Worstall’s Corollary

We require an update, a corollary perhaps, to Gell-Mann Amnesia. This is the observation by the American physicist that when we read a newspaper article on something we know about then we know — because we know about it we can observe that — the journalist is wrong. But then we turn the page and assume that the next piece, the subject of which we know not, is correct. That update needs to expand the observation to politics. Whenever the Lanyards do propose something in our own areas of expertise we know — because we know about it we can observe that — they’re about to micturate the wealth of the nation up against the nearest set of vertical bricks. But then we go on, too often, to debate the next policy as if it’s not obviously a gross waste of time, money and effort. 

Nominations are accepted, obviously, for the naming — and I’m not so modest that I’d object to Worstall’s Corollary. I do also regard this as a basic truth about politics and political decision making even if even I have not wholly and fully absorbed the point. Even I think that some governmental spending is not obviously ludicrous, so I still have work to do.

The spark for this particular version of the observation is something in my neck of the woods — rare earths. I’ve discussed the basic background here before, including the proof that I got this all right 15 years back last time China played hooky with the market. The essential point made there is that sure you can have a monopoly in something but if it’s contestable — people could, with effort, compete with you — then that doesn’t do you much good. As soon as you try to flex your monopolistic muscles then people will compete with you and you’ll lose the monopoly.

This is in fact happening, and it’s happening to every level of that rare earths business. Mines are opening, processing plants being built, separation plants, purification, magnet factories and all the rest. We’d all prefer not to have had to do this but we do, so we are. Some of that western (to use a single word for non-China) production will go on to make good profits too. There is new knowledge in the field, there are better deposits we didn’t know — or perhaps care — about before, interesting separation methods and so on. After all, one of the things we know about monopolies — especially state influenced ones — is that they tend not to be all that innovative. There are now decades of advances in other sciences to be applied to the rare earths world and they are being applied. I’ve even done my little bit in sparking a couple of them which is nice for my ego, even if not my wallet.

But this idea that government planning will always get these things wrong. We require a little technical knowledge here. A useful mental model — all models are wrong, but some are useful — is that chemistry is about the electrons in the outermost orbit around an atom, physics is about the nucleus where the neutrons and protons live*. The lanthanides (that bar along the bottom of the periodic table, La and then Ce to Lu) have the same number of electrons in that outer orbit and different numbers in inner shells and obviously different nuclei. Chemistry is therefore not all that hugely useful in separating them although it can, at expense and with difficulty, be done. One recent suggestion for a separation plant on Teesside was £200 million to build, then $15 a kilo material running costs. But it’s the physics of them that we are interested in. The fun way the different lanthanides colour lights for example, or for phosphors on screens, the very different magnetic properties across them and so on. So, difficult though it may be we do have to separate them into their individual elements, we very rarely use just a mixture (well, actually, we do, but that’s for lighter flints and that’s not a high value occupation, making lighter flints).

This separation process though. The way this works is that we have to take them out in order (again, models, useful) so we get the lanthanum first, the cerium, then the praseodymium, the neodymium, the dysprosium is number 9 on the list, terbium is 10. Those third and fourth, Pr and Nd, are very useful in magnets. 9 and 10, Tb and Dy, are essential for magnets that will work at higher temperatures. Great! But, as we’ve already said the chemistry is very similar, So all ores will contain all 15 of them. In varying quantities, to be sure, but we get all of them from any mine because natural processes have just as much difficulty separating them as we do. 

This means that in order to get the magnet ones we want we’ve got to process out the La and Ce first. We’ve said we need a £200 million plant plus $15 a kg to do this separation. But La and Ce currently sell for $1 a kg. We want lots and lots of those magnet ones later in the series, but that means having to produce those earlier ones to get to them. The truth of rare earth mining is that you make a huge loss on the “lights” and a good profit on the “heavies” (from the atomic numbers, La is 57, Ce 58, Nd 60, Tb 65 — getting heavier, d’ye see?). And, because nature is a bitch, there are always more of the lights in a mine or concentrate than there are of the heavies. 

But, and here’s the thing: different types of ore can be more or less light heavy (sorry). We, today, want more of the heavies than we do the lights – thus the price differences, La and Ce at $1, Dy and Tb at $350 and $1,700 a kg (prices from before the most recent excitements) — so we’d like to be mining the ores which are heavies heavy, not the light heavies. These do exist, too, globally, in a way that decades back we didn’t think was true. See above about greater knowledge accumulating while China was running their little contestable monopoly.

See above about greater knowledge accumulating while China was running their little contestable monopoly

This could be said to be a bit of a lead up but then explaining a global industry in 1,000 words is indeed possible and you’ve understood it all perfectly — which means you’re cleverer than politics and the political system. For, they’ve now grasped that rare earths are interesting, even that perhaps they might do something about them. At which point: The mine at the heart of Trump and Xi’s battle for superpower dominance

Gosh, really? A rural corner of Turkey is at the centre of a race between China and the US to secure critical rare earth minerals? 

That is interesting. So the superpowers are snarling at each other over who gets to be involved in developing the second largest — after one in Inner Mongolia — deposit of rare earths in the world, eh? This is going to be about superpower dominance, is it? My word. So, people with industry knowledge decide to look up this potential mine, this deposit:

Notably, the Beylikova deposit in Eskişehir is reported to contain all 17 rare earth elements. This deposit is particularly rich in light REEs such as lanthanum (La), cerium (Ce)

That is, this is heavy in the lights, the very type of ore we’re not, in fact, interested in anyway. But the superpowers are playing a dominance game over it because politics doesn’t grasp what you now do as a result of a thousand words from me. We want the heavy rich deposits — these are called ionic clays — and they’re found all over the place: Oz, Argentina, Brazil, all sorts of places. There are even some in the US.  

This superpower rivalry is two bald men fighting over a comb

This superpower rivalry is two bald men fighting over a comb. They’re arguing about who gets to mine the very form of rare earth ore we don’t care about. Even, in fact, we would prefer not to be mining at all. 

So, what do we call this? Worstall’s Corollary to Gell-Mann Amnesia works for me, but given the size of my ego, it would. We really do need some word or phrase for the proof that even when politics does get involved with something important it still gets it wrong. 

That this shows that the planned economy never will work is obvious but we’ve also other proofs of that contention. But, well, what do we then call this proof?

* This is a really bad model an awful lot of the time but it suits us just fine here.