THE STABILIZATION AND ASSOCIATION AGREEMENT (SAA) BETWEEN BOSNIA AND HERZEGOVINA (BIH) AND THE EUROPEAN UNION: Impacts of the Adaptation Protocol (AP) on the Agriculture and Food Sector The Stabilization and Association Agreement (SAA) between Bosnia and Herzegovina (BiH) and the European Union: Impacts of the Adaptation Protocol (AP) on the Agriculture and Food Sector The World Bank Group1 KEY MESSAGES: In the short term, the Adaptation Protocol (AP) of the SAA will result in tariff revenue losses for BiH and may place additional pressure on some sectors. The World Bank carried out a trade analysis to identify potential effects of this AP on the imports from the EU on consumption and local production of agricultural and food products. The maximum tariff revenue loss for agriculture and food imports from the EU-28 was estimated to range between BAM 51 and 68 million annually, based on available production and trade data. This estimate provides a partial perspective on the trade flows; technology changes, as well as consumer benefits of reduced prices can offset these short term effects. Trade liberalization has been ongoing for a decade in BiH, as in other countries in the region, and will continue to influence trade flows. In June 2008 an Interim Stabilization and Association Agreement (SAA) on trade and trade-related issues enters into force. On February 1st 2017, the Adaptation Protocol (AP) adapting the Stabilization and Association Agreement (SAA) between BiH and the EU provisionally entered into force. The SAA AP affects 8 percent of BiH’s agriculture and food imports from EU-27 and Croatia. The majority of agricultural and food imports from the EU-27 and Croatia have already been liberalized. Improving the competitiveness of the agri-food sector in BiH is a good proposition in a transforming environment. This is a process that can be built around three key areas: (i) investments in the productive and entrepreneurial capacity of agricultural producers, linking them to markets and building their comparative advantages; (ii) investments in an integrated information system for decision making, where both compliance mechanisms (food safety measures, registries, etc.) to EU accession and other information (prices of important products, weather data, extension services, etc.) can build resilience and improve the capacity to respond to trade and climate pressures of small producers, hence making them more competitive in a rapidly changing agro-ecological and socio-politic environment; (iii) strengthening the capacity of the Government to provide an enabling environment for these investments to have impact. 1 The analysis was carried out by Alberto Portugal and Lulu Shui, with comments and suggestions received from Michael Ferrantino, Irina Schuman, Paolo Correa, Julian Lampietti, Daniel Gerber, Javier Suarez, and Svetlana Edmeades for their helpful comments and suggestions. 1 Background on the Stabilization and Association Agreement (SAA) and the Adaptation Protocol (AP) Trade relations between Bosnia and Herzegovina (BiH) and the European Union (EU) are governed by the Stabilization and Association Agreement (SAA). The SAA between the EU and Bosnia and Herzegovina - a key step towards EU accession –provides the formal mechanisms and timelines to bring BiH closer to EU standards, supporting movement towards eventual EU accession. Full and effective implementation of the SAA is required for any further assessment by the EU of the country’s accession prospects. The entry into force of the SAA, and its effective implementation, provides a positive signal of progress towards EU accession. As such it can increase the confidence of investors, domestic and international in the country. A key element of the SAA is the liberalization of bilateral trade, which was already put in place with the entry into force in June 2008 of the Interim Agreement on trade and trade-related issues, On February 1st 2017, the Adaptation Protocol (AP) adapting the Stabilization and Association Agreement between BiH and the EU provisionally entered into force. BiH had been required to adapt the SAA after Croatia became an EU member in July 20132. Croatia, along with BiH, was a member of the Central European Free Trade Agreement (CEFTA), but was obligated to leave CEFTA upon its accession to the EU. As a result, the EU demanded that the concessions between the two sides be adjusted according to the trade flows that existed between BiH and Croatia prior to the accession of Croatia to the EU. The BiH authorities refused to do so out of concerns about the impact that this would have on the country's agricultural sector. Subsequently, the EU suspended BiH’s preferential trade arrangements (Autonomous Trade Measures, ATM3) as of January 1, 2016. In 2016 the adaptation of the SAA, through the implementation of the AP, was made