Modelling Africa’s trade routes

People talk a lot about how development aid might be used to improve a country’s attractiveness as a trade partner. (Mostly the World Trade Organisation, but not exclusively!)

“Aid for Trade” is a controversial project because it has a distinctly globalisation-friendly vibe about it, and a fundamental belief in the kind of trickle-down economics so beloved of market-oriented people and organisations.

But one thing that is never discussed when the possibility is raised of improving a country’s export competitiveness, is that in the absence of additional global demand, any increase in export due to an Aid for Trade programme must be accompanied by a reduction in exports for somebody else.

With the global economic model I’ve built as part of my PhD (and some fabulously bold assumptions about how trade works), I can have a stab at modelling which countries stand to gain and which to lose from a particular Aid for Trade project.

This picture, drawn using ESRI’s flashy new ArcGIS Pro shows the modelled results of improving Ethiopia’s export infrastructure. (It’s “inspired” (for which read, pinched) from this nice flight paths visualisation.)

Flows which increase are in blue, and those which decrease are in red.


Here are the boring but very necessary caveats:
– Only those between African countries are shown.
– Because the increased flows are much bigger in magnitude than the decreased ones (at least within Africa) I’ve had to compromise on the line thicknesses, leading to an overstatement of the decreases!! Caveat emptor!!
– This is based on a gravity-type trade model. Their use in predicting trade is controversial.
– The economic descriptions used in this model are based on estimates, since most African countries don’t publish the kind of economic data you’d need to build a proper description.