Field of Science

Secularisation in the US will be swamped by religious fertility and immigration

There was a lot of noise recently about the ARIS survey, which showed a dramatic increase over the past decade in the numbers of non-religious Americans.

So, what does the future hold? Ever more secularisation? Perhaps, if people continue to switch out of religion.

But the demographic picture is more complicated than that. Religious people have more children than non-religious. Sure some of their children will lose their faith as they reach adulthood, but only a minority.

Then too the USA is a nation of immigrants, and immigration continues at a high rate. Immigrants tend to be more religious than the natives (they tend to come from poorer, more religious countries).

The Association for the Study of Religion, Economics, and Culture is just concluding its 2009 conference in Arlington, Virginia. Among the presentations was one by Prof Eric Kaufmann of Birkbeck College, London.

Using estimates of switching, fertility, and immigration for the religious types, he's put together a projection for what the religious picture will look like in the USA in 2050. Here's what he concludes:

  • The main drivers of religious affiliation to 2043 are immigration and secularization. However, fertility matters more in the long term.
  • Muslims will outnumber Jews by approximately 2020
  • Jews, white Catholics and liberal Protestants will decline
  • Protestants will decline from a majority in 2003 to 40 percent by 2043; Catholics may outnumber Protestants by mid-century
  • The non-religious will increase their share of the white population but not of the total population
  • Secularization will plateau by 2043 and will reverse thereafter.

The 'culture wars' that play such an important role in US politics will be affected. Kaufmann predicts that opinion on abortion is likely to become more pro-life, but that attitudes regarding homosexuality will be stable, reflecting more liberal attitudes among younger cohorts but more conservative attitudes among demographically-growing groups.

Of course, there are a number of assumptions that have to be made to make these kinds of predictions. Will the switching out rates change? Will the composition, never mind the scale, of migration change? I reckon so.

Nevertheless, it seems reasonable that the broad picture will not look so different. There is no impending mass secularization of the USA. Fertility rates and immigration will overwhelm conversions.


  1. Religious people have more children than non-religious. Sure some of their children will lose their faith as they reach adulthood, but only a minority.


    Remember the scene in Idiocracy where the educated couple keeps postponing having kids while the white trash couple has 15? That's the reality of America for you.

  2. It's not the number of children that matters, it's the reproductive success. More educated populations appear to correlate with greater wealth and less religiosity. Education and wealth also appear to correlate with better health and success of the children.

    This is rather obvious from reason. More children mean that each child receives a smaller portion of available resources, including parental attention. If one is sufficiently wealthy, this may not affect overall reproductive success. This implies that the poorer one is, the fewer children one should have so as to reduce division of the slim resources. But this has to be balanced with the increased reproductive success of having more children in the first place.

    It appears, but is complicated to study, that the optimum point for poorer, less-educated tends towards higher number of children. Conversely, for middle-class it appears the optimum point is more towards a few, well-educated children to make best use of available resources. For the wealthy, the optimum size appears to grow again as the resources-per-child at that point have little to no overall effect on reproductive success.

    The question then is which group has the greatest overall reproductive success. It gets even more complicated here because the population diversity becomes important. A more educated, less religious populations produce more useful medical and technological assistance to reproductive success, and it matters if their combined total wealth is redistributed somewhat through socialized medicine or not.

    After all that, my point is that the estimates presented here far oversimplify the issue. If religious populations grow, their reproductive success may drop whereas the more secular, educated populations may succeed more.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS