Abstract: In this paper we try to shed new light on the aggregate religion puzzle: the fact that the positive relationship between religiosity and cognitive subjective well-being that has been consistently found at the individual level does not emerge at the aggregate level. We posit the hypothesis that the motivation underlying the observance of religious prescriptions may be confounding the relationship of interest. Using data from the Integrated Values Survey 1981?2014, we select some variables as possible proxies for the prevalent religious motivation and discuss this novel interpretation. Then we extend a standard cross-country life satisfaction model including controls for both the average level of religiosity and the prevalent motivation. Results show that the average level of religiosity is ceteris paribus positive and significantly associated with average life satisfaction. Moreover, our main proxy for the prevalent religious motivation also emerges highly significantly associated with average life satisfaction. The estimations are subjected to a wide variety of robustness checks and results remain largely unchanged. Moreover we carry out a thoroughly discussion of the endogeneity issues that may affect our results exploiting the panel nature of the data. We conclude that our results seem reasonably reliable, are consistent with findings of the previous literature, and point to a possible explanation of the different evolution of the religious phenomenon across countries.

Authorship

Download reference