Must-Read: Olivier Blanchard: Further Thoughts on DSGE Models
Must-Read: Very smart from the extremely sharp Olivier Blanchard. But, as he knows well, I disagree. I am very impressed by the fact that the further an economist was from the gravitational pulls of the influence of Prescott, Lucas, and even Friedman, the fewer stupid and the more smart things the economist had to say about the post-2005 world. That seems to me to be worth registering, and to have implications for the enterprise:
Olivier Blanchard: Further Thoughts on DSGE Models:
I believe… there is wide agreement… [that]:
- Macroeconomics is about general equilibrium….
- For exploration and pedagogy… transparency and simplicity… toy models…. For forecasting… accuracy, and… statistical models…. For conditional forecasting… more structural models… [that] must fit the data closely and do not need to be religious about micro foundations.
- Partial equilibrium modelling and estimation are essential to understanding the particular mechanisms of relevance….
I see two propositions as more controversial:
- The specific role of DSGEs… is to provide a… formal, analytical platform for discussion and integration of new elements… a base from which to explore [issues]….
- The only way in which DSGEs can play this role is if they are built on explicit micro foundations… because any other approach makes it difficult to integrate new elements and have a formal dialogue…. One wishes there were a short cut and a different starting point. I do not think either exists…
As Olivier knows, starting from DSGE means that you have to (a) put the TFP residual on the right-hand side, (b) start with a forward-looking Calvo-pricing Phillips Curve, and (c) tie the slope of the IS curve to the intertemporal elasticity of substitution of a representative agent.
None of those bring you any gifts worth having. All of those lay traps into which your analysis might fall.
And is a formal dialogue with those who start by putting the TFP residual on the right-hand side really one worth having? After 35 years, what has resulted from such a dialogue?