a crucial condition of BiH's EU membership application process. On July 18th 2016, BiH agreed to adapt the SAA to take into account bilateral trade with Croatia. On September 9th, the BiH government adopted a protocol on adjusting the SAA, which was signed in December 2016. The new AP provides preferential market access for both, EU agricultural and food products into BiH and BiH agricultural and food products into the EU market. This market access reflects trade preferences and trade flows that existed between BiH and Croatia under CEFTA. Under the terms of the new AP, agricultural and food producers from BiH benefit from access to the EU through the increase of relevant quotas for exports of wine, sugar and fish. EU agricultural and food producers will have market access to BiH through the creation of tariff rate-quotas (TRQs) for a number of products. Study Objectives and Limitations The objective of this note is to provide insights into the potential effect of the AP on imports of EU agricultural, food and fisheries products into BiH. We adapt and calibrate the Tariff Reform Impact 2 In connection with Croatia’s EU accession, the EU negotiated and signed adaptations to all its existing trade agreements with third countries, including BiH. 3 The EU’s “autonomous trade measures” (ATM ) grant exceptional and unlimited duty-free access to the EU market for almost all agricultural and processed food products originating in the Western Balkan countries. Between 2000 and 2016, ATM allowed BiH exporters to also be exempt from EU specific duties on imports of fresh fruits and vegetables related to the so-called entry price system which aims at stabilizing the EU market by preventing the price levels in non-EU countries from having repercussions on prices within the EU. ATM was reintroduced in relation to BiH on February 1, 2017 and will remain in force in relation to all Western Balkan countries until 2020. 2 Simulation Tool (TRIST)4, a partial equilibrium trade model developed by the World Bank, to estimate the impact of imports from the EU on consumption and local production of agricultural and food products for which disaggregated data is available. The impact of a change in tariffs is modeled in a simple partial equilibrium trade model with imperfect substitution between imports from different trading partners and domestic production. The team benefitted from the cooperation with the counterparts (MOFTER and Ministries) regarding data and inputs. This analysis has important limitations and the results should be considered as indicative estimates. This is a trade analysis that only partially reflects the overall impact and provides a very short term perspective on trade interactions. TIRST is a static model that depends on the availability of disaggregated agricultural data on both imports and production. The following aspects need to be kept in mind when drawing conclusions from this report:  The model does not take into account the impact of EU enhanced market access on BH exports;  The model does not estimate the potential impact of non-tariff barriers that may exist;  The model only considers products for which there is correspondence between production and import data, which was not the case for some products (e.g. processed food) and these were not considered in the analysis;  The model estimates rely on trade data available until 2015 and FAO production data available until 2013; data for 2016 was not available at the time the analysis was carried out;  The model provides estimates of direct revenue losses connected to the tariff; no indirect loss related to the value of the liberalized import is estimated;  The model does not consider changes in technology, which can influence productivity and hence production volumes. Descriptive analysis Total BiH’s agricultural imports have been relatively stable from 2011 to 2015. The EU-27, CEFTA and Croatia are three major trading partners with BiH, together accounting for over 80 percent of BiH’s agricultural imports (Figure 1). Beverages, cereal and animal products dominate BiH’s imports from its major trading partners, as shown in table A1 of the Appendix. The SAA AP affects 8 percent of BiH’s agriculture and food imports from EU-27 and Croatia. To estimate the share of imports from the EU and Croatia that will be affected by the implementation SAA AP, we focus on BiH imports in 2015, the last year with available imports data. Figure2 shows that the majority of agricultural and food imports from the EU- 27 and Croatia have already been liberalized.5 Imports that will be partially liberalized in 2017 under the new TRQs have a total value of just under BAM 120 million of equivalent 2015 imports. 