17

Republicans argue that those that benefit more from capitalism and want the government to do less to help others would be more likely to give that money away.

ABC's 2020 is one source of this claim. They claim that,

It turns out that this idea that liberals give more…is a myth. Of the top 25 states where people give an above average percent of their income, 24 were red states in the last presidential election.

And, from that they reason that Republicans give more. Is this good reasoning and is it in line with other data-points on this subject?

UPDATE I've refined the question as advised in the comments to be about Republicans and Democrats in the United States: I'd rather the question be broad about conservatives and non-conservatives, but I think it'd be extremely hard to answer it otherwise.

Mad Scientist
  • 43,643
  • 20
  • 173
  • 192
Evan Carroll
  • 28,401
  • 42
  • 129
  • 239
  • 27
    Oooh, I can see it will be tricky to extract the confounding variables here. Can wealthy people afford a greater percentage of their income while also being more likely to be politically conservative? Do the proportion of donations to churches that go towards advancing the religion/church itself count, and if so, are conservatives more likely to belong to such a church? – Oddthinking Mar 13 '13 at 07:45
  • 5
    Without providing a full referenced answer the argument from ABC 2020 is a fallacious, because it mixes people and states. For example, while blue states are on average richer than red states it still turns out that wealthy/rich people tend to lean more republican. See for example Gelman on this. – Erik Mar 13 '13 at 11:55
  • 3
    -1 Your language is argumentative and biased against the claim being true. I've suggested an edit with more neutral tone and I hope it will be accepted. I'll change to a +1 if you fix things. – William Grobman Mar 13 '13 at 15:23
  • @WilliamGrobman I'm confused: is it that I claim the rumor is widespread in the group that benefits from it that makes my claim of skepticism biased? Or, my statement that Capitalism places selective pressures against those that give altruistically, selecting for those that make the higher return on their capital instead. On those grounds, I'd not expect the most ardent supporters of capitalism to abandon the selfish-competition they cherish for humanitarian deeds. When calling out probable propaganda, it's hard not to be biased against the propagandizers. – Evan Carroll Mar 13 '13 at 15:33
  • 1
    @EvanCarroll charitable donations are effectively advertising that happens to also create other good things besides benefits of awareness of the donator. But capitalism is really secondary in the equation, giving people the freedom to spend their money is the real driving force. – Ryathal Mar 13 '13 at 17:12
  • Disagree on your opinions of capitalism, but I'm not referring to giving-for-promotion, or out of selfish desire: that's not my idea of charity. – Evan Carroll Mar 13 '13 at 17:13
  • 4
    -1 for "Contrary to even the most rudimentary understanding of Capitalism" bit. Good question otherwise. – user5341 Mar 13 '13 at 19:02
  • 2
    To get back to the question "Do American Rebublicans or Democrats give more?", it's still ambiguous. Does it mean in total $, or in median $/person, or in median fraction of disposable income? And, what about donation of time or goods? Not to mention relevance - so what if they do? – Mike Dunlavey Mar 14 '13 at 17:44
  • 1
    This question can not be answered in its current form because a person is not required to indicate their political affiliation when making a charitable donation and therefore there is no reliable comprehensive data on this topic. – Paul Mar 16 '13 at 21:45
  • I agree, @EvanCarroll, that's a stupid statement to make about the AIDS virus. The title question here about charity is a [still somewhat vague] question about populations where the most useful data that one would want to directly answer the question are either uncollected or not publicly available. The methodological questions ("Is this good reasoning?") raised in the body in this case are better, answerable questions. I'd suggest revision to asking the methodological issues in the title, e.g. "Can one conclude from state-level data that R's are more charitable than D's?" – Paul Mar 16 '13 at 23:55
  • Why from state-data? Don't restrict the question just leave it as is, let people attempt to answer it and critique their answer. – Evan Carroll Mar 17 '13 at 00:21
  • 2
    Removed a long discussion... According to our [Privileges section](http://skeptics.stackexchange.com/privileges/comment), you should only use comments to request clarification from the author or leave constructive criticism that guides the author in improving this post. Please review the **When shouldn't I comment?** section and act appropriately in the future. – Sklivvz Mar 17 '13 at 01:13
  • 2
    -1 trolling, clearly intended to incite arguments, fighting, and no doubt biased answers (of which op no doubt will pick one that favours his own preconceived "truth"). Suggest this be closed as non-constructive. – jwenting Mar 19 '13 at 14:48
  • 1
    @jwenting that's a bold accusation, all I did was ask a question. – Evan Carroll Mar 19 '13 at 15:05
  • This statement is based on deductions listed on tax returns (where else could you get actual **data**). Wealthier people give, in part, to lower their taxes and all of it appears on tax returns while lower income people rarely list donations on their returns since they don't benefit. – Hannover Fist Sep 29 '15 at 19:43
  • Is this a question about party affiliation, or ideology? – indigochild Jan 26 '16 at 04:03
  • Second clarification based on @Oddthinking's comment - do we need to remove the effects of confounding variables? It seems like this can be answered just by showing a correlation between party affiliation and charitable donations. – indigochild Jan 26 '16 at 04:04

