bilaterals.org has received the following four impact studies concerning the EU-West Africa Economic Partnership Agreement:
ITAQA 2008, "Scenario et Résultats du modèle d’équilibre général pour l’étude des accords de Partenariat Économique (APE) entre l’Union Européenne et les Pays de l’Afrique de l’Ouest", Vol I et Vol II, avril + commentary (EN/FR)
ITAQA 2012, "Service Order N° 2: Market Access with EPADP. HS6 Classification of products." April
University of Ibadan 2014, "CGE Modelling of Impact of European Union-West Africa Economic Partnership Agreement on Nigeria", 23 April
IFPRI 2016, Discussion Paper 01502, "The European Union–West Africa Economic Partnership Agreement: Small Impact and New Questions", January
We requested Jacques Berthelot to prepare a comment on these materials, which we publish below.
The EU Commission refuses to publish four reports on the West Africa EPA
Jacques Berthelot (email@example.com), SOL
According to Regulation (EC) No 1049/2001 of the EU Parliament and Council of 30 May 2001 regarding public access to European Parliament, Council and Commission documents, "In order to bring about greater openness in the work of the institutions, access to documents should be granted by the European Parliament, the Council and the Commission not only to documents drawn up by the institutions, but also to documents received by them. Sensitive documents… protect essential interests of the European Union or of one or more of its Member States in the areas covered by Article 4(1)(a), notably public security, defence and military matters". 
In that context one wonders why the DG Trade of the EU Commission has refused to circulate four reports on the impacts of the EU-West Africa (WA) EPA (Economic Partnership Agreement) that it has commissioned and financed, presumably because it did not share their conclusions. In that instance the issues of public security, defense and military matters were not at stake.
However these four reports have been cited and were posted on websites for some time, but have been removed in the meantime. The first two reports were made by the French Consultancy firm ITAQA in April 2008 and April 2012, the third by two Researchers of the University of Ibadan in April 2014 and the fourth by IFPRI in January 2016.
David Laborde of IFPRI (http://www.ifpri.org/profile/david-laborde-debucquet) provides the links to the two ITAQA reports in his curriculum vitae:
1) ITAQA 2008: “Regional CGE modeling for West Africa: an EPA Study”, written by Bernard Decaluwe, David Laborde, Helene Maisonnave and Veronique Robichaud. Report for the EC and ECOWAS secretariats done by ITAQA Sarl. Volumes 1, 2, 3 & 4. April 2008. French and English version. Available at http://epa-model.eu/site/ on request". However when you click at this link, one gets the message: "Sorry, The website will be back online as soon as possible" (http://epa-model.eu/cgi-sys/suspendedpage.cgi).
The 2008 report, based on 2004 trade data, concluded that:
- The trade diversion in favour of imports from the EU reduces the regional integration process, one of the main EPA objectives: at the end of the liberalization process Nigeria’s imports would be reduced by 8.7% from Mali, by 5.7% from Niger, by 5% from Ghana and by 4% from Ivory Coast.
- The losses of customs duties would be of €3.182 bn the last year of the liberalization period (-29.2%) for ECOWAS, of which €1.898 bn for Nigeria (-28.3%).
Nigeria’s investments would be reduced by 12% and GDP by 1.8% at the end of the liberalization period.
The same team made three other reports in 2009 and 2010, complementing the ITAQA report of 2008 and preparing that of 2012 but they cannot be downloaded either:
- “EPA in West Africa: implications of the Interim Agreements of Ivory Coast and Ghana for the both countries and the region” ;
- “Empirical and Theoretical improvements for the regional CGE model of West Africa;
- “Analysis of the Alternative Market Access Scenarios and EPADP for West Africa”.
2) ITAQA 2012: “A study with Market Access and EPADP scenarios using the HS6 model for the West Africa EPA” written by Bernard Decaluwe, David Laborde, Helene Maisonnave and Veronique Robichaud. Report for the EC and ECOWAS secretariats done by ITAQA Sarl. French and English version. Available at http://epamodel.eu/site/ on request", and the link tells you the same "Sorry, the website will be back online as soon as possible".
