Koos Jansen
BullionStar Blogs
Koos Jansen
Posted on 19 Jan 2014 by

China Not Only Imports Gold Through Hong Kong

There are some gold analysts that create presumptions about Chinese gold demand by looking at how much gold Hong Kong net exports to the mainland, as if the mainland only imports gold from Hong Kong. I think this a misconception; just because China doesn’t officially disclose their gold trade numbers doesn’t mean they only import gold through Hong Kong.

The mainland has 22,117 kilometers (13,743 miles) of border, and I can’t think of one reason why gold could not enter through ports located anywhere at the border. To give you a small example, this is an article from the China Gold Association which reports on gold ore imports from Kazachstan in 2012.

Translated by Soh Tiong Hum:

Gold Ore Imports Surge Through Port Of Alashankou, Xinjiang

Published: 2012-11-28

Between January and September 2012, 16,800 tons of gold ore worth USD 6.5 million entered Alashankou, Xinjiang port of entry, an increase of 1173 % over the same period in 2011. The sharp increase in volume is mainly due to rise in gold price and an increase in profit margin for imported inferior gold ore. China maintains its strong emphasis on the import of resource commodities and provides enterprises with encouraging policies. In the meanwhile port of entry inspection units and transport departments take convenient and speedy measures to ensure that goods clear customs quickly, in turn lowering cost of logistics. In the same year, the port commission attracted investments in a gold ore processing plant and a gold smelter so as to value-add to imported gold ore thereby increasing demand for the raw material.

Map of China

Alashankou (A on the map) in the Xinjiang province is the main port from Kazachstan to China. 16,800 tons of gold ore is approximately 4000 ounces of fine gold, not much. However, it’s an example of the Chinese buying all the gold they can get their hands on. If you read the website of the port Alashankou it states the import of metal ores is still increasing. Gold is not only coming in through Hong Kong, it can be imported from anywhere. CME started to recognize this in September 2013:

The vast majority of bullion inflows into China emanate from Hong Kong which still serves as the main conduit into the mainland and often serves as a proxy for Chinese demand (although direct imports through Shanghai are increasing).

According to my findings  the best way the calculate China’s total net gold import is by taking SGE withdrawals as a reference, read this for a full analysis.

In 2013 the exact amount of gold withdrawn from the SGE vaults was 2197 tons. This can only have been supplied by domestic mine production (430 tons), scrap (200 tons, my guess) and import. So..

Import = 2197 – 430 – 200 = 1567

China roughly (because I cant be sure on the scrap number) has imported 1567 ton of gold in 2013.

Some news outlets have been reporting on falling net imports by China in November, from 130 tons in October to 61 tons in November, thereby suggesting demand is fading.

Hong Kong - China gold trade monthly 11-2013

This suggestion ignores the fact that the mainland doesn’t exclusively import gold from Hong Kong. It’s true that most of China’s gold imports emanate from Hong Kong, but this is not an accurate reflection of Chinese demand. Let’s have a look at the next chart.

SGE vs mining + HK imports + scrap 2013

We can see big differences between the red line (SGE withdrawals = Chinese demand) and the height of the blue bars (all known supply). The gaps had to be filled by additional import (illustrated in November on the chart). Note, Honk Kong trade numbers from December haven’t been released yet.

By looking at  SGE withdrawals we can see that demand hasn’t been dropping since October, au contraire,  it has increased! From 139 tons in October, to 168 tons in November, reaching 218 tons in December. Chinese demand for physical gold was clearly visible on retail level around new year when there was a national shopping spree which I reported on here. The upward trend continues in January; 99 tons of gold were withdrawn from the SGE vaults in the first 10 days of 2014. One of my sources in the mainland notified me on scarcity in storage capacity for consumers, January 7:

HSBC, Bank of China, Dah Sing Bank, Bank of East Asia, Shanghai Commercial Bank, ANZ, Citibank and Hang Seng Bank; NONE have available Safe Deposit Boxes – all occupied and there is a waiting list.

Having said that, lets quickly go through to the other numbers from the Hong Kong Census and Statistics Department on gold trade up until November, as this is still valuable information.

