Geographical characteristics of countries plays a major role determining the level at which a country participate in international trade. Geographical location of a country is a factor that influences trade openness. Island is one of the geographical characteristics that we are interested in. Island variables have been used as a control variable in the gravity model. Despite divergent findings of island variable to trade, there has been no meta-analysis to resolve the diverging results. Following the Protocol outlined in Meta-Analysis in Economics Research-network (MAER-Net), we meta-analyse 48 primary studies with 628 diverse island coefficients collected from studies published between 2000 and 2009 inclusive. We investigated if being an island country will generate a positive or negative effect on international trade. Despite the island variable has been mostly used as a control variable, it still suffers from publication bias. Regardless of publication status, (i.e., whether peer-reviewed or not), the literature suffers from severe negative publication bias. The genuine effect size of being an island to trade for all studies was positive, with a magnitude of 0.217 and 0.223 for the peer-reviewed studies. Our findings show that island countries trade more due to their connectedness and cheap transport availability via sea.

, , , ,
Demena, Binyam Afewerk
hdl.handle.net/2105/55990
Economics of Development (ECD)
International Institute of Social Studies

Langat, Evans Kipkurui. (2020, December 18). The impact of being an island on trade: a meta-analysis evidence from gravity models. Economics of Development (ECD). Retrieved from http://hdl.handle.net/2105/55990