60022 Estimating the Fiscal Impact of the Côte d'Ivoire- Africa Trade Policy Notes EU Economic Partnership Note #4 Agreement Mombert Hoppe August, 2010 Introduction Côte dIvoire is the second largest economy in West Africa. It is one of eight member states of the UEMOA and is also a member simulation results of the impact of this state of ECOWAS. It applies the UEMOA Economic Partnership Agreement on fiscal Common External Tariff (CET) which revenues in Côte dIvoire. It finds that the consists of four tariff bands at zero, five, ten, interim EPA is likely to lead, at full and twenty percent. implementation, to a loss of annual fiscal To prevent trade disruption with the revenue in the order of less than CFA 43 European Union at the end of 2007 when the billion, representing 2.2 percent of total trade component of the Cotonou agreement 2007 government revenue. This reduction in ended, Côte dIvoire has negotiated and fiscal revenue will occur over 15 years and signed an interim Economic Partnership will be at least partially compensated by Agreement (EPA) with the European Union. growing import volumes as the economy Under the agreement, Côte dIvoire will grows. remove tariffs on 80.8 percent of the goods Data description originating in the European Union by 2023. The tariff removal has started in 2010 with For the estimation, we obtained data at tariff reductions for the first 576 lines, for transaction level from customs for the year 406 lines from 5 to 0 percent, for 160 lines 2007 covering imports for consumption from 10 to 0 percent, and for 10 lines from (excluding imports into warehouses and 20 to 0 percent. government imports). According to these Using the Tariff Reform Impact Simulation Tool (TRIST) developed by the World Bank1, this short note presents the implications of trade policy changes using TRIST (tariff reform impact simulation tool), World Bank 1 Policy Research Working Paper 5045, See Brenton, P; C. Saborowski; C. Staritz and E von http://go.worldbank.org/AMF45BV930. Uexkull (2009) ,,Assessing the adjustment 1 Table 1: Main import partners for Côte d'Ivoire, 2007, in CFA million Top 10 Import Partners Import Value Share Of Total Imports Nigeria 769,275 24.2% France 683,330 21.5% China, Peoples Republic 209,345 6.6% Venezuela 102,173 3.2% Thailand 88,894 2.8% Germany 86,459 2.7% U.S.A 85,802 2.7% India 79,928 2.5% Japan 78,421 2.5% Great Britain 69,977 2.2% Table 2: Imports from the EU, top 20 sectors in CFA million EU27 Total imports EU ISIC # ISIC Description (mln) (mln) share 292 Manufacture of special purpose machinery 156,136 200,102 78.0% 242 Manufacture of other chemical products 112,149 169,496 66.2% 341 Manufacture of motor vehicles 88,589 174,832 50.7% 151 Production, processing and preservation of meat, fish, fruit, vegetables, oils and fats 66,479 191,018 34.8% 011 Growing of crops; market gardening; horticulture 62,775 90,694 69.2% 291 Manufacture of general purpose machinery 57,279 92,775 61.7% 241 Manufacture of basic chemicals 54,221 169,444 32.0% 322 Manufacture of television and radio transmitters and apparatus for line telephony and line telegraphy 47,843 81,289 58.9% 353 Manufacture of aircraft and spacecraft 41,034 42,913 95.6% 271 Manufacture of basic iron and steel 34,079 95,681 35.6% 343 Manufacture of parts and accessories for motor vehicles and their engines 33,754 39,183 86.1% 289 Manufacture of other fabricated metal products; metal working service activities 30,983 49,239 62.9% 210 Manufacture of paper and paper products 26,869 47,172 57.0% 154 Manufacture of other food products 24,951 35,264 70.8% 232 Manufacture of refined petroleum products 24,366 60,885 40.0% 155 Manufacture of beverages 23,954 29,494 81.2% 153 Manufacture of grain mill products, starches and starch products, and prepared animal feeds 22,487 183,353 12.3% 331 Manufacture of medical appliances and instruments and appliances for measuring 18,510 27,231 68.0% 152 Manufacture of dairy products 15,680 24,184 64.8% 252 Manufacture of plastics products 15,044 26,624 56.5% Total of to 20 sectors for EU imports 957,182 1,830,872 52.3% Total all imports 1,142,297 3,180,734 35.9% 2 data, Côte dIvoire imported a total of CFA generally face lower tariffs than those from 3.18 trillion in 2007. 