Ahead of tomorrow’s Google+ Hangout, Jan Skoyles provides a brief look at the research series she will be discussing, along with the key findings.
In May we began a series of articles that examined the gold market and how the price of gold is determined globally.
It was clear from the outset of the gold price smash that there is a divergence happening in the gold market. Inventories of ETFs, funds, and futures market depositories collapsed by over 5.5 million ounces ($7 billion), physical removals were reported by the COMEX of 1.4 million ounces and the SPDR Gold Trust (GLD) reported total inventory removal of nearly 4 million ounces.
However, investors in the East and emerging world were grabbing all of this unwanted gold.
Deliveries on the SGE reached levels never previously seen and delivery numbers remain elevated.
Yet despite this the gold price remains low. It was clear that demand for physical gold was not enough anymore, so we set about looking into the elements of the gold market.
Our first article in this series took an overview of the gold market; from the different financial instruments that allow investors to gain exposure to the gold price, to the supply of gold from refineries.
Whilst a very high-level look at the gold market, the research showed that the gold price is not the sum of several parts but is instead controlled by dominating elements – namely COMEX and the London Gold Market.
This finding was despite the size of the ETF market which had taken much of the blame for the gold price smash. The media had clearly confused ‘gold-backed’ ETFs’ effect on setting gold prices.
When comparing paper markets (GLD and COMEX) to the physical market we found that the volume and capacity of the physical market was significantly lower than that of the paper products, despite the comparable market caps. When we drilled down further we found that COMEX was twenty times larger than the GLD, in traded volumes.
We took a quick look at refineries and their annual capacity. Unsurprisingly we found that Switzerland dominated this scene, playing home to the top three largest refineries. The top refineries in the world have a total annual capacity of 8,000 tonnes, twice current gold supply. Despite this vast size, in the week that we examined COMEX, the volumes traded were 50% of annual refining capacity.
It was not difficult for us to conclude where we believed was the beating heart of price discovery – COMEX. However, we were quick to conclude with the thoughts of Kyle Bass who once explained, if the notional value of open interest at COMEX is too large compared to physical bullion lying in the COMEX warehouses, then ‘you go get your gold.’ It’s a confidence game… like most financial markets.
Inevitably COMEX was where we had to next turn our attentions to. Given the huge volumes that pass through COMEX, even though such a small percentage of this paper can be delivered, it is understandable that many feel this one exchange distorts the price of gold and the realities of the gold market
The recent decrease in inventories, particularly from the JPMorgan warehouse, has attracted much scrutiny of late. Discussion of a COMEX default and settling large gold contracts with cash, rather than gold, i.e. defaulting, continue to animate many a gold market discussion.
Whilst we agree that the price derived from COMEX is not a true reflection of gold demand or even the true price of gold, we were unable to find evidence that the levels of open interest were currently a cause for concern.
In terms of market participants we found few seemed to have changed their positions on COMEX. Commercials short positions have even been falling in 2013, whilst their long positions have been growing. We also found neither commercial net positions nor non-commercial net position market participant was showing uncharted behaviour in an example of withdrawing liquidity from COMEX.
When readers realised that we failed to find COMEX under stress some were clearly upset that we had not been able to expose the market as one based on manipulation. In truth, we were only able to present the data available.
One of the key takeaways from our COMEX research was the rapid decline of warehouse stocks, something which has been the source of recent discussion. Jesse of jessecrossroadscafe blog wrote last week that registered gold (that which is eligible for delivery) continues to decline. It was these statistics that were leading many to believe that COMEX was under stress.
However when we looked at both cover and delivery ratios we found that we would need the number of deliveries need to increase by over 250% to put COMEX under stress.
We concluded that it is necessary to keep an eye on the delivery ratio exceeding the cover ratio. Should this happen the amount of ounces being demanded for delivery will exceed the stock of registered gold ounces at COMEX.
Being a fractional market COMEX is vulnerable to runs, as banks are, but at this time COMEX continues to function and hold the liquidity monopoly in the gold markets accessible to Western investors.
We were prompted to look at the London Gold Market thanks to Bron Sucheki who, after reading our research into COMEX, brought to our attention previous studies that found the centre of price discovery fluctuated between COMEX and London.
Whilst aware of the opaque nature of the London Market, we were still surprised at how little information we were able to gather.
The two markets, whilst clearly dominant forces, are tricky to compare given the London Market is inaccessible to individuals and retail trade. At first glance it appears that London is the significantly more powerful market, in 2011 it accounted for 86% of total activity within the gold market, whilst COMEX accounted for just 10% of total volume.
Whilst America’s lack of physical force in the market may suggest that it cannot compete with London as the heart of price discovery, a 2012 paper from Lucey et al suggests that the transparent nature of COMEX gives it its ability to compete with the OTC market. At present COMEX carries more influence, but as Lucey et al explained, this changes throughout recent history.
Compared to COMEX we found that activity in the London Gold Market is far less sustainable, in terms of gold bars available, should buyers and central banks choose to take delivery of their unallocated gold.
