The extinction of gold derivatives

Aug 5, 2021·Alasdair Macleod
This month is the fiftieth anniversary of the Nixon shock, when the Bretton Woods agreement was suspended. And the expansion of commercial banking into credit for purely financial activities became central to the promotion of the dollar as the international replacement for gold.

With the introduction of Basel 3, commercial banking enters a new era of diminishing involvement in derivatives. The nominal value of all derivatives at the end of last year amounted to seven times world GDP. While we can obsess about the effects on precious metals markets, they are just a very small part of the big Basel 3 picture.

However, gold remains central to global money and credit and the impact on gold markets should concern us all. In this article I quantify gold forwards and futures derivatives to estimate the impact of reversing anti-gold policies that date back to the Nixon shock in 1971.

We are considering nothing less than the effects of ending fifty years of gold price suppression. Through leases, swaps, and loans central banks have fed physical bullion into derivative markets from time to time to keep prices from rising and breaking the banks who are always short of synthetic gold to their customers.

To summarise, bullion banks withdrawing from derivative markets is bound to create replacement demand for physical gold that can only drive up the price and further undermine fragile confidence in fiat currencies at a time of rapidly increasing monetary inflation.


The introduction of the net stable funding ratio (NSFR) as a central feature of Basel 3 regulations will have a major impact on derivative markets. For the purposes of this article, we are interested in how it will affect the exchange value for gold.

Derivatives break down into two broad categories. There are derivatives traded on regulated exchanges, for which there are publicly available data, principally futures contracts and options on futures. But these are the tip of an iceberg that consists of over-the-counter derivatives, multiples larger in outstanding obligations. They consist of forwards, swaps, loans, leases, and options, for which collective data is scarce.

Officially, the purpose of derivatives is to hedge risk. And since we turn to banks to finance our activities either by drawing down on our deposits or obtaining bank credit, they are the usual originators of derivatives, expanding their quantity as the demand for underlying assets, such as gold, increases. Consequently, they have become primarily a source of paper equivalents, because banks rarely deal in physical commodities.

While risk management was the original function, banks have turned trading in derivatives to lucrative profit centres. The banks have evolved products that permit speculators and investors to acquire exposure without having to access underlying products. It is a sophisticated version of betting, whereby you can buy and sell paper from a computer terminal without having to touch the referenced asset. And when you have markets populated by punters fuelled by growing quantities of paper currencies which in turn fuel financial asset values, it is natural that banks increasingly operate as their bookies.

For this and other reasons in recent decades the world of OTC derivatives has exploded in size, as banks have diversified from expanding credit for manufacturing and non-financial service businesses into expanding credit for purely financial activities. Consequently, at the end of last year, according to the Bank for International Settlements notional amounts of paper derivatives outstanding were $582 trillion with a gross value of $15.8 trillion.[i] That represents gearing between notional amounts and their value of over 36 times, and nearly seven times the World Bank’s estimate of global GDP.

As the chart from The Bank for International Settlements above shows, total derivatives expanded rapidly ahead of the Lehman crisis in August 2008. After a brief wobble they continued expanding into 2014 before declining into 2016, since when the uptrend has been gently rising.

Following the Lehman crisis and while admitting the usefulness of derivatives for the purpose of risk management, as part of their overhaul of banking regulations the BIS would have been concerned at the systemic risks to commercial banks from increased position taking. This has led to two new definitions: high quality liquid assets, which can be readily realised in a crisis; and the net stable funding ratio, which ensures that a bank’s assets are suitably funded by its liabilities.

The Basel Committee on Bank Regulation finalised its NSFR rules in October 2014, coinciding with peak derivatives in the BIS chart above. The publication of these future regulations might have been one reason behind the subsequent 2014-16 decline. But another factor was the introduction of written bilateral agreements between counterparties netting off common derivatives into one position. This has the effect of reducing apparent outstanding derivatives, which had hit record levels in 2008 before netting agreements were in place.

