Koos Jansen
BullionStar Blogs
Koos Jansen
Posted on 3 May 2014 by

The Round Tripping Myth And Why It Doesn’t Hurt Chinese Gold Demand

Much has been written lately about the influence of Chinese commodity financing deals (CCFD) on Chinese gold demand. An often perceived analysis is that CCFD have been inflating Chinese demand/import figures and thus balance the surplus of physical gold supply in China.

In 2013 the mainland net imported 1,158 metric tonnes through Hong Kong – the Chinese do not disclose total net gold import – and domestic mining output was 428 tonnes. Without counting scrap supply and net gold import through other ports than Hong Kong, total supply was 1,581 tonnes (according to me total supply was 2,197 tonnes, as Shanghai Gold Exchange withdrawals equal total supply and demand, more on that later). Most institutions, like the World Gold Council (WGC), the China Gold Association (CGA), and GFMS state Chinese demand was lower than 1,581 tonnes in 2013, forcing themselves to explain where the rest of the supply ended up. I disagree with their explanations that are often based on CCFD. CCFD can be either done through round tripping or gold leasing. This post is solely on round tripping.

Many reports I read on CCFD simply don’t take into account Chinese laws on gold trade and the structure of the Chinese gold market. These are a few of the pieces I stumbled upon regarding this matter, from:

Goldman Sachs
Zero Hedge
Monetary Metals
The World Gold Council
FT Alphaville

What are Chinese commodity financing deals?

Simply put Chinese commodity financing deals are trades to acquire low cost capital using a commodity as collateral. There can be many ways to do this, but for now we will focus on round tripping as I would like to demonstrate this does not influence Chinese net gold import or demand. Let’s walk through the reports linked to above and get an understanding of such deals with regard to gold in comparison to other commodities. Goldman Sachs stated a few important facts:

While commodity financing deals are very complicated, the general idea is that arbitrageurs borrow short-term FX loans from onshore banks in the form of LC (letter of credit) to import commodities and then re-export the warrants (a document issued by logistic companies which represent the ownership of the underlying asset) to bring in the low cost foreign capital (hot money) and then circulate the whole process several times per year. As a result, the total outstanding FX loans associated with these commodity financing deals is determined by:

– the volume of physical inventories that is involved

– commodity prices

– the number of circulations

Our understanding is that the commodities that are involved in the financing deals include gold, copper, iron ore, and to a lesser extent, nickel, zinc, aluminum, soybean, palm oil and rubber.

…Chinese gold financing deals are processed in a different way compared with copper financing deals, though both are aimed at facilitating low cost foreign capital inflow to China. Specifically, gold financing deals involve the physical import of gold and export of gold semi-fabricated products to bring the FX into China; as a result, China’s trade data does reflect, at least partially, the scale of China gold financing deals. In contrast, Chinese copper financing deals do not need to physically move the physical copper in and out of China, so it is not shown in trade data published by China customs. In detail, Chinese gold financing deals includes four steps:

1. Onshore gold manufacturers pay LCs to offshore subsidiaries and import gold from Hong Kong to mainland China – inflating import numbers

2. offshore subsidiaries borrow USD from offshore banks via collaterizing LCs received

3. onshore manufacturers get paid by USD from offshore subsidiaries and export the gold semi-fabricated products – inflating export numbers

4. repeat step 1-3

Making a clear distinction between gold and copper round tripping is very important. Though Goldman Sachs mentions the difference – gold needs to be physically round tripped, copper not – they desist to explain what’s causing this difference. The reason is that the trade laws for gold are different than for every other commodity (because the Chinese government has chosen gold to be a part of China’s economic backbone and likes to have firm grip on this metal). A quote from myself from a post I wrote a few weeks ago:

A very long and complicated story short: In the mainland there are two types of trade; general trade and processing trade. General trade can be considered as normal trade. If gold is imported in general trade this is required to be sold through the Shanghai Gold Exchange. Only 12 banks have general trade licenses for gold from the PBOC (though for every shipment they need anew approval).

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

It’s not likely the PBOC would approve bullion gold to be exported in general trade.

Additionally there are a few jewelry companies that have PBOC licenses, but these also have to ask for permission for every trade they conduct. The PBOC has a very firm grip on gold trade.

Processing trade is something else. In this trade form raw materials from abroad are imported, processed into products and then these products are required to be exported again. This processing is usually done (there can be exceptions) in Customs Specially Supervised Areas, or CSSA (or Free Trade Zones). Processing trade doesn’t require a permit from the PBOC, as the gold that is imported will be exported after being processed. To export gold from a CSSA to a non-CSSA (that’s the rest of the mainland) a PBOC license is required.

