Beyond Decheng Mining

Feature | 2 April 2015
Qingdao phantom metals

Stephen Tricks and Philip Prowse outline the issues raised by the Qingdao phantom metals scandal
and identify potential solutions

It has been around 12 months since the name Qingdao entered trade finance terminology worldwide, not as the home of a fine beer to accompany a bowl of noodles, but as the location for a warehousing scandal that has rocked the commodity finance community.

Much has been written in the meantime on the problems at Qingdao;1 several banks and trading companies have had to instruct their lawyers,2 but getting to the bottom of events in Qingdao is by no means easy.
The Chinese authorities have been reluctant to release much information, saying that the investigations continue. It may fall to the courts outside China to try to work out the details of what went wrong - indeed, at the time of writing, the High Court in London is due to deliver judgment on the dispute between Citi and Mercuria over the early repayment of repos following the discovery of the Qingdao fraud. Details are summarised in the box-out on page 50.

On 10 December, Reuters reported, "China has been investigating whether private metals trading firm Decheng Mining and its related companies used fake warehouse receipts to obtain multiple loans secured against a single cargo of metal, including metal owned by Mercuria."3 Whatever the outcome of that case, warehouse receipts appear to have played
a key role.

Using goods in store as collateral is as old as trade finance itself. The short point to take away is that this is by no means the first time that lending against warehouse receipts has come unstuck and one doubts that it will be the last. But two aspects deserve further attention. The first is the use of letters of credit (LC) to provide finance against the security of warehouse receipts. The second is the management of stocks in store.

Musical warehouse financing LCs

Even in China, where warehouse financing has become a major source of liquidity for investment in the general economy in recent years, this type of problem is not new. Indeed a case on fraudulent financing through LCs against multiple warehouse receipts issued on the same consignments reached the Chinese Supreme Court in December 2013. At the time, that decision did not penetrate western consciousness, but in the light of the Qingdao scandal, it is worth taking a closer look.

The vanishing Shi Ming

Between 2003 and 2008 Shi Ming, a plastics and steel tycoon in Ningbo, China, borrowed more than US$200m from Chinese and western banks against copper said to be worth up to US$26m. When copper prices plummeted in the global financial crisis, Shi had no option but to run, leaving the banks and other creditors to clear up the mess. More than 80 lawsuits were filed. Most of them foundered in the wake of his disappearance, but one test case involving the Shanghai branch of ANZ Bank (China) Co Ltd eventually reached the People's Supreme Court.

According to the Court's judgment, Shi controlled a string of companies both onshore (i.e. in mainland China) and offshore (in Hong Kong). Shi arranged for Ningbo Construction Bank to issue 90 days acceptance LCs at the request of his onshore company, Shengtong, in favour of his offshore company, Lianchuang. The main documents required under the LCs were warehouse receipts issued by a C. Steinweg. Lianchuang would present the documents to ANZ Shanghai, which would negotiate the credit and pass the documents to the issuing bank. In turn the issuing bank would release the documents to Haitian, an import agent, which would then pass them to Shengtong.

So far, so good. But rather than simply tender the receipts to the warehouse to collect the metal, Shengtong would sometimes ask Steinweg to issue new receipts against the surrender of the existing receipt.4 Sometimes Shengtong replaced the genuine receipts with fraudulent receipts issued for non-existent cargo, and sometimes Shengtong simply recycled the existing receipts. Whichever route was chosen, the receipts ended up back in the hands of Lianchuang, who would tender them as a compliant presentation against the next LC. And so the cycle went on and on until it came crashing down in late 2008.

Fraud unravels all

When the music stopped, ANZ Shanghai was holding several LCs on which it had made discounted payments to Lianchuang, and on which it demanded reimbursement from Ningbo Construction Bank as issuing bank. Ningbo refused to pay on the grounds that the presentations under the LCs were fraudulent and that ANZ Shanghai was implicated in the fraud. There had been some evidence in the lower courts that employees of ANZ Shanghai had advised Shi Ming on setting up the arrangements
for generating finance on goods in store
and had assisted him in registering the offshore companies.5

The Supreme Court agreed with Ningbo on both counts. The presentations were fraudulent and ANZ Shanghai was not acting bona fide when negotiating the credits. That decision in itself would be unremarkable, although a warning to other negotiating banks who might find themselves in a similar position. However, one comment from the Supreme Court has wider implications. The Court said that if a letter of credit was being used as a pure financing tool and was not being used in its traditional role of a payment method for international trade, the normal protection given to the LC under Chinese law as an instrument independent of the underlying transaction would be lost. This meant that, in the event of a scheme being fraudulent because of a lack of a genuine underlying transaction, the LC would be unenforceable, irrespective of who triggered the fraud. On this basis, a bona fide negotiating bank could find itself unable to enforce a reimbursement obligation on the part of
a Chinese issuing bank in relation to a pure financing transaction which turns out to
be fraudulent.