4 For more information on TRIST, please refer to Brenton et al. (2009) as well as the World Bank’s TRIST homepage: http://go.worldbank.org/2X1IC75J40 5 Article 34.1 of the SAA states: “From the date of entry into force of this Agreement, no new customs duties on imports or exports or charges having equivalent effect shall be introduced, nor shall those already applied be increased, in trade between the Community and Bosnia and Herzegovina”. Annex III of the SAA has five sections, each one describing a modality for tariff reduction from the date of entry into force of the Agreement, including Annex III(f) that was introduced by the Adaptation Protocol. 3 Figure 1: Origin of agricultural and food imports in BH 3,500 3,000 2,500 Millions BAM 2,000 1,500 1,000 500 - 2011 2012 2013 2014 2015 Turkey Others Croatia CEFTA EU-27 Source: Staff calculations using data shared by the BH Government. Note: Agricultural and food imports refer to products under chapters 1-24 of the Harmonized System (HS). Figure 2: Value of BH agriculture and food imports (BAM million, 2015), by origin and tariff status (as applicable at the entry into force of the protocol adapting the SAA) Fully liberalized (2017) No change in SAA regime (2017) Subject to SAA Additional Protocol (2017) ROW 986 318 CROATIA 402 59 EU-27 1,004 59 60 Source: Staff calculations using data shared by the BH Government. Notes: 1. Agricultural and food imports refer to products under chapters 1-24 of the Harmonized System (HS); 2. The SAA AP (entered into force on 02/01/2017) provides for increases in tariff rate quotas (TRQs) for 68 HS-10 digit tariff lines; 3. RoW stands for Rest of the World. Tariff levels on imports into BiH from EU-27 and Croatia differ greatly across product groups. Figure 3 shows the domestic production for different products, the imports of these products from Croatia and EU-27, as well as the weighted average these imports were confronted before and after the entry into force of the SAA AP.6 Most categories have tariffs close to zero once the SAA AP is in force, except for chicken and a lower extent for potatoes, grapes and milk. Chicken meat has the highest protection level equal to an ad valorem equivalent (AVE) rate of over 200 percent, mainly because the relative high-specific 6 The details of TRQs in place on Feb 2017, as part of the SAA and AP are included in table A2 in the Appendix 4 tariffs included in its compound tariffs.7 As a consequence, EU-27 imports are relatively small compared to domestic production. It is worth mentioning that the AVE of compound tariffs varies in time if there are changes in the unit price of the trade good. Some products of Croatian origin are likely to be liberalized by the SAA AP, such as potatoes and milk. Figure 3 Domestic production, imports from the EU, and tariffs for different products. Source: Staff calculations using data compiled from FAO and shared by the BH Government. Note: Production and imports are in BAM million, whereas the weighted tariffs are in percentage. Data constraints limited the extent of the modelling exercise, and in particular, the impact assessment on domestic production. Production data was taken from the FAO, which was only available for some primary and fresh food products until 2013. We were able to simulate the impact of the SAA AP only on categories that had data on production. Yet, as shown in Figure 4, most of imports are not affected by the SAA AP either because they have been already liberalized by Jan 2017 or because the SAA regime does not specify yet a change in the tariff regime. Figure 4Figure shows the amount of EU imports that have not been liberalized (in green). Among imports having a change in tariffs or subject to SAA (grey bars in Figure 2), Figure 4 differentiates products for which no production data is available (blue bars) from products with FAO production data (orange bars). 7 Compound tariffs have two components: (i) an ad valorem tariff that is calculated as a percentage of the value of the product, and (ii) a specific tariff expressed as monetary terms per physical unit of the good being imported. 5 Figure 4 Value of BH agriculture and food imports (BAM million, 2015), by effect of the SAA AP Not liberalization from the SAA AP Subject to SAA AP (w/o production data) Subject to SAA AP (w/ production data) ROW 1,304 CROATIA 405 50.9 8.1 EU-27 1,063 52.2 7.8 Source: Staff calculations using data shared by the BiH Government. Notes: 1. Agricultural and food imports refer to products under chapters 1-24 of the Harmonized System (HS); 2. The SAA Adaptation Protocol (entered into force on 02/01/2017) provides for increases in tariff rate quotas (TRQs) for 68 HS-10-digit tariff lines. TRIST Model and Simulation Strategy The Tariff Reform Impact Simulation Tool (TRIST), a partial equilibrium model, is used to estimate the effects of SAA AP adoption on imports and domestic production in BH. The World Bank has developed TRIST, a simulation tool that can be used by policy makers