Generosity Index v. Income and Poverty

In my recent post, Generosity index: what about those rich, greedy Republicans?, I displayed a table showing states that voted Republican are more generous according to the “generosity index.” I received the following response from Art in the comments.

In politics, the opposing parties are forever trying to paint their opponents as morally lacking as compaired [sic] to themselves. Listen to conservative talk show radio, and liberal talk show radio and you will see this as a central theme. To me, this chart simply shows something that has been long well known: The more money people have, the less (as a percentage) they give to charity. You see the rich states at the bottom and the poorer states at the top.

Is this true? Let’s look at the facts using the same measures of “richness” the U.S. Census Bureau does — income and poverty. Using information found in tables 7 and 8 in “Income, Poverty, and Health Insurance Coverage in the United States: 2003,” I created a table showing the 50 states ranked by generosity index, money income, and percent in poverty. The first two columns are sorted in ascending order and the third in descending. In other words, depending on column, the states listed at the top of the table are more generous, have less income, and more poverty.

In general, the table supports Art’s statement. The states with higher incomes are also the states with the lower generosity indexes.

Interestingly, the blue states also have less poverty, though this is not as clear as the money income statistic. What is clear is that the states with the most poverty are all red and the states with the least poverty are all blue. In between the extremes, the red and blue states appear evenly mixed.

Can we draw any conclusions from the poverty distribution?

Generosity index: what about those rich, greedy Republicans?

A couple of days ago, I responded to Fred Clark’s claim at slacktivist that “[i]t is not possible to endorse the work of charitable agencies [* * *] while simultaneously working to eliminate the estate tax.” The discussion has continued in the comments at slacktivist.

In the comments, I used the “generosity index” to support my position that conservatives are not less generous than liberals. In fact, the generosity index suggests that red states are more generous than blue states.

Beth attempted to explain the apparent discrepancy between red states’ and blue states’ generosity indexes by stating that a small number of rich Republicans giving significantly less than the majority Democrats would lower the generosity index for blue states as a “statistical quirk” of calculating averages. Beth makes a valid point and I’m doing a lousy job of summarizing it. I recommend that you read the comments at slacktivist. I’ll place a link there to bring you back here.

Fortunately, the people at Catalogue for Philanthropy (“CFP,”1 the ones who created the generosity index) already thought of this issue. They calculate the generosity index for each state four times using only tax returns with adjusted gross income (“AGI”) in the following ranges:

  • All returns
  • $75,000 to $100,000
  • $100,000 to $200,000
  • $200,000+

By limiting the data to narrow ranges, CFP is able to avoid the effect of statistical outliers on the averages.

For convenience, I summarized the generosity index rank by adjusted gross income levels and added color coding for whether the state voted “red” or “blue.”

When we eliminate those greedy, fat cat Republicans by limiting our inquiry to those in the $75,000 to $100,000 AGI range, the red states are even more dominant on the top of the generosity index ranking than if we look at the data for all returns. This seems to be the opposite of what Beth expected.

Interestingly enough, if we look at the data for those with AGIs of at least $200,000 (which according to [this site( should be 63% Republican voters, 35% Democrat voters), the red states still rise to the top of the generosity index rankings.

  1. I’d like to thank CFP for posting the information and a phone number for questions. I’d also like to thank Marty Cohn for answering the phone (three times!), directing me to the data already posted on their website, and answering my questions.