Andrew Sullivan and Kevin Drum both link to this Pew Report on various worldwide opinions. Here is the graph that gets people talking, a plot of per capita GDP versus religiosity:
This looks like a curve that was drawn by hand, rather than fit by least-squares, but there is obviously a correlation: as a country gets wealthier, it gets less religious. The United States, obviously, is a whopping outlier. Why is that? What is it about the U.S. that makes it so different from our demographic cousins, even within the Anglosphere? (Kuwait is also an outlier, but the reasons are pretty straightforward.) I’ve heard various theories, but none has really been convincing.
(Looking closely, maybe a better fit to the data would be to horizontal line segments: one at 2.25, for GDP between 0 an 10,000, and one at 0.75, for all higher incomes. Perhaps there is a phase transition that countries undergo when their per capita GDP hits around 10,000. Or, even more likely, there is some hidden third variable that is highly correlated with both GDP and religiosity. That kind of curve would make the U.S. seem less exceptional.)
I’m sure if you split the US 50/50 down the line after sorting for religiosity. I’m sure the most religious 50% would have a lower GDP than the least religious by a long shot. The U.S. is a big and varied and multicultural place compared to most countries.
“but there is obviously a correlation: as a country gets wealthier, it gets less religious”
The second part of that statement could be read as implying the link is causal rather than a correlation. Who knows? I could just as easily see a country getting wealthier as it gets less religious.
The “phase transition” that I’d think of first is fertility. The US will deviate in terms of that variable as well. I keep seeing the Darwinian argument that secularists are so reproductively unprolific that they’ll eventually be swamped by religionists, especially if the secularists encourage the religionists to migrate into secularist regions (e.g. from a poorer country to a richer one).
The more complicated evolutionary reasoning may be spurious, but I don’t think the fertility relationship is.
I’d like to see another axis added for population density. I suspect the U.S. is comparatively sparsely-populated, compared to other wealthy nations. Perhaps that is a factor?
Of course, Mexico City comes to mind, and I’m not sure whether India and China count as wealthy nations. Perhaps it’s a combination of wealth and population density? The idea being: the denser your population is, the more likely you are to encounter (and befriend!) people with other beliefs, thus potentially eroding the idea that your religion makes you a fundamentally different sort of person.
Using GDP as an indicator of wealth is not the best idea.
http://en.wikipedia.org/wiki/Gross_domestic_product#Criticisms_and_limitations
I could get hit by a car tomorrow and do more to raise the GDP than I would in a whole year in my relatively low paying job.
Having emigrated from the UK to America, one reason for the US being an outlier is that it is a much more politically conservative nation than most of Western Europe. I don’t know if it’s cause or effect, but conservative nations do tend to be more religious.
Whatever the reasons, surveys are beginning to show that the US may well come back into line with the rest of the wealthy nations within a few decades regarding religiosity. The younger generations are markedly less religious than the baby boomers and pre-boomers, and trending further that way.
We’ve gone from 5% of a generation claiming to be non-religious to around 20% in 40 years. Given that most people tend not to change their beliefs once they have reached their 20s, this will naturally make America less religious as the balance of the population changes.
What beliefs the next generation holds is anyone’s guess, but I would be very surprised to see this trend towards the secular reversing. It certainly hasn’t happened anywhere in Western Europe.
But rather than worry about the individual beliefs of Americans, what I am hoping is that this trend marks the beginning of the end for the religious right’s ability to shape public policy. With the current backlash against the Republican party, we may have already seen the high-water mark in the past six years of the Bush administration. When they are finally able to recover from their recent setbacks, they may find their base is an awful lot smaller than it was before. I, for one, will not be terribly disappointed.
I’d like to see this graph done per country, plotting “religiosity” (or however one quantifies it) against annual income.
I believe there’s a pretty strong negative correlation between per capita income and fertility, so if you normalized to either, I’m guessing the fit would improve.