In the NSFR equation[ii], derivative liabilities net of matching derivative assets, if their liabilities exceed their asset values have an available stable funding (ASF) of zero. In other words, unlike more stable categories of balance sheet liability a bank cannot use them to fund balance sheet assets. And if derivative assets exceed associated derivative liabilities, they require an ASF of 100% to be applied as required stable funding; in other words, they must be funded totally by liabilities that qualify as available stable funding.

At the banks’ treasury level, net long and short derivative positions are an inefficient use of the balance sheet by curtailing more efficient uses. The same applies to uneven positions in equities and commodities, though different ASFs and RSFs may apply. Furthermore, note that no distinction is made between regulated and OTC derivatives. The overall effect is likely to stem and reverse the tide of bank credit expansion into purely financial activities. And given that banks have already reduced their lending to non-financial activities relative to their total lending, by hampering further expansion into financial activities Basel 3 appears to mark the peak of commercial banking.

It is against this background that we approach our examination of gold derivatives. At $834bn, gold OTC derivatives are too small to register on the BIS chart above. This article drills down into what is essentially a minor element of the Basel 3 revolution, making bullion banking the subject of far larger derivative issues.

Gold derivatives

There are two classes of gold derivative commonly dealt with, forwards and futures as well as options in both categories. The legal difference is that forward contracts are bilateral bespoke agreements, while future contracts are standardised and traded on registered exchanges.

Essentially, they serve two different markets. Futures are classified as regulated investments while forward contracts are not. As regulated investments, investing institutions have limitless access to futures contracts, whereas their access to unregulated OTC forwards is strictly limited, if permitted at all. Therefore, the speculating traders in Comex futures, for example, can be anyone. Traders in forward contracts in the London market are predominantly acting as principals not requiring regulation, which therefore excludes nearly all collective investment schemes and investment managers. These principals include banks, family offices and their ultra-rich principals, privately-owned corporations, sovereign wealth funds and central banks.

We shall start by examining the OTC market, whose forwards and swaps are predominantly dealt in by the members of the London Bullion Market Association (LBMA) for settlement either in London (loco London) or in Switzerland (loco Zurich).[iii]

Figure 2 shows that since gold bottomed against the dollar in December 2015 outstanding forwards and swaps had more than doubled from $351bn to $834bn and OTC options had tripled from $101bn to $304bn by December 2020. And while the multiyear shifts appear to be greater in percentage terms than for the whole OTC universe, the expansion of outstanding obligations before the Lehman crisis shared with the general trend, and the expansion from late 2015 does as well.

According to Dr Fergal O’Connor writing in Issue 99 of The Alchemist, about 60% of daily settlement volume in gold in London’s bullion market is for spot settlement, while swaps and forwards accounted for about 30% of the total. Bear in mind that these figures cannot be compared with those of the BIS derivatives, which are of outstanding positions, including those of non-LBMA members. O’Connor’s figures appear to be of settlement between LBMA members only and exclude trades with non-members.[iv] This point was brought out in research by Paul Mylchreest for Hardman & Co, where he points out that turnover volumes reported in 2011 were approximately five times higher than in 2018.[v] However, it seems likely that O’Connor’s figures, which is of LBMA members’ transactions only and provided by the LBMA itself, formed the basis of the LBMA’s assumption in evidence to the PRA consultation earlier this year that unallocated gold is spot physical by another name.

With LBMA customer transactions now excluded from the LBMA’s turnover figures, we can therefore ignore the apparent mismatch between the proportions that are shown as spot transactions and forward trades which seems to suggest that physical settlements dominate the market. And we know from the BIS figures that at the end of last year swaps and forwards totalling $530bn existed. In practice, forwards will be reflected on bank balance sheets as liabilities in the form of unallocated gold accounts (where in normal banking parlance there would be customer deposits) while swaps are almost certainly off-balance sheet leading to on-balance sheet transactions. We do not know the split between swaps and forwards, but that does not matter.