in client countries to analyze the adjustment implications of trade reform. When appropriate data are available, it can provide information on the short-term relative vulnerability of different sectors in the domestic economy in terms of output and employment.8 The tool was developed to provide better estimates of the impact of changes in tariffs on government revenues, imports, protection and prices. It is flexible and can incorporate tariff liberalization scenarios involving any group of trading partners and any schedules of products, it runs in Excel, with formulas and calculation steps visible to the user; and is open-source and users are free to change, extend, or improve according to their needs.9 The trade model in TRIST is a partial equilibrium model that treats demand for each product in isolation from the rest of the economy. Hence, it does not take into account inter- and intra-sectoral linkages or the economy wide impacts of tariff changes. But this is not the primary objective of TRIST, which is designed so as to avoid the degree of aggregation of the data that would be necessary in order to implement economy wide computable equilibrium models and to remain simple and transparent in its 8 It can also be linked to household budget data to trace the influence of changes in prices following trade reform to household expenditures and the costs of attaining the given consumption bundle 9 The tool and extensive documentation can be found in the webpage: http://econ.worldbank.org/WBSITE/EXTERNAL/EXTDEC/EXTRESEARCH/EXTPROGRAMS/EXTTRADERESEARCH/0,,co ntentMDK:21537281~pagePK:210058~piPK:210062~theSitePK:544849,00.html 6 assumptions, with the flexibility to adjust the key parameters. Thus, TRIST has been designed with the specific task of providing policy makers with important insights into the short-term effects of trade reform. It has not been designed for making longer-term predictions about the broad economy wide impact of trade reform. Indeed, no long term effects are estimated, such as growth and reallocation of production factors such as capital and labor. By its comparative static nature TRIST allows the comparison of two states - one in which the base values of policy instruments (such as tariffs) are unchanged and another in which these base values are exogenously changed. Thus no dynamic effects are considered. TRIST involves three stages to determine the impact of imports and production following a change in trade policy (Figure 5). In the first stage, TRIST allocates expenditure on imports of a product across different country suppliers. The allocation changes when tariffs and duties are amended10. The extent to which a given change in relative prices translates into a change in relative imports depends on a user-defined importers substitution elasticity. In order to isolate the importers substitution effect, total imports are held constant in this step. In the second step, total expenditure on a given product is allocated between domestic sources and imports. The domestic substitution effect allows for a demand shift between domestic production and imports when the relative price of imports changes. The extent to which the share of imports in domestic consumption changes depends on a user defined domestic substitution elasticity. The change in imports is then distributed across all importers according to their share of the import market. This calculation step can only be modeled if data on domestic production is available. Finally, the third step allows for an overall demand effect in response to the change in the average price of domestic consumption of the good. The average price change is computed as an average of the price change in imports and the price change in domestic production, weighted by their relative shares in domestic consumption. A decrease/increase in the average price of the product leads to a percentage increase/decrease in overall consumption of the product, proportionately distributed between imports and domestic production. The extent to which imports change for a given change in the overall price depends on a user-defined import demand elasticity.11 10 The exporter substitution effect defines how imports from exporter A are substituted for imports from exporter B when the price of imports from exporter A relative to B declines, for example following a preferential trade reform that includes exporter A but not exporter B. 11 Elasticities are crucial parameters of the model that are difficult to estimate and so detailed and robust estimates of the three elasticities (importers substitution, domestic substitution, demand) are not readily available in the literature. TRIST includes sensible default values for each of these three parameters that are common across products and import suppliers. The sensitivity of the results can be easily assessed by