The Qingdao scandal shows that the problems highlighted in this earlier case have not gone away and sheds a new light on the issue with warehouse financing.

What would the solution look like?

We believe there are a number of steps companies could consider putting in place that might avoid history being repeated
yet again.


One way in which a certain degree of control can be exercised over handling commodity collateral is registration of title for goods through a reliable registry. The Dubai Multi Commodities Centre (DMCC), a recognised name for commodity trading worldwide, provides users with a dedicated online platform for registering possession and ownership of commodities stored in UAE based storage facilities.

Through the DMCC Tradeflow system,6 owners of stored goods can request rated warehouse keepers to issue "tradeflow warrants" which represent the ownership of their goods. These warrants can be used by the owners to pledge beneficial ownership or transfer title of the stored goods to lenders as collateral. Similarly, the "LME warrants" are traded on the LME and can now be electronically transferred through the 'SWORD' automated system.7 Banks would generally recognise LME warrants as superior to ordinary warehouse receipts and will take a pledge over an LME warrant as security. However, the feasibility of adopting such electronic registration systems in China might be limited. This is because the ability of a warehouse-issued document to transfer title is not automatic but arises in law through trade custom or usage. China would need either legislative change or the creation of reliable commodities trade platforms, both of which would be costly and time consuming.


Given the nature of the problem and the need for a speedy cost-effective outcome, it might be worth investigating a more robust procedure for identifying both the commodities inside the warehouse and the fact that such goods are being used as collateral. Metal commodities are already labelled to identify them - something that would be more difficult for other types of commodities such as grain. However, although such labels will identify the metals, unless they indicate whether the goods are charged and in whose favour, anyone inspecting the goods will not know whether the metal can be used as security for a loan or in other ways freely disposed.

Better engagement between warehouses and collateral managers

The labelling option would normally be useful in preventing a Qingdao-type scenario, but one of the problems which may have arisen in Qingdao is that local regulations and practice made it difficult for collateral managers to gain unfettered access to the warehouses in order to control the existence and labelling of the metal. If the system is to function, it is important that warehouses work better with collateral managers. A system for approving warehouses that meets basic requirements is already in place at the DMCC and the LME. It may be some time before a similar system can be adopted in China, but in the short term there is room for improvement in the local practice and regulations. Lenders should be at the forefront of supporting such initiatives.

Lenders will have to find a way to ensure that reliable collateral managers are appointed and that the collateral managers have sufficient rights of access to the warehouses. They will then also have to be prepared to engage more fully with such collateral managers and ensure a good dialogue throughout the financing process.


In parallel to a better engagement with collateral managers, the banks can also show more caution in their financing.

Banks can use tranching as a means to balance the risk associated to a particular portion of the loan with appropriate collateral. This way, before releasing a new tranche the bank can check the borrower's compliance with pre-imposed conditions. Tranching would not remove the risk but would enable collateralised lending to proceed at a more balanced pace, allowing perhaps for fraud and risk management to become more visible and decreasing the size of any potential loss.


Finally, the lenders should always consider collateral protection insurance (CPI) or request that the commodity traders borrowing against the goods provide evidence of adequate cover. CPI premiums would be quite high in markets such as China, especially following the Qingdao incident, but it is a price worth considering when assessing the profitability of a deal.

Why does Qingdao matter?

China is the leading importer of commodities in the world, accounting for 40% of the global copper and steel consumption. Since the financial crisis, commodities such as copper, aluminium and iron ore have increasingly been used in China as collateral in borrowing from non-banking institutions.

It is estimated that about US$109bn of FX loans in China are backed by commodities as collateral. This is equivalent to about 31% of China's total short-term FX loans and 14% of China's total FX loans. Given its significant size, anything which impacts the commodities market in China has worldwide repercussions. In the three days following the initial reporting of the Qingdao incident the price of copper on the LME dropped by 4%. Such sudden price changes have a serious impact upon the quality of security for hundreds of millions of dollars' worth of FX loans to commodity traders in China. In addition, they affect how the world's largest commodity traders finance their day to day activities. The stability of the commodities market depends on the measures which both commodity traders and banks will take to ensure that collateral is kept, managed and disposed of in an adequate manner and/or that adequate title is held in repo situations.

