THE ASEAN–CHINA FREE TRADE AGREEMENT: POLITICAL ECONOMY IN INDONESIA

Abstract

The ASEAN–China Free Trade Agreement has been the most consequential trade agreement for ASEAN in the past decade. In this article, I use partial equilibrium modelling to estimate the direct impacts of the agreement on Indonesia and China, and from these estimates draw political economy implications for Indonesia. Although the agreement has contributed to a modest trade surplus for Indonesia overall, it has led to a larger bilateral deficit with China. In addition, the shifts to surplus for Indonesia have mostly been in resource-based sectors, while the shifts to deficit have occurred in many manufacturing sectors that the government would like to see grow. I argue that pushback against the agreement has contributed to a resurgence of non-tariff trade barriers in the country, although other political economy forces also have been at work. The agreement ultimately provides a cautionary tale: cutting regional import tariffs can lead to pressures for more complex and less transparent trade policies.

Perjanjian Perdagangan Bebas ASEAN-Cina (ASEAN–China Free Trade Agreement) merupakan perjanjian yang paling berdampak bagi negara-negara ASEAN pada dekade yang lampau. Dalam tulisan ini, penulis menggunakan model ekuilibrium parsial untuk mengestimasi dampak langsung dari perjanjian tersebut pada Indonesia dan Cina. Hasil estimasinya kemudian digunakan untuk menelaah implikasi ekonomi politik perjanjian tersebut bagi Indonesia. Meskipun perjanjian itu telah berkontribusi sedikit terhadap surplus perdagangan Indonesia, secara umum ia menambah defisit perdagangan Indonesia terhadap Cina. Selain itu, pergeseran ke arah surplus perdagangan bagi Indonesia sebagian besar terjadi pada sektor-sektor berbasis sumber daya, sementara pergeseran ke arah defisit terjadi pada banyak sektor industri pengolahan, yang justru ingin dikembangkan oleh pemerintah. Penulis berpendapat bahwa dampak yang negatif dari perjanjian tersebut telah ikut mendorong munculnya kembali hambatan-hambatan perdagangan non-tarif di Indonesia, walaupun kekuatan ekonomi politik lainnya juga ikut berperan. Perjanjian tersebut memberikan pelajaran bahwa pemangkasan tarif impor regional juga dapat membuka jalan bagi munculnya kebijakan-kebijakan perdagangan yang lebih kompleks dan kurang transparan.

Keywords:

• ASEAN
• China
• free-trade agreements
• political economy

JEL classification:

• F13
• F15

Notes

1. The SMART model was originally developed by Laird and Yates (1986). Jammes and Olarreaga (2005) offer a useful presentation of the algebra of the SMART model. Full details of the implementation used in this article are available from the author upon request.

2. This makes sense, in the absence of information on how much lower the prices of imports of the various products would have to be before imports started being demanded in positive quantities. In principle, such a problem would also arise in a general equilibrium (GE) model. With the broader commodity aggregates of a GE model there may not be zero imports, but then the problem would be the greater extent of aggregation bias than in a more disaggregated model. Zero imports also create dilemmas for gravity models.

3. This net change is calculated relative to the import demand curve. Part of the customs revenue lost to the tariff cuts is in effect transferred to domestic consumers.

4. One effort to produce a global NTM database is by Kee, Nicita, and Olarreaga (2009). Their estimates of the tariff-rate equivalents of non-tariff import restraints on sugar and rice in Indonesia, at only 0% and 1% respectively, are highly questionable: Marks and Rahardja (2012) offer much higher estimates, in a study of Indonesia alone.

5. Hillberry and Hummels (2013) critique the approach on which Broda, Greenfield, and Weinstein's analysis and its precursors are based.

6. Subtracting trade creation and trade diversion from the actual level of imports from the country being granted preferential status could cause its baseline imports to become negative. For other source countries, removing trade diversion caused by the 2010 ACFTA preferences would make the level of baseline imports higher than the actual level of imports, and so the non-negativity constraint would not bind.

