Anders Rahbek hædres af Nykredits Fond
Professor Anders Rahbek modtager Nykrdits Fonds..
Blondinens hævn v. Shu-bi-dua
Handlingen i Blondinens Hævn kan ikke bruges som argument - Shu-bi-dua er mere troværdig. Dimensioneringsreformen er måske en god løsning.

Debate: Searching for optimal policy not knowing the true model

Jeppe Druedahl
Af Jeppe Druedahl
31 maj 2012 16:14 - i kategorien Fagligt - #Kritiske Politter  #David Colander 

Critical Students of Economics (Kritske Politter) have invited David Colander, Professor of Economics at Middlebury College, to discuss some of his recent contributions to macroeconomic methodology.

In the first year we learn that there is a clear distinction between apositiveand anormativestatement. A positive statement is about how the economy is and about which effects different actions imply. A normative statement is about how the economy should be and, given the positive models developed by economists, what policy to implement. But the financial crisis has clearly shown that we do not know the true model – at least there is no agreement among economists. This fundamental uncertainty implies that the choice of models have a non- negligible normative impact.

In Critical Students of Economics we have invited David Colander, Professor of Economics at Middlebury College, to discuss this problem with us and other interested students. In recent articles he has argued that policy should not directly follow from models; it must follow from reasoned analysis which the use of models, but combined with institutional knowledge, intuition, and common sense. Among the issues we will discuss is the methodology in choosing and evaluating assumptions.

Another interesting claim made by Colander that will be discussed at the meeting is that economics should focus on solving imprecise real-world problems, and only secondarily relating that solution to abstract mathematical problems such as solving stochastic general equilibrium models. He holds we should rely more on the use of rules of thumb with large fudge factors, and should be open to using partial equilibrium models. Could this also be a more suitable way of integrating the complexity, heterogeneity and possibility of dynamic learning which we observe in the real world, but which is so hard incorporate in mainstream general Walrassian equilibrium models?

Practical info:

Time: 6th of June, 16:15 – 17:30.

Place: CSS 7.0.34.

No need to sign up. Everybody is welcome!

0 kommentarer

Tak for din kommentar!
Skriv venligst en kommentar der er længere end 5 tegn

Skriv en kommentar

Log ind for at kommentere - eller opret en bruger