15 May 2013

Why the price smash affected GLD and SLV stocks differently

A number of bloggers have observed the difference between GLD's gold stocks and SLV silver stocks in response to the April price smash. Sharelynx is reporting the following changes over the past four weeks:

GLD down 3,031,042oz (-8.23%), current stocks 33,811,468oz
SLV down 341,111oz (-0.10%), current stocks 335,666,675oz

Sharelynx also tracks all the other major ETFs, COMEX, TOCOM, Sprott, BMG, Central Fund, Bullion Vault and GoldMoney reported stocks. The change in the total of all those over the past four weeks is:

Gold down 5,576,479oz (-6.12%), current total 85,565,264oz
Silver up 912,541oz (0.11%), current total 855,911,574oz

Whether you look at GLD vs SLV or total gold stocks to silver stocks, silver is basically holding even with gold taking a 6-8% hit. The explanation I think has a lot to do with who is investing in GLD vs SLV (or gold vs silver more generally).
Latest figures from Reuters has GLD's ownership by institutions at 51.3% while SLV's is 19.6%. Deutsche Bank notes that "one-third of institutions holding bullion will probably keep it. We expect that the bulk of the drawdown comes from institutional investors rather than retail investors".

So GLD/gold holdings have dropped primarily due to institutional liquidations whereas SLV/silver holdings has held up because there are more individual "buy and hold" investors in SLV/silver.

My thesis is sort of supported by looking at Bullion Vault's numbers, as Bullion Vault is primarily a retail product (average account is $50k link). For gold over past four weeks they are only down 1.6%and for silver they are up 1.1%, which is very different to the general trend.

The investors in the Sprott funds are the strongest hands of all, with PHYS and PSLV showing zero change in ounces held (that is a joke, BTW).
PS - a couple of interesting facts from the Sharelynx numbers:
1. Both GLD and SLV have a "market share" of publically reported stocks of 39%
2. Ratio of silver oz to gold oz is almost exactly 10:1 (ie for every ounce of gold held, 10 ounces of silver are held)
3. Ratio of silver to gold by dollar value is 0.16:1 (ie for every dollar invested in gold, only 16 cents is invested in silver)


  1. "publicly reported stocks" of gold & silver ETPs I take it?

  2. and COMEX, TOCOM, Sprott, BMG, Central Fund, Bullion Vault and GoldMoney stocks as well.

  3. Thanks Bron,

    That leaves what? London? What's your best guess?

  4. I'm not sure I understand your theory. You're saying that the GLD inventory draw-down is due to institutional liquidation of GLD shares? Does that mean that when shares of GLD are sold, their physical inventory goes down by some equivalent/corresponding amount?

    Are you ruling out the possibility of share-holders taking physical delivery?

  5. If an institutional holder of GLD shares "liquidates," that just means they sell the shares to someone else, right? This does not cause any metal to leave the trust. Metal only leaves if an Authorized Participant (AP) decides to redeem a basket of 100,000 shares for the corresponding amount of metal. Are you suggesting that large volumes of institutional selling are causing the share value to fall below NAV, thus giving the APs an arbitrage opportunity, and that's why they are redeeming shares for metal?

  6. Not sure about this, but i think if you hold enough shares you can redeem them for physical and the shares are not sold but dissolved

  7. S Roche - total investment stocks are estimated at around 30,000t

    Anon - APs are the only ones who redeem ETF shares for physical and they are doing that because of net selling by holders. My post is speculating on who is doing most of the selling of the ETFs.

    I do not think it is ETF investors just deciding to take delivery and keep invested as this inventory drop coincided with the price smash.

  8. What do you mean by "net selling"? Suppose hedge fund A decides to sell 500,000 shares of GLD. There's someone (probably multiple someones) on the other side of the transaction buying these shares -- call these buyers B, C, and D. Selling by A is buying by B, C, and D.

    Now it's possible, even likely, that such a large sell order will drive the GLD price below NAV. So maybe buyer D is an AP who sees an arbitrage opportunity. D snaps up 200,000 shares below NAV and simultaneously puts in an offsetting short position at spot. D redeems the shares, gets the metal, closes out the short position with the metal, and pockets the arb amount.

    Is *this* the mechanism you have in mind? The idea with SLV then is that such a massive sell order that would drive the price below NAV is less likely, thus no arbitrage opportunity, thus no metal leaving SLV. Correct?

  9. I met with Juan Carlos Artigas, one of the Research contacts at the WGC. In the meeting the subject of share redemptions for physical by the public was mentioned, and he indicated that, if an individual/institution has 100,000 sh lots, they can petition an AP to redeem for them. The process was described as "laborious and costly",few actually bother, but can in fact happen, and has. FYI, dh

  10. What is the "joke" on PHYS and PSLV?

    Is that a rip on Sprott? Are his vaults being emptied like GLD and SLV's?

    I hear Sprott talking a lot of King World, and he seems very solid. But I'm just looking in through the windows.


  11. If A = B + C + D then that is not net selling. If D is an AP then that is net selling of an amount equal to A - B - C; as D is not a real buyer as they have just hedged themselves by selling metal in OTC market.

    The joke about the Sprott funds is because they are closed end funds.

  12. Bron:
    Perhaps you can help me, hopefully with a post on GoldChat.

    There is one thing I cannot quite understand, despite having thought about it for some time.

    Who, besides those individuals and some institutions that own physical or paper gold, actually loose money, or incur opportunity costs, when the price of precious metals declines, as has happened recently?

    I understand that the Perth Mint does not, as it owns no precious metal. It stores allocated metal for account holders, and it backs its unallocated accounts with metal that is being refined, or fabricated, or is for sale. Being a Mint account holder, both allocated and unallocated, I know full well who is exposed to changes in the prices, both up and down.

    However, the Perth Mint’s ‘business model’ is apparently unique in that it doesn’t involve hedges. How about other refiners, fabricators, and purveyors of precious metal products, particularly at the wholesale level?

    I know, from past experience, that Local Coin Shops always know what the current going wholesale prices are, and will still phone a wholesaler when a large transaction is in the offing in order to lock in a guaranteed price that they can make a profit on. Fair enough, or they couldn’t stay in business.

    But how about the larger operations? How do they hedge their stock against price movements, particularly to the downside, as they will profit from price increases on stock they hold but cannot let themselves be unprotected from downside risks if they want to remain in business.

    I know that spreads tend to increase when ‘spot’ prices go down, but only temporarily and sooner or later adjust to lower prevailing prices. I also understand that, eventually at least, miners will only be able to sell the partially refined metal that they produce at the lower prevailing prices. But there are lags at all stages from mine output to retail sales.

    If one looks at the COMEX, which is not really intended to be a major vehicle for delivery of large amounts of physical metal, it is basically a ‘zero sum game’, or speculators’ market , with clear winners and losers on every contract. Do large bullion buyers and sellers hedge their holdings of physical metal there with paper contracts? Who, then, would be their losing counterparties?

    Or is most of the necessary hedging done on the LBM Over-the-Counter unallocated market, where there are presumably also clear winners and losers, at least over time?

    But the main players on the LBM are the Bullion Banks, are they not? Do they hedge against price declines with short forward contracts? If so, who are their counter parties, other Bullion Banks? Or Central Banks? Or just big speculators, like hedge funds?

    Again, someone has to loose when prices go down. So who are the losers when the evil manipulators crash the COMEX derived ‘spot’ price? Or alternatively, do efficient markets just naturally balance excess supply and declining demand with lower prices?