Tag Archives: silver bars

LBMA Silver Price Benchmark – Changes, but no Wider Participation

On 21 September, ICE Benchmark Administration (IBA) announced that it will take over the administration of the daily LBMA Silver Price benchmark auction beginning Monday 2 October. This LBMA Silver Price auction is the successor to the former London Silver Fix auction. The auction takes the form of trading unallocated silver positions on an electronic platform. The resulting price from the daily auction provides a daily silver price reference rate or benchmark which is used widely throughout the global precious metals industry. It is also now a Regulated Benchmark, regulated by the UK Financial Conduct Authority.

Bizarrely, even though it has now been more than 3 years since this new LBMA Silver Price auction was launched, there are still only 7 direct participants in the auction, a fact which flies in the face of all the previous promises from the LBMA that the rejuvenated silver auction would allow dramatically wider auction participation. These 7 participants are HSBC, JPMorgan, Morgan Stanley, Bank of Nova Scotia – ScotiaMocatta, UBS  Toronto Dominion Bank, and China Construction Bank.

Even more surprisingly, from 2 October, ICE states that only 5 of these 7 bullion banks, namely HSBC, JP Morgan, the Bank of Nova Scotia, Toronto Dominion Bank, and Morgan Stanley, will continue to participate, with UBS and China Construction Bank staying on these sidelines because they do not currently have the IT systems in place to process cleared auction trades, a clearing procedure which ICE will be introducing to the auction. Two other commodity trading companies INTL FCStone and Jane Street, will however, join the auction on 2 October. INTL FCStone and Jane Street also recently joined the LBMA Gold Price auction as direct participants.

Beyond the continued exclusion of the vast majority of global silver participants from the auction, the very fact that a new administrator has had to be drafted in to run this LBMA Silver Price auction is itself noteworthy, as is the ultra-secretive way in which ICE has been selected as the new auction administrator.

CME / Thomson Reuters – Exit Stage Left

In early March this year, the London Bullion Market Association (LBMA) announced that CME Benchmark Europe Ltd and Thomson Reuters Benchmark Services Ltd were pulling out of their roles as administrator and calculation agent of the  daily auction.

This news was somewhat surprising given that the CME – Thomson Reuters duo had only taken up responsibility for the silver auction in August 2014 and were just 2.5 years, or halfway through their 5-year contract providing this service. See BullionStar article “More Bad News for the LBMA Silver Price, but an Opportunity for Overhaul” from 7 March for more details.

While there have been various theories put forward as to why CME and Thomson Reuters decided to pull out of the new London silver auction, there has never been any official explanation forthcoming from either the LBMA, the CME Group nor from Thomson Reuters. with all parties remaining tight-lipped about the motive for the departure.

Notably, over its short life span, the new silver auction has on occasion suffered from a number of embarrassing glitches that both delayed its run time and skewed its auction price calculation, for example in January 2016, and even in April 2017 after the CME – Thomson Partners had announced their decision to exit the process. See “Death Spiral for the LBMA Gold and Silver auctions?”, dated 14 April 2017, for more details.

There were also rumours that CME and Thomson Reuters were exiting oversight of the auction due to the advent of more onerous European benchmark regulations. Whatever the real reason, the lack of clarification from the LBMA – CME – Thomson Reuters is strange given that this new silver auction was supposed to usher in an era of transparency to this critical and globally used silver pricing benchmark.

Stranger still is that the process initiated by the LBMA to secure a replacement provider for the silver auction has been itself run with the utmost level of secrecy and a total lack of consultation with the global silver market.

When news of the CME – Thomson Reuters departure broke on 3 March, the LBMA was quick to confirm, via Reuters, that it would ‘shortly’ launch a new tender to find a replacement provider for the auction process, and that the alternative provider would be identified by ‘the summer’, before taking up the new position ‘in the autumn’.

Then following this 3 March statement from the LBMA, there was zero communication with the global silver market on this issue. No updates, no news of what the tender process consisted of, no updates on whether there was a short-list of applicants, no information on how many companies had applied to the tender nor their identities, and no publication of the proposed auction solutions of any of the tender applicants. In short, there appeared to be a news blackout by the LBMA, and also little interest in the issue from the London financial media.

It was only 3 months later on 8 June that Reuters revisited the issue, saying that ICE Benchmark Administration (which runs the LBMA Gold Price auction) and the LME (which runs the LBMA platinum and palladium auctions) were “vying for control of the London silver benchmark price”. Reuters also commented that “the LBMA …had no comment on the bidding process”.

Remembering that the LBMA Silver Price is a globally used and FCA regulated benchmark which determines silver prices for myriad silver industry participants and investors around the world, the secretive stance of the LBMA in 2017 is even harder to fathom. In contrast, back in 2014 when this LBMA silver auction was initially launched, there was at least an element of transparency about how the administrator selection process was conducted.

The 2014 Process – Transparent Lip-Service

In May 2014, when London Silver Market Fixing Limited, the operator of the former London Silver Fixing benchmark auction, announced that it would step down from running the silver auction, the LBMA moved quickly to launch a ‘consultation’ to ensure that it and its bullion bank members retained full control over the real estate of the London Silver Fix and the selection and introduction of a replacement silver benchmark auction.

The consultation, launched in mid May 2014 included an online survey which could be completed by any interested silver market participants, not just LBMA members. This survey allowed the global silver market to provide feedback on what an ideal replacement auction should look like, and at least on paper, appeared inclusive and collaborative with regards to worldwide silver stakeholders.

When the results of this survey were published on 5 June 2014, it revealed that 440 participants of the silver market globally had completed the survey, with 25% of the respondents (i.e. 110 participants) indicating that they would be interested in acting as a contributor, and another 33% (or 145 respondents) indicating that they were ‘maybe’ interested in acting as a contributor in the auction. The general consensus was also that the industry wanted “an increased number of direct participants” in the silver auction.

The LBMA then launched a semi-transparent “Request for Proposals” process for any solution provider companies that wished to apply to become the new administrator of the silver price auction.

Ten companies expressed interest in becoming the new auction administrator, and from this group the LBMA choose a short-list of 7 interested providers and organised a seminar in London on 20 June 2014 at which this short-list of providers presented their proposed solutions. This seminar was, however, only open to LBMA members, so even at this point, the reluctance of the LBMA to really consult with and include the broad global silver market was apparent.

This short-list consisted of the CME Group – Thomson Reuters, the LME, Bloomberg, ETF Securities, ICE Benchmark Administration, Platts, and Autilla (in conjunction with Cinnober Financial Technology). The LBMA even went as far as publishing some of the slide presentations and / or executive summaries of these 7 proposals.

There was then a second survey of seminar attendees and LBMA full members in which they voted on which of the proposals of the short-list candidates they would most like to see implemented. Following this on 11 July 2014, the LBMA announced that the joint bid by CME And Thomson Reuters had been selected to become and administrator and auction platform provider for new replacement silver auction.

There then followed a number of seminars from CME Group, Thomson Reuters and the LBMA in late July and early August 2014 in which they promised the world in terms of vastly increased direct participation and central clearing in the new silver auction, promises which unfortunately never came to pass. See BullionStar article “The LBMA Silver Price – Broken Promises on Wider Participation and Central Clearing”, dated February 2016, for full details of these broken promises.

The point of covering the above is not so much to rehash the auction selection process from 2014, but to illustrate that while it ended up being more of a lip-service to consultation with the broader worldwide silver market, at least there was an element of communication from the LBMA through each step of the process during which the LBMA successfully retained dominant over the control of this key Silver Pricing benchmark.

Communication and Transparency  – Out the Window

Fast forward to 2017, and it becomes apparent that for whatever reason, the LBMA’s experiment with communication and semi-transparency (as of 2014) was thrown out the window, with the LBMA Board reverting to its characteristic secrecy and opacity.

With ICE Benchmark Administration about to embark on administering the LBMA Silver Price, it’s pertinent to ask what actually happened between early March 2017 and the present to lead to this outcome? Well, its hard to say actually, precisely because there is very little information available.

On 8 June, Reuters revealed that it had been told that the only 2 candidates were in serious contention to run the auction were ICE and the LME. Then on 14 July, the LBMA announced that ICE was being lined up as the chosen provider. See “ICE Benchmark Administration to Become Administrator for the LBMA Silver Price”.

The news page of Issue 86 of the LBMA’s magazine The Alchemist, from mid-August 2017, provides a clue into how the selection process that chose ICE was probably run.

“The Board has also been closely involved in the recent decision to appoint ICE Benchmark Administration as the new administrator for the LBMA Silver Price.”

This Board refers to the LBMA Board, which is a new name for what was formerly known as the LBMA Management Committee. This LBMA Board is a 10 person committee and includes representatives from bullion banks and precious metals refineries. Interestingly, of the three bullion banks currently represented on the LBMA Board, two of them, namely HSBC, and JP Morgan are direct participants in the LBMA Silver Price auction.

So it appears that this secretive and opaque ‘tender’ process to appoint a successor administrator to the LBMA Silver Price auction was controlled and run by the LBMA Board, and not, as should have been the case, by a consensus approach involving all participants in the vast global silver industry.

Central Clearing – One Step Forward, Two Steps Back

When ICE secured the silver auction mandate on 14 July, it released a statement in which it referred to its administration of the LBMA Gold Price as a model that it seeks to follow when it takes over the administration of the LBMA Silver Price:

“Our centrally cleared model has already enabled broader participation and we continue to expand the gold auction. We anticipate this will support expanded participation in silver as well.”

However, there are still only 15 entities currently authorized to directly participate in the LBMA Gold Price. Nearly all of these entities are bullion banks, and four of these banks are still suspended from the daily gold auctions because they have not implemented internal system changes to allow the processing of cleared auction trades. The excluded banks for the gold auction are UBS, Standard Chartered, China Construction Bank, and Société Générale.

As per Reuters article from 24 May 2017 “London’s gold benchmark hit by volatility after banks exit”:

“trading volumes [in the gold auction] fell sharply after April 10, when four of the 14 participating banks and brokers stopped taking part after the auction’s administrator, Intercontinental Exchange (ICE), introduced a requirement to clear that meant participants had to modify their own IT systems and procedures.”

In essence, the introduction of central clearing into the gold auction by ICE was intended to facilitate broader auction participation. However in reality, the changes have done the opposite and actually shrunken the list of active participants.

The same pattern is now playing out in the silver auction, with 2 of the 7 existing direct participants in the LBMA Silver Price, namely UBS and China Construction Bank, now dropping out precisely for the same reason that they don’t have the internal IT changes in place to process cleared auction trades.

There has even been a delay in ICE taking over the silver auction, because in late August, ICE said that it was planning to commence administration of the LBMA Silver Price on 25 September. See Platts article here for details. Then on 21 September, 4 days before the 25 Sept earmarked launch date, ICE pushed back the launch another week until 2 October. What caused this delay is unclear, but it may have been related to other participants not being ready in time to process these new cleared auction trades.

ICE Silver Futures – to Facilitate Central Clearing

So how exactly does ICE implement central clearing in the daily London gold and silver auctions. In summary, it implements a model that involves trading ICE Gold Daily Futures contracts and Silver Daily Futures contracts. Previously in the auctions, all of the direct auction participants had to maintain large bilateral credit lines with each other. Under ICE’s central clearing model, ICE now offers Exchange for Physical (EFP) transactions, with the EFPs exchanging into these futures contracts positions which trade on the ICE Clear US platform.

In the world of LBMA unallocated positions, these futures can be ‘physically settled’ into either gold and silver respectively, however, it is not actually physical gold or physical silver that is being settled, but more correctly unallocated gold and unallocated silver (i.e. paper gold and paper silver). ICE even states this when it says the futures are:

For gold – “A physically settled daily futures contract for gold delivered loco London in unallocated vault accounts.

For silver – “A physically settled daily futures contract for silver delivered loco London in unallocated vault accounts.

ICE launched its daily gold futures on 30 January. More recently, ICE launched its daily silver futures on 5 September. Although these silver futures have been available for trading for 3 weeks now, they have not traded at all according the the trading volume reports on the ICE market data website. This was similar to the ICE daily gold futures, which only started to see actual trades when the LBMA Gold Price auction began to allow central clearing. So expect some small volume trading of these silver futures from 2 October onwards.

An added bonus for ICE is that the gold and silver auctions kickstart its futures contracts, however at the same time it has forced some of the direct participants in the gold and silver auctions to drop out, thus reducing the already meagre numbers of direct participants in these very influential benchmarks and also reducing liquidity in the auctions.

Conclusion

Currently, only market making members and full members of the LBMA can directly participate in the LBMA Silver Price auction. This is because full or market making membership of the LBMA is a stipulation of the LBMA’s “Benchmark Participant” criteria.

In a document released in August titled “Administration of the Silver Price“, ICE states that it will seek to:

focus on increasing the number of participants and bringing the benchmark under IBA’s IOSCO-compliant governance and oversight framework.

IOSCO here refers to International Organisation of Securities Commissions. Following the regulatory investigations into the manipulation of LIBOR and other interest rate benchmarks, IOSCO established a task force to devise a best practice guidance framework for financial benchmark related activities. In July 2013, this task force published their guidance in a final report called ‘Principles for Financial Benchmarks’.

One of the IOSCO benchmark principles states that a financial benchmark should be a reliable representation of interest, in other words, that it should be representative of the market it is trying to measure using metrics such as market concentration.

According to Section 4.2 of the Thomson Reuters “Commodities Benchmark Methodologies LBMA Silver Price” document published in August 2014, there are an estimated 500-1000 active trading entities in the global silver market.

Therefore,  the current handful of LBMA bullion banks that will directly participate in the LBMA Silver Price auction from 2 October, i.e.  HSBC, JP Morgan, the Bank of Nova Scotia, Toronto Dominion Bank, and Morgan Stanley, in addition to 2 commodity trading companies INTL FCStone and Jane Street, is in no way representative of these 500-1000 active trading entities in the global silver market.

Therefore, yet again, with the LBMA acting as gatekeeper on who is allowed to be a direct participant in the LBMA Silver Price auction, ICE has its hands tied on meeting IOSCO’s requirement that the should be a reliable representation of interest, and there is zero chance that this silver auction will ever see the many 100s of silver trading entities taking part and zero chance that the auction will ever reflect the silver price discovery that these 100s of silver trading entities would bring to the table.

How many Silver Bars are in the LBMA Vaults in London?

Sometime in the coming days, the London Bullion Market Association (LBMA) plans to begin publishing gold and silver vault holding totals covering the network of commercial precious vault operators in London that fall under its remit. This follows an announcement made by the LBMA on 8 May.

There are seven commercial vault operators (custodians) in the LBMA custodian vault network namely, HSBC, JP Morgan, Brinks, Malca Amit, ICBC Standard Bank, Loomis (formerly Viamat), and G4S. Note that ICBC Standard Bank has a vault which is operated by Brinks on behalf of ICBC Standard. It is also quite possible that some of the HSBC vaults, such as the famous GLD gold vault, are located within Brinks facilities.

