Poor Numbers

Poor Numbers is the first analysis of the production and use of African economic development statistics.

Morten Jerven is Assistant Professor in the School for International Studies at Simon Fraser University.

The book is remarkable given that it is largely the result of the efforts of a single individual gaining access to NSOs.... Highly recommended to all those interested in the SSA region and in the measurement of economic activity in developing countries. Its publication has already started a much-needed lively discussion, which is a precondition for improving the quality of macro-economic statistics.

This important book attempts to systematize what most quantitative practitioners in Africa generally understand: African macroeconomic data are poor.... Using a variety of sources that include current surveys of in-country statistical collection agencies and firsthand historical accounts, Jerven outlines several root causes of the data problem, which include Africa's colonial heritage and the more recent, structural adjustment policies. He continues his analysis by exploring how data are consciously shaped by both local and international politics and international aid agencies. Specifically, Jerven is critical of World Bank transparency and its unwillingness to provide him with quantitative methodologies of its official data compilation.... This volume opens up a venue for a research paradigm that could lead to much-needed improvements in the collection of African data. Summing Up: Highly recommended.

"Increasingly, scientists turn to the large statistical databases of international bodies when testing favoured hypotheses to control for growth and economic development. They might hesitate after reading Poor Numbers.... This book offers fascinating, disturbing insights for anyone interested in the role of numbers in the social sciences. For those using global economic databases, it should be required reading."

Poor Numbers is a powerful little book..., highlighting the risks of making political inferences solely based on statistical analysis...Although an economist by training, Jerven's clear prose without jargon helps make Poor Numbers reach a wider readership. It is imperative to note that his is not a simple criticism of quantitative methodology, but of the confidence one has in the findings of quantitative analysis without due attention to the quality of the data. In this sense, even those who have no scholarly interest in African development economics would find the findings and conclusions pertinent to the foundational debates on the role of methodology and theory in political science.

[Poor Numbers]is a useful reminder of the dubious information content of economic indicators generated by national accounting systems of sub-Saharan African states. I recommend the book to all scholars and researchers who contemplate the use of data generated by national accounting systems of sub-Saharan African countries.

Nicolas van de Walle, Cornell University, author of African Economies and the Politics of Permanent Crisis, 1979–1999:

In Poor Numbers, Morten Jerven takes on the issue of inaccurate macroeconomic data in Sub-Saharan Africa. First, by describing collection methods, he shows quite convincingly that the data are pretty dreadful, and perhaps more damning, that they may include systematic and predictable flaws linked to the way in which they are collected and aggregated. Jerven demonstrates that basic national accounts data are too poor to assess very basic characteristics of African economic performance since independence. This short elegant book is fascinating and strikes me as a must-read for any social scientist interested in African political economy and policy.

Carol Lancaster, Dean of the School of Foreign Service and Professor of PoliticsGeorgetown University, author of Aid to Africa: So Little Done, So Much to Do:

I found Poor Numbers illuminating and disturbing at the same time—I think that is exactly what Morten Jerven intended. It is well written, even elegant in some places. Jerven's recommendation that more funding be put into statistical services to do baseline surveys and field-based data collection makes a lot of sense.