For once, a senior economist has had the humility to admit how little economists know. If only there were more. Nobel Laureate Sir James Mirrlees gave a talk at the ANU http://crawford.anu.edu.au/events/content/video/?year=2012&id=2341 which goes a long way to exposing the truth that the conventional economics Emperor has no clothes, or at least not the kind of clothes that cover up the important parts of the anatomy. His talk is notable for its rare honesty, the kind of honesty that becomes possible, presumably, after you have won a Nobel prize and career success is assured.
Mirrlees starts by identifying the intersubjective nature (that is collective perception) of economic valuation. Valuation is not, and cannot be, objective. He compares it to a competition to identify the most beautiful face, decided by what most people thought was the most beautiful face. If everyone thinks the face is the most beautiful, then it is the most beautiful.
That is what economic valuation typically is, he says. So called “fundamental” analysis is just the “lowest sensible price”. That is why forecasts don’t work, because they purport to be objective. They cannot track intersubjectivity.
Mirrlees then makes a somewhat startling claim. The conventional economic model (including the theory of General Equilibrium):
“doesn’t, for example, give any reason for the existence of money.”
Hmmm. That is a pretty big “example.” It tells us why economics is so poorly placed to tell us anything about what is happening in the conventional capital markets, let alone the massive amount of meta money (what Mirrlees describes as “those funny assets called derivatives”) that are piled on top of the conventional debt and equity markets.
What Mirrlees is wrestling with is a basic issue that faces economics. The discipline is not a science, neither can it be objective. So what is it? My best guess is that it is a specialised language for talking about transactions. It possesses the weaknesses and strengths of many specialised languages. It can be extremely precise in ways more general idioms cannot. But it necessarily entails assumptions about what it is describing that cannot be examined from within. That invariably leads to narrowness or bias.
The narrowness in economics is that it can only really measure repetitive transactions related to the exchange of goods and services. So it is well equipped to measure things like GDP. And, as Mirrlees comments, it has the basic metric of the interest rate to value that. So it is reasonably effective in tracking the relationship of interest rates to supply and demand patterns with GDP (hence that being the focus of central bankers; it is what they can measure).
But as soon as we are in the intersubjective world of asset valuation, economics starts to founder, because the transactions are not regular enough. That is why central bankers struggled so much with house price inflation in the lead up to the GFC. Because houses are not exchanged as often as consumer products, the transactions are more intermittent, it is necessary to employ more notional values. Such notional values are harder to measure, so the easier thing to do is to ignore them.
Mirrlees comments that it is possible to value property by discounting future rents or shares by discounting future dividends. But any speculation on property values is outside economics purview and if companies decide to retain earnings rather than paying dividends then economics likewise falls over. In other words, as soon as the transactions become unpredictable, the whole thing falls over. Currencies, he says “you have to say the fundamental value is zero”. The reason being that currency valuation is purely intersubjective. Currencies can only be valued in relation to other currencies.
“I’m not a fan of standard economic theory,” he concludes, with glorious understatement.
One is reminded of the joke about the fellow who lost his keys in the dark. But he spends his time looking for them under a streetlight. When asked why he is doing it when he knows the keys are not there he replies: “But I can see what I am doing here.”
Mirrlees makes a further comment about derivatives that is worth noting.
“Disallowing the derivatives can go a long way. I don’t understand why that has not been done already.”
It has not been done, I suspect, for two reasons. One is that the same people who have benefited from the derivatives and other meta money debauches have either had their hooks into government or actually been in government. It is nicely detailed in the documentary Inside Job http://en.wikipedia.org/wiki/Inside_Job_%28film%29
Another reason is that regulators have been trained, or are advised by people trained, in the standard economic theory whose inadequacies are described by Mirrlees. When faced with the perils of meta-money, the $700 trillion of derivatives or the increasingly absurd speed of high frequency trading, they have no framework to deal with it.
Two issues must be addressed. First, the relationship between intersubjective “conventional” transactions such as buying or selling shares, property and currencies, and the development of computer models that are deployed to exploit that intersubjectivity. Thus, traders exchange shares, conventional buying and selling. Then the meta money traders come in and shift the values by using derivatives (futures, warrants etc) usually with the help of complex mathematical models. The same with high frequency trading, which is purely computerised.
What is the right way to think about this relationship? I believe the best starting point is to see it as a system of signs being interpreted at two levels. Conventional traders try to read it as a mix of real value and perception. Meta traders try to read it as pure perception (high frequency trading, for instance, is purely about reading behaviour, the underlying value of the asset is irrelevant.)
The second issue is the problem of infinite regress. With conventional trading, the number of trades is limited: by the number of shares on offer, the number of properties on the market, even the amount of a nation’s currency on issue. Because there is scarcity, it makes sense to see them as some sort of balance between supply and demand, although it is a very different balance to that which applies in consumer product markets because it involves shared perception.
But in the world of meta money there are no limits. Derivatives can be constructed forever. The only limit seems to be getting into shorter trades than nano-seconds.
This is self evidently dangerous and little has been learned. Some better understanding for regulators could be derived by looking at the system as cybernetic: a system of signs. Even more advantage could be gained from imposing limits on traders, before they send the system into a tailspin, an infinite regress that threatens money itself.