Adding in the Bank of England gold vaults under the Bank of England’s head office in the City of London, the LBMA vaulting network comprises eight sets of vaults. However, the Bank of England vaults do not store silver, or at least there is no evidence that the Bank of England stores silver. However, the other 7 vault operators can and do store silver, or at least most of them do. It’s unclear whether the G4S vault stores anything on behalf of anyone, but that’s a different story.

The forthcoming LBMA vault data will represent actual physical gold and silver holdings, i.e. real tangible precious metals, as opposed to the intangible and gargantuan paper gold and paper silver trading volumes generated each day in the London precious metals markets.

The LBMA will report physical holdings data on an aggregated basis for each of gold and silver, i.e. one quantity number will be reported each month for vaulted gold, and one quantity number will be reported each month for vaulted silver. The LBMA data will be on a 3-month lagged basis. For example, if the LBMA begins reporting this data in early July (which it probably will), then the first set of data will refer to the end of March period.

The uncertainty as to when the LBMA will begin to publish its vault holdings data is purely because the LBMA has not provided a specific publication commencement date. At first, the LBMA announced that the reporting would begin “in the summer”. Subsequently, it announced that it’s vault reporting would begin in July.

As to whether the LBMA vault holdings numbers published each month will include or exclude the Bank of England gold vaults holdings is also unclear. At the end of April, the Bank of England went ahead and separately began to publish vault holdings numbers for its own gold vaults, also on a 3-month lagged basis. More information on this Bank of England initiative can be read in BullionStar blog “Bank of England releases new data on its gold vault holdings

Incidentally, the Bank of England has now updated its website (updated 30 June) with the gold holdings figure for its vaults as of the end of March, and is reporting total physical gold holdings of 163.36 million troy ounces, which equates to 5081 tonnes of gold.

When the LBMA begins to publish its numbers, it will be clear as to whether the LBMA gold number includes the Bank of England gold holdings or not, and this will probably even be specified in a footnote of the report. Excluding the Bank of England vaults (or at least the non-loaned gold in the Bank of England vaults which is not under the title of bullion banks), the remaining lion’s share of the LBMA’s gold holdings number comprises gold held by Exchange Traded Funds (ETFs) in London.

In early May, in the article “Summer of 17: LBMA Confirms Upcoming Publication of London Gold Vault Holdings”, I calculated that these gold-backed ETFs which store their gold in the LBMA vaults in London accounted for 1510 tonnes of gold. Specifically:

The HSBC vault in London holds gold on behalf of the SPDR Gold Trust (currently 853 tonnes) and ETF Securities (about 215 tonnes). The JP Morgan gold vault in London holds gold on behalf of ETFs run by iShares (about 210 tonnes in London), Deutsche Bank (95 tonnes), and Source (100 tonnes). An ABSA ETF holds about 36 tonnes of gold with Brinks in London. In total, these ETFs represent about 1510 tonnes of gold.

The approach used to calculate the gold stored by these ETFs in the London vaults can be seen in the article “Tracking the gold held in London: An update on ETF and BoE holdings”. To this 1510 tonnes gold figure we can add gold held on behalf of customers of BullionVault and GoldMoney – which is roughly 12 tonnes of gold between them (4.75 tonnes for GoldMoney, and 7.2 tonnes of gold for BullionVault).

When the LBMA publishes its first gold total for gold held in its vault network, it will also be clear as to whether the LBMA vaults hold any significant amount of physical gold above and beyond the gold allocated within the gold-backed ETFs. There may be some gold tonnage held on an allocated basis by the LBMA bullion banks as a ‘float’, and also some gold held in allocated form by various institutional investors such as hedge funds, but my hunch is that this residual gold will be at most a respectable fraction of the amount of gold stored on behalf of ETFs in London.

However, the silver holdings in the LBMA vault network are a different kettle of fish entirely, and in addition to ETF holdings (which are reported), there could be significant silver holdings in the London vaults which have gone unreported up until now (unreported silver in the form of what consultancy GFMS calls ‘Custodian Vault’ holdings).

Reported Silver

Although gold usually generates the most headlines, it’s important not to forget about silver, and the fact that this new LBMA reporting will also provide a monthly aggregated total for the amount of physical silver held in the LBMA vaulting network in London. The silver stored in these LBMA vaults is in the form of variable weight London Good Delivery silver bars.

The recommended weight range for a Good Delivery silver bar is between 900 troy ozs and 1100 troy ozs, however, these bars will often weigh in the region of about 1000 troy ounces each. The minimum purity of a London Good Delivery silver bars is 99.9% pure silver. For example, on the BullionStar website there is a Heraeus 0.999 silver bar weighing 947.75 troy ounces. This Heraeus silver bar is an example of a Good Delivery silver bar.

Since silver has a lower value to weight ratio than gold and is bulkier to store, silver a) takes up more room and b) can be stored in secure warehouses rather than ultra-high secure vaults that are used to store gold. This is particularly true in expensive cities such as London where it is more economical to store silver in locations with lower commercial rental values.

In the LBMA vaulting network, London Good Delivery silver bars are stored 30 bars per pallet, i.e. a formation of 10 bars stacked 3 bars high. Since each bar weighs approximately 1000 oz, each pallet will weigh about 30,000 ozs, i.e. each pallet would weigh about 1 tonne.

Silver bars stored on pallets
‘1000 oz’ Silver Bars – 30 bars per pallet

At this stage, can we arrive at an estimate of the minimum amount of silver currently held in the LBMA London vaulting network? The answer is yes, for the simple reason that, in a similar manner to gold-backed ETFs, a substantial number of silver-backed ETFs also hold their silver in the vaults of London-based precious metals vaulting custodians, and these ETFs publicly report their silver bar holdings.

In addition, BullionVault and GoldMoney (which are not ETFs), both hold silver with one of the custodians in the LBMA vaulting network – Loomis. But I have included the BullionVault and GoldMoney silver totals below purely because even though they are non-ETF custodian vault holdings, both companies’ silver holdings are publicly reported on their websites.

However, there is probably also a lot more additional silver held in the London vaults above and beyond the silver bars allocated to ETFs and the known silver stored by GoldMoney and BullionVault. Some of this additional silver falls under what Thomson Reuters GFMS classify as ‘Custodian Vault‘ silver, which is silver that is basically in an ‘Unreported’ category but which Thomson Reuters GFMS seems to think it knows about through its own ‘proprietary surveys’ and ‘field research’. This ‘Custodian Vault’ silver probably accounts for a substantial amount of silver in the London vaults. However, it is difficult to know because GFMS does not provide granularity on its numbers beyond an overall ‘Europe’ number. But I have made some assumptions about this ‘Custodian Vault’ silver in London, which is discussed in a final section below.

Silver ETFs

For the silver-backed ETFs, the first step is to identify which silver ETFs hold silver bars in the LBMA vaults in London. Using the list of silver ETF providers specified on Nick Laird’s GoldChartsRUs website (subscription only), the platform providers and their ETFs which hold silver in the LBMA vaults in London are as follows:

  • iShares: 1 ETF
  • ETF Securities: 6 ETFs
  • SOURCE : 1 ETF
  • Deutsche Bank: 3 ETFs

Between them, these four providers offer 11 ETFs that hold some or all of their silver in LBMA London vaults. This silver is held with custodians JP Morgan and HSBC, and with sub-custodians, Brinks and Malca Amit. Note, that GoldMoney and BullionVault store silver in London with Loomis as custodian.

As publicly traded vehicles, most of these ETFs publish daily silver bar weight lists or holdings files and they also undergo twice yearly physical audits by independent auditors. These weight lists and audits documents are helpful in pinpointing who the custodians and sub-custodians are, which locations these silver ETF’s store their silver in, and how much silver (in silver bar form) is stored in each location.

iShares Silver Trust (SLV)

The iShares Silver Trust, ticker code SLV, is the world’s largest silver-backed ETF.  It’s probably best to think of SLV as the silver equivalent of the mammoth SPDR Gold Trust (GLD).

The custodian for SLV is JP Morgan Chase Bank (London Branch), and Brinks also acts as a sub-custodian for SLV. SLV holds silver in vaults across both London and New York. According to the SLV daily silver bar weight list, SLV’s silver bars are held in two Brinks vaults in London, one JP Morgan vault in London, and one JP Morgan vault in New York.

As of 29 June 2017, SLV reported that it was holding 348,841 Good Delivery silver bars containing a total of 339.89 million troy ounces of silver, or a colossal 10,572 tonnes of silver. The actual SLV bar list, which is uploaded to a JP Morgan website in pdf format using the same filename each day, can be seen here, but be warned that the file is about 5370 pages long, so there’s no real need to open it unless you are curious. A screenshot of the top of the first page is provided below

Silver bars held in the iShares Silver |Trust (SLV), JP Morgan London custodian Morgan
Silver bars held in the iShares Silver |Trust (SLV), JP Morgan London, SLV custodian. Source: SLV weight list, JP Morgan website – Click to Enlarge

The SLV weight list specifies that the SLV silver is held in a ‘Brinks London‘ vault, a ‘Brinks London C‘ vault, a ‘JPM London V‘ vault, and a ‘JPM New York‘ vault. Between them, 2 Brinks vaults in London hold 55% of SLV’s silver bars representing 5753 tonnes, or 54% of the silver held in SLV. Adding in the ‘JPM London V‘ vault means that 289,053 silver bars, weighing 8720 tonnes (or 82% of SLV’s entire silver holdings) are held in LBMA London vaults.

iShares Silver Trust (SLV) - Silver bars and Ounces by location
iShares Silver Trust (SLV) – Silver bars and Ounces by location – Click to Enlarge

The auditor for SLV is Inspectorate. Interestingly, the latest Inspectorate letter for SLV, for record date 10 February 2017, does not make a distinction between the 2 Brinks vaults in London and just reports that SLV’s silver is in:

“Three vaults located in and around London and New York:

– two vaults owned and operated by JP Morgan Chase Bank N.A. with 124,054 bars

one vault owned and operated by Brinks, as a sub-custodian for JP Morgan Chase Bank N.A. with 220,066 bars

 This would suggest that Inspectorate does not see the need to distinguish between the “Brinks London” vault and the “Brinks London C” vault, presumably because both Brinks vaults are in the same building in the Brinks facility (which is beside Heathrow Airport).

Even though the official custodian for SLV is JP Morgan Chase Bank N.A., London Branch (see original SLV Custodian Agreement filed April 2006 here), since it’s launch in 2006 SLV has at different times used quite a diverse group of sub-custodian vaults as well as at least 3 JP Morgan vaults. For example, over the 3 year period from early 2010 to early 2013, SLV stored silver in the following vaults:

  • Johnston Matthey, Royston
  • Brinks London
  • Brinks London A
  • Brinks London C
  • Viamat (now known as Loomis)
  •  JP Morgan London A
  • JP Morgan London V
  • JP Morgan New York

Royston is about 50 miles north of central London. The above list is taken from the following chart which is from the ScrewTape Files website.

SLV vaults used, April 2013 and prior. Source: Screwtape Files
SLV vaults: Feb 2010 – April 2013. Source: Screwtape Files – Click to Enlarge

Given that there are Brinks vaults in London named ‘Brinks London‘, ‘Brinks London A‘, and ‘Brinks London C‘, this would most likely imply that there is or was also a ‘Brinks London B‘ vault, which, for whatever reason, doesn’t show up in any ETF custodian documentation.

The naming convention of the JP Morgan vaults in London as ‘JPM London A‘ and ‘JPM London V‘ is also interesting. SLV silver started being taken out of the ‘JPM London A’ vault in February 2012, and this vault was depleted of 100 million ounces of SLV silver (~ 3100 tonnes) by October 2012 (blue line in above chart). At the same time, the SLV silver inventory in the ‘Brinks London’ vault ramped up by 100 million ounces of SLV silver also between February 2012 and October 2012.

JPM London A could be JP  Morgan’s original vault in the City of London. This would then make the JPM London V vault a separate location. My pet theory (pet rock theory) is that the V in the ‘JPM London V’ could refer to Viamat International, which is now known as Loomis. JP Morgan could have outsourced storage of silver to Viamat by ring-fencing some vault space. JP Morgan could then call this space a JP Morgan vault, even though it would be physically within a location managed by one of the security storage / transport providers.

I now think on balance that HSBC probably took the same approach with its gold vault and has it located in a Brinks facility, but that it calls it a HSBC vault. This could also mean that HSBC uses Brinks to store silver, while referring to it as HSBC storage. As to whether HSBC and JP Morgan store gold at the Bank of England while labelling it as a HSBC or JP Morgan storage area is another interesting question, but is beyond the scope of discussion here.

Note, there is also an iShares Silver Bullion Fund known as SVR which uses Scotia Mocatta as a custodian, which as of 29 June held 2,154 silver bars, however, SVR mostly holds its silver bars mostly in Toronto with Scotia, with a small number of silver bars stored with Scotia in New York. SVR therefore does not store any silver bars in London. See latest SVR weight list here.

ETF Securities – 6 ETFs

Keeping track of all the silver-backed ETFs offered by ETF Securities is challenging to say the least, but in the below discussion I’ve tried to devise a system which will make things at least a little clearer.

ETF Securities operates 6 ETFs which hold physical silver bars that are stored in the LBMA precious metals vaulting network in London. Of these 6 ETFS, 3 of them hold silver bars and nothing else. The other 3 ETFs are precious metals baskets which hold ‘physical’ gold, silver, platinum and palladium. Two of these ETFs are domiciled in the UK, 2 are domiciled in Australia, and the other 2 are domiciled in the US. In each of the UK, Australia and the US, ETF Securities offers 1 silver ETF and 1 precious metals basket ETF.

It’s most convenient to refer to the codes of these ETFs when discussing them. The 2 UK domiciled ETFs, with codes PHAG (silver) and PHPM (precious metals basket), are positioned under a company called ETFS Metal Securities Limited (MSL). The 2 ETFs domiciled in Australia, with codes PMAG (silver) and PMPM (precious metals basket), fall under a company called ETFS Metal Securities Australia Limited (MSAL). The final 2, which are US domiciled, are known as SILV (silver) and GLTR (precious metals basket).

ETFS Metal Securities Limited (MSL) – PHAG and PHPM

ETFS Physical Silver (PHAG) has a primary listing on the London Stock Exchange (LSE) and trades in USD. It’s NAV is also in USD. The custodian for PHAG is HSBC Bank Plc, with a listed vault location of London. Note: There is also another variant of PHAG called PHSP. It’s the same security as PHAG (same ISIN) but its trades in GBP (and its NAV is calculated in GBP). Its best to ignore PHSP as it’s literally the same fund.