Pocessing trade

An example for a processing trade would be; gold from Hong Kong is exported to Shenzhen (a CSSA just across the border from Hong Kong and well known for its vast jewelry fabrication industry; 4000 manufacturers), then the gold is fabricated into jewelry and imported back into Hong Kong. This trade would show up in Hong Kong’s customs report, but it would not affect Hong Kong net export to the mainland.

Chinese gold trade laws
The rest of the world can trade in the same manner with China mainland as Hong Kong.

Copper trade in China is also done through general or processing trade, the difference being that for copper general trade there is no PBOC license required and imported copper is not required to be sold through the Shanghai Gold Exchange (SGE), or any other exchange. This explains the difference in copper and gold round tripping.

Now let’s jump to what the World Gold Council had to say in it’s latest report China gold market progress and prospects on round tripping. In my opinion they made some conflicting statements, not only regarding CCFD:

Imported gold is being used via gold loans and letters of credit to raise low cost funds for business investment and speculation.

Most of this has been built up since 2011, when gold has been increasingly used as the basis for a variety of financial operations [CCFD] in China that have required the importation of very large quantities of physical bullion. … , there is the use of gold for financial arbitrage operations that will also be based upon gold loans or LCs. In most cases the gold is quickly re-exported to Hong Kong, often as very crude jewellery or ornaments to get round tight controls on bullion exports. (This is the practise commonly referred to as ‘round-tripping’. Moreover, because nearly all gold flowing into China goes through the SGE, round-tripping can inflate the SGE delivery figures.)

In theory only jewellery manufacturers or others active in the gold trade can take out gold loans, although it appears that the rules can be circumvented either by setting up a ‘gold enterprise’ or by using the services of an existing company in the gold trade. …

Gold loans or LCs used to import gold therefore offer wealthy individuals and companies a form of cheap short-term financing either for business or speculation.

The WGC is correct when it states the CCFD are conducted by gold enterprises (and wealthy individuals). These are allowed in processing trade to import gold (in a CSSA), later to be exported. But then they claim this round tripping inflates SGE deliveries, which is false!

First of all the term they should use is SGE withdrawals, that is the amount of gold leaving the SGE vaults entering the Chinese market place, whereas deliveries merely relate to the settlement between longs and shorts after a trading day, that is the amount of physical gold changing ownership in the SGE vaults. Second, although they got the term wrong it’s quite remarkable they don’t spent any more attention to SGE withdrawals; they don’t even disclose withdrawal numbers. How can it be it’s common knowledge in China what the significance is of SGE withdrawals, yet the WGC refrains from discussing this topic? Last but not least, only the 12 banks mentioned above can import gold in general trade, which is required to be sold through the SGE (supply) and subsequently can influence withdrawal numbers (demand). These banks are not involved in round tripping, they don’t need round tripping because they have direct access to the cheapest funding in the world. Gold enterprises are not involved in SGE withdrawals, as gold enterprises can only import gold in processing trade that doesn’t flow through the SGE.

For me there was also some great information in the WGC report, but when it comes to supply and demand I found numerous errors. One other example:

Only banks with PBOC-issued import licenses can import gold. They can only import LBMA good delivery bars and these must be traded through the SGE.

This is true except for the part that they can only import LBMA good delivery bars. The specifications for LBMA good delivery bars is that the weight must be in between 10.9 and 13.4 Kg, while PBOC license holders have imported at least 1,100 metric tonnes in 1 Kg bars in 2013. The correct statement should be: Only banks with PBOC-issued import licenses can import gold. They can import gold bars made by LBMA-approved refiners that meet SGE specifications and these bars must be traded through the SGE”.

Next, GFMS:

Looking at demand by country China tops the list for the first time (based on our data series) as Indian consumption struggled under the government’s import restrictions, higher taxes, a weak rupee and subsequent high domestic premium. Indeed, China consumed 1,283 tonnes of gold in 2013, 28.2% of global physical demand and substantially higher than India at 987 tonnes and 21.7% of global demand. Chinese imports of gold, and deliveries from the Shanghai Gold Exchange (SGE) were also considerably higher than this owing to growth in pipeline stock, increased holdings by commercial banks to back paper products (a legal requirement in China) and some double counting of gold that was involved in round-tripping to Hong Kong ….

