Trade Reform in the CEMAC: Developments and Opportunities

Contributor Notes

Authors’ E-Mail Addresses:;

This paper provides an update on the main elements of the reform agenda concerning the CEMAC trade regime as well as a tentative quantitative assessment of selected effects on tariff revenues and trade patterns. Notwithstanding data limitations, the key messages from the analysis are as follows. First, there is a need for a renewed political commitment to regional integration. In addition, key measures for improving compliance with the requirements for a customs union need to be introduced, including limiting tariff exemptions, phasing out remaining surcharges, strengthening the determination of products' country origin, and enhancing customs administration. There is also a need to improve transportation infrastructure and organization. Finally, there is a strong case for tariff reduction, with or without an EPA. Trade liberalization would help boost economic growth and poverty alleviation and limit risks of trade diversion with an EPA. Tariff reform should be complemented by improvements in domestic revenue mobilization.


This paper provides an update on the main elements of the reform agenda concerning the CEMAC trade regime as well as a tentative quantitative assessment of selected effects on tariff revenues and trade patterns. Notwithstanding data limitations, the key messages from the analysis are as follows. First, there is a need for a renewed political commitment to regional integration. In addition, key measures for improving compliance with the requirements for a customs union need to be introduced, including limiting tariff exemptions, phasing out remaining surcharges, strengthening the determination of products' country origin, and enhancing customs administration. There is also a need to improve transportation infrastructure and organization. Finally, there is a strong case for tariff reduction, with or without an EPA. Trade liberalization would help boost economic growth and poverty alleviation and limit risks of trade diversion with an EPA. Tariff reform should be complemented by improvements in domestic revenue mobilization.

I. Introduction

The creation, in 1994, of the CEMAC customs union, was a major step in the regional integration process in central Africa. The reform was part of a wider initiative to boost regional integration and policy effectiveness in conjunction with the devaluation of the common exchange rate relative to the French franc. The CEMAC replaced the UDEAC (Union Douanière des Etats d’Afrique Centrale), which was created in 1964 but had remained largely ineffective. The 1994 reforms introduced (i) a common external tariff (CET), (ii) the gradual removal of tariffs on intra-regional trade (completed in 1998), (iii) the harmonization of indirect taxation (with the introduction of a VAT in 1999); and (iv) the replacement of quantitative import barriers by temporary import surcharges (to be phased out by 2000—see below).

The implementation of the agreed regime by the member countries, however, has remained unsatisfactory. Trade is hampered and distorted by cumbersome and costly border procedures, as well as national restrictions and exemptions in defiance of the common rules. Further problems stem from the lack of transportation infrastructure, security problems and, for trade with the rest of the world, high common external tariffs.

A 2002/03 initiative to improve policy implementation has largely stalled. In 2002, French experts prepared a report on customs procedures (the “Steenlandt” report), and an October 2003 workshop in Brazzaville resulted in a proposed roadmap for further reform. These efforts were supported by the EU, France, and the World Bank. However, since then the initiative has lost momentum, even though some of the actions identified in the roadmap have still moved forward. The June 2005 summit of the CEMAC Heads of State in Malabo confirmed the need for improved implementation of the CEMAC trade regime.

Several recent initiatives concerning external trade provide new challenges and opportunities for welfare-enhancing reforms. First, the 2003 workshop suggested reducing the number of CET rates from four to three, while lowering the top rate from 30 percent to 20 percent, in line with the WAEMU regime. The importance of external liberalization was reaffirmed by the IMF’s Managing Director in a March 2006 speech.2 CEMAC representatives have indicated that changes to the CET should be based on both a review of experience with the changes introduced in 1994 and a forward-looking assessment of proposed new rates. Second, an Economic Partnership Agreement (EPA) with the EU could result in far-reaching reciprocal trade liberalization between the two regions. And, finally, global trade liberalization, for example in the context of the Doha Round, would affect market access and world market prices, in particular for agricultural products.

This paper provides an update on the main elements of the reform agenda concerning the CEMAC trade regime as well as a quantitative assessment of selected potential effects on tariff revenues and trade patterns. Unfortunately, the latter analysis is hampered by severe data limitations: sufficiently detailed recent balance of payments data for most exercises in this paper were available for Cameroon, Gabon, and the Central African Republic only.3 The next section gives an overview of the main deficiencies of the current regime as well as the proposed remedies. Section III discusses the scope for boosting intra-community trade, and Section IV the implications of tariff reductions or an EPA. Finally, Section V concludes and provides policy recommendations.

II. The Current Regime: Characteristics, Deficiencies, and Reforms

A. Trade Characteristics

Trade restrictions and an uneven application of CEMAC rules constrain external and intraregional trade. Further, structural reforms and the region’s tariff policy have had very little positive impact on trade. While overall trade as a share of GDP increased slightly during 1994-2005, this is broadly reflective of developments in the oil sector, which is insulated from the domestic tariff regime.4 In line with the increased oil output, total goods exports to GDP increased from about 34 percent of GDP in 1994 to about 51 percent of GDP in 2005 (Figure 1), non-oil exports fell only slightly from 12 percent of GDP to 8 percent over the same period. It follows that CEMAC countries’ share of non-oil exports in total regional exports declined substantially during 1994-2005, from 35 percent to 16 percent. Figure 2 suggests that since about 2004, EU countries are no longer the primary destination for CEMAC’s exports. Exports to the US and to Asia have been increasing in the last two years (at 33 and 29 percent in 2005, respectively). On the import side, goods trade has remained fairly constant at about 18 percent of GDP in 1994 and 2005. The diverging trends of imports and exports are reflected in the trade balance, which improved from 5 percent of GDP to 21 percent of GDP over this period. While EU countries accounted for about 52 percent of total CEMAC imports in 2005, their share has followed a declining trend (Figure 3 and Table 1). At the same time, imports from Asian countries more than doubled in the last three years.