For clarity, the following charts are all based on trade numbers from Hong Kong. With these numbers we know how much gold ends up in Hong Kong itself (import minus export) and how much gold Hong Kong trades with other countries (net import or export). The “China net inflow charts” are only about the amount of gold that China mainland net imports through Hong Kong.

Hong Kong net exported 1017 tons of gold to the mainland in 11 months.

Hong Kong - China gold trade 11-2013

Hong Kong itself net imported 573 tons of gold in 11 months. I hope I can write more on this in the future because there is mainland demand hidden in these numbers.

Hong Kong gold trade 11-2013

Hong Kong net imported 66 tons from Switzerland in November, down from 85 tons in October, – 22 % m/m. A second monthly decrease, this could signal that the main vein (the gold route from the UK to Shanghai) is drying up. Year to date (nov-2013) Switzerland has net exported 848 tons of gold to Hong Kong.

HK Swiss gold trade 11-2013

From January – November Hong Kong and the mainland net imported 1590 tons.

Hong Kong + China net gold inflow 11-2013

Koos Jansen
E-mail Koos Jansen on:

E-mail address:

Tell us about yourself and your existing or proposed blog:

  • 4n1k

    Surprise, surprise…

  • matt

    For what its worth I don’t think China imports much gold from other countries – if you look at their exports they don’t report much, and what they do seems to be double-counting. That coin import/export looks misstated as well.

    • In Gold We Trust

      Thanks for your comment. Do you have any evidence to back up your statements. I would love to see them. Koos

      • Matt

        Well starting with the coins, 15t was worth 290m yuan – so $50m? That’s $100/oz.

        • In Gold We Trust

          The currency is HKD. Nevertheless, the value of the gold coins does not match the weight. I did the same check for 2012, again a mismatch. This would imply the coins contain very little gold, although the customs departmanent has assured me in the past this category is about pure gold coins. I will ask the customs dep one more time about these coins. Thnaks for your sharp eye!

        • In Gold We Trust

          I got an email back from HK customs:

          The following is the detailed definition of the above commodity item for your reference:

          Quote-

          The HKHS code 7118 9000 (Hence SITC 98002) applies to coins of any metal (including precious metals) of officially prescribed weight and design, issued under government control for use as legal tender. Consignments of individual coins or of sets of coins which are legal tender in the country of issue are classified in this heading even if they are put up for general sale in presentation cases.

          Coins are made by stamping out blanks from sheet metal; these are then “struck” with the appropriate dies to produce simultaneously the designs on the two faces.

          The heading does not cover:

          – Medals even if “struck” in the same way as coins;

          – Coins mounted in brooches, tie-pins or other objects of personal adornment;

          – Broken, cut or battered coins of a kind usable only as scrap or waste metal.

          – en quote

          Conclusion: the coins are NOT pure gold. In the future I will not count them as gold import or export, and I will remove the segment from the post.

        • In Gold We Trust

          I got an email back from HK customs:

          quote-

          The following is the detailed definition of the above commodity item for your reference:

          The SITC 98002 code applies to coins of any metal (including precious metals) of officially prescribed weight and design, issued under government control for use as legal tender. Consignments of individual coins or of sets of coins which are legal tender in the country of issue are classified in this heading even if they are put up for general sale in presentation cases.

          Coins are made by stamping out blanks from sheet metal; these are then “struck” with the appropriate dies to produce simultaneously the designs on the two faces.

          The heading does not cover:

          – Medals even if “struck” in the same way as coins;

          – Coins mounted in brooches, tie-pins or other objects of personal adornment;

          – Broken, cut or battered coins of a kind usable only as scrap or waste metal.

          -end quote

          Conclusion: the coins are NOT pure gold. In the future I will not count them as gold import or export, and I will remove the coin segment from the post.