35.9 percent of these the rest of the world. The trade-weighted imports originated in the European Union. actually applied tariff rate on EU imports Non-oil imports were CFA 2.26 trillion in was 5.3 percent in 2007, while the average 2007, 50.4 percent of which originated in tariff for imports from the rest of the world the European Union. Other major trading (excluding other ECOWAS member states) partners were Nigeria (of which 99.5 percent was 7.5 percent. Given that the statutory were petroleum imports), China, Venezuela trade-weighted tariff rates were 10.7 and 8.8 (only petroleum imports), and Thailand (see percent, respectively, the detailed data show Table 1). that imports originating in the EU were generally granted more exemptions (half of Table 2 shows imports from the European the duty payments on imports from the EU Union by the main industrial (ISIC) were exempted) than imports originating in categories. These imports are concentrated other non-ECOWAS countries (where only in machinery and chemicals, as well as 15 percent of payments were exempted). cereals and processed foods. This already indicates that removing tariffs on EU imports is likely to have a The data obtained also contain values for the substantially smaller effect on fiscal revenue actually collected tariff revenue at the than statutory tariff data, which has often borders (accounting for exemptions been used in similar studies in the past, granted), as well as data on revenue would indicate. These exemptions are collected through excise taxes and the Value granted on a large variety of products but a Added Tax (VAT, at 18 percent), which is substantial share has been granted on collected at the border for imports. Table 3 imported vehicles. summarizes the amount of taxes collected at the border. These data allow the estimations Table 4 further details the treatment of of revenue changes to be based on actual imports originating in the EU. It groups rather than hypothetical revenue data. imports by tariff rates that should have been and that were actually levied. While only 6.4 Assessing the revenue collection in greater percent of imports originating in the EU detail, the data show that Côte dIvoire should have entered Côte dIvoire duty free, collected CFA 61 billion or 39.8 percent of actually 34.7 percent did not pay any duties. its customs revenue from imports At the same time, using data aggregated at originating in the EU.2 Imports from the EU the product line level, only 15.6 percent of 2 Revenues of CFA 479 million were collected from goods imported from the European Union imports originating within UEMOA. This figure is paid more than 10 percent tariff duties, small but should in principle be zero given that while according to statutory tariff rates, 32.6 UEMOA has established a free trade area, indicating percent should have paid these higher rates. that issue relating to rules of origin may be important but also concerns with the recording of data collected at the borders. In particular, some goods that originate outside the customs union and pay duty upon import are recorded as originating within the be able to adequately simulate the effects of customs union. While these cases can be easily preferential trade liberalization on these trade flows, identified, it raises the concern that the origin of although given the relative magnitude of these flows products could also be inadequately recorded in the the impact on the the results of the simulations is customs database in other cases. Where the origin of likely to be small. trade flows is incorrectly recorded, the model will not 3 Table 3: Border taxes collected by Côte d'Ivoire, 2007, in CFA million Statutory tariff Collected tariff Excise tax VAT Total Value 239,335 152,331 9,257 200,316 Share Of Total 42.1% 2.6% 55.4% Simple Average 13.7% 11.8% 0.3% 14.8% Weighted Average 7.5% 4.8% 0.3% 6.0% Table 4: Distribution of tariffs applied to imports from EU Imports from EU (CFA million) falling into categories according to: Share of EU imports statutory tariff statutory applied Tariff category rate applied tariff rate tariff rate tariff rate 0 73,131 396,023 6.4% 34.7% 0<=5% 426,162 418,565 37.3% 36.6% 5%20% 0 490 0.0% 0.0% Source: database and own computations Based on the trade data and Côte dIvoires currently levying a zero tariff on some of the interim EPA exclusion list used for this lines that would be excluded from simulation exercise, Côte dIvoire would liberalization under the West African EPA liberalize 59.3 percent of imports from the market access offer, resulting in an effective EU by January 2013 (of which 10.8 percent exclusion from liberalization of 23.1 percent already attracted a statutory tariff of zero of current imports