The ITAQA 2012 report discloses the results of the market access offer simulation proposed by the ECOWAS and communicated to ITAQA in February 2012 and compared the results with a Business as usual (BAU) situation with EU GSP (General System of Preferences) tariffs for the three DCs (Ivory Coast, Ghana, Nigeria) and DFQF (duty free-quota free) EBA (Everything But Arms) for the LDCs (least developed countries).
- Average ECOWAS tariffs applied on products originating from the EU will drop from an average of 10.7% to 5%, a decline by half in 30 years.
- The most significant decreases are shown by Ivory Coast (64%) and Nigeria (55%).
- By the end of the liberation process in 2040, the imports from the EU zone increase by 12%, the same for Nigeria.
- Trade liberalization barely contributes to an improvement in performance of exports from the countries in this zone. The only countries benefitting are Ivory Coast and Ghana because of the suppression of the GSP duties but Nigeria’s situation would deteriorate in the long term.
- There is a decline in intra-regional exchanges, reflecting trade diversion in favor of the EU and to the detriment of regional partners. For products equal in quality and characteristics, the decline in the prices of European imports due to the tariff reductions pushed African importers away from regional producers because they became less competitive than before the tariff cuts.
- For the whole ECOWAS, the loss of total fiscal revenue would be of 8% at the end of the tariff reduction process, of which of 14.8% for Nigeria.
- The impact on real GDP is a direct reflection of the change in investments, with a drop at 2.8% below the BAU (business as usual) level in 2040 for Nigeria.
3) The third report on the EPA impact on Nigeria was prepared by Adeola Adenikinju and Abiodun Bankole of the University of Ibadan and published on April 23, 2014: CGE modelling of impact of European Union-West Africa Economic Partnership Agreement on Nigeria. It took into account the last data on the ECOWAS CET (Common external tariff) and the EPA text, with the following conclusions for Nigeria:
- Customs duties would fall from 10% at the start of liberalization to 35% at the end (2035).
- Total budget revenues would fall by 0.5% at the start and by 5% at the end.
- GDP would fall by 2% ($8.3 bn), which is considerable.
- Unemployment would rise by 15%, with reductions in wages and consumption.
- The interest rate would increase to 25% in 2035, with a negative impact on both public and private investment.
- Trade diversion mentioned in a 2011 study is confirmed, with imports from the rest of ECOWAS falling by 3% to 4%, which would reduce regional integration that the EPA claims to promote.
4) The fourth report was prepared by Antoine Bouet, David Laborde and Fousseini Traoré of IFPRI which put it on its website in January 2016 or early February. It was picked-up by the Technical Centre for Agricultural and Rural Cooperation (CTA) in Brussels which announced it in its Newsletter that provided the following link on 22 February 2016: http://www.ifpri.org/publication/european-unionwest-africa-economic-partnership-agreement-small-impact-and-new-questions
It is likely that only few people were able to download it as the link worked only for a few days. However J. Berthelot did and, unaware of its withdrawal, on 23 April made some comments on excerpts of the report (on 4 pages) and circulated them widely together with the report itself, assuming that the report was already largely known. The more so as the IFPRI report was published with the following notice: "IFPRI Discussion Papers contain preliminary material and research results and are circulated in order to stimulate discussion and critical comment. They have not been subject to a formal external review via IFPRI’s Publications Review Committee. Any opinions stated herein are those of the author(s) and are not necessarily representative of or endorsed by the International Food Policy Research Institute".
What follows are a very short summary of the report and Berthelot’s more extensive comments.
In short, the authors’ conclusions are the following:
- "Public revenues from import duties are negatively affected by the reform, from –7.5 percent in Benin to –25.8 percent in Burkina Faso, which can affect the ability to finance public services.
- The results concerning welfare are negative for Nigeria, Senegal, Benin, the Rest of ECOWAS region, and Togo and positive for Burkina Faso and Côte d’Ivoire.
- The benefits of the EPA between the EU and WA’s countries appear small, if not negative".
Here are the more extensive comments based on excerpts of the IFPRI Report:
IFPRI Discussion Paper 01502, January 2016
The European Union–West Africa Economic Partnership Agreement
Small Impact and New Questions
By Antoine Bouët, David Laborde, Fousseini Traoré
This study has been prepared at the request of the Chief Economist Unit of DG Trade to provide economic analysis in support of trade negotiations and trade policy issues related to the Economic Partnership Agreement between the EU and the West Africa–Economic Partnership Agreement group.