We can get a feel for the use of gold derivatives by looking at the BIS’s OTC Derivatives by Maturity tables (Table D9).[vi] Of the $834bn gold derivatives outstanding, $757bn matured in one year or less, $62bn between one and five years, and $20bn in five years or more. The remaining $5bn error may be due to rounding or different data reporting methods.

Under the new British regulatory interpretation of the Basel 3 NSFR, either customer unallocated gold accounts will end up being closed, or alternatively banks will have to acquire unencumbered physical gold to back them to avoid financing disadvantages for their balance sheets. The alternative of banks imposing charges on unallocated accounts to offset the regulatory burden seems unlikely to happen, because bank customers are likely to demand off-balance sheet custodial ownership instead, costing as little as 8 basis points for the largest accounts. We must therefore conclude that as the bullion banks rearrange their affairs ahead of the year-end, demand for physical gold among LBMA customers is likely to increase substantially, as precious metals business movers towards custody arrangements.

Availability of physical gold in London

At end-December 2020, $530bn —the total of forwards and swaps mostly between members of the LBMA — was the equivalent of 8,676 tonnes of gold. Total holdings of vaulted gold reported by the LBMA at end-June was 9,587 tonnes, which at first sight and as promoted by the LBMA’s reporting might be taken for the market’s underlying liquidity. But more than half of it is the Bank of England’s 5,756 tonnes, which is not a LBMA member and nearly all its gold is earmarked for central banks. The true figure for LBMA member vaults is therefore 3,831 tonnes. And of that, approximately 1,500 tonnes are custodial gold for ETFs, leaving 2,331 tonnes. This balance is held between allocated accounts on behalf of large investors around the world and as backing for unallocated accounts on bank balance sheets including the four owners of the LBMA’s settlement system. The LBMA and vault operators do not provide breakdowns of these categories but given that allocated bullion can be stored and insured for less than ten basis points annually, custodial gold could easily exceed 1,500 tonnes, leaving a pool of physical gold of under 1,000 tonnes.

From the bullion banks’ point of view, preserving physical liquidity is vital to their operations. And the obvious variable is ETF custodial gold. It explains why discouraging the wider public from becoming bullish is paramount, and in times of illiquidity the best policy is to just tough it out, knowing that investors always turn sellers when the panic of the day subsides, thereby releasing LBMA-vaulted bullion. And the LBMA trumpeting headline vaulting figures ten times the true liquidity adds to the deception. Let us not forget that the LBMA primarily represents the interests of the ringmasters in this market.

Basle 3 will fundamentally change the London market, removing lucrative gold trading under strictly managed conditions from the bullion banking cartel. This is why the LBMA vociferously opposed its introduction. They were relying on an approach of calling out the regulators for not recognising that gold is a High Quality Liquid Asset as defined in Basel 3: on this point they are certainly correct. But they then appear to have relied on the preponderance of spot transactions between settling LBMA members, published in The Alchemist Issue 99 and referenced in footnote iv to this article, to claim that unallocated is simply a convenient form of physical gold.[vii]

The PRA was not taken in by the LBMA’s lobbying efforts and would have looked at the underlying contract templates between LBMA banks and their customers and decided that these accounts were backed by derivatives as defined by Basel 3 and correctly recorded in the BIS’s derivative statistics. One wonders at the damage that the World Gold Council, which is meant to represent the wider case for gold while relying on its income from a physical gold ETF, would have done to its own reputation at the BoE by being sucked in to support the LBMA’s self-interested lobbying.

Obviously, the Bank of England (which hosts the PRA) was looking at gold trading in a wider context. The unallocated jig was up, and the Chinese were making a good fist of gaining control over global physical trading. That had to be countered. And in the near-final version of the PRA’s article 428f, not only was the London Precious Metals Clearing Limited put onto a proper clearing house footing (#1), but bullion banks were told that unallocated accounts must become pooled accounts backed entirely by unencumbered physical gold (#s 2 and 3).[vii]

Clearly, the BoE and its regulator desire a future for gold trading in London. It is likely that the UK’s Treasury and even the US authorities might have been consulted, given the US’s financial interest in curbing China’s strategic expansion into physical gold. The wider strategic implications of creating a rival to China’s growing global dominance of physical gold trading will not have been lost on the highest levels of government.