changing the values of the elasticities. When detailed local knowledge on these elasticities is available, TRIST allows users to define trading partner and product specific elasticities. Furthermore, there is an option to include the most well-known estimates of elasticities in the literature. First, the user can choose to incorporate the import demand elasticities estimated in Kee et al (2005). However, these elasticities are not available for all product groups (HS 6 digit). In addition, the user can choose to use the product specific import demand elasticities used in SMART. For exporter substitution elasticities or domestic substitution elasticities there are no estimates available at the level of product detail that TRIST uses. 7 Figure 5: The three steps in the TRIST model. A synthetic baseline scenario is constructed to take account of events between 2013 and 2016 and address other data constraints. TRIST is a static model that is calibrated in a baseline scenario and once the trade policy variables are shocked, it would estimate the new post-shock scenario. Ideally, we would have liked to obtain trade and production data for 2016 to estimate the effect of the SAA AP implemented in February 2017. Yet, FAO production data is only available until 2013 and trade data until 2015. In addition, other events need to be considered to construct a baseline scenario. First, Croatia joined the EU in July 2013 and BH imposed quotas on their imports from this country after that. Second, the floods in May 2014 and the drought of the summer 2015 caused losses to agriculture production in BH. Figure 6 describes the simulation strategy step-by-step. First, we calibrate the model using average 2011/12 values for production, trade flows and trade policy data (tariffs and TRQs), which mimics a scenario where there are no natural disasters affecting domestic production. Then, we apply a policy chock by setting all BIH tariffs and TRQs on imports from the EU-27 and Croatia to the levels of January 2017, just before the entry into force of the SAA AP. The simulated trade flows and production levels constitute what we call the synthetic baseline scenario. Finally, we shock the model by changing tariffs and TRQs on imports from the EU to the levels of February 2017, after the SAA AP enters into force, which would simulate the production and imports levels after the SAA AP. Figure 6: Steps of the TRIST simulation strategy 8 Results FAOSTAT production data allowed simulations only for 8 product groups and only products with large, non-binding quotas in the TRQs are expected to have higher imports and lower production for local market. To estimate whether a TRQ will be binding once the SAA AP is implemented, we used 2015 import data (the last year with available import data) and checked whether the same TRQ will be binding if applied in 2015. The underlying assumption is that the demand for imports from the EU-27 and Croatia in 2017 onwards will be at least as big as in 2015. SAA AP was found to affect only local production of poultry and honey12 and no adverse effect on local production was found for the following: (i) potatoes; (ii) cabbages and other brassicas; (iii) carrots and turnips; (iv) sour cherries; (v) grapes, and (vi) whole, fresh cow milk. The quotas of TRQs applied to these products under the SAA AP will become binding. Thus, only the first units before reaching the quota enter the BiH market duty-free, and imports beyond the quota enter the BiH paying the MFN tariff that was prevalent in these imports before the implementation of the SAA AP. Thus, there is no actual change on imports and production as the constraining TRQs shelter these products from EU competition. The only change is a loss of tariff revenue for the first units and the rent may be distributed between traders and consumers. As explained before, the choice of elasticities is fundamental on the magnitude of results. We run a few simulations using different vectors of elasticities. Simulation results when using high elasticities for substitution between imports and for domestic-foreign substitution, drawn from the GTAP model13, suggest an increase in the negative effect on domestic producers as local production of poultry and honey falls by 1.2 M BAM and 0.2 M BAM respectively (Figure 7). These simulations can be considered as an extreme unfavorable case scenario for BiH producers. Consumer welfare increases following the reduction of tariffs on EU products, because they reduce local prices; consumer welfare gains are likely to offsets tariff revenue loss plus producers’ welfare loss in the model. Simulations do not take into account the effects on BiH exports to the EU due to enhanced marked access under the SAA AP that could also increase revenue for local producers exporting to the EU. Figure 7: Simulation results: (units in BAM) Change in: Imports from EU- Production for Imports from ROW 27 and Croatia local market Chicken meat 1,837,289 51,823 -1,182,538 Honey 276,647 26,406 -223,289 Source: Staff calculations using TRIST model. Note: Chicken meat: Elasticity for imports substitution is 8.8, elasticity for domestic/foreign substitution is 4.4, elasticity for demand is 0.5. Honey: Elasticity for imports substitution is 2.6, elasticity for domestic substitution is 1.3, elasticity for demand effect is 0.5. 