete. For if they would plainly confess it was incomplete, future business could be severely damaged.

What I blame these firms is that they’ve created a meme that the gold market is as large as annual mine supply. This has caused all sorts of misconceptions. Often I read analyses based on a comparison between quantitative demand and mine output. Such analyses are likely to jump erroneous conclusions.

H/t Ronan Manly, Bron Suchecki, Nick Laird from Goldchartsrus.com


Simplified overview gold flows 2015:


Spectacular Chinese Gold Demand 2015 Fully Denied By GFMS And Mainstream Media

Debunking the Thomson Reuters GFMS Gold Survey 2016 report. New information provides a more detailed perspective on the Chinese domestic gold market. 

In the Gold Survey 2016 report by GFMS that covers the global gold market for calendar year 2015 Chinese gold consumption was assessed at 867 tonnes. As Chinese wholesale demand, measured by withdrawals from Shanghai Gold Exchange designated vaults, accounted for 2,596 tonnes in 2015 the difference reached an extraordinary peak for the year. In an attempt to explain the 1,729 tonne gap GFMS presents three brand new (misleading) arguments in the Gold Survey 2016 and reused one old argument, while it abandoned five arguments previously put forward in Gold Survey reports and by GFMS employees at forums. Very few of all these arguments have ever proven to be valid, illustrated by the fact that GFMS perpetually keeps making up new ones, and thus gold investors around the world continue to be fooled about Chinese gold demand. For some reason GFMS is restrained in disclosing that any individual or institution in China can directly buy and withdraw gold at the Shanghai Gold Exchange, which is the most significant reason for the discrepancy in question.

According to my estimates true Chinese gold demand in 2015 must have been north of 2,250 tonnes.

The reason I keep writing about this subject (the discrepancy in question) is that it eventually will enable me to show that global physical gold supply and demand as presented by GFMS is just the tip of the iceberg. And, as stated in my previous post true physical supply and demand is far more relevant to the gold price than the numbers by GFMS.

New Information has enabled me to shine a fresh light on the Chinese domestic gold market, so we’ll zoom in once again to get the best assessment of the mechanics of this market. This post is part two of an overview of the Chinese gold market for 2015. In the first part we focused on the (paper) volumes traded on the Shanghai Gold Exchange (SGE) and Shanghai International Gold Exchange (SGEI). In this post we’ll focus on the size and mechanics of the Chinese physical gold market, while at the same time addressing the fallacious information in the Gold Survey 2016 (GS2016).

The Gold Surplus In China According to GFMS

First, let’s have a look at an overview of the key supply and demand data points for 2013, 2014 and 2015, as disclosed in Gold Survey reports by GFMS.

GFMS SD 2013-14-15
Exhibit 1.

Without GFMS mentioning the volume of SGE withdrawals for 2015 (2,596 tonnes) in the GS2016 they disclose apparent supply in the Chinese domestic gold market at 2,293 tonnes. Mine output accounted for 458 tonnes (page 22), scrap supply for 225 tonnes (page 36) and net import was 1,610 tonnes (page 54). The latter is incorrect because GFMS has double counted 63 tonnes Australia exported to China, as demonstrated in my post Australia Customs Department Confirms BullionStar’s Analysis On Gold Export To China, but the let’s not nitpick.

On other pages in the GS2016 we read total (consumer) demand for 2015 was 867 tonnes (page 52), consisting of retail bar demand at 199 tonnes (page 52) and gold fabrication at 668 tonnes (page 41). According to their own data there was a surplus of 1,426 tonnes (2,293 – 867) in the Chinese gold market. Whilst, in 2013 the surplus accounted for 826 tonnes and in 2014 for 917 tonnes, according to data disclosed in previous Gold Survey reports. Meaning, in the past three years GFMS has observed 3,169 tonnes (826 + 917 + 1,426) that were supplied to China not to meet demand, but for reasons that are constantly changing - wait till we get to the plea.

Remarkably, in the GS2016 report GFMS writes:

Hong Kong remained the primary conduit of Chinese gold imports, though its share has been contracting since 2013 … Gold import from this conduit was traditionally regarded as a simple proxy to estimate Chinese consumption … The declining dominance of Hong Kong and the increasing proportion directly routed into Beijing and Shanghai therefore points to the necessity of changes on methodology to calculate Chinese gold demand.

Exhibit 2.

GFMS states that when all Chinese imports came in through Hong Kong this inflow was “regarded as a simple proxy to estimate Chinese consumption”, but now gold is also being imported directly from countries like Australia, the UK and Switzerland, such inflow “points to the necessity of changes on methodology to calculate Chinese gold demand”. How can it be that a couple of years ago Chinese gold import from Hong Kong reflected demand, but a few years later direct massive additional import from the UK and Australia does not reflect demand?

As you probably know (otherwise you can read it here) most of the gold supply in China flows through the SGE. Consequently wholesale demand can be measured by the amount of gold withdrawn from SGE designated vaults. Comparable to the difference between apparent supply and consumer demand shown in exhibit 1, is the difference between SGE withdrawals and consumer demand – the latter being even wider.

Chinese domestic gold market S&D 2015
Exhibit 3.

In the GS2016 GFMS has written a chapter fully dedicated to the humongous difference between SGE withdrawals and their assessment of demand. The chapter is titled “A Review And Explanation Of How China’s SGE’s Withdraw Numbers Are Impacted By Other Trading Activities”. In this post we’ll only briefly discuss whether the arguments are valid, as one of them has to do with China’s highly complex VAT system and I like to expand on this subject in detail in a separated post. However, we’ll expose more of the mechanics of the Chinese domestic gold market in this post, which conveniently demonstrates why nearly all the arguments by GFMS that will be discussed later on are bogus.

This might surprise you, but I actually had fruitful correspondence in the past months with a Senior Precious Metals Analyst at GFMS and a Senior Analyst at Metals Focus (MF). Both gentlemen have been very helpful in sharing their methodology for computing (Chinese) physical supply and demand data. I have to say both of them have answered all my questions. This service is seldom provided by the the World Gold Council, the Bank Of England or the London Bullion Market Association. Based on the information shared by GFMS and MF I’ve refined my view on our on-going disagreement with respect to the Chinese gold demand.

The Mechanics Of The Chinese Domestic Gold Market And Estimating True Chinese gold demand.

Let us refresh our memory regarding the structure of this market. In the Chinese domestic gold market nearly all physical gold supply and demand flows through the SGE because all bullion import1 into the domestic market is required to be sold first through the SGE and there are rules and tax incentives that funnel nearly all domestic mine output and scrap supply through the central bourse. As gold in the Chinese domestic market is not allowed to be exported1, the amount of gold withdrawn from SGE designated vaults therefore serves as a decent indicator for wholesale demand.

However, there are a few possibilities through which SGE withdrawals can be distorted for measuring demand.

  • If metal is in some manner recycled2 through the central bourse. When gold is bought and withdrawn from the SGE vaults and promptly sold and deposited into SGE vaults (for example though process scrap), these flows would inflate SGE withdrawals while not having a net effect on the price of gold, hence the related supply and demand volumes would be deceiving. Although article 23 from the Detailed Rules for Physical Delivery Of the Shanghai Gold Exchange states that bars withdrawn from SGE designated vaults are not allowed to re-enter these vaults, this rule does not fully prevent gold from being recycled through the exchange. If bars withdrawn are re-melted and assayed by an SGE approved refinery they are allowed back into the vaults. And thus, some recycled gold can inflate SGE withdrawals as a measure for true demand.
  • Another example that could distort SGE withdrawals is when international members would withdraw gold from vaults of the SGEI in the Shanghai Free Trade Zone2 (SFTZ), these withdrawals are included in the total SGE withdrawal figure, to store elsewhere in the SFTZ or export abroad1.

For ease of reference we’ll label the amount of gold recycled through the SGE that has no net effect on the price, and gold withdrawn from SGEI vaults that is not imported into the Chinese domestic market as distortion2.

Therefor, in order for us to make the best estimate of true Chinese gold demand we should subtract the amount of distortion from SGE withdrawals. The crux of true Chinese gold demand is establishing the amount of distortion, that’s it.

Previously I assumed the scrap numbers by GFMS mainly reflected gold that was making it’s way back to the SGE and these flows included disinvestment. Both assumptions appeared to be false.

  1. Scrap numbers from GFMS and MF, although they’re certainly not equal, are collected from refiners that are not all SGE members. Implying not all refineries scrap is making its way to the SGE, but is sold through other channels.
  2. Scrap numbers from GFMS and MF include jewelry and industrial products sold back from consumers, they do not include disinvestment that flows directly through refineries to the SGE. GFMS does measure disinvestment at retail level, for example, when people sell bars back to banks these will get netted out to compute net retail bar demand. But if an affluent investor or institution wants to sell (disinvest) 500 Kg they’re likely to approach a refinery directly.

In my nomenclature “distortion2” is the part that inflates SGE withdrawals as a measure for demand, “scrap” is supply from sold fabricated products like old jewelry, and “disinvestment” is supply coming from investment bars sold directly to refineries making its way to the SGE.

As a consequence, these new insights regarding scrap and disinvestment supply have changed my perspective on the Chinese supply and demand balance.

To reach a more clear understanding of what was just described, I’ve conceived an exemplar graph to visually interpret the Chinese physical gold supply and demand balance. Have a look.

Example Chinese domestic gold market S&D
Exhibit 4.

As you can see in the graph above total supply and total demand are exactly equal, this is because one cannot sell gold without a buyer or buy gold without a seller. Consequently we can gauge demand by measuring supply. Please note, in the supply and demand balance shown above, and in our further investigation, two elements are left out. On the supply side I left out stock carry over in SGE vaults from previous years, as this information is not publicly available. On the demand side I left out gold bought at the SGE that was not withdrawn from the vaults, as this information is also unknown.

In all its simplicity the example chart shows that the difference between consumer demand and true Chinese gold demand is caused by direct purchases from individual and institutional clients at the SGE. While GFMS merely counts demand at retail level, by jewelry and bar sales at shops and banks, the real action is at wholesale level, at the SGE.

GFMS fully neglects direct purchases at the SGE (demand) and any corresponding disinvestment to the SGE (supply). Hence our disagreement.

Estimate of SGE withdrawals distribution
Exhibit 5. According to my analysis the PBOC does not buy its gold through the SGE. For more information read my post PBOC Gold Purchases: Separating Facts from Speculation.

At the moment the SGE has almost 10,000 institutional and over 8.3 million individual clients that are able to buy gold directly at the SGE. There is even an SGE smartphone application called “Yijintong” that allows anyone with an internet connection to open an SGE account and trade directly on the SGE wholesale platform enjoying the lowest spreads in China. Furthermore, the SGE counts 183 domestic members and 63 international members.

Exhibit 6. Screenshot Yijintong.

