What if Africa were to liberalise the intra-Africa automotive components trade? A forthcoming tralac working paper (by this author) presents the results of a simulation of the complete liberalisation of the intra-African automotive components trade[1]. The analysis is an attempt to simulate the outcomes of the establishment of an African free trade area (FTA) for this specific sector only. This would be the case once the African Continental Free Trade Area (AfCFTA) has been fully implemented. As such the paper looks forward into the near future to assess what the potential gains and losses would be for this sector from the establishment of the AfCFTA.

One striking result of this exercise is the conclusion that if continental liberalisation were to take place in the near future, almost all of the gains in this sector would accrue to one country – South Africa. However a continental industrial policy for this sector could ensure that over the medium to longer term, the gains would be more evenly spread.

Tariff-based market access

To understand the full gamut of potential gains and losses resulting from liberalisation, it is first necessary to overview the market access profile of the automotive components trade in Africa. Automotive component imports are subject to tariffs in Africa, except where preferential arrangements exist in customs unions or free trade areas. The table below presents tariff and trade data for Africa’s top 26 automotive components importers, ranked from the top by magnitude of the effectively applied (weighted) tariff to this product group (HS 8708). As can be seen, the aggregate level of tariffs is quite high – with twelve countries (almost half of the group) having applied tariffs equal to or above 10%. The balance of the countries have tariffs below 10%, which are sometimes referred to as ‘nuisance tariffs’.

Source: Author’s calculations on data sourced from World Integrated Trade Solution (WITS)

When further disaggregating by Africa/ROW split, the data shows that in general, African nations’ weighted average tariff for imports from all other African nations is less than the weighted average for imports from the ROW[2]. This is because for most African nations, preferential tariffs in favour of REC partners reduce the weighted average for imports from all African nations. Another observable pattern is that most African nations import far more from the ROW than from Africa. This generally applies throughout, while there are exceptions in that the SACU members excluding Botswana import nearly as much or as much from South Africa as they do from the ROW.

Another notable pattern is that the nations of North Africa – Tunisia, Morocco and Egypt tariff African imports at the same or greater rates than imports from the ROW. This is most striking in the case of Egypt, which applies a preferential tariff to about 15% of its imports from the ROW. The trade flows data also indicates that the North African automotive exporters are focused on Europe rather than Africa, whereas South Africa is far more integrated into Africa, especially Southern Africa.

These patterns of protection will have a bearing on the results of liberalisation – on which countries benefit and which bear the cost. Countries that tariff African imports more than those from the ROW (such as Egypt) may appear to have more to lose than others, but what is also pertinent is the amount of existing imports from Africa. The North African nations may tariff African automotive imports at similar or higher rates than imports from the ROW but they also do not import any great volume from Africa. This means they will not feel any significant negative impacts from the liberalisation.

Simulating liberalisation

The simulation exercise produces a large amount of data, only a small portion of this data is reproduced here – in the figures below.

What is striking about these results is the extent to which the benefits are captured by South Africa. As pointed out at the outset, South Africa dominates the auto components trade in Africa, being by far the largest supplier to the continent. It is not surprising therefore that when intra-African liberalisation takes place; South Africa is best positioned to capture most of the gains. Small gains in trade are enjoyed by Tunisia, Egypt, Namibia and Morocco, but these are proportional to these nations’ original trade volumes with the rest of Africa.

About half of South Africa’s trade gains are the result of displacing the imports from the ROW (trade diversion). The ROW nations that lose out most are Belgium, China, the United Arab Emirates and the United States, but there is a group of 22 nations in total that experience some losses. Where gains are concentrated and losses are distributed (shared), policy action like this is less likely to cause retaliation but this can never be guaranteed.

The trade effects were attributable to the trade partners of Africa – both other African nations as well as ROW nations. However the revenue effects are born by the African importers of automotive components. These effects are visualised in the right hand chart in the above figure. There are also welfare gains as a result of the liberalisation – the gains experienced by the consumers in the importing nation, as a result of being able to acquire the product at a reduced price.

There is an inverse relationship between the revenue effects and the welfare effects – the countries that forgo the greatest revenue previously had a higher volume of imports at a high tariff. When these tariffs are cut to zero, the customs revenue is lost but the consumers of the product benefit from lower prices and experience welfare gains, or ‘consumer surplus’ gains: the benefits of the price drop in local currency. In this case, Mozambique and the DRC are the two countries that forgo the most revenue/enjoy the largest welfare gain. Since these countries do not have domestic automotive components industries, their tariffs were in place for revenue collection and so this revenue will have to be made up somewhere else. However, the advantage of liberalising an intermediate good is that the costs of final production of downstream goods are reduced. These countries (and the other liberalising importers) could benefit from being able to obtain cheaper intermediate goods and produce final goods more competitively.

Rules of origin

What cannot be simulated using the SMART methodology employed are the various non-tariff barriers, including the variation in rules of origin (RoO) requirements. RoO can be a significant trade barrier where the local content requirement is high, particularly where the product is quite complex and relies on multiple diverse components. Since the predominant automotive component trade flow in Africa is from South Africa to SADC countries, the SADC RoO for HS87 is relevant. For built up vehicles, the requirement is 40%, which is lower than that for components. The minimum local content for components varies between 50 and 60%, depending on the item.

These local content requirements do constitute a non-tariff barrier to deepening trade in this sector and the industry body in South Africa (NAAMSA) have expressed concern over this in the past[3]. Any true liberalisation within Southern Africa or Africa more generally would need to address RoO as an important non-tariff barrier to trade.

Conclusion

The AfCFTA would also eventually see all or most intra-African trade flowing across border unhindered by tariff or non-tariff barriers. The results of a simulation of complete intra-African liberalisation of trade in automotive components are striking in that the primary beneficiary is South Africa. The losses due to trade diversion are distributed over a set of European, Asian and American exporters and the gains from lower prices are enjoyed by the set of African importers. There is some loss of customs revenue to African governments but a net gain in trade leading to increased intra-African integration. However, the effects are concentrated in Southern Africa among the members of SADC and little impacts are felt elsewhere in Africa. This is due to the skewed nature of the trade in HS8708 and the far higher level of intra-trade in Southern Africa and between South Africa and its SACU and SADC trade partners.

In the near future, the markets for many products will be liberalised across Africa, as the AfCFTA is implemented. Simulations such as the one presented here can contribute to understanding the impacts of liberalisation across countries and markets. This will help in designing mitigating policies and adjusting industrial and NTB policies to better contribute to an equitable distribution of gains.

---

[1] Stuart, J. 2020. The Automotive Components Trade in Africa: Its Place and Potential. tralac Working Paper. Stellenbosch: tralac.

[2] This disaggregated data is not presented here but is available in the working paper cited in footnote 1.

[3] TSG. 2004. The potential for regional integration of the SADC motor industry. Report prepared for SADC Secretariat. Arlington: The Services Group