ABSTRACT
During the Economic Partnership Agreement (EPA) negotiations with ACP countries a range of fears were raised about the possible adverse effects on their economies (e.g. Bilal and Rampa, 2006; Delpeuch, 2007; Oxfam, 2006). These included the possible welfare-reducing effects of trade diversion from more efficient third-country suppliers to EU exporters, the possibility that more competitive EU imports would undermine local industry and lead to a process of de-industrialization, the impact of competition from subsidized EU food production on domestic food security, the potential impact of the loss of tariff revenue on EU imports for the provision of government services and, in general, the risk that these agreements would exacerbate rather than reduce overall poverty levels. Recent literature has used a variety of modelling approaches to examine these concerns, without any overall consensus emerging (see ODI, 2006, and the references cited therein). This chapter addresses these concerns for the case of Uganda. We intend to analyse the poverty impact of the trade provisions of a full EPA between the EU and the East African Community (EAC) of which Uganda is a member. For the interim EPA signed in December 2007 to be compatible with WTO rules, ‘substantially all trade’ between the EU and the EAC must be liberalized. The European Commission interprets this phrase to mean that 90 per cent of the bilateral trade value must be liberalized where the liberalization can occur asymmetrically.1 As under the interim EPA the EU will abolish 100 per cent of its tariffs on EAC imports (with transitional periods for rice and sugar), the EAC has to liberalize 82.1 per cent of its imports.2 The EAC’s tariffs on EU imports will be phased out gradually in three tranches starting in 2010 and completed in 2033; the first non-zero tariffs will be eliminated in 2015. The ultimate list of those sensitive tariff lines which will be exempt from liberalization at the conclusion of the transition period is not yet known. Stevens et al. (2008) find that 593 tariff lines (Harmonized System) where Uganda had actual imports – corresponding to 42.9 per cent of Ugandan import value from the EU – are not listed in the interim EPA, either for liberalization or exemption. Hence, the agreement is rather preliminary at this stage. In the absence of an elaborated EAC tariff liberalization schedule, we have constructed three alternative scenarios for such possible schedules, reflecting
imports. We use the nominal 2006 revenue from each tariff line as a reflection of the sensitivity the government has attached to the product in the past. The EPA-EAC scenario optimizes the list of sensitive products with respect to overall EAC interests, i.e. the tariff revenue for the EAC as a whole is maximized; the EPA-UGA scenario gives higher weight to Uganda’s interests; and finally the EPA-AG scenario prioritizes the protection of Uganda’s agricultural sectors. We examine the likely impact of each scenario on the poverty headcount, and the depth and severity of poverty, in Uganda. Uganda as a least developed country (LDC) has enjoyed duty-free access to the EU market under the EU’s Cotonou Agreement until its cessation and replacement through EPAs and, since 2001, under the Everything but Arms (EBA) scheme, which remains in place indefinitely. We thus simplify the analysis of measuring the poverty impacts of the EPA’s trade provisions by examining solely the requirement that Uganda as an EAC member reduces over time its tariffs on EU imports.3 We perform our simulations using a combined computable general equilibrium (CGE) microsimulation model which enables the quantification of the adjustment impacts on the economy following EPA liberalization and the impacts on overall poverty. The Uganda CGE model is a static, nonmonetary model based on a 1999 Uganda social accounting matrix (SAM). The SAM has been updated to better match the factor income shares as observed in the household survey. A pre-experiment generates a synthetic SAM for 2006 which takes the 2006 import tariffs into account, in particular Uganda’s implementation of the EAC custom union’s common external tariff (CET) and internal EAC tariff elimination. Starting from this synthetic baseline SAM, the CGE model provides the post-simulation factor returns and commodity prices for the microsimulation that follows in a second step. The microsimulation projects these figures onto a detailed household income distribution derived from the 2002/3 Uganda National Household Survey (UNHS), generating a counterfactual income distribution for poverty analysis. While recognizing the limitations of the model used in this chapter, we conclude that the introduction of the trade provisions of an EPA between Uganda and the EU would have a very small poverty effect which may be positive or not depending on the list of sensitive products exempt from liberalization. The small magnitude of the impacts is driven, in part, by the relatively low share of the EU in current Ugandan imports and the relatively low average tariff which these imports currently face. Much of the effect would be due to a relative shift of resources out of import-competing sectors and into coffee production. But despite the possibility of reducing the poverty headcount, the very poorest would lose income in all of our EPA scenarios. It is important to underline that these are comparative static results, and that these effects in practice would play out over an extended 25-year time scale in which the nature of the Ugandan economy will undergo substantial change which will likely dwarf the poverty impacts which we identify.
