African Statistics: Poor Numbers and Political Wrangling

Morten Jerven, an academic at Canada’s Simon Fraser University, published an important book for those who believe that decision-making should be informed by facts. His book, Poor Numbers: How We are Misled by African Development Statistics and What to Do About It, was published earlier this year by Cornell University Press.

Jerven analyzes the production of African development statistics, showing that the process is in disarray, not least because of historical factors dating back to the colonial period, as well as more recent resource and management issues. His bottom line is that African statistical data is unreliable. Africa may be “rising,” but that conclusion should not be based on official statistics. The book is an academic study, based on field research, and has been peer reviewed. It has been favorably reviewed and has generated widespread discussion in the foreign policy and development community. Unlikely for a study of statistics, the book is also a good read.

But some African reactions to the book have been bitter. Jerven was uninvited from giving the keynote speech at September’s Addis Ababa meeting of the UN Economic Commission for Africa. Simon Allison in South Africa’s Daily Maverick reports that South Africa statistician-general, Pali Lehohla, the head of Statistics South Africa, threatened to withdraw South Africa’s delegation if Jerven spoke. Allison expresses surprise, because Statistics South Africa is usually regarded as the continent’s best statics office, and Jerven discusses it little in the book. According to the media, Lehohla claims that Jerven’s scholarship is faulty. Not so, according to a wide range of experts. Jerven’s work was replicated by teams at the IMF, African Development Bank, and UNECA. They came to the same conclusions as he had. Among others, Allison quotes Richard Watts of Government Spending Watch as saying, “Morten Jerven is of course right–the capacity and resources going into statistical offices in Africa is generally low, and a real issue.”

Allison’s interview with Lehohla, with accompanying discussion and commentary, provides a fascinating insight into the persistence of “an old-fashioned pan-Africanist of the Thabo Mbeki school.” He concludes that “there is perverse encouragement in all this is that African data is being debated at all.” Allison’s interview is fascinating and deserves a blog post of its own.