From the SGE (December 2015):

Yijintong is the first professional mobile terminal of state-level gold market jointly researched and developed by the Gold Exchange and all its members …

Yijintong has comprehensive functions and advanced systems, which are compatible with various Android and iOS operating systems. Right now, it possesses market, transaction, search and information functions, so investors can conduct transactions via mobile phones … In early 2016, Yijintong will support mobile phone online account opening. After that, new users will be able to establish Shanghai Gold Exchange’s “Gold Account” business on their mobile phones directly, and avoid the step of visiting stores. It has brought convenience for personal investors to participate in gold and silver transactions.

Investors can log into Yijintong through mobile phones to conduct daily and nightly market transactions and search, utilizing all-day mobile phone services for gold and silver transactions, allowing Yijintong to become a mobile phone gold and silver investment edge tool that integrates functions and practicability, which also helps investors to do well in both work and financial management.

Exhibit 7.

SGE app qr code

Exhibit 8. Download methods: iOS and Android mobile phone users can scan the QR code and open it in the browser to download and install directly.

The China Gold Association (CGA) makes yearly estimates of direct purchases at the SGE. In their Gold Yearbook 2013 direct purchases (net investment) were assessed at 1,022 tonnes, computed as SGE withdrawals minus consumer demand. The CGA neglects any distortion flowing through the SGE hence I stopped using their methodology. Have a look at the screenshot below.

CGA Gold Yearbook 2013 demand
Exhibit 9. CGA Yearbook 2013: framed in black is total demand 2013 in tonnes (Red = jewelry manufacturing, blue = small bar production (sold at ie banks) , purple = industrial material, turquoise = coin manufacturing, yellow = other, green = net investment)

Unfortunately me personally can’t exactly compute true Chinese gold demand, as I don’t have business relationships with all Chinese refineries to gauge disinvestment supply flowing to the SGE. In any case, these are the formulas:

True Chinese demand = net import into the Chinese domestic market1 + scrap + disinvestment + domestic mine output

Although a tad complex, the exact formula including SGE withdrawals is:

SGE withdrawals = net import into the Chinese domestic market1 + (domestic mine output – domestic mine output that not flows to the SGE) + (scrap – scrap that not flows to the SGE) + disinvestment + distortion + (an amount equal to “domestic mine output that not flows to the SGE + scrap that not flows to the SGE” being disinvestment or distortion)

Although not all scrap as disclosed by GFMS ends up at the SGE, it’s definitely all genuine supply and therefore useful in the first formula above. Same goes for domestic mine output.

The part of scrap and domestic mine output that doesn’t travel to the SGE (although being genuine supply) must be replaced by either disinvestment or distortion at the SGE (exhibit 4). Note, in the knowledge direct purchases from the SGE are immense in China (exhibit 9) we can safely assume that disinvestment flows to the SGE are sizable as well.

My new insights unfortunately do not imply that we can make a more precise estimate of true Chinese gold demand. However, I think the best approach is to set the lower bound of true Chinese gold demand at net import1 + mine output + scrap. While I think true demand is likely higher because disinvestment to the SGE can be significant.

Sadly because disinvestment is unknown, distortion is also unknown (exhibit 4)

Let’s return to our discussion with GFMS. The big question is of course, how can total Chinese gold demand by GFMS be 867 tonnes, in a market where mining output accounted for 450 tonnes (source), net imports by my calculations accounted for 1,575 tonnes1, and there is also scrap and disinvestment supply, but export is prohibited and the premium on gold in China was positive throughout the whole year?! This cannot be.

Shanghai Gold Exchange SGE gold premium 2015
Exhibit 10.

I would like to show a real life example to illustrate what’s going on the Chinese gold market: In 2015 the Chinese stock market (the Shanghai Composite Index) declined by 40 % from June till August. Seeking for a safe haven the Chinese bought physical gold en masse directly at the SGE; some weekly withdrawals in July, August and September transcended 70 tonnes. The gold was of course sourced by imports (look at the premium in exhibit 10), yet GFMS doesn’t consider this to be demand.

SCI vs SGE withdrawals
Exhibit 11. 

The Arguments

Although true Chinese demand cannot be less than SGE withdrawals minus distortion, GFMS pretends their arguments can explain the gigantic gap between SGE withdrawals and consumer demand. Illustrated in the chart below.

Chinese domestic gold market S&D 2013 - 2015
Exhibit 12.

All arguments presented can only explain the size of distortion (exhibit 4), not the difference between SGE withdrawals and consumer demand! Actually, I should stop writing here, but I won’t. Let’s briefly go through these arguments to see if they make any sense.

The chapter in question, “A Review And Explanation Of How China’s SGE’s Withdraw Numbers Are Impacted By Other Trading Activities” (Gold Survey 2016), surprisingly lists three new arguments…

  • Tax avoidance (page 56).
  • Financial statement window dressing (page 58).
  • Retailers selling unsold inventories directly to refiners (page 58)

…and one old argument:

  • Gold leasing activities and arbitrage opportunities (in China gold is money at lower cost) (Gold Survey 2016, page 57, Gold Survey 2015, page 78)

Given the fact GFMS has gone all out in this chapter one would assume it to be complete. But strangely, arguments presented in prior Gold Survey reports and at forums have been abandoned. The following arguments were presented by GFMS in recent years:

  • Wholesale stock inventory growth (Augustus 2013) (Gold Survey 2014, page 88)
  • Arbitrage refining (Gold Survey 2014, page 88) (Reuters Global Gold Forum 2015)
  • Round tripping (Gold Survey 2014, page 88) (Gold Survey 2015, page 78, 82)
  • Chinese commercial bank assets to back investment products. “The higher levels of imports, and withdrawals, are boosted by a number of factors, but notably by gold’s use as an asset class and the requirement for commercial banks to hold physical gold to support investment products.” (Gold Survey 2015, page 78).
  • Defaulting gold enterprises sent inventory directly to refiners and SGE (Gold Survey 2015 Q2, page 7)

What happened to arbitrage refining as described by GFMS Senior Precious Metals Analyst Samson Li at the Reuters Gold Forum in 2015? Has this arbitrage opportunity ever existed or did the market change and now the opportunity is closed? I never thought this argument was very compelling. Maybe GFMS changed its mind on arbitrage refining.

What happened to the round tripping of gold between Hong Kong and Shenzhen, put forward in the Gold Survey 2014 and 2015 as a reason that inflated SGE withdrawals? Did criminals stop using this scheme, or did GFMS find out it never inflated withdrawals because gold flows through Free Trade Zones are separated from the Chinese domestic gold market and the SGE system1? In several posts I’ve extensively shown round tripping does not inflate SGE withdrawals, for more information click here.

What happened to the argument Chinese commercial banks buy and withdraw gold at the SGE to back investment products they offer to customers, a practice which boosts import and withdrawals but was not considered demand by GFMS? Or is it demand now, as GFMS dropped this argument from the list? Ok, gotcha.

Now briefly about the new arguments listed by GFMS in the GS2016:

  • Tax avoidance.

The definition of tax avoidance is that it’s a legal way to pay as little tax as possible. However, the scheme GFMS describes in the GS2016 report is tax evasion, which is highly illegal, and worst case the perpetrator can suffer life imprisonment. This is not some legal loophole as GFMS purports (page 56).

We initially became aware of the scheme in 2013 when it first emerged, but based on information gathered from our contacts, the number of industry participants mushroomed in 2014 and 2015 as other traders became aware of the potential loophole.

By writing the scheme is a way of tax avoidance and a loophole GFSM is misleading their readers. In addition, this illegal scheme did not emerge in 2013. The tax rules are now the same as when the SGE was erected in 2002. In fact, if you click here, you can read an article about the same crimes in 2009. But as mentioned before, we’ll save the details for a forthcoming post, when we’ll also address “financial statement window dressing” and “retailers selling unsold inventories directly to refiners”.

About gold leasing that would inflate SGE withdrawals, I’ve written numerous blog posts about this in the past. Best you can read my post Chinese Commodity Financing Deals Explained. In all the posts I’ve written over the years on the subject I’ve stated that the gold leased is not likely to leave the SGE vaults except when the gold will be used for jewelry manufacturing (which is genuine demand). Effectively, all the gold leasing by enterprises, investors and speculators to acquire cheap funding happens within the SGE system and do not inflate withdrawals. Ironically, in the latest World Gold Council (WGC) report it’s written [brackets added by me]:

Over recent years we have observed a rising number of commercial banks participating in the gold leasing market. … It’s estimated that around 10% of the leased gold leaves the SGE’s vaults. The majority is for financing purposes and is sold at the SGE [and stays within the SGE vaults] for cash settlement.

So, I hope to have clarified why according to my estimates true Chinese gold demand in 2015 must have been north of 2,250 tonnes (import 1,575 tonnes, mine output 450 tonnes, scrap supply 225 tonnes). More details in the next post when we will discuss the tax scheme.


1. Estimating China’s net gold import is difficult. For one, because China’s customs department doesn’t publicly disclose its cross-border trade statistics for gold so we depend on bullion export data (HS code 7108) from the rest of the world. Data from Hong Kong, the UK, Switzerland, the US, Canada and Australia is publicly available, but for example data from South Africa is not. Therefor provisional data on China’s net import is not always fully accurate. Only when the CGA publishes the import amount in their Gold Yearbook can we know for sure. My estimate is 1,575 tonnes for 2015.

Net bullion exports to China in 2015: Hong Kong 861 tonnes, Switzerland 292 tonnes, the UK 285 tonnes, the US 6 tonnes, Japan 5 tonnes, Australia 124 tonnes, Canada 3 tonnes.

In China gold is not allowed to be exported from the domestic market (SGE Main Board). However, gold is allowed to be imported into / exported from China through processing trade, usually done in Free Trade Zones. This is the only way gold can be exported from China. Note, processing trade flows are completely separated from the Chinese domestic gold market. For detailed information read my post Chinese Cross-Border Gold Trade Rules.

In order to track how much gold China is net importing, it’s necessary to net out bullion export to China by foreign countries, with import from China by foreign countries (HS code 7108). Although, it’s also possible that bullion is imported into China through processing trade and exported as jewelry (China has a vast jewelry manufacturing industry), which falls under a separated trade category (HS code 7113). Suppose, a jewelry manufacturer in Shenzhen import 2 tonnes of gold from Hong Kong under HS code 7108 through processing trade, processes the gold into jewelry to subsequently export the finished products back to Hong Kong under HS code 7113. This would blur our view on net bullion import by China, however I neglect this phenomenon in my calculations.

The fine gold content in jewelry exported from China (HS code 7113) is very difficult to measure as the total value of the products shipped also contain other precious metals, gems and includes the fabrication costs. Hence, the value and weight of jewelry exported from China does not reveal the fine gold content. The reason why I do not adjust net bullion inflows into China by jewelry outflow is because the gold content in jewelry exported from China is roughly offset by imports of gold doré or gold as a by product in ores and concentrates.

For example, the most recent CGA Yearbook in my possession, covering calendar year 2014 (exhibit 13), states “Chinese domestic and overseas gold mining output” was 512.775 tonnes. In the same report it’s mentioned “domestic mining output” accounted for 451.799 tonnes, implying overseas mine supply accounted for 60.976 tonnes. And thus, I net out overseas mining imported into China (60.976 tonnes) against jewelry exported from China. If I find more information on Chinese cross-border gold trade flows I will adjust my methodology accordingly.

