Centerfield

Thanks for pointing out that article. Food for thought.

A key point is that if cost of living is higher in urban areas, the higher income does not provide better living standards. But progressive taxation is based on income, not income compared to cost of living, so it is an extra burden on urban areas.

This affect should make urban areas sympathetic to a low tax regime. But a more powerful factor favoring government is simply the need to have more rules when people live in proximity. Rugged individualism is simply not appropriate in an apartment building with thin walls.

Tully,

As always, you have done your homework...great share. Rick makes some good points as well...I'm not sure we discussed taxes in the previous post. I wonder if blue states have a higher tax structure...my guess would be that it's entirely possible. Therefore, high incomes are not necessairly indicitave of a higher standard of living.

Rick's point about rugged individualism in an apartment building is also humorous. Point taken, but there's no place like the highway to witness American individualism in its purest form. If that apartment building is in the South or Midwest, where public transportation is not as sophisticated as the Northeast, chances are that the apartment dweller is climing into their Ford Escape or Honda Civic. Put them behind the wheel and those American instincts take over...lol. That's where road rage comes in.

This provides an interesting explanation for why the GOP favors private charity where the democrats favor government programs. Rural Americans are more conviced of the appropriateness and efficacy of charity, helping your neighbor out, and so on. Urban dwellers not so much. Generally, this philosophical difference is thought to be cultural, and to some extent it is, but the cultural schism may be an artifact of the human response to some understanding of scale. And this cuts both ways, for both sides are convinced of the efficacy of their approach, but few consider that in fact, even social approaches need to vary depending on the scale.

I meant to mention progressive taxation as a skew factor, Rick, but got in a rush this morning. You're dead on. It also plays a part in that net payor/payee of government transfers from state-to-state.

Densely populated urban areas have a lesser sense of community, which in turn leads to more reliance on government mandate as compared to voluntary and "peer pressure" behaviors. It's a lot easier to cooperate with three neighbors than 300. I'm not going to try to quantify that, but experience tells me it's true.

Tully,

I don't want to go around in circles on this one, and, before I continue, I want it to be clear that I don't dispute your assertion about population density and voting patterns. In fact, I whole-heartedly agree.

What I'm having trouble understanding is the characterization of false assumptions and the discussion of causation versus correlation. How do those apply to the previous thread? I thought it (at least the original post) was simply about reconciling apparently contradictory statistical analyses.

After going through it, it seemed very similar to the popular-vote-versus-electoral-college-outcome scenario, which is just a simple matter of numbers.

It is a simple matter of numbers. The (apparent) paradox is that while individual higher-income earners tend to vote GOP, states with higher household income levels go Dem. But it's only a "paradox" because of incomplete info not presented in full context. The correlation of income and voting is very weak in the vast area of income where most individual voters fall, $15-$50K, and the given income figures for state are by households, not individuals. It is only at the income extremes that the correlation is strong, but the number of voters in those categories is much smaller.

This is aggravated by a lack of other, stonger, factors in the data presentation (context). Namely, population density and ethnicity (and cost-of-living and household size and incomes per household). The relatively weak correlation between income and voting is simply much smaller than the known correlation of other factors for the bulk of the population. This suggests that the income correlation itself is to some extent an artifact of the other factors.

So the apparent paradox arises from the confusing of aggregate state income levels with individual incomes, and the resulting (implicit and false) assumption that a higher aggregate household income level should imply a higher GOP vote rate. The causal disconnect is in the presentation of household income levels as indicative of individual income levels, when the vast bulk of voters fall in the area where the relationship is weak and other factors are much stronger.

Doe that make sense to you? "Causation" isn't quite the right concept, but the confusion between the ideas of causation and correlation is, as is the the relative strength of the correlations. It only looks like a paradox because of the hidden assumptions and lack of context.

I think I'm smelling your coffee now. Thanks.

if possible go down deeper than the states. here in Pa. for example Phila ( a county) has historically voted Dem (in part because of a legacy of a good political machine), but the surrounding counties (bucks, montgomery,chester, and delaware) have historically higher income levels per household and traditionally skew republican. This pattern holds true throughout the state. Pgh is dem but the surrounding communities are repub. harrisburg is similar.The higher income level does skew repub but in the blue states numbers matter as well.

Okay, the economic mystery has been solved. However, one must also keep in mind that cultural issues are not explained by this same methodology. Otherwise, the areas with the highest population density would also be the ones with the greatest societal controls (e.g., anti-abortion laws and such). To the contrary, cultural issues are based primarily on, get ready for this, the culture, which tends to be much more secular and modern in areas with dense populations and more religious and traditional in sparsely populated areas. This is probably because a heterogenous population such as those that live in big cities and suburbs leads to a greater level of understanding between religions, sexual orientations, etc. Rural areas, OTOH, tend to be homogenous, a trait that leads to the maintenance of traditions and orthodoxies of old. This also explains why states like Vermont with no real urban centers are borderline socialist while urbane states like Texas, complete w/ Houston, Austin, etc, are redder than red.

I took the median income by state and divided it by the cost of living to come up with the "monetary standard of living" by state. The results are in my article at
http://vdare.com/sailer/050508_family.htm .

The reality is that income only moderately correlates with Kerry's share of the vote by state, and when adjusted for cost of living, there's basically no correlation at all.

What actually correlates strongly is what I call "Affordable Family Formation." Bush did well in states where there is lots of land for suburban development, low housing prices, people get married fairly young and stay married, and have more children.

Steve, that's a most impressive analysis and affirms many things I've said for years. It hits home on a personal level as well--we moved away from Denver to red-state Kansas when we decided to start a family, at the same time as half of California and New York seemed to be moving to Colorado to escape the high prices and taxes. Which of course drove up the cost of living in Colorado by leaps and bounds.

I'm going to link your analysis for a new thread, just as I did your article on the Baby Gap last December.