When we first set out to investigate the gold market we had not planned to look at the Shanghai Gold Exchange, however as the weeks went on there was little way we could avoid looking at it. The weekly graphs by @KoosJansen were a clear sign that if there was ever going to be a divergence in price discovery between paper and physical gold, the SGE could be the wedge that would force this division. In contrast to the 30% fall in the gold price, HK exports to China have increased by 68% YoY.
Unsurprisingly the volumes traded on the SGE, compared to both COMEX and London were so small we couldn’t compare them on the same scale. We believe that the future role of SGE in really impacting price discovery could be some way off.
Yet the levels of physical gold being removed from the exchange suggest that it will have a significant role to play in the supply of gold to the market. Mainly in constraining it.
In terms of physical gold demand the Shanghai Gold Exchange is heads and shoulders above the other gold exchanges. The delivery ratio on COMEX, in the last four years, has been below 5%, in contrast it has been consistently above 25% on the SGE.
But even these numbers may be misleading. Whilst both COMEX and the SGE allow participants to take delivery, the two take very different approaches.
There is a key difference in the term ‘delivery’ as our friend @koosjansen verified late last week.
‘…deliveries do not necessarily leave the COMEX vaults. When delivery is settled the buyer receives a warrant after which he can decide to take the metal out of the vault, or leave it in and resell it at a later date through going short. We don’t know how much metal is drained from the COMEX by looking at delivery settlements. Data on warehouse stock is available but doesn’t include numbers on withdraws and restocks, only the change in total stock.’
In contrast, he explains, ‘It’s 100 % certain that all SGE delivery numbers I publish are about physical gold leaving the vaults.’
Physical gold demand in the most populated country on earth does not seem to be subsiding, yet neither do COMEX and futures volumes generally.
We believe that as increasing numbers of Westerners look to take delivery of their gold from Western exchanges, there will be more pressure on the physical market. China’s primary physical exchange, as we see from delivery data, is the liquidity hub with the best reputation for such demand.
Anecdotal and media reports show that gold premiums are an ongoing phenomenon. Unlike investors in the West, Chinese housewives just want physical gold regardless of the price. This is clearly evident in the delivery data on the SGE, unlike in the West where we see gold as an investment which we hope will climb in price, the Chinese see gold as a way to free up devalued cash and place it into a real store of value.
ETFs bore the brunt of the blame for the fall in the gold price, yet our research suggested that ETF outflows showed a change in the nature of gold demand and not a decline.
Interestingly we noted that ETF demand wasn’t down in all funds. In fact in Japan’s and India’s demand had remained constant throughout the gold price drop. This was yet more proof of our Eastern contemporaries approaching gold investment over the long-term, rather than for quick profits.
The recent ETF outflows in the West suggest that we too might be changing our approach to gold investment, rather than owning a financial instrument removed from the safe-haven, we want to own the physical safe-haven. Given the inaccessibility of owning gold through GLD, and taking delivery, this is not an unreasonable phenomenon.
ETF outflows for Q1 2013 were minimal compared to total demand for gold in the same period. We believe much of this unwanted Western gold is making its way to China, and possibly India.
Outflows from GLD (which accounts for over 70% of the world’s largest gold-backed ETFs) were nearly three times those from COMEX warehouses, but in percentage terms COMEX suffered greater outflows.
Whilst weekly volumes on GLD are around one-twentieth of those on COMEX, the market value of the top ETFs compared to open interest on COMEX is almost the same, hence the perception that ETFs impact the gold price.
Concluding comments
We draw two concluding points from our series of research.
- The gold price is not determined by demand and supply of physical gold. Instead it has become the victim of a highly speculative paper market where financial instruments have been created in order to derive benefit from the inherent desire felt by investors to own, trade and punt gold.
- The nature of gold demand is changing. As Eastern nations and emerging economies take their place on the global economic stage their investment decisions will also make an impact on markets. These countries do not wish to hold these paper products when they can own physical gold instead.
Since we started writing these articles gold market developments and other research have taken a series of twists and turns. Namely, we have seen a change in the way the mainstream look at gold, many now pay attention to levels of delivery on physical exchanges as well as premiums being paid in both China and India.
Since April we have also seen significant controls on gold come into play in the world’s largest buyer – India. Many analysts have used this development as a sign that the gold bull market is over. But China is steal India’s crown, and whilst official demand may subside in India we have little doubt that demand in neighbouring countries is going to increase…
As Terrence Duffy (President and Executive Chairman of CME Group), told Bloomberg following the April’s gold price fall: “…we saw all the gold stocks trade down significantly, we saw all the gold products trade down significantly, but one thing that did not trade down, was gold coins, tangible real gold. That’s going to show you, people don’t want certificates, they don’t want anything else. They want the real product.”
The World Gold Council referred to the ‘dichotomous nature’ of the gold market – this is clear as the paper gold market continues to tell a different story to the underlying physical market.
Jan invites gold investors from around the world to join her for a Google+ Hangout Wednesday 7th August 2013, 20:00 UK time to discuss the findings and key takeaways from her latest research
How to join?
1) Circle The Real Asset Company to get your Hangout invite.
2) Visit the Hangout event page and click ‘Yes’ to join.
3) Become a member of our Google Plus community, be part of the build-up and keep up with Jan’s work.
