Binding Overhang and Wave-like Expansion of Free Trade

• In Event: Networks and Policy on a Global Scale

Abstract

Why does trade liberalization proceed in fits and starts manner? Why do many countries devote much of their efforts to negotiating preferential trade agreements (PTA) while maintaining their commitments to global free trade? Under what conditions do countries alter their free trade policies? Joining recent theoretical and methodological development, this study elucidates and tests a co-evolving network model of FTA formation to reconcile the inconsistency between policy commitments and observed practices in countries’ FTA approaches. By clarifying why countries generally seek out partners with high binding overhang to extend PTA as a way to grow free trade benefits at present in order to induce initially uncooperative neighbors to liberalize their trade barriers in the future, we show, combining evidences from change-point and network analysis, how free trade can proliferate and why such phenomenon tends to occur sporadically.

Existing scholarships are divided over the effects of PTA on trade liberalization: the “stumbling block” perspective views PTA as locking in intra-member trade that hinders their paths to global free trade, while the “building block” argument contends that PTA generates competitive trade liberalization pressure from third countries that expands free trade to the rest of the world. Empirical observation since the Uruguay Round also shows that free trade expansion often occurred in short bursts.

To reconcile these competing claims, we argue that pro-free trade countries can achieve further trade liberalization by bypassing uncooperative geographic neighbors to negotiate PTA with more cooperative partners, a strategy we term “partner switching”; in so doing, the enlarged free trade benefits emanating from these “hub” countries can induce initially uncooperative partners to liberalize their trade barriers in order to exploit the former’s growing free trade benefits, thus spurring a new round of bilateral PTA that expands the scope of free trade, a mechanism we call “strategy-updating.” In line with recent literature that sees binding overhang (the depth between bound and applied tariff rates) as inversely related to a country’s import market power, we show that pro-free trade countries tend to prioritize partners with binding overhang greater than themselves to negotiate PTA as this allows them to leverage their market power to elicit trade liberalization from the latter – the more intense the “partner-switching,” the larger the free trade benefits they can accumulate. Because the proliferation of PTA reduces the equilibrium binding overhang for the rest of the world, it becomes less costly for initially uncooperative partners to liberalize their trade barriers, resulting in the expansion of free trade until the gap in binding overhang is insufficient to sustain such dynamics. Yet contrary to conventional wisdom, the model also suggests that trade liberalization, particularly in the form of inter-regional PTA, is most observable during global economic recession that generates additional push to divert trade flows to new markets where higher binding overhangs have hithertofore hindered bilateral trade.

Using WTO PTA and UN Comtrade data, we test the empirical implications of the above claims. Our Bayesian change-point analysis detects four structural breaks in global PTA network between 2000 and 2019 during which global PTA activities gradually morphed from three regional clusters (Europe, America, and Africa) into two inter-regional clusters (Asia-Pacific, Europe-Latin America), with Eastern European and Latin American states saw the largest growth in the probabilities of being connected to PTA in the periods after the 2008 financial crisis. We estimate the longitudinal changes in PTA network topography as a function of country-dyadic gap in binding overhang using temporal exponential random graph model (TERGM), controlling for bilateral trade flows, shared third country PTA ties and international organizations memberships, regime types, and other dyadic and monadic attributes. The result suggests that, conditional on previous state of the PTA network, the “strategy-updating” effect is most pronounced on current network formation through the generation of new PTA ties before and immediately after the 2008 financial crisis, but are insignificant in periods when global economy resumed its normal state under which regional PTA clusters may become the stumbling block to global free trade. The finding is further corroborated by evidence from community detection analysis that unveils the clustering of highly-connected free trade partners from different regions after 2008.

This study contributes to the trade liberalization literature by clarifying the mechanisms that generate the observed phenomena as well as identifying the conditions under which PTA is most likely to promote (hinder) global free trade.

Authors

• Hsiang-Chuan Sha, National Taiwan University

• Huan-Kai Tseng, National Taiwan University

• Hsuan-Wei Lee, University of North Carolina at Chapel Hill