12 Honey production is 1.4 percent of the value of agriculture production recorded by FAO, whereas poultry production is 5.2 percent. Also, import volumes of honey and poultry by BH are quite significant (more than 200 tons of honey and above 14,000 tons of poultry in 2013), which implies that domestic producers are already facing external competition. Furthermore, productivity of the poultry sector is higher than in neighboring countries, which indicates existent capacity to respond to competition. 13 The standard GTAP Model is a multiregional, multisector, computable general equilibrium model, with perfect competition and constant returns to scale that is also a source of data on elasticities. For reference, see: https://www.gtap.agecon.purdue.edu/models/ 9 Upon effectiveness of the SAA AP, the maximum tariff revenue loss estimated for agriculture and food imports from the EU-28 can be expected to range between BAM 51 and 68 million annually.14 There are many products for which the TRQs of the SAA AP will become binding, thereby import levels, and thus local production, are not affected. Yet the first imported units before quotas become binding do not pay tariffs and provoke a tariff-revenue loss for the government, while imports beyond quota will pay the most favored nation (MFN) tariff. The estimates focus on the HS tariff lines that will experience changes in TRQs following the adoption of the SAA AP and that are likely to become binding. Thus, the “maximum revenue loss” is estimated assuming that TRQs will be binding once the SAA AP is implemented, assuming the demand will be larger than the quota. Figure 8 reports our estimates of the maximum tariff revenue loss for different categories of products. The yearly variation follows the TRQ schedules that tend to increase quota levels. Figure 8: Maximum tariff revenue loss estimated for agriculture and food products. 2017 2018 2019 23 23 21 21 16 15 9.9 8.8 7.8 7.8 7.8 5.8 3.1 3.1 3.1 1.5 1.5 1.5 0.1 0.1 0.1 0.2 0.2 0.2 0.1 0.1 0.1 0.6 0.6 0.6 0.4 0.4 0.4 Source: Staff calculations Despite the data limitations, the results from the modelling exercise imply that some agricultural producers might be adversely affected by the SAA AP in the short term. However, at the aggregate, tariff revenue losses estimated under this exercise need to be weighed against other structural factors of the BH’s agricultural sector that influence the magnitude of SAA impacts. Challenges with the existing farm structure, institutional and regulatory reforms that have not yet been fully aligned with EU requirements to allow for free flow of goods to the EU, as well as difficulties in accessing credit and capital by farmers, food safety considerations, as well a technology packages to improve productivity, remain and continue to inhibit the development of a more competitive agricultural sector in BH. 14 The contribution of the agriculture sector to BH GDP is about 7 percent in 2014, accounting for BAM 1,840 million. Thus the revenue loss is about 3 percent of the latter figure. 10 Looking forward Measuring the impact of tariff changes provides a partial view of the larger issues in the performance of the agricultural sector. Among the potential entry points for strengthening the sector’s performance vis- à-vis trade (and other) shocks are: 1. Improving the performance of the domestic agri-food producers and adding value to agricultural production. This would require actions at the policy level, as well as at the producer level, such as: i) Alignment of existing regulatory frameworks on food safety to EU requirements, access to financial support, etc.; ii) Ability to differentiate between targeting support that has social objectives (providing income support) from measures supporting investments that foster the modernization and competitiveness of the sector; iii) Building the resilience of the sector to weather and price shocks, hence smoothing production cycles. There are both financial and climate-smart production measures that could be taken to manage risk, while improving productivity, mitigating vulnerabilities and improving longer term competitiveness through modernization of systems. 