I'm a big Paul Krugman fan, and have been for many years, but I was quite unimpressed by his column today. He intended to rebut Obama's claim about religion in his famous "bitter" comment, but he ended up rebutting a series of claims about religion that Obama didn't actually make. Obama claimed that people in the Midwest have become bitter about the lack of jobs, and that this has led them cling to religion.
To rebut this claim, Krugman pointed out that poor people in Montana and Maine (two poor non-Midwestern states) go to church less than people in Connecticut. Okay, but that has nothing to do with the relationship between religion and economic circumstances in the Midwest. Obama didn't claim that people outside the Midwest weren't as religious as Midwesterners, so citing data from non-Midwestern states doesn't make any sense. Krugman also noted that while people in poor states go to church more than people in rich states, "this result largely reflects the fact that southern states are both church-going and poor." Again, this has nothing to do with whether people in Midwestern states have reacted to their worsening economic circumstances by clinging to religion.
Moreover, the graphs that Krugman provided to support his arguments only showed the relationship between church attendance and income for 2006. You have to look at the data on a time-series to see if worsening economic circumstances have led Midwesterners to cling to religion. Looking at the church attendance rates for just 2006 won't do any good, because you don't know whether the Midwest church attendance rates in 2006 are higher or lower than they were during good economic times.
To that end, I very quickly pulled up a couple of graphs from the National Opinion Research Center website, using data from the annual General Social Survey. These are very, very rough indicators, and I'm using them to make any positive claim; they do, however, paint a slightly different picture than Krugman did. I couldn't actually find a graph of median household income in the Midwest because the GSS database on the NORC website is absolutely awful, so I'll just use the one Krugman posted on his blog:
Next, here's the percentage of people in Midwestern states who attended church "every week":
As you can see, the percentage of regular church-goers in the Midwest isn't a perfect fit with Krugman's graph, but it does very roughly track median household incomes.
Finally, this graph shows the percentage of Midwesterners who said on the survey that their financial situation had been "getting worse" during the "last few years." This is probably a much better indicator of "bitterness" than household income, because people who have been poor for a long time are less likely to be bitter about their economic circumstances than people who have just recently become poor.
The percentage of people who feel their financial situation has been "getting worse" tracks the percentage of regular church-goers more closely than does the median household income. Also, notice the huge spikes in both regular church attendance and worsening financial situations around 2003. I don't really know what to make of that, but it's interesting.
Again, I'm not saying Obama is right that Midwesterners cling to religion because of their worsening economic circumstances. I'm just saying that it's probably not as cut-and-dry as Krugman seemed to imply. In fact, minus the part about religion, Harvard economist Benjamin Friedman made essentially the same argument as Obama in a meticulously researched book, The Moral Consequences of Economic Growth. Friedman broadly argued that during periods when real incomes across the board have been rising (e.g., the late '90s), people are significantly more open on social issues like immigration, affirmative action, trade, etc. So while I don't think Obama was 100% right (especially about religion), I'm willing to cut him some slack on this one.