Rent seeking and hubs : towards a new economic geography of sub saharan africa

This thesis is a modest attempt to uncover the main features and the underlying drives of Sub Saharan urbanization process. We begin by describing the stylized facts of urbanization in that region. Sub Saharan Africa’s urban pattern features urban bias and urban primacy. Locational advantages and political effects have induced a bias in favor of political capitals and ports. Thus, literature seems to emphasize rent-seeking and hubs determinants at the expense of agglomeration economies strengthening the prejudice against Sub Saharan Africa biggest cities. We question that preconception by performing an empirical analysis investigating the relationship between urban concentration and economic growth. The novelty of our approach lies in the use of semiparametric methods along with a missing data imputation algorithm. Our empirical results point out the specificity of the urbanization patterns in different regions. This prompts us to better characterize the underlying drives of Sub-Saharan Africa urbanization process. This is the challenge of our theoretical investigation which is twofold. On one hand we have attempted to explain urban agglomeration in Sub Saharan Africa by both economic and political factors. We find that this rent-seeking behavior fuels the formation of large urban agglomerations in developing countries, via mechanisms of interregional income transfers. Such a finding may serve to explain a seemingly paradoxical aspect of urban development in SSA: agglomeration despite high trade costs. On another hand, we rely on international trade and hub effects to explain the formation of agglomerations. We are thus able to explain the evidence of the persistence and the fostering of huge agglomerations in coastal locations of Sub-Saharan Africa. We find that openness is likely to trigger agglomeration in the hub especially when transport costs are low. This result is consistent with Weber’s theory of location (Beckmann and Thisse, 1986) which states that in a star-shaped network without any dominant location, entry points are the optimal locations.(ECON 3) -- UCL, 200