2. Strengthening institutions. BH still suffers from a fragmented and unclear institutional and regulatory set up that needs harmonizing to ensure that policies, regulations and enforcement work as a single unified system. This is particularly so as it relates to food safety in line with EU requirements that point towards the need for clear mandates and responsibilities between all the public actors. While considerable efforts were made in trying to harmonize these aspects with amendments to legislation at State and Entity level over the last couple of years, implementation on the ground of some of these aspects remains pending. 3. Leveraging resources. Some of the tariff revenue loses can be mitigated by attracting alternative funds for the development of the agri-food sector. These resources may come in the form of grants to the government or could be mobilized with the donor community or IFIs. In order to optimize their impact at producer level, such financial assistance would need to be delivered in time with yearly programing and seasonal cycles, be selective and simple in terms of the type of support given with minimum administrative costs and a maximization absorption by beneficiaries. For that to happen all government structures and stakeholders need understand their role in making the support system work. A Rural Development Strategy with a blue print of the paying structure for the country would provide the guidance for setting up such an efficient administrative system to transparently guide such investments down to the producer/farm level. 4. Effective management of a support program depends on effective agricultural information system. While significant effort shave been made in developing databases and information systems under various donor funded programs, these systems remain incomplete or are hampered by institutional hurdles that prevent broad based information sharing within country or with the exterior. EU accession requires a high degree of agriculture related information that forms the basis for policy making. In converging towards more EU compliant support programs such as area based payments, a number of existing data bases such as the farm and livestock register need integration with the Land Parcel and Identification system that then ties into the IACS that will provide the basis for more accurate sector impact analysis and better policy targeting at all administrative levels of BiH. 11 References Brenton, P., Saborowski, C., Staritz, C. and von Uexkull, E. (2009), “Assessing the adjustment implications of trade policy changes using TRIST (Tariff Reform Impact Simulation Tool)”, Policy Research Working Paper 5045, Washington D.C.: The World Bank. Kee, H. L., Nicita, A. and Olarreaga, M. (2005), “Import demand elasticities and trade distortions”, Policy Research Working Paper 3452, Washington D.C.: The World Bank, published in The Review of Economics and Statistics (2008) 90(4): 666–82. 12 Appendix: Additional tables Table A1. Top 10 imports in BH at the HS-4 level of aggregation from selected origins EU27 Croatia % in total CEFTA % in total agricultur % in total Value agricultur Value al imports Value agricultur (BAM al imports (BAM from (BAM al imports million) from EU27 million) Croatia million) from CEFTA Meat of bovine animals, fresh or chilled 143.5 12.8% Beer made from malt 54.3 11.7% Beer made from malt 66.8 7.8% Food preparations not elsewhere specified or included 76.2 6.8% Wheat and meslin 36.6 7.9% Bread, pastry, cakes, biscuits and other bakers' wares 66.5 7.8% Wheat and meslin 65.8 5.9% Chocolate and other food preparations containing cocoa 30.4 6.5% Maize 66.2 7.7% Chocolate and other food preparations containing cocoa 62.2 5.5% Waters, including mineral waters and aerated waters 28.7 6.2% Sunflower-seed, safflower or cotton-seed oil 60.3 7.0% Sunflower-seed, safflower or cotton-seed oil 49.1 4.4% Preparations of a kind used in animal feeding 26.8 5.8% Live bovine animals 56.4 6.6% Meat of swine, fresh, chilled or frozen 48.4 4.3% Food preparations not elsewhere specified or included 22.7 4.9% Preparations of a kind used in animal feeding 43.1 5.0% Cheese and curd 45.7 4.1% Cigars, cheroots, cigarillos and cigarettes 19.5 4.2% Waters, including mineral waters and aerated waters 42.4 5.0% Preparations of a kind used in animal feeding 41.1 3.7% Maize 17.9 3.9% Sunflower seeds 40.1 4.7% Waters, including mineral waters and aerated waters 37.4 3.3% Bread, pastry, cakes, biscuits and other bakers' wares 16.2 3.5% Chocolate and other food preparations containing cocoa 36.2 4.2% Live bovine animals 30.2 2.7% Sauces and preparations therefor 12.8 2.8% Wheat and meslin 31.6 3.7% Total 53.4% Total 57.3% Total 59.6% 13 Figure A2: Detail of Tariff rate quotas (TRQs) applying to EU imports and in place after Feb 2017 TRQs are TRQs are expected expected TRQ TRQ TRQ to be TRQ TRQ TRQ to be (tonnes) (tonnes) (tonnes) binding in (tonnes) (tonnes) (tonnes) binding in CN-8 Description Feb 2017 Jan 2018 Jan 2019 2017? CN-8 Description Feb 2017 Jan 2018 Jan 2019 2017? 