Are Conservatives More Generous? [UPDATE]

A while back, we discussed Arthur Brooks' book, Who Really Cares?, in which Brooks claims to have found an increased tendency among "conservatives" to donate money to charities of all sorts and to volunteer.  After offering some preliminary thoughts based on second-hand accounts of the book, I promised to read the book and post some thoughts. I'm working on a full review of the book, and so I don't want to spoil too much. But my read did answer a few of the questions I had based on the news accounts of the book's findings.

First, what do the numbers look like for religious liberals?  Here's what Brooks says on page 50:

    "Religious liberals bear a resemblance to religious conservatives in their giving habits.  They are almost as likely to give (91%), but give away about 10% less money than religious conservatives each year.  On both these measures, they greatly exceed population averages.  They are about as likely to give to secular causes as religious conservatives.  Two-thirds volunteer each year.  They are a bit less likely than religious conservatives to volunteer for religious causes, and a bit more likely to volunteer for nonreligious causes."

Second, what does he say about secular conservatives?

    "Like secular liberals, secular conservatives are not civically involved, and when it comes to giving and volunteering, secular conservatives are the least charitable group of the four. . . . Secular conservatives behave less charitably than others in informal ways.   For example, they are less likely than average to let someone in front of them in a line, to give up a seat on the bus, to give a stranger directions, and to help a homeless person."

Third, does he take into account the differential spatial distribution for liberals and conservatives to account for different costs of living and tax burdens?  Not that I could tell, and this may go some length towards explaining the differences between religious conservatives and religious liberals.  He also rules out any consideration of taxation and government spending as an expression of generosity towards the poor for reasons I find dubious, but that he explains at length.  I'll save this discussion for the book review.

So, the giving habits of religious liberals are pretty much the same as those of religious conservatives (maybe not quite as generous, but the differences are small, and liberals tend to live in more expensive places with higher taxes).  Secular conservatives, on the other hand, are less generous and less likely to volunteer than practically everyone else, though, again, secular liberals give them a run for their money.  It seems like the obvious conclusion to draw here is that religion is doing all the work.  But after acknowledging that religious liberals act pretty much like religious conservatives and that secular conservatives act pretty much like secular liberals when it comes to charitable behavior, and after acknowledging that religious views are by far the strongest predictor he has of charitable behavior, he continues to argue throughout the book as if liberal economic positions CAUSE people to avoid charitable behavior while conservative economic positions CAUSE people to give to charity.  Given the data he has about secular conservatives and religious liberals, this argument makes very little sense to me. 

The book makes some very good points about the importance of charity as a virtue, points that resonate very nicely with Catholic thoughts on the same topic.  But these observations are overshadowed, in my view, by his need to color his analysis primarily in liberal-conservative terms.  I'm not sure whether this way of framing his findings was his choice or the choice of his publisher, Basic Books.  I've looked at some of Brooks's academic writings, and they are far more measured in their tone and conclusions.  I can understand that it sells many more books to say that conservatives are more charitable than it does to say that religious people are more charitable.  But his own data just does not seem to support the point.