Assessing the scale of metal warehouse trades

Assessing the scale of metal warehouse trades

Izabella Kaminska | Aug 08 08:59 | 1 comment | Share

Earlier this week Morgan Stanley published an in depth look into the financing warehouse trades in metals — the ones most analysts have been in denial about (at least publicly) for at least five years — and why they are now, thanks to new LME proposals, finally easing. The note is titled: “Beginning of the end in warehouse trades: A game changer for base metals”. There were three notable observations.

First, it’s not just banks that should be blamed for fuelling the queue and inventory over-financing problems. Part of the problem is related to the general demise of independent warehouse operators in the metals industry. That is to say, there aren’t enough warehouse owners who do not have conflicting interests as traders or bankers on top of their warehousing businesses: As has subsequently become clear, domination of a good delivery point through the ownership of the majority of multiple warehouse units licensed to receive a particular metal in a given location is a necessary condition for controlling metal inflow, outflow and associated rental cash flow that is the foundation of warehouse rental and financing deals. In addition, a sizeable balance sheet and an active involvement in physical trading as a means of engaging in inventory finance or warehouse incentive deals became a sufficient condition of this domination. Not surprisingly, the lessons and opportunities provided by the Metro acquisition were not lost on the big physical commodity trading houses, which, in a short space of time from March 2010 onwards, bought the majority of the remaining independent LME warehousing businesses.

The most important of these acquisitions were Glencore Xstrata’s purchase of the Pacorini Group, Trafigura’s acquisition of North European Marine Services (NEMS), the Noble Group’s purchase of World Wide Warehouse Solutions (WWS) and Louis Dreyfus Commodities’ acquisition of GKE. Because of these ownership changes and a small number of subsequent deals, “the number of independent LME warehousing companies has accordingly steadily declined to around 35 per cent of the total as of July 2013,” according to Thomson Reuters. In this process, with the notable exception of the C.Steinweg Group, most of the remaining independent LME warehouse companies have become small and location-specific operations.

Second, the number of cancelled (and un-cancelled) warrants in the system was a good clue to the scale of the issue for a long time (note the difference between the blue and the yellow):

Indeed, as the Morgan Stanley analysts note:

The financing deals were physically manifested by growing concentrations of aluminium in Detroit and Vlissingen, zinc in New Orleans, and copper in Johor and Antwerp. Subsequently, cancelled warrants have blown out as customers scrambled to access metal. Exhibits 9-11 illustrate this trend by showing first the total inventory reported in all LME warehouses (blue line) – then that number less cancelled warrants (yellow line). Note that the rate of warrants being cancelled has slowed since the LME’s proposed changes in July, suggesting the panic over metal consolidation is easing.

Consolidated inventory created higher location premiums
Record high physical premiums for immediate metal delivery accompanied the surge in cancelled warrants. This was a very unusual phenomenon for metals that were already in significant supply surplus, such as aluminium and zinc – which prompted the plethora of consumer complaints in the last two years.

Third, all of this in many ways contributes to the scarcity amid plenty problem.

That is… inventory games are by and large confusing curve signals. And above all, it is the yield-related incentive to lock inventory away for significant periods of time which is creating a situation in which it pays to store and hoard, even if the curve moves into backwardation on occasion (since everyone’s storage and financing costs are not the same).

But also, and more relevantly, as the LME moves to unclog inventory and the financing trade unwinds, that we may end up in a situation in which the curve overreacts on the backwardation front, even despite ongoing surpluses in the market. Confusing and counterintuitive to say the least.

But as Morgan Stanley’s analysts explain this is all because of the potential impact on shorts (our emphasis):

In the shorter term, we think that there could be other countervailing effects on these generally bearish implications for prices. As the LME itself recognizes, if the industry believes that these proposals will be effective, they could result in greater speed of access to LME metal and an increase in cancellation of LME warrants, to obtain a scheduled delivery-dated position. This could paradoxically lead to queue lengthening in the short term, even if the longer-term impact of these new requirements would be to reduce queues to or below the 100-day Affected Warehouse threshold.

As the LME also recognizes, the potential on-exchange impact on LME price discovery will be greater than previous load-out rate proposals because on-exchange short position holders might be denied the ability to warrant LME metal at certain locations if the warehousing company feels the warranting will cause an unsustainable increase in the subsequent quarter’s delivery out requirements for that warehouse company in that location. This might well have the effect of artificially exacerbating backward-dated markets or creating a backward-dated market even where surplus market conditions might indicate a fundamental contango term structure as shorts seek to delay delivery on exchange. In short, this policy has the potential to create an unintentional artificial supply and demand dynamic in LME markets that would be at odds with longer-term results from these measures.

Furthermore, the proposed policy is almost certain to result in higher LME warehouse rents as a result of increased costs of compliance and business modeling required under this linked proposal.

Which leads us to wonder… if LME load-out rate proposals are enough to create something of a false backwardation/squeeze in metals — by simply denying shorts the ability to warrant metals when needed — could a similar phenomenon, perhaps one fuelled by rising financing costs rather than LME rules per se, be having an impact on shorts in other commodity markets as well?