ETFS Physical PM Basket (PHPM) is a precious metals Basket ETF that also holds gold, platinum, and palladium, in addition to silver. The custodian is HSBC Bank Plc with a vault location in London. There is also a GBP variant of PHPM called PHPP. Again, just ignore PHPP in this analysis.

ETFS Metal Securities Limited (MSL) officially reports all of its precious metals holdings in the same report (which it reports on each trading day). Since PHAG and PHPM are part of MSL, PHAG and PHPM silver bar holdings are reported together. According to the MSL weight list, as of 30 June 2017, MSL held 62,427 London Good delivery silver bars containing 60,280,155 troy ounces of silver (1875 tonnes). The individual ETFs within MSL also report their own holdings. However, there is a slight mismatch between dates on the individual fund pages and the date in the MSL spreadsheet with PHAG and PHPM reporting 29 June, while MSL has reported 30 June.

It’s not a big deal though. As of 29 June, PHAG held 58,777,148 troy ozs of silver (1828.2 tonnes) and PHPM held 1,480,037 troy ozs of silver (46 tonnes), which together is 60,257,185 troy ounces of silver (1874.25 tonnes), which is very close to the MSL reported number. Overall, PHAG holds 97.5% of the silver that is held in MSL, and PHPM only holds about 2.5% of the silver held in MSL.

Now, here’s the crux. While MSL uses HSBC Bank Plc in London as custodian for its silver, HSBC also uses Malca Amit London as sub-custodian, and the Malca Amit vault holds more than twice the amount of MSL silver (i.e. predominantly PHAG silver) than the HSBC vault. MSL’s reported silver holding are distributed as per the following table:

ETF Securities -
ETF Securities PHAG and PHPM – Silver bars and Ounces by location – Click to Enlarge

MSL holds 62,427 London Good Delivery silver bars in LBMA vaults in London, containing 60.28 million ounces of silver (1875 tonnes of silver). The Malca Amit vault stores 42,917 of these bars (1283 tonnes), and a HSBC vault stores another 19,510 silver bars (592 tonnes).

Inspectorate is also the independent auditor for the silver held by MSL. According to the latest Inspectorate audit letter, dated 3 March 2017 but referring to an end audit date of 31 December 2016, the silver in MSL was held in the vaults of HSBC Bank plc, London and at the vaults of Malca-Amit London.

ETFS Metal Secs. Australia Ltd (MSAL) – PMAG  & PMPM

ETFS Physical Silver (PMAG), domiciled in Australia, is an ETF which only holds silver, and holds this silver in London with custodian HSBC Bank plc at a vault location in London. Note: ETF Securities officially refers to PMAG as ETPMAG.

ETFS Physical PM basket (PMPM) is a precious metals Basket ETF that also holds gold, platinum, and palladium, in addition to silver. The custodian of PMPM is HSBC Bank plc with a vault location in London. Note: ETF Securities officially refers to PMPM as ETPMPM.

In a similar way to UK domiciled MSL, MSAL (the ETFS Australian company) reports on all of its precious metals holdings in one daily spreadsheet including the silver in PMAG and PMPM. As of 30 June 2017, MSAL held 2754 silver bars in a HSBC vault in London, containing 2,664,690 troy ounces of silver (82.88 tonnes of silver).

Of the 2,664,690 ounces of silver held by MSAL, over 98%, or 2617,229 ounces, is held by PMAG, with less than 2% held in PMPM (47,362 ounces). The actual figures are 98.22% vs 1.78%. This means that PMAG roughly holds 2705 silver bars, and PMPM holds 49 silver bars.

Inspectorate is, not surprisingly, also the independent auditor for MSAL’s metal holdings, and as per the latest audit letter for record date 31 December 2016, the silver bars audit location is stated as having been “HSBC Bank plc, London“.

ETF Securities US domiciled ETFs: SIVR and GLTR

The final two ETF Securities ETFs which hold silver bars are the ETFS Silver Trust (SIVR), and the ETFS Precious Metals Basket Trust (GLTR). HSBC bank plc is the custodian of SIVR and JP Morgan is the custodian of GLTR. However, GLTR also uses Brinks as a sub-custodian.

The latest silver bar weight list spreadsheet for the ETFS Silver Trust (SIVR), dated 29 June, which is titled “HSBC US Silver Bar List”, states that the SIVR Trust holds 21,437 silver bars containing 20,363,315 troy ozs of silver (633.4 tonnes of silver). There is no mention of SIVR holding any of its silver with a sub-custodian. The latest independent audit report for SIRV, by Inspectorate, for an audit reference date of 31 December 2016, states that the audit took place “at the vault of HSBC Bank plc, London (the “Custodian”)“, where Inspectorate found “20,108 London Good Delivery Silver Bars with a weight of 19,171,492.300 troy ounces.

The latest silver bar weight list for the ETFS Precious Metals Basket Trust (GLTR), also dated 29 June, and which is titled “JPM Precious Metals Basket Bar List“,  states that the GLTR Trust holds 5,670 silver bars containing 5,496,035 ozs of silver (~ 171 tonnes of silver).

However, while 85% of these bars (144.5 tones of silver) are stored in the ‘JP Morgan V‘ vault, 15% of the silver bars (26.5 tonnes of silver) are stored in a ‘Brinks 2‘ vault. So according to GLTR naming convention, as there is a ‘Brinks 2’ vault, presumably when it was first named, there was also a ‘Brinks 1’. ‘Brinks 2’ could possibly be referring to the same location as the ‘Brinks London A’ vault.

GLTR
ETF Securities (GLTR) –Silver bars and Ounces by location – Click to Enlarge

Inspectorate is also the independent auditor for the precious metals held by GLTR. In the latest Inspectorate audit letter for GLTR, with an audit reference date of 31 December 2016, Inspectorate states that its audit was only conducted “at the vault of J.P. Morgan Chase N.A, London (the “Custodian”)” where it counted “4,873 London Good Delivery Silver Bars“. This probably means that GLTR’s holdings of silver bars in the ‘Brinks 2’ vault are quite recent, i.e. they have been acquired since 31 December 2016.

SOURCE – Physical Silver P-ETC

A silver-backed ETF offered by the ETF provider ‘SOURCE’, which is named the Physical Silver P-ETC, holds its silver bars in a London vault of  custodian JP Morgan. The SOURCE ETF platform was originally established in 2008 as a joint venture between Goldman Sachs, Morgan Stanley, and Merrill Lynch.

The latest silver bar weight list for the Physical Silver P-ETC (dated 23 June) states that it holds 3,129,326 troy ounces of silver (97.34 tonnes of silver). The list does not state an exact bar count, but looking at the weight list, there are about 3,237 silver bars listed.

Inspectorate is also the independent auditor for the Physical silver P-ETC. The latest Inspectorate audit letter, conducted on 4 January 2017, states that at that time, this ETF held 2,048 silver bars containing 1,982,343 troy ounces of silver. This is interesting because about a week ago, this SOURCE Physical silver P-ETC held about 4 million ozs of silver. Now it holds 3.1 million ounces of silver, and at the start of the year it held under 2 million ounces of silver. So the quantity of silver held in this SOURCE silver ETF fluctuates quite dramatically.

Deutsche Bank ETFs

There are 3 ETCs listed on the Exchange Traded Commodity (ETC) section of the Deutsche Asset Management website which hold physical silver in London. These 3 ETCs are as follows:

  • db Physical Silver ETC
  • db Physical Silver ETC (EUR)
  • db Physical Silver Euro hedged ETC

The Factsheets for these 3 Deutsche ETCs all list the custodian as “Deutsche Bank”, but list the sub-custodian as “JP Morgan Chase Bank”. For example, the Factsheet for the db Physical Silver ETC states

“Custodian/Sub-custodian:       Deutsche Bank AG/JP Morgan Chase Bank N.A.”

Shockingly, there do not seem to be any recent independent audit documents for any of these Deutsche ETCs anywhere on the Deutsche Asset Management website. The latest ‘Inventory Audit’ document in the ‘Download Center’ of the website is dated November 2012. That audit document can be viewed here. The old audit document stated that on 25 September 2012, ‘DB ETC Plc’ held 13,314 silver bars containing 13,040,194.3 troy ounces of silver (405.6 tonnes of silver), and that the audit was conducted at ‘Custodian and Location‘ of ‘JP Morgan Chase Bank, N.A. London‘. I have scanned the entire website and there is no sign of any other audit documents or any silver bar weight list.

The initial metal entitlement for units issued in each of these 3 ETCs was 10 troy ounces per unit. The latest units issued figures from Deutsche (dated 22 June 2017) for these ETCs is as follows:

  • db Physical Silver ETC: 277, 500 units issued
  • db Physical Silver ETC (EUR): 533,000 units issued
  • db Physical Silver Euro hedged ETC: 878,000 units issued
  • Total units issued for silver-backed db ETCs = 1,688,500 units

This would mean that in total, these 3 ETCs would have had an initial metal entitlement of 16,885,000 troy ounces of silver. However, due to what looks like operational fees being offset against the metal in these ETCs (i.e. selling silver to pay fund expenses), the effective metal entitlement for each of the 3 ETCs is now stated on the Deutsche website as being less than 10 troy ounces.

For db Physical Silver ETC, the entitlement is 9.6841 ounces. For db Physical Silver ETC, the entitlement is 9.6930 ounces and for db Physical Silver Euro hedged ETC the metal entitlement is a very low 7.9893 ounces.

Therefore, the amount of silver backing these ETCs looks to be (277500 * 9.6841) + (533000 * 9.693) + (878000 * 7.9893) = 14,868,312 troy ounces = 462.5 tonnes. Since there is no bar count, an approximate bar count assuming each bar weighs 1000 oz would be 14,870 Good Delivery silver bars.

Since there are no audit reports and no silver bar weight list for these ETCs, it’s difficult to know if real allocated silver in the form of London Good Delivery silver bars is backing these Deutsche Bank db ETCs, let alone trying to figure how many silver bars are in a JP Morgan vault in London backing these Deutsche products. We can therefore use 462.5 tonne for Deutche but with a caveat that there is no current silver bar weight lists or independent audit documents.

Total ETF Silver held in London LBMA Vaults

Adding up the silver held in the 11 ETFs profiled above yields the following table. In total, the 11 ETFs hold approximately 12,041 tonnes of silver (387.2 million troy ounces) across 4 vault operators. Brinks vaults hold 48% of the total, and JP Morgan vaults hold another 30%. HSBC and Malca Amit hold about 11% each of the remainder.

ETF Silver Holdings -  Tonnes, Held in London Vaults

ETF Silver Holdings – Tonnes, for Silver stored in London LBMA Vaults

In terms of London Good Delivery silver bars, these 11 ETFs hold approximately 400,000 of these silver bars. Since the 3 Deutsche ETFs (ETCs) don’t have an available bar list, I converted the assumed troy ounce holdings to bar totals by assuming each bar held weighs 1000 ozs. Brinks stores over 191,000 of these Good delivery silver bars. That equates to nearly 6,400 pallets with 30 silver bars per pallet. If the pallets were stacked 6 high, and arranged in a square, that would be an area 32 pallets long by about 33 pallets wide. In addition, Brinks may also store silver on behalf of HSBC, or even on behalf of JP Morgan. Who knows?

Number of ETF held Good Delivery Silver Bars stored in London LBAM vaults
Number of ETF held Good Delivery Silver Bars stored in London LBMA vaults

BullionVault

According to the latest numbers on the BullionVault website (Daily Audit), BullionVault has 349,939.57 kgs of silver stored in London. That equates to 11,250,557 troy ozs of silver, or 350 tonnes of silver. This silver is stored in the form of London Good Delivery Silver Bars. According to the BullionVault website, BullionVault use Loomis as a custodian for storing silver bars in London:

“The London (UK) vault is run on our behalf by Loomis International

Those with a BullionVault login can go in and view BullionVault’s latest silver bar weight list which has been generated by Loomis, but BullionVault don’t allow this list to be published externally. Suffice to say, the latest list, dated 11 May, lists 11,544 silver bars which are stored across nearly 400 pallets.

GoldMoney

The GoldMoney website has a real-time audit page which currently states that GoldMoney has 202,057.614 kgs of silver. That equates to 6,496,153 troy ozs of silver, or 202 tonnes of silver stored in London. This silver is also stored with Loomis. At least some of this silver and probably a lot of it is in the form of London Good Delivery silver bars. Without being able to log in to the site properly, it’s not possible to see a bar list.

So between them, BullionVault and GoldMoney have 550 tonnes of silver stored in Loomis vaults in London. My guess is that Loomis (formerly Viamat) store precious metals in a warehouse in Shepperton Business Park, Govett Avenue, Shepperton, a warehouse which is in the corner of the business park, beside the railway track.

Adding this 550 tonnes of silver to the 12040 tonnes of silver held by the 11 ETFs above gives a figure of 12,590 tonnes. Let’s call it 12,600 tonnes. This is then the lower bound on the amount of physical silver in the LBMA vaults in London.

Thomson Reuters GFMS – “Custodian Vault” silver

On its ‘Silver Supply’ web page, the Silver Institute website has an interesting data table titled “Identifiable Above-Ground Silver Bullion Stocks” which lists 5 categories of above-ground silver stocks, namely ‘Custodian vaults’, ‘ETPs’, ‘Exchanges’, ‘Government’, and ‘Industry’.

What’s notable and striking about this table is that the ‘Custodian Vaults‘ category for 2016 amounts to a very large 1571.2 million troy ounces of silver (50,440 tonnes), and also the fact that this ‘Custodian vaults’ category is distinct from silver held in ‘Exchanges’ (such as COMEX and TOCOM) and ETPs / ETFs (such as the ETF products discussed above). The ‘Custodian Vaults’ category also does not include ‘Government’ stockpiles or ‘Industry’ inventories. The actual table and the data in the table are sourced from the Thomson Reuters GFMS “World Silver Survey” 2017 edition. As you will see below, this ‘Custodian Category’ refers to holdings of silver which are not reported, but which are stored in custodian vaults, including in the London vaults. This category therefore needs to be examined in the context of the LBMA’s imminent reporting of silver holdings in the LBMA London vaulting system.

GFMS World Silver Survey - Identifiable silver stocks
Above-ground Identifiable Silver Stocks –  Source: GFMS World Silver Survey 2016. Click to Enlarge

You can also see from the above table that this 2016 Custodian Vaults figure of 1571.2 million ozs (50,440 tonnes) grew from a 2008 total of 615.6 million ozs (19,148 tonnes), so in eight years has risen more than 250%.

On pages 37-38 of this GFMS World Silver Survey 2017 (pdf – large file), GFMS makes some very interesting assertions. GFMS starts by defining what it calls Identifiable silver bullion stocks. It states:

‘Identifiable bullion stocks can be split into two categories: unreported GFMS stock estimates that are based on confidential surveys and field research;  [and secondly] stocks that are reported.