For gold trading, this is the other bookend to London’s big bang of the mid-1980s, which gave birth to the banking industry’s love affair with derivatives. That episode is now ending with Basel 3. London must adjust to predominantly physical trading. And the demand for physical gold is growing, not only as implied by Basel 3 regulatory changes, but more obviously driven by the increasing inflation of fiat currencies. It is no wonder that the most active Chinese bank in London’s gold market, ICBI Standard Bank, one of the four owners of London Precious Metals Clearing Limited, recently bought Barclay’s 2,000 tonne vault, because vaulting capacity for London’s physical demand in the wake of Basel 3 and considering current monetary developments will be extremely scarce.

Regulated gold futures contracts

Statistics on regulated gold futures are freely available, and fortunately for analysts, Comex’s regular gold futures GC contract dwarfs all others, allowing them to confine their comments to this one market.

It is far smaller than OTC forwards. At the end of 2020, outstanding futures contracts were for 1,739 tonnes equivalent, which compares with the 8,676 tonnes equivalent for forwards and swaps in the BIS’s statistics. The relationship between the two major gold markets has usually been described as Comex providing a hedging facility for dealers in London. This is meant to occur through two mechanisms. The first is that net short positions on Comex hedge net long positions in London, where there is a drip-feed of global mine supply of about 70 tonnes a week. And the second is the exchange for physical facility which allows for positions to be transferred between markets.

Besides the EFP facility (a misnomer because it is an exchange between futures and forwards — no physical is involved), the relationship between the two markets is very different from the conventional story. If anything, the hedging requirements out of London are for LBMA member banks to access net longs on Comex, because the bullion banks are short to their depositor customers’ unallocated gold accounts to a far larger extent than the drip-feed of physical bullion coming into the market. In London, banks job on the short tack, just as they do on Comex.

Bullion bank trading desks are included in the Swaps category, which is defined by the CFTC as, “an entity that deals primarily in swaps for a commodity and uses the futures markets to manage or hedge the risk associated with those swap transactions”. In other words, not exclusively the Swaps category includes to bullion bank trading desks. Referring to the CFTC’s Bank Participation Report, we see that on 6 July (the last report available) banks accounted for 59% of the Swaps category longs, and 73% of the Swap’s shorts. Of these, 31 are foreign, likely to be LBMA members.

Together with the Producer/Merchant/Processor/User category, swap dealers on Comex are designated as non-speculators, while Money Managers and Other Reportables are designated as speculators. And finally, there are a minority classified as Non-Reportables included in the speculator category.

In effect, the two non-speculator categories provide market liquidity, and they record both long and short positions, almost always being net short. The table below shows the recent position in Comex, excluding spreads, which are matching contracts arbitraging price differentials.

Based on the weekly Commitment of Traders Report issued by the CFTC, the table is arranged to show the non-speculator categories separated from the speculator categories, along with the position changes from the previous week (the panel to the right). Of the 500,187 contracts of open interest on 27 July, 403,286 were not spreads (81,518 + 321,768). Of these, the Swaps were liable for 77% of the net short position, representing 536.6 tonnes equivalent.

For the purposes of Basel 3 and its NSFR, regulated derivatives are little different from OTC derivatives, so we can expect banks dealing on Comex to reduce their involvement as soon as they can practicably exit, especially as the same banks run books in London as well. But the overall Swap category’s position is proving difficult to reduce, as Figure 3 illustrates.

Total swap net shorts stand at $31bn. According to the most recent Bank Participation Report, the banks share of this works out at $24.4bn. There are two problems standing in the way of these banks eliminating their short futures position. The first is growing interest from the speculator managed money category in selling dollars for gold futures, and the second is a new trend whereby the Producer/Merchant category is reducing its net short position, increasingly throwing the onus for supplying longs to speculators onto the swaps.

