David Cameron's call to 'stand up' for Christian values has led to the sort of fact-free posturing that religious debate usually provides. It confirms my prejudice that when someone says 'I (don't) believe in God', the emphasis is entirely on the 'I'.

But there is an alternative reaction. We can ask: what are the practical, social effects of religion?

It is, of course, trite to say that one can be 'good' without being religious. As Ken Binmore, among others, has shown, moral codes - reciprocal altruism - can emerge from natural selection, without invoking God.

Nevertheless, it is a legitimate question of social science to ask if the behaviour of the religious differs, on average, from that of the non-religious.

And there is evidence for this. A laboratory experiment has found that Christians are more likely than atheists to trust other Christians, and to justify this trust by behaving more trustworthily - though other research suggests such an effect might be confined to more liberal denominations. This needn't mean that Christians are inherently morally better than us; it might be an example of stereotype threat, the tendency of people to live up or down to their stereotype.

But here's a thing. We know that trust is good for an economy, because it helps overcome problems of incomplete contracts and markets for lemons; as Rabbi Jonathan Sacks said recently:

Try running a bank, a business or an economy in the absence of confidence and trust and you will know it can't be done.

Which raises the question: could it be that religious attitudes promote economic growth?

Yes - to some degree. A paper (pdf) by Luigi Giuso, Paola Sapienza and Luigi Zingales concludes:

We found on average that religion is good for the development of attitudes that are conducive to economic growth'
On average, Christian religions are more positively associated with attitudes that are conducive to economic growth, while Islam is negatively associated.

Not just Christian, though. Some economists have found that, among Buddhist Tibetan herders, religiosity is associated (pdf) with higher individual incomes.

It does not follow, however, that religion's effects are wholly benign. Religiosity is strongly correlated (pdf) with gender inequality,and Giuso's paper found that (on average) 'religious people are more intolerant.'

All of which raises a paradox. From the point of view of this liberal atheist, the practical effect of religious belief is the exact opposite of what people, including Cameron, suppose it to be. Whilst its moral effects are, I think, to be regretted in many ways, its economic effects are more benign.