Regional economic communities (RECs) come in various shapes and sizes. From the highly integrated European Union (EU) to the long-lived Southern African Customs Union (SACU), to the pre-FTA Community of Sahel-Saharan States (CEN-SAD); each grouping has a role to play in strengthening economic and political ties between member states. RECs are first and foremost about trade- the promotion of intra-REC trade and the removal of trade barriers between states, but as economic integration progresses, deeper levels of integration develop and cooperation can extend into other areas. Investment, trade in services, policy harmonisation and financial integration can all grow within the economic space of a REC but these deeper forms of integration rarely happen before trade integration has progressed.

For this reason, the primary question relating to the success of a REC is if it is able to measurably increase trade between member states. As Africa embarks on its journey to a continent-wide free trade area (the AfCFTA), reference has to be made to the experience of the African RECs and their track record of promoting intra-REC trade. There are thirteen African RECs as at the date of writing (excluding the supra-REC Tripartite FTA) and each of them are at varying stages of progression in terms of integration.

The EAC makes a good African case study in integration because it has progressed further than any of the other eight AU-recognised RECs. However, the admission as members of Rwanda and Burundi in 2009 also presents a case for analysis since trade and tariff data is available and straddles 2009 for at least one of the EAC member states – Uganda. The method of analysis that is appropriate is a treatment/control group where one attempts to answer the question “did entry into the REC boost trade for of the entering country?” Only if the new REC member experiences an increase in trade with its REC partners, relative to its non-REC trading partners, can we infer that REC membership is trade-boosting.

This type of analysis has been used before to measure the impact of REC admission on trade[1], but with reference to Africa’s free trade agreements with the European Union and the United States. A 2010 study by Frazer and Van Biesebroeck found that admission to the AGOA programme had trade-boosting effects for apparel exports as well as agricultural and manufactured products. They also found that the preference margin – i.e. the difference between the preferential tariff and the most favoured nation tariff – for products had a positive impact on trade. In other words, the greater the relative advantage of having access to AGOA preferences, for any particular product, the greater the boost to trade.

In a similar study that examined the impact of European GSP and EBA preferences granted to developing countries, Thelle et al found that the LDCs had enjoyed bigger gains than more developed economies and that the EBA preference scheme had brought greater gains than the GSP scheme.

These two cited studies were attempting to determine the effectiveness of preference schemes on the designated developing country groups’ trade with the preference granting nation/s. These are unilateral preference programmes as opposed to the communities among equals that form the African RECs. However, the method can be as well applied to intra-African trade in an attempt to determine if membership of multilateral preference areas is trade-enhancing.

In an exploratory econometrics exercise, this author used a treatment/control group method to determine the impact of the entry of Rwanda and Burundi into the EAC on Uganda’s imports from these and other countries. The data was accessed from the World Bank’s world integrated trade solution (WITS), which has tariff and trade data for multiple years for several African countries. The exercise is made interesting by the evolution of trade preference arrangements over the period of analysis (2000-2016). Specifically:

• Uganda was a part of the EAC free trade area (FTA) for the entire period of analysis.

• In 2008, COMESA began the journey to a free trade area, that is, they began the process of tariff phase-down. All the EAC member states belong to COMESA, and thus participating in this process. The data shows that during the period of analysis Uganda did give preferential access to members of COMESA.

• Rwanda and Burundi joined the COMESA grouping in 2004 but only joined the EAC in 2009.

This means that Uganda participated in preferential and non-preferential trade with many African trade partners over the period 2000-2016. In order to attempt to separate out the effects of preferential trade area membership on Uganda’s imports from other African countries, two drivers of increased trade are modelled:

• The year 2010 is used as a turning point in the data, by creating a ‘time dummy’ that captures the effect on Uganda’s imports of its new EAC trading partners, Rwanda and Burundi

• The preference margin for each imported product classification (HS four and two digit) is used to capture the effect, if any, of a preference margin in Uganda’s imports.

In addition, non-FTA related trade shifts are controlled for and by including African (REC fellow members and non-REC fellow members) and non-African trade partners in the data. The technique used is a panel estimation, allowing for heterogeneity (differences) among trade partners and product groups through using a cross-sectional fixed effects specification.

The quantitative results will not be presented here, but rather a verbal explanation of the results will be given. The regressions were able to explain about 80% of the variation in Uganda’s imports (weighted). The results showed that:

• The preference margin had a significant impact on Uganda’s imports over the period. In other words, EAC membership and COMESA’s free trade arrangement from 2008 was important for trade. Among the trading partners that enjoyed preferential access to Uganda’s markets and showed increased trade, in this case via COMESA, were Egypt, Eswatini and Mauritius.

• The interaction of the treatment group (Rwanda and Burundi) and the period after 2009 was significant. That is, although Uganda’s imports increased quite rapidly over the period (rising trend), there was an increase in imports from Rwanda and Burundi over and above the preference margin effect or the trend effect.

• The preference margin effect for Rwanda and Burundi was not very significant[2], meaning that Uganda’s other preferential trading partners were more likely to be motivated by the preference margin they enjoyed.

The results show that the mere entry in a REC can have trade-boosting effects, besides the effects of preference margins. This most likely results from the other trade-enhancing drivers of REC admission, for example reduced border friction, customs cooperation, harmonisation of standards, expansion of cross border investment and enhanced trade in services. REC membership is a type of ‘glue’ that creates an impetus for deepening integration over time.

This is good news for the AfCFTA, which faces a daunting task due to the timeline for integration, the number of REC members as well as the sheer geographic size and diversity of the member states of the AU. The lesson seems to be that it is enough to initiate the free trade of goods, with preferences granted but not necessarily at 100% ambition. Based on our quantitative analysis (and to the extent they are generalizable), African RECs are indeed trade-enhancing and both the preference margin and the ‘glue’ effect are able to bring about increasing levels of intra-African trade.

---

[1] See for the case of the EU and the developing countries:
Thelle, M.H., Jeppesen, T., Gjodesen-Lund, C. and Van Biesebroeck, J., 2015. Assessment of economic benefits generated by the EU Trade Regimes towards developing countries. European Commission, DG-DEVCO.

Also, for the case of the AGOA countries and the US:
Frazer, G. and Van Biesebroeck, J., 2010. Trade growth under the African growth and opportunity act. The Review of Economics and Statistics, 92(1), pp. 128-144.

[2] Note however that this will probably change over time.