This volume critically addresses the empirical growth literature on Africa that, according to its author, is largely based on cross-section regressions, is grounded in faulty statistics, and provides a static narrative and dubious, stylized facts. Morten Jerven presents a plea for studying African economies from a longitudinal and individual perspective by focusing on four experiences in East Africa (Kenya and Tanzania) and Central-South Africa (Botswana and Zambia). Rather than trying to explain persistent and widespread stagnation, the research question to be addressed is, in Jerven's opinion, this: How did African economies perform and why did some countries perform better than others?

Jerven stresses that sustained growth took place from independence to the mid-1970s and that the negative differential with respect to the world growth happened only after 1975. Moreover, he highlights the wide variance in growth performance across Africa, avoiding the oversimplification of treating all African countries as parts of a single unit.

Jerven dismisses the “subtraction” method used by empirical economists, where the performance of developed countries provides the yardstick against which Africa's (poor) performance is measured and explained. This approach results in a depiction of Africa's entire postcolonial era as chronic growth failure or, to use William Easterly's expression, “lost decades” (“The Lost Decades: Developing Countries' Stagnation in Spite of Policy Reform 1980–1998,” Journal of Economic Growth [2001]: 135–57).

Throughout the book, Jerven pays detailed attention to the quality of data underlying national accounts, its internal consistency and biases (specifically, its low and variable coverage of subsistence activities over time), and the lack of reliable population figures. The large growth variance among different sets of gross domestic product (GDP) estimates for post-independence Africa proves their lack of reliability. According to Jerven, the misleading appraisal of African economic performance by empirical economists can be explained largely by their neglect of the quality of national accounts.

The growth experience of the four countries provides Jerven with an opportunity to challenge the relevant contributions in the volume Country Case Studies (vol. 2 of The Political Economy of Economic Growth in Africa, 1960–2000, ed. Benno J. Ndulu et al. [2008]). Jerven highlights the inconsistency between “good” policies and good economic outcomes, arguing further that poor-quality data with large margins of error may blur any difference between good and bad performances. He points out that such expressions as “bad policies” and “mismanagement” imply that there exists an objectively “good” and feasible policy. Further, he rejects the conventional view that associates Tanzania's and Zambia's poor performances with interventionist policies, in contrast to the view attributing Botswana's and Kenya's stronger performances to market-oriented policies. Export growth (diamonds) and decline (copper) would explain, according to Jerven, the opposite fate of Botswana and Zambia while “socialist” Tanzania and “capitalist” Kenya actually performed similarly meaning that their ideological choice did not really matter.

Jerven's thorough and thought-provoking contribution deserves some critiques. For instance, he fails to mention the continuity between colonial and postcolonial, pre-1975 growth; thus, I question the consistency of his association between interventionist policies and growth. It is also worth noting that Jerven fails to consider that while pre-1975 growth in Africa took place at a pace similar to that of other world regions, it started from a much lower level. In a Solovian growth model, one could argue, this lack of convergence would imply underperformance.

In Jerven's account, growth was exogenously driven with institutions and policies playing a minor role. However, the fact that countries respond differently to similar shocks reveals that growth is partly endogenous. He also doubts that a causal link can be established between “bad” policies and “bad” performance. However, my own estimates show that “bad” policies correlate negatively with human development. Furthermore, policymaking mistakes are largely excused as a result of limits to the information available to decision makers.

Jerven's emphasis on a lack of data and a frequent discontinuity in sources is somewhat surprising coming from a professional economic historian. He does not take advantage of the cumulated experience of other economic historians, specifically their reconstruction of historical national accounts in currently advanced economies, in approaching his study of post-independence African economic history.

The lack of continuity in national accounts in African countries, while not a unique situation, presents a more important problem. Jerven describes how the introduction of a new base year implies a new and wider coverage of economic activities, but he does not mention that it also represents a significant upward revision of the level of GDP in the year in which the new and the old series overlap (often the new benchmark year). The author tries to circumvent the obstacle by studying periods in which continuous short series exist, resorting to the use of growth rates in order to compare performance tendencies between these periods. Jerven wrongly argues, “Since the comparison is between growth rates, it is not necessary to harmonize” the discontinuous national accounts series (p. 42). Unfortunately, this is no solution, as the undercoverage of subsistence activities, which varies over time, is not addressed in his comparative exercise. In fact, what Jerven calls “statistical” growth—that is, the growth that results artificially from a comparison of estimates that include the subsistence sector with estimates that underreport subsistence activities—is not prevented unless different national accounts series are properly spliced. What is actually needed is an adequate splicing strategy to combine and homogenize the available national accounts series into a single, consistent one. In fact, one needs to establish homogeneous GDP levels before estimating growth rates; only then can a proper assessment of the performance of African countries be carried out.

Thus, by failing to provide an alternative to the available quantitative evidence, that is, the available set of national accounts estimates, Jerven falls short of constructing an alternative and superior narrative of post-independence performance and instead does a neat demolition job on national accounts of post-1960 African countries. Nevertheless, both economic historians and development students would profit much—and certainly become wiser—from reading this volume.