Exhibit 13. Screenshot taken from my hard copy of the CGA Yearbook 2014. We can see framed in blue is “domestic and overseas mine supply”, framed in red is “gold import” and framed in green is “total supply”. Total supply, as disclosed by the CGA, equals SGE withdrawals for 2014, likewise it did for the years 2007, 2008, 2009, 2010, 2011, 2012 and 2013. 

Last but not least, gold can be imported through processing trade into the Shanghai Free Trade Zone (SFTZ) where the Shanghai International God Exchange (SGEI) vaults are located. Potentially, this gold in SGEI vaults, once sold to foreigners is withdrawn and exported abroad (inflating SGE withdrawals). However, a source at ICBC has indicated to me that regarding physical flows the SGEI is mainly used by Chinese domestic banks to import gold into the Chinese domestic market, at least this was the case until December 2015. So I don’t see a possibility there were exorbitant large volumes of gold in SGEI vaults in 2015, or have been withdrawn and exported.

The only noteworthy imports from China (the SGEI) I have observed are by India, which has taken in 370 Kg during 2015 (source Zauba), and by Thailand that presumably bought 7 tonnes (source COMTRADE).

Screen Shot 2016-08-13 at 11.20.26 am Screen Shot 2016-08-13 at 11.21.56 am Screen Shot 2016-08-13 at 11.22.20 am Screen Shot 2016-08-13 at 12.11.32 pmExhibit 14.

iAu9999 & iAu9999 OTC SGE
Exhibit 15. Weekly trading volume of the most popular SGEI contract iAu99.99, traded on-exchange and in the OTC market.

For more information on the SGEI read my posts, Workings Of The Shanghai International Gold Exchange and What Happened To The Shanghai International Gold Exchange?

2. For the sake of simplicity I have categorized under “distortion” everything that is not true demand, namely: process scrap, stock inventory change, arbitrage refining (if it exists), the VAT scheme, smuggling and SGEI withdrawals.

The Mechanics Of The Chinese Domestic Gold Market

This post is part one of the Chinese Gold Market essentials series. Click here to go to an overview of all Chinese Gold Market Essentials for a comprehensive understanding of this gigantic physical gold market.

The Shanghai Gold Exchange (SGE) is by far the largest physical gold bourse in the world. China is the largest gold importer and gold mine producer globally and both of these supply channels flow through the SGE. The yearly amount of physical gold withdrawn from the SGE vaults has exceeded 2,100 tonnes since 2013. In reports hidden from public eyes the China Gold Association (CGA) states annual Chinese gold demand equals SGE withdrawalswhile Western consultancy firms like the World Gold Council and Thomson Reuters GFMS report Chinese gold demand is roughly half this tonnage. The mysterious difference can only in part can be explained by contrasting metrics. In this post we’ll examine the mechanics of the Chinese domestic gold market and why nearly all physical gold in China flows through the SGE, in order to estimate true Chinese gold demand as precisely as we can. 

Kindly note, the definition of SGE withdrawals has changed late 2014. How it was changed can be read in the subsequent posts of the Chinese Gold Market Essentials (Workings Of The Shanghai International Gold ExchangeSGE Withdrawals In Perspective)

According to my analysis the structure of the Chinese domestic gold market with the SGE at its core has been designed by the People’s Bank Of China (PBOC), (i) to provide the Chinese citizenry direct access to the gold wholesale market, (ii) to grant all gold traded in the Chinese wholesale market to be of the highest quality, (iii) to be able to monitor gold traded in the Chinese gold market, (iiii) keep track of the amount of physical gold added to Chinese (non-government) gold reserves. Sprouted from the centrally minded Chinese authorities the SGE system was conceived in 2002 to facilitate the citizenry to buy physical gold, strengthen the Chinese economy and develop the Chinese gold market in order to support China’s internationalization.

For my analysis of the Chinese domestic gold market I’ve relied on Chinese laws, annual reports drafted by the China Gold Association (CGA), SGE, PBOC and Shanghai Futures Exchange (SHFE), next to sources in China at commercial banks and individual traders. The aforementioned reports are:

  • China Gold Association (CGA) Gold Yearbook 2006, 2007, 2008, 2013 (Chinese)
  • SGE Annual Report 2007, 2008, 2009, 2010, 2011 (English and Chinese)
  • China Gold Market Report 2008, 2009, 2010, 2011 (English and Chinese)

Most of these reports have been written in conjunction by the CGA, SGE, PBOC and SHFE.

Screen Shot 2015-03-10 at 3.25.12 PM
Screenshot from the China Gold Market Report 2011.

All the English reports that were available on the SGE website (until 2014), but now have been taken offline. Nowadays most information is only published in Chinese print.


Prior to 2002 the Chinese gold market was practically non-existent. The PBOC had the monopoly in trading gold and Chinese people were only allowed to buy jewelry in designated shops. In 2002 the PBOC erected its subsidiary the SGE to allow the free market to take over the the pricing and allocation of gold. However, the Chinese domestic gold market didn’t change over night.

To a certain extent the Chinese domestic gold market functioned as was planned starting in 2007, as for the first time the amount of physical gold withdrawn from SGE designated vaults equaled Chinese (wholesale) gold demand that year. All supply and demand was matched at the SGE, without the direct interference of the PBOC in 2007. In the CGA Gold Yearbook 2008 we can read:

2007年,上海黄金交易所黄金出库量363.194 吨,即我国当年的黄金需求量,比2006年增长了48.02%,低于供给增长率8.82个百分点。

In 2007, the amount of gold withdrawn from the warehouses of the Shanghai Gold Exchange, the total gold demand of that year, was 363.194 tonnes of gold, …

From 2002 until 2007 Chinese gold demand did not equal SGE withdrawals, to which I conclude the reform of the market wasn’t fully worked out in those years. From 2007 until 2011 SGE withdrawals exactly matched total Chinese gold demand, according to the metrics used by the CGA used in the Gold Yearbooks published in Chinese. From 2012 until present SGE withdrawals by approximation match total Chinese gold demand – the details will be discussed later on.

SGE foreign exchange gold system
Shanghai Gold Exchange, China Foreign Exchange Trade System

There are a few basic rules with respect to the Chinese (domestic) gold market that determine why SGE withdrawals equal Chinese (wholesale) gold demand, these rules compound to the mechanics of this market. The first we’ll discuss are import rules.

Chinese Cross-Border Gold Trade Rules

Gold bullion import into the Chinese domestic gold market can be done by banks that enjoy a PBOC gold import license – though for every shipment anew approval must be submitted at the Chinese central bank. Bullion export from the Chinese domestic gold market is prohibited as far as I know. At this stage there are fifteen banks that enjoy a PBOC import license:

  1. Shenzhen Development Bank / Ping An Bank
  2. Industrial and Commercial Bank of China
  3. Shanghai Pudong Development Bank
  4. Agricultural Bank of China
  5. China Construction Bank
  6. Bank of Communications
  7. China Merchants Bank
  8. China Minsheng Bank
  9. Standard Chartered
  10. Bank of Shanghai
  11. Industrial Bank
  12. Bank of China
  13. Everbright
  14. HSBC
  15. ANZ

The Chinese domestic gold market with the SGE at its core is separated from Chinese Free Trade Zones (Customs Specially Supervised Areas) where different cross-border trade rules apply. For more detailed information please read my post Chinese Cross-Border Gold Trade Rules.

All bullion imported into the Chinese domestic gold market by one of the fifteen banks is required to be standard gold and sold first through the SGE before entering the Chinese market place. Standard gold in China is bullion casted by an LBMA or SGE approved refinery in bars of 50g, 100g, 1Kg, 3Kg or 12.5Kg, with a fineness of Au9999, Au9995, Au999 or Au995. Solely standard gold is allowed into SGE designated vaults to be traded through the SGE system. The Chinese gold trade rule book says [brackets added by me]

Gold to be imported … shall be registered at a spot gold exchange [SGE] approved by the State Council where the first trade shall be completed.

When standard gold is traded over the SGE or SHFE it’s exempt from Value Added Tax (VAT). When standard gold is not traded over the SGE or SHFE it is not exempt from VAT. In addition, when non-standard gold, like jewelry, powder, ore, and dore, any gold that doesn’t meet standard gold specifications, is traded in the Chinese domestic gold market it is exempt from VAT – although the value added part may be not VAT free. These VAT rules incentivize nearly all wholesale gold supply to be traded in the form of standard gold through the SGE system.

The Chinese Mint has an exception to export golden Panda coins from the Chinese domestic gold market and to my knowledge there are also a few  jewelry companies that are allowed to trade non-standard gold in and out of the Chinese domestic market, but this tonnage is insignificant.

Individuals can officially import and export 50 gram when traveling abroad. However, this rule isn’t very stringent on the import side. Many mainland tourists visit Hong Kong to buy jewelry and bring as much bracelets as they like across the border when they return home.

At the LBMA forum in Rome September 30, 2013, Vincent Chow (Chow Sang Sang Jewelry, Hong Kong) showed some interesting slides.

Screen Shot 2015-03-10 at 6.06.21 PM

In the slide above we can see that jewelry (non-standard gold) in China enjoys 17.5 % tax on added value and 5 % consumption tax. Whilst, in Hong Kong gold jewelry enjoys no VAT or consumption tax whatsoever. This is why many Chinese women travel to Hong Kong for jewelry bargains.

Chinese Gold Mining

China has the largest domestic gold mine production in the world with an output of roughly 450 tonnes in 2015. The vast majority of this output is sold first through the SGE. All Chinese gold miners are required to sell their standard gold over the SGE. From China Gold International Resources Corp. Ltd. we can read [page 15, brackets added by me]:

On October 30, 2002, the Shanghai Gold Exchange commenced operation under the supervision of the State Council. Thereafter, the PBOC ceased its gold allocation and gold purchase operations. All PRC [People’s Republic of China] gold producers are now required to sell their standard gold bullion through the Shanghai Gold Exchange, and prices of gold on the Shanghai Gold Exchange are determined by market demand and supply, which essentially converge with the price of gold in the international market.

In addition, overseas gold mining output can be imported into the Chinese domestic gold market as non standard gold, subsequently to be refined into standard gold by an SGE approved refinery and traded over the SGE.

Because the SGE has the best liquidity in China, gold mining companies are incentivized to cast their output in standard gold bars and sell it through the SGE. However, in theory miners can sell non-standard gold off-SGE. For example China National Gold Group (China Gold) is a miner that also has its own physical stores to sell bars and ornaments.