Figure 1.
Figure 1.

Oil and Non-Oil Exports to GDP

Citation: IMF Working Papers 2007, 137; 10.5089/9781451867015.001.A001

Source: IMF, World Economic Outlook and staff estimates.
Figure 2.
Figure 2.

Composition of CEMAC Exports to various partners

Citation: IMF Working Papers 2007, 137; 10.5089/9781451867015.001.A001

Source: IMF, Direction of Trade Statistics; and staff estimates.
Figure 3.
Figure 3.

Composition of CEMAC Imports from various partners

Citation: IMF Working Papers 2007, 137; 10.5089/9781451867015.001.A001

Source: IMF, Direction of Trade Statistics; and staff
Table 1.

CEMAC: Sources of Imports and Tariff Revenue by Region of Origin1

(in percent of total)

article image
Source: Data provided by the authorities.

Based on total imports by Cameroon (2004), Gabon (2004) and the CAR (2003).

Intra-community trade in the CEMAC is low, at less than 2 percent of GDP and about 1.5 percent of total trade. The share of intra-regional in total trade of the CEMAC countries has remained stagnant over the past five years, and even trended down somewhat since 1997—despite the 1994 initiative for creating an effective customs union (Figures 4 and 5). Internal trade is much lower than in other country groupings with free internal trade in Africa. Intra-regional trade in WAEMU amounts to more than 10 percent of total trade. For SADC and COMESA this share is between 5 and 10 percent of total trade. Official data may underestimate the magnitude of intra-regional trade due to large unrecorded trade flows—especially between neighboring countries. However, even with an adjustment for underreporting, the magnitude of intra-regional trade would still remain very low.5

Figure 4.
Figure 4.

Intraregional Exports Selected Regional Groups

Citation: IMF Working Papers 2007, 137; 10.5089/9781451867015.001.A001

Source: IMF, Direction of Trade Statistics; and staff estimates.
Figure 5.
Figure 5.

Intraregional Imports Selected Regional Groups

Citation: IMF Working Papers 2007, 137; 10.5089/9781451867015.001.A001

Source: IMF, Direction of Trade Statistics; and staff estimates.

B. Deficiencies and Reforms

While the unweighted average MFN tariff rate is close to 20 percent, most imports enter under the 10 percent rate, as shown in Table 2. CEMAC import tariffs are ad valorem and were revised in 2001 in order to ensure conformity with the 2002 version of the WTO’s Harmonized System. The CET comprises four rates—5 percent for basic necessities, 10 percent for raw materials and capital goods, 20 percent for intermediate and other goods, and 30 percent for general consumer goods.

Table 2.

CEMAC: Sources of Imports and Tariff Revenue by Import Tariff Band1

(in percent of total)

article image
Source: Data provided by the authorities.

Based on total imports by Cameroon (2004), Gabon (2004) and the CAR (2003).

While the 1994 reforms that created the CEMAC customs union were a major step forward, the trade regime remains plagued by poor implementation. The customs union suffers from a range of implementation problems—listed below—as well as red tape, weak and inefficient customs administration, involving the misclassification of imports, and corruption problems, including theft.

The Executive Secretariat, located in Bangui (CAR), lacks both the authority and the resources that would be needed for effective oversight of the arrangement. Data reporting to the secretariat by the member countries has been incomplete and, as a result, the secretariat has only a very partial picture of regional trade flows and of country-specific procedures and regulations.

National trade regulations are not always in compliance with community rules.6

  • Discrepancies between the CET and actual tariffs rates applied by member countries have been introduced to provide additional protection to national industries.7 The 2003 roadmap proposed a detailed review of both the nomenclature and the tariff rates in order to restore compliance with the CET. However, even in the absence of such a comprehensive exercise, the planned gradual move to the Automated System for Customs Data (ASYCUDA) by five member countries (excluding Equatorial Guinea) should still be beneficial to proper CET implementation.

  • Temporary tariff surcharges of up to 30 percent were introduced in 1994 for a period of six years or less, to help cushion the impact of abolishing quantitative restrictions. However, not all surcharges have been abolished as scheduled. Reportedly, the surcharges have been removed in the CAR, Cameroon, the Republic of Congo, and Gabon. However, Equatorial Guinea and Chad still maintain surcharges on selected products (see Box 1).

  • Several, but not all, CEMAC members have phased out the granting of firm-level tariff exemptions in the oil, mining, and tourism sectors, through bilateral Conventions d’Etablissement. In addition to renegotiating remaining existing exemptions, a lasting solution should involve aligning the national investment codes with the CEMAC Investment Charter, which strictly limits tariff exceptions.

  • A range on taxes and charges on top of the CET severely complicate the tariff regime (see Oliva, 2006). Some of these charges have been initiated at the community level: the community integration tax (1 percent) introduced in 2002 for financing the community institutions, and a statistical fee (up to 2 percent). Additional fees have been introduced at the national level, and their legal status under the common regime seems unclear. These include levies related to other international organizations, in particular ECCAS and OHADA.8 Box 2 illustrates these various duties for the Republic of Congo.

Problems in Implementing the Agreed Trade Regime in Chad

Experience in Chad, as revealed in the preparation of a Diagnostic Trade Integration Study, illustrates the myriad of implementation problems concerning the CEMAC customs union (Integrated Framework, 2005). (However, Chad should not be seen as a representative example.)

  • While the ASYCUDA customs database has been introduced in the main customs offices, it remains to be extended to those in the provinces. Introduction is hampered by severe planning problems, lack of training, and lack of network wiring. At present, the information provided in the statements issued by the customs offices varies—except concerning oil and cattle, which are handled by separate posts—with many smaller offices not reporting the tariff line and/or the country of origin.

  • A large range of exemptions from import tariffs are granted, including for imports under the Doba oil project, for the French military base, and for specific firms through Conventions d’Etablissement. In 2002-03 the actual import tax base was only 23 percent of total imports.

