We study the co-evolution of religion, science, and politics. We first uncover, in international and U.S. data, a robust negative relationship between religiosity and patents per capita. The model then combines: (1) scientific discoveries that raise productivity but sometimes erode religious beliefs; (2) a government that allows innovations to diffuse, or blocks them; (3) religious institutions that can invest in doctrinal reform. Three long-term outcomes emerge. The “Western-European Secularization” regime has declining religiosity, unimpeded science, and high taxes and transfers. The “Theocratic” regime involves knowledge stagnation, unquestioned dogma, and high religious-public-goods spending. The “American” regime combines scientific progress and stable religiosity through doctrinal adaptations, with low taxes and some fiscal-legal advantages for religious activities. Rising income inequality can, however, empower a Religious-Right alliance that starts blocking belief-eroding ideas.

In earlier work we identified a robust negative association between religiosity and patents per capita, holding across countries as well as US states. In this paper we relate 11 indicators of individual openness to innovation (e.g., attitudes toward science and technology, new versus old ideas, change, risk taking, agency, imagination, and independence in children) to 5 measures of religiosity, including beliefs and attendance. We use five waves of the World Values Survey and control for sociodemographics, country and year fixed effects. Across the 52 regressions, greater religiosity is almost uniformly associated to less favorable views of innovation, with high significance.