There Is an Old Joke About Economists, Keys, and Lampposts That Comes to Mind Here…
This is an interesting, if an Aesopian, article by Olivier Blanchard…
He says that we need five kinds of macroeconomic models. He then gives examples of the five kinds. The examples of four of the kinds track—foundational, policy, toy, and forecasting—and we do indeed need those four kinds.
Then we come to the fifth kind, which Blanchard introduces by saying that we need:
DSGE models. The purpose of these models is to explore the macro implications of distortions or set of distortions. To allow for a productive discussion, they must be built around a largely agreed upon common core, with each model then exploring additional distortions…
But two paragraphs down he writes, of all existing DSGE models:
The current core, roughly an RBC (real business cycle) structure with one main distortion, nominal rigidities, seems too much at odds with reality… the Euler equation for consumers and the pricing equation for price-setters…
What is he saying? That all existing DSGE models are worthless. We should throw them away, and start over with respect to building this fifth kind of model. We can call it “DSGE”, but it will not be what has—hitherto at least—been “DSGE”. In his view, it will have to have “nominal rigidities, bounded rationality and limited horizons, incomplete markets and the role of debt”—i.e., real rather than fake microfoundations.
The problem, of course, is that we cannot (yet) set out such a model in a sufficiently tractable form. Thus I think that one inescapable conclusion that we need to draw from Blanchard is that, for now, we need to keep using our four useful kinds of models—foundational, policy, toy, and forecasting. And I think a second inescapable conclusion we need to draw is that we should stop doing DSGE models which involve looking under the lamppost for the key where it is not. We need, rather, to disassemble the lamppost. And then some of us can use its pieces to try to build another lamppost, but this time locate it in a less silly and useless place.
Olivier Blanchard: On the Need for (At Least) Five Classes of Macro Models https://piie.com/blogs/realtime-economic-issues-watch/need-least-five-classes-macro-models: “We need different… five types…
…[of] general equilibrium models….
- Foundational models… the consumption-loan model of Paul Samuelson, the overlapping generations model of Peter Diamond…
DSGE models… to explore the macro implications of distortions or set of distortions… built around a largely agreed upon common core, with each model then exploring additional distortions…. The current core, roughly an RBC… seems too much at odds with reality to be the best starting point…. How close [should] these models… be to reality[?]… They should obviously aim to be close, but not through ad-hoc additions and repairs, such as arbitrary and undocumented higher-order costs introduced only to deliver more realistic lag structures…
Policy models. (Simon Wren-Lewis prefers to call them structural econometric models.)… For this class of models, the rules of the game must be different than for DSGEs. Does the model fit well, for example, in the sense of being consistent with the dynamics of a VAR characterization? Does it capture well the effects of past policies? Does it allow one to think about alternative policies?
Toy models… IS-LM… Mundell-Fleming… RBC… New Keynesian model…. [To] allow for a quick first pass at some question, or present the essence of the answer from a more complicated model or class of models….
Forecasting models. The purpose of these models is straightforward: Give the best forecasts…
All models should be built on solid partial equilibrium foundations and empirical evidence.