Trade Negotiations Insights • Volume 9 • Number 6 • July 2010

Addressing the fiscal challenges of an EPA: Some preliminary considerations

by ECDPM

Discuss this articleShare your views with other visitors, and read what they have to say

The debate around the impact of EPA tariff liberalisation on government revenues in ACP countries has been particularly controversial throughout the negotiations. A great deal of the opposition to EPAs has focused on the argument that the agreements are likely to have serious effects on government revenues, and therefore on social expenditures geared towards the achievement of broader development objectives, such as the Millennium Development Goals. The financial and economic crisis, combined with recurrent food and energy crises, has only heightened these concerns. Attempting to quantify the fiscal impacts of EPAs has therefore been an important, yet sometimes difficult, exercise.

In terms of reaching a comprehensive assessment of the fiscal impacts of EPAs, the picture remains relatively unclear despite a growing number of regional and country-level studies on the subject. Differences in methodology, data and assumptions have contributed to a mixed picture, although it is also important to acknowledge that comparison and debate between various studies is healthy, and no forecast is likely to be definitively accurate in light of the uncertainties involved in making them. One overall conclusion, however, is that there has been a general tendency to overestimate the direct fiscal losses resulting from an EPA, at least in some of the earlier studies as compared to more recent analysis. Nonetheless, the implementation of an EPA can have significant, and in some countries very serious, consequences for government revenues.

Beyond efforts of researchers to quantify the tariff revenue losses that might result from EPAs, surprisingly little discussion has taken place thus far on the more practical and policy-related questions of how to address the fiscal impact of the agreements. While giving estimates of the impact that liberalisation will have on revenues, most studies do not delve deeper into implications of this work, and ask how trade taxes might be replaced as implementation proceeds. For some countries, replacing remaining trade taxes might become a pressing concern in the next few years, especially given the substantial liberalisation that has already taken place over the last few decades.

In terms of the policy responses that will be needed, these are likely to be fairly context specific, depending naturally on the scale of expected losses but also on a range of other factors, such as the existing tax mix, the strength of the economy and breadth of the tax base, tax-raising capacity (in terms of the ease of implementing new taxes or strengthening the collection of others), or even whether there is scope for current spending levels to be reduced as government services are delivered more efficiently. The debate should thus shift away both from theory and rhetoric, towards more concrete actions to address the potential revenue shortfalls, with a role for all stakeholders, including governments, the private sector, researchers and donors. What strategies can ACP countries and their partners adopt to mitigate the negative consequences of liberalisation and take advantage of opportunities for wider reform or consolidation in the area of taxation? Below are some preliminary considerations, with illustrations from Tanzania and Mozambique that highlight the differences and similarities at the country level between different ACP countries on the revenue consequences of EPAs.

The reform context: varying expectations and perceptions

Given that the various models of EPA revenue losses lack perfect foresight and can offer only tentative indications on what to expect, it is perhaps unsurprising that opinions amongst officials on the issue also tend to differ amongst and within ACP countries. In some parts, the views of officials and stakeholders at the country level often continue to reflect long-held positions on the consequences of EPAs more generally, with ‘EPA sceptics’ highlighting the potential negative effects of EPAs on revenues and competitiveness. In contrast, ‘EPA supporters’ put greater emphasis on the dynamic opportunities and gradual transition process. Underpinning such views, however, is the fact that the context of reform is very different in ACP countries, depending on factors like the initial baseline position, ongoing dynamics of regional integration, and the available options for reform.

The case of Tanzania

With regard to fiscal losses as a result of implementing EPAs, Tanzanian officials familiar with the negotiations emphasise the significant proportion of revenue that continues to be derived from taxes of all kinds (i.e. not just customs duties) on imported goods: currently some 43% of total revenues originate from imported goods. Not all of this revenue is under threat however, since much of the tax collected comes from taxes other than import duties, such as VAT and excise taxes also levied on imported goods, which are not affected by EPA liberalisation requirements, at least directly. Furthermore, flexibility in the EPA liberalisation schedule means that some goods are excluded from liberalisation - in Tanzania’s case, these amount to around 17% of imports, including some ‘big ticket’ revenue-generating products on which import duty will continue to be applied.

Conversely, it is important to understand that the elimination of duties will not just lead to lower customs revenue itself, but also reduce the basis for calculations of other ad valorem taxes, most notably VAT, which are generally levied after duties have already been applied. In addition, the issue of VAT has in recent years been a sensitive item on the reform agenda in Tanzania, with a reduction in the rate from 20% to 18% to bring it closer into line with regional EAC partners, further reducing fiscal revenues.