  • Imports of sugar, cigarettes, and drinks from non-member countries are still subject to a “temporary” 25 percent surcharge, which was scheduled to be removed by 2000. In addition, imports from all countries—including other CEMAC members—are subject to a range of charges, including a statistical fee (2 percent) and a storage fee (depending on number of days in customs storage).

  • Chad agreed in 2002 to apply transaction values as the basis for import valuation, but minimum values are still used for imports from non-CEMAC bordering countries.

  • There are no taxes on export products, except for cattle (FCFA 1,500 per head) and other livestock, and some plants. However, the 2 percent statistical fee also applies to exports. Exports are free from quantitative restrictions, apart from some restrictions for reproductive cows.

Goods produced in member countries are, in principle, eligible for a zero-rate preferential tariff, but qualification has been subject to problems. A new study commissioned by the CEMAC secretariat in 2006 seeks to address problems concerning the application of the 1993 rules of origin. The existing regime confers community origin on raw (mining and agricultural) products originating from the CEMAC, and also on goods manufactured in the CEMAC—with a local context of at least 40 percent or of at least 50 percent of the raw materials (in value terms) used in the production stem from the CEMAC. Certificates of origin are to be issued by the customs office closest to a producer. However, problems have arisen as not all customs offices have the required expertise, resulting in improper certification which, in turn, is frequently rejected by other member countries (recent examples concern powdered milk and wine). Also, a pre-certification procedure for frequent exporters is hardly used. In order to address these problems, in 2006 the CEMAC secretariat initiated a study by a consultant on simplifying the rules and improving compliance.9

Implementation of the Trade Regime in the Republic of Congo

The various problems in implementing the agreed CEMAC trade regime are illustrated by experience in the Republic of Congo, which has recently been discussed in Oliva (2006) and WTO (2006).

  • Imports are subject to a large number of additional duties and fees, raising the de-facto import tariff from 19 percent to about 22 percent, and hampering transparency. These additional charges are: an automation fee (2 percent), the Community Integration Tax (1 percent), a statistical tax (0.2 percent), an OHADA levy (0.05 percent), an (ECCA) Community Integration levy (0.04 percent), and inspection fees for qualifying transactions. There is also a 15 percent tax on imports of woods products. The customs also collect a 5 percent levy as an advance payment on income taxes or the flat-rate tax.

  • For certain goods, including cement, customs valuation is still based on reference prices.

  • Widespread structural and ad hoc exemptions greatly undermine customs duty collection. Exemptions include products, flour imports, and imports for public investments.

  • Nontariff barriers include price controls and domestic monopolies for sugar and wheat flour, that maintain high domestic prices, while imports are constrained by quotas import licensing procedures.

  • The export regime includes the automation fee (2 percent), a supplementary exit duty (2 percent), diamond royalties (2 percent), and various fixed and variable levies on wood.

Three of the six CEMAC member countries (Cameroon, Republic of Congo, and the CAR), in principle, apply transaction-based import valuation—as was the case already in 2003. This valuation basis has been mandatory since 2000, under the WTO agreement on customs valuation. In Chad, the move from using standardized reference prices to transaction-based valuation has been hampered by training problems. Gabon, which has finished the training and has updated its regulations, has pointed to revenue concerns as a reason for delaying the adjustment. Moreover, even in those countries that, in principle, apply transaction-based import valuation, implementation has been incomplete.

Problems concerning overly costly and cumbersome transit procedures and double taxation are, reportedly, being resolved. The existing transit regime includes a costly guarantee system. The system requires importers of goods to acquire a bond at the point of entry into the union, which aims to guarantee the—generally, coastal—transit country that the required duties are paid and that the goods are indeed re-exported to the—generally, landlocked—country of destination. However, the bonds can be very costly and reimbursement subject to long delays. In order to resolve these problems, a CAR customs official was recently stationed in Douala to collect import duties at the point of entry into the region—in line with a suggestion in the 2003 Roadmap. A similar procedure is under consideration for clearing imports through the Douala port destined for Chad. A more efficient transit regime would help reduce the incidence of double taxation of customs duties (as well as excises and VAT) on goods imported from a third country and then reexported within the CEMAC.

Trade is also affected by a relatively burdensome overall regulatory environment for doing business. The World Bank’s Doing Business 2006 database provides comparable measures of business regulations and their enforcement across 175 economies including the WAEMU and CEMAC member countries. Following Oliva (2006), we summarize some of these measures for CEMAC and compare with other groups/regions. Regarding trade-related impediments, the number of documents required for importing or exporting in CEMAC countries is comparable with that in WAEMU and in SSA more in general (Table 3).10 However, trade is also hampered by regulations that are not directly trade related but that work as barriers to trade by increasing the cost of doing business and hampering entrepreneurship. These obstacles are summarized in Table 4. Comparing WAEMU and CEMAC averages with SSA and other regions, structural impediments to developing a competitive private sector are relatively high for both the WAEMU and CEMAC countries. Out of the 175 countries, CEMAC and WAEMU’s overall rank is in the bottom tier (153 for WAEMU and 157 for CEMAC), worse than SSA and other regions.

Table 3.

Doing Business Survey (2006): Trading Across Borders

article image
Source: World Bank, Doing Business Database 2006, and IMF staff calculations.
Table 4.

Doing Business Survey (2006)

article image
Source: World Bank, Doing Business Database 2006, and IMF staff calculations.

Generally, quantitative restrictions do not appear to provide a main barrier to trade. Most qualitative restrictions were converted into tariffs in 1994. Imports of sugar are restricted in Gabon and the CAR. Exports of timber and logs are subject to prohibitions and quota restrictions in several member countries, for environmental reasons.

