The goal of this thesis is to present a meta-analysis of studies that are focused on the relation between the international trade flow and the trade costs. The effect of trade costshasbecomeoneofthekeyelementstoresolvethesixmajor puzzles in the bilateral trade flow. I examine 1,090 estimates of the trade cost effect reported in 58 studies, codify 51 aspects of study design that may influencetheestimates.Iuse meta-regression analysis to investigate why trade costs effects vary. The results suggest that different methods and mainly data characteristics systematically affect the estimated trade costs effects. I find evidence for publication selection bias by using the appropriate tests. The authors of primary studies tend to report preferentially large estimates of the elasticity of trade with respect to trade costs. The evidence for publication selection bias is stronger for studies reported in peer-reviewed journals thanfor unpublished studies.

No description available.

No description available.

Magister Legum - LLM

Magister Legum - LLM / South Africa