LME warehouse changes come closer into view

The LME has released it's somewhat delayed modified set of rule changes on warehousing. Rather than providing concessions to warehouse load-out rules it has made them more stringent. As a background as to why the LME are proposing rule change, please see this post on what has gone awry with the current system.

If these rules are to become binding by next year, current trend in stock building at some warehouses will change given the risk that warehouses will be forced to deliver huge tonnages in mid-2014.

In this post I will use the example of the warehouses in Vlissingen, which is most symptomatic of what has happened in Aluminium. This analysis could easily also be applied to Detroit in the case of Aluminium or New Orleans in the case of zinc.  Over the next few weeks, ill expand this to include other "affected warehouses" and what it might mean for total inventory delivery.

There have been few changes to the original proposal as laid out earlier in the year.  Those warehouses holding a certain amount of stock will be subject to minimum load out rules as shown in the table on the left.

On top of this, those warehouses that have "excessive" queues for delivery are subject to further requirements for minimum load-out rates.  The key change in this most recent proposal is the lowering the queue length to trigger additional load out requirements from 100 days to 50 days of warrants cancelled.

The rules determining minimum load-out rules kick in if the limits are reached during any business day during given calculation periods. The first of these began as of the 1st of July this year and will end on 31 March 2014 as a preliminary period.  The first period of even more stringent rules is from 1 April to 31 June 2014 and is then calculated every 3 months.

The easiest way to demonstrate these calculations is via a working example, with the example provided by the LME very similar to the current situation at Vlissingen.  Currently at this warehouse there is a little over 2mt of aluminium on warrant, of which ~1mt are cancelled.

Because of the tonnage held, the minimum load out rate is 3,000tpd.  The 1mt of cancelled warrants calculates to a queue of over 330 days, so it easily meets the "affect warehouse" definition.

At the conclusion of the preliminary period, Vlissingen will have to deliver the minimum 3,000tpd plus the difference between total load-ins and the higher of actual load-outs or the minimum requirement over a the preliminary period.

For example In the period from 1 July to today, cumulative load-ins has been ~336kt, while outgoing metal has been a little over the 3,000tpd minimum at 274kt.  So if the preliminary period ended today, they would have to deliver over a 3 month period a daily average 3,000 + (62kt/65 days) = 3,950 tpd

Lets for example assume that the trend of load-in vs. load-outs continues until the end of the preliminary period.  That would mean total load ins would be a whooping 136.5kt larger, raising the minimum load-out requirment to a huge 5,100tpd.

This potential requirements is a large disincentive for the status quo to continue over the next 100 or so business days.  If anything, I would expect the cumulative increase in stock at Vlissingen to be 0 by the end of the preliminary period, so they can avoid the additional load out requirement in the first discharge period.

The requirements on affected warehouses get more difficult again after the end of the preliminary period.  Not only do they have to load-out the cumulative net gain in inventory, but the also have to load out 50% of the amount loaded in during the 3 month period, up to including the minimum load out rate.

So for example, if you are an "affected warehouse" and you don't build any additional stock over that 3 month period, i.e. your total load-ins equals your minimum requirement of 3,000tpd.   At the end of this period, the daily average requirement will still go up even if there is no additional stockbuild by the calculation 3,000tpd + (0.5*195kt/65) = 4,500tpd.  If stock is built on top of this, that must be withdrawn in the following period as well.

The diagram below should the requirements at different points in time for the calculation periods and discharge periods.

Ultimately, there is a large incentive for affected warehouse not to take any additional stock, as it will only lift their minimum requirements over the following 3 months.

This should solve the problem of LME warehouses bidding for material, as the minimum load out requirements will mean queues will have to drop.  As queues drop, so does their ability to cover premiums via rents earned.

Because the minimum-load out rules are heavily skewed towards not building additional stock, it seems likely that warehouses like Vlissingen will deliver at 3,000tpd and not take in any additional material once the first calculation period has been reached.

That means it will still take quite a long time for metal to clear wareshouses.  Its also unclear where this will go, given there are still incentives to move excess inventory into non-LME warehouses for carry trades.

But premiums are likely to start to fall more meaningfully.  Premiums had been relatively stable recently as market participants waited to see if this proposal was delayed or modified.  But now its been released, it seems likely that the willingness of warehouse's to attract more tonnage will shift dramatically.