Redstate

then taxes should continue to drop everywhere :-)

that correlation does not equal causation. But in this case, even the correlation is suspect. Put dark blue, high-tax Rhode Island's growth (Census 1990-2000) in the context of some red/purple, low-tax states, and the idea that people are running from blue states to red states because of lower taxes seems less convincing.

North Dakota: 0.5%

Alabama: 1.0%

New Hampshire: 1.1%

South Carolina: 1.5%

Tennessee: 1.7%

Rhode Island: 4.5%

We can only cheer this news that some Blue States are realizing that in order to raise tax revenues, you must cut tax rates.

This isn't a criticism of your diary, which explicitly only addresses state taxes, but I did want to pedantically note that it's an entirely different thing to "get up and leave" at the national level, and I believe the general argument then is that people/businesses won't try to earn money if it is overtaxed by the feds. Also, what do you suppose the relationship is between state tax rate and test score average?

It matters who's moving and why.  "Growth rate", if that means population growth rate, only helps if the right people are moving.  Poor people, for example, are more likely to move from New Hampshire to Massachusetts, rich ones the other way, because of welfare systems.  Poor people really don't care about taxation, especially income tax.  They don't pay any.  And, considering the urbanization of Rhode Island, it seems likely they're attracting a lot of low-skilled aliens as well, who put extra burdens on welfare services.  Sure Boston is more culturally exciting than Providence, or southern New Hampshire, but you can live in New Hampshire, if you have money, and get to downtown Boston in an hour, on a good day.  Portsmouth's a nice town.  Comparative per capita income growth, after taxes, might be an interesting stat.

Didn't see anything about people fleeing to red states, more like moving from higher tax blue states to slightly lower tax blue states.

However, it makes sense that if blue states raise taxes high enough, they will loose people to red states, assuming the reds maintain low taxes (shouldn't the commie states be red?).

Different parts of the country. Weather plays a pretty big role in where people want to live. I can see why ND isn't high on the list with the undesirable weather, no large metropolitan area, no mountains, and no ocean. Doesn't exactly have a lot going for it. It's a really cheap place to live, though.

Asf6 makes a helpful point:

"It's already been pointed out to you that correlation does not equal causation. But in this case, even the correlation is suspect."

We certainly agree that the matter is debatable. That the sky turns blue after sunrise is not evidence that sunrise causes the sky to turn blue. We concede the point. However, we also believe a causal relationship is plausible. As regards correlation and causation, we have in this diary offered three academic studies which suggest that the correlation is in fact causal. We offer the links here:

Do Taxes Affect Migration?

Do Local Fiscal Policies Affect the Migration of Human Capital?

Do State Taxes Affect the Migration of Human Capital?

These three studies all suggest that there is a measurable correlation between migration and fiscal policies at the local and state level, and all three answer their titular questions in the affirmative.

That there are exceptions (growing, lower tax Blue States, and shrinking, higher tax Red States) is also affirmed here. Generally speaking, however, Redness correlates to a lower state-level tax rate, and Blueness correlates to a higher state-level tax rate. We affirm the opposite correlation as regards abortion rates.

In any case, as food for thought, consider this table showing highest and lowest State and Local Tax Burden in Major U.S. Cities. The top 5 are Blue. The bottom 5 are Red.

That many of the middle 40 might be purplish, is a truism.

RedShift

your civil response. Let's agree the matter is debatable. I'm not ruling it out. I'd just like to see some more research first; right now I'm inclined to believe it has more to do with folks leaving the Rust Belt for the Sun Belt, and not much else.

Brendanm98 observes,

"This isn't a criticism of your diary, which explicitly only addresses state taxes, but I did want to pedantically note that it's an entirely different thing to "get up and leave" at the national level, and I believe the general argument then is that people/businesses won't try to earn money if it is overtaxed by the feds."

We certainly concur with this. In Daniel Henninger's excellent article on this point, Blue State Pols are Emptying Their Own States, he observed that the US is unique in this very sense:

"In the world's stalest states, such as Germany or Japan, people faced with cost-of-living waters rising to choking levels turn numb and go nowhere. But here in the U.S. moving on is a tradition, and today we have Web sites to reveal a suitable refuge from state political cultures intent on keeping the spending and tax spigot open."

Certainly, beyond the state level, if federal taxes punish productivity too severely, people merely stop performing the punished activity-- namely, productivity.