from the EU. percent), an additional 13.4 percent by January 2018, and an additional 7.7 percent Simulating trade policy changes by January 2023, leading to a removal of Using the trade flow pattern and actual import duties on imports from the EU of revenue collected in 2007, this section 80.3 percent based on the imports structure describes the simulation results of a number of 2007. Using the latest available market of tariff liberalization scenarios using the access offer presented by ECOWAS, which Tariff Reform Impact Simulation Tool differs from that negotiated by Cote (TRIST). The model uses the detailed trade dIvoire, in the context of the regional EPA and revenue data described above to negotiations, 26.7 percent of Côte dIvoires estimate the effects of tariff rate changes on imports from the EU would remain excluded trade flows and revenue collection. from liberalization at the end of the transition period. However, Côte dIvoire is 4 Using a partial equilibrium framework, the Revenue effects of the Côte d'Ivoire-EU model first estimates the trade diversion interim EPA effects resulting from relative price changes following trade liberalization. The removal The first scenario used for simulating of customs duties on products originating in reforms assumes that Côte dIvoire removes selected countries makes these products tariffs on 80.3 percent of imports from the cheaper in relative terms when compared to European Union. We use the exclusion list the same products originating in the rest of that forms an integral part of the interim EU- the world and importers substitute these Côte dIvoire Economic Partnership products for those that are now relatively Agreement to identify these products that more expensive. There is no substitution are excluded from liberalization. The between different products and because the simulations are undertaken using standard model uses elasticities for its calculations, elasticities, as well as higher elasticities for zero trade flows will remain zero. In a substitution among various import sources second step, as average prices for goods fall for sensitivity analysis.3 in the domestic market, consumers demand Based on these simulations, overall imports more of those goods, and the model by Côte dIvoire would increase by CFA 18 estimates a trade creation effect as effective billion, or 0.6 percent. Tariff revenue demand for cheaper products increases. Both collected would decrease with CFA 38 effects are modeled using the existing billion or 25.0 percent. When including structure of trade flows, price changes as a secondary effects on VAT and excise result of tariff reductions, and elasticities for collections, the overall loss in border tax the substitution effect and the demand revenues would equal CFA 41 billion or effect. The aggregate changes in trade flows 11.3 percent of revenues from tariffs. are then used to calculate the changes in Doubling the elasticity of substitution customs duties and border taxes (customs among various sources of imports, which duties, excise duties, and VAT) collected. leads to a greater trade diversion effect, The results should be interpreted as short- results in estimated tariff revenue losses of term simulations, as the model does not CFA 40 billion or 26.0 percent, while capture effects in the real sector, i.e. overall border taxes would fall by CFA 42.5 responses in output, competitiveness gains billion or 11.7 percent (see columns titled from cheaper access to intermediate inputs "Interim EPA" in table 5). that make specific domestic industries more competitive, or effects resulting from changes in export performances. The model 3 does not consider general increases in TRIST uses two elasticities during the simulation. import flows as a result of economic growth The elasticity of substitution describes the and increased demand. Consequently, the responsiveness of importers to relative price changes of identical goods following preferential tariff simulation results can be seen as describing liberalization and consequently models the degree to only the change on tariff revenues as a result which importers switch to new sources as a result of of tariff reform and represent the marginal these price changes. It is set at 1.5 in the standard changes as compared to a baseline scenario scenarios, and at 3 for the sensitivity analysis. The without an EPA. second elasticity models consumers responsiveness in demand for a given product to price reductions of products following tariff reform. This demand elasticity is set at 0.5 in all scenarios. 