I’m reprinting this here because in the ten years I’ve been studying and researching this political dysfunction, it only gets worse. And I rarely read or hear any rational analysis – only one-sided arguments. The media has really misguided America because they write narratives and don’t study hard data.
Red-faced or Blue-blooded: Exploding the Myths of American Party Politics
In recent years American politics has become highly polarized, making democratic governance less amenable to compromise and more gridlocked. After a generation of conflict and heightened partisanship during the Obama presidency, as we careen from budget battles to periodic government shutdowns, we seem no closer to bridging the gap. One reason for this impasse stems from a misunderstanding of our politics driven by a popular media narrative that perpetuates cultural stereotypes, political myths, and partisan hyperbole.
The narrative appears to have emerged during the 2000 and 2004 presidential elections, when the media colored the states red and blue in their visual props in order to better represent the Electoral College races. This was actually instructive because there was an obvious clustering of red states and blue states: blue in the Northeast, along the coasts and the Great Lakes regions; red in the south and throughout most of the Midwest. The media explanations for such clustering focused on exit poll survey data that was based on voter identities and preferences, including race, ethnicity, gender, age, church attendance, etc. And therein lies the problem: analysis based on identity is going to come up with identity-based answers, yet the very fact that the pattern is geographic (red and blue states) means that a spatial factor has to be at work.
Urban vs. Rural
Of course, media pundits had a ready answer for the geographic state pattern: red states in the South represented a racial bias while those in the Midwest demonstrated a fundamentalist Christian bias. Thenceforth, the narrative of party polarization in American politics settled on the cultural and personal attributes of voters, claiming that party affiliations and voting patterns were being driven by race, gender, ethnicity, religious belief, and lifestyle preferences, such as the type of car one drove or the music one listened to. Closer inspection of the hard data, however, exposes the fallacy of this narrative.
If one looks at a more accurate map that breaks down voting patterns into smaller units, of counties or Congressional districts, a more nuanced geographic pattern reveals itself. Blue voters are shown to be concentrated in urban areas, while red voters dominate rural areas. The suburbs represent the swing voters or the purplish middle ground, with inner suburbs voting more blue and outer exurbs more red.
The advantage of county voting data is that we can compare it with county census data on race, ethnicity, age, gender, income, household formation, population, etc. Comparative statistical analysis then shows which of these factors most accurately explains voting patterns. What we find is that the most significant factors are the population density of the community and the number of married households vs. female heads of household. All other factors turn out to be relatively insignificant, including, surprisingly, race.
One might ask, “How can that be? We know that there’s a racial divide where blacks predominantly vote Democratic and whites vote Republican.” (Actually, that’s only half true; whites are much more evenly split.) Certainly race had an impact on the 2008 election, but the novelty of voting for the first non-white president was less of a motivating factor in 2012. Exit polls do show that black voters vote Democratic by overwhelming margins, but the fact that they live in urban areas and have a high correlation with Female Heads of Household (.8 in the 2000 census) means that these two other factors trump their racial identity.
If we take black voters out of the analysis completely, and thus remove any racial bias, we find the same results are even stronger among the non-black population: urbanites mostly vote blue, rural residents mostly red, while suburbanites are mixed. Married households lean red and single households lean blue.
There is a self-selection process that reinforces these results because single voters tend to live in large cities and married couples choose to move out of the city into suburban and rural communities in order to raise families. These two factors—population density and family formation—help explain the majority of these red-blue voting patterns. (We’ll address a third factor when we get to ideology and religion.)
Rural–urban splits in American politics are nothing new and are driven by a natural divergence of economic interests. The first regional party divide was between Hamiltonian Federalists (cities in the North and East) and Jeffersonian Democrats (farming interests across the South and near West). The next battle, over the Second National Bank of the U.S, pitted Jacksonian Democrats against the Whigs. Then, of course, we had the Civil War between North and South over slavery and tariffs. This was followed by another split over banking as McKinley Republicans fought against the Bryan Democrats over gold and silver-backed money. (The 1896 Electoral College map looks almost identical to that of 2004, but with reversed colors: the South and Midwest favored the Democrats, while the Northeast and coasts went Republican.)
Parties and Media
This begs the next question: “Why do certain regions now vote Democrat or Republican so consistently instead of mixing their ideological preferences?” The fixed pattern is an outgrowth of the two party platforms that were set in the 1950s and 60s, when Democrats began to appeal to urban voters with Great Society social programs and identity-group politics, while Republicans targeted rural and suburban voters with lower taxes and family-oriented policies. Consider the fact that the South was once solidly Democrat and is now solidly Republican. The agrarian South has been consistently traditional and conservative; it’s the parties that have flipped, not the voters. The parties have organized their platforms and campaigns to appeal to constituencies based on geography. And by pursuing these electoral strategies, the parties have become ideologically more pure, reinforcing the natural policy divide.
