Microsoft Word Cheongaftaeffects doc

---

 

 

 

Running Head:  AFTA Effects Disaggregated

 

 

 

 

 

The Effects of AFTA:  

A Disaggregated Analysis 

 

David Cheong† 

Johns Hopkins University, SAIS Bologna Center 

This Version: August 2008 

 

 

 

 

 

 

 

 

†

Johns Hopkins University, SAIS Bologna Center, Via Belmeloro 11, 40125 Bologna, 

Italy. Tel: [phone], Email: [e-mail] 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

---

 

 

 

 

 

Abstract 

 

 

 

 

In this paper, using a gravity model, I study changes in trade patterns of ASEAN  

(Association of South-East Asian Nations) at the Harmonized System (HS) six-digit level 

in the period 2001 to 2003. These three years saw the acceleration of trade liberalization 

among ASEAN countries as they moved towards the ASEAN Free-Trade Area (AFTA).  

To avoid inconsistent estimates due to a log-linear specification of the gravity model, I 

use the Fixed Effects Poisson Quasi-Maximum Likelihood estimator. The estimates show 

that: in general, ASEAN preferential margins had a trade-creating effect at the product 

level; the majority of ASEAN countries benefited significantly from this trade creation; 

and there was net trade creation in the HS4 and HS8 product categories. These results 

suggest that ASEAN trade liberalization in the early 2000’s had positive welfare effects.    

 

 

 

 

 

JEL Codes: C23, F15

 

 

 

Keywords AFTA, ASEAN, free trade area, gravity model, panel data

 

 

---

 

 

 

1

 

Introduction 

 

The number of preferential trading arrangements (PTAs) has increased fourfold since the  

early 1990’s (World Bank, 2005). This proliferation has occurred in parallel to, and 

perhaps out of frustration with, attempts at multilateral trade liberalization within the 

World Trade Organization. The fact that there are fewer participants in a PTA facilitates 

reciprocal reduction in tariffs and non-tariff barriers and provides a framework for 

coordination of trade-related activities such as customs operations and transportation 

networks within the PTA. By creating a more integrated trading area, a PTA could have 

other positive effects like higher foreign direct investment. As a commitment mechanism, 

a PTA could also improve the ability of member countries to reform their domestic 

policies and enhance the overall economic climate. 

 

As shown by Viner (1950), the discriminatory nature of a PTA creates one major pitfall. 

The margins of preference that PTA members enjoy could induce trade diversion, when a 

PTA member substitutes its imports from non-member countries with imports from 

within the PTA. Trade diversion creates a negative welfare effect because imports are not 

sourced from the most efficient supplier-country. In contrast, a PTA could also induce 

trade creation, when a member country substitutes its domestic production with imports 

from member countries. This creates a positive welfare effect because PTA members 

alter their production structures to fit better with their respective comparative advantages. 

 

In this paper, I use a gravity model to assess the impact of trade liberalization according 

to the ASEAN Free-Trade Agreement (AFTA). The main research question is whether 

preferential tariffs have created or diverted trade. Most empirical studies of the effects of 

preferential trading arrangements are conducted with aggregate data. The innovation in 

this paper is to conduct the analysis at the product level. By doing so, I am able to avoid 

the aggregation bias in estimating gravity models as noted by Anderson and Wincoop 

(2004). The rationale for studying the period 2001-2003 is that ASEAN countries, in 

response to the shocks of the Asian Crisis, decided in 1998 to accelerate tariff reductions 

towards AFTA, thus creating a natural experiment for studying the trade effects of 

AFTA.  In the next section, I describe the evolution of AFTA and the trade liberalization 

provisions within this ASEAN agreement. In the third section, I discuss the phenomenon 

of production sharing in the ASEAN region and its impact on intra-regional trade. In the 

fourth section, I describe the empirical model used for my assessment of AFTA and 

discuss the results. In the last section, I provide some conclusions and policy 

implications.