Most, but not all, export taxes have been phased out. In contrast to the joint tariffs on imports, export taxes are set at the national level. They are applied on exports to all countries, including other members. Reportedly, the taxes concerned are generally low and relate to only few products (especially logs and timber products).11

Finally, it is important to note the wide range of overlapping trade agreements that shape the trade regime.12 The CEMAC countries are also members of the Economic Community of Central African States (ECCAS), which aims to create a free trade area by the end of 2007 (WTO, 2006).13 The CEMAC countries benefit from several arrangements providing preferential access to developed countries. These include the Cotonou and Everything But Arms (EBA) agreements with the European Union, preferences under the African Growth and Opportunity Act (AGOA) for exports to the United States market, and under the Generalized System of Preference (GSP) for exports to several other countries.

III. Prospects for Enhancing Intra-community Trade

The lack of functional transportation corridors across the CEMAC countries is likely the main impediment to intra-regional trade—although several helpful initiatives are under way. Several of the weaknesses of the trade regime discussed in Section II hamper trade within the CEMAC—in particular, the problems concerning the rules of origin and double taxation. However, these concerns are probably overshadowed by the severe deficiencies of transportation infrastructure in the region. The resulting high costs and long delays in the transportation of goods are discussed in Box 3. In line with an action plan adopted at a June 2004 workshop in Douala, several projects for improving cross-country road infrastructure are currently being implemented—supported by the European Union, the African Development Fund, and the World Bank. These projects aim at connecting, in particular, Douala (Cameroon) with Bangui (CAR) and with N’djamena (Chad). A still unpaved section in the CAR of the Douala-Bangui corridor poses a serious bottleneck during the rainy season. However, part of the project has been delayed as a result of the CAR’s suspension from EU funding. The projects also include a trade facilitation component, with the construction of border posts and checkpoints (on both sides of the border) for administrative services (including customs, police, and forestry services). Transportation is also hampered by serious security problems, involving highway robbers, armed cross-border rebels, and internal rebel groups.

Transportation Delays and Costs

Administrative delays due to transit and customs procedures, roadblocks, and poor transportation services result in delays and excessive transportation costs and thereby hinder intra-regional trade. For example, a cargo journey from the port of Douala (Cameroon) to Bangui (CAR) and N’djamena (Chad) takes 2 and 5 weeks, respectively, on average. In principle, in the absence of any administrative or organizational hurdles, transportation from Douala to N’djamena could be completed in one week (Integrated Framework, 2005). In addition, the average clearing and transportation costs for a 20 foot container from Douala to Bangui or N’djamena are about €4,000—about four times the cost of maritime transport from Europe to Douala (World Bank, 2004). Port expenses account for about 25 percent of that amount (which is almost equivalent to the cost of maritime transport from Europe to Douala), and the remaining 75 percent is spend on road transportation.

The 2005 Diagnostic Trade Integration Study for Chad suggests that, although high, these costs and delays are comparable to those in West African countries. For example, average clearance time for imported containers in the ports of Douala is about 19 days, which was second only to Senegal (30 days), in a review of 8 countries along the west coast of Africa (SAATP 2004). However, costs and delays are about 30 percent lower for landlocked countries in East Africa; for example for Kigali (Rwanda) which is within the same distance from its parts as N’djamena. In addition, for Uganda and Tanzania, 72 percent and 85 percent of roads (respectively, are classified as in good or fair condition compared to 30 percent in Cameroon (World Bank SAATP indicators).

The World Bank’s Doing Business indicators also corroborate the high transportation costs, focusing on the costs and delays related to procedural requirements for importing and exporting (Table 3). On average, it costs about $1800 to export and almost US$2100 to import a 20 foot container load—which exceeds to costs in the WAEMU by about 60 percent and 20 percent, respectively, largely as a result of the high costs of trade in Gabon. Reflecting very lengthy procedures in Chad, the time required to export and import is, on average, about 20 percent more for CEMAC countries than in the WAEMU.

A structural determinant of the low level of intra-community trade may be the lack of complementarities in the production structures of the member countries. There is substantial similarity in the natural comparative advantages of the CEMAC countries, and, as a result, in their production and trade patterns (Table 5), which limits the scope for internal trade. In particular, oil is the main source of exports for all countries except the CAR. Other exports relate to wood products and cotton. But there is very little diversification, especially in manufacturing. Of the six members only Cameroon has a significant industrial base.

Table 5.

CEMAC: Sources of Imports and Tariff Revenue by Product Type1

(in percent of total)

article image
Source: Data provided by the authorities.

Based on total imports by Cameroon (2004), Gabon (2004) and the CAR (2003).

The existence of only limited product complementarities is confirmed by a more formal analysis. Trade complementarity between a pair of countries is traditionally measured by the Trade Complementarity Index (TCI). The index is a measure of similarities between the export basket of one country and the import basket of another; therefore, it can provide useful information on prospects for both intraregional and external trade. We define the bilateral product complementarity index between country i exports and country j imports based on Michaely (1996), as follows:


where Xik is country i’s total exports of product k, and Mjk is country j’s total imports of product k. The higher the index between two countries, the greater the product complementarity. The index is zero when no product exported by one country is imported by the other; the index is 100 when the export-import shares match exactly.

We calculate and compare bilateral product complementarity indices for CEMAC and WAEMU member countries using UN-COMTRADE data at the two-digit classification level (Tables 6 and 7). Following Tsikata (1999) and Khandelwal (2004), we consider TCI’s above 25 as indicative of strong complementarity. Our estimates suggest a much higher degree of average complementarity within the WAEMU compared to CEMAC (30.4 compared to 16.9). In addition, the highest average complementarity in the CEMAC (CAR imports to rest of CEMAC exports) is lower than the smallest average complementarity in the WAEMU (Senegal imports to rest of WAEMU exports). These findings help explain the higher volume of intra-regional trade within WAEMU compared to CEMAC.

Table 6.

Trade Complementarity Index in the CEMAC (two-digit)

article image
Source: United Nations COMTRADE database.
Table 7.