Recent Red-Blue patterns have become hardened for two additional reasons that stem from party and media incentives. First, both parties benefit from a polarization myth based on identity. Why? Because if voters identify strongly as either Republican or Democrat, liberal or conservative, their votes are virtually guaranteed and party candidates don’t need to spend campaign funds trying to win them over. This is why presidential candidates don’t bother to campaign in solidly red or blue states, but focus all their energies on swing states. (More pernicious is the fact that national Democrat candidates can pay lip service to the minority vote and suffer no real consequence, while Republicans have chosen to essentially write off minority voters.)
The second reason that Red-Blue patterns have come to be reinforced lies with the media. Mainstream media is centered in metropolitan areas and must appeal to these audiences or go out of business. Their reporters live in the same urban communities where they work, so their worldview is largely colored according to urban interests. These facts do not necessarily imply a deliberate political bias, but since mainstream media’s audiences have been heavily politicized (liberals read the big city papers and watch the three major urban networks), its coverage of politics has tended to exhibit a strong leftward, or urban, bias. As a consequence, alternative media has expanded to meet the preferences of non-urban audiences with a rightward bias. These media biases reinforce polarization in a vicious feedback cycle as people only tune in to the news that confirms their political views. For these two reasons—party and media incentives—we can’t expect that political polarization will be reduced by either of these institutions.
We need to understand that voting patterns are correlated with lifestyles because geography is also associated with lifestyle choices. But lifestyle choices do not determine political identities. This is an important insight that many of our national politicians consistently misread. We saw it when Barack Obama claimed that small town folks “cling to guns or religion” because they are “bitter” about their economic “frustrations.” On the Republican side, we witnessed a similar gaffe when Mitt Romney claimed that the 47% of Americans who “don’t pay income taxes…[and] were dependent on government” would always vote for the party promising them more benefits paid for by taxes on someone else. Both of these statements are gross caricatures of our political culture.
Religion and Ideology
This brings us to the last issue that colors our political divide: the thorny question of religious faith and practice. The proxy for religious faith, which is difficult to measure, is frequency of church attendance. Residents of red states go to church more frequently than residents of blue states. Thus, churchgoers are generally thought to represent the ideological Religious Right that votes Republican. This view concludes that religious faith is politically opposed to secularism.
This is not quite accurate, however, and contributes to yet another misreading of our politics and religion. Churches provide different functions within rural and urban communities. In rural areas the church is a communal meeting place where people gather for social purposes as well as to observe their faith. In an urban secular environment, this social need is more likely to be met by reading the Sunday paper at Starbucks.
What frequent church attendance does do is create an organizational structure for political messaging, wherein large numbers of voters can be reached efficiently and effectively. In this sense, churches have played the same role for the Right as industrial unions have played for the Left. Republican political strategists have taken full advantage of the growing evangelical movement as Democrats have suffered the decline of unionization. It is a mistake for the secular Left to confound religious attendance with conservative orthodoxy and attack political opponents for their faith instead of their politics. Religious belief is highly pluralistic in America, meaning there are many faiths and denominations that disagree on almost all aspects of religious doctrine and political preference. Attacking people of religious faith merely unites them in defense of that faith, even when they may not agree on much else.
The most salient split for our politics lies in the ideological differences between orthodox and unorthodox belief. This applies not only to religious belief, but to the secular world as well. Another way to put this is that in our society we have secular fundamentalism as well as religious fundamentalism, and they both influence our ideological preferences. Fundamentalism is defined as strict adherence to orthodox doctrines. We might categorize fundamentalists politically as one-issue voters. So, on the Right, we have cultural conservatives, evangelicals, pro-life, creationist, pro-family groups, while on the Left, we have environmentalists, liberation theologians, pro-choice feminists, and same-sex marriage advocates. Moderates of all stripes are those who occupy the middle ground of our politics. Politically, the focus on ideology becomes too complex to really draw any hard conclusions. For instance, where are the conservationists? Do they side with environmentalists or with traditional naturalists? And where are the libertarians?
The ideological divide has often been characterized as a split between traditionalists and modernists, but I don’t think such a classification fits American politics very well. Instead, I would suggest a different label to describe the majority of Americans who are neither conservative, liberal, nor radical.; They are “tolerant traditionalists,” rooted in the past but willing to embrace change at their own pace. They are bound by faith but amenable to reason. They are not really polarized on the big issues, but they do have different political and policy preferences based on geography, family formation, and ideology. These differences need to be negotiated through democratic politics with the goal of reaching acceptable compromises and creating a more coherent political society—much more cohesive and functional than we have witnessed in recent years.
The real picture: Between 2000 and 2012 not much has changed, only gotten worse…
2004 Election by Congressional District
Maps are by Robert J. Vanderbei and can be found here.