31 January 2011

Fraud on GLD holders?

Further to my post yesterday, FOFOA left this comment on his blog discussing two risks he sees with GLD:

1. That the gold in GLD has multiple claims on it. Quote: “GLD's gold bars originated as reserves in the mainstream bullion banking system. That is, they are essentially reserves on loan to the ETF from the bullion bank's fractional reserves. And the lending of anything always creates a synthetic supply” and “that lent its reserves to paper GLD investors in the first place”

2. That shorting of GLD results in multiple claims on the gold: Quote: “The other side is the lending of this "synthetic gold supply" that creates an additional synthetic supply of synthetic gold. When someone shorts GLD they borrow the shares from someone else. And that same share can potentially be borrowed again and again.”

My short response is I disagree with 1 in respect of “lending” but agree with 2. Now for the long response.

The idea in point 1 that GLD has encumbered gold in it was raised in Catherine Fitts’ “Precious Metals Puzzle Palace” and also Hinde Capital’s “Precious Metals ETF Alchemy GLD – the new CDO in disguise?”

I discussed this issue in this post. If Authorised Participants borrowed physical (or used physical backing their unallocated liabilities to their clients, which is the same thing) and delivered that to GLD, there is no claim or encumbrance by the original lender to the Authorised Participant on those bars held by GLD.

To claim otherwise is to question the entire basis of Allocated gold that the market (and “giants”) operate and rely on, as no giant with a couple of tons of gold can just bury it in their backyard. It would also question the integrity of GATA consultant James Turk’s GoldMoney as well.

The second part of FOFOA’s comment is that any delivery to GLD by a bullion bank of physical gold that was supporting/backing the bullion bank’s fractional unallocated liabilities is a “synthetic supply” that effectively suppresses price by “divert[ing] growing investment demand away from the tightening physical market.”

I would note that for this statement to be true the bullion bank(s) in question must be naked short. Not all Authorised Participants for GLD would have access to the physical to do this, nor would they all be willing to take on such a financial exposure. Question to FOFOA: how many tonnes of GLD do you think are short?

A final point (and FOFOA probably won’t like this conclusion). I would claim that those at risk from this activity are not holders of GLD, but bullion bank unallocated clients because the legal title to the metal in GLD is held by the Trust, not the Authorised Participants or bullion bank unallocated clients.

Now this is a simple legal fact. It does not mean that in any meltdown when fractional claims come home to roost, holders of GLD will survive while bullion bank clients will not. It may happen that the custodians do “take” the gold behind GLD and give it to their unallocated clients first and use the get out clauses in GLD to say they “lost it” or some such other fraud if they are unable to subsequently replace it. Some believe that this is likely behaviour, others that no matter how ruthless bullion banks may be, that they would not engage in such outright fraud. I'll leave that to you, the reader, to have an opinion on.

FOFOA seems to think this is a possibility because he thinks that the bullion banks somehow have a special right to the gold in GLD, see this comment:

A place to park your unallocated deposits and sell them for dollars that can then chase an ROI, knowing that the gold will still be there for you to buy back any time it is needed, because you are an Authorized Participant with special rights to the gold.

I do not see how Authorized Participants have any special rights. Once they deliver gold and get GLD shares and then sell them for cash which they then "chase an ROI", they give up an rights to the GLD share or gold. Yes an Authorized Participant has special ability to redeem GLD shares for gold, but they have to tender the GLD share first, which in the situation above, they can only obtain by buying GLD shares off investors, thus pushing the price up.

If you, the reader, think that it is likely that bullion banks would steal Allocated gold, I would then argue that if a bullion bank was to consider engaging in such fraud, it would not do so with GLD’s Allocated metal because that is an exchange listed product with regulatory, audit and client visibility (through the bar list). Using Allocated metal held by other clients with the bullion bank would be a far lower risk as that is an over-the-counter market arrangement, subject to far less oversight as it is in the "dark". Again, GLD represents the least risky paper gold – the last to fall, so to speak.

Remember that gold ETFs are only 2,000t out of 30,000t of privately held gold, a fair bit of which is Allocated with bullion banks and other custodians. I think it is a more believable thesis that any short covering, fraud etc is more likely to occur in the over-the-counter "dark" market first, leaving the visible ETFs to maintain the fa├žade that everything is OK.

The point of my discussion above is not to suggest GLD is a safe investment. It is just to introduce a little more nuance beyond a simplistic “GLD is bad because bullion banks involved.”

In respect of point 2 about the shorting of GLD, I would like to see some numbers on that. I am not sure it is as pervasive as implied. This article on ETF shorting in general and ETF settlement fails give a sense of the extent of the issue, but unfortunately GLD is not mentioned.

One final comment by FOFOA:

From my perspective, GLD had the opposite effect on the price of gold (and may have been intended for just that purpose) as it diverted growing investment demand away from the tightening physical market.

