Estimating the Benefits of Trade Reform: Why Numbers Change

Much attention has been paid to the World Bank’s analyses of multilateraltrade reform. According to our latest estimates, full liberalization of worldmerchandise trade would increase global income in 2015 relative to the baseline by $290 to $460 billion, with higher percentage gains for developing countries (0.8 to 2.0 percent) than for high-income countries (Anderson and others 2005). These numbers are significantly lower than earlier World Bank estimates (World Bank 2001, 2003), which put the global gains of full merchandise trade reform at some $400 billion per year in 2015. This note examines why the numbers of have changed and, more generally, why estimates of gains from reform can vary widely. The answer lies in a mixture of factors: new data on tariffs, incorporation of recent major reforms in trade policy (notably in China), inclusion of preferential trade arrangements, and new poverty elasticities with respect to growth. Beyond these methodological issues, it is important to distinguish the scenario under analysis—whether it is full liberalization (the basis of all calculations and the benchmarks of this note) or partial reforms; whether the analysis is static or dynamic (and includes a productivity response); and whether the scenario includes service