I’m guessing that there would be a correlating graph for GDP and education, that as the general education of the population rises, the GDP rises, with a blip to one side for the US, indicating that it’s easier to be bone-ignorant in America, yet still make a good living.
Not to imply that religious people are bone-ignorant, or anything.
Beren:
I know one counterexample: Sweden is extremely sparsely populated too, even more than the US: 20 people per km^2 compared to the US’ 31. It is also the least religious country in Europe.
If one divides the sample into Europe and ‘all others’ one could make the case that
the social safety net in most European countries explains the reduced need for religion.
Well, I can’t remember his exact quote, but paraphrasing James Dobson of Focus on the Family (and other crazy xTian political zealotry): Attacks on American Christianity are attacks on capitalism and our way of life!
It is interesting that the graph shows the US somewhat higher in per capita GDP than either the IMF or even the CIA. Norway, Finland, Ireland, and tiny but principled Luxembourg have higher GDP/capita. Ireland therefore makes another interesting case re: religiosity versus over capitalism.
Here’s one reason per capita GDP is not a good way to study the correlation between wealth and religiosity. The same reason applies to average income or any measure of wealth where you total up some amount of wealth and divide by the number of people.
Say you have a country with one multii-zillionaire and everyone else dirt poor. If poor people tend to be religious, this country will have a lot of religious people. But, its average income could be high, thanks to that one multi-zillionaire.
In short: even if the proposed correlation between religiosity and low income is true, this way of looking for it won’t work for countries that have a high variation in income – a few people
much richer than the rest.
And, of course, the US has a higher variation in income than most European countries. Maybe that’s why it looks exceptional in this graph. I can’t tell.
So, this graph is based on a crappy methodology – unless the hypothesis we’re interested in is really that high per capita GDP, per se, makes people less religious. (Maybe one multi-zillionaire on the block is enough to make all the poor folk stop believing in god. Who knows? But this seems a bit far-fetched.)
I seems to me that the graph does not account for the vast scope of religion. The concept of religion itself is as varried as the number of religions. This graph treats religion as a uniform concept, while even a mere student of religion will quickly learn that not all religions are moral equivalent.
So I guess to finish my point, could the US be such an outlier as a result of the primary religious belief being various forms of Christianity? Whereas Africa is much more varried in religious beliefs including Islam, Christianity, and a host of other less mainstream beliefs. Could it be that this graph is really displaying the differences in religion? Perhaps some religions are beneficial to the development of society, and others are not?
So where are China & Russia in that graph?
Seems that atheism or lack of religion per se, hasn’t made them all that ‘wealthy’ in the past century.
I would say having faith in their own currency, adopting ‘capitalist’ economics – being willing & able to hold vast sums of dollar debt from the US – and the velocity of ‘money’ have contributed to the Chinese economic miracle.
But then again, in Capitalism money is god.
Ironically per capita or per head, means that some of the poorest and most densely populated parts of earth have the potential for the highest GDP. Imagine if and when average incomes in India, China, Brazil, Mexico (and Africa) reach US & EU levels – their GDP will grow rapidly. Whereas their ‘inherited’ wealth (infrastructure & property) still lags a little behind.
Life is a series of repetitive rituals (and consummerism):
get up and have breakfast, a cup of coffee or tea (and brush your teeth?)
get up and have the three esses – a sh*t, a shower, and a shave?
get up and read the daily papers, and/or learn something ‘new’
get up and travel to work, and/or take the kids to school on the way?
get up and make money (pray for more paying customers to visit your shop?)
get up and get drunk, and hope to have some fun – or at least the ilusion
or simply get up, and enjoy another day.
So which daily rituals float your boat?
But hey, it’s all about whatever gets you through the day.
So Enjoy and “have a nice day”
China and Russia are probably not good data points as their governments actively crack down on religion (or used to, in the case of the former Soviet Union). There was a rebound in religious observance all over the ex-communist countries when the yoke of communism was lifted, showing that even though they were supposed to be atheist, many people never abandoned the faith of their fathers.
