Working Papers

Thilo R. Huning

Research fellow in economics

Geography matters.

We only begin to understand how economic activity is shaped by geography, as standard growth models cannot explain why some nations fall behind, and some prosper. Why do cities emerge? How did all these 'good' institutions come along? How did the Monstrum Germanicum become the core of the European Union? What role does trade play? What are the implications for today's economic policy?

Working Papers

"You Reap What You Know: Observability of Soil Quality, and Political Fragmentation", joint with Fabian Wahl

Berkeley Economic History Labl (BEHL) Working Paper 2016-05 and European Historical Economics Society Working Paper No. 101

Abstract. We provide a theoretical model linking limits to the observability of soil quality to state rulers' ability to tax agricultural output, which leads to a higher political fragmentation. We introduce a spatial measure to quantify state planners' observability in an agricultural society. The model is applied to spatial variation in the 1378 Holy Roman Empire, the area with the highest political fragmentation in European history. We find that differences in the observability of agricultural output explain the size and capacity of states as well as the emergence and longevity of city states. Grid cells with higher observability of agricultural output intersect with a significantly lower number of territories within them. Our results highlight the role of agriculture and geography, for size, political, and economic organization of states. This sheds light on early, though persistent, determinants of industrial development within Germany, and also within Central Europe.


"How Britain Unified Germany: Endogenous Trade Costs, and the Formation of a Customs

Union", joint with Nikolaus Wolf

Abstract. We analyze the foundation of the German Zollverein as an example how geography can shape institutional change. We show how the redrawing of the European map at the Congress of Vienna 1815–notably Prussia's control over the Rhineland and Westphalia–affected the incentives for policymakers to cooperate. Our argument comes in three steps. First, we show that the new borders were not endogenous to trade. They were at odds with the strategy of Prussia in 1815, but followed from Britain's intervention at Vienna regarding the Polish-Saxon question. Second, we develop a theoretical framework, where state planners set tariffs on imports and transits to maximize revenue. We show that in a world with transit tariffs a revenue-maximizing state planner faces a trade-off between benefits from cooperation and the cost of loosing geographical advantage. In a third step we calibrate the model combining historical data on tariffs, freight rates, market sizes with GIS data on lowest costs routes under endogenous tariffs. We then run counterfactuals to show how borders affected incentives: if Prussia would have succeeded with her strategy to gain the entire Kingdom of Saxony instead of the western provinces, the Zollverein would not have formed. We conclude that geography can shape institutional change. To put it differently, as a collateral damage to her intervention at Vienna ``Britain unified Germany''.


"Infant-Industry Protection With Endogenous Trade Costs: Evidence From Pre-Zollverein Prussia"

This paper argues that the net effects of infant-industry protecion cannot be analyzed neglecting geography. Assymetric geography, relative size and position, both limits the ability of states to impose reciprocal tariffs, and allows state to have other states's consumers bear the welfare-loss induced by tariffs. This model is then applied to Pre-Zollverein Prussia and shows how the Prussian customs policy allowed to keep British products out of the Prussian hinterland. It could explain how the ability to impose endogenous trade costs can decide whether a tariff is welfare-neutral for the domestic consumers, and why infant-industry protecton did work in the 19th century but would fail in today's transport technology and institutional framework.