Unreported stocks include the lion’s share of our government category and our custodian vault category.”

Reported inventories are predominantly held in ETPs..but also include some of the government and industry stockpiles.”

However, in the accompanying commentary to the above table, GFMS classifies all ETP, Exchange and Industry holdings as “Reported“, and all Custodian Vaults and Government holdings as “Unreported“. Therefore, it is useful to regroup the 2016 figures from the above table into a Reported category and an Unreported category, as the GFMS commentary then makes more sense. A regrouped table of the 2016 data is as follows, and illustrates that ‘Custodian Vault’ holdings of silver (none of which are reported) account for a whopping 61% of all above ground silver:

Identifiable Above_ground silver stocks grouped by Reported and Unreported
Identifiable Above-ground silver stockpiles, 2016  – grouped by Reported and Unreported categories

A GFMS bar chart in the 2017 World Silver Survey also underscores the dominant position of these (unreported) ‘Custodian Vault’ holdings:

Above
Identifiable above-ground Silver grouped into 5 categories, 2007 – 2016. Largest % is ‘Custodian Vaults’. Source: GFMS World Silver Survey 2017

GFMS goes on to say that in 2016 “Reported stocks were 36% of identifiable stocks“. Conversely, we can see that ‘Unreported’ silver stocks (Custodian Vaults and Government) were 64% of identifiable stocks.

GFMS says that for 2016 “71% of reported stocks were ETPs“, the rest being Exchange and Industry classifications. Exchanges refers to silver held in warehouses of COMEX (NY), TOCOM (Japan) and the SGE and SHFE (China). COMEX is currently reporting 209 million ouzs of silver in its approved warehouses in New York, of which 172 million ozs in Eligible and 37 million ozs is in the Registered category.

Interesting, but on a side note, GFMS also states  in its 2017 silver report that as regards COMEX silver inventories:

“Eligible stocks reported by COMEX contain a portion that is allocated to ETPs”.

“At the end of 2016, the portion of COMEX Eligible stocks that was allocated to ETPs was around 16% of total COMEX eligible stocks.”

This will probably be an eye opener for those interested in COMEX silver warehouse stocks.

Addressing ‘Custodian Vault‘ stocks of silver, GFMS says that Europe’s share of Custodian Vault stocks was 488.7 million ozs (15,201 tonnes) in 2016 and accounted for 31% of total Custodian Vault stocks. Asian ‘Custodian Vault’ stocks of silver were just over 50% of the total with the remainder in North America (Canada and US).

Silver holdings in Custodian Vaults by Region

Silver holdings in Custodian Vaults by Region, 2007 -2016. Source – GFMS World Silver Survey 2017

But what do these ‘Custodian Vault’ stocks of silver refer to?

GFMS does not provide a detailed answer, but merely mentions a number of examples, which themselves vary by region. For Asia GFMS says “the bulk of these stocks are located in China, and reflects stocks held in vaults at banks“, and also ” other parts of Asia, such as Singapore, have been increasing in popularity for storage of bars and coins in recent years“, while in India “global bullion banks increasingly seeking this location as a strategic point for silver vaulting in case the need arises.” There are also silver “stocks in Japan”. From a BullionStar perspective, we certainly are aware that there is a lot of silver bullion in vault storage in Singapore, so the GFMS statement is accurate here.

In North America, GFMS attributes the “growth in silver custodian vaulted stocks not allocated to ETPs” to a “drop in coin sales in North America last year“. In the 2016 edition of the World Silver Survey, GFMS said that the growth in custodian vault holdings was partially due to “the reallocation by some North American investors from their ETP holdings” [into custodian holdings].

Turning to Europe, GFMS says that the growth in Custodian vault silver holdings “can be attributed to increased institutional investor interest“. Therefore, according to GFMS, institutional investors in Europe are buying silver and holding real physical silver in Custodian vaults.

With 488.7 million ozs (15,201 tonnes) of silver held in Europe in ‘Custodian vaults’ that is not reported anywhere, at least some of this silver must be held in London, which is one of the world’s largest financial centers and the world’s highest trading volume silver market.

GFMS states:

“Custodian vault stock data excludes ETP Holdings, but it is important to note that most custodians of ETP silver stocks also store silver in vaults that are not allocated to ETPs. the same is true of futures exchange warehouses.” 

So how much of this 15,201 tonnes of ‘Custodian Vaults’ silver that is said to be in Europe is actually in London vaults? Apart from London, there would presumably also be significant physical silver holdings vaulted in Switzerland and to a lessor extent in countries such as Germany, the Netherlands and maybe Austria etc. So whats’s a suitable percentage for London? Given London’s extensive vaulting network and prominence as a hedge fund and institutional investment centre, a 40-50% share of the European ‘custodian vault’ silver holdings would not be unrealistic, with the other big percentage probably vaulted in Switzerland. This would therefore put previously ‘Unreported’ silver holdings in the London vaults at between 6080 tonnes and 7600 tonnes (or an additional 182,000 to 230,000 Good Delivery Silver bars).

Adding this range of 6080 – 7600 tonnes to the 12,040 tonne figure that the 11 ETFs above hold, gives a total figure of 18,120 – 19,640 tonnes of silver stored in the LBMA vaults in London (545,000 – 585,000 Good Delivery silver bars).

Note, BullionVault and GoldMoney silver is technically part of the ‘Custodian Vault’ figure, so can’t be counted twice.

ps: In its 2017 World Silver Survey, GFMS also stated that in 2016, ETPs (ETFs) held 664.8 million ounces of silver “with 75% of the total custodian vaulted stocks [that were] allocated to ETPs held in Europe and 24% in North America. Asia makes up the balance of less than 1%.“. This would mean that as of the date of the GFMS calculation for 2016, 498.6 million ounces of ETF silver was vaulted in Europe.

Above, I have accounted for 387.1 million ounces of silver that is currently stored in London on behalf of 11 ETFs. There are also 3 Swiss Silver ETFs which store their silver in Switzerland. These are ZKB (currently with 74.9 million ozs), Julius Baer (currently with 13.7 million ozs) and UBS (currently with 5.89 million ozs), giving a total of 94.49 million ozs of silver for these 3 Swiss based platforms. Therefore, between London vaults and vaults in Switzerland, there are currently 14 ETFs that together hold 481.6 million ounces of vaulted silver (14,980 tonnes of vaulted silver).

Conclusion

When the LBMA finally manages to publish its first report on the silver and gold stored in the LBMA vaults in London in the coming days, we will have a clearer picture of how much physical silver is actually in these mysterious and opaque vaults.

A lower bound based on ETF holdings and BullionVault and GoldMoney holdings would be about 12600 tonnes of silver. A higher bound that also reflects ‘Custodian Vault’ holdings could be in the region of 18120 – 19640 tonnes of silver. There would probably also be some LBMA bullion bank float, which may or may not be included in ‘Custodian Vault’ figures, that could push the silver total to over 20,000 tonnes or more.

The LBMA perennially claims that it wants to bring transparency to the London precious metals market. This has been a very hollow mantra for a long time now. However, while some of the LBMA members may want this transparency, others, possibly some of the powerful bullion banks or their clients, certainly don’t want transparency. Take a case in point. At the Asia Pacific Precious Metals Conference (APPMC) in Singapore in early June, the LBMA CEO in a speech to the conference talked about the difficulty of even getting a press release out about the upcoming publication of gold and silver vault holdings data. She said (fast forward to 8:37 in the below video):

It was actually a huge achievement just to get the press release out.”

For what is supposed to be a mature and efficient financial marketplace, this is a truly bizarre occurrence, and it must be pretty obvious that some of the vested interests in the London gold and silver markets needed to be dragged kicking and screaming over the finish line as regards being in any way open about how much gold and silver is actually in these LBMA London vaults.

But now, according to the LBMA CEO in the same part of her speech, even so-called “credible investors” (as opposed to uncredible investors?) also “find it a little odd that as a marketplace, there’s no data“, which may explain the vampires within the LBMA being dragged into the daylight.

Hopefully with the above analysis and the upcoming aggregated LBMA silver vaulting numbers, these “credible investors” (and the hundreds of millions of other silver investors around the world) will now be less in the dark about the amount of silver in the London LBMA vaulting network, and will now have better information with which to make investment decisions when buying silver and selling silver.

German and Swiss Precious Metals Refiners join forces as Heraeus acquires Argor-Heraeus

German precious metals group Heraeus Precious Metals (HPM), part of the Heraeus industrial group, has just announced the full acquisition of Swiss precious metals refining group Argor-Heraeus. Heraeus is headquartered in Hanau, just outside Frankfurt. Argor-Heraeus is headquartered in Mendrisio in the Swiss Canton of Ticino, beside the Italian border.

The Heraeus takeover announcement, on 3 April 2017, continues a noticeable acquisition trend in the Swiss precious metals refining sector and follows the July 2016 acquisition of Neuchâtel based Metalor Technologies by Japanese Tanaka Precious Metals, and the takeover of Ticino-based Swiss gold and silver refiner Valcambi by Indian group Rajesh Exports in July 2015. The Heraeus press release from Monday 3 April can be read here in English.

singapore-germany-silver-bullionstar-bar-1kg-2015-refined

A Deal Telegraphed in November

In early November 2016, BullionStar was among the first to report that Swiss Argor-Heraeus was indeed an acquisition target. At the time, market sources had indicated that the most likely acquirer was a private equity company Capinvest, with other suitors said to be Japanese group Asahi and Swiss based MKS-PAMP.

In late 2016, S&P Global Platts reported that Swiss private equity company “Capvis” was in talks to acquire Argor-Heraeus, with one of Platts sources quoting a purchase price in the region of €200 million with completion in Q1 2017, while another source said €200 million was too high a figure. At the end of the day, a Capvis takeover did not materialise and earlier this year market sources said that Argor-Heraeus was no longer for sale (externally). In hindsight, it was probably at this stage that Heraeus decided to make its move. Alternatively, the discussions with external buyers may have just been conducted so as to gauge sentiment and establish a series of potential valuations for the Swiss refiner.

As a reminder, Argor-Hereaus had an unusual ownership structure in that it was jointly owned by 4 shareholders, namely German group Heraeus, German bank Commerzbank, the Austrian Mint, and Argor-Heraeus management. Prior to the takeover, Heraeus was the largest shareholder holding 33% of Argor-Heraeus shares, with Commerzbank holding a further 32.7% of the equity, the Austrian Mint holding another 30%, and Argor-Heraeus’s management holding the balance of shares.

As an existing shareholder and board member of Argor-Heraeus, the Heraeus group would have been privy to all of Argor-Heraeus’s financial and operational details, and so would have been in an advantageous position to negotiate purchase price details with Commerzbank and the Austrian Mint, which would have been a natural advantage relative to external potential acquirers.

Purchase Price

However, the exact purchase price Argor-Heraeus is not known, since, according to the Heraeus press release “the parties have agreed not to disclose financial details of the deal”. Notwithstanding this, German newspaper Handelsblatt is claiming that the Heraeus takeover values Argor-Heraeus at “half a billion Swiss Francs“, since according to Handelsblatt’s sources, Heraeus paid “few hundred million euros for the remaining Argor shares“. With CHF 500 million equal to approximately €468 million, the Handelsblatt claim would mean that Heraeus may have paid €313 million for the 67% of Argor-Heraeus that it did not own. This would be far higher than the €200 million figure that S&P Platts mentioned in December.

singapore-gold-bullionstar-bar-100g-back

Motivations for Acquisition

According to Heraeus, one of its motivations in acquiring Argor-Heraeus is to strengthen its capabilities in gold and silver refining by tapping “Argor’s expertise and processing capacity for gold and silver”, since Heraeus considers itself strongest in platinum group metals. Heraeus states that another driver of the acquisition is geographical diversification given that Argor-Heraeus has facilities on the ground in Chile, as well as in Italy, Germany and of course Switzerland, while Heraeus has a strong presence in Asia, North America and India in addition to Germany.

BullionStar Bars by Heraeus and Argor-Heraeus

BullionStar is very familiar with both Heraeus and Argor-Heraeus and has a good working relationship with both refinery groups. In fact, the popular BullionStar 1 kg silver bars are produced by Heraeus on behalf of BullionStar, while BullionStar’s own branded 100 gram gold bars are minted by Argor-Heraeus on behalf of  BullionStar. See bar images above. Argor-Heraeus is one of the five refineries appointed to the LBMA’s Good Delivery List referee panel for Gold and Silver. In addition, BullionStar carries a wide range of other Heraeus silver bars, Heraeus gold bars, and Argor-Heraeus gold bars.

Conclusion

With 3 of the 4 giant Swiss precious metals refineries having now been acquired by new owners within less than 2 years of each other, this leaves the PAMP refinery, owned by MKS PAMP, as the only one of the “Big 4” Swiss refineries to have bypassed this recent flurry of corporate control activity. As to whether MKS PAMP will itself become a takeover target is debatable, but it would be surprising if MKS isn’t thinking about this very question right now.

Further information

For more information on the Heraeus group and its precious metals activities, see BullionStar Gold University profile of Heraeus https://www.bullionstar.com/gold-university/heraeus-refinery

For a full overview of Swiss refiner Argor-Heraeus, please see BullionStar Gold University profile of Argor-Heraeus https://www.bullionstar.com/gold-university/argor-heraeus-refinery

For more information on the Tanaka acquisition of Metalor Technologies, see specific section of Metalor profile on BullionStar Gold University https://www.bullionstar.com/gold-university/metalor-refinery#heading-4

For more information on the acquisition of Swiss Valcambi by Rajesh Exports, see BullionStar blog https://www.bullionstar.com/blogs/ronan-manly/swiss-gold-refineries-and-the-sale-of-valcambi

For analysis of initial news (last November) about Argor-Heraeus being acquired, see BullionStar blog https://www.bullionstar.com/blogs/ronan-manly/swiss-gold-refinery-argor-heraeus-to-be-acquired-by-private-equity-investors

More Bad News for the LBMA Silver Price, but an Opportunity for Overhaul

On Friday 3 March 2017, in a surprise announcement with implications for the global silver market, the London Bullion Market Association (LBMA) informed its members that the current administrator and calculation agent of its recently launched LBMA Silver Price auction, Thomson Reuters and the CME Group respectively, will be pulling out of providing their services to the problematic London-based silver price benchmark within the near future. Thomson Reuters and the CME Group issued identical statements.

This is surprising because Thomson Reuters and the CME Group only began administering / calculating the LBMA Silver Price auction two and a half years ago in August 2014, when, amid much hubris, the duo were awarded the contract after a long-drawn-out and high-profile tender process. Notably, the Thomson Reuters  / CME contract with the LBMA was for a 5-year term running up to and into 2019. So the duo are now pulling out mid-way through a contract cycle.