01022961 Live cows for slaughter 1,935 1,935 1,935 No 16023119 Processed Meat 40 40 40 No 01022991 Live cows for slaughter 190 190 190 No 16023211 Processed Meat 130 130 130 No 01039211 Live swine 575 575 575 No 16023219 Processed Meat 30 30 30 Yes 01039219 Live swine 1,755 1,755 1,755 No 16023230 Processed Meat 170 170 170 No 01039290 Live swine 195 195 195 No 16023290 Processed Meat 230 230 230 No 01059400 Live chicken 1,455 1,455 1,455 No 16024110 Processed Meat 360 360 360 No 02071290 Meat: chicken 80 80 80 No 16024915 Processed Meat 150 150 150 No 02071310 Meat: chicken 90 90 90 No 16024930 Processed Meat 445 445 445 No 02071330 Meat: chicken 55 55 55 No 16024950 Processed Meat 60 60 60 No 02071360 Meat: chicken 320 320 320 No 16025031 Processed Meat 70 70 70 No 02071399 Meat: chicken 25 25 25 No 16025095 Processed Meat 295 295 295 No 02071420 Meat: chicken 30 30 30 No 17019100 Sugar 55 55 55 Yes 02071460 Meat: chicken 130 130 130 No 17019910 Sugar 3,470 3,470 3,470 No 02071499 Meat: chicken 50 50 50 No 20011000 Cucumbers 265 265 265 No 04014010 Milk 80 80 80 Yes 20019070 Sweet Pepper 70 70 70 No 04015011 Milk 30 30 30 Yes 20059950 Mixed Vege 245 245 245 No 04022118 Milk 25 25 25 Yes 20059960 Mixed Vege 40 40 40 Yes 04039051 Milk 500 500 500 No 04012011 Milk 5,432 9,506 13,580 No 04039053 Milk 290 290 290 Yes 04012091 Milk 720 1,440 1,440 No 04051011 Butter 160 160 160 Yes 04031011 Yogurt 1,515 3,030 3,030 Yes 04051019 Butter 200 200 200 No 04031013 Yogurt 1,520 3,040 3,040 No 04061030 Cheese No 04039059 Yogurt 1,763 3,525 3,525 No 355 355 355 04061050 Cheese Yes 16010099 Sausage 1,693 3,385 3,385 No 04061080 Cheese 165 165 165 Yes 04031091 Milk 480 480 480 Yes 04090000 Honey 165 165 165 No 04031093 Milk 130 130 130 Yes 07019050 Potatoes 50 50 50 Yes 04031099 Milk 25 25 25 No 07019090 Potatoes 1,265 1,265 1,265 Yes 04039091 Milk 530 530 530 No 07049010 White Cabbage 280 280 280 Yes 04039093 Milk 55 55 55 Yes 07061000 Carrots And Turnips 50 50 50 Yes 19053119 Bread 365 365 365 No 08061010 Table Grapes 45 45 45 Yes 19053199 Bread 600 600 600 No 08092100 Fresh Sour Cherries 410 410 410 Yes 19053219 Bread 300 300 300 No 08119075 Processed Sour Cherries 70 70 70 No 19059045 Bread 35 35 35 Yes 16010091 Processed Meat 285 285 285 No 22082029 Brandy 85 85 85 No 16021000 Processed Meat 75 75 75 No 24022090 Tobacco 3,200 3,200 3,200 No 16022090 Processed Meat 140 140 140 No 220410 Quality sparkling wine No 13,765 19,530 220421 Wine of fresh grapes Yes 14 Figure A3: Correspondence between production (FAO) classification and trade classification (HS) data Harmonized Harmonized Harmonized FAO FAO FAO decscriptions System FAO decscriptions System FAO code FAO decscriptions System code code Classification Classification Classification 100111 240110 020311 100119 24011060 02031211 15 Wheat 100191 24011080 02031219 100199 24011090 02031290 100310 826 Tobacco, unmanufactured 240120 020319 44 Barley 100390 24012060 02031915 1035 Meat, pig 100510 24012080 020321 56 Maize 100590 24012090 02032290 100210 240130 02032219 71 Rye 100290 0201100000 02032211 100410 0201100010 020329 75 Oats 100400 0201202000 02032955 100490 0201202010 020711 97 Triticale 100860 0201203000 020712 070110 0201203010 02071310 116 Potatoes 07019050 0201205000 020713 07019090 0201205010 02071330 071331 0201209000 02071360 0713320000 0201209010 02071399 176 Beans, dry 1058 Meat, chicken 0713320010 020210 020714 07133310 867 Meat, cattle 0202100010 02071420 07133390 0202100020 02071460 071339 0202201000 0207149900 222 Walnuts, with shell 080231 0202201010 0207141020 120110 0202201020 236 Soybeans 120190 0202203000 020760 267 Sunflower seed 120600 0202203010 Cabbages and other 070420 0202203090 358 brassicas 070490 0202205000 040711 388 Tomatoes 070200 0202205010 040721 1062 Eggs, hen, in shell 397 Cucumbers and gherkins 070700 0202205020 04079010 07096010 0202209000 04079090 07096091 0202209010 1182 Honey, natural 040900 401 Chillies and peppers, green 07096095 0202209020 02032955 07096099 04012011 020711 07031011 04012019 020712 403 Onions, dry 07031019 04012091 02071310 07031090 04012099 020713 406 Garlic 070320 04014090 02071330 414 Beans, green 070820 04015019 02071360 882 Milk, whole fresh cow 426 Carrots and turnips 070610 04015039 02071399 497 Lemons and limes 080550 04015099 020714 1058 Meat, chicken 515 Apples 080810 04014010 02071420 521 Pears 080830 04015011 02071460 523 Quinces 080840 04015031 0207149900 526 Apricots 080910 04015091 0207141020 530 Cherries, sour 080921 020410 531 Cherries 080929 020421 020760 534 Peaches and nectarines 080930 020422 536 Plums and sloes 080940 020423 977 Meat, sheep 544 Strawberries 081010 020430 040711 547 Raspberries 081020 020441 040721 1062 Eggs, hen, in shell 560 Grapes 080610 020442 04079010 567 Watermelons 080711 020443 04079090 569 Figs 080420 1182 Honey, natural 040900 15