Wednesday, September 21, 2011

My Supposed Hubris

I do macroeconometrics, if pushed to define one of my research areas. This is a classic area for criticism by those towards the right of the political spectrum, not least Arnold Kling. Kling has an essay entitled Science of Hubris, in which he carries out his usual attack on what he calls scientism, i.e. any attempt to put really precise numbers on things that we can't be precise about. In Kling's mind, we just can't be precise about any kind of economic aggregate, notably GDP, or investment, because there's just too much going into them, and these aggregates are made up of very different things too.

So essentially, he dismisses the entire field of applied macroeconomics because statistical agencies aren't able to add things up particularly well, and because, well, economists shouldn't be thinking in such broad terms; they should only be thinking at the microeconomic level. Those who practice is, apparently, are the very epitome of hubris; that is, to have "excessive pride or self confidence" as my Mac's Dashboard dictionary tells me.

Of course, anyone who puts their head up above the parapet is liable to accusations of hypocrisy, but folk like Kling at EconLog, and Boudreaux at Cafe Hayek are remarkable examples of precisely this. Kling and Boudreax (and others on their respective blogs) have remarkable confidence in one thing, and one thing only: The price mechanism in a fully free market. They are scornful and incredibly bitingly sarcastic (esp. Boudreaux) of those who dare to suggest anything otherwise, who dare to imagine that government intervention could possibly be a useful thing. To be that requires a heck of a lot of hubris, from what I can see.

And it's not limited to that; Russ Roberts and David Henderson at EconLog had a number of posts about Ideological Turing Tests, and how those on each side of a debate characterise the other, declaring (of course), that those on the right, those nearer to the libertarian school of thinking, were much better than those anywhere else on the spectrum at accurately describing the position of their opponents. Then just today, a post about confidence, where the application is not to Austrians with their remarkable over-confidence in the price mechanism in many inappropriate contexts but, yes, you guessed it, Keynesians (or at least, their crude caricature of them). Utterly stunning the amount of times I think about pot calling kettle when I read Hayekian/Austrians on their blogs.

But back to macroeconometric models. What I struggle to understand is this. If an econometric model is built that happens to include all the relevant explanatory variables for an aggregate variable (e.g. inflation) such that the residuals are white noise, then to all intents and purposes, that aggregate variable has been explained. A humble presentation of such a model would not make bold predictions about the future (forecasting is entirely different to macroeconometrics), but would instead make suggestions at what had been learnt from this exercise. But Kling and those in his camp would disregard it entirely.

Kling et al suggest that one of the most dangerous things about Keynes and his teaching was that he let loose governments and convinced the common man that there was intellectual rigour behind their own whims and desires. Equivalently though, via their scepticism of absolutely everything other than what they previously believed in, Kling et al promote an unhelpful atmosphere of scepticism through which genuine academic progress is hindered - all because of their ideology rather than any desire to be scientific in their pursuit of knowledge.


Post a Comment

<< Home