Field of Science

Religion causes wealth inequalities

Back in 2008, I wondered whether the reason that the religious give more to charity than the non-religious might be that they get more out of it. They get an extra benefit from charity (an anticipated reward from their God) that atheists don't. When atheists want to redistribute money, they would prefer to do it via taxation (because that minimises the risk of free-riders - people who don't contribute their fair share).

Well it turns out that Ceyhun Elgin, at Bogazici University in Istanbul, and colleagues have been thinking along the same lines. They were wondering whether it might explain why religious countries tend to have higher income inequality than non-religious countries. Their analysis has just been released - you can download the working paper here.

The first thing they did was to develop a mathematical model, which showed that the idea was plausible. Agents in their model get more satisfaction from charity if they are religious, and so (they found) prefer a lower tax rate. That, in turn, means lower levels of income redistribution. It's all complex stuff and I have to confess that I haven't taken the time to try to understand it, but I'll take their word that this is what they've shown!

Then they to some multiple-regression analyses of real-world data. They found that, after controlling for GDP and also for type of religion (Protestant/Catholic/Muslim), countries with high levels of belief in the afterlife also have high levels of income inequality, low levels of tax, and low levels of government spending.

All this tallies well with what other research has shown. In my own analyses, I found that high levels of prayer are associated with low levels of welfare payments for the unemployed, even after controlling for a range of factors.

Of course, there are other reasons why religion might be linked to income inequality. There's good evidence that religious people aren't as anxious about losing their jobs, because God (and their Church) will provide.

More importantly, high levels of income inequality are linked to all sorts of societal problems, as well as high levels of anxiety (for rich and poor, although mostly for the poor). That could increase religious beliefs. There's good theoretical reasons to think this, and also some evidence from studies comparing countries (including my own).

Elgin suggests that religious countries might not be as unequal as the statistics show. That's because charitable benefits are often not counted as income. However, given that charitable donations are dwarfed by government welfare (even in low welfare countries like the USA), I don't think this is likely.

What's more, it's not true that the religious prefer small government. There is one area of government spending that they are in favour of - and that's defence spending! So the effect of religion on government welfare spending may be even more dramatic than Elgin's figures suggest.

Creative Commons License This article by Tom Rees was first published on Epiphenom. It is licensed under Creative Commons.


  1. You write "cause" in the headline, but does the study really show that?

    If you read DS Wilsons "Darwin's Cathedral" and Rodney Starks books on the rise of Christianity, you get the impression that early Christianity was rather a reaction to the otherwise miserable society, and that it created a miniature well-fare society that worked better than the roman society. So the correlation between dysfunctional societies and religiosity could be due to religion as a well-fare organization when a state is not possible or less likely to happen.

    Then again, there's a lot of diversity in what function religion has in a society, so this cannot be generalized just any way.

    If the church is the well-fare state for it's members, then the real state would be a competitor against something more resilient and durable, and also much closer to the members. The Amish would be an extreme of this. Then it wouldn't be strange that the religious disliked big government while religion did not directly causing dysfunctional societies. The question is really at what level the community is reliably working: clan, village, church, state?

    Then again, it's odd that they support military defense, which means they rely on national/state level rather than church. Maybe it has to do with a mental division of external threats and internal problems.

  2. Tom seems to like provocative titles for his blog posts :) The post itself appears to me to be sufficiently hedged, even if the title is not.

    The mathematical model they built is the thing that interests me the most. I never really quite thought of it that way, but it does make a lot of sense. Joacim came quite close to summing it up nicely (though I question why you would assert that church membership would be more "resilient and durable" than the state!) by characterizing the welfare state as a direct competitor to the charitable aspects of religion. With my biases, I see it the opposite way -- that the welfare state option is likely to be more resilient and durable, hence why churches would be especially concerned about it. In any case, it's food for thought.

    It was incidental to the post, but I can't help but reflecting on the now well-established finding that income equality -- regardless of absolute income -- increases anxiety, crime, and a whole host of societal woes. This was a counterintuitive result to me when I first found out about it, and it helped shape my political opinions. Though already fairly liberal, I was sympathetic to the idea that "a rising tide raises all ships" -- that it would be better to be poor in a rich country with a high standard of living than to be average in a poor country with a low standard of living. Perhaps this is not as true as we would guess...

  3. Well, I've done a post on inequality causing religion, and one on "Religion causes inequality (or is it the other way around?)" - so I thouhgt it was time to complete the circle :)

    Seriously, I think that the effect is mostly in the direction of inequality => more religion. But in the course of researching that topic, I found a lot of evidence that the opposite can happen.

    It can happen because the Church is a kind of parallel state for members. It can happen because religion reduces the anxiety that would otherwise be felt by people in insecure situations (so they don't see the need for a protective state). And it can happen for the reasons proposed in this latest study.

    Which makes me think it likely that it's a feedback loop.

    Now that's interesting because feedback loops create a situation where you can have multiple "stable states" with rapid transitions between them. So a high inequality, high religion state like the US could be stable and resist changes in religiosity and inequality - until small changes reach a tipping point.

    Maybe that's what happened in Europe after the war. The introduction of highly socialised welfare states as a result of the wartime experience forced societies over some kind of social cliff.

    Just speculation, but interesting to think about.

    James: have you seen the book "The Spirit Level". I can't remember what they called it in the US. The website is here:

  4. Nope, hadn't heard of it. I added it to my future reading list, though honestly it will probably be a while before I get to it... I often find that sort of thing horribly depressing. heh...

  5. James: I called churches more resilient and durable because I read some study that showed that religious communities often last longer than secular communities (be they hippie or socialist or whatever). I think it looked at 19th century US communities. Perhaps Tom can tell if this is correct. Now, the state is not really comparable to a community, it's much too loose. Still, it can mean so much for the well-fare level of a society, like in Sweden were I live.

    Also, I have too agree with Tom that there's probably a feed-back mechanism here. Inequality -> churches -> resists well-fare state -> inequality. In this aspect, religious support and resistance against the health care bill would be interesting to study, as well as attitudes towards inequalities within the church.

    I'm starting to suspect that the best way to reduce religion's influence is acutally through social democratic policies rather than humanism. To speculate, I think most people care about a secure environment, and they'll adapt in any way to those that can provide it.

    Btw Tom, have you done any review on Rodney Starks books on Christian history? His latest is Cities of God from 2006. They're really fascinating since he uses statistical testing on historical data. I'm intrigued by his arguments about how Christianity offered a way of not having to obey the laws for hellenized Jews, while still being Jewish. It's like they fall into an stable point, reducing psychological and social tension among themselves and the surrounding society.

    You can probably get pretty far with the analogy of culture as a multidimensional forcefield with stable energy wells and such.

  6. I'd rather claim „religion allows inequality“. But before I elaborate on this, please James can you provide some evidence for „the now well-established finding that income equality -- regardless of absolute income -- increases anxiety, crime, and a whole host of societal woes.“

  7. Paul N, The best recent book is Spirit Level - it just so happens the NYT just published a commentary on it.

    I made a modest contribution in a paper published back in 2009, showing (among other things) that income inequality correlates with a lot of other indicators of societal problems. There's a summary in an earlier blog post.

  8. I well understand your point, and this is something I will comment on. Maybe it's a typo only, but James claimes the opposite talking about income equality not income inequality.


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