Furthermore, the shortage of physical liquidity combined with the increase in monetary inflation for all the major currencies and the dollar especially is turning Comex into a physical delivery market. So far this year, users of the futures market have stood for delivery for a total of 123,100 contracts, representing 383 tonnes, in a market where in previous years deliveries of physical were not common.

Central bank gold leasing could become a major issue

In 2002, Frank Veneroso, a respected analyst concluded that central banks had leased anything between 10,000—16,000 tonnes of gold. It is now largely forgotten, but just as there is little public evidence of continued central bank leasing there is no evidence that it has declined or been reversed. For a central bank the purpose of leasing was to earn interest on an otherwise non-yielding asset to pay for storage costs and to become a profit centre. For an arranger of leasing, such as the Bank of England, it ensures that physical bullion is available to manage markets. For example, last August saw the Bank of England listed as a sub-custodian for the GLD ETF. This was no doubt a filing disclosure of a leasing arrangement to make up for a severe bullion shortage in the markets driven by rising public demand.

Concealing arrangements whereby gold is made available by a central bank to the market has been facilitated by the IMF in its accounting treatment of central bank gold. It is the IMF that collects the figures. Gold swaps are recorded as collateralised loans, with the gold remaining on the central bank’s balance sheet. A gold loan, which is treated in the same manner as a repo, also remains on the central bank’s balance sheet. In its latest guidance, the subject of gold leasing is omitted. But reclassifying a lease as a loan which remains on a central bank’s balance sheet gets round the problem by simply redefining it.[ix]

There is little doubt that government-owned gold has been used to “manage” the gold price. In the 1970s, the US Treasury openly sold bullion by auction, but stopped doing so when they merely stimulated demand. Regulatory encouragement for the expansion of derivative markets has subsequently absorbed demand that would otherwise have driven bullion prices even higher. And there is the stubborn refusal by the US Treasury to quash rumours of missing gold by appointing an independent metal audit of monetary gold in its possession.

But perhaps the most damning evidence was the refusal of the New York Fed, responsible for storing earmarked gold for foreign central banks, to firstly let Bundesbank officials inspect its earmarked gold, and then to refuse to deliver it to Germany as requested. After some haggling, in 18 months the Bundesbank only managed to get back 37 tonnes of the 1,500 tonnes held in New York on its behalf. There should have been no argument about it: earmarked gold is gold held in custody and as custodian the New York Fed should have responded immediately to the Bundesbank’s instructions.

They didn’t. Did the Bundesbank suspect the US authorities were misappropriating its gold, and that was why it wanted it back? And why were its authorised representatives denied access? And the tonnage that came back from America was melted down immediately to “bring the bullion up to the current bar standard”. Unsurprisingly, this action stoked speculation that it was to eliminate bar details which didn’t match the Bundesbank’s records.

This treatment of Germany’s undisputed property by the US authorities sparked a similar withdrawal request from the Netherlands, which was satisfied in a timelier manner. And even Austria thought it wise to send a team of auditors to check on its gold in the Bank of England’s vaults.

The message seems to be that the central banks, which have between them declared ownership of 35,544 tonnes, are concealing old leases, more recent swaps and loans, and outright misappropriation of earmarked gold. It is a market deception that has been brewing for half a century. It was a problem likely to remain hidden so long as two conditions continued to be fulfilled: derivatives would continue to expand to soak up excess demand synthetically, and global money-printing would not spiral out of control. Even back in 2002, Frank Veneroso reckoned up to 50% of monetary gold in the central banks might have vanished in leases. Circumstantial evidence, such as the Bundesbank’s experience, suggests it is a trend that has continued, in which case a major portion of the West’s monetary gold has simply vanished.

At a time of mounting evidence that the destruction of purchasing power for fiat currencies is approaching, for some central banks the backstop of turning their fiat currencies into credible gold substitutes may not be available, and nowhere is this more of an issue than the future backing for the dollar itself.

