Religiosity and Science: An Oxymoron? Evidence from the Great Influenza Pandemic

The paper is co-authored with Enrico Berkes, Davide Coluccia and Gaia Dossi.

Abstract:
This paper studies the relationship between religiosity and science over time. Focusing on the United States during the 1900 and 1930 period, we define a novel indicator of revealed religiosity that leverages naming patterns of newborn babies, and measure scientific progress through the universe of patents granted in the US over this period. We document that the relationship between religiosity and science is negative before 1918, and positive thereafter. Exploiting plausibly exogenous variation in exposure to the Spanish Influenza pandemic (1918-1920), we find that relatively more affected counties become both more religious and more innovative. We use individual-level data to shed light on the potential mechanisms. We show that in counties affected by the pandemic: i) pre-existing differences in religiosity increased, leading to a polarization of religious beliefs; ii) individuals in science-related fields, who were less religious before the shock, became even less religious than the rest of the population.