This is a short squib, based on some notes for a training concept that I’m currently working on (I suppose I should say at this point to various readers: I mean, one that I’m working on in spare, snatched moments from my main task of whatever it is that I promised to you). It’s also slightly occasioned by this genuinely excellent piece by Charles Goodhart, which concludes:
Finally, the advent of the GFC made it abundantly clear that neither central banks nor economists fully understood the working of the economic system. We, central bankers and economists, no doubt have learnt many lessons as a result of the GFC. But, do we yet really understand enough about the workings of the financial system as a whole? Probably not! The very fact, as outlined above, that there are several narratives about the main causes of the GFC, indicate that there remain many uncertainties about the way our system works. We cannot be sure that we have learnt the right lessons; uncertainty remains endemic.
That’s certainly true. But … I very much doubt whether the kind of “understanding” that central banks are going to try to develop is really the kind of understanding that I want them to have. They’re going to go back to their existing models, include all sorts of “frictions” and “financial variables” in them, then solve the model, hold a few conferences, wait until someone in academia comes up with a reasonably tractable version and declare that to be the consensus. Maybe I’m misjudging them; maybe they’re going to throw all the existing models out and make new strides into agent-based modelling. Doesn’t really matter in my opinion — however good the model is, that isn’t the kind of understanding I’m looking for.
It would be somewhat better if the central bankers decided to leave behind their beloved economics degrees, and to “understand the financial system” in the sense of learning a lot of things about how it actually works. One of the great shames of the global financial crisis was not how the economics profession failed to see it coming, but how many professional analysts of the financial system needed a glossary to find out what a CDO was, or how repo haircuts worked. There’s certainly a lot of room for those in charge of the financial system to develop much better factual knowledge of the institutional arrangements and of legal and accounting issues, and more importantly to ensure that this knowledge is spread throughout the central banking institutions, rather than locked up in the minds of one or two “technical specialists”. But even that isn’t the kind of understanding I am thinking of.
The issue is this. I have spoken to several regulators who have helped to draft regulations relating to margin calls. They have written survey articles on margining practices, and economic articles on the theoretically optimal strategies for margin calls. But there is one sense in which I understand margin calls a lot better than they do.
I understand margin calls because about fifteen years ago, I was young and stupid and had too much spare money. So I opened a spread betting account. Within a couple of months, I was just young and stupid. Having the experience of receiving a message telling you that you have not only got things wrong and all your money is gone, but that you have got things so wrong that you have to send some more money and lose that too, is a particular feeling.
I want margin calls to be regulated by someone who gets it. Someone for whom the simple two-word phrase “margin call” produces a faint visceral reaction. Someone who understands that it’s not just a risk management tool, or a means of enforcing mark-to-market policies, or a stochastic variable subject to boundary conditions — it’s a margin call, and it’s one of the most frightening things in the financial world.
That’s why my modest and presumably entirely unworkable proposal is that as part of their training programme, everyone working in a financial supervisory authority ought to be required to open an account with a spread-betting company and to trade it actively, with position sizes in material amounts relative to their salary, for a period of eight weeks. That’s how you understand things.