February 20, 2009

Red-State Blue State Rich State Poor State Why Americans Vote the Way They do-Part One Summary

This is a fantastic and powerful work of political science edited Andrew Gelman (with many contributors but I will refer to "Gelman" for convenience rather than to slight his colleagues) Gelman has also (with some of these colleagues) set up an excellent blog that deals with many of the same issues.

.This book I think it's fair to say is both a work designed to expose myths commonly held about US politics by laymen and a collection of a great deal of their own political science work.

This is obviously useful for experts in the field. I also think it or a similar work is essential to have a well thought out view on the US electorate in general. I should add it was written before the 2008 election though most of its conclusions will remain solid. Though having said that Obama seems to have actually won among voters earning more than $200,0000 unlike those earing $100,000-$200,000 (though frankly this may partly be due to sampling)-which cuts against quite a lot of their comments on US voting . I'd like to briefly point out some of the stuff they show so you can see why it's so valuable

Even though poorer states vote more republican ( a fairly well known fact) , at least till this election richer voters in every state voted more Republican (though often narrowly)

Religiosity (in the sense of churchgoing) makes more of a difference t for better off voters-n - i.e. the gap in how they vote, between rich people who go to church and don't is much bigger than those who don't. This tends to be true in western democracies though not universally-including states as different from each as Sweden, France and Israel (obviously Israel it's synagogue going-but I’m using it generically as place of religious worship)

Linked to this and it should to the amazement of authors despite them being experts, when one cross references income and state-the big difference between "Red" and "Blue" states is between rich people in poor states vs. rich states. So in 2004 in Mississippi and Connecticut (the richest and poorest states) the people with the lowest income voted similarly. By contrast the gap among the best off in the two states state was colossal around 40% or so- The result was Bush easily won "poor" (by American standards adjusting for PPP it's income is probably around the same as the United Kingdom) Mississippi and easily lost rich Connecticut (as did Mccain)

Linked to this in poor states the better off you are the more likely you are to attend church in the. IN rich states the opposite is true- the poor are more pious. This makes a lot of sense of the above-It suggests essentially in states in the rural or South "moral" issues if anything increase the divided in voting created by economics while the opposite is true in the North East and Western Coast. So in State's like Connecticut poor people voting Democratic for higher benefits and rich people voting Democratic for legal abortion have the effect of almost cancelling each other out- that is a caricature but one with some truth.

It should be noted that the degree to which religion effects voting is different from which party it benefits more. I personally believe this has tended to help the Republicans more but this book doesn't really comment on that.

They also quote powerful evidence that church attendance actually is a better predictor of voting for the right in a large number of political systems than in the United States. E.g. Sweden, Germany .It's also worth noting that in every state they (briefly) examine, religious attendance is either essentially not a predictor at all or is correlated more right-wing voting behaviour. In contrast in a few countries such as Israel and Ireland being richer actually means you’re less likely to vote for the right.

Another interesting point -clashing with a forest of journalistic articles, America is actually a state where income is a better predictor of voting than most-in France and Germany for example it barely predicts at all.

In the United States. Among religious types in the United States overall Mormons are massively Republican, evangelical Christians fairly Republican (it's important to remember a very large % of evangelicals in the US are black so this is despite that), non-evangelical protestants very close to even. Catholics fairly Democratic and Jews as massively Democratic as Mormons are republican.

. All these groups are more likely to support Republicans if they go to church normally as opposed to virtually never. However the difference in effect is huge. For mainstream Protestants it makes next to no difference while for Catholics the effect is slight though significant (my understanding is it's much bigger for Hispanic Catholics. The effect is however huge for evangelicals- and even larger for Mormons and interestingly Jews.

While Jews vote about 7-2 for Democrats Jews who attend Synagogues every week is about evenly split. To look at it another way while among non attendees non evangelical Protestants are about 5 times as likely to vote republican as Jews there is virtually no difference between Jews and non-evangelical Protestants who go to church every week. This incidentally would seem to suggest problems for explanations of the left-wing behaviour of American Jews that root it in the influence of the old Jewish tradition.

I hope I have shown just a few of the ways this book is valuable. Anyone who has any real interest in the politics of the most powerful nation on earth should find it useful and informative. I will post some more on this invaluable book-including some of my problems with some of their conclusions but it really is a great work.


Vino S said...

I do think the voting behaviour of US voters does differ from the "logical" one of poorer votes voting Democrat and richer ones voting Republican.

The noticeable thing about the former slave states is that - in presidential elections - 70 or 80 percent of white voters vote Republican. This is more than the % of white voters who vote Republican in other states _despite the fact that per capita income for white Americans is lower in the South than nationwide_. I do think that is rather odd - and indicates that lots of poor people are voting on 'social' and 'cultural' issues rather than economic ones.

Sulla said...

Vilno I thoroughly recommend you read the book I think you'll love it.

On your point US politics has huge exceptions to income voting. On them totally agree with you (i think the authors would too but they don't go far enough as hopefully future posts will show) the point is merely that this is not the whole story. However this is true in nearly all western nations- and arguably more in most than the United States-for example your statement is more true in Germany than the US contrary to media wisdom.

Their do show that in the US (unlike many countries e.g Germany) churchgoing is not the explanatory factor it works much better as an explanation of the behaviour of rich voters than poor voters. It’s of course quite possible for poor voters to think republican policies will help them and for rich voters to think Democratic voters- the redistribution differences are not that large and if you think either policies are better for the economy you can see them benefitting you

. And of course income is a poor proxy for even direct income effects- a domestic servant with a rich employer may want high taxes for the rich for straightforward economic reasons, a life insurance worker may strongly support the tax on inheritances which economically benefits that industry enormously.

I do think they underplay cultural factors a bit- after all a) religion and churchgoing are not quuite the same thing b) they are plenty of other cultural factors e.g race and crime which are not obviously linked with religious views even casually. Even issues such as aboriton and homosexuality go well beyond religiosity never mind churchgoing.

White southerners may be an example for this (though its important to note there's a huge class divide among them-much bigger than white northerners so it's middle class whites who behave particularly differently in the South not working class one's )

I'd add middle class African Americans the US over would also be an example of similar types of voting to what you identify in sections of the white working class-it cuts both ways.