Effective rates of protection revisited for Indonesia

Abstract

This paper calculates nominal and effective rates of protection for Indonesian tradables sectors in early 2008, and compares these figures with previous calculations for 1987 and 1995. Such a review is overdue. Many non-tariff barriers to imports and exports have been abolished, though new import restraints on rice and sugar are notable exceptions to this trend. Import tariffs have been lowered, particularly through regional preferential trade arrangements. We account for such arrangements in two different ways. Export taxes persist in certain natural resources sectors, but most rates have been reduced. We find that more than half of the effective support provided to tradable products sectors now comes from subsidies on fuels, fertiliser, electricity and liquefied petroleum gas, rather than from trade policies per se. Duty drawbacks and exemptions for exporters boost the effective rate of protection for tradables sectors overall by a small fraction of 1%, and for no input–output sector by more than 3%.

Keywords:

• ASEAN Free Trade Area
• trade liberalisation
• preferential trade agreements

Acknowledgements

This work was initiated in 2007 while Marks was consultant to the World Bank, Jakarta, in the Trade Research and Development Agency, Ministry of Trade, Republic of Indonesia. The views in this paper are those of the authors alone, and not necessarily those of the World Bank or the Ministry of Trade. The authors are grateful to Pomona College for research support, to George Fane and Sam Laird for helpful discussions, and to participants in an August 2007 seminar in Jakarta and an October 2007 seminar at the Australian National University for their comments.

Notes

¹For an imported item, the border price is the price of the item just before it enters customs. For an exported item, it is the price of the item just after it exits customs.

²The latter calculation is based on comparison of domestic producer prices with border prices, and thus takes precedence over the official duty rate: if the percentage price differential exceeds the official duty rate, then evidently the non-tariff measure is the binding restraint on imports; if it is less than the official duty rate, then evidently leakages of some sort have reduced the effectiveness of the import duty.

³This assumes that the export tax rate is applied to an external market price rather than a domestic price of the product to calculate the amount of tax per unit. It is consistent with the Indonesian practice of setting directive export prices (harga pedoman ekspor) based on external prices.

⁴Output weights are also distorted by trade and other policies, though presumably less consequentially than trade weights.

⁵George Fane noted in a personal communication that price comparisons such as were used in his earlier studies of Indonesia can often lead to negative implied rates of protection, even for imported items subject to import tariffs or non-tariff measures. One reason is that many prices are not posted, but must be negotiated by the buyer, and the seller understands that the researcher does not intend to buy the product. Also, small distribution margins in relatively low-wage countries like Indonesia may push down domestic prices relative to their foreign counterparts.

⁶The alternative method distinguishes between import-oriented and export-oriented sectors, much as FPC did, except that we use global trade weights instead of output weights. Details of the alternative calculations are available upon request from the authors.

⁷For institutional details on Indonesian trade policies, World Trade Organization (2007) is a useful reference.

⁸The Asian Development Bank's Asia Regional Integration Center has a useful database on the status of the regional trade agreements of a number of countries; see <http://aric.adb.org/FTAbyCountryAll.php>.

⁹Marks (2011) uses this approach to examine multiple tariff schedules in the Lao People's Democratic Republic in 2008.

¹⁰This assumes away cross-price effects among products that could lead to a significant reduction in trade in the product.

¹¹To obtain the broader product aggregates from IO sectors, we use the same IO-sector free-trade value added weights that we use below to aggregate ERPs, using the Balassa (1982) method to estimate value added per unit under free trade, and accounting for all the policies examined in this paper. The use of the free-trade weights reduces the weight on heavily protected sectors such as rice, and boosts it in sectors subject to export restraints. Our mixed approach, in which trade (in this case, import) weights are used to aggregate tariff lines to the level of IO sectors, but IO sectors are aggregated to broader sectors using free trade value added, seemed a reasonable way to minimise biases in our presentation, given particularly that tariff rates are more uniform within than among sectors.

¹²This held for 186 tariff lines for ASEAN–China trade, 67 for ASEAN–Korea trade and six for the CEPT program (the latter for rice and sugar lines only, as noted earlier). Indonesia's MFN rates were reduced for a number of products in 2008. For about 80% of the tariff lines for which Indonesia's 2008 ASEAN–China and ASEAN–Korea tariff rate commitments were above its 2008 MFN rates, those commitments were at or below its 2007 MFN tariff rate levels.

¹³Exporters are also allowed drawbacks for value added taxes paid on imported inputs, but in this paper we will not account for value added taxes in any way. Marks (2005) applies an effective value added tax rate analysis to Indonesia based on the framework of Gottfried and Wiegard (1991). We also do not account for sales taxes on luxury goods. Most of the goods targeted are durable capital goods, and in any case these taxes would have a negligible impact on costs of production in the economy.

¹⁴Muir (1986) provides a useful overview and evaluation of the duty drawbacks at the time of their introduction.

¹⁵Minister of Finance Decree 129/KMK.04/2003. For only one IO sector theoretically eligible for duty drawbacks, LNG (sector 105), is 125% of exports more than total output. We do not treat this constraint as binding. As table 6 below shows, however, the duty facilities provide a negligible benefit to the sector anyway.

¹⁶The appendix is available at <http://rspas.anu.edu.au/~gfane/Indon%20trade%20policy_Appendix.PDF>.

¹⁷Refiners gain from export taxes on crude palm oil, but are harmed by export taxes on refined products such as cooking oil.