[Excerpts from this Discussion Paper and comments by J. Berthelot (SOL), 23 April 2016]
[These excerpts are sufficient to understand why the EU Commission has refused to publish this report that it commissioned and financed. SOL’s comments add more reasons to refuse the WA EPA, based on the biases and contradictions of the report itself. The excerpts are in black and the comments are in blue within square brackets]
African countries are well known for having a restricted domestic fiscal base, so changes to public revenues due to the EPA could endanger the government’s ability to provide public goods.
In our evaluation we implement the application of the GSP tariffs by the EU on products coming from West African countries in the baseline, while these tariffs are removed in the scenario describing the implementation of the EPA.
The EPA between West Africa and the EU… includes… (4) no subsidization of any European agricultural exports to WA
[which is a huge misrepresentation as all EU agricultural products receive large domestic subsidies which benefit as well to the exported products. In 2015 the EU cereals exported to WA have received €232 million of subsidies and the dairy products €72 million, whereas meats and eggs received €162 million in 2014.]
… and (6) a development assistance package (EPADP) to complement market-opening efforts.
[A misrepresentation as it is only a repackaging of already traditional EU cooperation funds].
Unlike in the agreements with other ACP regions, the EU also has committed to support West African countries with a €6.5 billion development assistance package covering the period from 2015 to 2020.
[Not true, as attested by the EU Commission DEVCO brochure of July 2015: "From 2014-2020, €6.5 billion will be delivered to support PAPED… The funds are drawn from the existing EU financial instruments: 11th European Development Fund National Indicative Programmes (NIP), Regional Indicative Programme (RIP), intra-ACP programme, and relevant EU thematic budget lines" (https://ec.europa.eu/europeaid/sites/devco/files/epa-brochure_en.pdf). With contradictory statements about the length of the fund: 5 years, 6 years or 7 years.]
The EU also has committed to assisting its African partners in the fiscal transition process, as the loss of tariff revenues is a key challenge for these countries.
[Indeed article 60.3 of the EPA agreement foresees "a significant contribution to absorbing net fiscal impact in full complementarity with the fiscal reforms" but this is not credible as there is not any additional euro in the EU Budget for that.]
EU exporters thus appear to be the main beneficiaries of the agreement.
One of the main findings of these studies is the loss of public tariff revenues for ACP countries as the agreement enters into force. This loss of tariff revenues is due to both the elimination of custom duties on imports from EU and trade diversion effects (replacement of imports from the rest of the world by untaxed imports from the EU). The impact on public revenues depends on the initial share of tariff revenues in overall government revenue; the average loss of tariff revenues is projected to be high, at 25 percent in the long run for ACP countries and 38 percent for ECOWAS countries (Fontagné, Laborde, and Mitaritonna 2011).
Most African economies have a relatively low rate of effective tax collection; thus, considering the tariff revenue to be equal to the product of nominal tariff and trade flows is a huge overestimation. When available, we use a country-sector-specific efficiency ratio; otherwise, we use a country-level efficiency ratio and aim to duplicate effective tariff revenues as indicated in International Monetary Fund financial reports.
[But assuming that there is a low rate of effective import duties collection assumes also that the ECOWAS authorities are unable to improve the effectiveness and integrity of customs services and officers and it is above all an easy way to underestimate the detrimental impact of the losses of import duties of the EPA.]
The Development Package In addition to having a market access component, the EPA between WA and the EU includes a DP known as EPADP. This involves a combination of aid-for-trade, infrastructure improvements, and upgrades to the region’s production capacities. These different measures are mapped into key structural variables (for example, productivity increase and reduction of trade costs) of the model. We propose the following approach to represent this program: a DP is included in both the baseline and the scenario since the European Commission intends to adopt a DP regardless of the result of the negotiations.
[Another easy way to hide that there won’t be any additional euro for the EPADP above the traditional EDF, EIB loans and EU Cooperation funds].
