Recent Posts

Monday, March 27, 2017

Latest Blowup Demonstrates Weakness Of DSGE Macro

The post "No Criticising Economics is not Regressive" by "Unlearning Economics" has once again re-awakened the macro wars. I am agnostic on the big picture story about economics in general, but the austerity corner of the debate highlights an obvious weakness of DSGE macro in particular: the mathematics that is used has not achieved what it set out to achieve.

Professor Brad DeLong was critical of a lot of the claims made by "Unlearning," I am not the person to discuss those topics with. I would summarise the original argument as: free market driven economics has caused a lot of problems. However, I would note that wishing that the other half of the political spectrum would disappear is a common premise for op-eds, but it is hardly helpful.

Whether such claims about economics have merit depends upon what you think is happening to Western Civilisation. If you believe that we are a declining civilisation, courtesy of feckless elites, you pretty much have to accept that economists are part of the problem. However, if you think the outlook is less dire, you can come away with a better view of our economist class. That debate is well out of scope of this article.

However, if we focus on DSGE macro in particular, there is no way of looking at what Brad DeLong wrote, as well as this article by Professor Simon Wren-Lewis, and not conclude that the basic premise of the approach has failed. The use of "sophisticated" mathematics was supposed to bury these purely verbal arguments, and replace them with statements about the mathematical properties of the macro-economy.

What happened? The pro-austerity forces won the political battle after the Financial Crisis, relying on slipshod spreadsheet analysis, as well as City Economists drawing charts with scary lines on them.

If DSGE macro had a plausible model of fiscal policy, they could have easily blown the other side of the water. But such models were noticeable by their absence.

DSGE models had one job -- and they failed.

I was trained as an applied mathematician. We did not adjudicate on the correctness of proofs by forming committees from the right schools; you just followed the logic, and see whether it works. The necessity of asserting authority is a sign that economics has taken a very wrong turn.

(c) Brian Romanchuk 2017


  1. I still think the best strategy is to ring fence these people and let them carry on talking amongst themselves.

    We're just better off building the modelling from scratch using system and control techniques with domain knowledge input from the better parts of social science and the real world.

    1. As I wrote on the previous article, I don't see the value of writing about mainstream macro. To what extent it comes up in debates that I discuss, I need to cover it.

      However, this latest argument was just too ridiculous to ignore. If DSGE macro was even one-tenth as useful as its practitioners said, they would have crushed those austerity arguments.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.