Do ETF outflows mean the end of the gold bull market, or just the end for paper gold?
Amidst declining gold prices in 2013, some analysts have been quick to point to outflows from gold ETFs as evidence that the gold bull market is over.
We think this is wrong-headed thinking as we explain below, where in the next part of our research into the gold market we turn our attention to gold-backed Exchange Traded Funds.
Here we take a look at the size of the ETF market compared to other gold markets and ask if ETF outflows indicate the end of gold demand or a change for the better.
Gold-backed ETFs are sometimes (wrongly we think) seen as a cheaper way of gaining exposure to the gold price and cheaper than holding physical bullion. But we believe that the attraction of owning ETFs is quickly disappearing in favour of owning physical gold. Investors are no longer interested in owning a financial instrument that removed from the safe-haven instead they want to own the safe-haven itself.
ETF outflows and the gold price
Since mid-April outflows from gold-backed ETFs have been blamed for gold’s inability to recover from the price smash. In the first quarter of 2013 holdings have fallen by nearly 180 tonnes. Goldman Sachs and Société Générale (to name just two) cited ETF outflows as evidence that confidence in the gold market was disappearing.
According to the World Gold Council, ETFs account for a small proportion of gold demand, just 6.5% on average over the last three years.
Despite representing such an insignificant portion in the global gold market, the World Gold Council believe ETFs (along with futures) have a direct impact on ‘price formation and the gold spot price.’
In the fortnights both preceding and following the gold price fall in April, outflows from gold backed ETFs totalled 176 tonnes, topping February’s record of 111 tons. The world’s largest gold-backed ETF, SPDR Gold Trust’s GLD saw its biggest monthly outflow in 8 ½ years.
These serious outflows, which according to Reuters average 20 tonnes per week this year, may have come as a surprise to many as 2013 appeared promising for gold ETFs, as at the beginning of the year holdings reached a record 2,767 tonnes.
Performance this year of top gold ETFs
Marcus Grubb, MD of the World Gold Council estimated in a recent interview that gold-ETF outflows amounted to 600 tonnes for 2013 alone.
Currently assets in the world’s largest ETFs exceed $56 billion, as we reported previously, GLD holdings account for over 70% of this.
When outflows of gold ounces are the point of comparison it is clear to see why the many believe ETF outflows are responsible for the fall in the gold price. GLD has seen three times the amount of outflows than COMEX depository drawdowns.
Yet when comparing percentages it is clear that COMEX has been much more heavily affected. Drawdowns in inventory are nearly 6 times higher than GLD outflows. GLD outflows, compared to other ETFS are only marginally higher.
This suggests that whilst confidence in both futures and ETFs could be waning, COMEX is bearing the brunt of this move from paper to physical.
Gold price discovery
Although ETFs account for a significantly small amount of gold demand, they appear to hold an outsized effect on gold price discovery, given their diminutive size next to COMEX volumes. Below we compare the largest gold futures market’s open interest to the largest gold-backed ETF, in terms of dollar value traded in the last week.
Given the volumes of GLD traded compared to COMEX one may be surprised over how much the former is held accountable for the gold price weakness.
However, whilst it may seem that the futures market dominates the gold market in terms of trading volumes, it’s a different story when you compare the notional value of open interest on COMEX (in the last week) to market value of the top gold-backed ETFs.
Last week total holdings in the ETF fell to 935 tonnes, a level not seen since February 2009. This has fallen from a December peak of 1,353, over a 30% drop – nearly equivalent to the fall in the gold price since the September 2011 high.
If Marcus Grubb’s estimations are correct then GLD alone accounts for 50% of total gold ETF outflows seen in 2013.
iShares COMEX Gold Trust ETF, the second largest ETF in this analysis, is clearly significantly smaller than GLD, but it hasn’t suffered nearly as much. Holdings are now seen at a similar level to August 2012. From a peak of 221 tonnes in November last year, holdings have fallen to just 178 tonnes, a fall of 19% – significantly less than GLD.
Physical gold demand eclipsing ETF outflows
There is little reason to suggest ETF outflows mean we are witnessing the end of gold’s phenomenal run. However, the gold is clearly called for elsewhere.