¹⁸Under Minister of Finance Regulation 94/PMK.011/2007, the export tax rate schedule that applied for the various palm oil products depended on the level of a palm oil reference price, which was formed as a composite of the prices of these products in international markets. In January 2008 the reference price was high enough for the highest export tax rate schedule allowed by the regulation to apply. The reference price, set on a monthly basis, was given in Minister of Trade Regulation 48/M-DAG/PER/12/2007. The export tax scheme was revised further in February 2008, with an additional higher tax rate schedule for the highest reference prices. Palm oil products are included in IO sector 56.

¹⁹Minister of Finance Regulation 92/PMK.02/2005. The hides are included in sector 81.

²⁰Minister of Finance Regulation 51/PMK.02/2006. Unprocessed rattan is included in sector 30, wood chips and processed wood in sector 84 and veneers in sector 85.

²¹The sugar import policy regime was established through a series of decrees of the Minister of Industry and Trade between 2002 and 2004, culminating in Minister of Industry and Trade Decree 527/MPP/Kep/9/2004. Stapleton (2006) offers a useful analysis of sugar trade policies during this period.

²²The seasonal ban was initially established by Minister of Industry and Trade Decree 9/ MPP/Kep/1/2004. Minister of Industry and Trade Decree 357/MPP/Kep/5/2004 subsequently allowed the ban to be extended at the discretion of the Minister of Industry and Trade, in consultation with the Minister of Agriculture.

²³Minister of Trade Regulation 12/M-DAG/PER/4/2008. (The former Ministry of Industry and Trade was split into the Ministry of Industry and the Ministry of Trade in 2004.) Imports by Bulog remained subject to a seasonal ban, and the authority for Bulog to import medium-quality rice in practice required a cabinet decision or a recommendation from the Food Stability Team in the Coordinating Ministry of Economic Affairs.

²⁴Sugar cost parameters come mainly from the Ministry of Trade and from sugar importers. Parameters for rice were kindly provided by George Fane as part of the research for Fane and Warr (2008). The rice import barriers affect sectors 1 (field rice), 57 (milled, polished rice) and 59 (other flour). The sugar import barriers all apply to sector 62.

²⁵Business owners have stated that the ban on imports of used capital equipment can raise the cost of importing spare parts and other new equipment because of corrupt behaviour by customs officials. There have been a number of reforms of the policy in recent years, but the basic structure remains intact.

²⁶Resosudarmo and Yusuf (2006) note that log exports were first banned in 1981. The ban was replaced by high export taxes in 1992. These export taxes were reduced sharply under the terms of the agreements with the International Monetary Fund in 1998.

²⁷Malaysia applies a quota on log exports from the state of Sarawak on the island of Borneo, and thus the price of logs may be depressed there as well, which would imply that the actual rate of protection may be more negative than we calculate based on a comparison of log prices in Indonesia and Sarawak. Perhaps a more apt comparison would be between log prices in China and in Indonesia, adjusted for transport and other costs.

²⁸Minister of Industry and Trade Decree 385/MPP/KEP/6/2004.

²⁹Minister of Industry and Trade Decree 118/MPP/KEP/2/2003.

³⁰Minister of Trade Regulation 02/M-DAG/PER/1/2007.

³¹Minister of Industry and Trade Decree 443/MPP/KEP/5/2002. Enforcement of the ban evidently presented problems, and this led to the enactment of Minister of Trade Regulation 19/M-DAG/PER/4/2007, whose purpose was to improve regulation of inter-island trade so as to reduce smuggling.

³²Minister of Trade Regulation 07/M-DAG/PER/4/2005.

³³Minister of Energy and Mineral Resources Regulation 34/2009 and Government Regulation 23/2010.

³⁴The trade restrictions go back at least to 1998, with Minister of Industry and Trade Decree 558/MPP/Kep/12/1998. The most recent export policy stipulation along these lines is in Minister of Trade Regulation 01/M-DAG/PER/1/2007.

³⁵The price of gasoline in Indonesia, with value added tax removed, was slightly higher than the price in Singapore. Singapore price data are daily spot prices for conventional regular gasoline, from the Energy Information Administration, US Department of Energy.

³⁶Electricity and LPG are aggregated into IO sector 142 (electricity and gas), which we treat as non-tradable. We draw on data from a variety of sources for both products, and can provide details of the data and calculations upon request.

³⁷We use the sector numbers used by the government in the 175-sector IO table. Sector 34 is not included in our tables because it consists of non-tradable agricultural services.

³⁸Some finished motor vehicles, such as ambulances and specialty trucks, have much lower tariff rates.

³⁹Details of the effective protection calculations are available from the authors upon request.

⁴⁰Maryadi, Yusuf and Mulyana (2004) discuss how these contractual arrangements work in South Sumatra, as part of a policy analysis matrix study.

⁴¹The fuel subsidy does not accrue directly to the sugar sector, but instead comes primarily through its use of road transport services.

⁴²To make the figures more comparable, we omit the effects of subsidies from our own calculations, and follow FPC by calculating effective rates of protection based on the Humphrey (1969) method, as described earlier. In practice, effective rates of protection calculated with the Humphrey method (ERPH) tend to be close to, but smaller (less positive or more negative) than, rates calculated with the Balassa (1982) method. To accommodate preferential arrangements, we use the MRP import duty rates for 2008.

⁴³Full details of the 175-sector calculations are available from the authors upon request.

⁴⁴Fane and Condon (1996) arbitrarily set their estimate of the ERP in some sectors to 600, because it would otherwise have been very large, or even negative if the policies were found to be so protective that a sector would have had negative value added if all policy distortions were removed.

⁴⁵Arguably it would be better for the government to apply resource rent taxes to extract as much as possible of the rent component of natural resource exploitation, but without heavily taxing the value added contributed by labour and capital.