Lies, damned lies, and African statistics

“Figures often beguile me,” wrote Mark Twain, “particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: ‘There are three kinds of lies: lies, damned lies, and statistics.’” There is some dispute as to whether Disraeli actually used this phrase, but the it has stuck as a common colloquialism on the dubious value of statistics.

The role of statistics in making economic and other policy decisions, including those that affect support from donor and developing agencies, has re-emerged recently with a public spat between Morten Jerven an Associate Professor at the Simon Fraser University, School for International Studies in Vancouver and the author of “Poor Numbers: how we are misled by African development statistics and what to do about it”  (Cornell University Press) and Pali Lehohla, the South African Statistician-General. It seems that Jerven was on his way to Addis Ababa to deliver a keynote speech at a meeting of the United Nations Economic Commission for Africa (UNECA), when he was informed he was no longer invited. The reason for Jerven being uninvited to the meeting, it seems, was Lehohla. who was reported as being so angry at Jerven’s appearance on the agenda that he threatened to withdraw the South African delegation. A good analysis on this episode by Simon Alison was published by South Africa’s Daily Maverick.

Jerven argues that it is time to invest in better GDP statistics in Africa: The “lack of firm facts leaves convenient room for negotiation of the numbers when it is needed. Currently the data, if they are available, are not timely or of the quality needed, and hamper any serious agenda for economic governance or development planning”.

There is nothing new about a call for better statistics in Africa, and other developing countries. Agencies such as the UNECA’s African Centre for Statistics (ACS), aim to “revitalise and better coordinate statistics at UNECA Secretariat, enhance its statistical coordination function and help to strengthen capacities of statistical systems in Africa”. The ACS programme of work includes the promotion, coordination of and advocacy for statistical activities in Africa, the building of a data hub for the provision of development data on Africa, measurement of informal sector and informal employment, support for statistical training, strengthening household surveys capabilities in Africa, and providing technical assistance.

The Partnership in Statistics for Development in the 21st Century, otherwise known as PARIS21, was founded in November 1999 by the United Nations, the European Commission, the Organisation for Economic Co-operation and Development, theInternational Monetary Fund, and the World Bank, in response to the UN Economic and Social Council resolution on the goals of the UN Conference on Development. PARIS21 focuses its efforts on encouraging and assisting all low-income and lower middle income countries to design, implement, and monitor National Strategies for the Development of Statistics (NSDS) and to have nationally owned and produced data for all MDG indicators. An NSDS is expected to provide a country with a strategy for developing statistical capacity across the entire national statistical system. It would provide a vision for where the national statistical system should be in five to ten years. It would also provide a framework for mobilising, harnessing, and leveraging resources (both national and international) and a basis for effective and results-oriented strategic management of the national statistical system. 

Jerven argues that there is no shortage of plans out there to improve statistical services in Africa. The trouble, he says, is that each of these have their own protagonists who are trying to gain support for ‘their’ program or ‘agenda’ at the coast of actually improving measurements. Instead, we have a situation of “governance by ignorance”.

It’s not surprising that Jerven’s message sticks in the craw of African statistic agencies. Why, when all this work is being done over many years by various African and international agencies, does it take an academic from Vancouver to point these issues out?  While Lehohla contests the rigour of Jerven’s work, others attest to it strengths and it seems to have been generally well reviewed. This ruckus seems to be another example of how an attempt to silence a critic has had the opposite result. I suspect Jerven will sell more books as a result of this incident.

Lehohla argues:

He says it’s been peer reviewed. Of course it’s been peer-reviewed in the US, and not in Africa. That whole issue of writing Africa’s narrative by others who least understand it, it’s tragic.

He has a point, of course. Jerven’s work would be more compelling if he was able to show a closer engagement with the reform efforts underway in Africa. However, the decision to un-invite Jerven from a forum for such engagement, seems petty and a missed opportunity. Hopefully, the resulting public debate on African statistic systems will lead to an improvement in measurements and a greater capacity of statistic agencies to manage statistic systems. Everyone seems to agree this is necessary.