Trade Complementarity Index in the WAEMU (two-digit)

article image
Source: United Nations COMTRADE database.

Within the CEMAC, the more developed and diversified economies of Cameroon and Gabon appear to be in a much better position to market their exports in the region than the smaller and less developed CAR. Exports from Cameroon to CAR exhibit the highest degree of complementarity (28.6), and exports from CAR to Gabon the lowest (7.8).14

A similar pattern emerges for the WAEMU: with an average TCI of 51.3, Senegal and to a lesser extent Cote d’Ivoire—i.e., the largest and most developed member countries—are in the best position to market their exports to the region.15 There is also evidence of asymmetric complementarities for the smaller economies, in particular, Benin and Niger: product complementarities between Niger and Benin exports with the region’s imports are the lowest while at the same time these countries’ imports have the highest complementarity with the region’s exports.

Nonetheless, a closer look at the product composition of internal and external trade suggests significant potential for an expansion of trade within the union. The above exercise gives an impression of the (limited) scope for diverting existing trade flows towards the regional partners. However, this assumes unchanged production patterns. The key question may be whether intra-regional trade could be boosted, also, by stimulating the production in the CEMAC of those goods—especially in the manufacturing sector—for which an effective market in other member countries could be opened up as a result of a substantial reduction in transport-related and administrative barriers within the region. To help answer this question, we calculated for three CEMAC members, for what part of imports (at the 6 digit level) from non-CEMAC countries, any other CEMAC country already had significant exports to any country—indicating that there is, at least, a potential for production within the union that could satisfy the demand for imports.16 This exercise indicates that for 21 percent of total CEMAC imports, there may be a potential for the development of internal sources.

IV. Recent Trade Initiatives

A. Reducing the CET

With tariff rates up to 30 percent, and an unweighted average of about 19 percent, the CEMAC CET rates are high in comparison with other countries and country groups, including in Africa (Table 8). In particular, with an average rate of about 13 percent, WAEMU tariffs are substantially lower. As shown in Table 9, average CEMAC tariffs exceed WAEMU tariffs for all sectors. The highest rates apply to footwear, wood products, and agriculture—with average rates of 23 percent or more—which are also produced domestically. This pattern is suggestive of a protectionist bias. Evidence of tariff escalation is mixed, although for most industries, tariffs are higher on finished than on semi-finished products (WTO, 2006). Finally, with its large variation in tariff rates, CEMAC tariffs are also more dispersed than those of the WAEMU—which complicates customs administration and entails price distortions across imported goods.

Table 8.

Simple Average MFN Tariffs, 2005

(in percent)

article image
Source: IMF Staff calculations.
Table 9.

CEMAC and WAEMU MFN import tariff rates by sector

article image
Source: Oliva [2006]

Revenues from external tariffs have provided a significant source of fiscal revenues—although declining in relative terms (Table 10). From 2000 to 2005, import duty receipts have remained stagnant in nominal terms, declining in proportion to total fiscal revenues and GDP. Tariff revenue amounted to 1.5 percent of GDP in 2004, and 1.3 percent in 2005, while declining from 9 percent of total revenues to 7 percent, as higher oil prices have boosted GDP and fiscal receipts. Actual tariff revenues have remained well below the level that would result from combining the schedule-based rates to imports, reflecting the eroding effect on exemptions on the import tax base.17

Table 10.

CEMAC: Tariff Receipts in percent of GDP1

article image
Source: WEO and staff estimates, and data provided by the authorities.

Based on data for 2004, except for the CAR (2003 data).

Nonetheless, there are strong arguments for tariff reduction. Extensive empirical evidence supports the positive impact of trade liberalization on economic growth and, thereby, on poverty alleviation (Berg and Krueger, 2003). For the CEMAC countries, trade liberalization could, in particular, help boost the development of a more diversified non-oil export base, which is of vital importance for long-term growth and external sustainability, as oil-resources are expected to be largely depleted over the coming decade and a half. For the CEMAC, trade liberalization could result both from lower rates and from reducing the high degree of tariff dispersion (which would result in more uniform price incentives across products and curb tariff escalation). Accordingly, a move from the current tariff schedule to WAEMU rates—which are capped at 20 percent rather than 30 percent—could serve as a useful first step.

Trade liberalization should be accompanied by measures to compensate for its impact on fiscal revenues, as needed. Table 11 suggests that the revenue impact of adopting WAEMU tariff rates could be as large as 40 percent—reducing the calculated receipts (disregarding exemptions) from 2.2 percent of GDP to 1.3 percent. This simulation, however, does not incorporate any positive volume response of imports to the lower domestic prices of imported goods, and therefore overestimates the actual revenue effect.18 Reducing the many tariff exemptions that have narrowed the current tariff base provides a first option for offsetting such losses. Further revenue measures should aim to strengthen income and indirect tax policies and administration. In principle, shifting from trade taxes to taxes on domestic consumption would help offset the revenue loss, while preserving the efficiency gains from the tariff cuts, as domestic producers would now face world market prices.

Table 11.

CEMAC: Scenarios for Tariff Receipts in percent of GDP1

article image
Source: WEO and staff estimates, and data provided by the authorities.

Based on data for 2004, except for the CAR (2003 data).

B. Economic Partnership Agreement

Discussions with the EU on an EPA started in 2003 and an agreement could be enacted as of January 1, 2008. São Tomé and Principe and the DRC have joined the CEMAC in the preparations of an EPA (which would not involve their membership of the CEMAC as such). The first phase of the negotiations was completed in December 2005.19 The second phase involves drafting the specific agreements. Specific issues at this stage concern product exemptions, rules of origin, and the pace of liberalization. On rules of origin, discussions have focused on a 50 percent local content requirement, in the context of wider talks of all ACP countries with the EU.20

An EPA could be instrumental in addressing the serious institutional weaknesses of the CEMAC trade regime. The proper functioning of an EPA would require the uniform application of trade rules by all CEMAC countries, in particular concerning rules of origin. More generally, an EPA could provide both the area-wide political momentum and the technical assistance needed for empowering the regional authorities, harmonizing the rules and procedures, and strengthening their implementation, and reforming the investment climate.