More surprisingly, in their statements of 3 March, the LBMA / Thomson Reuters and CME allude to the European Benchmark Regulation being in some way responsible for the hasty departure. However, given that the units of CME and Thomson Reuters that are parties to the LBMA contract are their specialist benchmark units “CME Benchmark Europe Limited” and “Thomson Reuters Benchmark Services Limited”, which specialise in administering and calculating benchmarks, this excuse makes no sense.

In essence, this development is an embarrassment for all concerned and could lead to further reputational damage for the parties involved. It also now re-focuses market scrutiny on an area which the LBMA and its associates could well wish to forget, i.e. the former London silver fixing run by the infamous London Silver Market Fixing Limited, a company which itself is still one of the defendants, along with HSBC, Bank of Nova Scotia and Deutsche Bank, in a live New York class action suit that is scrutinizing the manipulation of the London silver price.

LBMA Silver Price: A Regulated Benchmark

Note that the LBMA Silver Price benchmark is now a “Regulated Benchmark” under United Kingdom HM Treasury Legislation, and is one of 8 financial market benchmarks regulated by the UK’s Financial Conduct Authority (FCA). So this is not some backwater obscure benchmark that we are talking about here. This is a benchmark with far-reaching effects on the global precious metals markets and a sister of the LBMA Gold Price benchmark. The reference prices from these benchmarks are used from everything from valuing Exchange Traded Funds (ETFs) to being the price reference points in ISDA swaps and bullion bank structured products such as barrier options.

According to the LBMA’s usual public relations mouthpiece Reuters, which relayed the news to the broader market on 3 March, the LBMA will be:

“looking to identify a new provider in the summer, and have the new platform up and running in the autumn”

This dramatic “exit stage right” by Thomson Reuters and the CME Group is a far cry from their initial and continued corporate spin of being committed to the silver price auction, which they claimed both at auction launch in August 2014, and also as recently as 2016 when they grovelled with promises of process improvement and wider participation in the auction in the wake of the silver price manipulation fiasco in the LBMA Silver Price auction on 28 January 2016.

It was on 28 january 2016 that the midday auction took a whopping 29 rounds to complete and the price derived in the auction was manipulated down by a massive 6% under where silver spot and silver futures prices were trading at that time. See the beginning of BullionStar blog “The LBMA Silver Price – Broken Promises on Wider Participation and Central Clearing” for further details about the 28 January auction.

TRCMEsquare

 

Where is the Commitment?

On 15 August 2014, the day the LBMA Silver Price auction was launched, William Knottenbelt, MD at CME Group stated:

“Through our existing relationships with market participants and the broader silver marketplace we are uniquely positioned to provide a seamless transition for the spot silver benchmark in London.” 

“CME Group has a long and successful history of offering benchmark risk management and price discovery solutions for the global precious metals markets.” 

Then, on 22 March 2016, when CME and Thomson Reuters introduced some changes to the auction in the wake of the 28 January 2016 auction price manipulation, both parties released more spin on their continued commitment to the auction. Thomson Reuters’ Head of Benchmark Services, Tobias Sproehnle, in a statement that now looks to be hollow, said:

“these changes together with a comprehensive consultation with the broader silver community – producers, intermediaries and consumers – are a further demonstration of Thomson Reuters and CME Group’s commitment to providing innovative, market leading benchmarks for the Silver market.

While Gavin Lee, the head of CME Benchmark Services, led with an equally hubristic statement that:

“in consultation with Silver market participants, we are always looking for new ways to develop this benchmark further

These statements from CME and Thomson Reuters, less than a year ago, run totally contrary to the fact that the duo are now going to abandon the LBMA Silver Price auction ship, which will necessitate the appointment of a replacement administrator and calculation agent. Where is the continued “commitment” to the silver benchmark and the silver market that they were we eager to espouse last March?

Why the Hasty Departure?

According to the Reuters news report last Friday 3 March:

A spokesman for Thomson Reuters confirmed the company was stepping down from the process. CME could not immediately be reached for comment.

Not very informative or cooperative from either party when one of the providers was not even available to explain its exit rationale, and the other merely confirms a fact to its in-house news arm, a fact which the LBMA had already announced earlier that day to its members.

However, if you look at the CME Group website, a short announcement was added to its website on 3 March 2017, which states:

The forthcoming European Benchmark Regulation, due to be implemented in January 2018, prompted a review of the existing LBMA Silver Price administration arrangements and, in consultation with the LBMA, CME Group and Thomson Reuters have decided to step down from their respective roles in relation to the LBMA Silver Price auction.

This statement was also added to the Thomson Reuters website on 3 March.

Before briefly looking at the relevance of this “European Benchmark Regulation”, which the Reuters news article even failed to mention, its notable that the CME / Thomson Reuters early withdrawal was also covered on 3 March by the MetalBulletin website.

According to MetalBulletin (subscription site), the above statement by CME is apparently part of an identical statement which the LBMA released to it members on Friday 3 March (the LBMA statement).

MetalBulletin adds in its commentary that:

“CME is looking to streamline its precious metals division, with contracts in this area being its fastest growing asset. The exchange wants to focus on its core products, Metal Bulletin understands.”

What MetalBulletin means by this I don’t know. The logic doesn’t make any sense. The sentence doesn’t even make sense. Benchmarks are a core product of CME group. CME even states that it offers:

“the widest range of global benchmark products across all major asset classes”

CME Benchmark Europe Limited was specifically set up in 2014 to provide the calculation platform for the LBMA Silver Price. Furthermore, CME has just launched a suite of silver and gold futures contracts for the London market (launched in late January 2017), the silver contract being the “London Spot Silver Futures (code SSP)“. Even though these CME contracts have had no trading interest so far, the CME claims that it is currently “working with major banks to synchronize their systems to start trading” these contracts (London Spot Silver Futures and London Spot Gold Futures).

So why would CME want to voluntarily ditch the provision of a high-profile London silver benchmark, when it could attain trading synergies between the LBMA Silver Price and its new London silver futures contracts, or at the very least improve brand recognition in the market?  And not to forget CME and Thomson Reuters claim a”commitment to providing innovative, market leading benchmarks for the Silver market“.

European Benchmark Regulation

Turning to the new “European Benchmark Regulation”, what exactly is it, and why would it be relevant for the LBMA and CME and Thomson Reuters to mention the European benchmark Regulation in the context CME and Thomson Reuters pulling out of the LBMA Silver Price auction?

At its outset, the European Benchmark Regulation was proposed by the European Commission. The Commission’s proposal was also issued in coordination with a range of entities and initiatives such as MiFID, the Market Abuse Directive, the benchmark setting processes of the  European Securities and Markets Authority (ESMA) and European Banking Authority (EBA), and also the IOSCO financial benchmark principles.

According to law firm Clifford Chance:

The new [EU] Regulation is a key part of the EU’s response to the LIBOR scandal and
the allegations of manipulation of foreign exchange and commodity benchmarks

“The Regulation imposes new requirements on firms that provide, contribute to or use a wide range of interest rate, currency, securities, commodity and other indices and reference prices.”

“Most of the new rules will not apply until 1 January 2018”

“The new Regulation imposes broad ranging and exacting requirements
on a wide range of market participants. It may reinforce the trend to discontinue benchmarks and reference prices

According to law firm Simmons & Simmons:

The Regulation seeks to:

  • improve governance and controls over the benchmark process, in particular to ensure that administrators avoid conflicts of interest, or at least manage them adequately
  • improve the quality of input data and methodologies used by benchmark administrators
  • ensure that contributors to benchmarks and the data they provide are subject to adequate controls, in particular to avoid conflicts of interest
  • protect consumers and investors through greater transparency and adequate rights of redress.

The Regulation aims to address potential issues at each stage of the benchmark process and will apply in respect of:

  • the provision of benchmarks
  • the contribution of input data to a benchmark, and
  • the use of a benchmark within the EU.

All of these goals aspired to by the legislation of the European Benchmark Regulation seem reasonable and would benefit users of the LBMA Silver Price auction, so given the above, it seems very bizarre that CME and Thomson Reuters and the LBMA stated last Friday 3 March that:

The forthcoming European Benchmark Regulation, due to be implemented in January 2018, prompted a review of the existing LBMA Silver Price administration arrangements

Remember that the CME and Thomson Reuters service providers to the LBMA Silver Price are their specialist benchmark units “CME Benchmark Europe Limited” and “Thomson Reuters Benchmark Services Limited”. That is what these units do, administer and calculate benchmarks. This European benchmark Regulation has been known about for a few years. Especially known about by the benchmark units of CME and Thomson Reuters. The Regulation didn’t suddenly appear out of nowhere last week, as the above statement is appearing to hint at.

And why such a brief and unclear statement from CME, Thomson Reuters and the LBMA? Is this European Benchmark Regulation just an excuse being thrown out to distract from other issues that might really be behind CME and Thomson Reuters stepping down.

Or perhaps CME and Thomson Reuters are aware of issues within the current administration of the LBMA Silver Price that would make it difficult to comply with the new legislation or that would make it too onerous to comply? But such rationale doesn’t make sense either because why are CME and Thomson Reuters not bailing out of the all the benchmarks that they are involved in? Furthermore, if the European Benchmark Regulation is a factor, why would any other benchmark service provider such as ICE Benchmark Administration (IBA) bother to pitch in the LBMA’s forthcoming tender process to find a replacement for Thomson Reuters and CME?

Perhaps CME and Thomson Reuters are worried about future reputation damage of being associated with the LBMA Silver Price due to some brewing scandal? Or perhaps the powerful bullion banks within the LBMA wanted to scupper any change that there will ever be wider participation or central clearing in any future version of the auction?

I will leave it to readers to do their own research on this and draw their own conclusions.

A Banking Cartel vs. Wider Auction Participation

One issue which has dogged the LBMA Silver Price auction since launch is that it never gained any level of “wider participation” or market representative participation. There are only 7 bullion banks authorised by the LBMA to be direct participants in the auction, and there are zero direct participants from the silver mining, silver refineries, and silver sectors.

This is despite the LBMA, CME and Thomson Reuters all misleading the global silver market on this issue on many occasions, and claiming that there would be very wide participation in the auction after it was launched. See BullionStar blog “The LBMA Silver Price – Broken Promises on Wider Participation and Central Clearing” for a huge amount of factual evidence to back up this statement, including webcasts by CME, Thomson Reuters and the LBMA, and an interview by Reuters with LBMA consultant Jonathan Spall, formerly of Barclays. Here are a few examples:

The LBMA’s Ruth Crowell was claiming back in July and August 2014 that they were interested in having 111 direct participants:

“clear demand for increased direct participation, and we had 25% of those 444 coming back saying they would be interested, and we’re still interested in having all of those participants on board”

“The advantage with centralised clearing, particularly for the pricing mechanism, is that we can really exponentially grow the amount of direct participants

Jonathan Spall, LBMA Consultant stated that:

“The hope of course is that we get many more participants in the new benchmark process….while it is likely that we will start by having banks involved it is ultimately hoped that the wider market will participate, be they refiners, miners etc.

“Ultimately – and as I said before – the intention is that there is much wider participation. So yes, refiners, miners etc.

Harriett Hunnable, then of the CME Group, stated:

“So this is really the new world, this is not the old fixing…..this is wider participation…and the London bullion market is really encouraging that…this is the new world, or the LBMA Silver Price!”.

According to the CME / LBMA / Thomson Reuters presentations, there was supposed to be a “phase 3 introduction of centralised clearing

Central counterparty clearing will enable greater direct participation in the London Silver Price

In summary, central clearing would allow direct participants to participate directly in the auction without the need for bi-lateral credit lines. However, the plan for central clearing was quietly dropped. The CME and Thomson Reuters have now had 32 months in which to introduce central clearing into the silver auction and it hasn’t happened. Nor will it now. The fact of the matter is that the LBMA banks do not want wider participation and they don’t want central clearing of auction trades either. These banks, which at the end of the day are just costly intermediaries, essentially want to monopolise the silver auction and prevent wider participation, and prevent true silver price discovery. Could it be the banks through their LBMA front that have sabotaged the contract with CME and Thomson Reuters so as to reset the contract and re-start another tender process that will ensure that no wider participation can ever see the light of day?

It’s also important to note that there is no way for miners and refiners to be direct participants in the auction. This is because the LBMA has designed the auction participant rules to keep out refiners and miners (and anyone else that is not a bullion bank). The rules are specifically designed so that only bullion banks can satisfy the LBMA’s Benchmark Participant criteria. See section 3.13 of the LBMA Silver Price auction methodology document accessible here.

Currently only 7 bullion banks are direct participants in the auction, namely HSBC, JPMorgan Chase, Bank of Nova Scotia (ScotiaMocatta), Toronto Dominion, UBS, Morgan Stanley, and China Construction Bank.  Most of these banks are very influential on the LBMA Management Committee. HSBC, Scotia and Mitsui were in the auction from Day 1 on 15 August 2014. UBS joined the auction on 26 September 2014, JP Morgan Chase Bank joined on 14 October 2014, Toronto Dominion Bank joined on 6 November 2014. Mitsui left in either late 2015 or January 2016 (the exact date is unclear). China Construction Bank only joined the auction on 6 May 2016.

Lastly, Morgan Stanley only joined the LBMA Silver Price auction on 25 October 2016 (which is just 4 months ago), at which point the LBMA / CME and Thomson Reuters had the audacity to spin that 7 LBMA bullion banks trading in a shadowy auction of unallocated silver accounts in London somehow represents the global silver market:

CME: “The addition of another member brings greater depth and diversity to the market and underlines the ongoing globalisation of the Silver Price as a leading, liquid precious metals benchmark.”

Thomson Reuters: “With the addition of Morgan Stanley to the panel, the LBMA Silver Price provides even deeper insight into the global silver market. We continue to welcome new participants to this essential mechanism for the markets.”

LBMA: “They [Morgan Stanley] add depth and liquidity to the auction and I look forward to other market participants joining in the future.”

LBMA Silver Price is NOT Representative of Silver Market

But, to reiterate (and as was stated previously in this blog), the LBMA Silver Price auction is not representative of the global Silver Market whatsoever, and it does not meet some of the simplest IOSCO benchmark requirements:

“IOSCO benchmark principles state that a benchmark should be a reliable representation of interest, i.e. that it should be representative of the market it is trying to measure. Interest is measured on metrics such as market concentration. In the Thomson Reuters methodology document (linked above), on page 11 under benchmark design principles, the authors estimate that there are 500-1000 active trading entities in the global silver market.”

The Thomson Reuters methodology document from August 2014 also admitted that “volumes in the LBMA Silver Price are a fraction of the daily volume traded in the silver futures and OTC markets”.

Why then are 7 LBMA bullion banks allowed to monopolize the representation of 500 – 1000 active trading entities from the global silver market within the auction, an auction that its worth remembering generates a silver reference price which is used as a global silver price reference and pricing source?

BullionStar investment silver bars and coins

Refiners and Miners

Based on the current rules, the vast majority of the world’s silver refiners cannot directly take part in the LBMA Silver Price auction.