China National Gold Group store
China Gold store

Chinese Scrap And Disinvestment Supply

Scrap supply in the Chinese domestic gold market, as defined by old jewelry or industrial products sold, has a strong incentive to flow through the SGE as well because of the VAT rules regarding standard gold. Scrap is not required to be sold through the SGE, yet many refineries cast standard gold, which are the most commonly traded bars and the only bars allowed in the SGE system. Because standard gold traded off-SGE enjoys VAT and the SGE is the most liquid exchange in China most refinery bar production finds its way to the SGE. Below is a picture of a 100 gram standard gold bar:

SGE bar small hi-res
SGE standard gold bar 100 gram

1. 上海黄金交易所标准金条 SGE Standard Gold Bar.
2. 上海黄金交易所标志 SGE Logo.
3. 品牌标志 Brand Logo.
4. 金条品牌 Bar Brand (泰山 is Mount Tai, which is produced by Shandong Gold).
5. 成色 Fineness.
6. 重量 Weight.
7. 金条编号 Bar Number.

Important for understanding all metrics used to measure Chinese gold demand, there are three kinds of scrap flows that enter the SGE.

  1. The first is genuine scrap, as defined above, which is old gold products (jewelry or industrial) sold for cash and therefor true supply having an effect on the gold price. These scrap flows are included by GFMS in their scrap numbers.
  2. The second is disinvestment. Some large and smaller investors in China buy gold directly at the SGE. If the bars are sold back they can be sent directly through a refinery and then making their way to the SGE. An investor wanting to sell 1500 Kg in gold bars is not likely to walk into a jewelry store to sell its material, more likely he will approach a refinery. Disinvestment (in my nomenclature) is not measured by GFMS as scrap.
  3. The third kind is recycled gold (again, this is my nomenclature). Recycled gold can be for example process scrap, which is metal spill over from jewelry or industrial fabrication. Suppose, a jewelry manufacturer buys 1 tonne of gold at the SGE and starts fabricating jewelry. During production 800 Kg makes it into finished products while 200 Kg is scrap spill over. The spill over, called process scrap, is being sold back from the jewelry manufacturer to a refinery making its way back to the SGE. Effectively the 200 Kg have been recycled through the SGE, being both demand and supply, having no net effect on the price. Process scrap is not included in GFMS data. Next to process scarp there can be other forms of gold being recycled through the SGE, which will label as distortion.

Note, since the inception of the Shanghai International Gold Exchange (SGEI) in 2014, total SGE withdrawals as disclosed in the Chinese Market Data Reports include both withdrawals from SGE vaults in the Chinese domestic market as well as withdrawals from SGEI vaults in the Shanghai Free Trade Zone (SFTZ). The withdrawals from SGEI vaults that are not imported into the Chinese domestic market are also part of distortion.  

In the (CGA) Chinese annual gold reports I mentioned at the beginning of this post, scrap, disinvestment and distortion flows are all counted in one category “scrap supply”, in contrast to GFMS data. This partially explains the difference in Chinese gold demand.

As supply and demand are always equal, when CGA metrics measure more supply, logically they measure more demand as well. When, later on, we’ll estimate true Chinese gold demand, we should subtract distortion from SGE withdrawals

Just like the London Bullion Market Association (LBMA) the SGE respects a chain of integrity. Meaning, only SGE approved refineries can supply bars to the SGE system and once bars are withdrawn from the SGE vaults they leave the chain of integrity. To prevent fraud, hereafter, these bars are not allowed to re-enter the SGE vaults. The only way they can be sold through the SGE is if they’re recast into new bars by an SGE approved refinery. From the SGE rulebook, Detailed Rules for Physical Delivery of the Shanghai Gold Exchange we can read:

Article 23

Any gold bullion withdrawn by a member or customer shall not be loaded into any Certified Vault in the future.

The same rule is disclosed on the websites of China’s largest banks that offer customers SGE trading accounts. Read point 2 of a considerations segment from an ICBC gold product:

ICBC SGE rule gold bars

This rule is essential for comprehending the mechanics of the Chinese domestic gold market.

SGE Withdrawals Equal Chinese Wholesale Gold Demand 

If we put together the rules mentioned above we can understand the basic formula for the Chinese domestic gold market:

SGE withdrawals = Chinese Wholesale Gold Demand

As import + mine + scrap + disinvestment + distortion is total physical supply to the SGE and everything that is withdrawn is total demand, therefor:

Import + Mine + Scrap + Disinvestment + Distortion = Total Supply = SGE Withdrawals = Wholesale Demand 


True Chinese gold demand = Import + Mine + Scrap + Disinvestment

Example Chinese domestic gold market S&D
Chart 1.

As you can see in the graph above total supply and total demand are exactly equal, this is because one cannot sell gold without a buyer or buy gold without a seller. Consequently we can gauge demand by measuring supply. Please note, in the supply and demand balance shown above, and in our further investigation, two elements are left out. On the supply side I left out stock carry over in SGE vaults from previous years, as this information is not publicly available. On the demand side I left out gold bought at the SGE that was not withdrawn from the vaults, as this information is also unknown.

The formula is supported by reports from the CGA and SGE since 2007, as every year SGE withdrawals equal total (wholesale) gold demand in these documents. Let’s have a look at CGA demand data and SGE withdrawals since 2007:

2007: SGE Withdrawals 363.2 Tonnes

Screen Shot 2015-03-12 at 4.54.58 PM
SGE Annual Report 2007: framed in red are total SGE withdrawals 2007 in Kg’s (363.19365 Kg)
Screen Shot 2015-03-11 at 10.27.37 AM
CGA Gold Yearbook 2007: framed in red is total demand in Kg’s (363.19365 Kg)

2008: SGE Withdrawals 543.2 Tonnes 

SGE Annual Report 2008, total SGE withdrawals in Kg's
SGE Annual Report 2008, total SGE withdrawals in Kg’s
Screen Shot 2015-03-11 at 10.59.54 AM
CGA Gold Yearbook 2008: framed in red is total demand 2008 in tonnes

2010: SGE Withdrawals 837.2 Tonnes

SGE Annual Report 2010, total SGE withdrawals in Kg's
SGE Annual Report 2010, total SGE withdrawals in Kg’s
China Gold Market Report 2010, total demand in tonnes
China Gold Market Report 2010, total demand in tonnes

2013: SGE Withdrawals 2197 Tonnes

Screen Shot 2015-03-12 at 9.25.07 AM
SGE Monthly Report December 2013, framed in red is total 2013 SGE withdrawals in Kg
CGA Gold Yearbook 2013 demand
CGA Gold Yearbook 2013: framed in black is total demand 2013 in tonnes (Red = jewelry manufacturing, blue = small bar production, purple = industrial material, turquoise = coin manufacturing, yellow = other, green = net investment)
CGA Gold Yearbook 2013 supply
CGA Gold Yearbook 2013: framed in black is total supply 2013 in tonnes. (Purple = domestic and overseas mine output, green = recycled gold, blue = bullion import)

In the last screen shots (from the CGA Gold Yearbook 2013) we can see total supply/demand in 2013 was 2,198.84 tonnes, which is 1.88 tonnes higher than SGE withdrawals. This can be explained by jewelry import that was counted as demand, but not sold through the SGE.

Chinese Gold Demand Metrics

Chinese wholesale gold demand as disclosed in Chinese reports by the CGA is the widest measure of demand. Western consultancy firms like GFMS use different metrics, resulting in lower supply and demand figures. Though, contrasting metrics can not explain the full difference, which has aggregated to 5,030 tonnes from 2007 until 2015. In this post we’ll briefly discuss contrasting metrics, which explain the difference in part, a succeeding post is dedicated to an extensive study on these metrics and how we can measure genuine Chinese gold demand.

Shanghai Gold Exchange SGE withdrawals yearly vs GFMS demand 2007 - 2015
Chart 2: The difference in all its simplicity.

In the supply and demand tables above we could see the difference was labeled as net investment (in the CGA Gold Yearbook 2013 at 1,022.44 tonnes), which is calculated by the CGA as a residual between what is withdrawn from the SGE vaults and gold sold at retail level (jewelry shops and banks). GFMS does not include net investment on its demand balance, but merely includes the gold sold at retail level. Net investmentwhich roughly equals the difference, is caused by direct purchases from individual and institutional customers at the SGE that withdraw their metal (also see chart 3 below).

Wang Zhe Chairman & President of the wrote in 2010:

During the first four months of this year, the number of individual investors kept growing rapidly and now has exceeded two million. The Exchange has become the main channel of investment of physical precious metals, fulfilling the needs of domestic residents.

The quote above clearly states that in 2010 direct purchases at the SGE by individual clients (266 tonnes) already exceeded retail bar investment (142 tonnes).

Purchasing gold directly at the SGE is fairly simple in China. Every Chinese citizen can buy gold or trade derivatives at the SGE through a commercial bank. For 50 RMB a client can open an SGE account at his local commercial bank branch. Subsequently, he or she receives a unique 10-digit trading number that gives access to one SGE account, consisting of a Bullion Account and a Margin Account. The 10-digit trading number will stay with an individual forever, even if he or she switches banks. The process is illustrated in the picture below:

Screen Shot 2015-03-11 at 7.19.46 PM
Source: China Gold Market Report 2008

When a physical SGE gold contract is exchanged the full amount of funds is transferred from the buyer’s Margin Account to seller’s Margin Account, the gold is transferred from the seller’s Bullion Account to the buyer’s Bullion Account (settlement is T+0). Gold credited to a Bullion Account is allowed to be withdrawn from the vaults at any time.

Screen Shot 2015-03-12 at 6.01.39 PM
Source: China Gold Market Report 2011

The SGE has almost 10,000 institutional and over 8.3 million individual clients, next to 183 domestic members and 63 international members such as banks and refineries. Naturally, the metal withdrawn from the SGE vaults by individual and institutional clients is not sold at retail level (jewelry shops and banks).

Since early 2016 there is even an SGE smartphone application called “Yijintong” that allows anyone with an internet connection to open an SGE account and trade directly on the SGE wholesale platform enjoying the lowest spreads in China.

Any individual can buy bullion directly at the SGE, the only reason he or she would buy gold in a jewelry shop is because these bars are decorated and come in all sorts of shapes and sizes. Obviously, large investors would not buy retail gold but prefer relatively cheaper SGE bars.

chow tai fook gold bar 2
Chow Tai Fook gold bar.

I should mention, the CGA discloses Chinese gold demand for the English speaking world excluding net investment, to make it appear demand is far lower than it is in reality. From Reuters on Chinese gold demand 2013:

Gold consumption in China grew to 1,176.40 tonnes last year, with jewellery demand climbing 43 percent to 716.50 tonnes and bullion demand soaring 57 percent to 375.73 tonnes, the China Gold Association said on its website.

Jewelry and bar demand is exactly the same as in the screen shot from the CGA Gold Yearbook 2013 above, but net investment is excluded. Why the widest measure of the Chinese gold market is hidden from the English speaking world is left for speculation.

I shall give another example that confirms the more realistic size of Chinese gold demand. Na Liu, from CNC Asset Management Ltd, traveled to China in 2014 and spoke to The President of the SGE Transaction Department. Afterwards Na reported on Chinese gold demand in 2013 [brackets added by me]:

The President of SGE Transaction Department (The President) said: “This 2,200 tonnes of gold, after leaving our vaults, they entered thousands of Chinese households in the form of jewellery and investment purchases.”