Brendanm98 then introduces a valid question about the possible cause/effect relationship of standardized test scores and interstate migration patterns. The question is indeed intriguing, especially since this study, while acknowleding the significance of tax-based interstate migration patterns, also noted the significance of education funding in those patterns:

"The findings suggest that high-skilled labor is more mobile, and it is characterized by higher migration rates toward states with lower income-tax and sales-tax rates. At the same time, a higher share of government spending on education appears to attract higher rates of migration."

Brendanm98's question therefore pertains:

"Also, what do you suppose the relationship is between state tax rate and test score average?"

This an interesting question. If we take average ACT scores by state, the average scores show a dramatic correlation to a state's color, with Blue States crowding the top of the field, and Red States occupying the low end. Providing top and bottom ten, with average composite scores:

Top ten:              Bottom Ten (excluding DC):

Connecticut   22.8    Arkansas       20.3

Massachusetts 22.8    Alabama        20.2

Washington    22.7    Texas          20.2

Vermont       22.6    Colorado       20.2

Oregon        22.6    North Carolina 20.2

Maine         22.4    Georgia        20.0

New York      22.4    New Mexico     20.0

New Hampshire 22.3    Louisianna     19.8

Minnesota     22.3    South Carolina 19.4

Wisconsin     22.2    Mississippi    18.7

We concede a positive correlation based on test scores, with Blue states being higher. But before we send the Red States off to special ed, there is an additional piece of information that casts doubt on this particular correlation, and that is the student participation rate. We note that the higher a state's participation, the lower the average composite score. (i.e., Connecticut's participation rate was 10% and Mississippi's participation rate was 94%). Ordering states according to participation rate results in the exact opposite correlation:

Top ten:              Bottom Ten (excluding DC):

Illinois 100          Oregon 12

Colorado 100          Massachussetts 12

Mississippi 94        Maryland 12

Tennessee 92          New Hampshire 10

Louisanna 85          Maine 10

North Dakota 82       Connecticut 10

Alabama 77            Pennsylvania 9

South Dakota 76       Rhode Island 8

Nebraska 76           New Jersey 6

Arkansas 76           Delaware 4

Again, there is a strong correlation between the color of the state and the participation rate, with Red states being higher.

Although it is entirely possible that Connecticut's 10% were the least intelligent students in the state, and that the lack of participation is not a factor in Connecticut's #1 ranking, we doubt it. Unless there is an equalizing factor in the ranking system to compensate for lack of student participation, we have no inclination to take the average test score rankings at face value. After all, Illinois, a Blue State, had 100% participation, and an overall ranking of 40th in average test score. Because of this, we strongly suspect that if other Blue states had such a high participation rate, their average test scores would almost certainly drop as less talented students are added to the test pool.

Barring the introduction of additional and convincing evidence, there do not appear to be any hard fast conclusions to be drawn from the 2005 state-by-state average test scores. And if Red State tax rates tend to be lower, and Blue state tax rates tend to be higher, nothing so far suggests that there is a positive correlation between state tax rates and the average intelligence of graduating seniors by state.

RedShift

This discussion seems to miss one of the major economic stories of recent years, which is the massive increase in housing prices in major Blue State cities.  Bubble or not, there is huge demand to live in precisely these cities.  And since these cities tend to be compact & constrained by geography, the market responds not by increasing housing quantity, but by raising prices.  This forces lots of people who would like to stay in Blue states to leave, and surely has a bigger affect than state income taxes

As Yogi Berra once said, "Nobody goes there anymore; it's too crowded."  

So while the population growth of geographically small Blue states won't necessarily keep up with the fastest-growing cheap-land Red states, I can't see how those states could actually shrink.  

 

Regarding participation rates, you're quite right that this biases the results. The ACT isn't used much out East; looking at SAT results (PDF) shows pretty much the inverse of ACT results.

I suppose the proper thing to do would be to compute p*x+(1-p)*X for each state, where p=participation rate, x=score, and X=participation-weighted average score for all states. Sounds like a project for a rainy day...

The SAT is preferred on the coasts.

The ACT is preferred in between, but often accepted instead of the SAT.

SAT scores are meaningless nationwide, but ACT scores are pretty accurate. MS has high SAT scores, but low ACT scores because only students headed out-of-state take the SAT. SC has low SAT scores and low ACT scores because state schools accept both tests.

While they both "do" the same thing, they are completely different tests. The ACT tests what you have learned, the SAT tests how well you can think.