Poor Numbers: How we are misled by African development statistics and what to do about it
Abstract
This book's main asset is that the author, trained as an economic historian and initially an everyday user of macroeconomic statistics, examines how national income accounting data are put together. This exercise, albeit self-evident to me, is unfortunately a rarity among academics studying development in Africa. The author does not advocate throwing the baby out with the bathwater, on the basis that statistics are important and key to international aid policy decisions. What he does advocate is broader coverage of critical analysis of statistical production. I hope his book will drive such a move. He also asks for statisticians themselves to take a critical view of their own output and analyse the pertinence of the figures they produce. I support this call wholeheartedly. However, those who know Africa and its National Statistical Institutes (NSIs) will not learn a great deal from reading this volume. Although it is gratifying to see in black and white some of the misgivings development economists frequently have to deal with when comparing different statistics, this is not the place for a technical discussion on the relevance of a given national accounts statistic or concept in Africa. Surprisingly, the author does not touch on issues that are just as important to the estimation of Gross Domestic Product (GDP) as the elements discussed in the book. These are just some of the issues I would raise: the estimation of inflation and in-country price differentials, and how to estimate purchasing power parities for international comparisons.