A brief note on silver

So far as we are aware, central banks do not store silver. A possible exception is the Peoples Bank of China or through one of its agencies, which was appointed in 1983 to manage the acquisition of China’s gold and silver. But just as banking involvement in gold derivatives is set to eventually become little more than a topic for financial historians, the same is true for silver.

Silver was finally demonetised in the 1870s. Its replacement with gold in monetary standards for the few remaining European nations on silver standards led to its lower repricing as a predominantly industrial metal today, and the gold-silver ratio rose from 15—16 or so to over 70 currently.

We are so used to the state defining money for us that we forget that all theories of state money are fatally flawed, because its promoters can never resist actions that lead to its destruction. And when state currencies die it is the people who decide what will replace them as money, and the most reliable replacement that people have always returned to through the millennia is gold and silver. Following the chaos of a currency collapse the decision as to whether silver can circulate again as money is a decision for the people.

On a practical level, silver is more accessible than gold to most people. Today, a one ounce silver coin buys $25-worth of goods, and as fiat currencies slide, a silver ounce buying, for example, a value of one-twentieth of a gold ounce in silver is practical money for most people faced with exchanges of goods in the absence of fiat currencies.

These conditions are likely to arise following an accelerating expansion of state money and credit, such as that developing today. It is a separate consideration from the market changes that arise from bank regulation and the resulting diminution of paper silver. Suffice it to say that what applies to the effect of banks withdrawing from gold derivatives also applies to those of silver.


This month marks the fiftieth anniversary since the last vestiges of a gold standard for the US dollar were abandoned. We can surmise that it was by design that the banking reforms that followed in the 1980s contributed the suppression of gold by expanding the availability of derivatives substituting for physical bullion. And it became central to the promotion of the dollar as gold’s permanent monetary replacement.

Today’s policymakers repeatedly exhibit an ignorance of the underlying reasons behind the promotion of the dollar as the worlds reserve currency. Through the passage of time, they have probably come believe that gold has no future monetary role. And the few among the statists who are alarmed by their addiction to monetary inflation probably think salvation lies in technology —digital currencies issued by central banks, cutting out credit creation by commercial banks to increase central banking control over money and credit.

The escape route of digital currencies probably explains why reducing the role of commercial banks in financial markets by curtailing their derivative activities is not raising serious concerns at the central bank level, let alone over the likely impact on the gold price. We can only conclude that at the highest levels of government the authorities are no longer concerned that a rising gold price is a challenge to fiat currencies and can be simply dismissed — in which case they will have underestimated the likely consequences of reversing a fifty-year tide of gold suppression.

[i] See It should be noted that the gross market value is the sum of in-the-money valuations. Because for every derivative there is an equal and opposite position there are matching gross market liabilities which are not recorded in the BIS’s semi-annual surveys.

[ii] NSFR = Available Stable Funding/Required Stable Funding, and must always be greater than one

[iii] A swap contract is where two parties agree to exchange between them a variable rate for a fixed sum. It is principally a means of insuring against price volatility.


[v] See

[vi] See

[vii] See

[viii] For a detailed explanation of the relevant rules, see

[ix]See the IMF’s Monetary and Financial Statistics Manual and Compilation Guide available from

The views and opinions expressed in this article are those of the author(s) and do not reflect those of Goldmoney, unless expressly stated. The article is for general information purposes only and does not constitute either Goldmoney or the author(s) providing you with legal, financial, tax, investment, or accounting advice. You should not act or rely on any information contained in the article without first seeking independent professional advice. Care has been taken to ensure that the information in the article is reliable; however, Goldmoney does not represent that it is accurate, complete, up-to-date and/or to be taken as an indication of future results and it should not be relied upon as such. Goldmoney will not be held responsible for any claim, loss, damage, or inconvenience caused as a result of any information or opinion contained in this article and any action taken as a result of the opinions and information contained in this article is at your own risk.