Also GFMS briefly mentions SGE deliveries (withdrawals) and acknowledges physical supply transcended demand, which they inter alia explain by “double counting of gold that was involved in round-tripping to Hong Kong”. Though, round tripping has no effect on Chinese net gold import (read the Goldman Sachs quote above) or SGE Withdrawals. Round tripping is only done by Chinese gold enterprises that import gold into a CSSA and soon after export the gold.

The problem these institutions (WGC, GFMS, etc.) have is this:

SGE withdrawals vs WGC Chinese gold demnad

There has been so much gold flowing into the mainland since 2007 of which they don’t know were it ended up (the accumulative difference 2007-2013 was 1,989 metric tonnes in total).

When I called the SGE a few months ago to ask where the difference as shown in the chart above was going, they told me these are withdrawals from individual SGE account holders. If these individuals are secretly jewelers trying to avoid VAT rules or institutional buyers I don’t know. I do know these 1,989 metric tonnes haven’t been used for stock increases at jewelers or banks. In 2012 the accumulative difference accounted for 983 tonnes. With so much gold “in stock” why would they add another 1051 tonnes in 2013?

Another argument by GFMS is that, “holdings increased by commercial banks to back paper products (a legal requirement in China)”. Ok, so that has got nothing to do with demand? That gold is just brought into a vault, laying there soon to be obsolete? No, they’re referring to Chinese buying gold on paper that oddly enough has to backed by physical gold in China – I know, it’s unbelievable. According to me all arguments GFMS presents to explain the large amount of SGE withdrawals, of which they refuse to disclose the numbers, are untenable.    

Reuters disclosed a very important quote in their article:

“I do not think there is large scale gold trade-financing deals like in copper,” said Jiang Shu, an analyst with Industrial Bank, one of the few gold-importing banks in China. Import quotas and a limited number of banks allowed to import meant gold would not be a popular choice for commodity financing deals, he added.

This gentleman can know because he works at one of the banks that is allowed to import gold in general trade and it’s very likely he knows a thing or two about the Chinese gold market.

From Goldman Sachs:

…However, we don’t know how many tons of physical gold are used in the deals since we don’t know the number of circulations, though we believe it is much higher than that for copper financing deals.

Yes, the number of circulations can be in between 1 and X. But, for sure these circulations do not influence Chinese net gold import or SGE withdrawals.

Koos Jansen
E-mail Koos Jansen on:

  • geeee

    Really well done Koos, glad someone is holding these myths from mainstreamer sources to account. Too long they have gotten away with weak work that lacks depth but yet serves their self interest or supports their books.

    • In Gold We Trust

      Thanks geeee, I guess we can shake hands.

  • In Gold We Trust

    There is a debate about this post on Twitter, which started here:


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

      The WGC quote from an article refers to “cumulative” so I think they are talking about the accumulated round tripping, which has been misunderstood as saying that 1000t was tied up in these deals.

      In your tweets you note that only 660t has been imported into HK from mainland from 2011-2013. I don’t think it is unreasonable that another 340t may also have been exported as jewellery or semi-finished gold products to other countries.
      It is worth considering that the purpose of these deals is to access significantly cheaper financing, not to produce legitimate jewellery or other products. If you can sell the “value-added” exports for a small profit that is a bonus, but not essential.
      It is entirely possible that the exported products are simply melted and refined back into 1kg bars to be imported back into China for “processing” again. This is why it is often called round tripping.
      This sort of gaming has occured in the past, particularly in developing countries that insitute policies of giving export credits to encourage local jobs. People then game this by importing 400oz bars, doing as little fabrication as possible to get the export credit, and then melting the “product” into a 400oz bar and importing it again.
      While China is not offerring an export credit, the difference in lending rates creates the same sort of arbitrage incentives.
      The issue with these sorts of round tripping is the amount of gold tied up in the circular flow of gold into and out of the country (and back in and out etc etc). That is why the Goldman article refers to “number of circulations”. Unfortunately they do not know or are able to estimate it but based on the other metals refered to in their report it could be between 3 to 5 I’d guess.
      So if we had 300t of import/export round tripping with China and this was cycling 5 times then only 60t would be tied up – and when the arbitrage is closed, it would only be 60t released back into the market, so I don’t think this is a big issue.
      Koos, you have pointed out that HK is not the only way gold gets into China but on the other side of that argument you would have to also conceed that “processed” gold does not all have to go out via HK. Thus, to the extent that WGC is correct about the 1000t then that is only 340t not accounted for in HK stats and thus around 113t per year that net imports into China may be overstated by.

