Will Economic Partnership Agreements Increase Poverty? The Case of Uganda

• Author: Alan Matthews,Ole Boysen
• Date: 01 May 2017
• Published date: 01 May 2017
• DOI: http://doi.org/10.1111/rode.12272

Will Economic Partnership Agreements Increase

Poverty? The Case of Uganda

Ole Boysen and Alan Matthews*

Abstract

Economic Partnership Agreements (EPAs) between the EU and African Caribbean and Pacif‌ic countries

are frequently criticized because of fears about negative implications for economic development. Using

Uganda as a case study, this paper employs an integrated computable general equilibrium-

microsimulation model framework rich in household-level detail to assess the consequences of the East

African Community EPA for economic output and poverty in Uganda. Simulations of the agreement’s

tariff liberalization provisions indicate a very small negative economic impact and ambiguous outcomes

for poverty. The poverty results depend in size and sign on the poverty line, on the way the government

addresses tariff revenue losses and on labor market assumptions.

1. Introduction

African, Caribbean, and Pacif‌ic (ACP) countries have been negotiating Economic

Partnership Agreements (EPAs) with the EU since 2002 in what has been

described as a turbulent process (Hurt et al., 2013). EPAs were foreseen in the

Cotonou Agreement as a replacement for the non-reciprocal trade preferences

ACP countries long enjoyed on the EU market, but which required a World Trade

Organization (WTO) waiver because they did not comply with WTO rules.

Negotiations were due to be completed by the expiration of the WTO waiver at the

end of 2007 but by this date only one EPA had been concluded—with the

Caribbean ACP countries (CARIFORUM). Interim EPAs were initialed by 20

other ACP countries to allow them to benef‌it from duty-free and quota-free market

access under the EU Market Access Regulation (EC) 1528/2007, but 43 ACP

countries did not sign any agreement. These were mainly least-developed countries

(LDCs) that already enjoyed duty-free access to the EU market under the

Everything but Arms (EBA) arrangement. By the 1 October 2014 when the market

access benef‌its under Regulation (EC) 1528/2007 were due to expire,

comprehensive EPAs had been concluded with the Caribbean countries, West

Africa, the Southern African Development Community and the East African

Community (EAC) while interim EPAs continue in place with individual ACP

countries in other ACP regions (European Commission, 2015c).

The ACP reluctance to sign EPAs was based on a number of fears about possible

adverse effects on their economies, see, e.g. Hinkle and Schiff (2004), Bilal and

Rampa (2006), Busse (2010), or Oxfam (2006). These include the possible welfare-

*Boysen: Department of Agricultural and Food Policy, University of Hohenheim, Germany, E-mail:

boyseno@tcd.ie. Matthews: Department of Economics, Trinity College Dublin, Republic of Ireland. The

f‌irst author is grateful to the International Food Policy Research Institute (IFPRI) for facilitating access

to the household survey data during an extended research stay. Furthermore, we are especially grateful

to James Thurlow for providing the 2007 Uganda Social Accounting Matrix (SAM).

Review of Development Economics, 21(2), 353–382, 2017

DOI:10.1111/rode.12272

©2016 John Wiley & Sons Ltd

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

public goods and government services, the potential costs of acceding to EU

demands in the areas of government procurement, investment regulations, services

liberalization and intellectual property protection and, in general, the risk that these

agreements would exacerbate rather than reduce overall poverty levels. The

existing literature has used a variety of modeling approaches to examine these

concerns, without any overall consensus emerging (see, e.g. Babula and Baltzer,

2007; Bilal et al., 2012; Curran et al., 2008; Fontagn

e et al., 2011; ODI, 2006).

In this paper, we examine a sub-set of these issues for Uganda using an

integrated computable general equilibrium-microsimulation (CGE-MS) model.

Uganda concluded its EPA negotiations with the EU on 16 October 2014 as part of

the EAC.

1

The agreement is essentially a tariffs-only agreement, leaving “behind

the border” issues such as services, government procurement, competition and

intellectual property issues for further negotiations over the next 5 years (European

Commission, 2015b). Our focus is thus on the impact of the tariff provisions

included in the EPA. These give the EAC countries including Uganda duty-free

and quota-free access to the EU market for 100% of their exports, while reducing

duties to zero on a substantial share of EU imports into Uganda over a very

lengthy transition period. While we calculate the likely impacts on Ugandan

economic structure and output, we emphasize particularly the likely impacts on

poverty and the various factors that might inf‌luence this.

Because Uganda was entitled to duty-free access to the EU market anyway under

the EBA, signing the EPA represents a form of unilateral tariff liberalization by

Uganda. Basic trade theory predicts that unilateral import tariff liberalization will

increase welfare by reallocating resources to more eff‌icient uses. This result

depends on the standard assumptions regarding the operation of markets but also

on the absence of other trade distortions. A preferential agreement like the EPA

where only tariffs against selected importers are reduced may lead to an overall

decrease in welfare owing to trade diversion according to the theory of the second

best (Hinkle and Schiff, 2004). Similar second-best effects may occur if a country

liberalizes just some of its tariffs but maintains protection of sensitive sectors.

Liberalization of the more eff‌icient sectors can lead to production shifting to the

more ineff‌icient sectors and result in a welfare loss. Moreover, if the government

replaces the import revenue loss through other taxes, the new taxes could be even

more distorting leading again to a welfare loss. Thus, even the direction of the

national-level welfare impact of the partial liberalization agreed under the EPA is

not clear.

The trade liberalization impacts can be translated into poverty impacts via three

main channels as identif‌ied by McCulloch et al. (2001): the price transmission,

enterprise and government channels. The price transmission effect changes the

prices of the liberalized goods. If these price changes are ref‌lected in the prices of

the goods purchased by poor households, the direct effect will depend on whether

poor households are net consumers or producers of these goods. The enterprise

channel is where trade liberalization affects households through its impact on

enterprise prof‌its and hence employment and wages. The government channel is

where trade liberalization affects poverty through changes in the government’s

354 Ole Boysen and Alan Matthews

©2016 John Wiley & Sons Ltd