…when we asked why the China Gold Association’s number is so low [demand disclosed without net investment], the President said: “They mainly cover the gold sales through the gold shops. This is their main source of information. And their number is quite useful in that way. However, our system [SGE withdrawals] has broader coverage.”

Needless to say, the people that run the SGE, CGA, PBOC and SHFE are all related. Depending on the occasion these people choose to disclose net investment, and thus the true size of the Chinese gold market.

Let us briefly have a look at the compositions of the difference. In the chart below we can see apparent Chinese gold supply (import + mine + GFMS scrap = center column) versus SGE withdrawals and GFMS demand. Obviously, what is being supplied in the Chinese gold market is far more than what the GFMS reports as demand.

Chinese domestic gold market S&D 2015
Chart 3.

The gap between the height of the centre columns (apparent supply) and the red columns (SGE withdrawals) is something that will be discussed in following post.    

A legitimate reason that explains part of the difference (SGE withdrawals – GFMS demand) is stock inventory change. There is always gold in transit from being withdrawn from SGE vaults by jewelry and industrial fabricators to being sold at retail level. This is wholesale stock inventory is not being counted by GFMS as demand as this is likely hedged in the futures market and thus has no net effect on the gold price. Stock inventory change can’t make up the full difference between SGE withdrawals and consumer gold demand, but is part of distortion

In 2013 GFMS wrote me net investment is solely stock inventory change. Of course this can’t be true. When confronting them with all the evidence I had collected they wrote me:


We have checked with our Data Specialist and confirmed that we use a different methodology. Total Chinese demand used by Thomson Reuters GFMS only include jewelry, physical bullion bars/coins and all industrial demand. Any stock movement change (which is essentially the item 6 net investment) will not be included as underlying demand.


So according to you net investment is “stock movement change”? This would be gold added to the stocks from jewelers, the mint, industrial companies, etc? (this is a few hundred tons each year!)


That’s correct based on the resolution provided by our data specialist.

The World Gold Council (WGC) has tried to explain the difference in two reports released in 2014 dedicated to the Chinese gold market, which both contained fallacious and self-contradictory statements. Up until now the WGC and GFMS have presented very weak arguments that should explain the difference.

For this post we’ll wrap it up. In short, there are several metrics to measure Chinese gold demand:

  • Gold sold at retail level = GFMS gold demand
  • Import + mine = net gold added to Chinese (non-government) reserves
  • Import + mine + scrap = the lowest measure of Chinese gold demand
  • Import + mine + scrap + disinvestment = true Chinese gold demand
  • SGE withdrawals = total wholesale gold demand

SGE chairman, Xu Luode, said at the LBMA conference in 2014 [brackets added by me]

Last year [2013], China imported 1,540 tonnes of gold. Such imports, together with the 430 tonnes of gold we produced ourselves, means that we have, in effect, supplied approximately 2,000 tonnes of gold last year.

The 2,000 tonnes of gold were consumed by consumers in China. Of course, we all know that the Chinese ‘dama’ [middle-aged women] accounts for a significant proportion in purchasing gold. So last year, our gold exchange’s inventory reduced by nearly 2,200 tonnes, of which 200 tonnes was recycled gold.

Typically, Xu likes to measure import + mine as Chinese gold demand as this is the amount of gold added to Chinese (non-government) reserves. I prefer to consider all metrics to have the best understanding of the Chinese domestic gold market and keep track by how much Chinese reserves are increasing.

SGE Withdrawals 59t in week 6, YTD 374t: Chinese Gold Soap Extended Another Season

The latest numbers from the Shanghai Gold Exchange (SGE) demonstrate a little over 59 tonnes have been withdrawn from the vaults in week 6 of 2015, down 0.35 % w/w. SGE withdrawals, which are often used as a proxy for Chinese wholesale gold demand, account for an amazing 374 tonnes year to date, up 17 % y/y.

Screen Shot 2015-02-17 at 8.45.16 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.

Corrected by the volume traded on the Shanghai International Gold Exchange (SGEI), withdrawals in week 6 were at least 51 tonnes (read this post for a comprehensive explanation of the relationship between SGEI trading volume and withdrawals). Year to date withdrawals corrected by SGEI volume were at least 333 tonnes.

I can’t proof it at this stage, but I think domestic withdrawals are more likely to be 374 tonnes year to date than 333 tonnes. The Chinese import huge amounts of gold which obviously is not going through the SGEI. In addition, it seems unlikely every trade on the SGEI is withdrawn from the vaults in the Shanghai Free Trade Zone to be exported. The SGEI is not yet the place to buy gold.

Shanghai Gold Exchange SGE withdrawals delivery 2015 week 6

Shanghai Gold Exchange SGE withdrawals delivery only 2014 - 2015 week 6

According to my estimates China has imported 280 tonnes year to date (until February 13).

Here’s some more perspective; SGE withdrawals year to date are definitely not less than last year when we experienced how Chinese people can storm their local malls on a mission to buy as much gold as they can (perhaps that’s why this year lifeguards were on the scene to control the shoppers).

Lunar Year Withdrawals

Yes, demand has been huge in China, but not according to the price of gold. Technically speaking it doesn’t matter how much gold is bought to qualify demand, if supply is outstripping demand the price will go in one direction only, down. Officially, by tracking the price this should tell us everything about supply and demand of gold. A declining price means supply transcends demand, until a new price equilibrium is found, and vice versa.

If the price of gold is determined by physical supply and demand of gold we are supposed to believe that since April 2013 (see chart 1) there has been far more supply than demand as the price has come down substantially. Supposedly there has been so much supply, that Swiss refiners expanded their plants and worked in three shifts 24 hours a day to refine all this gold, which they fortunately could dump in China. The decline in the price was an enormous flood of supply and feeble demand!

Perhaps you can read my cynicism between the lines, as in my opinion there is more going on than the official story; for example, all leading Western consultancy firms have underreported 2013 and 2014 Chinese gold demand in both years by roughly 1,000 tonnes. As I have written a few hundred times on these pages: why the mainstream media isn’t covering this remains a mystery.

Most of the big potatoes in the gold realm sort of deny the amount of gold China is accumulating or present inaccurate arguments to defend their understated demand numbers.

On February 17, 2015, Mr Phillip Klapwijk, an analyst with Precious Metals Insights in Hong Kong, previous Executive Chairman of Thomson Reuters GFMS, was interviewed by Jan Harvey and Michael Wagner at the Reuters Global Gold Forum:

Harvey: The period in the run-up to the Lunar New Year is considered a peak season for gold demand in China. From what you have seen, how has buying been this year, compared to last year and the year before?

Klapwijk: In terms of the YoY, my sense is that 2015 is up moderately on 2014 but well below the exceptional demand seen in the same period in 2013.

Hmm, I beg to differ as we can see in chart 3.

Harvey: What sort of consumer trends have been noticeable this year – what sort of products are people favoring, and is gold buying overall as widespread as it has been in the past?

Klapwijk: This is partly a price and timing related phenomenon but also I think speaks to – by Chinese standards – a somewhat ‘soft’ environment for retails sales of jeweler this New Year.

I wrote a lengthy piece a couple of days ago stating retail sales do not make up the majority of gold sales in China.

Harvey: Do you consider withdrawals from the Shanghai Gold Exchange to be a useful measure of demand?

Klapwijk: The withdrawals show what the big picture is for physical “demand”. They are not per se an indication of the total demand for jeweler, investment products or industry.

This is because a good part of the withdrawals represent gold that is used purely for financing and other end-uses that are not equivalent to real consumption.

Wagner: Is financing with Gold a big deal in China?

Klapwijk: Therefore relying on SGE withdrawals to measure the size of and change in true demand is highly misleading. For example, withdrawals in 2013 came to 2,197t and in 2014 to 2,102t. This both overstates the true size of demand and, of course, completely understates the drop in jewelry and especially bar demand last year.

Yes, the difference is mainly the use of gold for financing. 

Harvey: How has that side of the market developed over the last few years? Is it still a growing area of business?

Klapwijk: The use of gold for financing… a kind of “gold carry trade”….has developed tremendously in recent years. Essentially borrowers in the shadow banking milieu are taking advantage of the availability of gold at comparatively very low rates of interest compared to straight forward RMB loans.

An indication of this is that at end 2013 the SGE reported gold loans by members totaled over 1,100 tonnes. My understanding is that number will have grown quite a bit in 2014.

There it is again, the amount of gold tied up in financing deals (CCFD’s) used as an argument to explain elevated SGE withdrawals. In my latest in-depth article on the Chinese gold market I expanded on why this can’t be the explanation as these deals only show up as SGE withdrawals if used for genuine gold business; jewelers leasing gold for production.

In a report the World Gold Council (WGC) released April 2014, China’s gold market: progress and prospects, it was stated:

… No statistics are available on the outstanding amount of gold tied up in financial operations linked to shadow banking but Precious Metals Insights believes it is feasible that by the end of 2013 this could have reached a cumulative 1,000t…

Right, so we’re talking about the same gold, 1,000 tonnes has grown into 1,100 tonnes, but now the source of this number (Precious Metals Insights, Phillip Klapwijk, February 17, 2015) says this is simply the amount leased at the SGE (not disclosing what is withdrawn or not). When I emailed the WGC in 2014 to ask what kind of “financial operations” the 1,000 tonnes actually were, they replied:

Gold leasing: Banks have built up this business to support China’s burgeoning gold industry. Miners, refiners and fabricators all have a requirement to borrow gold from time to time. For example, fabricators borrow gold to transform into jewelry, sell and then repay the bank with the proceeds. It is an effective way for the fabricator to use the bank’s balance sheet to fund its business. Banks have strict policies in place for who they can lend to, and these have been tightened over recent years, but during PMIs field research it identified that, in some instances, organizations other than genuine gold business had used this method to obtain gold, which it would then sell to obtain funding [in this case the gold wouldn’t be withdrawn from the SGE vaults]. It would then hedge its position. According to PMI, this can generate a lower cost of funding than borrowing directly from the bank. Our colleagues in China think this would be a very small part of total gold leasing; the majority of it would be used to meet the demands of genuine gold businesses.

Here the WGC admits leased gold that is withdrawn from the SGE vaults is used for genuine gold businesses and only a small part of total leases is used in shadow banking! I rest my case. 

Koos Jansen vs WGC/GFMS/CPM Update

Recently Thomson Reuters GFMS released its global gold demand figures for 2014 in which they stated Chinese demand was 866 tonnes. I briefly expanded on why I didn’t agree with this number by a humongous margin in my post from January 30 – as Chinese supply was at least 1,834 tonnes. Followed by a post released January 31 on the fact that GFMS again had double counted Chinese gold volume traded on the Shanghai Gold Exchange and the Shanghai Futures Exchange. However, this quarrel does not exclude the World Gold Council and CPM Group.

In Gold Demand Trends Full Year 2014 the World Gold Council (WGC) discloses Chinese gold demand at 814 tonnes, far below the tonnage we see being supplied to China. My previous extensive post on my disagreement with WGC demand figures I wrote months ago, hence I feel a strong propensity to update the blogosphere with my latest insights regarding the Chinese market.

