African Development Successes

December 11-12, 2009
Sebastian Edwards, Simon Johnson, and David N. Weil, Organizers
Research Conference #1 of 3 on Economic Research on African Development Successes
Cambridge, Massachusetts



Sponsored by the NBER Africa Project, with support from the Bill & Melinda Gates Foundation.

The NBER Africa Project was launched in Fall 2007 to study economic success stories in Africa. The primary goal of the project is to support high quality research that identifies and analyzes African economic development successes, ultimately to inform policy making at both the national and international levels. Of particular interest are the positive economic experiences in a number of African countries over the last decade. What are the roots of these developments? To what extent are they sustainable and transferable to other African countries?

Forty research projects have been commissioned. This first full research conference of the Africa Project included five final presentations, below.


Agriculture, Roads, and Economic Development in Uganda (NBER Working Paper No. 15863)
Douglas Gollin, Williams College
Richard Rogerson, Arizona State University and NBER

Development economists have long recognized that economic growth is accompanied by a structural transformation of economic activity. Perhaps the most evident change is a movement away from agriculture and into manufacturing and services, which seems to be a central feature of growth. In many parts of Africa, however, this transition appears to be occurring very slowly. A large fraction of the sub-Saharan population continues to earn a living from semi-subsistence agriculture: in some countries more than 75 percent of the population is employed primarily in agriculture, and much of their production is intended for home consumption, not for the market. Gollin and Rogerson focus on the high cost of transportation as one explanation for this. African road networks are notoriously poor, especially in rural areas. High transportation costs necessarily imply high input costs and low output prices for agriculture. Building on recent field work in Uganda, the researchers document some key facts about transportation costs and offer a simple general equilibrium model to explore some counterfactual experiments. Their results suggest that the fraction of the population in semi-subsistence agriculture is highly sensitive both to agricultural productivity levels and to transportation costs. Their model also suggests positive complementarities between simultaneous improvements in agricultural productivity and transportation.


Evaluating the Effects of Large Scale Health Interventions in Developing Countries: The Zambian Malaria Initiative (NBER Working Paper No. 16069)
Nava Ashraf, Harvard University and NBER
Gunther Fink, Harvard University
David N. Weil, Brown University and NBER

Since 2003, Zambia has been engaged in a large-scale, centrally coordinated national anti-Malaria campaign which has become a model in sub-Saharan Africa. This paper aims at quantifying the individual and macro level benefits of this campaign, which involved mass distribution of insecticide treated mosquito nets, intermittent preventive treatment for pregnant women, indoor residual spraying, rapid diagnostic tests, and artemisinin-based combination therapy. Ashraf, Fink, and Weil discuss the timing and regional coverage of the program, and critically review the available health and program rollout data. To estimate the health benefits associated with the program rollout, they use both population based morbidity measures from the Demographic and Health Surveys and health facility based mortality data as reported in the national Health Management Information System. While they find rather robust correlations between the rollout of bed nets and subsequent improvements in their health measures, the link between regional spraying and individual level health appears rather weak in the data.


Deals versus Rules: Policy Implementation Uncertainty and Why Firms Hate It (NBER Working Paper No. 16001)
Mary Hallward-Driemeier, The World Bank
Gita Khun-Jush, University of Chicago
Lant Pritchett, Harvard University

In surveys of firms in developing countries, "policy uncertainty" is a major complaint and is perceived as an obstacle to firm expansion. Several recent papers have argued that it is "institutions" and not policies that matter for long-run economic growth. Hallward-Driemeier, Khun-Jush, and Pritchett reconcile these two strands of the literature by showing that what matters for firms is not (just) inter-temporal uncertainty about shifts in notional policy but also uncertainty about policy implementation - how the realized policy actions taken will affect the profitability of their firm. What is meant by "institutional quality" is fundamentally related to the reliability with which the direct organizations and background institutions responsible for policy implementation translate notional into realized policy. Firms base their investment decisions by forming expectations over trajectories of profitability, which are influenced by their expectations of realized policy actions. Once these three inter-related elements of the link between policies and outcomes are acknowledged - policies are mappings; positive models of policy implementation not notional policy mappings matter; and, expectations determine responses - then everything about the performance of "policy reform" can be explained. In particular, in weak institutional environments, business is done through "deals" - individual firm/person specific accommodations with the policy implementation apparatus about how their specific businesses will be treated: big business makes big deals, little men make little deals (only losers (try to) follow the rules). These authors show how many dimensions of the growth experience in Africa can be understood in a framework in which policy implementation uncertainty plays a large role.


International and Intra-national Market Segmentation and Integration in West Africa
Jenny Aker, Tufts University
Michael Klein, Tufts University and NBER
Stephen O'Connell, Swarthmore College

Borders can segment markets, with important consequences for market performance and welfare. While international borders separate markets in different countries, the geographic concentration of ethnic groups may give rise to economically-relevant intranational borders within countries in sub-Saharan Africa. Aker, Klein, and O'Connell assess the extent to which international and intranational borders segment markets for staple food and cash crops between Niger and Nigeria, and also within Niger. They find that the international border effect is statistically significant, yet of limited economic magnitude for grains and cash crops. The border effect is more pronounced for an intranational border in Niger that separates the Hausa and Zarma regions.


Cape Verde and Mozambique as Development Successes in West and Southern Africa (NBER Working Paper No. 16552)
Jorge Braga de Macedo, Universidade Nova de Lisboa and NBER
Luis Brites Pereira, Universidade Nova de Lisboa

de Macedo and Pereira assess the extent to which Cape Verde and Mozambique reveal a positive interaction between globalization and governance in the orientation and predictability of economic policies, as well as the accompanying institutional arrangements. Economic success under globalization involves market perceptions regarding outcomes such as export diversification and narrowing of the income gap with the frontier. Economic success is in turn sustained by good governance and political and economic freedom. The context is provided by cooperation agreements to which Cape Verde and Mozambique belong, notably in their respective sub region of Africa, where a convergence-diversification regime and a divergence-specialization regime can be defined. Comparing regimes across sub regions, the researchers find that West Africa countries are becoming more diversified while in Southern Africa they are becoming more specialized. Opening up to trade is also an important driver of both convergence and diversification for the former, especially in the range of 45-75 percent of GDP, but not for the latter. Finally, in the Southern Africa convergence-diversification regime, economic and political freedom drive convergence, suggesting effective institutional arrangements.