At the heart of urban economics are agglomeration economies, which drive the existence and extent of cities and are also central to structural transformation and the urbanization process. This paper evaluates the use of different measures of economic density in assessing urban agglomeration effects, by examining how well they explain household income differences across cities and neighborhoods in six African countries. We examine simple scale and density measures and more nuanced ones which capture in second moments the extent of clustering within cities. The evidence suggests that more nuanced measures attempting to capture within-city differences in the extent of clustering do no better than a simple density measure in explaining income differences across cities, at least for the current degree of accuracy in measuring clustering. However, simple city scale measures such as total population are inferior to density measures and to some degree misleading. We find large household income premiums from being in bigger and particularly denser cities over rural areas in Africa, indicating that migration pull forces remain very strong in the structural transformation process. Moreover, the marginal effects of increases in urban density on household income are very large, with density elasticities of 0.6. In addition to strong city level density effects, we find strong neighbourhood effects. For household incomes, both overall city density and density of the own neighborhood matter.
Africa’s demand for urban housing is soaring, even as it faces a proliferation of slums. In this setting, can modest infrastructure investments in greenfield areas where people subsequently build their own homes lead to better quality neighborhoods in the long run? We study "Sites and Services" projects implemented in seven Tanzanian cities during the 1970s and 1980s, and we compare greenfield areas that received basic infrastructure investment (de novo areas) to geographically proximate greenfield areas that did not (control areas). Using satellite images, surveys, and census data from the 2010s, we find that de novo areas developed into neighborhoods with much better housing quality. Specifically, de novo neighborhoods are more orderly and their buildings have larger footprint areas and are more likely to have multiple stories and better amenities, due not only to the persistence of initial investments but also to private complementary investments. We also document the role of sorting of owners and residents, which only partly accounts for the differences in housing quality across neighborhoods. Finally, we study initially squatted areas that were also upgraded as part of “Sites and Services”, and our descriptive evidence suggests that they are now if anything worse than the control areas. We conclude that preemptive infrastructure investments can lead to neighborhoods with significantly better housing in the long run.
It is often harder than it seems to measure and trace how much productivity is increasing in a place. It becomes even harder in countries where national, let alone subnational statistics, are poor. In such countries, it is already difficult to tell where people live and how fast the population is growing. It is even harder to answer relevant policy questions regarding urban planning and transportation needs. Night Lights data can potentially help us find answers to these questions.