Exporters’ Response to FTA Exporters’ Response to FTA Tariff Preferences: Tariff Preferences:

Evidence from ThailandEvidence from Thailand

byby

Archanun KohpaiboonArchanun Kohpaiboon

Faculty of Economics, Thammasat UniversityFaculty of Economics, Thammasat University

IssuesIssues

The proliferation of free trade agreements (FTAs) The proliferation of free trade agreements (FTAs) implies that FTAs would engender far-reaching implies that FTAs would engender far-reaching implications, not only for the philosophy and operation implications, not only for the philosophy and operation of the multilateral trading system, but also for the day-of the multilateral trading system, but also for the day-to-day conduct of cross-border trade.to-day conduct of cross-border trade.

While the number of FTAs notified to the World Trade While the number of FTAs notified to the World Trade Organization (WTO) jumped to 462 by February 2010, Organization (WTO) jumped to 462 by February 2010, half of them are in the Asia and Pacific region, the half of them are in the Asia and Pacific region, the center of global trade dynamism.center of global trade dynamism.

So far there has not been any systematic analysis of So far there has not been any systematic analysis of trade-flow effects of FTAs because of the limited access trade-flow effects of FTAs because of the limited access to administrative records of FTA implementation. to administrative records of FTA implementation.

Two approaches are used to examine the FTA effect on Two approaches are used to examine the FTA effect on trade.trade.

1.1. Gravity Model Gravity Model with a binary dummy variable to with a binary dummy variable to distinguish between FTA members countries distinguish between FTA members countries from non-members. from non-members.

2.2. Firm Survey Approach, some advances to inter-Firm Survey Approach, some advances to inter-firm econometric analysisfirm econometric analysis

The gravity model approach ignores the ‘conditioning The gravity model approach ignores the ‘conditioning effects’’ of RoOs by implicitly assuming that tariff effects’’ of RoOs by implicitly assuming that tariff concessions offered by FTAs are readily available to the concessions offered by FTAs are readily available to the exporters. Such an assumption is rather restrictive: The exporters. Such an assumption is rather restrictive: The few available studies of the actual utilization of FTA few available studies of the actual utilization of FTA concessions suggest that the actual utilization rates differ concessions suggest that the actual utilization rates differ considerably among FTAs.considerably among FTAs.

While useful, the firm survey approach is subject to the While useful, the firm survey approach is subject to the major shortcoming of limited sample sizes, e.g. major shortcoming of limited sample sizes, e.g. Takahashi & Urata (2009) -1,688 Japanese firms; Takahashi & Urata (2009) -1,688 Japanese firms; Wignaraja et al. (2010) -221 firms Thai firms; Yunling Wignaraja et al. (2010) -221 firms Thai firms; Yunling (2010)-226/436 Chinese firms. (2010)-226/436 Chinese firms.

Against this backdrop, this paper sets out to examine Against this backdrop, this paper sets out to examine the response of exporters to tariff concessions offered the response of exporters to tariff concessions offered under signed Free Trade Agreements (FTAs) of under signed Free Trade Agreements (FTAs) of Thailand with a view to informing the debate on how to Thailand with a view to informing the debate on how to design ROOs and administrative procedures for design ROOs and administrative procedures for enhancing the effect of FTA-induced trade. enhancing the effect of FTA-induced trade.

In order to indicate the response of the private sector to In order to indicate the response of the private sector to

FTA tariff preference, FTA utilization (FTAU), the FTA tariff preference, FTA utilization (FTAU), the ratio between the administrative records and actual ratio between the administrative records and actual trade, is calculated. trade, is calculated.

The calculated FTAU is further used as the dependent The calculated FTAU is further used as the dependent variable in an variable in an inter-product panel-data econometric inter-product panel-data econometric analysisanalysis in order to gain a better understanding of the in order to gain a better understanding of the patterns of FTAU across products. patterns of FTAU across products.

Why Thailand?Why Thailand? Thailand is suitable as a case study of this subject for Thailand is suitable as a case study of this subject for

three reasons;three reasons; From the new millennium onwards, Thailand has From the new millennium onwards, Thailand has

been enthusiastic in doing FTAs with countries been enthusiastic in doing FTAs with countries around the world. around the world.

Administrative records for FTA implementation of Administrative records for FTA implementation of Thai exporters are available for the period 2003-08. Thai exporters are available for the period 2003-08.

Thai exporters have the potential to utilize tariff Thai exporters have the potential to utilize tariff concessions offered by FTA because the Thai concessions offered by FTA because the Thai manufacturing sector is relatively broad based, manufacturing sector is relatively broad based, compared to neighbouring countries. compared to neighbouring countries.

