Measuring Convergences and Divergences in APEC RTAs/FTAs : A Text-mining Approach

In this paper, I suggest that text mining analysis of regional trade agreements (RTAs) can be a suitable methodology to develop a tangible measure of convergence between RTAs. By utilizing well-established text similarity concepts in related literatures, I attempt to investigate how much RTAs in APEC are converging in terms of how much similar RTA texts are. Furthermore, which areas of RTA converge more or less will be examined. The main results of the study are as follows. The RTAs signed by APEC members are gradually converging over time, and they converged (in terms of 5-gram Jaccard similarity) at an annual average of 8% for all RTAs (both inter- and intra-regional) while 9.7% for intra-regional RTAs only. The areas that converged the most are service and transparency chapters, which show 2.2 times and 1.6 times higher level of convergence than the average, respectively. The objective and intuitive indicators of regional norm convergence are expected to provide a common understanding for setting goals and strategies for regional economic integration in the future