All I will say to this is that is was not intended by the WGC for that purpose and there was obsession about creating a product that resulted in "physical offtake". Whether the vehicle that resulted was the best design is another matter. There seems to be this view that the WGC is part of the "bullion bank conspiracy". I suggest one look at the membership of the WGC and some of the companies and their involvement in other activities supportive of gold. WGC wanting the gold price to drop doesn't stack up in my view.

29 January 2011

Funny business in GLD

FOFOA has an interesting speculation on the movements in GLD:

So now I offer up a scenario, not as a statement of fact, but as fodder for thought and discussion. In this scenario I am not assuming that the drain on GLD to date has been the direct redemption of ETF shares by Giants. I presume it is simply redemptions by Bullion Banks in order to meet the delivery demands of "important clients," real Giants, perhaps from Asia and the Middle East. And because the BBs would normally have better options than plundering GLD, I am assuming those options are either gone or far more problematic than legalized looting.

Also, following Lance Lewis' "puke indicator," one could be forgiven for suspecting that the Bullion Banks have some way to temporarily "pound" the price of gold down on the COMEX in order to buy back ETF shares during a "good price window" with the intention of redeeming those shares into deliverable gold for clients that purchased it at a higher price.

I left a comment, which I post below FYI:

The reason one cannot correlate gold price and GLD holdings is because authorized participants (AP) don't have to create and redeem GLD shares on a daily basis in response to investor activity in GLD.

For example, if you're an AP and have a view that the market is bullish, then you expect over time to see net buying of GLD. Therefore, if on one day there is net selling of GLD, then you can:

1. Buy GLD shares
2. Immediately lease gold and sell it (or just short futures).
3. AP is now long GLD and short unallocated gold or futures. Important to note they have no exposure to gold price movements.
4. Sit on the GLD shares and when investor bullish sentiment returns
5. Sell your GLD shares
6. Buy unallocated gold and repay your lease (or close out your short future).

The above process means that the AP avoids GLD share redemption and creation costs.

Same thing happens in the face of net buying - an AP borrows gold and delivers it to GLD for shares, which they sit on an over a period of time sell into demand for GLD.

This is a way of minimising transaction costs when making a market in GLD or SLV or any other ETF.

The end result is that we see lumpy creation and redemptions, reflecting accumulated buying or selling activity over a number of days.

In the case of large lumpy redemptions, that can reflect an AP who held on to GLD shares in the expectation they would be able to offload them later into expected buying. If that buying does not eventuate, then the AP offloads the lump of GLD share they have as they are incurring ongoing funding costs.

You are correct in that redemptions of GLD cannot really be used to infer too much about what is going on re investor sentiment. The GLD bought back by an AP and gold redeemed is just sold by the AP to someone else, ultimately.

All GLD holding movements tell us is the sentiment of GLD holders. All that futures tell us is the sentiment of futures traders. Are these markets representative of the all private investors in gold. Maybe, maybe not.

What commentators miss is the OTC "dark pool". Consider that ETFs + Futures only represent less than 10% of estimate privately held gold (see this post).

In that case, we should not get too excited by the activity we see with ETFs and futures as it is not where the real giants are.

Consider also that bullion banks know their activities in ETFs and futures can be seen/deduced in some way. Therefore you must assume they let you see what they want you to see, with their real position and activities hidden in the "dark" OTC market.

19 January 2011

Australian Housing Prices

I've been following an excellent series of posts on the high cost of Australian housing and land from The Unconventional Economist blog. Worth a look if you are interested in the reasons why this is the case - Leith van Onselen's puts it down to restrictive government policies on land use.

18 January 2011

Reserve Bank of Australia Gold Sale

Further to The Australian's 11 January article "Reserve Bank's gold sale cost us $5bn", I thought my readers may be interested in the then Perth Mint CEO's response to the 1997 sale.

Consider these quotes:

"It is important to remember that we are evaluating gold now in a low-inflation environment, in which other financial assets are currently performing extremely well, the most important of which is the stockmarket. However, current conditions are not going to last forever. In my experience, in the long term, politicians cannot help themselves. When they are faced with a really tough decision, such as the one creeping up on them now - that of how to create jobs - they will take the easy option and let inflation go."


"When the crunch comes, governments will want to be re-elected. Rather than take the tough decisions, they will turn the inflationary tap back on. For that reason, I believe that gold will again have its day. I have not lost faith in gold."

The foresight of these statements come from the fact that Mr Mackay-Coghill worked for International Gold Corporation (Intergold) from 1971 to 1986. Intergold represented South African miners in the production and selling of Krugerrands and Mr Mackay-Coghill was responsible for the introduction of the Krugerrand to world markets in his role as CEO of Intergold.