Only 8 precious metals refiners are Full Members of the LBMA while 25 refiners are associates of the LBMA. Of the 8 full members, 5 of these refiners are on the LBMA refiner Referee panel, namely, Argor-heraeus, Metalor and PAMP from Switzerland, Rand Refinery from South Africa, and Tanaka Kikinzoki Kogyo from Japan. These refiners were added to the panel as LBMA Associates in 2003, and were only made Full Members in 2012. The only reason they happened to be fast-tracked as full members of the LBMA was due to their status as Referees for the LBMA good delivery list. Even the other major Swiss based refinery Valcambi is still not a full member of the LBMA.

Based on the current participant criteria of the Silver auction, where only full LBMA members could conceivably become direct participants, 25 of the refiners that are LBMA Associates cannot directly take part in the auction even if they wanted to. Candidates for Full LBMA Membership also have to jump through a number of hoops based on sponsorship by existing members, business relationships, due diligence, and involvement in the precious metals markets.

For a refiner to even become a LBMA associate, the refiner must have already attained Good Delivery Status for its silver or gold bars. There are about 80 refineries on the LBMA’s current Good Delivery List for silver. The chance of the vast majority of these refiners taking part in the LBMA silver auction is nil since not only are they not LBMA full members, they aren’t even LBMA associates.

Based on the current auction criteria, it’s without doubt literally impossible for nearly all silver producers / miners on the planet to directly participate in the LBMA Silver Price auction. Precious metal mining companies are not normally officially connected to the LBMA, and would more naturally be members of the Silver Institute or World Gold Council or another mining sector organization. So it’s confusing as to why the LBMA even mentions mining companies as possible auction participants since there are no mining companies that are Full Members of the LBMA, so they cannot be participants in the silver auction. The only mining companies that are even “Associates” of the LBMA are Anglogold Ashanti and Coeur Mining.

In 2014, Coeur Mining’s treasurer, referring to the LBMA Silver auction said:

“We hope to have the opportunity to become a direct participant down the road and look forward to working with the LBMA, CME and other silver producers to drive the evolution of this market.”

The unfortunate Coeur Mining now looks like it has been strung along by the LBMA with empty promises that it can somehow someday participate in the silver auction, but this is literally a fiction given the way the auction rules are currently set up.

Conclusion

In its announcement on 3 March, the LBMA said that it will shortly launch a tender process to appoint a replacement provider. The LBMA told Reuters News:

“We would be looking to identify a new provider in the summer, and have the new platform up and running in the autumn”

However, given the abysmal track record of the LBMA Silver Price, the question that should really be asked at this time is why is the bullion bank controlled LBMA even allowed to be in charge of such an important “Regulated Benchmark” as a global silver price benchmark, a benchmark that has far-reaching effects on global buyers and sellers of silver.

Take a brief look back at how the last tender process run by the LBMA for the London silver price was handled.

A Silver Price Seminar held by the LBMA on 19 June 2014 was not even open to the wider bullion market. As Ruth Crowell, CEO of the LBMA, told the publication MetalBulletin in an October 2014 interview:

“Not just our members, but ISDA members, and any legitimate members of the market were invited to the seminar. We also had observers from the FCA and the Bank of England. We wanted to keep [attendance] as wide-ranging as possible but to avoid anyone who perhaps would be disruptive

What is this supposed to mean? To prevent anyone attending the seminar who might have a different view on how the global silver price benchmark should be operated that doesn’t align with the view of the LBMA?

The actual process of selecting the winning bid from the shortlist of tender applicants was only open to LBMA Full members and Seminar attendees via a 2nd round voting survey. The independent consultant review that was part of the selection process, was conducted by someone, Jonathan Spall, who was not independent of the former fixings and so should not have been involved in the process.

Promises of wider participation involving refiners and miners were abandoned. Promises of central clearing of auction traded were thrown out the window. Prior to launch, the auction platform was hastily built by Thomson Reuters and CME without an adequate market-wide solution for clearing silver trades. Another of the bidders, Autilla/LME, had a working auction solution which would have allowed wider market participation at August 15 2014 go-live, but this solution was rejected by the LBMA Management Committee, LBMA Market Makers and the LBMA Data Working Group, the groups which had the ultimate say in which applicant won the tender.

There were only 3 participants in the LBMA Silver Price auction (all of them banks) when it was launched in August 2014, and two of which, HSBC and Scotia, were parties to the former London Silver Fixing. The LBMA Silver Price auction was therefore an example of same old wine in a new bottle. The same 2 banks, HSBC and Scotia are now defendants in a silver price manipulation class action suit in New York. There are now only 7 direct participants in the LBMA Silver Price. These are all bullion banks. This is 32 months after the auction has been launched. The LBMA accreditation process specifically prevents refiners and miners from joining the auction. As there are 500 – 1000 trading entities of silver globally, the LBMA Silver Price mechanism is totally unrepresentative of the silver market.

The defection of CME and Thomson Reuters now provides a one-off opportunity for the global silver market to insist that the current scandal ridden current auction be scrapped and taken out of the hands of the bullion bank controlled London Bullion Market Association (LBMA). It is also an opportunity to introduce a proper silver price auction in its place that is structured to allow direct participation by hundreds of silver trading entities such as the world’s silver refiners and miners, an auction that employs central clearing to allow this wider participation, and an auction that is based on trading real physical silver and not the paper credits representing unallocated claims that the participating London bullion banks shunt around between themselves. This could help lead to real silver price discovery in the global silver market. However, the chances of this happening with the LBMA still involved in the new tender process are nil.

Guest Post: How to Trigger a Silver Avalanche by a Pebble: “Smash(ed) it Good”

UBS and other precious metals traders on how to wreak havoc in silver markets

Written by Allan Flynn, specialist researcher in aspects of gold and silver.

avalanche

“An avalanche can be triggered by a pebble if you get the timing right” 

Earlier this year at April’s hearings for London Silver and Gold Fix lawsuits, the judge and defendant’s attorney quipped about trader chats named “the mafia” and “the bandits” published in prosecutors findings of Forex investigations but conspicuously absent from precious metals investigation findings, and the silver and gold antitrust lawsuits under consideration.

THE COURT: “Those were bad facts for the defendants.”

LACOVARA: “I think, your Honor, that if we had chat rooms that said “The Cartel”, we might be having a different focus to oral argument today.”

THE COURT: “I think that is correct.”

Given the judges skepticism of the allegations described in an earlier article, it came as a surprise early October when the banks listed were ordered by magistrate Valerie E. Caproni to face charges. More surprising perhaps was the exemption granted Swiss bank UBS, which despite having been found guilty and fined for “precious metals misconduct” by the Swiss Financial Market Supervisory Authority FINMA in November 2014, was granted motion to dismiss from both silver and gold lawsuits.

All that may be about to change according to documents filed in a New York district court December 7th, where plaintiffs claim that transcripts showing conspiracy to manipulate silver, provided by Deutsche Bank as part of an April settlement agreement, includes extensive smoking gun evidence involving UBS and other banks. Plaintiffs describe a “multi-year, well-coordinated and wide-ranging conspiracy to rig the prices of silver and silver financial instruments that far surpasses” that of the previous complaint, including potentially incriminating evidence of UBS precious metals traders allegedly conspiring with other banks.

Five additional banks to the remaining defendants HSBC and Bank of Nova Scotia are mentioned including Barclays Bank, BNP Paribas, Standard Chartered Bank, Bank of America and Merrill Lynch. The Memorandum of Law signed by Vincent Briganti on behalf of Lowey Dannenberg Cohen & Hart for plaintiffs on Wednesday 7th December seeks leave to amend the existing complaint filed with the United States District Court Southern District of New York.

Included in the memo are numerous astounding transcripts indicating coordination between UBS and other banks of “pushing,” ”smashing,” ”bending,” ”hammering,” ”blading,” ”muscling,” and “ramping” the prices of silver and silver financial instruments.

In support of claims of conspiracy to manipulate the price of silver downward the following gem is attributed to UBS Trader A: “so we both went short” “f*cking hell it just kept going higher” “63,65, then my guy falls asleep, it goes to 69 paid!” “then finally another reinforcement came in.

Discussions supposedly of coordination between UBS and their competitors about fixing the price of physical silver by offering only wide spreads between the bid and ask (where a “lac” is reference to an Indian measure equaling 100,000 units) go like this:

UBS Trader B: “what did u quote let me check”

Deutsche Bank Silver Fix Trader-Submitter A: “44/49”

UBS Trader A: “just quote wider if they call me in 1 lac I will quote 7-8 cents”

Deutsche Bank Trader B: “how wide u making 1 lac today 5 cents?”

UBS Trader A: “silver actually steadier than gold i would make 5-6 cents wide in silver”

UBS Trader A: how wide would you quote 5 lacs silver?”

Deutsche Bank Trader B: “10cu>?”

Deutsche Bank Trader B:”how wide u quote for 3 lacs?”

UBS Trader A: 10 cents”).

Manipulation of the Silver Fix price to benefit their silver trading positions in derivatives by UBS is claimed in the following exchanges:

Deutsche Bank Trader B: “u guys short some funky options” “well you told me to no one u just said you sold on fix”

UBS Trader A: “we smashed it good.”

Deutsche Bank Silver Fix Trader-Submitter A: “UBS boring the market again”…”just like them to bid it up before the fix then go in as a seller…they sell to try and push it back.”

It’s further alleged by plaintiffs that UBS implemented an “11 oclock rule” where both UBS and Deutsche Bank would short silver at 11A.M.

As examples of the comparative ease by which UBS moved the silver market the memo reveals Deutsche Bank Trader B added UBS Trader A to a chat with HSBC Trader B, which UBS Trader A deemed “the mother of all chats,” and leading to the trader’s own analysis:

UBS Trader A to Deutsche Bank Trader B: “if we are correct and do it together, we screw other people harder”

UBS Trader A: “an avalanche can be triggered by a pebble if you get the timing right” and “silver still here, u can easily manipulate silver”, and in reference to UBS supposed manipulative influence by an unnamed party: “u guys WERE THE SILVER MKT.”

hobo

UBS intended to reap financial rewards by manipulation of the price of physical silver and associated financial instruments, the memo says as UBS Trader A suggested: “go make your millions now jedi master…”pls write me a check when u aer a billionare,” and “i teach u a fun trick with silver” to which Deutsche Bank Trader B replied: “show me the money.

Confident of their ability to manipulate UBS made bold predictions according to the following alleged extracts:

UBS Trader A: “gonna bend this silver lower”; “i will bend it lower told u”; ”hah cool its gonna get ugly”; “use the blade on silver rg tnow it’ll hold it up,

Deutsche Bank Trader B: “yeah,

UBS Trader A: “gona blade silver now.

Of course all the secrecy in the world about the operations was required of the chat groups by UBS Trader A stating: “pls keep all these trick to yourself,” “btw keep it to yourself…,” and “ok rule of thumb EVERYTHING here stays here.

Examples of other banks alleged transcripts are included in the following:

Barclays

Deutsche Bank Trader B instructing Barclays trader A: “today u smash,

Barclays Trader A: “yeah” and “10k silver” “im short.

It’s alleged that Barclays and Deutsche Bank shared information so often that Barclays Trader A remarked “we are one team one dream.”

Materials in the memo even include the Deutsche Bank and Barclays precious metals traders agreeing at one stage to “stay away” from silver for a week.

The traders of course knew it was terribly wrong with Barclays Trader A responding to Deutsche Bank’s Trader B instruction to “push silver”: “HAHAHA lol i don’t think this is politically correct leh on chat.

Merrill Lynch

Allegedly fixing the bid-ask spread they offered clients on silver:

Merrill Lynch Trader A: “How wide r u on spot? Id assume 10 cents for a few lacs?

Deutsche Bank Silver Fix Trade-Submitter A: “im getting ntg but stops”

…Merrill Lynch Trader A: “we had similar” “I sweep them…Fuk these guys.

Showing disregard to global regulators even after noting their activities the two continued to “sweep” the silver market, allegedly observing at one stage: “Someone got stopped messily.

smash

BNP Paribas Fortis

Fortis Bank Trader B allegedly conspired with Deutsche Bank to manipulate silver prices, using what he termed a “bulldozer” on the silver market.

Standard Chartered

Conversations between Deutsche Bank Silver Fix Trade-Submitter A and Standard Chartered Trader A as follows:

 “Yeh” “small long out of the fix…” “ok where to sell sivler then?

23.40 thru that use it as a stop profit and let it runnnnnnnnnnnnn

were on the same wavelength

im long silver”…”ilke both [silver and gold] to get the absolute sht squeezed out of them” “im longer silver than i am gold

Conclusion

Assuming the transcripts submitted are accepted and plaintiffs are permitted to file their Third Amended Complaint, the possible pending “avalanche” of settlements in silver lawsuits will speak volumes for the investigative prowess of the CFTC and the DOJ, both of which were commissioned to investigate long running allegations of silver and precious metals market manipulation over recent years, and came up completely empty.

It appears Judge Caproni, former FBI General Counsel, was on the money when considering the potential of ineptitude in government investigations of precious metals markets at April’s gold hearing: “I don’t put a lot of stock in the fact that there are investigations because I was a government lawyer for a long time and I know what you need to open an investigation. By the same token, the fact that they closed it without charging anybody doesn’t mean that everybody is innocent. So I don’t put a lot of stock in it one way or the other.”

The CFTC proudly announced in September 2013 they had spent five years and seven thousand enforcement hours investigating complaints of manipulation in the silver market, including with assistance by the Commission’s Division of Market Oversight, the Commission’s Office of Chief Economist, and outside experts, but yet found nothing.

The Department of Justice Antitrust Division which were so confident of their investigation of collusion in precious metals they went to the extraordinary lengths in January of this year of providing a letter to silver and gold lawsuit defendants advising they had closed their investigation without findings of wrongdoing.

The Swiss Financial Services watchdog FINMA investigated, published and prosecuted UBS for forex and precious metals trading misconduct but yet said so little about precious metals findings in their November 2014 investigation report, it was impossible for the court to withstand UBS motion to dismiss in both metals.

And finally of the ability of authorities to reign in rogue banks in the precious metals or any other markets, the memorandum flags a fact that should draw the attention of those trying to figure out if they can indeed trust that their bullion bank has their best interests at heart simply by banning participation in trader chat rooms.

“The chats contained in the DB material are just the tip of the iceberg, as evidence suggests that Defendants intentionally communicated in undocumented ways to keep their manipulation hidden.”

For example the memo includes the salient reminder that banks will always find a way “to evade detection,” in this case where two traders are described as also communicating “via email and personal cell phone.”

The above article was first published at Allan Flynn’s website here.

Allan Flynn is a specialist researcher in aspects of gold and silver. He is currently investigating for future publication on the same topic and works in property and commercial architecture when he needs to eat. He holds shares in precious metals producers and banks.