2 Answers2

13

There is at least one study disputing that interpretation:

Voluntary contributions from individuals are the lifeblood of nonprofit organizations, which in turn fund a large portion of social services in the United States. Given this reliance donor generosity, it is important to understand who contributes, and to where. In this paper, we argue against the conventional wisdom that political conservatives are inherently more generous toward private charities than liberals. At the individual level, the large bivariate relationship between giving and conservatism vanishes after adjusting for differences in income and religiosity. At the state level, we find no evidence of a relationship between charitable giving and Republican presidential voteshare. Finally, we show that any remaining differences in giving are an artifact of Republicans' greater propensity to give to religious causes, particularly their own church. Taken together, our results counter the notion that political conservatives compensate for their opposition to governmental intervention by supporting private charities.

However, note that this adjusts for "religiosity". So what they seem to be saying is that conservatives/Republicans give more because they are more religious, not because they are Republican or more conservative. So in other words, they found the same relationship as Brooks but counted it differently. Whether you call this insightful analysis or sophistry is probably going to be heavily influenced by your own views.

The link in the question is citing the work of Arthur C. Brooks, a professor at Syracuse University, who wrote Who Really Cares: The Surprising Truth About Compassionate Conservatism. This article by conservative columnist George Will cites some of his findings, including:

-- Although liberal families' incomes average 6 percent higher than those of conservative families, conservative-headed households give, on average, 30 percent more to charity than the average liberal-headed household ($1,600 per year vs. $1,227).

-- Conservatives also donate more time and give more blood.

-- People who reject the idea that "government has a responsibility to reduce income inequality" give an average of four times more than people who accept that proposition.

Note that the statements in the article mix findings from examining state level data with results from polls. Presumably Brooks' book gives more detailed citations.

One reason why I included these is that they directly dispute the assertion that conservatives give more as a percentage of income because they are poorer. As the first factoid shows, conservative households give more in raw dollar terms even though they have lower incomes. As the second factoid shows, conservatives give more in non-monetary resources. Note that this doesn't dispute that these things are more true because of religion than political partisanship. But "religiosity" itself is predictable from political partisanship.

Another source. Again, they find that religious people give more and conservatives/Republicans are more religious than liberals/Democrats.

TLDR: The statistic is real but the interpretation can be contested. Religious people give more. Conservatives/Republicans are more religious. Conservatives/Republicans give more (because they are more religious). All statistically true by every source that checked. Subject to different interpretations.