As with all complex problems, there is no simple solution and all parties involved in commodity financing will need to show willingness and commitment to improve the current situation. Before and until a solution is reached through legislation or government intervention, commodity traders, warehouse managers and banks can use tools currently available at their disposal. In our view, physical labelling of collateralised goods, better use of collateral managers, tranching against covenant compliance and enhanced insurance could, taken together, ensure that the risk of any potential loss is minimised, if not eradicated.

Stephen Tricks is a consultant and
Philip Prowse a partner at Clyde & Co

Box-out 1

Citi v Mercuria - legal analysis

One of the issues the English High Court may have to address in the dispute between Citi and Mercuria is the status of warehouse receipts as documents of title. Does the transfer of a warehouse receipt under a 'repo' agreement actually transfer constructive possession of the goods? Unlike the transfer of an original bill of lading, the transfer of a warehouse receipt will not normally transfer the right to take delivery of the goods from the warehouse. Only the party named on the warehouse receipt can take delivery unless the warehouse keeper has expressly confirmed to the third party that it holds the goods to that third party's order. In English law such a confirmation is known as an attornment. There are exceptions which allow the automatic transfer of the right to delivery, such as the DMCC Tradeflow system and the LME SWORD system.

In some countries local law may recognise the transfer of a warehouse receipt as transferring the right to take delivery. Furthermore, in the event of a default and multiple claims to the same goods, the local courts may accept jurisdiction over the ensuing disputes. It is therefore good practice to investigate the local law and practice of the place of the warehouse rather than rely only on the law applicable to the finance or 'repo' agreement.

Citigroup Global Markets Limited & Anr v. Mercuria Energy Trading PTELimitedAnr,
High Court of Justice, Queen's Bench Division, Commercial Court, 14-884.

Box-out 2

Other cases of warehouse fraud

Singapore Tin Industries (STI)

This case involved a claim by ABN AMRO, a category II member of the London Metal Exchange, against CWT Commodities, under a collateral management agreement (CMA). Under the agreement, CWT was obliged to issue warehouse receipts and certificates of quality for tin that STI was using as security for trade finance as provided by ABN AMRO. To store the metal, CWT leased a warehouse at STI's premises.

CWT issued certificates of quality and warehouse receipts in relation to tin ingots, concentrates, slag and also seven batches of tin dross. The bank advanced over US$22m to STI of which at least US$10m was not repaid. It transpired that STI had been acting fraudulently by, as the judge put it, "round-tripping" the tin dross inventory. The seven batches of tin dross were reportedly purchased from, and sold on to, third parties in transactions financed by the bank. In reality, no actual sales of the tin dross were made by STI and it was left in the warehouse and mixed with additional tin dross produced by STI to provide security for additional rounds of financing by the bank. As the Singapore judge commented: "The reality was that the bank was advancing money on the security of tin dross produced entirely by STI and left to accumulate in the warehouse."

Stone & Rolls

Another notable example of warehouse fraud was the losses incurred byKomerční banka, a Czech bank, through finance provided to Stone & Rolls, a Geneva-based trading company. In that case, estimated losses of around US$250m were incurred under 30 letters of credit issued in relation to large quantities of Russian and Ukrainian agricultural products sold by Stone & Rolls.

Komerční's problem was that there were no genuine sales of produce stored in Russian warehouses; the invoices and warrant lists presented were sham and Stone & Rolls was found to have participated in a dishonest scheme designed to defraud the Czech bank.

A United Nations working paper on warehousing issues has also highlighted similar problems in other cases. In one example, several Hungarian banks incurred losses where they had provided finance against warehouse receipts, which, it was eventually discovered, had been issued by private rather than public warehouses. In another case, US banks provided finance for imports of commodities into Russia, but faced losses when it was discovered that the warehouse receipts provided for the goods were fake.

Source: 'Warehouse woes - how can you protect your commodities supply chain from fraud?' by John Whittaker of Clyde & Co. See

  1. See Bernard Goyder's comment in TFR at

  2. See Bloomberg's report on 14 July 2014 at

  3. The UK court declined to make an immediate ruling when the case came to the High Court on 3 December 2014. See

  4. There is no suggestion that Steinweg was at fault.

  5. For further discussion of how fraud unravels all see


  7. See

Already registered? Login to access premium content

Give Feedback