This article was written by Steven Spencer. All views and opinions expressed are strictly his own.
The LME today (4th April) issued a notice on Nickel trading which included a point that will raise eyebrows in all corners of the metal industry. Point 8 states: “One aspect that the LME has already identified as having contributed to the situation was sizable positions in the OTC market and the LME’s lack of direct visibility of such activity”.
Of course, since the creation of LME Clear and the establishment of cash margining in the wake of the takeover by the Hong Kong Exchange, several brokers have found their ability to offer credit to clients while being bound to have their bank accounts automatically drained by LME Clear every time they executed on the LME an unworkable situation. The fact that LME Clear could do this via their Protected Payment System (PPS) instantly but not be in a position to return the cash on the same day, or on occasion not till 36 hours later, effectively prevented many brokers from trading at all once their capital had vanished.
The result was the departure of some Ring Dealing companies and also a change from Category Two to Category Four or Five status for the smaller brokers, who now funnelled their trades through the bigger financial institutions. The latter, although allowed more favourable terms on cash margining, began to look at LME Clear as an unnecessary burden on their cashflow. Why should they leave funds at a clearing house several levels of magnitude smaller than themselves?
While their trades had previously been cleared through the London Clearing House (LCH), this had a number of benefits. The larger clearing members tended to be active on other markets, allowing for a broad view of risk to be taken; it could fit within a single default fund structure that covered all the London markets (rather than being measured market by market) which is why LME Clear will now be doubling its default fund to $2bn. LCH’s operations allowed for economies of scale to influence the clearing fees, and in the LME’s case kept them at a much lower fee than a stand alone service would have allowed.
But, in line with most, if not all, markets the lure of cash flow and profitability accruing to the exchange rather than to a stand alone independent clearing house (which acted for profit but not for profit maximisation) was too attractive to ignore. LCH’s unique cross market offering, which was hugely in the clearing members’ interest, was broken up by commercialised exchanges, over which members had largely lost any real influence in the demutualised world.
The effect, of course, was to drive major broker/dealer banks and other financial institutions, backed by their substantial balance sheets, to create their own off market offerings to clients; structured trades, credit lines, complex option packages driven by algorithms – and all out of sight of the LME with no cashflow implications. Their contracts included the brand value of the LME, mostly written under the rules of the LME and including LME dispute clauses, and it was this that eventually led the LME to take a stand and demand a “copyright” fee for the privilege of using their brand. But it was not too onerous a cost and as we have seen, the LME seems not to have obtained the transparency it expected. Their visibility of over-the-counter business was very limited at best.
While the major LME metals such as Copper and Aluminium are generally traded in deliverable form, or at least in such quantities that any non LME-registered metal is insignificant in comparison to the overall volume, this cannot be said of Nickel, where, depending on the ore body, a large volume is produced as ferro nickel or nickel pig-iron, for example. With stainless steel the major demand, these two forms are the principal nickel constituents, but with their 5-35% nickel content, the value of the nickel at current high prices has to be hedged even though neither form is deliverable.
This brings us back to the current problem. In their drive to add volume to the exchange turnover the LME established their electronic trading platform and opened it to overnight trading. I’ve been thinking about this for a long time and it seems to me that the whole concept of electronic overnight trading is a disaster created by the drive for volume. The only volume between the close of Comex and the opening in London is frankly Chinese in one form or another. There is no meaningful two way market when the Chinese get the bit between their teeth. They generally either all buy or all sell. Allowing electronic trading (ET) overnight in that region in a very low liquidity market was asking for trouble, and it came in spades. Not having visibility of the OTC volume was adding fuel to the fire.
OTC trades offered by large financial institutions to producers, for example, can be structured as a mix of options that at the contract date have a delta (the volume that needs to be hedged) of practically zero. In a high liquidity market like copper, that’s a manageable risk as the price will move steadily in high volume. But in a low liquidity market like nickel trading electronically overnight, that delta can increase exponentially as the price moves violently, triggering more delta as the price moves rapidly through successive strike prices.
I’ve never been a fan of ET for a number of reasons. First, it has too many subtle complexities in the form of ghost orders, icebergs, algorithms, spoofing etc. The Flash Crash of 2010 should have been a warning to all markets and exchanges. Open outcry, “outdated” as many people believe, is still the best provider of transparent pricing and liquidity and what happened in nickel would not have happened in London daytime.
The biggest miss-step in the whole Select/etrading saga was the issue of electronic input into the system. At the point where orders are input, it should be done manually, rather than offering direct electronic access. That would – I think – prevent HFTs dominating the trade, since it would negate the advantages offered by the milliseconds on which their business depends.
The LME, as a low volume, low open interest, market providing a unique, essential service to the global base metals market is highly specialised and not particularly suited to HFT ‘bots.
HFTs offer no economic advantage to the overall metals business and the appearance of liquidity that they claim is false – their liquidity is largely inaccessible to the rest of the market, as it is mainly based on using computer programme speed to insert themselves into a trade that was going to happen anyway. This point was made repeatedly to various LME board members at the time, but on the whole they didn’t want to know. They were seduced by the volume…….well, ride the wind and reap the whirlwind.
The main argument about the coalescing of exchange and clearing is that, in governance terms, it will not always follow that the marketing aims of the exchange will match the risk management responsibilities of the clearing house. If, for example, the exchange is intent on listing a new contract with low liquidity and poor price formation, will an exchange owned Central Counterparty be in a position to refuse to clear: any clearing house needs ‘get out ability’ in the event of a default. Clearly the idea of common ownership of exchange and clearing facilities is less than optimal in such situations. This is a regulatory conundrum that FSA/FCA have never really tackled – perhaps post-Nickel they might?
