Tony Yates hits the nail on the head. But he should have hit a stake…

In a blogpost about the new economics curriculum and the magnificence of his own ideas John Kay (member of the advisory board of the INET economics curriculum commission) states:

Yet much of the “heterodox economics” the Manchester students suggest including is flaky, the creation of people with their own political agenda, whether Marxist or neoliberal; or of those who cannot do the mathematics the dominant rational choice paradigm requires. Their professors reject the introduction of these alternative schemes for the same good reasons their science colleagues would reject phlogiston theory or creationism

In a twitterstream about this blogpost Tony Yates (@t0nyyates) however hits the nail on the head (quote, not out of context):

“so much in economic theory unobservable. Utility, TFP, discount factor, elasticities, even capital“.

Which is right (well, not entirely, we do actually measure TFP (Total Factor Productivity) and capital, see below a little about capital). But Yates seems happy with this situation. Which he shouldn’t be. 

And he shouldn’t be happy as economic statisticians do measure lots and lots of variables – variables eagerly used by heterodox economists. Maybe the data are not measured in a perfect way – but that’s another discussion. Look here for the Diane Coyle book about GDP, which states that, as GDP changed over the years the way it was and is measured changed, too. But we do observe economic ()historically contingent) variables like unemployment, inflation, the stock of debt, the flow of lending and borrowing, the amount of money, capital and whatever. Yes, capital too, look here and here , though the last link is actually not about ‘capital’ but about ‘land and other non produced resources’ which means that the ‘costs of production approach’ can’t be used to estimate its value – we should not mix up solar panels with solar radiation. And we spend hundreds of millions doing this. The most glorious achievement of post war macro-economics is, head and shoulders, the improvement in macro-economic measurement. But students of economics are not taught how this about this and if it’s done in the right way. The (admittedly: ultra boring) manuals who explain how the science of economics estimates its variables are not part of the curriculum – to the detriment of the status of academic economics as a science.

Students are, however, still taught about outdated unobservables like ‘utility’. So, who is teaching and writing about ‘phlogiston’? People like Kay and Yates, who base their ideas on ‘unobservables’, or somebody like the late Wayne Godley, who wrote about observed fact. Or the very alive Steve Keen? Or Ha-Joon Chang? Or economists like the ‘heterodox’ Georgists, who promote the idea that ‘produced capital’ and ‘land and other non-produced resources’ are analytically and empirically distinct concepts (just like Ricardo already stated, by the way)?

And no, this is not naïve empiricism – economic measurements very much dependent on the theoretical ideas embedded in the concepts and operationalisations. Here you can find a not so flaky recent blogpost by J.W. Mason about this. Which means that we should our students exactly this – economic measurement is a difficult science and an art and you can’t be an economic scientists if you do not have a deep knowledge about the process of measurement of this moving target. And we have to confront them with this wisdom again and again and to teach them that we have to muddle through with whatever imperfect data we have – but we should not, never construct models which are unobservable to their core. Kay and Yates, however, do not seem to be aware of these problems, and seem more than happy with unobservables and the fuzzy models based upon phlogs. Talk about flaky…