      • Salacious Monk

        What Koos was trying to debunk was “Round Tripping can inflate the SGE delivery”. Page 56 of the WGC report. Imports into Mainland China = imports into the CSSA+imports into the domestic market(SGE). Round tripping is only relevant to the imports into the CSSA. It is NOT relevant to the imports into the domestic market. The SGE is about the real demand for gold in China’s domestic market.
        What are you trying to do? Mixing up the two import types to mislead readers?

      • In Gold We Trust

        Hi Bron,

        In the article I focus on the fact that the base of the WGC’s analysis of the Chinese gold market is false, which makes everything on top of that incorrect.

        Before I continue, this was the original quote from the WGC:

        “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, equal to a nominal value of nearly $40bn.86.”

        To your opinion the 1000 tonnes is accumulative and the gold tied up in these deals is far less?

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

          I find that WGC wording confusing, using “outstanding amount” and then “cumulative” implies they mean 1000t is tied up.

          Looking at the Goldman 18th March paper “Days numbered for Chinese commodity financing deals” they estimate the total notional value of PM financing deals is $60 billion (page 13, Exhibit 23, with a range of $50-80b), which is higher than WGC’s $40b but same ballpark.

          However, they say that “we don’t know how many tons of physical gold are used in the deals since we don’t know the number of circulations, though we believe it is much higher than that for copper financing deals we don’t know how many tons of physical gold are used in the deals since we don’t know the number of circulations, though we believe it is much higher than that for copper financing deals”

          Goldman’s circulation estimate for copper is 5 with a maximum of 10, so lets say by “much higher” they mean 7. So working back from their numbers $60b / 7 / $1350 / 32151 = approx 200t tied up in these deals.

          Now if WGC do actually mean that 1000t is tied up then taking off my rough 200t Goldman sourced round tripping figure we have 800t left. Now we know that gold is borrowed and used as collateral http://www.ingoldwetrust.ch/chinese-gold-leasing-hidden-danger within China and certainly some gold would be tied up in these deals (FYI this would show up as stock increase, that is difference between wholesale SGE volumes and WGC measured retail demand as you and Jeff debated here http://www.ingoldwetrust.ch/open-letter-to-cpm-group) but 800t worth?
          While I don’t have any basis on which to estimate how much gold may be used in these leasing/collateral deals, I just find 800t hard to believe, partly based on the feeling that 1000t of such deals, if mostly hedged, would have had a significant impact on the gold forward curve and that doesn’t seem to be the case (but then again it is hard to know the impact on the curve as these financing business have built up over a few years).

          Maybe just ask WGC if they mean 1000t to be accumulative or outstanding?
          By the way, I find your comment about Jiang Shu at Industrial Bank a bit naive. Gold importing banks would profit from increased volumes of gold flowing from round tripping and the leasing/collateral deals, so they are going to play down the size of this business publically so as not to attract the attention of regulators less they close the loopholes.

          • Ninja Robot

            Have you ever read Koos Jansen’s work attentively? Gold export under general trade is almost forbidden and that’s why Alasdair Macleod mentioned no one had ever seen a China-refined gold bar. If there were any round tripping under general trade then there would be some China-refined gold bars showing up outside China.
            Besides that, every shipment under general trade requires a separate PBoC permit. If gold importing banks tried to do round tripping under general trade, then that would require several PBoC permits. Do you think the PBoC would be stupid enough not to find deals of this kind suspicious?

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

            Doesn’t look like you read my comments attentively: “It is entirely possible that the exported products are simply melted and refined back into 1kg bars to be imported back into China for “processing” again.”
            That quote is saying the round tripping is done as a processing trade. Fabricating gold into products (not bars) and then melting down and importing the gold as bars has occured in other countries before when there is an arbitrage or export credit etc to be had.

          • Matt

            WGC figure is 1,000t tied up but it’s not an independently calculated figure, just a rough guess based on trade flows. Lots of different ways gold is used in financing deals – round-tripping only one, others actually see metal going into China and being hoarded. Large stockpiles of gold have been built up in China in recent years. Also note China exports huge amounts of gold jewellery and other products to Hong Kong, much of which doesn’t show up in HK data as far as I can see.

          • In Gold We Trust

            Then how do you know it’s exported?

          • Matt

            Er, because it shows up in China’s data?