The surges in demand we are seeing across the globe appear to predominantly come from two camps, the first is central banks and the second is the household in investor – Mrs Wong as she has been labelled, in China.
The gold seen previously as part of ETF holdings may well be making its way to China, and to a lesser extent India, where they continue to import significant amounts of gold.
In previous research we have highlighted the elevated levels of delivery taking place on the Shanghai Gold Exchange. As has since been widely reported, SGE deliveries continue to match or at least reach towards global gold mine supply. However, whilst mine supply is an interesting statistic to measure against, it shows that much of this gold must be coming from elsewhere.
We believe a significant proportion is thanks to ETF outflows. In April, GLD outflows quadrupled which could have potentially accounted for over half of the SGE’s April deliveries.
Whilst ETF outflows may have totalled 180 tonnes in the first quarter of the year, this was a tiny proportion compared to the gold that was bought up in Shanghai in the same period.
Ultimately these ETF outflows should be measured against significantly larger and elevated physical demand around the world, especially Asian demand. By doing this we see that ETF outflows do not represent a bearish gold market, instead they indicate the changing nature of gold demand.
Are all ETFs suffering?
The eschewing of paper gold, in favour of physical bullion is clearly shown in China where the launches of two gold-backed ETFs have failed to raise as much money as expected. Meanwhile on the Shanghai Gold Exchange their levels of delivery beat or continue to match global gold mine supply.
The physically focussed Chinese gold investor is not interested in financial products that over-complicate gold.
Elsewhere in Asia, where demand is perhaps more consistent than in the West, ETFs have had a slightly more promising year – in Japan gold ETF assets have grown 10% in the last year. Whilst they are relatively new to the East, they didn’t experience the panic outflows, instead during the first quarter of 2013 they received $33.5 million in in-flows. Despite the fact that appetite for owning such instruments is clearly not as great as seen in the West, in India there are 14 gold ETFs. These Indian gold ETFs hold only 40 tonnes between them, a fraction of the 860 tonnes imported last year.
Do ETF outflows mean the end for gold?
Gold positions may well be being ‘dumped’ by those holding ETFs, but crucially we believe this is investors exiting one inferior investment product, in favour of a superior one – physical gold bullion.
GLD cannot be argued to be accessible if you are looking to own physical gold. Every ten shares of GLD represents 1 ounce of gold. However, in order to redeem shares for physical gold you must own a minimum of 100,000 shares, i.e. 10,000 ounces. At the time of writing this is over $13,000,000.
GLD is only an efficient way to buy gold in physical form if you’re a hedge fund – like David Einhorn’s Greenlight Capital – and can afford to buy baskets of shares and redeem them.
The reason the aforementioned fund manager, David Einhorn, did not continue to hold his gold position in the form of ETF shares, was due to worries about counterparty risk. There are at least 3 counterparties involved in the delivery of an ETF product.
Very few Mrs Wong’s have $13m to allocate to gold. The demand for physical gold is clearly driven by the East, they want to take delivery of affordable bars of gold with reduced counterparty risk.
It is the same situation at both COMEX and the London Gold Market where contract sizes and minimum delivery requirements are inaccessible for the majority of investors and the counterparty risk is significantly higher than owning physical gold.
Conclusions on gold ETFs
We do not agree with the mainstream and those analysts who believe ETF outflows signal the end of the gold bull market. These analyses are overly broad brush in their approach and also plain wrong.
There is clearly no doubt that some speculators in the West have been lowering their gold exposure – sometimes held in the form of ETF positions – but we do not believe that these ETF outflows signal worldwide withdrawals of capital from the market. Instead we believe it signals a change in the nature of the global gold market.
We are also convinced that the on-going debate about ‘paper gold’ has grown in its size and audience, meaning that more market participants opt for physical instead of owning a security or debt instrument.
As the gold market more widely wakes up to ‘paper gold’, we believe this will continue to negatively affect participation in gold securities and derivatives.
However, it is important not to mistake this for a mass exit of capital from gold and the end of the bull market.
Like our research? Get it delivered to you with our fortnightly newsletter.