          Thanks,

          KJ

  • http://goldchat.blogspot.com/ Bron Suchecki

    Hi Koos,

    I think your estimate of 200t of scrap is understated, but then we really don’t know (I would note that worldwide the WGC’s figures since 2008 have scrap running at 57% of mine supply and 76%of jewellery production).
    Also, I would doubt that it would be evenly spread out over the year, maybe pick a percentage of mine supply and apply that consistently. Certainly there is a regular base amount of scrap as a result of jewellery & industry manufacturing, but at the Perth Mint we see that scrap flows are price related, with Asian being a bit smarter than Westerners and selling into strength.

    I would suggest calculating the “unknown” part (that is, SGE withdrawals less known mine production and known imports) and graphing that against price. However, eyeballing your SGE gold price in RMB http://www.ingoldwetrust.ch/sge-gold-withdrawals-down-silver-premiums-5 I can’t see much of a correlation.
    The big key months with unknown gaps are Jan, Apr, May, Jul & Nov. Interesting, try and graph the unknown against the SGE premium over international price. That to my eye looks like a better fit. My theory would be that as the premium increases then maybe the banks are willing to offer better buyback premiums to entice sales back so they can sell that into the demand and capture the higher premiums.

    • In Gold We Trust

      Hi Bron,

      According to GFMS Chinese scrap supply in 2011 was 125 tons, and in 2012 it was 120 tons. This scrap supply was measured during high gold prices. The Chinese didn’t buy as much gold in those years as in 2013, so presumably also didn’t sell as much in 2013 as in the previous years.

      Globally scrap supply is a large part of total annual supply, but this is not all coming from China. Escpecially not in 2013, when the Chinese bought unprecedented amounts of physical.

      We do have to bear in mind the Chinese have a different way of calculating scrap, they count the recasting of bars (that supply the SGE!) as scrap, where GFMS doesn’t. Hence my estimate of 200 tons.

      Going back to withdrawals from the SGE vaults, do you really think this is all supplied by scrap?

      Koos Jansen

      • http://goldchat.blogspot.com/ Bron Suchecki

        Koos,

        If the SGE doesn’t allow the same bar # to come back on the exchange, then the only way to trade gold and sell it back at a profit would be to recast it, and the classify such recasting as scrap, then yes the difference could all be “scrap”.

        While I can agree that there may be some unofficial importing, I don’t find it surprising at all that there would be a lot of two way trading going on within China of investment bars. Indeed I would find it highly incredible that Chinese are only buying gold and not one investor in China ever sells back for a profit.

        For 2013 you have 2197t of withdrawals which is demand (ie buying). We then have 430 mine and 1017 of imports. Based on your WGC figures, lets say real scrap from industry/jewellery is 150t. I could agree that say 10% of imports aren’t report, that’s another 100t. The unaccounted for difference is then 500t.

        As a percentage of my estimte of 5000t held by investors within China, that is only 10% which is sold/turned over each year, or 2t each trading day. That is entirely realistic in my opinion and well below gold’s worldwide turnover and turnover of currencies (see http://goldchat.blogspot.com.au/2009/10/king-of-currencies.html)

        • Salacious Monk

          The 1017 tonnes of gold is the import from Hong Kong. Who told you that China could only import gold through Hong Kong? Gold can be shipped directly to Shanghai

          • In Gold We Trust

            Google Translate from Chinese media:

        • In Gold We Trust

          Hi Bron,

          I think “missing SGE supply” is not scrap :s Most buyers of PHYSICAL gold aren’t traders, they’re hoarders. Traders stay in the futures realm.

          Of course some physical owners sell sometimes; although in 2013 there hasn’t really been an occasion of “taking profit” as prices kept falling (but buying remained insatiable = trend).

          Your comparing world PAPER turnover with 500 tons of theoretically PHYSICAL SGE scrap supply? Based on your estimate of 5000t held by investors within China?

          I guess you have your analysis and I have mine…

          • http://goldchat.blogspot.com/ Bron Suchecki

            I’m not comparing paper to physical. Perth Mint sees a lot of physical turnover/selling – we refine scrap and sold bars as well as new mine production and we have often refined bars being sold from the Asia region.
            Just in small old Perth there is a good business of investors who buy and sell physical bars with us as they prefer to use non-leaveraged and simple ways to trade. Traders do not exclusively use the futures realm.
            The sale back of a mere 2 tonnes of physical per day in China is not unrealistic. Whatever the figure is, you cannot ignore the fact that there is some selling back by Chinese physical buyers.