          • In Gold We Trust

            Hi Bron,

            The fact that people borrow gold in the domestic market (after the SGE) also has got nothing to do with Chinese gold demand. I don’t think Chinese banks buy gold specifically for loans. So when these domestic market loans are repaid there is nothing unwound in the sense that the market is flood with gold that was “tied up” in these gold loans. If a central bank gold lease is repaid, is the market then flood with unwound gold? But maybe Salacious Monk knows more about this..

            I don’t find it naive to listen to an analyst of Industrial Bank. I value the words of an analyst of a bank like this higher than that of Goldman Sachs, that has repeatedly shown to consciously lie to anybody for their own benefit, or the WGC that has proven to know nothing about the structure of the Chinese gold market.

            I will sent an email to the WGC to ask for clarification on the 1000 tonnes.

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

            If you read your own article Chinese Gold Leasing Hidden Danger this behaviour is a potential problem and I believe that a bank analyst does know that “every rule has loopholes” and that “gold lease has become a fashionable spread arbitrage game among the enterprise owners” and that the bank has a conflict of interest in that they profit from the volumes this activity brings and so would downplay the activity and would try and “flexibly” get around regulations (“Some bank insiders say, gold lease is an off-sheet lease activity. When authorities are putting tight controls on on-sheet lending, this kind of off-sheet lease business is flexible”) Just as you can’t trust everything that comes out of a Western banker’s mouth, neither can you trust a Chinese banker IMO.

          • In Gold We Trust

            Reply from the WGC:

            Thanks for your email. Sorry you hadn’t received a response, I know Louise and Albert have been very busy with Gold Demand Trends over recent weeks.

            In China’s gold market: progress & prospects, Precious Metals Insight (PMI) estimate that a cumulative 1,000t could be tied up in financial transactions. Our view is this does not represent a 1,000t pool of gold which can be quickly released to the market and in all likelihood the amount of gold used in these financial transactions is lower than the upper limit. Considered in the context of China’s stellar demand over recent years, we believe this represent a small fraction – cumulative jewellery and investment demand dwarf this number and consumers have been the driving force of the recent growth in China’s gold demand.

            For reference, the financial transactions referred to include:

            Round-tripping: Gold may have been used in round- tripping over recent years, but this is part of the market that comes and goes depending on prevailing market conditions. It is difficult to measure the amount of round-tripping, but one thing for certain is that the upper limit is capped at the volume of exports from China to HK. The recent fall in the yuan will likely have dampened this trade. And as you rightly point out , this should not be conflated with Chinese demand; round-tripping merely inflates trade statistics.

            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 jewellery, 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, organisations other than genuine gold business had used this method to obtain gold, which it would then sell to obtain funding. It would then hedge it’s 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.

            Gold pledging: Many banks accept gold as collateral for loans. You might find details of ICBC’s gold pledging scheme interesting. There are very clear rules about the gold that can be pledged.

            Please do let us know if we can be of any other help.

            Best regards,

          • In Gold We Trust

            This is a fragment of the email I got back from the WGC on round tripping:

            In “China’s gold market: progress & prospects”, Precious Metals Insight (PMI) estimate that a cumulative 1,000t could be tied up in financial transactions. Our view is this does not represent a 1,000t pool of gold which can be quickly released to the market and in all likelihood the amount of gold used in these financial transactions is lower than the upper limit. Considered in the context of China’s stellar demand over recent years, we believe this represent a small fraction – cumulative jewellery and investment demand dwarf this number and consumers have been the driving force of the recent growth in China’s gold demand.

  • In Gold We Trust

    What I meant was that those gold loans in the domestic market (which are settled after the SGE in terms of gold flows, round tripping happens before the SGE) do not necessarily have a downward pressure on the gold price. If a central bank gold lease to a bullion bank is unwound does that lower the gold price?

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

      It all depends on how much of the leased gold has NOT been hedged. The CCFD stuff seems to have been, but your article indicates that at the small business and retail level it is not. How much of that there is we don’t know, but consider that as that leasing stuff was being done it would have pushed the gold price down (one of the factors in gold price weakness over past few years?) and if it has to unwound then the gold has to be bought back to repay the leases, putting a support under the price.
      One has to dig deeper to 2nd and 3rd order effects if one is to understand the drivers of the gold price.

      • In Gold We Trust

        Anyway.. THIS article your commenting on is about the fact that CCFD did not influence the amount physical gold flowing into the Chinese market place (as the WGC has stated incorrectly).

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.