Third-country effects on the formation of free trade agreements ¬リニ

a r t i c l e i n f o The recent proliferation of free trade agreements (FTAs) has resulted in an increasingly complex network of preferential trading relationships. The economics literature has generally examined the formation of FTAs as a function of the participating countries' economic characteristics alone. In this paper, we show both theoretically and empirically that the decision to enter into an FTA is also crucially dependent on the participating countries' existing FTA relationships with third countries. Accounting for the interdependence of FTAs helps to explain a significant fraction of FTA formations that would not otherwise be predicted by countries' economic characteristics. The international trading system has experienced a dramatic increase in the number of free trade agreements (FTAs) in recent decades. Fig. 1(a) shows that new FTAs went into force every year during the period of 1991–2005. In 2004 alone, eighteen new FTAs were established. A parallel development is the increasing number of FTA partners for each country (Fig. 1(b)). In 1991, each nation had on average 1.8 FTA partners. In 2005, the average had risen to 9.9. In this paper, we examine how existing FTA relationships affect countries' incentives to form new FTAs. Previous studies have generally viewed the decision to enter into an FTA as a function of the participating countries' economic characteristics alone (e.g., market size, production cost, and distance), ignoring any potential effect of existing FTAs. Our analysis shows that a country pair's incentives to establish an FTA with each other depend crucially on their existing FTA relationships with third countries. We first develop a three-country theoretical model to highlight the importance of third-country effects. In this model, we examine how the incentives of a country pair to enter into an FTA with each other vary depending on whether the two countries already have existing FTAs with the third country. We begin with a benchmark " no-FTA " case in which neither country in the pair has an FTA with the third country. This benchmark case is then compared with two alternative scenarios: (a) a " one-FTA " case in which only one country in the pair has an FTA with the third country, and (b) a " two-FTA " case in which both countries in the pair have FTAs with the third country. 2 This comparison enables us to show how existing FTAs influence a country pair's decision to establish …