If you’ve ever taken a class in statistics you learned about the trap of drawing conclusions about cause and effect simply because there is a correlation:
“Correlation does not imply causation” is a phrase used in science and statistics to emphasize that correlation between two variables does not automatically imply that one causes the other (though correlation is necessary for linear causation, and can indicate possible causes or areas for further investigation… in other words, correlation can be a hint).
I suspect that right-wing conservative Michael Barone and his fan Daniel J. Mitchell ( of the Center for Freedom and Prosperity and a fellow at Cato)know that. Yet their rush to see cause and effect in a large overview of demographics – what the Census shows us about population shifts to the south – is built on a foundation of correlation and speculation – According to Census Data, People Vote with their Feet for Less Government
The world is a laboratory and different nations are public policy experiments. Not surprisingly, the evidence from these experiments is that nations with more freedom tend to grow faster and enjoy more prosperity. Nations with big governments, by contrast, are more likely to suffer from stagnation.
The same thing happens inside the United States. The 50 states are experiments, and they generate considerable data showing that small government states enjoy better economic performance. But because migration between states is so easy (whereas migration between nations is more complicated), we also get very good evidence based on people “voting with their feet.” Taxation and jobs are two big factors that drive this process.
Looking at the census data and matching migration data with state tax systems, here’s what Michael Barone wrote. He finds (not that anyone should be surprised) that the absence of a state income tax is correlated with faster growth, which attracts people from high-tax states.
…growth tends to be stronger where taxes are lower. Seven of the nine states that do not levy an income tax grew faster than the national average. The other two, South Dakota and New Hampshire, had the fastest growth in their regions, the Midwest and New England. Altogether, 35 percent of the nation’s total population growth occurred in these nine non-taxing states, which accounted for just 19 percent of total population at the beginning of the decade.
And here’s Diana Furtchtgott-Roth, writing for Realclearmarkets.com. She uses the presence of right-to-work laws (which prohibit union membership as a condition of employment) as a proxy for the degree to which big government and big labor are imposing restrictions on efficient employment markets. Not surprisingly, the states that have a market-friendly approach create more jobs and therefore attract more workers.
The recent release of population trends does indeed say some red states are growing – partly from birth rates and partly from migration from blue or purple states. Republicans were all over this – claiming the only conclusion one could possibly reach is the country is moving hard Right. There is no empirical data to support that conclusion. If Barone or Mitchell want to speculate that population trends favor conservatives and right-wing libertarians, that would be fine if they labeled their speculation as such. There is at the very least a reality based alternative interpretation – that people moving from blue states will turn some marginal red congressional districts blue. Democrats won the 2008 election cycle with a 50 state strategy –
Obama won states in every part of the country. He bridged the red state/blue state divide with a message based on unity and the economy. For the first time since 1996, we have a president that has a true mandate. He won states in every part of the country. He broke through the Republican South by winning Virginia, Florida, and possibly North Carolina. He added Iowa, Indiana, and Ohio in the Midwest. He added the Southwest in New Mexico and Nevada, and the key Mountain state of Colorado.
There seem to be two different reasons why some states flipped for Obama. In Indiana and Ohio, I think the economy powered these voters towards Obama. They were moved directly by the economic crisis and McCain’s seeming inability to handle it. The second key was Hispanic/Latino and young Cuban voters. In Florida, Colorado, New Mexico, and Nevada, these voters played a huge role with their support of Obama. Nationally, Hispanic voters supported Obama by a 2-1 margin. This was a backlash against the GOP’s anti-immigrant stance.
While the economy and Republican rule of the Beltway were fresh in everyone’s mind in 2008 – benefiting Democrats – you cannot swing red and purple states and districts unless there has been both a population shift favoring Democrats as well as a shift in the electorate towards moderates and the center-left. Christopher Beam does some more centered speculation in this article and uses some plain old rational thinking – The new census data may favor Republicans, but long-term demographic trends favor Democrats
The census data released yesterday counted people, but it didn’t say anything about who they are. The figures show that the populations of traditionally Republican states like Texas, Georgia, Arizona, and Utah are growing, while those of Democratic states like New York, Massachusetts, Pennsylvania, and New Jersey are shrinking.
Barone and Mitchell give the impression – hell they emphatically state – they do know who these people are – they’re obviously far Right conservatives fleeing blue states. They present no evidence that is the case.
But trends over the last decade or so suggest the country is becoming bluer. When we talk about population growth in the United States, we’re almost invariably talking about a group that votes Democratic. Political scientist Ruy Teixeira, who co-authored The Emerging Democratic Majority back in 2006, points out that minority voters have grown by 11 percent over the last 20 years while relatively conservative white working-class voters have decreased by 15 points. Emory University’s Alan Abramowitz projects that nonwhite voters will constitute one-third of the electorate in 2020. (By 2042, the entire U.S. population will be more than half minority.) College-educated women, 65 percent of whom supported Obama in 2008, went from 8 percent of the over-25 female population to 28 percent over the last 40 years. Young voters, who went for Obama 66-32, add 4 million new members to their ranks every year. Professionals, 68 percent of whom voted for Obama, are the “fastest-growing occupational group,” according to Teixeira. And, adding insult to injury, the fastest-growing religious population is “unaffiliated” voters, three-quarters of whom voted for Obama.
