Abstract:
Quantification of welfare changes due to trade liberalization plays a crucial role for political decision making. However, significant differences in simulated gains from liberalization do not serve to increase confidence in quantitative assessments based on trade models. A meta-analysis of trade simulations under the WTO Doha Round is employed to identify model characteristics that influence the magnitude of simulation results. Findings from a simple regression model are plausible and show that each simulation experiment represents a complex interaction of model characteristics and experimental settings that may not easily be communicated to nonexperts. Meta-analysis proves to be useful for empirically assessing this complexity. Copyright 2007 International Association of Agricultural Economists.