Insider critiques of neoclassical macro models
Paul Romer has just published a devastating critique of DSGE (or, in his parlance, ‘Post Real’) macro models. He’s not the first important insider to write an article like this. Look here for Paul Krugman, ‘How did economists get it so wrong‘. Look here for Willem Buiter, ‘The unfortunate uselessness of most ‘state of the art’ academic monetary economics’. Look here for Charles Goodhart, ‘Whatever became of the monetary aggregates‘. And look here for the insider of insiders, Olivier Blanchard, ‘Do DSGE models have a future‘ (His analysis: NO! His conclusion: yes). Especially Krugman, Buiter en Goodhart are extremely eloquent and their pieces are a joy to read.
Two questions: is there a common denominator to these fierce critiques? And does ‘your humble narrator’ have something to add?
The answer to the first question: yes, there is. All authors mention a contempt for reality. All authors mention obfuscating math. And the impossibility to ask questions about monetary instability when even ‘monetary’ models do not model money (and debt). My summary: the models in question disable economists to analyse economic reality instead of enabling them to do this.
Do I have something to add? Yes. Of the authors above, Romer is clearest (not the same as: clear) about the fact that a scientific paradigm does not only need theory but also needs a matching system of measurement, though Goodhart also clearly mentions (in 2007!) that not paying attention to the monetary aggregates (money but, in his view, also credit and debt) was quite a mistake. For quite some time, economic measurement and theory developed more or less in tandem. Veblen’s best students became the heads of organizations, the National Bureau of Economic Research and the Bureau of Labor Statistics and developed the Flow of Funds statistics. Keynes himself established the British Office of National Statistics (yes, the present day ONS), as he needed national accounts data which were not available. In the Netherlands, Tinbergen established the Centraal Planbureau, a ‘fiscal watchdog’ with a strong emphasis on empirics. Look here and here for more information about this. Instead of at least trying to match these efforts and directly measure the variables they model, like ‘the natural rate of interest’, ‘natural unemployment’ and ‘utility’ they chose to assume that these variables are emergent properties not of the economy – but of the models (read Romer). My point: earlier generations of economists did a much better job and did estimate variables consistent with their ideas and models. Which made their efforts scientific. Let’s stand upon their shoulders!