Microsoft Word Cheongaftaeffects doc

Table 2: The Effects of AFTA on ASEAN Imports by Country (2001 to 2003) 

 

  

Dependent Variable: ASEAN Country i's Import Value of Product k from Country j in Year t 

  

(1) (2)  (3)  (4)  (5) 

(6) (7) 

Explanatory Variables 

Brunei Cambodia

Indonesia

Malaysia

Philippines 

Thailand

Vietnam

1.970* 2.693**  0.265**  0.180  0.201* 

0.174** 0.002 

ln(Country's Preference Margin 

(in %) on Product k in Year t) * 

(D

NONASEAN

=1) 

(0.903) (0.842)  (0.090)  (0.109) (0.086)  (0.055) (0.117) 

0.770 4.039** 0.215+ 0.669** 

0.283+  0.220* 0.491* 

ln(Country's Preference Margin 

(in %) on Product k in Year t) * 

(D

ASEAN

=1) 

(1.198) (0.696)  (0.120)  (0.138) (0.152)  (0.099) (0.194) 

-2.051 -4.471** -0.080  0.099  0.267 

-0.095 -0.530** 

ln(Country j's GDP in Year t) 

(1.580) (1.034)  (0.457)  (0.273) (0.222)  (0.094) (0.176) 

year 2002 

0.398* -0.758** -0.050+ 0.081*  -0.021 

0.136** 0.151** 

  

(0.172) (0.220)  (0.027)  (0.037) (0.024)  (0.023) (0.058) 

year 2003 

  

-0.960** 

-0.198 

-0.064 

-0.092* 

0.189** 

0.326** 

  

  

(0.372) 

(0.135) 

(0.063) 

(0.039) 

(0.034) 

(0.069) 

Observations 5282 

6406 

69564 

81548 

66739 

149934 

33645 

Number of groups (by HS6 

classification) 

2641 2655  26848  29389 24443  54546 12783 

Estimation Method 

Fixed 

Effects 

Poisson 

Fixed 

Effects 

Poisson 

Fixed 

Effects 

Poisson 

Fixed 

Effects 

Poisson 

Fixed Effects 

Poisson 

Fixed 

Effects 

Poisson 

Fixed 

Effects 

Poisson 

Notes:  

 

 

 

 

 

 

 

1) Statistical significance indicated as: + significant at 10%; * significant at 5%; ** significant at 1% 

 

 

2) Reported standard errors are robust and calculated according to Wooldridge (1999). 

 

 

 

3) As these regressions are by ASEAN country, the GDP variable for Country i is subsumed into the year variables. 

 

4) The model for Singapore was not estimated because Singapore's preferential margins were all zero in the sample. 

 

 

 

 

 

 

 

 

---

 

  

14 

 

Table 3: The Effects of AFTA on ASEAN Imports by HS 1-digit Category (2001 to 2003) 

 

Dependent Variable: ASEAN Country i 's Import Value of Product k  from Country j  in Year t

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

Explanatory Variables

HS0

HS1

HS2

HS3

HS4

HS5

HS6

HS7

HS8

HS9

0.112

-0.390**

-0.037

-0.053

-0.141*

0.037

-0.028

-0.002

-0.077

0.057

(0.330)

(0.128)

(0.106)

(0.044)

(0.062)

(0.096)

(0.159)

(0.061)

(0.094)

(0.132)

1.379

0.010

-0.127

0.072

0.212+

-0.006

0.213

-0.104

0.283*

-0.275

(1.270)

(0.164)

(0.123)

(0.073)

(0.117)

(0.327)

(0.402)

(0.124)

(0.124)

(0.188)

-0.914*

2.778**

-0.619+

-0.673**

-0.269

-1.710**

-0.256

0.285

0.078

-0.510

(0.446)

(0.700)

(0.363)

(0.191)

(0.335)

(0.278)

(0.526)

(0.352)

(0.448)

(0.608)

-0.082

0.156*

-0.243*

0.100

0.271

0.007

0.299

0.006

-0.225

0.625*

(0.134)

(0.065)

(0.124)

(0.147)

(0.179)

(0.168)

(0.489)

(0.219)

(0.266)

(0.287)

year 2002

-0.000

-0.113+

0.021

0.182**

0.086*

0.173**

-0.034

0.037

0.080

0.121*

(0.253)