          • In Gold We Trust

            Of course there are sellers, but there is no Chinese law that says this is required to be sold through the SGE. Only “fresh gold” is reguired to be sold through the SGE. After imported and mined gold (“fresh gold”) is sold through the SGE (and leaves the vaults) its free to be traded amongst Chinese in any way, shape or form. Owners of SGE bars can sell it to ANYONE. Why would they sell it at a discount to a refinery/SGE?

          • http://goldchat.blogspot.com/ Bron Suchecki

            Because that is where the pricing liquidity is, that is the point of having an exchange. If there are periods of net selling by retail clients of course the banks will look to trade that metal back into the SGE as that is how they hedge their price exposure. I find it hard to believe that either a) Chinese investors are only buying and never selling or b) of those that sell, they somehow only sell directly to other retail investors via some ebay and never back to the bank where they bought the gold from.

            I will put some equiries out to see what the physical turnover situation is like within China.

          • In Gold We Trust

            The SGE is not the only physical exchange “point”. Investors can also buy from refineries and alike. In the US the main “pricing liquidity” takes place at the COMEX. But certainly not all physical is traded through the COMEX.

    • rowingboat

      Hi Bron, do the WGC/GFMS recycled figures include estimates for divestment by big money investors beyond retail coins/bars?

      I suspect they don’t…. just recycled jewelry/industrial use and coins. If so, we have no idea about one of the key price drivers, institutional demand/supply. Divestment by large investors is largely ignored in the gold community and erroneously leads to theories about central banks or GLD filling a supply gap, IMO.

      One of Koos’ referenced reports was very clear for 2011…. this Chinese report derived the total recycled component (400 tonnes from memory) by subtracting imports/mine production from SGE deliveries. So the SGE is unable to measure recycled supply directly.

      I think we all agree that investor buying/divestment is price sensitive. Eye-balling Koos’ charts, HK imports appear to have dropped significantly after the spike to $1900, while SGE deliveries didn’t drop as much… the inference being some wise investors in China sold meaning fewer imports required.

      • http://goldchat.blogspot.com/ Bron Suchecki

        The problem with any analysis, including WGC/GFMS, is that the turnover of private investor stocks is very hard to get a handle on. The recycling/scrap figures I believe come from reports from refineries as what they have refined and melted that is not newly mined gold. This figure would include jewellery and industry scrap but also bars and coins.
        The figures would not include bars sold back by investors to a dealer who then resells them back to another buyer. Such sale and resale occurs at each level of the distribution chain – small local coin dealer, larger dealers, distributors, mints/refineries. Only if there is net selling at one level (eg a level of sustained selling that result in accumulating stocks) does that level then look to offload the excess to the next level further up. A distributor may be able to net that off and not need to offload their excess to a refiner. However, such net selling would generally not be just localised to a single coin dealer but country wide – then you would see flows back up the chain appearing at a refinery.
        I would find it hard to believe that the same process does not occur in China, and thus in a net selling situation a bank will look to the SGE as the physical market to offload and hedge any such net selling.
        IMO, Koos needs to let the data speak to him rather than trying to make it fit to his thesis. Import/export flows compared to price, Chinese premiums to the “gap” etc are valid areas of research to see if there is a pattern that may explain the “gap”.

        • In Gold We Trust

          It’s not my thesis. From the CGA yearbook 2007:

          “In 2007, the amount of gold withdrawn from the warehouses of the Shanghai Gold Exchange, gold demand of that year, was 363.194 tons of gold.”

          From the China Gold Market Report 2008:

          “Physical gold withdrawals on 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.”

          It’s their design, not mine. I’m just analyzing their design.

      • LaBastard

        “If so, we have no idea about one of the key price drivers, institutional demand/supply.”

        I don’t get this meme. Physical demand has nothing to do with gold price discovery. The LBMA paper market is where and how price , such as it is these days, is discovered.
        Who gives shit about physical demand. It’s big. Sweet. Next.

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