This paper evaluates the relative importance of a "culture of cooperation," understood as the implicit reward from cooperating in prisoner's dilemma and investment types of interactions, and "inclusive political institutions," which enable voters to check the power delegated to their representatives. I divide Europe into 120km × 120km grids and exploit exogenous variation in both institutions driven by persistent Medieval history. To elaborate, I document strong first-stage relationships

more »

... present-day norms of trust and respect and the forces that produced consumption risk-i.e., climate volatility-over the 1000-1600 period and between present-day regional political autonomy and the factors that shaped the returns from elite-citizenry investments in the Middle Ages, i.e., terrain ruggedness and direct access to the coast. Using this instrumental variables approach, I show that only culture has a first order effect on development, even after controlling for country fixed effects, Medieval innovations, and other norms of conduct. Conditional on these observables, the excluded instruments have no direct impact on development and the effect of culture holds within pairs of adjacent grids with different Medieval climate volatility. An explanation for these results is that culture but not democracy is necessary to produce public-spirited politicians and push voters to punish political malfeasance. Micro-evidence supports this idea.