The effect of an EPA on the manufacturing sector in the CEMAC will likely hinge on liberal rules of origin. Available evidence refutes earlier notions that restrictive rules of origin could support industrialization by promoting the development of local supply chains that would provide inputs for domestic industries serving the European markets (so-called backward integration). Instead, a simple and low value added criterion is more likely to be beneficial, by attracting foreign investors into new export industries that would use third-country inputs—befitting from inexpensive local labor and tariff preferences. The potential for success of such approach is evidenced by the United States’ AGOA initiative. Under AGOA, eligible African countries are granted tariff free access to the United States market, with relatively liberal rules of origin.21 Several African countries, including Lesotho and Swaziland, have seen large increases in manufacturing production triggered by AGOA aimed at the American market, often using cheap inputs from outside the region.

In the context of an EPA, Hinkle and Schiff (2004) argue that SSA’s gains from liberalization of trade in the service sector are likely to come from the import side, due to the less developed nature of the export service sector in SSA and the constraints on expanding the employment of temporary workers in the EU.22 This argument also extends to the case of CEMAC where exports are dominated by the oil sector. Further, Hinkle and Schiff (2004) suggest that imports of services should be liberalized on both an MFN and an intra-SSA regional bloc basis at the same time as they are liberalized vis-à-vis the EU in order to attract investment by the most efficient service providers and benefit from economies of scale. At the same time, the timing of the liberalization in various service sectors depends on the capacity of the SSA countries to implement the required accompanying regulatory reforms in these sectors.

An overall evaluation of the costs and benefits of an EPA should also incorporate the likely incidence of trade diversion and revenue losses.

  • An EPA would entail serious risks of large-scale trade diversion. The relatively high CET would entail a clear incentive in case of selective trade liberalization for, not only, trade creation, but also trade diversion. A second indicator for the risks of trade diversion is provided in Table 12, which shows the degree to which goods (at the six-digit classification level) currently imported from non-European sources are also, already, imported from the EU. It appears that this is the case for almost all such imports—while imports from non-European sources accounted for 47 percent of total imports, those concerning goods that were not also imported from the EU amounted to only 1 percent of total imports.23 This finding is suggestive of a large scope for CEMAC countries to turn to the EU for products that are now, at least in part, sourced from other, cheaper suppliers.

  • The eventual revenue impact of an EPA would be substantial, as more than 50 percent of CEMAC imports originate in the EU. An EPA would involve an implementation period that may well extend to 2020, at the end of which EU goods could enter the CEMAC free of import tariffs or quota—albeit with exemptions for a limited range of products. As a benchmark, the EPA column in Table 11 provides a tentative approximation of the possible revenue impact in case of full import liberalization for imports from the EU. The lower revenue levels in comparison with those in the first column indicate that revenue could decline by more than half, which reflects the current tariff revenue relating to imports from the EU. The actual eventual tariff loss may be overestimated in this simple exercise by ignoring product exemptions, or underestimated by ignoring the further tariff loss that would result with trade diversion. A noteworthy result is also that the tariff loss would be largest by far for Gabon, reflecting the high EU share in its imports, which amounts to about 65 percent of the total. This is well above the average for the CEMAC as a whole, at 53 percent.

Table 12.

CEMAC: Indications of the Scope for Trade Diversion1

(in percent of total imports)

article image
Source: data provided by the authorities and IMF staff calculations.

Based on total imports by Cameroon (2004), Gabon (2004) and the CAR (2003) and exports by Cameroon (2004), Gabon (2004) and the CAR (2003), Chad (1995), Equatorial Guinea (1995), and Republic of Congo (1995).

The risks of trade diversion calls for parallel tariff cuts on an MFN basis—to be accompanied by a timely further strengthening of domestic revenue mobilization. The revenue loss would, however, be compounded in case an EPA were combined with tariff reductions on an MFN basis, thereby limiting trade diversion. A scenario for the tariff loss in case with parallel tariff cuts to the WAEMU tariff rates is shown in the final column of Table 9, which suggests that revenues from import tariffs could drop by about three-quarters.

C. Global Trade Liberalization

The significance of global trade liberalization for the CEMAC member countries can be illustrated by the effects of a successful conclusion of the Doha Round. Several studies found that that under a possible Doha round scenario, increased market access for—agricultural and nonagricultural—merchandise exports to industrial countries alone would not bring substantial benefits to many countries in Africa.24 This is, in part, because existing preferences for Africa (such as those under the Everything But Arms Initiative in the European Union (EU)) would be eroded (see Box 4). This argument, however, has little immediate relevance for the CEMAC, as its members have, so far, largely failed to take advantage of the existing preferences for developing a significant non-oil export sector. In addition, rising world agricultural prices would lead to worsening terms of trade for Africa’s net food importers—including the CEMAC. However, over time, higher world prices may turn some net food importers in Africa into net exporters.

The Doha Round and Africa’s Preference Erosion1

African countries have emphasized the potential losses from preference erosion. The estimated impact of preference erosion is, however, often overstated, because of the failure to take account of the underutilization of preferences and export increases in sectors not affected by preference erosion. Moreover, preference erosion usually represents a permanent shock occurring over a long period of time which can be anticipated.

Yang (2005) notes that what is overlooked is that, over the past two and a half decades, African countries have become increasingly dependent on other developing countries for trade, despite ever-expanding trade preferences they have received in industrial countries. Further, if recent trends continue, it is likely that by the time any Doha round liberalization is implemented (say, 2015), developing countries may account for an even larger share of Africa’s exports. Liberalization in developing countries will lead to little preference erosion because no major preferences for Africa are in place in these countries. Moreover, trade barriers against African exports (especially those against manufactures) remain significantly higher in developing countries than in industrial countries, and, hence, the potential market access gains are larger.