7. In this case, trade diversion is set as the smaller of that calculated in the model and the initial level of imports from a particular source country, to avoid imports becoming negative, as shown in equation (13) of Jammes and Olarreaga (2005). This is the only algebraic difference between my analysis and the SMART model, which instead modifies the functional form for trade diversion in order to prevent imports from being negative, as shown in equation (14) of Jammes and Olarreaga. The authors note that this modification causes a downward bias in the calculation of trade diversion.

8. In forming these weighted-average tariff rates as of 2010, and in estimating customsrevenue impacts for Indonesia and China, I accounted for the following preferential trade agreements as well as the MFN tariff schedules. For Indonesia: the ASEAN, ASEAN– Korea, ASEAN–Australia and New Zealand, and ASEAN–India free-trade agreements, as well as the Indonesia–Japan Economic Partnership Agreement. No other agreements were as yet in effect for Indonesia. In some cases, official preferential tariffs, including those under the ACFTA, have been higher than MFN rates, but in these cases Indonesia applies the lower of the two rates, according to Ministry of Finance officials, and this is the rate that I use. For China: the free-trade agreements with Bangladesh (under the Asia-Pacific Trade Agreement), Chile, Hong Kong, Macau, New Zealand, and Pakistan that are included in the TRAINS trade-policy database from the United Nations Conference on Trade and Development. I omitted certain regional agreements with minor trading partners and narrower commodity coverage.

9. In a simple sensitivity analysis, if one assumes an elasticity of substitution among import source countries of 2.5 for all products (more than the default value of 1.5 in the SMART model), the effects on trade flows are generally smaller, particularly for trade diversion. For example, for China the substitution elasticity estimates of Broda, Greenfield, and Weinstein (2006) range from 1.3 to 108.2, and an elasticity of 2.5 is typically on the low side. This will have a direct effect on trade diversion. It also affects trade creation and trade diversion indirectly, however, because the baseline import estimates are affected by the magnitude of the trade diversion that is initially removed from the actual 2010 import values. The size of these baseline values then affects subsequent trade creation and diversion as the final ACFTA tariffs are applied. Full details of the calculations using the alternative elasticity of substitution are available upon request.

10. Marks and Rahardja (2012) characterise trade and related policies in Indonesia as of early 2008, and quantify their impacts.

11. The inflation adjustment is done using the US producer price index, from the IMF's International Financial Statistics database. The trade data are from the UN Comtrade database.

12. In a 2013 visit to Indonesia, Chinese President Xi Jinping proposed the establishment of an infrastructure investment bank that would help fund development projects in Southeast Asia. The progeny of this proposal, the China-directed Asian Infrastructure Investment Bank, is expected to begin operations by the end of 2015.

13. For instance, in a later public forum, a member of the national legislature from the Golkar political party stated that the minister had ‘prioritized the interests of her ancestors’ in an aircraft purchase from China in 2011 (Jakarta Post, 21 May 2011).

14. Sri Mulyani became a managing director of the World Bank Group in June 2010, and Mari Pangestu was nominated by the Indonesian government in December 2012 to succeed Pascal Lamy as director-general of the World Trade Organization, but was not among the finalists for the position.

15. In May 2013, however, President Yudhoyono appointed Muhammad Chatib Basri, a well-regarded younger economist with a PhD from The Australian National University, as minister of finance, after Basri's predecessor became governor of Bank Indonesia.

16. More complete data that support these characterisations are available from the author upon request.

17. The indices of the nominal and real effective exchange rates (2010 = 100) show the value of the rupiah against a basket of the euro, US dollar, Japanese yen, Chinese renminbi, Malaysian ringgit, and Thai baht, formed using geometric total trade weights. The real index uses consumer prices in the respective countries.

18. Quoted by Burke and Resosudarmo (2012) on the basis of a Bank Indonesia press release on measures being taken to address the trade imbalance.