Agricultural Trade Policy Modelling: Insights from a Meta-Analysis of Doha Development Agenda Outcomes

In a meta-analysis of trade policy models, Hess and von Cramon-Taubadel (2008) use over 5800 simulated welfare effects from 110 studies of potential Doha Development Agenda outcomes to identify characteristics of models, data and policy experiments that influence simulation results. This meta-analysis, which is recapitulated here, produces plausible results and explains a significant proportion of the variation in simulated welfare effects. However, due to insufficient documentation and the complexity of the general and partial equilibrium models in the literature sample, many explanatory variables employed in this analysis are binary. This precludes more detailed analysis of their impacts across models. Therefore, a partial equilibrium model and a single country CGE for Canada are employed to generate synthetic meta-data. Simulation scenarios are based on random combinations of base data, elasticities and tariff changes selected from plausible ranges obtained from the literature sample. The synthetic meta-data has the advantage that the values of explanatory variables are measured exactly. This makes it possible to explore more complex issues of functional form and interaction between variables in the meta-analysis. The results indicate for both models that first- and second-order polynomials provide sufficient approximations of the model response. Especially in the CGE model, interaction terms between elasticities and policy variables are important. We conclude that meta-analysis can provide insights into the behaviour of trade policy models beyond what is possible with conventional sensitivity analysis and qualitative reviews.


Issue Date:
2008-05
Publication Type:
Working or Discussion Paper
PURL Identifier:
http://purl.umn.edu/43466
Total Pages:
25
Series Statement:
CATPRN Commissioned Paper
CP 2008-02




 Record created 2017-04-01, last modified 2017-07-06

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)