1 Based on Box 2.4 Regional Economic Outlook: Sub-Saharan Africa (September 2005), and Yang (2005).

A beneficial result from the Doha round for the CEMAC countries would crucially depend on the extent to which these countries themselves would also liberalize their own imports.25 Anderson, Martin, and Mensbrugghe (2005) find that Sub-Saharan Africa would gain provided that the countries engage fully in the Doha reform process. This result mirrors the earlier discussion of the gains from trade liberalization, with relatively high initial tariff levels. This would imply reconsidering the opportunity for developing countries to liberalize less than middle and high income countries. These results are largely confirmed by a simulation of the poverty impact of the Doha Round for Cameroon (Emeni, Cockburg, and Decaluwe, 2005), which finds a small adverse welfare impact of a Doha Round scenario. However, positive results emerge in case of full liberalization by Cameroon, although the size of these gains crucially depends on a possible reduction in Cameroon’s export prices, with increased export volumes of cash crops, wood processing, and forestry.

V. Concluding Remarks and Policy Recommendations

The manifold weaknesses in the implementation by the member countries of the agreed customs union regime highlight a need for a renewed political commitment to regional integration. While recent political declarations on strengthening integration policies are welcome, policy measures at the national level have progressed only slowly. The incipient momentum in 2003 for coordinated action appears to have lost steam.

Progress also requires effective central oversight. The CEMAC secretariat, which is responsible for overseeing compliance with the agreed trade regime, lacks the means and the authority for carrying out this role in an effective manner. In this context, recent proposals for converting the secretariat into a commission, with adequate means and authority, should be an important step. A first requirement for an effective central organization concerns data collection—both on trade flows and on country-specific regulations and practices.

Key measures for improving compliance with the requirements for a customs union should be adopted without delay, at both the regional and—most importantly—the national level. The deficiencies of existing regulations and of implementation by member countries have already been identified in the context of the 2003 roadmap for reform. Key actions include limiting tariff exemptions, phasing out remaining surcharges, strengthening (and, maybe, simplifying) the determination of products’ origin, and enhancing customs administration more generally, including through ASYCUDA. A further key requirement is the need to improve transportation infrastructure and organization.

The ongoing preparations for an EPA should be buttressed by comprehensive analyses of its likely effects. An EPA would reduce fiscal revenues from international trade—possibly by about half—and even more in the case of Gabon. It would also entail serious risks for trade diversion, given the wide range of goods imported from the EU at present. Preparations for an EPA should also include poverty and social impact analyses, taking into account the differential impact on import and export prices of goods produced and consumed by urban and rural households. The main benefits from an EPA may well be those that stem from the opportunities for improving the consistent application of trade regulations and possible EU assistance in upgrading and marketing CEMAC exports.

There is a strong case for tariff reduction, with or without an EPA. CEMAC tariffs are relatively high, especially the top rate of 30 percent rate for consumer goods. Trade liberalization would help boost economic growth and poverty alleviation and limits risks of trade diversion with an EPA. The returns to trade liberalization will be larger if they take place in tandem with other structural reforms (e.g., labor market reform) to enhance market functioning.

Finally, tariff reform should be complemented by improvements in domestic revenue mobilization. Both the introduction of an EPA and tariff reductions on an MFN basis would likely result in a significant loss of fiscal revenues that could amount to 1 percent of GDP or even more. Moreover, the gradual loss of fiscal revenues during the implementation period of an EPA would broadly coincide with the loss of oil-based revenues, as most oil resources are expected to be depleted over the coming decade, or so. The introduction of VAT systems in the 1990s has provided a helpful foundation for shifting from trade-based to domestic taxation.


  • Anderson, K., W. Martin, and D. van der Mensbrugghe, 2005, “Would Multilateral Trade Reform Benefit Sub-Saharan Africans,” World Bank Policy Research Working Paper 3616 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Berg, A., and A. Krueger, 2003, “Trade, Growth, and Poverty: A Selective Survey,” IMF Working Paper 03/30 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Bouet, A., J.C. Bureau, Y. Decreux, and S. Jean, 2005, “Multilateral Agricultural Trade Liberalization: The Contrasting Fortunes of Developing Countries in the Doha Round,” The World Economy, Vol. 28 (No. 9), pp. 1329-1354.

    • Search Google Scholar
    • Export Citation
  • Dufy, C., J. Puons, and M. Steenlandt, 2002, “Vers L’Union Douanière en Afrique Centrale,” November (“Steenlandt Report”).

  • Emini, C. A., J. Cockburn, and B. Decaluwe, 2005, “The Poverty Impacts of the Doha Round in Cameroon: The Role of Tax Policy,” World Bank Policy Research Working Paper 3746 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Hinkle, L., and M. Schiff, 2004, “Economic Partnership Agreements Between Sub-Saharan Africa and the EU: A Development Perspective,” The World Economy, 27, pp. 1321-33.

    • Search Google Scholar
    • Export Citation
  • Integrated Framework, 2005, “Diagnostic Trade Integration Study—Chad” (Draft)

  • Khandelwal, P., 2004, “COMESA and SADC: Prospects and Challenges for Regional Trade Integration, IMF Working Paper 04/227 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Michaely, M., 1996, “Trade Preferential Agreements in Latin America: an ex-ante Assessment,” World Bank Policy Research Working Paper #1583.

    • Search Google Scholar
    • Export Citation
  • Oliva, Maria O., 2007, “Trade Policy and Reform in Congo,” IMF Working Paper (Washington: International Monetary Fund, forthcoming).

    • Search Google Scholar
    • Export Citation
  • Tsikata, Y., 1999, “Southern Africa: Trade, Liberalization and Implications for a Free Trade Area,” TIPS Annual Forum.

