Last 5 posts


30 July 2014

Benchmarks and Reference Prices

The topic du jour amongst the forensic element of the financial community, if I judge by the number of requests for explanations and reports I have been receiving, is the vexed question of benchmarks, reference prices and the different ways in which they can be established. Lawyers and accountants, in the private sector, and governments and regulatory authorities, in the public one, are currently much exercised by the need to be able to demonstrate the independence, transparency and honesty of the pegs on which we all hang the pricing of our financial products and services. 

The Catalyst

It’s not difficult to see what the catalyst for all this interest was - the revelation of the rigging going on around the setting of the LIBOR rate. With hindsight, that was an accident waiting to happen; let’s face it, if you are asked to indicate where you think you should be able to clear your overnight cash credit or deficit, the likelihood is that you will give an answer that is slanted in the direction of your book (or, if your own book is neutral, perhaps to help out your friends). At some point, that moves from being a view of the market into being a misrepresentation of reality. Add to that a situation where some banks were anxious not to imply that they were perceived as a poor risk by their peers (which would have implications for the rate they would be required to pay) and it’s  - always with hindsight - pretty obvious that it could all go horribly wrong. Incidentally, the CFTC have just put out a press release with extracts of some of the conversations between the rate submitters from certain banks; it’s an entertaining read, on one level, and a real indictment of the system and its participants on another.

Varying Styles

Metal markets use a number of different ways of establishing a price. The closest to the LIBOR method is probably what happens in some minor metals - a call from the publisher of the price, a request for details of your recent trades, take out the outliers, average the rest together, and bingo! the market’s got its reference. I can’t be alone in suspecting that if I were involved, I’d probably submit details of the trades that best helped the position of my book. But maybe I’m being unjust, perhaps a tad cynical. With the LME’s foray into minor metals, there is an alternative, at least for cobalt and moly. That hasn’t caught on, though, with the majority of the market still preferring not to use it. I think the reason why it has failed to excite the business is the perceived lack of volume. Open outcry is - in my view, anyway - a pretty effective way of establishing a price; but, to work satisfactorily, the volume and liquidity of the market have to be sufficient to deter manipulation. So it’s a fine line for cobalt, say; on the one hand, the ring-round pricing model can encourage biased submissions, and on the other, inadequate liquidity in the LME ring would risk potential serious aberrations. We should recall how long it took for LME aluminium prices to become dominant - so the door to cobalt may still be open. 

The position of the silver fix became untenable as soon as Deutsche announced their withdrawal - a closed fix, with only two members, would strain the belief of even its strongest advocates. But it was probably a flawed model anyway; closing the doors to outside observation is hardly going to inspire confidence in market users. I can recall, when at a bullion bank some years ago, how the silver dealer used to pick up his book at the relevant time in the morning, waltz off into a closed room, and then come back later with the day’s reference price just announced across Reuters. And that’s pretty much all anybody knew about what was going on. Anyway, that one is condemned to the dustbin of history, and silver heads off into the brave new electronic world in a few weeks. I would add that liquidity is an essential of an open electronic auction, as well. Let’s hope it’s there.

The fixes for gold and pgms differ from silver in that they are open - in other words, rather than the dealers talking to each other with no outside participation, there is effectively a cascade, disseminating the details of the fixes as they happen. Customers can place and withdraw orders at any time during the process, and the price is only declared when buying and selling is in balance. Just to emphasise, that balance can be changed during the course of the fix by those who actually place the orders. That seems like the basis of a pretty healthy model; being reviewed at the moment, at least in the case of gold, hopefully those conducting the review will maintain the inherent transparency of what is there.

So, the LME and open outcry. Undeniably it’s a more costly option, but it has the great merit of being conducted in the full glare of publicity and as long as sufficient liquidity is available, it’s difficult to see a significantly better alternative.

I’ve laboured that a bit, but it’s important to understand the differences in the way reference prices are discovered. What they have in common, though, is that they are all used to a greater or lesser extent in pricing physical deals. That’s why the lawyers, accountants and regulators are so interested in them. In contractual disputes if the basis for the price is universally accepted as fair and binding, then the contract has a solid base. Once that starts being questioned, then other things are also called into doubt. That’s why the LIBOR scam was so important.


But there’s another set of numbers which seem to me less reliable, in a way. Those are ‘benchmarks’, as opposed to reference prices, which is what I have described so far. So, for example, there is the Codelco copper premium, or the Asian benchmark for copper concentrates. But they are qualitatively different, in the sense that they are not (theoretically) independently established. Thus Codelco publish their basis premium for the year and other copper producers benchmark themselves against it, setting their own level at so much below Codelco’s number. But that is effectively the flexible part of the price - the LME price is established publicly and broadcast by a ‘disinterested’ body; the premium level, although theoretically ‘benchmarked’, is a matter for individual contract negotiation. It all works swimmingly until there is a dispute - then the lawyers try to pin down the ‘real’ benchmark, only to find it’s a moveable feast.

So there, I think, is the difference between reference prices and benchmarks. The former are an essential tool in the functioning of the market, the latter - well, useful in defending performance vis-a-vis colleagues or competitors, but perhaps less so when it comes to contractual commitments. It’s the integrity of the former that is vital for the industry to guard. How that is done is a debate which is going to run on for quite a while, with the gold fix being the next subject (or target?).

The CFTC Press Release containing examples of misconduct from written communications can be found at: 

comments powered by Disqus