The Dark Side of the Generalized System of Preferences

The Generalized System of Preferences (GSP) was established to facilitate economic growth and development of low-income economies by fostering their exports to industrialized countries. As the design of these schemes is rather complex and controversial, we test \u85ve hypotheses regarding the e ¢ ciency of GSP. We employ an extensive dataset covering almost world trade and estimate a gravity model using Poisson maximum likelihood estimation. We nd that GSP might foster developing countries exports in the short-run (H1), but GSP seems to cause economic distortions over time that deter these exports in the long-run (H2). Additionally, GSP might be used by GSP granting countries to promote their own exports (H3). However, their exports also decrease in the long-run due to the distortions of GSP recipientseconomic structure (H4). Additionally, we \u85nd that GSP receiving countries with an advanced development stage generally seem to bene\u85t from GSP (H5). Taken 1 together, our empirical results imply that non-reciprocal arrangements such as GSP should be substituted by reciprocal agreements under GATT/WTO rules