  • World Bank, 2004, “CEMAC—Transport-Trade Facilitation,” Project Information Document, (August 2004).

  • World Bank, 2004, “Sub-Saharan Africa Transport Policy Program (SSATP),”SSATP Transport Indicators Initiative.

  • World Bank, 2004, “Rapport sur les indicateurs de performance du secteur des transports au Cameroun”.

  • World Trade Organization, 2001, “Trade Policy Review Cameroon,” Report by the CEMAC Secretariat.

  • World Trade Organization, 2001, “Trade Policy Review Gabon,” Report by the CEMAC Secretariat.

  • World Trade Organization, 2006, “Trade Policy Review Republic of Congo,” Report by the Secretariat.

  • Yang, Y., 2005, “Africa in the Doha Round: Dealing with Preference Erosion and Beyond,” IMF Policy Discussion Paper 05/08 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Yang, Y., and S. Gupta, 2005, “Regional Trade Arrangements in Africa: Past Performance and the Way Forward,” IMF Working Paper 05/36 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

This paper is based on a June 2006 Selected Issues Paper (Country Report no. 06/309) prepared by Jan Kees Martijn and Charalambos Tsangarides in the context of the annual Article IV discussions in relation to the CEMAC. Research assistance was provided by Dustin Smith and Gustavo Ramirez. Helpful comments were provided by Stephane Cosse, Catherine Pattillo, Yongzheng Yang, and Hans Weisfeld.


“Making a Blessing of Oil: Sources of Growth in the CEMAC Region,” Bata, March 15, 2006 (


Chad, Equatorial Guinea, and the Republic of Congo reported their latest external trade statistics for 1995. More recent data are available in the UN-COMTRADE database, derived from mirror flows (i.e., flows declared by partner countries), but these are partial data, which do not add up to total imports as estimated by the monetary authorities.


Increased trade and exports during this period can be attributed to both petroleum prices, which experienced a cumulative increase from $16 per barrel in 1994 to $70 per barrel in 2005, and production of oil, which went up by 85 percent.


Anecdotal evidence suggests that the amount of unrecorded informal trade within the region could be about 50 to 60 percent of recorded trade.


The CEMAC secretariat—which has provided much of the information underlying this overview—does not have up-to-date information on the incidence of the various concerns listed below across countries. As a result, this assessment should be seen as tentative and subject to further confirmation.


For example, the CAR provides temporary preferential tariff treatment on imports of heavy machinery and some vehicles for investment purposes of 8 percent.


The CEMAC member states are also members of the Economic Community of Central African States (ECCAS) and of the Organization for Business Harmonization in Africa (OHADA).


Including consideration of a suggestion in the Steenlandt report for moving to a criterion of “sufficient transformation” as evidenced by a change of product classification based on the SH nomenclature. In addition, proposals should be compatible with the requirements of an EPA (see below).


See Section III and Box 3 for a discussion of transportation costs.


The CAR maintains export taxes of 2.25 percent, 4.25 percent, 4.05 percent, and 10.5 percent on gold, diamond, processed wood, and timber respectively. Equatorial Guinea currently has export taxes that range between 1 percent for (coffee and cocoa) to 15.8 percent (logs), with intermediate rates for other goods (re-exports, plywood and sawn wood). Cameroon and the Republic of Congo maintain export taxes on timber and logs only. Chad imposes export taxes on cattle and some plants.


See Yang and Gupta (2005), for an analysis of the key features of the many trade arrangements in Africa, and the related concerns of weak implementation, lack of transparency, and protectionism.


The ECCAS also includes Angola, Burundi, Rwanda, Sao Tome and Principe, and the DRC.


We also calculated of TCI’s at the one-digit level as it allows us to increase the sample of CEMAC countries (results available upon request). While TCI’s at the one-digit level, are higher by construction, the conclusions still remain unchanged: complementarities within CEMAC are low, especially for exports by the CAR and Equatorial Guinea, and exports from Cameroon to Equatorial Guinea exhibiting the highest degree of complementarity.


However, product complementarity between Senegal and Cote d’Ivoire is high.


While the (2003 or 2004) import data relate to Cameroon, Gabon and the CAR, the export data include all six member countries, using older (1995) data for Equatorial Guinea, the Republic of Congo and Chad. After all, the existence of exports a decade ago still provides evidence of a potential for production. Exports were included only if they exceeded US$ 1 million.


The higher level of actual tariff receipts compared with the computed level for Gabon likely reflects misclassification in the fiscal data.


However, a positive volume response is unlikely to be large enough to maintain the original revenue level, given that CEMAC tariff rates are already below levels considered revenue maximizing (IMF, 2005).


The Schedule and principles for these discussions are presented in a joint 2004 document (Feuille de route des négociations des Accords de Partenariat Economique Entre l’Afrique Centrale et l’Union Européenne).


As discussed above, the current local content requirement within the CEMAC is 40 percent. CEMAC representatives noted that this internal requirement would need to be harmonized with the one under an EPA.


The regular requirement is that domestic content must equal at least 35 percent of a product’s value. Moreover, further exemptions apply (until September 2007) for apparel made in low-income countries using oustide fabric.


Possible priority sectors for liberalization are transportation, telecommunications and finance.


The part of imports from non-EU countries concerning goods not also imported from EU sources is smaller for the CEMAC as a whole than for individual countries because the calculation for the CEMAC as a whole considers EU exports to any CEMAC country.


The study finds that for Sub-Saharan Africa (excluding Southern Africa), the net effect could be a net decline in real income by 0.1 percent. However, the net result would switch to a 0.3 percentage point improvement in case developing countries undertook the same reductions in bound (but not necessarily applied) tariffs as the high income countries.

Trade Reform in the CEMAC: Developments and Opportunities
Author: Mr. Charalambos G Tsangarides and Mr. Jan Kees Martijn