(0.066)

(0.064)

(0.029)

(0.037)

(0.036)

(0.079)

(0.047)

(0.058)

(0.060)

year 2003

0.574

-0.295*

0.239*

0.364**

0.186*

0.323**

0.064

0.155+

0.193*

0.236+

(0.466)

(0.145)

(0.102)

(0.054)

(0.075)

(0.064)

(0.126)

(0.094)

(0.098)

(0.143)

Observations

11987

11046

21118

44591

38235

46819

51008

51215

104409

32690

Number of groups (by HS6 

classification)

4481

4046

7718

16100

14020

17042

18988

18553

40143

12214

Estimation Method

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Fixed Effects 

Poisson

Notes: 

1) Statistical significance indicated as: + significant at 10%; * significant at 5%; ** significant at 1%

5) The sample period is 2001-2003.

2) For the Fixed-Effects Estimator, the dependent variable is the natural logarithm of 

3) For the Fixed Effects Poisson estimator, the dependent variable is the level of import 

4) All regressions exclude Laos, Myanmar, and Singapore. Singapore is not included 

ln(Country's Preference Margin 

(in %) on Product k  in Year t) * 

(D

NONASEAN

=1)

ln(Country's Preference Margin 

(in %) on Product k  in Year t) * 

(D

ASEAN

=1)

ln(Country i 's GDP in Year t)

ln(Country j 's GDP in Year t)

 

 

 

 

 

 

 

---

 

  

15 

 

 

                                                 

Endnotes 

 

i

 Although the initial AFTA framework agreements were general, senior officials formulated 

specific details over the course of the year and drew up three implementation documents: 1) The 

Operational Procedures for CEPT, 2) Rules of Origin for CEPT, and 3) Interpretative Notes to the 

Agreement on the CEPT Scheme for AFTA.  The AFTA approach of working out details after 

launching the project has been characterized as “Agree First Talk After”. 

ii

 For example, in 2000, Malaysia invoked the Protocol Regarding the Implementation of the CEPT 

Scheme Temporary Exclusion List and delayed tariff reductions on completely-built-up 

automobiles and automobile knock-down kits in order to protect its local auto industry. 

iii

 Joint Press Statement by the 5th. Meeting of the AFTA Council, Chiang Mai, Thailand, 

September 21, 1994. 

iv

 The AICO was intended to be an important feature of ASEAN economic cooperation. It was 

designed to encourage technology-based investments in ASEAN, and was open to any ASEAN-

based company meeting the following requirements: 1) incorporated in and operating in an ASEAN 

country, 2) a minimum of 30 percent ASEAN equity, 3) the company engaged in some form of 

resource sharing (such as sharing of technology, market sharing, or consolidated purchases of raw 

materials). Raw materials, intermediate products, and output of approved AICO projects would 

enjoy 0-5% tariffs immediately. Such products would also enjoy local content accreditation and 

non-tariff incentives. 

v

 However, given the annual publication of tariff rates, some updating may have taken place 

between the publication year and the year the tariffs came into effect. 

vi

 Manchin and Pelkmans-Balaoing illustrate this by showing that the import content (sum of direct 

and indirect inputs) of key manufacturing sectors in ASEAN in 2001 ranged from 31% (other 

manufacturing) to 67.8% (electronics). 

vii

 To benefit from preferences, ASEAN exporters are required to submit a cost analysis of the 

product to be exported with accompanying invoices for scrutiny and approval by the local trade 

ministries. If approved, the local trade ministry provides the exporter with Form D, which is a 

certificate that the product qualifies for trade preferences. The exporter may then export the good to 

another ASEAN country at the preferential rate, but the goods are still subject to random checks. It 

is likely that, instead of facilitating intra-regional trade, the process has turned out to be more time-

consuming and costly in practice relative to ordinary customs clearance. 

viii

 Intra-industry trade, which is the overlap of imports and exports within a product category, is 

classified as vertical if the unit value of imports in the product category differs by at least 15% from 

the unit value of exports in the same product category. 

ix

 The importer-exporter-product fixed effects also help control for all factors related to import 

decisions made prior to 2001. These could be, for example, preannounced tariff rates and procedural 

changes in customs or FDI flows into an industry that were planned before 2001 but implemented 

after.  

x

 Test results from these robustness checks are not reported but are available from the author 

([e-mail]).