The MIRAGRODEP model is mainly built on the GTAP database. We first rely on the GTAP 8.1 database, which is the latest version available. This database contains the social accounting matrixes for eight West African countries (Benin, Burkina Faso, Ghana, Côte d’Ivoire, Niger, Nigeria, Senegal, and Togo).
[Which implies already that this model misrepresents all the 16 WA countries.]
In the baseline, GSP prevails for Ghana, Côte d’Ivoire, Cape Verde (after the phasing out of the LDC status), and Nigeria. Other countries benefit from the EBA preference.
[Wrong: the GSP+ status of Cape Verde since January 2014 grants it the same DFQF (duty free-quota free) market access to the EU as to the LDCs, particularly for its processed tuna exports when the GSP countries have to pay a 20.5% ad valorem duty. Thus Cape Verde exported 1,200 tonnes of processed tuna to the EU in 2015 at zero duty when Ivory Coast would have had to pay under GSP if the EPA is not implemented €25.8 M for its 26,097 tonnes and Ghana €35 M for its 33,679 tonnes]
EPADP consists of an “envelope” of €1.3 billion per year from 2015 to 2035 given by the European Commission.
[Not politically feasible as the Cotonou agreement expires in 2020 and the Commission cannot guarantee that the EU will continue to grant as much funds to WA, which would furthermore take into account the WA high population increase].
This envelope is given to ECOWAS countries both in the baseline and in the scenario
[which is another way to confirm and to hide that the EPADP will not add a single euro to the already programmed funds of EDF, EIB and EU cooperation funds].
This trade agreement entails an asymmetric opening of trade borders. West African countries open their borders to European products more than the EU opens its borders to West African countries… This confirms that the trade agreement entails an asymmetric opening of trade borders, to the disadvantage of West African countries.
Concerning the protection applied by Europe on products from West African countries by sector (Table 4.5), the sectors in which West African countries benefit most from improvements in access to European markets are dairy products (Côte d’Ivoire, Ghana, Nigeria), sugar (Côte d’Ivoire), rice (Côte d’Ivoire, Nigeria), and red meat (Côte d’Ivoire, Ghana, Nigeria); these three countries (Côte d’Ivoire, Ghana, and Nigeria) are the non-LDC countries from ECOWAS.
[A ridiculous statement as these products are not exported at all by WA countries which, to the contrary, have to import huge amounts of them from the EU and from the rest of the world!]
Senegal does not see any improvement in access to either EU or non-EU destinations; rather, the EPA leads Senegal to open its borders to European products. This increases Senegal’s imports and could deteriorate the current account balance. Since the model assumes that Senegal’s current account balance remains constant, this implies a depreciation of the real exchange rate, which increases Senegal’s competitiveness toward other destinations. Due to the EPA, in 2035, Senegal’s exports to Asia increase by 3.8 percent, to the Middle East and North Africa by 2.5 percent, and to the North American Free Trade Agreement by 2.9 percent; these figures are not presented in Table 4.7.
[But assuming that Senegal’s current account remains constant is not feasible! And that its deterioration implies a depreciation of its real exchange rate is not feasible either, first because the CFA franc has a fixed exchange rate with the euro (655.957 CFA F for one euro), which would then imply that this depreciation would take the form of lower production costs, particularly of wages, which would not be a real benefit for Senegal!]
On a large-sector level (agro food, industry, services), Table 4.8 shows how trade is affected for the main countries/regions of concern. Any rate of variation may be misleading since it can correspond to a small initial value: for example, Nigerian exports in the agro food sector augment by 7.2 percent in 2035 due to the implementation of EPA, but in the baseline in 2035, agro food exports represent only 0.5 percent of total Nigerian exports of goods and services.
[This type of argument that Nigerian food exports could increase forgets completely the huge population increase of Nigeria (by 61.3% from 2015 to 2035) and of all WA countries (by 64.4%) and their actual large food deficit, despite that Ivory Coast and Ghana have a food surplus with the EU owing to their exports of cocoa, bananas and processed tuna, but Nigeria has a food deficit with the EU.]
It is worth noting that Côte d’Ivoire’s and Ghana’s agro food exports are significantly raised by the EPA, as are industrial imports in Burkina Faso and Côte d’Ivoire.
