On March 21, just weeks ahead of a scheduled election, a band of soldiers seized the presidential palace in Bamako, Mali, and toppled the government of President Amadou Toumani Touré. What started as a mutiny over lackluster support for the armed forces in their fight against Tuareg insurgents in the country’s desert north wound up puncturing a two-decade run of elected government.
The coup in Mali shocked many observers, including those with deep knowledge of the region. “Mali’s [coup] was a complete surprise; nobody predicted it,” Alex Vines of London’s Chatham House think tank told GlobalPost. On the African Arguments blog, Jolyon Ford of Oxford Analytica wrote, “One can be confident saying that many in the business of risk analytics in Africa would not have predicted the Mali coup.” The Failed States Index is not meant to predict coups, but it’s worth noting that in 2011 Mali ranked 76th, right in the middle of the global pack.
That the Mali coup surprised so many Africa watchers and professional risk analysts does not mean it was inherently unpredictable. True, the subjective forecasts of even knowledgeable people typically aren’t much more accurate than chance, as psychologist Philip Tetlock showed in his tour-de-force 20-year study of expert political judgment. Statistical models, however, can often assess risks fairly accurately, even when the problem is complex and the models are simple.
As it happens, Mali appeared ripe for a coup in forecasts I published at the start of 2012, a couple of months before Capt. Amadou Sanogo and his colleagues seized power in Bamako. According to those forecasts, Mali ranked among the 10 countries most likely to experience a coup attempt this year. Second on that list was Guinea-Bissau, a neighbor of Mali that went on to suffer a successful coup of its own just a few weeks later.
Using statistics to forecast political events isn’t as complicated as it sounds. At its core, it’s all about recognizing the right patterns. Generally speaking, you start by building a list of similar events in relevant cases in the past; then you assemble data on likely risk factors; next, you use statistical techniques to generate an algorithm that captures useful patterns in those data; and, finally, you apply that algorithm to current data to get a forecast.
My 2012 forecasts pushed Mali toward the top of the list because Mali’s structural conditions at the end of 2011 looked a lot like conditions in other countries that have recently suffered coup attempts: in this case, a poor country with a partially democratic government and an ongoing insurgency in a tumultuous region. Read more on FP