ETHNIC NETWORKS BOOST TRADE BETWEEN COUNTRIES: Evidence from Africa

16 Apr 2019

The presence of ethnic networks between countries can lead to trade between these countries almost doubling. That is the implication of research by Conrad Copeland, to be presented at the Royal Economic Society's annual conference at the University of Warwick in April 2019.

Using an entirely new database for ethnic groups on the continent of Africa, the study shows that when two countries have a common ethnic group, trade between them nearly doubles.

Importantly, this effect is not based on the relative sizes of the ethnic groups in the two countries, which means that the increase in trade is likely to be due to the effect of local networks developed within these ethnic groups.

What’s more, the research demonstrates, the strength of these networks is persistent over long historical periods. Networks for ethnic groups that were historically prominent in trade across Africa continue to have a disproportionate impact on trade between countries.

While the existence of local networks, particularly ethnic networks, has been researched within individual countries, this is the first attempt to analyse the effect they have on bilateral trade relations at a continent-wide level.

These networks can substitute for formal contractual and logistical relationships where official infrastructure may be lacking. These networks have been found to have a significant effect on local markets and goods, but the question of scalability and the effect on national trade levels has, until now, been left unaddressed.

The main result of the new study is the dramatic increase in trade caused by the presence of the same ethnic group in two partner countries. The author theorises that this effect is caused by the interaction of individual networks at the local level, with the trust and contract enforcement mechanisms of ethnic networks substituting for weak state enforcement mechanisms.

The key to this analysis is demonstrating that the size of the ethnic population in each country does not matter: it is merely that the group is present in both countries. This leads to the conclusion that these effects are not driven by similarities in the population, but are rather a result of the local networks cross-border contacts can create.

Furthermore, the analysis finds that these effects are robust across long time horizons. This is illustrated by the fact that the networks for ethnic groups that were historically prominent in trade across Africa continue to have a disproportionate impact on trade between countries. While the precise drivers of this historical persistence need to be further investigated, the new research marks an important first step in this direction.

In order to investigate questions of ethnicity and trade in Africa, the research uses an entirely new dataset developed specifically for this project. This dataset brings together information from disparate sources to catalogue fully both the ethnic composition of partner countries and their trade relations. The author then develops several unique indices of ethnic commonality for countries and tests their impact on trade flows.

With this new database, the author is able to demonstrate the strength of the effect of common ethnic links on trade between pairs of countries in Africa. Fostering these links and the trade networks around them can significantly help encourage trade integration across the continent. Similarly, encouraging institutional development in the areas that these networks replace will go a long way towards strengthening trade networks between countries.

Conrad Copeland

University of Bristol | +44 7850 576976 | conrad.copeland@bristol.ac.uk