Empirical ModelEmpirical Model

, ,i j tFTAU FTA utilization of industry i, FTA j at time t

,, i j

FTAi jt t = tariff margin, the difference between applied rate and preferential rate of

industry i, FTA j

, ,i j tBLI = Backward Linkages index based on Leontief-inter-industry accounting

framework of industry I, FTA j at time t

, ,2 3 , ,

, ,

2i j ti j t

i j t

FTAUBLI

BLI

,i tFOR / = Output share of foreign firms in industry i at time t

,i tCON / = Output share of Thai conglomerate firms in industry i at time t

, ,i j tINT = Initial export value averaged in the past three years of industry I, export

destination j, at time t

,i tPC = Share of parts and components in total export of industry i, at time t.

Result (RE.Tobit Estimation Eq 5.4)Result (RE.Tobit Estimation Eq 5.4)

*** *** *** *** ***

*** *** ***

2, , , , , , ,

2.73 31.95 9.10 9.81 9.82

, , , ,58.14 9.47 4.83

0.03 0.71 0.17 0.1 0.07

0.01 0.09 0.04

4,184.

FTAi j t i j i j i t i t i t

i j t i t i t

FTAU t t BLI BLI FOR

INT CON PC

Log likelihood

272; 5350.67;# 21,097Wald obs

Cost of Compiling ROOCost of Compiling ROO

Re-estimate Equation 1 above in different sub-samples Re-estimate Equation 1 above in different sub-samples according to the tariff margin (according to the tariff margin (XX).).

Search for the tariff margin X that makes the statistical Search for the tariff margin X that makes the statistical significance of alpha 1 disappears.significance of alpha 1 disappears.

2, , 0 1 , , 2 , 3 , 4 ,

/

5 , , 6 , 7 , , ,/

FTAi j t i j i j i t i t i t

i j t i t i t i j t

FTAU t t BLI BLI FOR

INT CON PC

ConclusionsConclusions The product coverage in FTA trade is limited. Products The product coverage in FTA trade is limited. Products

that have benefited FTA tariff preferences so far are that have benefited FTA tariff preferences so far are highly concentrated. highly concentrated.

Our econometric analysis suggests that rules of origin Our econometric analysis suggests that rules of origin constraints are binding empirically. Both ability to constraints are binding empirically. Both ability to comply with ROO as well as tariff margin do matter in comply with ROO as well as tariff margin do matter in firms’ decision to use FTA. firms’ decision to use FTA.

The estimated cost in compiling ROO is The estimated cost in compiling ROO is equivalent to 5 and 8 per cent tariff. equivalent to 5 and 8 per cent tariff.

The FTA impact on export is conditioned on trade The FTA impact on export is conditioned on trade volume during the pre-signing FTA.volume during the pre-signing FTA.

So far it has been predominantly local firms, in So far it has been predominantly local firms, in particular large local conglomerates, which utilize FTA particular large local conglomerates, which utilize FTA tariff concessions, compared to foreign firms and small tariff concessions, compared to foreign firms and small and medium enterprises (SMEs). and medium enterprises (SMEs).

Products traded under global production sharing are Products traded under global production sharing are less likely to apply for FTA tariff preference because of less likely to apply for FTA tariff preference because of both dollar cost incurred by ROO imposition and both dollar cost incurred by ROO imposition and limited tariff margin. limited tariff margin.

Policy InferencesPolicy Inferences It is unlikely to promote export by maximizing a It is unlikely to promote export by maximizing a

number of FTAs signed. The nature of FTA partner number of FTAs signed. The nature of FTA partner does matter.does matter.

A FTA alone is unlikely to be an universal bullet to A FTA alone is unlikely to be an universal bullet to open up markets for products yet to be traded. open up markets for products yet to be traded.

If there were no trade records in pre-FTA period, it is If there were no trade records in pre-FTA period, it is unlikely for FTAs to generate fresh trade between FTA unlikely for FTAs to generate fresh trade between FTA partner. partner.

For those who believe in FTAs (not me), policy emphasis For those who believe in FTAs (not me), policy emphasis to harness the trade-induced effects of signed FTAs to harness the trade-induced effects of signed FTAs should be on reducing costs incurred from the presence should be on reducing costs incurred from the presence of ROO. of ROO.

There is room for inter-government cooperation to There is room for inter-government cooperation to mitigate any cumbersome obstacles preventing firms mitigate any cumbersome obstacles preventing firms from making use of FTAs.from making use of FTAs.

Examples of Rooms for improvement Examples of Rooms for improvement Cost structure DeclarationCost structure Declaration Self-issuing?Self-issuing? Problems at the border of importing countriesProblems at the border of importing countries

Figure 1

Flow Chart for firms to Receive Certificate of Origin

Source: Developed by Author

Exporters/

Representatives

Bureau of Imports, Department of Foreign Trade, Ministry of Commerce

Provide necessary documents to prove goods origin; (1) Manufacturing Process (2) Cost Structure

Exporters/

Representatives

Bureau of Trade, Department of Foreign Trade, Ministry of Commerce

STEP 1

STEP 2 Fill c/o request (either manual/electronics together with other documents such as reference numbers, shipment information

Issue c/o within a day and a half or less

Receiving the reference number within

5 working days if submitted documents

are satisfactory.