Recap; in China gold demand and thus import is greatly stimulated by the government; gold export is prohibited (in general trade), except for Panda coins; all imported and domestically mined gold is required to be sold first through the Shanghai Gold Exchange (SGE) before entering the Chinese marketplace; once gold is withdrawn from the vaults of the SGE it’s not allowed to re-enter these vaults before it’s re-cast into a new bars by an SGE approved refiner; tax incentives push scrap supply to be sold through the central bourse as well, if casted as new bars. Because of the structure of the Chinese gold market many analysts, including myself, use SGE withdrawals as a proxy for Chinese wholesale demand. Read these three posts (I)(II)(III) for more detailed information on the structure of this market. A simplified equation for the SGE is: 

SGE withdrawals = import + mine + scrap

When I started publishing SGE withdrawals and later connected them to Chinese wholesale demand (on September 18, 2013, December 12, 2013 and April 15, 2014) I was merely using numbers published in Chinese – from reports that are now all taken offline (CGA Gold Yearbook 2007-2009 & 2013, SGE Annual Report 2007-2011, China Gold Market Report 2007, 2008, 2010, 2011), the only two dots I connected was aggregated demand and SGE withdrawals, the rest became self-evident. In the next chart we can see total demand from 2004 to 2013 as disclosed by the China Gold Association (CGA) Gold Yearbook 2013, which is written in Chinese and only published in hard copies; the blue bars represent total demand in tonnes (LHS), the red line y/y growth in percentages (RHS).

Screen Shot 2015-02-12 at 11.45.59 PM
Exhibit 1. Chart total Chinese gold demand from the CGA Gold Yearbook 2013, scanned from page 50.

When I say Chinese wholesale demand in 2013 was nearly 2,200 tonnes I’m not making this up, these are metrics used by the CGA (SGE/PBOC).

Gold supply in China has been outstripping the WGC’s demand numbers by literally thousands of tonnes in the past years, to which the WGC is continuously seeking to find arguments that should explain the difference. In vain the Council has published two reports dedicated to the Chinese gold market that were permeated with incorrect statements. Jumping from one argument to the next hasn’t strengthened their case either. The oncoming chart gives you a rough approximation of the difference.

WGC demand vs SGE withdrawals
Exhibit 2. Chinese wholesale demand (SGE withdrawals) vs WGC demand

Kindly note the WGC has revised their Chinese demand numbers 2013 from 1,066 tonnes to 1,312 tonnes. The revision rebutted their initial data by 23 %.

Screen Shot 2015-02-12 at 1.45.02 PM
Exhibit 3.

For this post we’ll use data from 2013 as a model to examine the Chinese gold market.

Chinese Gold Supply And Demand Metrics

According to SGE chairman Xu Luode Chinese consumer demand in 2013 was 2,000 tonnes. In his speech at the LBMA forum in Singapore June 25, 2014, he noted:

Last year, China imported 1,540 tonnes of gold. Such imports, together with the 430 tonnes of gold we produced ourselves, means that we have, in effect, supplied approximately 2,000 tonnes of gold last year.

The 2,000 tonnes of gold were consumed by consumers in China. Of course, we all know that the Chinese ‘dama’ [middle-aged women] accounts for a significant proportion in purchasing gold. So last year, our gold exchange’s inventory was reduced by nearly 2,200 tonnes, of which 200 tonnes was recycled gold. 

The SGE chairman makes a distinction between SGE withdrawals (2,200 tonnes) and his measure of consumer demand (2,000 tonnes), which he calculates by subtracting recycled gold (200 tonnes) from SGE withdrawals. Xu’s consumer demand equates exactly to how much gold was added to Chinese non-government reserves; typically this is what Chinese leaders track, the amount of gold held among the population. Total demand (equals supply) minus scrap = import plus mined gold. Scrap supply doesn’t add to a country’s reserves, only import and mine supply increase reserves.

All players in this game use different metrics to produce supply and demand figures, partially this explains the difference. In this post we’ll examine these metrics and go through the arguments the authority on gold, as the WGC has crowned itself, has brought up to enlighten us in the past years.

To capture a nation’s gold demand the authority only publishes end-user demand data (consumer demand measured at retail level) in the Demand Trends Reports (page 26).

Consumer demand 

The sum of jewelry and total bar and coin purchases for a country i.e. the amount of gold acquired directly by individuals.

If you read the report you’ll notice the WGC measures end-user demand in the form of jewelry, bar and coin demand per country at retail level, next to “ETFs, OTC investment and stock flows” to compute global total investment. But, what if Chinese individuals buy physical gold at the SGE and have the gold withdrawn from the vaults to store the metal at their own discretion? Would this show up anywhere in demand numbers from the WGC? Fact is the SGE has over 5 million individual clients, according to the chairman of the SGE, next to 8,000 institutional clients!

The reason the WGC only discloses end-user demand is because that’s all that matters to the price of gold, if a jeweler has 2 tonnes of inventory this is hedged and therefor has no net effect on the price. But measuring Chinese gold demand at retail level is inaccurate as every single Chinese citizen can open an SGE account, buy gold and withdraw. Most SGE products are denominated is small lots, consumer sizes: 1 Kg, 100 grams, 50 grams, and 10 grams. There is one product to trade 12.5 Kg bars, though this contract has only been active on three days since its inception (total volume since its inception accounts for 3 tonnes, measured unilaterally). SGE customers (potentially every single Chinese citizen) can buy and withdraw bars as small as 50 grams at the SGE, granted of the highest quality.

Screen Shot 2015-02-15 at 12.23.33 PM
Exhibit 4.

This is one of the reasons there is such a huge discrepancy between retail demand (WGC) and SGE withdrawals. Not all Chinese citizens buy their bars at a jewelry shop, au contraire.

The CGA Gold Yearbook 2013 shows us how demand is composed.

Screen Shot 2014-11-12 at 3.44.10 PM
Exhibit 5. Overview Chinese gold demand 2013, from the CGA Gold Yearbook 2013.
  • Red = Jewelry manufacturing (716.5)
  • Blue = Small gold bar production (375.73)
  • Purple = Industrial material (48.74)
  • Turquoise = Gold coin manufacturing (25.03)
  • Yellow = Other (10.4)
  • Green = Net investment (1,022.44)
  • Black = Total (2,198.84)

I think Small gold bar production is the amount of non-SGE bars manufactured by jewelry companies, banks, etc.

If we add jewelry, bar, coin, industrial and other we get 1,176.4 tonnes. This is demand measured at retail level – which is close to what the WGC initially disclosed as demand. Than there is net investment of 1,022.44 tonnes. An SGE employee once told me this simply is a residual of SGE withdrawals minus what is measured at retail level, or, everything bought directly at the SGE that wasn’t sold at retail level. The huge difference between WGC demand and SGE withdrawals is mainly caused by net investment. The Council’s mission seems to be explaining why net investment is irrelevant, coming up with feeble arguments that should support this. Later on we’ll go through the argument list and what type of demand net investment can be.

Now, let’s have a look at the supply side.

Screen Shot 2015-01-30 at 5.12.24 PM
Exhibit 6. Overview Chinese gold supply 2013, from the CGA Gold Yearbook 2013.
  • Purple = Domestic and overseas mine output (445.417)
  • Green = Recycled gold (246.923)
  • Blue = Bullion import (1,506.50)
  • Black = Total (2,198.840)

From this article we know in 2013 China domestically mined 428.16 tonnes and 17.25 came from offshore activities, so actually 1,523.75 tonnes were imported. SGE withdrawals accounted for 2,196.96 tonnes in 2013, meaning 1.88 tonnes (2,198.4 – 2196.96) were imported in the form of ie jewelry that wasn’t required to be sold through the SGE.

In Understanding China’s Gold Market the WGC notes there are two types of recycled gold in China.

Recycling: There are two types of recycled gold: i) Gold-for-cash and ii) gold-for-gold. At a retail level consumers can sell gold for cash or, especially in the case of jewelry, exchange old pieces of gold for new pieces of gold. It may be that in tonnage terms, gold-for-gold recycling is similar in size to gold-for-cash recycling: we recently surveyed 1,000 consumers and found that 8% of gold owners had sold gold-for-cash, while 10% had exchanged gold-for-gold. At a wholesale level banks and jewelers may also sell or exchange gold stock with other suppliers.

I fully agree, again we see the difference in metrics; gold-for-cash is what GFMS measures (150 tonnes in 2013), gold-for-gold is measured by the CGA as a residual.

SGE withdrawals – import – mine – gold for cash supplied to the SGE = 96.923 tonnes

Hence total recycled gold in Exhibit 6 is disclosed at 246.923, this is gold-for-cash and gold-for-gold. More information on recycled gold in China according to CGA metrics we can read in the China Gold Market Report 2008 and 2009:

… At present, gold scrap in China mainly is in two major forms: repurchase of gold bars, only applicable to brand gold bars in reference to real-time gold price, and repurchase of gold jewelry through retailers.

… China’s actual gold recycling likely reached 100 tons in 2008. Physical gold withdrawals on the Shanghai Gold Exchange (SGE) topped 543.19 tons in the year, including gold imports of 81.44 tons by commercial banks, stock carry-over of 31.661 tons from 2007 and 282.007 tons of gold produced in the year. In theory, the gap of 148.082 tons was filled by recycled gold. Therefore the gold recycled in China in 2008 should have amounted to more than 100 tons in 2008.

A few years later (2014) from the WGC in Understanding China’s Gold Market:

Given the complexity of the market there could be many reasons, but the most likely explanation is that the SGE delivery figure includes the flow of recycled gold-for-gold as well as gold-for-cash. As explained previously, while recycled gold-for-gold will increase supply and demand, the net effect is market neutral. For this reason, demand and recycling estimates as reported in Gold Demand Trends and GFMS, Thomson Reuters’ Gold Surveys exclude recycled gold-for-gold. But because the structure of the Chinese gold market requires refined and recast recycled gold to be sold through the SGE [this is not mandatory, tax incentives stimulate refined gold to be sold through the SGE], it is likely the delivery figure captures this circulation of recycled gold-for-gold.

The next flow chart from the WGC is also quite helpful:

Screen Shot 2015-02-14 at 11.59.23 PM
Exhibit 7. Courtesy of the WGC. The red box is painted by me.

I hope it’s clear now what different metrics are used for both supply and demand. If we take another look at the Exhibit-2-chart, supplemented by import, mining and scrap data, it becomes more clear what the difference is.

WGC demand vs SGE withdrawals 2
Exhibit 8. The WGC mainly relies on data from GFMS for its Demand Trends Reports, though it doesn’t publish as detailed information on scrap supply as GFMS, hence I used GFMS’ scrap numbers.

Even if we take out gold-for-gold scrap supply (the green bars represent gold-for-cash scrap) total aggregated supply from 2007 to 2014 was 7,872 tonnes, while total aggregated WGC demand over this period was 5,469 tonnes. Leaving 2,403 tonnes to be explained by the Council in their argument list, which is focused on the ‘surplus’ in the Chinese gold market, but indirectly aimed at debunking the size of SGE withdrawals.

