Normally I start an article with a premise ... expecting this, testing that ... but today I'll come out with it - I'm disenfranchised with the 'silver shortage' people because nearly everything I look at with the data points the other way. Don't get me wrong - we have only begun to scratch the surface of what is possible with debunking, debate and analysis, and 'time will prove all things', but increasingly the popularized silver stories (particularly of the 2010-2011 vintage) all appear as solid as warm pig swill.
Part of my research challenge is to bring the complex stories to an easily digestible level, here's a quick look at what we've been discussing internally (I apologize to my fellow Screwtape Files writers who have already seen this story). SLV has a new vault - in New York. This detail has been superbly written up by the folk at about.ag. We became interested in the many SLV vaults, here's a chart of that historical data:
|Red and Blue lines indicate the most interesting event.|
The keen observer will notice a very specific trend - the "JPM London A" vault has basically been emptied over most of 2012, and the inventory has been shifted to "Brinks London". Was it the same inventory? Glad you asked, yes it was - or at least 99.5% of it, which we were able to verify with a match on serial number, refiner and gross weight (matches were measured between 01-Feb-2012 and most recent).
Can this be independently verified? Yes it can. I'm not the only one processing the SLV bar lists - if you visit about.ag/slv you'll see a text summary of the various vault movements (at a weekly level). There are a lot of '... to brinks London' statements - i.e. our graph is just the pictorial version of some of these observations.
During internal discussion the best theory we had for the movement is the possibility that Brinks is offering a competitive storage rate for their new vault (in order to utilize the new capacity), and that the deal probably includes a good rate for shipping - Brinks is (after all) a bullion transport company.¹
Regardless of the reasons behind it, the Brinks vault now holds a lot of metal, some of which used to be in JPM London A. Because we have a database, we can also easily extract more detail and look at metal shipped over time. The following graph shows the inventory which 'moves' across.
|Movements to 'Brinks London'. Daily average seems to be ~800,000 oz.|
Week of 25-July-2012 seems to be 'a perfect week' with even deliveries Mon-Fri.
We established through the data that all of the changes are between each bar list issue - what I mean by that is that in one list we see the bar marked as 'JPM London A' and in the following day it is marked 'Brinks London' (i.e. the bar is never 'missing'). In theory this simply means 'committed to Brinks' since one would have to assume that the re-vaulting process takes longer than a day? In any case, the deliveries are consistent, and if you look closely you'll be able to spot the 5-day working week for the times when we get a bar list produced every day. It is safe to suggest that there is a minimum batch size here - we see approximately 20 tonnes get moved at a time - the larger amounts are possibly a reflection of more than one truck/trip per day or simply a build-up for where we don't see a new bar list for a few days. The 20 tonne figure roughly represents the highest amount that can be insured in a single trip.
A bullion transport company (who have just built a really massive bullion vault), spent 10 months relocating nearly 100 million ounces of silver from another major vault, the movements of which is clearly visible in the data and is consistent in regards to practical batch movements. Identify the best response:
(a) The silver is not real - Brinks are actually the same lizard people who built the Denver airport facility and the vault is a new bunker that few have seen. The staggered "deliveries" are just part of the cunning and calculated deception.
(b) Brinks (like RCM) are also running a fractional paper scam. JPM needed to "relocate" the silver because they were being audited. Blythe has Brinks employees under mind control.
(c) Oh okay, that 100 Moz is probably real silver - but the rest isn't. Buy a silver oz, crash JP Morgan!
(d) Uh, seems like a legitimate movement of silver.
My primary concern though, has been trying to look for dark bullion, which is the subject of the next article I am (still) writing. We'll go into this more next, but just so you know what I'm talking about - the ViaMat inventory that gets closed out in mid-2011 --> doesn't seem to resurface again anywhere in the data. That is about 50 million ounces in round figures. The same happens for the 'Johnson Matthew UK. Royston' vault just recently - as observed at about.ag, these vaults do not re-enter SLV.
We don't know where that inventory goes² - that's the point of it being 'dark', we've got no data ... maybe it got sold to mints who melted them down for coins or maybe they are just in the hands of private investors. In any case, it's a lot of metal which needs to be factored into the narrative of 'silver shortages'. What I mean by this is that we've only just looked at two vaults and we can say that there's about (rought figures) 59 million ounces out there which we've seen once, but are no longer accounted for. Once we start the full tally, we'll see a lot more going missing. Where is it exactly? There's still several possibilities due to limit of reading error with the data (which I'll save for next post) but in the meantime anyone still peddling the 'shortage of silver' meme is not really looking at the core data.
* With flooding and power outages, the standard for Brisbane these days it seems.
¹ We're still not generally satisfied on the 'reasons why'. Other possibilities for the vault movement may be related to insurance levels for vault inventories - apparently as the value of the metal has increased, the threshold and premiums for covering existing vaults is changing (just some comment I saw on the internet from some fund manager).
² The database currently holds data for 16 silver ETF's.NOTE: we don't currently process the data for ZKB fund because they don't publish the refiner codes.