[This type of reasoning forgets completely the expected loss of competitiveness rising from the other FTAs recently signed by the EU with 3 Andean countries (Colombia, Peru, Ecuador), 6 Central American countries, and its on-going negotiations with the US (TTIP), Canada (CETA), not to forget with India, Mercosur, the Philippines. Not to speak of the climate change which could lower the production of bananas and plantains according to the IPCC report of 2014.]
Public revenues coming from the collection of import duties are negatively affected by the reform, from –7.5 percent in Benin to –25.8 percent in Burkina Faso. This is a key implication of this trade agreement since it can potentially affect the ability of Africa’s public sectors to finance public services.
Let us remember that in this central scenario, we do assume that the public sold is constant in proportion of GDP and that a lump-sum tax is implemented to offset the loss of custom duties, such that when public revenues are cut, real public expenses per capita remain constant. This lump-sum tax is in a range of $2.41 (Burkina Faso) to $17.51 (Ghana); it is $5.02 in Nigeria, $14.7 in Senegal, $11.79 in Benin, $4.95 in Côte d’Ivoire, $6.02 in the Rest of ECOWAS region, and $12.37 in Togo.
We focus on unskilled labor since this is the most important productive factor for a poverty analysis. In Nigeria, Senegal, Benin, and Burkina Faso, the nominal remuneration of this factor is reduced by the reform due to less demand for unskilled labor. In Nigeria, this is due to a reduction in the production of the other crops sector (Table 4.12) by 4.5 percent, as this sector absorbs 4.4 percent of total unskilled labor (2035—baseline), and the demand for this factor is reduced by 4.7 percent.
The results concerning welfare are negative for Nigeria, Senegal, Benin, the Rest of ECOWAS region, and Togo and positive for Burkina Faso and Côte d’Ivoire. In all West African countries, households are positively affected by a decreasing consumption price index (see Table 4.13) but negatively affected by a lump-sum tax needed to maintain public expenses and public sold constant.
For non-LDCs, the implementation of the EPA improves access to European markets for local exporters.
[Given that WA exports to the EU are predominantly food products, the report forgets to mention that the EU population would at best stabilize (the UN expects that it will decline after 2030) and will get older, implying that food consumption per capita will decline so that there is little chance that EU imports would rise, even more from WA which would become less competitive than other countries with which the EU has just signed FTAs or will do in the coming years.]
European goods are cheaper for local consumers and thus improve their purchasing power; as a result, local consumers may increase their consumption and demand for local products.
[Again contradictory statements: if imports from the EU rise and WA consumers’ purchasing power increases as a result, this won’t be true for all: the lower demand for regional products would impoverish the producers, which will be observed in particular for most farmers having to suffer the competition of increased imports of wheat, maize, barley and milk powder.]
In the EXP scenario, public expenses are decreased (by 8.5 percent in Senegal and Togo in 2035 and by 4.5 percent in Benin in 2035) to maintain the public deficit in proportion of GDP constant after the loss of custom duties… We may assume that such a reduction could jeopardize the provision of public goods in all West African countries. In other scenarios, we impose a new tax in the model that is levied to maintain real public per capita constant. Even if the lump-sum tax is not politically realistic, it clearly illustrates how much each citizen has to pay for public good to be maintained. This tax decreases individual private available income such that in six of eight countries/regions, welfare is reduced. Compensating for the loss of custom duties by an additional income tax results in a similar scenario.
If consumption prices are reduced due to the reform, the loss of custom duties requires the implementation of a new tax, which necessarily reduces welfare.
Third, the EPA agreement raises the issue of a fiscal adjustment. Custom tariffs represent an important part of public revenues in ECOWAS countries. However, the EPA entails a significant reduction of custom duties, since the EU is an important and significant trading partner in the region. To maintain public expenses and the provision of public goods constant, ECOWAS governments will have to find an alternative source of public revenues.
In conclusion, the benefits of the EPA between the EU and WA’s countries appear small, if not negative. West African countries should find a source of increased growth from other trade agreements, either multilateral or regional.
[Solidarité (now SOL) has just published an alternative assessment of the EU-West Africa EPA: The folly to implement the EU-West Africa Economic Partnership Agreement (EPA), based on 2015 trade data, 22 April 2016]