Using Data To Predict Civil Unrest In Africa

Avatar
Written by Andrew Friedman

There is a debate among scholars of war and peace as to whether we live in an unprecedented era of peace or an era when violence is consistent with historic norms but has moved to different venues.

The reason for this debate is an amazing lack of interstate (e.g. U.S. v. England) wars, at levels unseen since the beginnings of the modern state system. However, while there are now comparatively few state-versus-state wars, civil wars or insurrections within states are experiencing new highs. The causes are as diverse as the world itself; however, more of these wars tend to take place in Sub-Saharan Africa than anywhere else in the world. Half of all current U.N. peacekeeping missions are on the African continent and by some counts there are as many as 24 ongoing conflicts on the continent that are predominantly intrastate in nature.

Coups, civil wars and violent upheaval are not only responsible for untold human misery; they are also problematic for businesses operating in-country. Interruption in business activities, lack of consumer attention and difficulty with supply lines are just a few issues. It behooves businesses to understand when and where the next such event will take place. In a business environment that requires a multinational approach, it has never been more important to be able to predict instability of all sorts when deciding where to operate and what protections are necessary.

The last few years of instability in Egypt have been particularly telling. The economy has been in free fall. The country’s once-vaunted tourism industry is no longer attractive. This has predictably affected other areas of the economy. The difficulties of civil turmoil are not limited to tourism. The civil war that would ultimately oust former Libyan strongman Muammar Gaddafi caused a dent in oil production during the upheaval and even now, with a challenging security situation, has oil production at about one tenth of pre-war levels.

With this in mind, data-driven quantitative trends in political science make the field more useful to international investors than ever before.

Dr. Jay Ulfelder, an independent researcher who writes at the blog Dart-Throwing Chimp, compiles an annual list meant to predict the potential for coups throughout the world. In Ulfelder’s 2012 predictions, he did remarkably well. While coup attempts are excessively rare events, they took place in Guinea-Bissau and Mali, both of which were in the top 10 most likely states and both of which are in sub-Saharan Africa.

The updated predictions for 2013 paint a similarly bleak picture for the continent. Nine of the top 10 most likely states are sub-Saharan, along with 22 of the top 30 (roughly the top quintile of countries and what Ulfelder regards as the high-risk group). Additionally, the top five are all sub-Saharan, including Guinea-Bissau, Sudan, Mali, Madagascar and Mauritania.

While Ulfelder said the models “…were not meant to anticipate civil wars, non-violent uprisings, voluntary transfers of executive authority, (self-determined dissolutions of government), or interventions by foreign forces, all of which are better thought of (and modeled) as different forms of political crisis,” there is, both intuitively and empirically, a correlation between tendency for coups and civil upheaval of most other sorts.

Ulfelder spoke to AFKInsider via email. He said few of the statistics do the “heavy lifting” for predictions in all types of civil unrest so there is some correlation between the coup predictions and other types of civil unrest. Perhaps the most prominent of these, even greater than governance type or gross domestic product per capita, is the infant mortality rate.

“Infant mortality seems to be a really powerful summary measure of a bunch of things, from wealth to governance to disease burden and the like. It usually has more predictive power than gross domestic product per capita,” he said.

Great damage can be done to businesses operating within states that see massive civil upheaval. For this reason businesses, when choosing where to operate, should look to new and more advanced predictions on civil unrest from quantitative political science, or at least examine the indices that do the best job predicting such upheaval.

Infant mortality seems to be the most prominent of such statistics. Examining these predictions along with such statistics will make a number of decisions easier, including whether to begin operations in-country and whether to wait for insurance from the World Bank’s Multilateral Investment Guarantee Agency or another insurance company for civil upheaval.

Andrew Friedman is a human rights attorney and consultant who works and writes on legal reform and constitutional law with an emphasis on Africa. He can be reached via email at afriedm2@gmail.com or via twitter @AndrewBFriedman.