Heritability of religion and fertility

The United States is one of the few developed countries in the world with a fertility rate close to the replacement rate – that is, the rate of fertility required to maintain existing population levels. The two reasons most often cited for this is are high levels of fertility in the Hispanic immigrant population and the high level of fertility of religious people. Even when you control for income and education, religious people have more children than non-religious people (on average).

The higher fertility among religious people raises a couple of questions. As religion is heritable (that is, the predisposition to be religious and not the specific religion itself), will religion spread through society and what will the consequences of this be? What will happen to the underlying alleles (alleles are the different variants of a particular gene)?

If we had a situation where the religious have higher fertility and a religious genotype is determinative, it is clear that they will eventually form the largest group in the population and the religiosity allele will dominate the gene pool. However, what if there are more complicated dynamics such as defections between groups, with the religiosity allele(s) giving a predisposition as opposed to being determinative?

Robert Rowthorn addressed this question in a paper published earlier this year in which he explored the dynamic consequences of heritable religiosity and the higher fertility of religious people. Razib at Gene Expression wrote a great post on this paper when it was first released, but as is often the case, writing a blog post myself is the best way to get my head around it.

Rowthorn noted that a natural result of defections from a high-fertility religious group is that the religious group would be smaller than it would otherwise be. However, he also points out that defection from the religious group allows the religious allele(s) to spread to and within the non-religious group. If this religiosity allele affects the number of people from the non-religious group who become religious and as a result, boosts their fertility, the religious allele may come to dominate the population regardless.

To explore this issue, Rowthorn constructed two models (one haploid, in which predisposition is determined by a single gene, one diploid, in which predisposition is determined by two genes, one from each parent). In these models, culture (religion) determined fertility, while religious predisposition has a genetic component. On becoming an adult, one can defect from their religious or non-religious group, with their underlying genotype giving the probability of switching.

Some of the outcomes of the model are unsurprising. Firstly, the pace of evolution is heavily dependent on the fertility differential between religious and non-religious types. With a differential of two or three to one, large changes in population structure occur within five to 10 generations. Where the fertility differential is 1.3 to one, it might take several hundred generations for the share of the religious gene in the population to dominate from a low base.

With defection added, there are two main results. First, the speed of the spread of the religiosity allele is reduced, although it still spreads through the population. Second, the total number of religious people in the population is reduced, possibly significantly. Defection changes the expression of religiosity, but not the eventual spread of the gene.

For high fertility sects such as the Amish, with fertility rates two to three times above the national average, this would imply that they would come to dominate the population in five to ten generations, unless there is a particularly high level of defection. If they do have a high defection rate, while reducing the size of the Amish population, the religious gene will still spread. Even with a 50 per cent defection rate, the religious gene still spread to most of the population within 20 generations in Rowthorn’s model. From this, Rowthorn notes that while secularisation might reduce the growth of high-fertility sects, the importing of the religious gene into the non-religious population results in the religious allele ultimately spreading through the entire population.

Rowthorn does not venture far into the implications of his findings beyond the spread of religion. In the last paragraph, he makes the following observation:

It is interesting to speculate how such a predisposition might manifest itself in a secular context. The findings of Koenig & Bouchard suggest that a genetic predisposition towards religion is associated with obedience to authority and conservatism. If this is correct, then the diffusion of religiosity genes into the rest of society should see an increase in the number of secular people who are genetically inclined towards such values. The implications of such a development are beyond the scope of this paper to consider.

Is the growing conservatism among youth, which seems to be a common media topic in recent years, an early reflection this effect?

Rowthorn does not specifically explore the question of overall population size. If the religious allele(s) spread through society, fertility would be expected to be higher and overall population higher. This could have significant implications for the debate on population size I have posted about in the last week (such as here, here and here).

Further, Rowthorn’s findings show the general result that any genetically based predisposition that increases fertility can be expected to spread through the population. Whether that predisposition is religion, dislike of contraception, urge to have a larger family or some other trait, the effect is the same. The alleles responsible for higher fertility spread and, barring further environmental shocks, population growth increases. The only question is how long this will take.

*As an extension to this idea, I have a working paper that examines the consequences of heritability of fertility more generally.

Rowthorn, R. (2011). Religion, fertility and genes: a dual inheritance model Proceedings of the Royal Society B: Biological Sciences DOI: 10.1098/rspb.2010.2504


  1. Is the growing conservatism among youth, … an early reflection this effect?

    What “growing conservatism” other than that existing in the minds of right-wing pundits. Voters in the 18 to 24 age group broke 68 percent for Obama to 30 percent for John McCain. Reagan had much more support among youths than Republicans do today.

    1. I’m open as to whether the evidence supports growing conservatism, but a pairwise comparison between Reagan-Bush and McCain-Palin does not give us a useful perspective on how conservative youth are. The growing conservatism that tends to be reported on includes reduced levels of binge drinking, changed attitudes to sex, warmer attitudes to religion, increased intention to marry and have children, and reduced rates of teen pregnancy.

    2. You still haven’t cited any scientific sources for your “growing conservatism” assumption which still appears to only exist as wishful thinking among right-wing pundits. Some of your “conservative” examples are telling because they are framed in a way that presumes Liberals are not in favor of say reducing teen-pregnancy or binge drinking. It is hard to call these conservative values rather than sensible values unless they are confused with particular conservative moralistic approaches towards solving these problems.

      As to “warmer attitudes to religion” whatever that means I find it hard to reconcile with the steady marked decline of religiosity in the US.

      1. I’m not talking about conservatism on a political spectrum (nor was Rowthorn on my reading of the paper), but conservatism of thought and action.

        As for religion, this was the survey in the back of my mind when I wrote the original post (this survey is a few years old now – the link to the original report does not work). However, I’m open as to the precise direction of the overall trend at the moment as there are a range of conflating factors (such as immigration) and I expect that it is only the surveys with interesting results that make the news – that is why I added the qualifier “a common media topic” (or at least, I intended it to be a qualifier). However, one thing I am reasonably confident of is that the bible belt, with its high fertility rate, is not exactly a liberal production line.

Comments welcome

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s