5 Table 5: Summary of estimation results Standard elasticities Alternative elasticities Interim ECOWAS ECOWA Interim ECOWA ECOWA EPA EPA S EPA + EPA S EPA S EPA + Impact on imports: Imports pre 3,180.9 3,180.9 3,180.9 3,180.9 3,180.9 3,180.9 Imports post 3,198.7 3,195.7 3,198.1 3,199.0 3,195.9 3,198.4 Change in imports 17.7 14.8 17.1 18.1 15.0 17.4 % change in imports 0.6% 0.5% 0.5% 0.6% 0.5% 0.5% Impact on revenue: Tariff revenue pre 152.3 152.3 152.3 152.3 152.3 152.3 Tariff revenue post 114.2 121.1 115.9 112.7 120.1 114.7 Change in tariff revenue -38.1 -31.2 -36.4 -39.6 -32.3 -37.7 % change in tariff revenue -25.0% -20.5% -23.9% -26.0% -21.2% -24.7% Total Tax Revenues on Imports Total revenue pre 361.9 361.9 361.9 361.9 361.9 361.9 Total revenue post 321.1 328.1 322.9 319.4 327.0 321.5 Change in Total revenue -40.8 -33.8 -39.0 -42.5 -34.9 -40.4 % change in Total revenue -11.3% -9.3% -10.8% -11.7% -9.6% -11.2% Collected Tariff rate: Collected applied tariff rate pre 4.8% 4.8% 4.8% 4.8% 4.8% 4.8% Collected applied tariff rate post 3.6% 3.8% 3.6% 3.5% 3.8% 3.6% % change in collected applied tariff rate -25.4% -20.8% -24.3% -26.4% -21.6% -25.1% note: standard elasticieties are 1.5 for substitution among sources of imports and 0.5 for the demand elasticity; alternative elasticities assume substitution elasticity of 3 (and unchanged demand elasticity) The results indicate that the implementation revenue from border taxes decline with an of the EPA would have a very limited additional CFA 14.8 billion or 4.1 percent as impact on collected revenue from border compared to the original tax revenues. The taxes. effects during the last phase of the liberalization by 2023 would be relatively We also break down the simulations to small, with imports increasing by another analyze the effects of each of the three tariff CFA 1.6 billion, tariff revenue falling with liberalization phases. With standard CFA 3.4 billion, and overall revenue from elasticities, removing import duties during border taxes declining with an additional the first five years according to the CFA 3.6 billion or one percent. Again, these liberalization calendar would lead to an effects are estimated as relative to the increase in imports of CFA 10 billion or baseline scenario and do not take general 0.3%, a reduction in annual tariff revenue of import growth effects into consideration. CFA 21 billion, and a total reduction in They should be seen as indicating the annually collected border taxes of CFA 22.4 relative impact on fiscal revenue during the billion, or 6.2 percent of total border taxes three phases of liberalization rather than by 2013. Following the second liberalization specific estimates. To summarize, more than phase (after 10 years, by 2018), imports half of the total fiscal revenue loss from would increase with an additional CFA 6 border taxes relative to the baseline scenario billion, tariff revenue would decrease with will occur during the first liberalization an additional CFA 14 billion, and overall 6 phase, another 36 percent will occur during of a FTA among all ECOWAS member the second liberalization phase, while the states and tariffs would be removed on intra- remaining nine percent will occur during the regional trade. Again, in the absence of the last phase. From a revenue point of view, the new CET rates, the application of an EPA liberalization is strongly frontloaded. with the EU and the simultaneous removal of import duties on all intra-regional trade Revenue effects of an ECOWAS-EU EPA would lead to a reduction in tariff revenues of CFA 36 billion, and a reduction in overall To compare these results tentatively with the revenues collected at the border of CFA 39 likely effects of a regional Economic billion, or 10.8 percent. Again, effects would Partnership Agreement between West Africa be slightly larger using alternative (ECOWAS and Mauritania) and the elasticities (see columns titled "ECOWAS European Union, we also simulate the effect EPA +" in Table 5. of the latest market access offer submitted by ECOWAS on fiscal revenues in Côte Conclusion dIvoire. Here, we simulate the impact of removing tariffs on only 26.7 percent of Both the interim EPA and a potential imports from the European Union of which ECOWAS EPA will have limited effects on 3.6 percentage points were already facing a revenue collected at the borders in Côte statutory duty rate of 0 percent in 2007 in dIvoire. To a large extent, this reflects that Côte dIvoire. The simulation consequently many imports from the European Union are effectively continues to apply positive already entering duty free. Simulation statutory tariff rates on 23.1 percent of results based on a partial equilibrium imports originating in the European Union. framework estimate that fully implementing Many of those already faced applied tariff the interim EPA would lead to a loss of rates of less than the statutory rates. The fiscal revenues at the border of between simulation estimates that the application of CFA 40.8 and 42.5 billion, representing less this market access offer would lead to an than 2.2 percent of government revenue in increase of imports of CFA 15 billion, a 2007. Tariff revenue losses under a potential decrease in tariff duty revenues of CFA 31 regional EPA would be smaller as 26.8 billion, or 20.5 percent, and an overall percent of imports from the EU would not decrease in collected border taxes of CFA see their tariffs removed (as opposed to 19.7 34 billion, or 9.3 percent. These figures are under the interim EPA). Fiscal revenue somewhat higher using alternative losses at the border are estimated between elasticities (see columns titled "ECOWAS CFA 33.8 and 34.9 billion (or 9.3 to 9.6 EPA" in table 5). However, as part of a percent of current border revenues) in that regional EPA, Côte dIvoire would also case. The reduction in fiscal revenue, have to modify its statutory tariff rates by however, will occur over 15 years and will adopting the new ECOWAS Common be at least partially compensated by growing External Tariff (CET), which would see import volumes as the economy grows. import duties for a number of products However, changes to the ECOWAS-wide increase to 35%. As the new CET has not CET and the parallel completion of the yet been agreed in detail, we are unable to internal West African market would have a run a simulation of moving to the new further impact on fiscal revenues. In the ECOWAS CET. At the same time, the absence of a finalized West African CET, regional EPA would also lead to the creation these effects have not been modeled. Once 7 progress is made in this regard TRIST could resources to flow into those sectors that are be used also in neighboring countries to most competitive, and to the most efficient assess the wider impacts of full regional firms within them, will be essential to integration and the regional EPA. maximize the benefits from the EPA. Even if these losses are relatively small, government will either have to cut spending or find ways to replace these revenues. One About the Author way could be the expansion of the existing Mombert Hoppe is a Consultant in the Africa Poverty VAT system or the strengthening of other Reduction and Economic Management unit. This domestic sources of revenue. Another work is funded by the Multi-Donor Trust Fund for opportunity would be the establishment of a Trade and Development supported by the compensation fund as part of the (regional) governments of the United Kingdom, Finland, EPA. A similar mechanism has been Sweden and Norway. The views expressed in this paper reflect solely those of the authors and not established in COMESA and it has been necessarily the views of the funders, the World Bank used to compensate Burundi and Rwanda for Group or its Executive Directors. losses incurred when moving to the Common External Tariff of the East Africa Community. Both countries used TRIST to estimate the likely revenue losses at the beginning of the year to access the compensation fund. Changes to tariff revenue reflect only one dimension of effects resulting from the implementation of a free trade agreement. Trade liberalization with the EU may lead to lower consumer prices and reduced prices of intermediate inputs for some companies, increasing their capacity to integrate into global supply chains and increasing their competitiveness to compete globally with potentially large dynamic effects on output and employment growth. Nevertheless, these economic benefits would be enhanced if an EPA were to be accompanied by liberalization of imports from all sources. Accompanying the implementation of an EPA with good economic policies that address current constraints to growth and the development of businesses, reduce the costs of doing business, support those negatively affected by trade reforms and helping them to adjust, and increasing the overall flexibility of the economy, allowing 8