Abstract:We look at the effect of religiosity on social trust, defined as the share of a population that thinks that people in general can be trusted. This is important since social trust is related to many desired outcomes, such as growth, education, democratic stability and subjective well-being. The effect of religiosity is theoretically unclear: while all major religions call for behaving well to others, religious groups may primarily trust people in their own groups and distrust others, as well as cause division in the broader population. We make use of new data from the Gallup World Poll for 105 countries and the U.S. states, measuring religiosity by the share of the population that answers yes to the question "Is religion an important part of your daily life?". Our empirical results, making use of regression analysis whereby we control for other possible determinants of social trust and, by using instrumental variables, for the risk of reverse causality, indicate a robust, negative effect of religiosity, both internationally and within the US. When interacting religiosity with the degree of religious fractionalization, we find that the negative effect occurs above a rather low threshold level of fractionalization, where it increases monotonically, in line with our theoretical prediction.