[I posted this at Crooked Timber a day or two ago and got quite a lively response].
While the aesthetic defence of religion offered by Terry Eagleton might appeal to a small fraction of the intelligentsia, a far more common belief is that, regardless of truth value, religious belief makes people better citizens, and should therefore be encouraged.
Although this claim has various components, the most obvious social benefits of religious belief, and the biggest source of concern about the adverse consequences of unbelief, is the doctrine of an afterlife in which good actions will be rewarded and bad ones punished. Back in the 19th century, lots of people were really worried about this and, even in the 21st it’s a common theme in US discussions of religion.
But do we really need religion for this?
It seems to me that, if I were a social planner designing a religion for a modern capitalist society (any variant), I’d go for a vague belief in karma, embodied in a semi-comprehensible slogan like “what goes around, comes around”. The general idea is that, if you behave in ways that are bad, bad things will happen to you (note the air of paradox in the title of the bestseller “When Bad Things happen to Good People).
Compared to a full-scale religion, there are a bunch of benefits. First, there’s no need for a costly set of professionals to elaborate and expound the idea. Relatedly, there’s little or no room for theological disputes that can turn into social divisions.
Third, although it might seem problematic that there is no definition of “bad” here, that’s actually an asset. Full-scale religions tend to prohibit things like usury, or being gay or working on the Sabbath, and it often turns out that these prohibitions become obsolete or maybe were never useful in the first place, and that new lists of good and bad behavior are needed.
Finally, there’s a big gain in plausibility. Admittedly, some people appear to behave badly and live well, but it’s easy enough to believe that they are deeply deprived in some unobservable way, or that they would have lived even better if they had done better. At any rate, this seems more plausible than an elaborate afterlife with a gradation of eternal punishments and rewards
Of course, there’s an obvious problem with bias here. I’m an agnostic with a vague belief in karma, so it’s maybe not surprising that I would conclude that my beliefs are socially optimal. Ekelund, Hebert and Tollison, economists who appear to be mainline Protestants, have a whole book showing how mainline Protestantism is more economically efficient than Catholicism or (IIRC) fundamentalism.