The effects of income on religiosity are theoretically ambiguous. Higher income could lead to an increase in consumption of goods and services or allocation of time to activities that are complements or substitutes for religion. I exploit natural experiments to estimate the causal effect of income on religiosity among two different populations. Using data drawn from the Health and Retirement Survey, I explore the exogenous changes in Social Security income generated by the 1972 and 1977 amendments to the Social Security Act to estimate the causal effect of retirement benefits on religiosity among retired workers. Naïve cross-section estimates suggest that Social Security income is positively related to religious attendance and (insignificantly) positively related to self-assessed religious importance. However, after controlling for endogeneity, I find income has no effect on religious attendance among the elderly yet a marginally significant religious importance for the elderly women. I then draw data from the General Social Survey to the study the effects of labor income, wages and labor force participation on religiosity among adults between the ages of 18-64 in the U.S. Using per capita state GDP as an exogenous shocks to labor supply and income, my findings suggest that increases in hours worked conditional on employment and earnings reduce both private and public religious participation.