Chinese Commodity Financing Deals

Gold is used in two sorts of Chinese Commodity Financing Deals (CCFD) to raise cheap funds: round tripping and gold leasing (read this post on the details of round-tripping, this and this on gold leasing). CCFD’s are definitely on the WGC’s argument list.

As I’ve mentioned above the WGC has published two China specials to explicate where a few thousands tonnes of gold went missing somewhere on the Asian continent, next to all quarterly Gold Demand Trends.

1) China’s gold market: progress and prospects (April 2014)

2) Understanding China’s Gold Market (July 2014)

In the first report on page 56 it’s stated, “round-tripping can inflate the SGE delivery figures”. This is absolutely not true, as I have extensively written about in The Round Tripping Myth And Why It Doesn’t Hurt Chinese Gold Demand. No need to repeat myself.

Also on page 56:

The ‘surplus’ in the Chinese market mainly stems from either possible official purchases or the extensive use of gold for financial operations.

… No statistics are available on the outstanding amount of gold tied up in financial operations linked to shadow banking but Precious Metals Insights believes it is feasible that by the end of 2013 this could have reached a cumulative 1,000t…

Although the mainstream media got completely carried away with the assumption 1,000t was tied up in CCFD’s, in reality it’s not so worrisome. Because it doesn’t involve round- tripping it can only be gold leasing. Somewhat simplified, only three parties engage in leasing gold to raise cheap funds: jewelers, miners and speculators. In China all gold leasing is settled through the SGE. If miners or speculators lease gold, they would instantaneously sell it spot at the SGE – they would not withdraw – and use the proceeds for investments. Only jewelers would withdrawal leased gold from the vaults to fabricate jewelry, subsequently sell the jewelry and use the proceeds to repay the gold loan. These withdrawals would eventually end up at retail sales. Additionally, it’s not likely a jeweler purchases gold off-SGE to have it refined and bring it to the SGE vaults to settle the lease, because the jeweler can simply buy directly at the SGE to settle the lease. This means gold leasing (CCFD’s in general) does not inflate SGE recycled gold supply.

In my opinion CCFD’s cannot substantially distort Chinese wholesale demand (SGE withdrawals). Occasionally I read numbers on the size of the Chinese gold lease market, though for me it is impossible to say what share is withdrawn or processed within the SGE system. I only know the part that is withdrawn is mostly genuine gold business.

In Gold Demand Trends Full Year 2014 the Council states:

The flow of gold into China has far exceeded the amount needed to meet domestic jewelry and investment demand in recent years [please tell us gold investors how much!]. The role of the commercial banks in using this gold for financing purposes has been well documented, including in our report Understanding China’s Gold Market and this activity expanded in 2014. To some extent, this helps explain why Shanghai Gold Exchange delivery figures are significantly higher than consumer demand.

If the only document they disclose on CCFD’s is Understanding China’s Gold Market, there is nothing new for us to learn. Yes, jewelers can lease gold, have it withdrawn and produce it into jewelry while it’s pending to be sold in retail. But this gold will not return to the SGE, so eventually it will be end-user demand. The part that is pending can partially explain the difference.

Commercial banks can have all sorts of gold (leases, pledges, ETF’s, gold savings accounts, etc) on their balance sheet, this doesn’t mean this gold has left the SGE vaults. In China gold ETF’s are backed by SGE contracts.

Official Purchases

We just read on page 56 from China’s gold market: progress and prospects, the difference might be explained by PBOC purchases. However, on page 9 from Understanding China’s Gold Market:

China’s authorities have a range of options when purchasing gold. They may acquire some of the gold which flows into China; there has been no shortage of that. But there are reasons why they may prefer to buy gold on international markets: gold sold on the SGE is priced in yuan and prospective buyers – for example, the PBOC with large multi-currency reserves – may rather use US dollars than purchasing domestically-priced gold. The international market would have a lot more liquidity too.

That’s what I wrote in this post (April 2014), it seems the WGC turned 1800 and agrees the PBOC doesn’t buy gold through the SGE.

PBOC purchases remain a very tough subject to analyze. I know they buy gold, but I have weak evidence of how much. The theory of China implementing the Turkish model to build official reserves is interesting, yet one of many problems with this theory is that the PBOC has a very strong incentive to diversify $4 trillion in FX reserves from US dollars into physical gold. So, until I bump into any evidence that would suggest otherwise, the PBOC does not buy any gold through the SGE.


There are incredible amounts of gold imported into China. The Council has often noted on the argument list exact data is hard to get by because round tripping and scrap flows may distort import figures. From page 14, Gold Demand Trends Q1 2014:

Trade flows: illustrated last year when gold flowed out of western ETFs, through refineries in Switzerland and to consumers in the East, official trade data can provide insights into global gold flows. Global Trade International Services provide access to a wide range of countries’ trade data and we also monitor individual countries’ trade data, particularly from the Hong Kong Census and Statistics Department. However, looking purely at trade data can be misleading. It can include scrap, doré and concentrates, which would be captured in supply rather than demand. Nuances such as ‘round-tripping’ can affect the data too. So, while trade data plays a valued role in informing a view on global gold flows, it is an imperfect measure of gold demand.

I’ve written an extensive post on why this is not true. In the CGA Gold Yearbook 2013, it’s precisely disclosed how much gold is imported. For other years data is available as well.

Stock Inventory

Next on the argument list is stock inventory. As demonstrated above, if a jeweler has 2 tonnes of inventory this is likely hedged, which makes it for the Council’s metrics not count as demand. From Understanding China’s Gold Market:

… It is, however, indicative that as jewelers expanded, so too did their inventory levels and it is our judgment that across the industry between 75t to 125t may have been absorbed in the supply chain since 2009.

Ok, let’s go along with this, 125 tonnes of gold (at max) is held by jewelers and alike as inventory.

The Shanghai International Gold Exchange

Something the WGC hasn’t listed on the argument list, which they should, is trading on the Shanghai International Gold Exchange (SGEI). Because SGE withdrawals at this stage also capture potential withdrawals from the SGEI in the Shanghai Free Trade Zone, this can distort withdrawal numbers in the domestic market. However, trading on the SGEI has been faint since its launch, so in my opinion it doesn’t play a substantial role now. Read this post for a detailed explanation.


The difference in metrics should be clear; SGE withdrawals are the widest measure to capture demand, whilst WGC demand is said to capture retail demand. Though, I think there is a lot of demand from individuals and institutions that buy directly at the SGE that is overlooked by the WGC.

If we stick to WGC metrics (not count gold-for-gold supply) the only legitimate argument that can explain the difference is stock inventory, either purchased at the SGE or leased at the SGE by wholesalers. According to the Council this accounts for 125 tonnes. Exhibit 8 showed us the gap between supply from 2007 to 2014 (7,872 tonnes) and WGC demand over this period (5,469 tonnes) is 2,403 tonnes. So, the difference is 2,403 tonnes and the only real WGC argument we can find to fill this void is 125 tonnes of stock inventory.

In my humble opinion stock inventory can be higher than 125 tonnes, say 300 tonnes, but this would still not fill the gap. Therefor the WGC massively understates Chinese gold demand. 2014 Chinese gold demand disclosed by the WGC is 814 tonnes, though supply was approximately 1,850 tonnes (excluding gold-for-gold), SGE withdrawals accounted for 2,102 tonnes.

Screen Shot 2015-02-16 at 6.01.17 PM
Exhibit 9. Boxed in red is the total amount of gold withdrawn from SGE vaults in 2014 (in Kg’s)

Gold-for-cash scrap (GFMS’ scrap numbers) supply is not required to be sold through the SGE, theoretically it could be refined into, for example,  goldwire and sold directly to jewelers. Aggregated mine and import supply from 2007 to 2014 is 6,934 tonnes. 

For anyone who is interested in analyzing the Chinese gold market, here are some numbers to begin with:

Chinese supply and demand numbers 2007 2014 SGE WGC GFMS
Exhibit 10.

One last point I want to make; if we look below at the weekly SGE withdrawal chart since 2010 we can see spikes around every New Year / Lunar Year (and from April – September 2013 due to the crash in the gold price) when traditionally the Chinese people buy presents for each other, often gold. Would this chart look the same if SGE withdrawals also contained CCFD’s or official purchases? Or would the chart than look more seasonally independent? In my opinion, IF the PBOC would acquire gold through the SGE the chart would look more seasonally independent.

Shanghai Gold Exchange SGE withdrawals delivery 2015 week 5, dips
Exhibit 11.

CPM Group

Somewhere in November 2014, Jeffrey Christian, Managing Partner of CPM Group, wrote an email to one of my colleagues, arguing SGE withdrawals are not what they seem. The post in which my colleague published the email and responded is taken offline (dead link), but here’s the email in full:


Someone sent me a piece you wrote on gold, in China but also elsewhere.

One comment: You use SGE gold deliveries seemingly as a surrogate of gold demand. A significant amount of the gold purchased by jewelry and electronics manufacturers and used in making products is re-refined and returned to dealers as new or process scrap. I presume you know that. The scrappage rate for most manufacturing of gold products is between 50% and 70%, depending on the processes used in manufacturing various jewelry, electronics, dental implants, catalysts, and other products. This means that the ‘end demand’ of gold that goes off into products to customers is substantially less than the amount delivered, via the SGE or any other supplier anywhere in the world.

For example, let us assume a manufacturer – it could be a jeweler or an electronics component maker – uses sputtering targets, as most of them do for the past many, many years. The new scrappage rate in the manufacturing process is 70%. In this example, let’s say that the manufacturer is using 10,000 ounces per month, buying 10,000 ounces per month through the SGE, having it made into sputtering targets, using the targets (or, in reality, selling them to clients who use them), and then collecting the process scrap, having it re-refined, and returning it to a dealer. In reality, there are various companies involved in this loop, but let’s keep it simple and say, the manufacturer. So, each month the manufacturer is buying 10,000 ounces of gold via the SGE, using 3,000 ounces in products, and recycling 7,000 ounces. At the end of the year, it will have taken delivery of 120,000 ounces of gold, but will have used 36,000 ounces of gold. The remaining 84,000 ounces of gold will have recirculated through the market repeatedly, each time being counted as an SGE delivery. So, if you are trying to guess how much gold is being used in jewelry, electronics, and other manufacturing processes in China by looking at SGE deliveries, you would be over-estimating actual demand or use by 233%.

I assume anyone writing about fabrication demand levels for gold knows this, but thought I would mention it to you in case it’s news to you. I know that when I mention it in presentations to mining executives, institutional investors, Eric Sprott, the WGC, and other gold market participants or observers, they often have no idea of this, and sometimes cannot even understand the processes I am describing.

I hope this helps. See you soon, next year, I hope.


Jeff Christian

The key takeaway of Mr Christian’s letter is that the scrap rate in China is much higher than we think and it’s all recycled through the SGE (note, in the process he describes the recycled gold is not required to go through the SGE). My reply is, if we look at Exhibit 6 we can see that by the widest measure SGE scrap supply is 247 tonnes. If we subtract all scrap (247) from SGE withdrawals (2,197), we get 1,950 tonnes. Mr Christian notes this over-estimates demand by 233 %. So actual demand was 585.6 tonnes in 2013??

