簽訂區域貿易協定對貿易的影響：以東協十加六為例

本篇論文主要探討區域貿易協定對貿易的影響，基於全球的區域經貿整合已成為主流，而台灣近期積極與東南亞國家發展經貿外交，故以東協十加六區域貿易協定作為研究對象，並以實證結果提出政策建議供為政者參考。本文的研究方法沿用Baier and Bergstrand (2007) 所提出之追蹤資料型態的結構引力模型，採用1990年至2018年東協十加六的出口貿易總值資料，將研究資料分為兩組樣本，一為東協十加六之16國樣本，另一為擴充至其他亞太地區共計43個國家的樣本，分別探討簽訂東協十加六區域貿易協定前後，以及有無簽訂此區域貿易協定對貿易流量的影響。經本文採取多項穩健性測試分析，包含在模型中加入不同貿易協定的虛擬變數、考慮貿易協定的落遲效果和控制貿易流量隨時間增長的趨勢等，藉由上述各項穩健性檢驗，皆可得到在東協十加六區域內的樣本中，簽訂東協十加六區域貿易協定之後，對這十六個簽訂國彼此之間的雙邊貿易流量相較於簽訂前有顯著的增進效果，由最適的模型估計結果可推估其可增加5.2%的貿易流量；而在擴大範圍(亞太地區)的樣本中，有加入東協十加六區域貿易協定的國家，相較於無簽訂此區域貿易協定的國家，則其與同為此區域貿易協定簽訂國之間則可增加3.87%的貿易流量。

關鍵字

區域貿易協定；貿易流量；東協十加六；引力模型；卜瓦松擬最大概似估計法

並列摘要

This paper aims to investigate the trade effects of regional trade agreements. As regional economic integration has become the mainstream worldwide, Taiwanese government recently has intended to develop deeper economic cooperation with Southeast Asian countries. As such, this study plans to investigate the trade effect of joining ASEAN plus six. We apply the panel-data version of “structural gravity” equation developed by Baier and Bergstrand (2007) to two sets of samples. The first one is the sample that covers 16 member countries in ASEAN plus six, and the other one is an expanded data to include totally 43 countries in the Asia-Pacific area. We collect the corresponding bilateral-trade-flows data for pairing countries in our samples over the 1990-2018 period. The main findings of this study are as follows: (1) for the 16-country sample, the regional trade agreement of ASEAN plus six has positive effect on trade flows inside the trade bloc. It is estimated that compared to the bilateral trade flows before singing the regional agreements, the bilateral trade flows between member countries gaining 5.2% after signing the regional trade agreement; (2) for the expanded sample, the estimate results show that the membership of ASEAN plus six can increase the bilateral trade flows with 3.87%, compared with those countries not signing this regional trade agreement. These results are robust to a battery of sensitivity tests, such as additionally considering different types of trade agreements, adding lagged effects of trade agreements, and controlling for the trend variables.