More generally, many Tanzanians fear the potentially adverse effects that trade liberalisation may have on the economy and tax base as a whole. They are worried that increased competition with EU suppliers may result in the collapse of some domestic industries, including some major tax-paying firms, leading to declining levels of corporate income tax collections, as well as lower collection of personal income taxes as a result of reduced levels of employment.

From a political point of view, the difficulty for hesitant governments facing the prospect of trade liberalisation is that the potential for revenue losses are immediately apparent, while the benefits will only be seen over time. While the dynamic effects of an EPA are unpredictable, opponents to an EPA can easily highlight the potential revenue losses through simple calculations of forgone import duties. Combined with parallel trade liberalisation processes at the regional level and global crises (economic, food, energy, etc.), some political leaders might be reluctant to engage in a far reaching agreement with the EU whose benefits are uncertain.

One potential first step in such cases would be to conduct more in-depth assessments of the impacts of EPAs - at the country level, perhaps jointly commissioned - with a focus not on specific estimates, but rather on the robustness or fragility of countries’ revenue base, or the overall economic costs and benefits.

The reform context in Mozambique

In contrast to their counterparts from Tanzania, officials from Mozambique who have followed EPAs tend to view the revenue consequences of the agreements for their country as being relatively limited, for a number of reasons that illustrate the diversity of countries’ experiences. In the first instance, Mozambique’s officials point to their lower reliance on tariff revenue over recent years - and the fact that regardless of the controversies surrounding an EPA more generally, the agreement fits well within the pursuit of economic reforms, including trade and tariff liberalisation and tax reform, resulting in a lower reliance on customs duties and a more open and liberalised economy, which is a key element of the government’s development policy.

Secondly, officials point out that regional integration, principally in the SADC context, will have an important influence on customs duty revenues in coming years, in many ways regardless of EPA liberalisation. SADC (and mainly South Africa) is now Mozambique’s foremost trading partner, with around 40% of imports coming from the region - by contrast imports from the EU represented 12% of total imports in 2007, falling from 15% in 2005.

Finally, Mozambique officials point out that many of the goods actually being liberalised under the EPAs - such as capital goods or intermediate inputs - typically enter Mozambique under special duty exemptions given to large investment or aid projects. One consequence of this is that while some theoretical models might point to Mozambique losing fairly significant amounts of revenue based on assumptions about how much customs duty is collected, and on data that does not reflect recent shifts in the structure of trade - in reality the losses could likely to be less severe. Even where liberalisation of such goods does result in revenue losses, officials point out that at the same time the reductions in prices are more likely to encourage development rather than create negative competition, in line again with Mozambique’s general development strategy.

Strategies to address fiscal losses related to EPA implementation

Solutions that have so far been identified for dealing with the revenue consequences of EPAs include those that have been proposed in the context of negotiations - seeking to establish close links between liberalisation commitments and some compensatory financial assistance combined with support to fiscal reforms, as part of Aid for Trade. Beyond this, an emphasis should also be put on more general reform measures aimed at increasing the tax-raising capacity of ACP countries, which are further removed from the specific issue of EPAs and form part of a broader fiscal reform and domestic resource mobilisation agenda. Measures here can range from building the administrative capacity of tax administrations to target specific groups of taxpayers, to tackling international tax issues - such as tax evasion by multinational actors and (illegal) capital flight - at a bilateral or global level. Different approaches are often complementary, and include:

Financing mechanisms for ‘direct compensation’ of EPA net fiscal losses: here it is interesting to note some precedents where the EC has undertaken to provide direct support to provide partial compensation for fiscal losses due to trade liberalisation, either in the context of EPAs - as in West Africa[2], or in the context of regional trade liberalisation - as in East Africa through the Regional Integration Support Mechanism (RISM) programme.

EPA ‘accompanying measures’ leading to higher tax collection through economic growth: Aid for trade for EPA implementations and adjustments will be key to allow ACP economies to reap the potential benefits of an EPA, which in turn would lead to higher economic growth and associated tax collections.

Wider tax reform to broaden the tax base and increase compliance: EPAs need also to act as a catalyst for or build on wider reforms in ACP countries, including on fiscal reforms, in particular as part of a strong emerging agenda on domestic resource mobilisation in ACP countries.[3]

Going forward, one key challenge will be the need for political leadership and commitment to address EPA-related fiscal adjustments and broader fiscal reforms, as an integral part of a domestic resources mobilisation agenda, with appropriate development and technical support.

[1] This article draws on some initial analysis conducted by ECDPM for a project funded by IrishAid on the fiscal adjustments resulting from EPAs. The final study, written by San Bilal, Dan Lui and Melissa Dalleau, will be published in the coming weeks. Comments and suggestions are welcome; contact: San Bilal: sb@ecdpm.org

[2] See the article by David Laborde in this issue.

[3] See the article by Henri-Bernard Solignac-Lecomte in this issue.