Does that mean the country will become permanently Democratic? Of course not. Both parties will adjust to accommodate the shifting demographics. But Republicans will be playing catchup.
Those trends are hard-core demographics. Mitch and the “experts” he cites simply roll out a litany of assertions. None of them supported by any data. They see correlations as a facts. Some red states have lower or no state income taxes, ah-ha, that is why the population is shifting. My neighbor got a new TV that must mean he is watching more cable news? Could mean he is watching more DVDs. Barone – using nothing more than his patented right-wing crystal ball says these population trends, which have been in motion for the last twenty years, are due to no income taxes in red states. If that is true those people of whom Barone is reading their minds( anecdotal comments under news articles do not count for obvious statistical reasons) are making some poorly thought out decisions. Taxes by State
Many people planning to retire use the presence or absence of a state income tax as a litmus test for a retirement destination. This is a serious miscalculation since higher sales and property taxes can more than offset the lack of a state income tax. The lack of a state income tax doesn’t necessarily ensure a low total tax burden.
States raise revenue in many ways including sales taxes, excise taxes, license taxes, income taxes, intangible taxes, property taxes, estate taxes and inheritance taxes. Depending on where you live, you may end up paying all of them or just a few.
This is my anecdotal experience. Some years ago I lived in a red state that had income taxes to one that did not. In the first state, with deductions and credits I got all my taxes back. The no income tax state ended up costing more because of higher sales taxes and fees ( like driver license renewal). Both states gave businesses tax breaks that included a tax holiday for years and other economic incentives to get businesses like car manufacturers to open new plants. Who paid for the loss revenue? Ordinary citizens. It was a win win for business and a kinda win for labor. More jobs, but a bigger share of the cost of infrastructure. Taxes are or should be only part of the equation. On average Per Capita Personal Income by State is higher in blue states. So let’s say I’m making $35k a year in a blue state – with all those terrible taxes. Doing the same work in a red state might mean I make $34.3k. With lower taxes I might come out about even, but with the high probability of higher sales taxes, property taxes and fees, probably not. Let’s fellow down the yellow brick road of economic nirvana of Barone, Mitchell and Furtchtgott-Roth’s thinking. Life is better in red states because of the economic incentives. Looking at the macroeconomic picture where is some of that money coming from. The money in real people’s pockets that buy toothpaste, tires and pays the rent. Some of it is coming from blue states – If Democrats are the big spenders, why do Republican states get the money?
It isn’t surprising that the more Republican a state leans, the more likely it is to be furious about government spending. But what is surprising is that states with the highest anti-spending sentiment appear to be the largest beneficiaries of government spending. Not only do red states swallow the lion’s share of government spending, but Richardson found a linear relationship between the extent of GOP support in a state—and, by implication, the fervor of its anti-government sentiment—and the amount of federal largesse the state receives.
Alaska, home to Sarah Palin, and where two fiscally conservative Republican candidates for Senate recently mopped up 75 percent of the vote between them, received $1.64 in federal benefits for every $1 the state contributed to the national kitty. Massachusetts, Richardson found last year, received 82 cents for every dollar it paid into the national pool. No doubt as compensation, liberals in Massachusetts and other “blue” states also received lots of vitriol for being such out-of-control spenders.
The 28 states where George W. Bush won more than 50 percent of the vote in 2004 received an average of $1.32 for every dollar contributed. The 19 states where Bush received less than 50 percent of the vote collected 93 cents on the dollar.
Let’s set aside the hypocrisy about red states getting a ride off blue states for a moment – maybe they genuinely need that money or red states senators and representatives are better at getting more pork. Just in terms of economic incentives – which is the major slant of the Right – why are so any red state residents reliant on federal dollars if they are an economic conservative/libertarian wunderland. There has been plenty of talk over the last 10 years especially about Democrats and how to frame the debate. That might be a like learning to run in waist deep mud since right-wing voters pay attention to the false and empty slogans of the Right, but very little attention to what they actually do. Republicans are pleased as pigs at the trough at government spending as long as they are the beneficiaries. It has been like this since Nixon in the late 1960s. Some counter-factuals to consider – Behind the Population Shift
The rise of the Sunbelt has two common explanations: one climatic and the other commercial. The climatic, obvious explanation is that it’s the weather, stupid. The commercial explanation, which has a proselytizing undertone, is that places like Texas and Nevada attract companies and people with their lower business taxes and fewer regulations.
The first view emphasizes the outdoors; the second right-to-work laws. If all that we knew was that Sunbelt populations were increasing, it would be impossible to distinguish among these and other theories, but we have evidence on wages, productivity and the price of housing that can help us make sense of the Census.
If economic productivity – created by low regulations or anything else – was causing the growth of Texas and Arizona and Georgia, then these places should have high per capita productivity and wages. Yet per capita state product in Arizona in 2009 was $35,300, 16 percent less than the national average. Per capita state products were $36,700 and $42,500 in Georgia and Texas, respectively.