Community
  • 1
Brythan
  • 10,162
  • 5
  • 46
  • 53
  • 1
    Interesting. Anybody know what Brooks' actual sources are? I'm guessing he got the "Liberals are Richer than Conservatives" bit by taking the data from [This Pew Study](http://www.people-press.org/2005/05/10/beyond-red-vs-blue/), cherry-picking out only the 17% marked "Liberal" from the Demo portion, and then comparing it to the entire Republican base (29%). If you instead compare with their "staunch conservatives", the income levels are similar. If you compare the two bases, the Republican one is richer. Either of those are more like-for-like comparisons. – T.E.D. Jun 03 '16 at 14:43
2

No, its not good reasoning. I believe we don't really know on a partisan basis who gives more. That article's claim is an ... well ... interesting way of dealing with the statistics it has.

The only recent studies on the demographics of charitable giving I'm aware of for the USA's general finding is that poorer people tend to donate a much higher percentage of their income. Income and % of income donated are practically an inverse relation.

That means any way you can find to statistically group yourself with a poorer group is going to make you look more generous.

"Red" states do tend to have a lower income than "Blue" states, which means when you break things down on a statewide basis, yes Red states are more generous. For example, peruse through the NCCS stats, and you'll find my Oklahoma shows the average giving at about $5k per taxpayer, vs. only a little over $3K per Massachusetts taxpayer. However, due to the difference in population Mass. still gave more than 3x what OK did.

However, the political affiliation of people within a state are not "winner take all" like this article (and perhaps the Electoral college) tries to imply. Even the most partisan states tend to have the other party garner 38-ish% in national elections. To take Oklahoma again, it was about the worst state for team blue, but they still got 32% of the vote in 2012. Demographic breakdowns show that the Democratic vote share is higher the lower you go down the income scale.

So a Democrat could point to this and make the exact opposite claim: "Since poorer people give more than rich people, and poorer people are also much more Democratic, clearly Democrats give more". This of course suffers from the same problem. Some Republicans are poor too. There are legit reasons why they might be more apt to donate (eg: Evangelical churches tend to demand tithing).

T.E.D.
  • 1,588
  • 1
  • 11
  • 16
  • 1
    Does your answer boil down to "So without a study on this exact topic, I'm not sure how you could say"? If so, I'd recommend just deleting the answer and letting this question continue to go unanswered. Leaving questions unanswered when we don't have the data yet is a reasonable skeptical position. –  Jun 02 '16 at 18:49
  • 1
    @Dawn - The actual question-marked sentence is "Is this good reasoning and is it in line with other data-points on this subject?" and the answer boils down to "No." – T.E.D. Jun 02 '16 at 19:03
  • Then, I'd suggest making it clear that that is your answer and removing the "I'm not sure how you could say". –  Jun 02 '16 at 19:10
  • @Dawn - How's this? – T.E.D. Jun 02 '16 at 19:11
  • Wait, you say again "I believe we don't really know on a partisan basis who gives more". But more important is not what you believe, but rather, what experts in this field have shown to be the case. Can you make your answer more focused on the evidence? If we really just don't know, you should simply outline what the best evidence is, point to some experts who agree that it is inconclusive, and leave the speculation (jumping to conclusions) out. All the stuff about how poor people donate compared to rich people isn't really relevant to this question. –  Jun 02 '16 at 19:11
  • @Dawn - Makes sense. How about this? – T.E.D. Jun 02 '16 at 19:16
  • You say a lot of things without references: "We don't currently have data specifically on party affiliation and giving", "it seems like good data on this would be tough to achieve", "a Democrat could point to this and make the exact opposite claim", "There are legit reasons why they might be more apt to donate", etc. I would remove all unreferenced lines of argument. –  Jun 02 '16 at 19:56
  • @Dawn - Toasted that paragraph too. I guess it isn't central. I'd hate to get rid of the one before it (new last para) though, because I think there's a real danger of someone reading this doing just that. I'm arguing for using what the data gives you, not cherry-picking whichever supports your dogma best. But technically, I suppose it could go too... – T.E.D. Jun 02 '16 at 20:11