| Summary: | <p>This thesis studies how the spatial distribution of economic activity matters for the transmission of, and is in turn affected by different shocks.</p>
<p>In Chapter 2, I show that the agglomeration-congestion cost trade-off studied by urban economists alters both the distributional effects of, and gains from trade. Correspondingly, I demonstrate that openness to trade affects the spatial distribution of economic activity across cities of different densities. First, I show that the export intensity of firms and sectors is higher in denser locations. I propose an open economy economic geography model that rationalizes these stylized facts. When testing the underlying mechanisms proposed by the model, I find that the higher export intensity in denser places is driven by differences in productivity across firms, differences in factor intensities across sectors and trade-specific gains from agglomeration that lead to lower variable export cost in denser places. In line with the model predictions, I further find that a decrease in trade cost, proxied by exogenous changes in export market access, leads to a reallocation of economic activity to denser places. In currently on-going work I am exploring whether ignoring the increase in aggregate congestion costs from the rise in spatial concentration leads to an overestimation of the gains from trade.</p>
<p>In Chapter 3 I study how an exogenous population shock affects the spatial distribution of population. To this end I exploit the end of movement restrictions for black South Africans at the end of Apartheid to generate an exogenous population shock. Until 1991, black South Africans were severely restricted in their location choices and many were forced to live in homelands. Following the abolition of apartheid, they were free to migrate. Given the general gravity structure of migration, a town closer to the homelands is expected to receive a larger inflow of migrants than towns further away. Using this exogenous variation, I find that on average there is no endogenous reallocation of population following this exogenous immigration. When separately looking at rural and urban places, I find that there is a displacement of incumbents in the former while there are additional inflows in urban areas.</p>
<p>Chapter 4 provides evidence for the importance of trading opportunities for settlement location during the Iron Age. It finds that regions in the Mediterranean that are better connected across the open sea have more economic activity proxied by the number of archaeological sites. This correlation becomes particularly strong around 750 BC when the Phoenicians started to regularly cross the open sea. Analysis at the global level corroborates the finding that more connected locations have a higher population density in 1 AD. Overall, I find that trade and migration shocks have important effects on the long-run distribution of population and economic activity in space.</p>
|
|---|