These figures are far below per capita state products in slow-growing places like Connecticut ($58,500), Massachusetts ($50,600) and New York ($50,200). According to the Census Bureau’s 2009 American Community Survey, median family incomes were $56,200, $60,800 and $56,600 in Georgia, Nevada and Texas, but $83,000, $81,000 and $66,900 in Connecticut, Massachusetts and New York.
Low incomes and productivity in the growing states of the Sunbelt strongly suggest that their expansion is not driven by outsized economic success.
Perhaps, sunshine really is behind Sunbelt growth. A superb climate is surely part of the appeal of Silicon Valley and Los Angeles, but what about fast-growing Houston, which has 99 days a year with the temperature above 90 degrees?
The economists’ creed that free lunches are rare does appear to apply to cities as well as stocks. If a place is pleasant, you end up paying for it, especially in the form of higher housing prices. That logic explains why the median sales price for a home in the San Jose area was $630,000 in the third quarter of 2010.
But housing prices in Texas and Georgia are neither high nor rising. The median sales price for a home in greater Houston area is $159,000 and in Atlanta $113,000. The comparable figures for New York and Boston are $470,000 and $367,000, respectively.
Housing in the growth regions is inexpensive, both in absolute terms and relative to those areas’ incomes. People, perhaps unsurprisingly, don’t seem to be putting great value on humid Houston weather.
But those low housing prices actually provide a vital clue about why Arizona, Georgia and Texas are growing.
If housing – the most basic and necessary living expense – seems cheaper or better stated as a consumer you get more for your dollar, that is going to motivate a lot of people to move. Even if that means they might get a small cut in gross income. Here is where the Right has a nugget of truth in their argument, but get the facts or direction of the red state Sunbelt wrong. Housing regulations are less stringent in the south. Professor Glaeser should have mentioned notable exceptions such as Miami ( hurricane and wind storm building standards) where the cost of living nearly matches many blue cities such as San Jose and Boston, but that aside, lack of higher standards in building codes – California has to meet earthquake standards – is driving down housing prices in the Sunbelt. And he should have mentioned that areas like San Jose, Manhattan Island, Boston proper and many other places in the northeast and large western cities reached pretty much reached their new building on open land capacity ( unless you don’t mind a hour long commute). The Sunbelt fellows the supply and demand laws. There is still a lot of open land and land developers willing to build massive suburban mini-cites. If housing prices remain suppressed the current population trends to move south might also lessen. So this might seem like a de-regulation feather in the Right’s cap – even if Barone, Mitchell and Furtchtgott-Roth don’t know what they’re talking about. They and the Professor are ignoring the role of population density on housing prices and the cost of living. Blue states such as New Jersey – which is losing population has 1170 people per square mile, California has a density of 236 people per square mile. While Texas has a population density of 93 people per square mile and red state Georgia has a density of 167 people per square mile. First, these numbers relate to the relatively rural nature of red states versus blue states. That means there is less demand for a relatively abundant resource – land. Fewer people means less demand for infrastructure: Fewer schools, police, fire, side walks, power lines, land easements for the city, fewer and smaller highways, fewer teachers, fewer universities, fewer public institutions like libraries and museums. Should the Sunbelt reach the population density of California or New Jersey it is not outside the realm of possibility those places will need the same infrastructure or want the same amenities of higher population higher density states. This is not a knock against Miami, but as a southern city with population boom issues – largely due to northern “snow birds”, you’re looking at the future of Texas and Georgia. Wrapped up in those number and cost of increased cost of living trends will be the kind of cultural diversity, youthful populace and higher education levels Beam referred to in his article. Something that does not bode well for the red future painted by conservative prognosticators.
This link might also be helpful. As seem already, the Right and libertarians have taken one isolated data point and extrapolated all kinds of nonsense. There are many factors involved in an individual’s tax burden by state, State and Local Tax Burdens: All Years, One State, 1977-2008. And let’s not forget to factor in the most obvious mitigating factor, income. Income is not the end all anymore than state taxes. But if we play the simplistic tax game as the Right has laid out one has to consider if taking a five percent pay cut would compensate someone for getting a lower tax rate.
About Tax Burden Data
People often ask how Tax Foundation rankings of state-local tax burdens compare to Census data, which include two popular state-by-state rankings. One of these popular Census tables covers only state-level taxes (click here to view tables). Local taxes are excluded, such as property taxes and local sales taxes. This exclusion allows Census to report up-to-date state-level collections, which would be impossible if Census waited for the time-consuming tally of tax collections by thousands of local governments. However, some states accomplish at the local level what other states accomplish at the state level, so a degree of comparability is lost as a result. For example, New York’s state sales tax rate is 4 percent, and its counties have local sales tax rates that range from 3 percent to 5.75 percent. Connecticut, on the other had, has a 6 percent state-level sales tax with no local add-ons. In a ranking that includes only state-level taxes, New York appears less taxed than it actually is, and Connecticut appears more taxed.