Does wealth imply secularization and longevity?

FOR MOST OF the world wealth is positively associated with longevity (Preston 1975) and negatively associated with religiosity (Barro and McCleary 2003). However, not all countries conform with the overall pattern. As shown in Figures 1 and 2, the United States presents a very visible exception. If wealth is measured by real per capita GDP, longevity is measured by average life expectancy at birth, and religiosity is measured by belief in an afterlife, then by these measures the United States is very wealthy but also very religious compared to its OECD peers, wealthy and religious but not especially long-lived. Although some aspects in Figures 1 and 2 have been explained by economists and sociologists, no satisfactory explanation exists for the general patterns and for the United States exception in each of the figures, and no explanation exists at all that connects the two figures. (1) We fill this gap by developing a simple life cycle model with endogenous longevity and endogenous religious beliefs or faith.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

At its core, this paper asks: what are the economic forces that drive secularization and longevity? The answer is not at all evident because, as Iannaccone (1998) argues, previous work on the socioeconomics of religion has left open the question of religious belief formation. For instance, theories following Azzi and Ehrenberg (1975) and Iannaccone (1990) emphasize the demand for religion by individuals who base their choices on the perceived returns to religion, which depend on a combination of earthly and afterlife benefits. These approaches have generally maintained that religious activity depends on exogenous lifetimes while downplaying the role of afterlife benefits. But it seems plausible that religious activity alters the returns to longevity and religious beliefs, just as longevity or beliefs alter the returns to religious activity. Other theories emphasizing the supply of religion by religious organizations have focused on the forces that influence the market structure for religious firms. While these supply-side approaches can explain why the United States differs from other countries in terms of religious activity (Stark and Iannaccone 1994), they only indirectly address Figure l because they also take belief formation as given. Furthermore, the supply-side theories have little to say about Figure 2 because they are silent about connections to other dimensions of individuals' life cycle choices such as their life expectancy. Some of the determinants of life expectancies have been explored in the longevity production approach by Philipson and Becker (1998). Their model implies that life expectancies vary positively with wealth, which leaves open why the United States has a lower life expectancy than its OECD peers despite higher wealth. Their model also leaves unexplored the connection between religiosity and longevity, despite some controversial research (Sloan, Bagiella, and Powell 1999) on the links between religion and health.