You Reap What You Know: Observability of Soil Quality, and Political Fragmentation

Thilo Hunig is  PhD student at
Humboldt University, Berlin

New EHES working paper

Geographic conditions limited medieval rulers in their attempts to extract their peasants’ agricultural product. Soil quality determines agricultural output, and a high spatial variation of the quality makes it hard to observe what peasants could potentially harvest.

If observability is bad, peasants can cheat rulers. Therefore, states with a lower soil observability should have relatively more volatile and lower  income, and in the end have a weaker military. We can show that this mechanism explains parts of the large differences in territories’ areas in the late medieval Holy Roman Empire. Among the over 600 territories we count on a map of the HRE in 1378, we find a robust and significant link between observability of soil quality, and the political fragmentation of an area.

Fabian Wahl is Assistant Professor
at the University of Hohenheim

We digitized a map of 1378 Holy Roman Empire featuring over 700 territories. Roman and Carolingian legacy did not leave a unified German state, and despite technologically determined economics of scale in public goods provision (Alesina and Spolaore, 1997), especially military, these territories persisted with a high degree of sovereignty until the 19th century. This, historically unprecedented political fragmentation is still visible ,e.g. in today’s Germany many different dialect areas, whose borders are still constitute barriers to trade by causing significant border effects (Lamelli et al., 2015). We provide a model why an equilibrium with many small states was stable for centuries, and why variation in soil quality can explain parts of the equilibrium variation in state sizes.

We extend the theoretical model on geography, transparency and institutions outlined by Mayshar et al. (2013).  We make the model empirically testable by proposing a continuous observability measure, and connect observability to state size, assuming that states’ income determines military spending. We hypothesize that states on soils which are more spatially homogenous concerning their soil quality are usually larger, despite the effect of the level of soil suitability. Furthermore, we argue that city states primarily occurred in areas with a high variation in soil quality, as the territorial states were weak in this area, and furthermore, the specialization of cities on trade and proto-industry made their tax revenue independent from agricultural conditions what enabled them to survive.

1378 Holy Roman Empire

This provides evidence for the foundations of state building during the middle ages and hence the origins of modern territorial states. We develop a GIS measure, the soil observability index, which captures this theoretical idea, and find a robust positive relationship between soil observability, and sizes of medieval German territories and between low observability and the existence of city states.

The blog post was written by Thilo Hunig and Fabian Wahl.
The working paper can be downloaded here:

Alesina, A. and Spolaore, E. (1997). On the number and size of nations. Quarterly Journal of Economics, 112(4):1027–1056.
Mayshar, J., Moav, O., and Neeman, Z. (2014). Geography, transparency and institutions. Mimeo.
Lameli, A., Nitsch, V., Südekum, J., and Wolf, N. (2015). Same same but different: Dialects and trade. German Economic Review, 16(3):290–306.