 

Is there any gold bullion stored at the US Mint in Denver?

Anyone with even a passing interest in US official gold reserves will probably recall that the US Treasury claims to hold its gold (8,133.5 tonnes) over four locations in continental United States, namely at three US Mint facilities in Fort Knox (Kentucky), West Point (upstate New York), Denver (Colorado), and at the New York Fed (Manhattan, New York City).

The claimed gold holding locations and summary quantities held appear in a never-changing monthly Treasury report titled “Status Report of U.S. Government Gold Reserve”.

This report states that 4,583 tonnes of US Treasury gold are stored in the US Mint’s bullion depository in Fort Knox, 1,682 tonnes at the West Point bullion storage facility, and 1,364 tonnes in the US Mint facility in Denver, for a total of 7,628 tonnes of gold. The US Treasury further claims that 418 additional tonnes of its official gold reserves are held at the Federal Reserve Bank vault in New York. An additional 87 tonnes, a working stock figure (which never changes), comprises the balance.

While Fort Knox and the NY Fed vaults regularly take the limelight in terms of volume of media coverage, and to a lesser extent the West Point vaults do so also, there is very little if anything devoted to coverage of the US Treasury gold supposedly held in Denver. It is therefore of interest that none other than the US Mint on its own website recently ceased claiming that it stores gold bullion at its Denver facility.

On August 11, 2014, the US Mint’s Denver web page contained the following statement:

“Today, the United States Mint at Denver manufactures all denominations of circulating coins, coin dies, the Denver “D” portion of the annual uncirculated coin sets and commemorative coins authorized by the U. S. Congress. It also stores gold and silver bullion.

Denver Aug 2014
US Mint – Denver page on website August 2014

Less than a month later, on September 8, 2014, the above paragraph had been subtly changed to the following, and the words ‘gold and’ had been removed:

“Today, the United States Mint at Denver manufactures all denominations of circulating coins, coin dies, the Denver “D” portion of the annual uncirculated coin sets and commemorative coins authorized by the U. S. Congress. It also stores silver bullion.

Denver Sept 2014
US Mint – Denver page on website September 2014

The amended wording remains on the Mint’s present day Denver web page i.e. with just the “It also stores silver bullion” sentence.

Denver Aug 2016
US Mint – Denver page on website August 2016

At the very least this change in wording between August and September 2014 is very unusual. Why would the Mint have authorized and made such a wording change and deleted the reference to gold bullion? I asked the US Mint to clarify but the query went unanswered:

Given that the Denver Mint does not produce any gold or silver coins, the Mint does not have a need to store either gold or silver bullion working stock in Denver, so the above wording cannot be referring to metal being stored for fabrication supplies. The only commemorative coin produced in Denver is an uncirculated clad half dollar made of copper and nickel.  While the above change of wording on the US Mint’s website could have an entirely different explanation, it does raise the possibility that there isn’t any US Treasury gold bullion stored in Denver. This possibility would also subscribe to a view that has been expressed for quite some time now by well-known gold author and commentator James Rickards. Since at least 2010, and probably prior to that, Rickards seems to think that the US gold reserves are nearly exclusively stored at West Point and Fort Knox. Some tweets of his illustrate the point:   

This view, that the US gold is kept at West Point and Fort Knox, actually makes quite a lot of practical sense and is entirely logical. It also makes Denver look like the odd man out.

The US Mint facilities at Fort Knox and West Point are located adjacent to US military installations, namely the US Army base, Fort Knox, and the US Military Academy, West Point. The Fort Knox bullion depository, which opened in 1936, was actually built on land that was previously part of the Fort Knox military base, and that had been deeded to the Treasury Department. The West Point bullion facility, which opened two years later in 1938, was built on land formerly occupied by the West Point military facility, and that had also been deeded to the Treasury Department.

Having large quantities of gold stored in facilities next door to US military facilities is a natural security advantage for protection and also as a deterrent against any would be gold heists. In contrast, the US Mint facility in Denver is located on a city block at 320 West Colfax Avenue, between Delaware St and Cherokee St. It’s near a court-house and a police station but no sign of any US military facilities in the immediate vicinity.

Denver 3D
US Mint Denver facility
Denver 3D2
US Mint Denver rear view

The US Mint in Denver is also the odd one out (of the three) in that it offers public tours of the facility, something unheard of at Fort Knox and West Point. Arguably, the NYFed offers a gold vault tour, but out of US Mint facilities that the Treasury claims to store gold at, Denver is the only one with a public tour. The supposed location of the gold vaults in Denver is also a complete mystery with no photos or images of any vaults or contents of vaults (as far as I can see) ever on the web. A review of the Denver Mint tour (here) mentions supposed gold storage in the lower decks of the building but this seems to be merely supposition as it is inferring that 3 gold bars on display in Denver came from the building’s vaults. However, they did not. These three gold bars actually came from West Point, as CoinWeek stated in May 2012:

“Denver Mint plant manager David Croft pointed out that the three bars were shipped in from the U.S. Mint’s working gold supply at West Point and did not come from the gold that is in deep storage in Denver.

Which begs the question, why? The US Mint would probably answer, so as not to ‘break the seals’ on the Denver vault doors, but this shipping of 3 gold bars from upstate New York to Denver would seem completely unnecessary if Denver was storing  a couple of tonnes of gold, let alone 1,362 tonnes. The more accurate answer may be that the US gold, if it even exists to the extent to which the US Treasury claims, is held adjacent to US military bases at West Point and Fort Knox. 

Moving the goalposts….The LBMA’s shifting stance on gold refinery production statistics

On Friday 31st July 2015, I released an article discussing the sale of Swiss gold refiner Valcambi to Indian jewellery company Rajesh Exports. In my report, in a section about Valcambi’s annual gold refining capacity, I made passing reference to 2013 gold refining production statistics that had been published by the London Bullion Market Association (LBMA) on 1st May 2015. These same gold refinery production statistics had also been quoted by the LBMA as recently as July 2015 in the news section of Issue 78 of its ‘Alchemist’ magazine, (published on 21st July 2015, just a week before my article).

The reference in my article to 2013 LBMA gold refiner production statistics discussed the unprecedented 6,601 tonnes of gold that was refined in 2013 by gold refineries on the LBMA’s Good Delivery List. My reference to this 6,601 tonnes on 31st July, including a short table of LBMA data, was as follows:

“Rajesh Exports just revealed in its press release that over the last 3 years, Valcambi has refined an annual average of 945 tonnes of gold and 325 tonnes of silver (2835 tonnes of gold and 975 tonnes of silver over 3 years). Presumably the last 3 years that Rajesh mentions refers to the last 3 calendar years of 2012-2014.

The London Bullion Market Association (LBMA) doesn’t reveal annual production data of its refinery members on an individual level, however, the LBMA recently published high level totals of the refined gold production of its accredited refiners (LBMA Good Delivery List) over the years 2006 to 2013. What was striking about the data was that total refined gold production of its refinery members reached 6,601 tonnes in 2013, which was 42% higher than total refined gold production in 2012, and also more than double global mine production of 3,016 tonnes of gold in 2013. See table below from LBMA publication:

Total annual refined gold and silver production by LBMA refiners 2006-2013 (tonnes)

Refinery output 2006-2013

So with Valcambi being the largest gold refinery in the world, it would be realistic to suggest that its annual average of 945 tonnes of refined gold output over the last 3 years probably hides the higher refined gold production that it too experienced in 2013 versus 2012.”

In the first quoted paragraph, above the table, I had hyperlinked the word ‘publication’ to a LBMA source document URL which pointed to a pdf document named ‘LBMA Brochure Final 20150501.pdf’.

The ‘LBMA Brochure Final 20150501.pdf’ file was a 4 page document titled “A guide to The London Bullion Market Association”, with the refinery production statistics appearing on page 3 under a page title “The LBMA Good Delivery List”. The file ‘LBMA Brochure Final 20120501.pdf’ was created on 2015-05-01 at 10:09:50 using the applications Adobe InDesign CS 5.5 (7.5) and Abode PDF Library 9.9.

In the refinery section of the LBMA Brochure Final 20120501.pdf’ document, the LBMA’s commentary first explained what the Good Delivery ‘List’ refers to, as well as listing the number of gold and silver refineries on the List, and then proceeded to comment on the ‘Total refined gold production of the refiners on the Listin 2013, which it stated was 6,601 tonnes. The LBMA commentary also highlighted that this 6,601 tonnes of refined gold production by the refiners on the List was ‘more than double‘ 2013 world mine production of 3,061 tonnes.

The ‘List’ specified 72 refineries which refined gold, and 83 refineries which refined silver. It also showed that 16 refineries which refined gold were in Europe, 43 in Asia, 11 in the Americas, and 1 each in Africa and Oceania. So the 6,601 tonnes of gold statistic for 2013 represented 72 refineries on the Good Delivery List which refined gold. And the LBMA made clear in its commentary that refiners on the Good Delivery List represent 85%-90% of world gold production:

original - GDL List

From Page 3 of 'LBMA Brochure Final 20120501.pdf'
From Page 3 of ‘LBMA Brochure Final 20120501.pdf’

As mentioned above, the LBMA also printed the same 2013 gold refining figure of 6,601 tonnes in Issue 78 of its magazine, ‘Alchemist’, which was published on 21st July 2015. Alchemist is published in both hard copy magazine format and on-line. In the ‘LBMA News’ section of Issue 78, viewable here and here(Alch78LBMANews), the LBMA Chief Executive, Ruth Crowell, provided a news update on the Association’s Physical Committee, stating:

Total refined gold production represented by the accredited refiners on the LBMA’s Good Delivery List was 6,601 tonnes in 2013, more than double mine production of 3,061 tonnes. For silver, refined production by listed refiners was 24,570 tonnes, marginally below the 25,494 tonnes of mine production in the same year.”

 [The full issue of Issue 78 of The Alchemist can be viewed here (large file)]

According to the LBMA, the ‘Physical Committee is made up of industry experts from the physical bullion market“, therefore this physical committee is well aware of the 6,601 tonnes of gold refinery production figure in 2013, not least because it’s printed in the committee’s news section in the latest edition of the Alchemist.

The explosion in gold refining activity in 2013, and the huge throughput of Good Delivery bars being transformed into smaller higher fineness bars for the Asian gold market was without doubt one of the biggest stories in the gold world during 2013. I had cited the 6,601 tonnes figure to help support a calculation about Valcambi refining capacity, and my reference wasn’t really central to the main topic of my Valcambi article. But it was a topic that I was planning to re-visit, and I tweeted about it on 4th June when I first read the LBMA report that contained the 6,601 tonnes data:

 

All of the above seems logical and easy to understand. It was therefore surprising to notice that on Wednesday 5th August 2015, three business days after my Valcambi articles was published, the LBMA substantially amended the gold refinery figures in the file ‘LBMA Brochure Final 20150501.pdf‘, and dramatically lowered the 2013 refined gold production figure from 6,601 tonnes to 4,600 tonnes, while substantially altering the wording and meaning of the paragraph commenting on the refined tonnage. The document content was amended and re-saved with the same file name LBMA Brochure Final 20150501.pdf‘, and left in the same web directory. So anyone viewing the LBMA document for the first time would not know that the gold refining figures in the report had been altered and substantially reduced. The file directory in question is here, and contains the altered report:

http://www.lbma.org.uk/assets/market/gdl/LBMA%20Brochure%20Final%2020150501.pdf

(The ‘%20’ instances are just space delimiters within the URL)

Luckily, the original version of ‘LBMA Brochure Final 20150501.pdf‘ from 1st May 2015 can be viewed here -> LBMA Brochure Final 20150501.

Let’s look at what was changed between the two versions. Here is the exact updated LBMA text and data table after the Wednesday 5th August changes, including the matrix displaying the number of gold and silver refineries on the ‘List’. The number of refineries remained unchanged. However, notice the 2013 gold refining figure became 4,600 tonnes:

GDL List in updated version - no change

Page 3 of the 5th August changed version of 'LBMA Brochure Final 20120501.pdf'
Page 3: changed version of ‘LBMA Brochure Final 20120501.pdf’ 5th Aug

If you compare the original and altered versions of this LBMA report, you will see substantial differences. Here is a description of the changes, which I have highlighted using italics, underline and bold in various places, and the LBMA’s text is indented:

a) For gold, the LBMA reduced the 2013 total refinery production figure from 6,601 tonnes to 4,600 tonnes, a reduction of 2,000 tonnes of gold. To put the sheer magnitude of 2,000 tonnes of gold into perspective, 2,000 tonnes of gold is nearly twice as much gold as the Swiss National Bank (SNB) officially reports that it holds. [The SNB claims to have 1,040 tonnes of gold].

The LBMA added that words ‘estimated to be‘ in front of the 4,600 tonnes figure, and the words ‘owing to recycling of scrap material‘ were added after the figure. The ‘more than double‘ reference to the 6,601 tonnes of gold being more than double world mine production, was deleted and replaced by the word ‘above‘. The words ‘source Thomson Reuters GFMS‘ were added in brackets at the end of the sentence. The wording of “total refined gold production by the refiners on the List was retained and not altered.

“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).”

b) For silver, the 2013 total refinery production figure of 24,570 tonnes was increased to 29,984 tonnes, an increase of 5,500 tonnes. The words ‘estimated to be‘ were also added in front of the 29,984 tonnes figure. Unlike gold, no wording was added about recycling of scrap material. Since the LBMA upped the 2013 silver total so much, it was now far above mine production, so the previous words ‘marginally below‘ were replaced by the word ‘above‘.  Again, the words ‘source: Thomson Reuters GFMS‘ were added in brackets at the end of the sentence.

For silver[,] refined production by listed refiners in 2013 was estimated to be 29,984 tonnes, above the 25,981 tonnes of mine production in 2013 (source: Thomson Reuters GFMS).

c) The altered text still retained all of the references to the Good delivery refiner ‘List’, and still stated that the figures in the table were for ‘estimated annual refined gold and silver production by the refiners on the List’.

“The Gold refined by refiners on the List make up about 85-90% of world production. Total estimated annual refined gold and silver production by the refiners on the List are shown in the table below (tonnes).”

 d) The years 2006 and 2007 were removed entirely from the table in the changed version from 5th August, with the revised version only covering the years 2008 – 2013 and not 2006 – 2013 as per the original.

As the number of gold refiners in the ‘List’ above remained the same in this altered version as in the original version, there can be no doubt that this refers to the same group of gold refiners which had combined production output of 6,601 tonnes of gold in 2013 yet also, simultaneously (and impossibly) according to this altered version of the report, had a combined 4,600 tonnes of gold production output in 2013.

Total refined gold production of the refiners on the List

Question: How does the LBMA know that “Total refined gold production of the refiners on the List” was 6,601 tonnes in 2013?

