Abstract. This paper quantifies the extent of preferential trade as a share of total world trade in different regions of the world and for two periods. Results show that: i) preferential trade represented 40% of world trade in the period 1988-1992 and it slightly increased to 42% during the period [1993][1994][1995][1996][1997]; ii) during the second period, agricultural products generally benefited more from the existence of preferential trade agreements than industrial products (maybe due to GATT-exemption); iii) the regional distribution of preferential trade is relatively uneven with a significant share of preferential We are indebted to Sanoussi Bilal, Patrick Low and Jaime de Melo for helpful comments. We are also grateful to Mariana Gagliardi and Daniel Morales from SISD-WTO for providing us with information on existing preferential trade agreements and to Philip Rauh for patient help with data collection. All errors are ours. The views expressed here are those of the authors and do not necessarily reflect those of the institutions with which they are associated.University of Geneva Some authors provide such estimates, but only for a particular region, as Sapir (1998) for the European Community. However, to provide an estimate for the share of preferential trade in world trade using this method faces the constraint that customs statistic are often difficult to obtain, if not confidential in many countries.The aim of this paper is to quantify the share of world trade that occurs under preferential trade agreements (PTAs) using a parallel method. The approach we followed is to proxy the actual share of trade that enters under preferential treatment by considering only imports within a PTA that take place under tariff lines where the MFN tariff is higher than 3 percent.The idea is that if the MFN tariff is below 3 percent, then incentives to satisfy rules of origin, and therefore benefit from preferential treatment vanish. For preferences under the 3 Generalized System of Preferences (GSP) we only included imports where there are full preferences (i.e., duty free) and no quantitative limitations. We identify PTA by using WTO notifications to article XXIV and the enabling clause by members and accession countries (e.g., China).Shares of preferential trade were calculated using two different averages: an import-weighted average of countries in the sample which indicates the "true" share of preferential trade in the world or region, and a non-weighted average, which indicates the extent of preferential trade in the "representative" country. Hence, the import-weighted average naturally tend to give more importance to large trading countries, whereas the non-weighted average gives more importance to small trading countries.