The structure of livestock trade in West Africa

Brief post with the main findings of our paper

Cattle value chains support approximately 70 million people in West Africa and exports of cattle from ECOWAS are estimated to surpass USD 800 Million each year. This makes cattle alone the most important food product intraregionally. Yet sheep, goats, and camels are also raised and traded in West Africa. Because most livestock are raised in arid and semi-arid areas near the Sahel and the consumption concentrates in urbanized and coastal areas, livestock are transported primarily south and west to consumer markets through highly intricate animal trade networks. In our recently published paper, we used network analysis to study the topological and geographic structure of this trade using an animal movement database collected by the CILSS (Interstate Committee for Drought Prevention in the Sahel). This post briefly explains our 3 main findings and their implications.

Well-defined trade corridors, where markets trade with others nearby and disassortively

On average, markets traded with 1-2 other markets every 4-10 days. Only 2% of all the possible trade relationships between markets existed, where the “possible” number of connections is defined as the number of links that would exist if each market traded with every market in the network but itself. 2 in 3 movements were international, and despite some changes in the number and location of data collection points, some clear spatial patterns of trade were observed. For example, Shipments from Burkina Faso towards Benin, Togo, and Nigeria; and flows from northern Cote d’Ivoire towards the coast. We also found that markets were more likely to trade with others near them, and with markets whose degree centrality was negatively correlated with their own (in other words, markets traded disassortively).

But what does this mean? If markets were people and trade relationships were exchanges of goods between them, this means that people with many trade partners traded goods with people with few or no friends. This is common in human-built systems as they evolve towards, in general, increased efficiency. However, this has some important implications that lead us to the second finding.

Regional trade hubs

Although markets traded with few others on average, some markets concentrated significantly more trade than most of the markets. These role of these “hub” markets were associated with their structural and geographic location. The presence of hubs implies that if a livestock vaccination campaign were to be carried out to control a specific livestock disease, for example, it might be more efficient to pilot or focus the project on selected markets instead of the whole region. On the other hand, hubs can also make a network less robust to disruptions. Because relatively few markets handle most of the trade, their targeted removal could cause disastrous regional interruptions in animal supply, affecting food and nutrition security in extended geographical areas and even in other countries, well beyond the location of the affected market(s). Consequently, regional trade hubs should be protected. In Northeastern Nigeria, for example, agricultural markets targeted by Boko Haram and surrounding areas experience reduced activity after violent attacks.

Key livestock markets

We identified 9 key livestock trade markets by their quantitative contribution to trade connections, shipments, and volumes. Those hubs were either border, urban, or export markets, and one-third of them within 60 km to Nigeria.

In general, our findings reflect the fragmented nature of the regional road network inherited from the colonial period, highlighting the important role of road infrastructure, particularly in border areas. We also found peaks in small ruminant shipments in the months before and after Tabaski most of the years, concurring with a similar study in Mauritania.

Limitations

Our study only used complete, geolocated records in CILSS’s database for 2013-2017. These do not represent the entirety of trade in the region, nor all of the entries in the database. Notably, livestock shipments into Nigeria were under-represented, partly because their records tended to not specify the destination of the shipments. Other important trade axes in the region (such as Senegambia-Mauritania and Cameroon) are not covered by CILSS’ database. Related scientific publications:

Read or download the paper here.

This paper is the result of a collaboration with the UF African Networks Lab and the work was funded by USAID and the USDA.

Valerie C. Valerio
Valerie C. Valerio
PhD Candidate in Agricultural and Biological Engineering

My interests include building data and simulation models of agricultural value chains in developing countries, learning and teaching data analytics and communicating science.

Related