Answer: Because the Good Delivery refiners provide annual refinery production figures to the LBMA. It’s as simple as that.

Every refiner on the LBMA’s Good Delivery list is required to provide production data to the LBMA on an annual basis. This information is required by the LBMA as part of its obligatory Pro-Active Monitoring (PAM) programme of Good Delivery refiners. The PAM programme is defined by the LBMA as follows:

“The PAM programme reviews the assaying competence of refiners on a three-yearly basis. In addition, it checks that they continue to meet the minimum requirements for refined production and tangible net worth on an annual basis.”

This production data was supplied to the LBMA on a three-yearly basis until 2011, but the rules were changed in 2011 to an annual basis. From ‘Alchemist’ issue 65, December 2011:

“Some important changes in the Rules have been agreed recently. The first is that refiners will have to provide data on their tangible net worth and production on an annual, rather than a three-yearly, basis.”

In the LBMA’s most recent Good Delivery Rules from March 2015, ‘Section 10 Pro-Active Monitoring’ states that:

All current Good Delivery refiners are also required to submit their production and audited financial data on an annual basis to the Executive. “

Annex A of the same document, clarifies the compliance date and states:

“With effect from 1st January, 2012, all current Good Delivery refiners are required to submit their refined production and audited financial data on an annual basis to the Executive.”

Additionally, refiners applying to be accepted on the LBMA’s Good Delivery list need to submit a three year operating history with three years of production figures as part of the application. Annex A (Application Form), addressing what to include with an application, states that required documents include:

  • Figures for the last three years’ annual production of refined * gold/silver in tonnes.
  • Estimate of next two years’ annual production of refined * gold/silver in tonnes.

The asterick (*) states that ‘refined’ refers to “metal which has gone through a refining process, such as electrolysis, Miller Process or Aqua Regia refining“. These processes would apply to Good Delivery bars that were being converted into 9999 fineness kilobars for the Asian gold market.

Therefore, the LBMA knows exactly, down to the exact tonne, the figure for “total refined gold production of the refiners on the List” in 2013.

In issue 74 of the LBMA’s ‘Alchemist’ published in June 2014, when the LBMA’s Physical Committee was providing a news update on ‘Pro-active Monitoring’, and reviewing the 2011 refinery production statistics which had just been finalised at that time, the Committee highlighted the following:

A number of issues arising from proactive monitoring of refiners on the list have also been discussed….Two very interesting numbers arising from this work are the figures for total refined production represented by the accredited refiners. Although it takes time to complete this data collection, the figures for 2011 are now complete. The total for gold is 4,695.8 tonnes and for silver is 28,395.5 tonnes, in both cases significantly above the respective world mine production of 2,838.1 tonnes and 23,545 tonnes.

It appeared that the writer of that paragraph thought that the two refining numbers were interesting enough to be comment worthy because the numbers were ‘significantly above‘ the world mine production figures.

The LBMA also administers a ‘Responsible Gold Audit Programme’ for gold refiners on its Good Delivery List. The audit seeks to determine whether a refiner complies with the LBMA’s ‘Responsible Gold Guidance’. The actual audits are carried out by independent auditors that have been approved by the LBMA, but the audit results are passed back to the LBMA. For example, in February 2014, the LBMA issued a press release announcing  that 4 refiners had successfully passed the audit. The announcement mentioned that:

the audit reviewed the refiners’ production over a 12 month period. The LBMA received a large volume of reports in late 2013, and will continue to report in the coming weeks as each batch is reviewed.”

Therefore, the audits are another way in which the LBMA keeps track of the refiners production, in addition to the reporting coming in from the Pro-active Monitoring programme. Either way, the LBMA knows the refinery production statistics of the Good Delivery refiners and does not need to get estimates from GFMS or any other body.

kilobars

Thomson Reuters GFMS

Given the above, then why the sudden need by the LBMA on 5th August 2015 to include a reference to  “source Thomson Reuters GFMS“? By including the reference to “owing to recycling of scrap material”, it is clear that the intention was to solely relate the 4,600 tonnes of gold quoted to just two sources, namely, gold mining and gold scrap recycling. Furthermore, why had the figure suddenly become an ‘estimate’ and who was responsible for the estimate? There is no need for estimates of refinery production when every refinery on the Good Delivery ‘Lists’ provides the exact real production figures to the LBMA.

Additionally, what was the reason for suddenly throwing a perfectly logical paragraph out the window which had referred to gold refinery production statistics for 2013 collected by the LBMA, and replace it with an estimate about gold mining and scrap recycling from a company, GFMS, which does not specialise in collecting gold refinery production statistics?

What is GFMS?

GFMS was a metals analysis consultancy firm, that was acquired by Thomson Reuters in August 2011. GFMS was formerly known as Gold Fields Mineral Services. The group within Thomson Reuters is now known as Thomson Reuters GFMS. GFMS gathers supply and demand figures for gold and other precious metals, and publishes an annual gold survey and related update reports.

GFMS’s supply data for gold mine production and gold scrap is not the same metric as gold refinery production output, and is not even close to being the same metric, especially in 2013 when there were huge amounts of Good Delivery gold bars re-smelted and re-cast into smaller gold bars in Switzerland and other places for onward shipment to Asia.

‘LBMA Overview Brochure.pdf’

The LBMA also makes reference to annual gold and silver refinery figures in another document on its website, in a file named ‘LBMA Overview Brochure.pdf‘. This document is located in a ‘presspack’ directory, presumably for use by the LBMA’s Fleet Street press contacts. This document has, in its various iterations, included a paragraph with identical phraseology  about refinery production statistics i.e. ‘Total refined gold production by the refiners on the List‘, and has also included a similar table of gold and silver production statistics of LBMA accredited refineries.

On Saturday 1st August, the version of this other file, the ‘LBMA Overview Brochure.pdf‘ document on the LBMA web site, had also been altered with some very strange temporary alterations inserted for 2013 gold and silver refinery production statistics. All of the annual refinery figures in the entire table had been blanked out of the table with the shorthand ‘n/a‘ substituted in each row. The text had also been changed, and 4,848 tonnes had been inserted as the gold refineries’ production figure for 2013, and 30,934 tonnes for silver, with the word ‘above‘ added before world mine production for both metals. The overwritten figures and text appeared in a  slightly scrawled text font (see below). GFMS was not mentioned in this file. The file, on 1st August, in its web directory (http://www.lbma.org.uk/assets/downloads/presspack/) rendered in a web browser as follows, when this image was recorded:

31st July 4848

 Why 4,848 tonnes?

So, where did this 4,848 tonnes figure for gold, and 30,934 tonnes for silver come from? These numbers are another entirely different set of figures for 2013, a third set if you will. To answer where these numbers came from, we need to turn to a presentation given by Stewart Murray, former LBMA CEO, at the LBMA’s Assaying and Refining Conference held in London between 8th – 10th March 2015. In a presentation titled ” The LBMA Good Delivery List, Recent and Future Changes“, on 9th March, Murray utilised slides which, on page 9 showed the following:

Murray slide 9 assaying and refining pres refinery prod stats

Notice, that for 2013, the figures are 4,848 tonnes for gold, and 30,934 tonnes for silver. This dataset also only goes from 2008 to 2013.

GFMS also makes another appearance in this slide, with a GFMS combined mine production and recycled scrap figure for 2013 being quoted as 4,302 tonnes for gold, and 31,460 tonnes for silver, respectively.

The next slide in that presentation, slide 10, even gives a regional breakdown of the 4,848 tonnes and 30,934 tonnes figures, attributing 1,790 tonnes of gold refining to Europe in 2013. Keep these figures in mind as we go through this maze of numbers.

Murray slide 10 regional refinery breakdown

Slide 6 of the same presentation showed a line graph of the Good Delivery gold refiners that were referred to in the production figures in slide 9. You can see that the numbers of refiners in each line as at 2014 equate to the numbers of gold refiners in the ‘List’ of the original ‘LBMA Brochure Final 20150501.pdf‘ file, i.e. 16 gold refiners in Europe, 43 in Asia, 11 in the Americas, 1 in Australia (which was Oceania in the List – the Perth Mint), and 1 in Africa (Rand Refinery). So again, there can be no doubt that they are the same refiners being referred to here that had a production output of 6,601 tonnes of gold in 2013, and at the same time 4,848 tonnes. So the same refiners have been at work in 3 parallel universes during 2013, or so it may seem.

GD gold refiners graph murray

By Wednesday 5th August, the ‘LBMA Overview Brochure.pdf‘ file had also been updated and re-saved, and contained the exact same commentary text and the exact same table of refinery production output figures as the altered ‘LBMA Brochure Final 20150501.pdf‘, i.e. the 4,848 tonnes figure was gone and was replaced by 4,600 tonnes, and the file was re-saved by the LBMA with the same file name, and left in the same file directory that it had been in, i.e.

 

Again, a first time viewer would not know by looking at the report that the gold and silver refinery production figures had been altered and the text edited.

What do the document properties of the re-saved ‘LBMA Brochure Final 20150501.pdf‘ and ‘LBMA Overview Brochure.pdf files, saved on Wednesday 5th August tell us?

LBMA Brochure Final 20150501.pdf‘ was saved at 15:49:48 on 5th August by author name Aelred Connelly. Then 29 seconds later at 15:50:17 on 5th august,  file LBMA Brochure Final 20150501.pdf‘ was also saved by author name Aelred Connelly. Aelred Connelly is the LBMA’s Public Relations Officer, ex Bank of England for more than 25 years, where he was a gold bullion analyst and a relationship manager for the Bank’s central bank and government customers.

brochure final 20150501 connelly 5th aug

overview brochure connelly 5th Aug

So, what is going on here?

Could it be that the LBMA’s original figure of 6,601 tonnes of refinery gold production in 2013 should not have been published for some reason, and needed to be quickly changed, for example, that the publication of this metric breached refiner confidentiality, or that it just made the GFMS supply numbers look way out of line with reality?

Previous LBMA documents discussing refined gold production

There are a number of other slightly older LBMA reports, brochures and other documents which discussed and recorded Good Delivery refinery annual production statistics. The interesting aspect of these other files, apart from the numbers, is that the syntax and wording is identical to the version from 1st May 2015 which I had quoted and which disappeared by 5th August. Furthermore, none of the older versions match (in style) the new versions that use ‘estimates’ and that refer to Thomson Reuters GFMS.

The previous syntax also seemed totally adequate for use by regulatory agencies such as the US SEC, and the UK Treasury’s Fair and Efficient Markets Review.

A file here refers to 2009 refinery figures. The same statistics were quoted in version created on 19th April 2012, for use in a LBMA meeting with the US Securities Exchange Commission (SEC) on 18th April 2012:

“Total refined gold production by the refiners on the List was more than 4,000 tonnes in 2009, well above world mine production of 2,611 tonnes. For silver, refined production by listed refiners of 22,800 tonnes was marginally greater than the 22,342 tonnes of mine production in the same year.”

Then there is another version that was saved as 23rd May 2014 but refers to 2011. It was also used in January 2015, in a letter from the LBMA to the Fair & Effective Markets Review, Bank of England:

“Total refined gold production by the refiners on the List was 4,695.8 tonnes in 2011, well above world mine production of 2,838.1. For silver, refined production by listed refiners of 28,395.5 tonnes was greater than the 23,545 tonnes of mine production in the same year.”

Another newer version  on 12th August 2015

There is also an even newer version of a file specifying “total refined gold production by the refiners on the List” now uploaded on the LBMA web site. This latest document, called “A guide to The London Bullion Market Association August 2015“,  is from 12th August 2015.

“Total refined gold production by the refiners on the List was estimated to be 4,579 tonnes in 2013, owing to recycling of scrap material, above world mine production of 3,061 tonnes (source: Thomson Reuters GFMS). For silver, refined production by listed refiners in 2013 was estimated to be 28,013 tonnes, above the 25,981 tonnes of mine production in 2013 (source: Thomson Reuters GFMS). The Gold refined by refiners on the List make up about 85-90% of world production. Total estimated annual refined gold and silver production by the refiners on the List are shown in the table below (tonnes).”

In this version, a refinery in China (Daye Nonferrous Metals Company) was accredited to the Good Delivery List for gold in June 2015, and so it was moved from the silver only category to the gold and silver category on the List. Why the 2013 gold production figure was then reduced again from 4600 tonnes to 4579 tonnes is unclear,  and even more mysterious is why the 2013 silver production figure became 28,103 tonnes,when in the two report versions from 5th August LBMA , the 2013 silver production total had been 29,984 tonnes. That’s a reduction of 1,871 tonnes of silver between 2 LBMA reports that were published a week apart.

Table: Comparisons – LBMA refinery production (1st May vs 5th August vs 9th March)

lbma data comparison table

It is not just 2013 where the refinery production statistics deviate significantly for both gold and silver. For gold, the altered figures were applied to 2012, 2011 and 2010 also. For 2009 and 2008, the revised data is actually higher for gold than the 1st May 2015 published version. The differences in 2010, 2011, and 2012, and indeed, 2008 and 2009 require explanations also.

For silver, the altered figure for 2013 is, as mentioned earlier, more than 5000 tonnes higher in the newer version. This article has focused on gold. I have not looked at the silver angle. Other people may wish to explore the silver angle.

The figures in the newer LBMA documents of 5th August are very close to the figures used by Stewart Murray in his 9th March presentation, except for 2013 in gold and silver, and in silver in 2012. There is still however, a 248 tonne difference between the 4,848 tonnes 2013 gold production figure quoted by Murray on 9th March, and the lower 2013 gold figure of 4,600 tonnes added into the LBMA documents on 5th March.

Conclusion

There are 2,300 tonnes of 2013 gold refining output in excess of combined mine production and scrap recycling being signalled within the  6,601 tonnes figure that was removed from the LBMA’s reports on 5th August 2015.

Could it be that this 6,601 tonne figure included refinery throughput for the huge number of London Good Delivery gold bars extracted from gold ETFs and LBMA and Bank of England vaults and converted into smaller gold bars in 2013, mainly using LBMA Good Delivery Swiss gold refineries? And that maybe this 6,601 tonne figure stood out as a statistical outlier for 2013 which no one wanted to talk about?

The objectives of HM Treasury’s Fair and Efficient Markets Review (FEMR) include transparency and openness. It would appear that altering already published gold refinery statistics, especially for 2013, seems not to be in the spirit of these FEMR objectives.

Part 2 of this analysis of the LBMA’s 2013 gold refinery statistics looks behind the 6,601 tonne number at the phenomenon of Good Delivery bars being processed through the Swiss gold refineries in 2013, the gold withdrawals from the London-based gold ETFs, and the huge shipments of gold from the UK to Switzerland in 2013. Part 2 also examines the 2013 withdrawal of gold from the Bank of England, and how GFMS and the World Gold Council tried to, or tried not to, explain the non-stop processing of Good Delivery gold bars into smaller finer kilobars during 2013.