GFMS Reports Chinese Gold Trade Volume Incorrect By 100%

Thomson Reuters GFMS, one of the leading consultancy firms regarding precious metals supply and demand data has recently released the GFMS Gold Survey 2014 – Update 2. From the report:

Thomson Reuters’ supply and demand data are collected and collated by our team of research analysts based in Australia, China, Europe, India and the USA within an extensive field research programme which involves interviewing stakeholders across the supply chain in every market and utilizing the unique data sets available to us after researching the market continuously since 1967.

… [etc]

All this information, including mine cost profiles, analysts “view of the field”, disaggregated supply and demand data back to 2000, as well as base case and two alternative scenarios underpin price forecast for one, three, and ten year periods and are now available on Thomson Reuters Eikon.

For the ones that don’t know, Thomson Reuters Eikon is a data terminal that costs you something in between $800 and $1,800 a month, depending on how many bells and whistles you prefer.

In a previous post I noted I didn’t agree with GFMS on Chinese gold demand 2014, disclosed by them at 866 tonnes while supply in China was 1,833 tonnes (import 1,200 + mine 451 + scrap 182), resulting in a gap of 967 tonnes. But I would like to save the demand discussion for another post to expand upon.

The section in the GFMS report that shows gold trading volume on the largest exchanges of the planet looks like this:

Screen Shot 2015-01-31 at 9.51.43 PM

The table is obviously meant so readers can compare the gold volumes traded on the major exchanges; all data is computed into metric tonnes. The COMEX data is correct, the CME publishes gold futures volume as number of contracts, when I multiply all contracts traded in 2014 by 100 ounce, which is the size of one contract, and divide the total amount of ounces by 32151, the total tonnage is 126,007. (Just about the same as GFMS reports.)

Then, the Shanghai Futures Exchange (SHFE), the primary gold futures exchange in China. One gold contract/lot on the SHFE equals 1 Kg. The SHFE publishes gold futures volume as number of contracts, when I add all contracts traded in 2014 and divide the total amount by 1,000, the total tonnage is 47,500 tonnes. Seemingly the same as GFMS reported. However, volume on the SHFE is counted double-sided, or bilaterally. From the SHFE:

  1. The unit for trading volume, open interest and the change of open interest is lot, herein are double-side counted; trading value herein is double-side counted.

There for the total tonnage has to be divided by 2 if compared to COMEX volumes. The actual total tonnage traded on the SHFE in 2014 was 23,750 – counted unilaterally. GFMS has effectively double counted SHFE gold trading volume. This way GFMS has also disclosed all data from the SGE incorrect – Au(T+D) and the spot contract. From the SGE

The Volume [Kg] and Amount are calculated bilaterally.

The open interest and turnover on the SHFE and SGE are counted bilaterally as well. Additionally, take note I’ve written on August 27, 2014, GFMS was making exactly the same mistake on their silver numbers, in the World Silver Survey 2014:


Imagine you pay $1,800 a month for an Eikon terminal that feeds you inaccurate Chinese volume and open interest data. Previously I noticed an error in the Bloomberg terminal that discloses the Chinese price of silver including 17 % VAT, and thus feeding false data.

Of course it’s anyone’s choice to decide what data to use or how to interpreted data, this post is merely meant to share my view on Chinese precious metals trade data in a effort to help investors to get a better perspective on global markets.

The World Gold Council Clueless on Chinese Gold Demand?

February 18, 2014 the World Gold Council released the Gold Demand Trends for 2013. According to this report total 2013 Chinese consumer demand was 1,065.8 tons. In my opinion this number is highly disputable.

Chinese Gold Market Essentials

The Chinese gold market is completely structured top down. The main physical (spot and deferred) exchange in China is the Shanghai Gold Exchange (SGE), that serves as the entrance point for imported and mined gold to the Chinese marketplace. Additionally the SGE is supplied by recycled gold. The Shanghai Futures Exchange (SHFE) facilitates the trading of gold futures contracts (to compete with the pricing power of Western markets).

The reason the PBOC requires imported gold to be sold first through the SGE (mined and recycled gold are stimulated to be sold through the SGE by tax incentives) is to keep track of how much gold is added to non-government reserves (jewelry, bar hoarding, etc). The setup is quite simple; to channel import and mine supply through one exchange the PBOC can efficiently supervise the quality and quantity of the gold that enters the Chinese market place. The PBOC likes to knows how many grains of fine gold are being held among the people.

Before gold is allowed to enter the vaults of the SGE it’s assayed by the National Quality Supervision and Inspection Center for Bullion & Its Products that grants a minimum fineness of 99.95. Gold withdrawn from the SGE vaults can only re-enter as recycled gold. After being remelted, by one of the associated refineries (see the list on page 40), and an assaying process the gold can move back into the SGE vaults. Just to make sure on SGE level gold of the highest purity is traded.


The consequence/purpose of the structure of the Chinese gold market is that SGE withdrawals equal wholesale demand. Confirmed by the fact that total demand reported by the China Gold Market Reports 2007 – 2011 exactly equal SGE withdrawals for the corresponding years. For an extensive analysis read this.

The Great Disparity

In 2013 SGE withdrawals accounted for 2197 tons. The next screen dump is from the last SGE report of 2013, the second number from the right (red – 本年累计交割量) is the total amount of gold withdrawn from the SGE vaults in Kg.

SGE total withdrawals 2013

The World Gold Council (WGC) claims to report on ALL sorts of demand; consumer, investment, industrial and central bank demand. This is from their February 18, 2014 press release:

Global consumer demand for gold at unprecedented levels in 2013 China the world’s largest gold market in 2013 

Consumers around the world bought gold in record amounts in 2013, led by demand in China and India, with China becoming the world’s biggest gold market, according to the latest World Gold Council Gold Demand Trends report. …

In 2013 the gold market saw 21% growth in demand from consumers which contrasted with outflows of 881t from ETFs. The net result was that global gold demand in 2013 was 15% lower than in 2012, with a full year total of 3,756t. …

Central banks

Although down 32% on 2012 they continued to be strong buyers of gold, a trend which began in 2009. 2013 saw net purchases in all four quarters, totalling 369t, meaning 12 consecutive quarters of net inflows. …


Demand reached 405t in 2013, virtually unchanged from the figure of 407t in 2012.

We can see the WGC tracks consumer (jewelry, bar and coin), ETF (and thus investment funds), central bank and industrial demand. Logic, if one wants to thoroughly analyze the gold market all facets of supply and demand must be taken into account.

So how come there is such a big difference between Chinese demand reported by the WGC, 1066 tons, and wholesale demand, 2197 tons? Why is the WGC missing 1132 tons? One reason is because the Chinese are hiding it. Since 2008 the Chinese have great interest to hoard in the dark in order to diversify their US dollar reserves, strengthen their economy and protect it from external shocks. The China Gold Association (CGA) changed the way they measure demand and all other Chinese gold institutions ceased publishing reports on demand since 2011. The only valuable information they continue to publish are SGE withdrawals. Not often, but sometimes the facts seep through the Chinese press:

China’s explosion in demand for physical gold in 2013 left a deep impression on international investors. The Shanghai Gold Exchange withdrawals for the year up till 27 December 2013 exceeded 2180 tons. Considering the exchange’s position as a hub for domestic gold circulation, in conjunction with a system that forbids withdrawn gold from re-entering inventory, to a large extent the withdrawals number can be treated as the best benchmark for physical gold demand in the Chinese market.  Not to mention that the entire 2013 global mined gold production does not exceed 2700 tons. China’s massive demand has to a large extent remade the world’s gold circulation system. Newly mined and stocked gold is moving through trade links in London – Switzerland – Hong Kong – into China in a large scale orientation towards the East. The impact of China’s demand on international gold price will inevitably increase.

Why consumer demand as presented to the world has been understated since 2008 is because the China Gold Association is manipulating the demand category net investment to suppress other categories like jewelry and bar.  This is an overview from the China Gold Market Report 2010.

Chinese gold demand 2010

According to the China Gold Market Report 2007 net investment was 18 tons. Thomson Reuters GFMS (which is the sole data provider for the WGC), assumed this was gold added to the stocks of banks, jewelers and the mint, as it was withdrawn from the SGE vaults but wasn’t sold on retail level. In 2007 this was probably true. But when Lehman fell the world changed, the Chinese began to overstate net investment to hide true demand. Up until today GFMS states net investment is stock movement changeLet’s have a look at how much net investment was in recent years. Note, in the chart below the “Difference” is approximately net investment, calculated as SGE withdrawals minus WGC consumer demand minus industrial demand.

SGE withdrawals vs WGC

Actual net investment published by the China Gold Market Reports: 2007 – 18 tons, 2008 – 129 tons, 2009 – 147 tons, 2010 – 266 tons, 2011 – 285 tons.

From 2007 – 2013 net investment was roughly 2000 tons. This definitely can not have been stock movement change at banks, jewelers and the mint. According to my analysis this gold was bought at the SGE by investment funds, individual investors and jewelers pretending to be individual investors. In the mainland there is 17 % VAT on jewelry, plus an additional 5 % consumption tax – you do the math.

WGC gold

I contacted the WGC and got in touch with one of their experts based in China. When I asked him what net investment was we had a brief debate after which he had to admit he was clueless on where this gold was going. He told me the Chinese will never disclose this information. I asked him if he would like to collaborate with me to research net investment. I got no response. A week later I asked him if he would grant me permission to publish our email correspondence. He responded to not publish our correspondence as it was meant to be private.


In the meantime I was emailing a precious metals analyst from Thomson Reuters GFMS about the Turkish gold market. This gentleman was extremely kind and helpful and explained to me in detail how GFMS measures Turkish gold coin production. Very valuable information for me! When I asked him what his take was on Chinese net investment, I got no response.

CPM Group

Jeff Christian from CPM Group did respond after I wrote him an open letter regarding SGE withdrawals. You can read our debate in the comment section of this post. Not surprisingly we couldn’t agree. What was interesting about our conversation was that he said total Chinese net gold import in 2013 was 1411 tons.


Of course I’ve written a million emails to the CGA, in English and Chinese, no response whatsoever. I called them speaking English, no luck. My friend in the mainland has called them numerous times, they always say the gentleman who wrote the China Gold Market Reports is on holiday. The message is clear…

The mainland officially net imported 1158 tons of gold from Hong Kong in 2013. Total net import according to Jeffrey Christian was 1411 tons (according to me it was 2000 tons), let’s take his number for an example. How can China import 1411 tons and mine 428 tons (that’s 1839 tons) but only demand 1066 tons? Did they import gold without asking for it? Did someone secretly pushed it across the border and now the Chinese are stuck with it? Or is there a lot of demand the WGC doesn’t disclose?

Anyway, I think 1066 tons Chinese consumer demand as reported by the WGC is highly underestimated. To be continued…