No doubt the decades-long crackdown has had some lasting effect, and these countries are never likely to be as religious as they once were, but they are probably not useful as comparisons with what’s happening in the US.
In the same way, some Islamic countries, like Saudi Arabia, will skew the data the other way since it is illegal for a citizen not to be a practicing Muslim.
But where, realistically, the average outnumber the super-rich by thousands, if not millions, to one, can a gilded-age economy (i.e. extreme economic polarization) skew per capita GDP all that much?
Sorry, I should have said mode rather than average.
excise the states that lost the Civil War and you have your answer.
the divides that existed in American society then are precisely reflected in that plot. If those states had seceded and the coutries were truly and formally divided, the “Confederate States of America” would be back amongst the yellow dots (they would have had, and would still have, a difficult time living without the northern states), whereas the US (rest of it) would be with western Europe.
The US has high median per capita GDP as well, so the billionare outlier isn’t that strong of an argument either. Also Ireland seems to be up there also on the GDP/Capita and they are fairly religious (or so I might have thought)
I’d venture to guess its just historical coincidence, America just so happens to have been a resource rich, fairly isolated area free of war that highly religious people colonized not so long ago on historical timescales.
Also we are far less religious today than we were even 200 years ago, so perhaps its just that the system hasn’t had time to relax yet.
I’m not particularly sure if the statement makes sense even at the individual level though. We could look for correlations on the Forbes 100 list for religiosity for instance, but Im not sure if we would see anything there -shrug-
Perhaps a clearer way to put it:
GDP = consumption + investment + govt. spending + (exports – imports).
Kind of a complicated measure compared to something like personal income or wealth.
Now, it’s true that wealth is distributed incredibly unevenly in the US (top 1% have something like a third of the wealth, whereas in the UK it’s more like a fourth or a fifth, and in Scandanavia, there’s even less inequality). It’s easy enough to see how such a distribution is going to separate, say, median income from the mean. But how well does unequal wealth distribution correlate to per capita GDP? My naïve comparison of per capita GDPs of European nations to the US would lead me to guess perhaps not much. What’s the true relationship btw. per capita GDP and measures of personal wealth or income?
I see some partial answers have come in since I started typing.
Sean, I am surprised no one has blogged about this interesting paper by Psaltis et al
0710.4564 . Maybe you could blog about it.
Thanks
I think that the suggested issue that the US is really made up of “two Americas” with a secular, wealthy upper class and a vast swarm of poor people who are very religious is misguided. Despite wealth inequality, median incomes in the US aren’t very low and consistent with those of other well-developed nations and often higher. That seems to undermine the idea that it wealth inequality is the driving force.
I think the more likely explanation would be found with the difference in the social function of religion in the US v. Europe. In many parts of the US, religious institutions also form the major social institutions, serve as centers for networking and thereby wealth creation. It is likely, then, that a great many more people to the north of the median income are religious, or profess to be religious, than in other countries because of the social and economic benefits of doing so. At the same time, however, aggregate church attendance (both in- and outside zones where religion serves as the prime social force) is fairly low while Americans frequently lie about their church attendance, inflating it dramatically. That indicates that appearing religious probably serves a social function, one deeply ingrained enough that Americans will lie about it to strangers.
That’s my answer: America isn’t really that much more religious, it just pays to say you are in the US, skewing the numbers.
No, I disagree. America has a much, much larger minority of deeply religious people than in Western Europe. For example, less than 10% of the UK attends church frequently, whereas its around 50% in the US, of which about half are characterized as evangelicals / fundamentalist / bible-believers.
While I am sure there is some lying going on, it’s not to the degree that would completely invalidate these survey results — there are many more well attended churches here than in the UK, for example.
Whether or not these religious people live according to the dictates of their faith is, of course, in doubt, but there is no doubting that a much larger proportion of the US population is seriously committed to their beliefs than in Western Europe.