Economic Statistics: problems of knowledge and manipulation
Morten Jerven
Routinely economic statistics inform and support big headlines, but statistics itself seldom makes the news. That was not so for 2014. The Financial Times recently called the GDP rebasing in Nigeria one of the biggest news stories out of Africa in 2014.
On one level the news of the jump in total GDP from Nigeria is good news. Of course it is a good thing that Nigeria is richer than we thought. It is a symptom that the now biggest economy in Africa is working to improve data needed for economic governance, but it also points to big knowledge problems and data gaps.
There has been a lot of writing on political motives and the veracity of the new numbers of Nigeria. They are indeed preliminary estimates, and have not yet been validated with data on the demand side of the economy (the new numbers are derived from estimating total output) and scholars will continue to struggle to square higher income and higher growth, with continued high levels of poverty.
The real startling fact about the rebasing is perhaps that technically there is nothing wrong about Nigeria’s GDP doubling from one day to another. Is the new number too high or too low? That one can debate without any prospect of final settlement. GDP is always an approximation. Any GDP estimate depends on a range of assumptions and data points that can all be contested.
All bathroom scales tells you the wrong total weight, by a few grams, or by several kilos. This does not matter if you are interested in whether you are gaining or losing. The problem appears when you compare your own weight to your neighbor’s. Or if the scale changed overnight and now tells different total weight. That is what happened in Nigeria. They changed the scales. The previous was using weights from 1990, the new one is based on 2010 data. The same happened in Ghana when they changed from a 1993 to a 2006 benchmark year.
Nigeria had used weights from 1990 to estimate economic gains until 2014. How could it take almost a quarter of a century before the data for economic governance was updated? The answer is a matter of priority. GDP numbers are a product of the national statistical system. Many national statistical offices have been neglected since the 1980s. There were more pressing concerns than investing in reliable and regular data on economic activities.
With the publication of Poor Numbers this gap in knowledge became clear for everyone to see. The African Development Bank responded by saying the rise of Africa was equally plain for everyone to see. The chief Economist for Africa at the World Bank reacted to the Ghana revision by declaring Africa’s “Statistical Tragedy.”
Data for economic governance has been neglected for too long, and as the information is updated, there is bound to be a few surprises. The credibility of official statistics will benefit from a transparent and open process as the supply of data from African economies on the rise is meeting the demand of an increasing number of observers that are suddenly paying attention.
Ed’s Note: Morten Jerven is Associate Professor at the Simon Fraser University, School for International Studies. His book Poor Numbers: how we are misled by African development statistics and what to do about it was published by Cornell University Press in 2013. In 2015, Jerven’s fourth book, Africa: Why Economists Get It Wrong will be published.