ABSTRACT: The widely believed claim that many voters in American elections are voting against their economic interests (“lower income Republicans versus affluent Democrats”) in favor of their social or cultural values is not supportable by the data concerning class voting patterns. American voters are polarized on both a class and cultural basis. Economic polarization takes place on a national level, and cuts across regional and local boundaries, with rich Americans overwhelmingly voting for the Republicans and poor Americans leaning strongly towards the Democrats. Cultural polarization represents intra-class conflict within the middle class, primarily the upper middle class, with affluent people in wealthier states voting for the Democrats and persons with a comparable class position in the poorer states voting Republican. Furthermore, the “red-state/blue-state” electoral map represents conflict not between states per se as much as conflict between ideologically polarized Congressional districts, local communities, counties and neighborhoods.

In recent years a stereotype has emerged in American politics. The picture

presented by much of the media is one of lower income persons voting Republican and

upper income persons voting Democratic. In other words, many people have started

voting against their own economic interests in favor of their cultural values, with upper

income, urban, educated, cosmopolitan elites voting for liberal social policies, and lower

income, rural, religious voters favoring conservative policies. This image is often

depicted on electoral maps as the “red state/blue state” divide with the socially

conservative red state poor and working class pitted against affluent but socially liberal

residents of the blue states.  This picture is widely accepted, but is it true? Is it an

accurate depiction of the class and cultural divisions among voters? The evidence

indicates that it is not. The available data shows that the voting patterns of the poor are

reliably Democratic. Instead, the red state/blue state divide is symptomatic of cultural

conflict among middle to upper-middle income persons, and of intra-class conflict

among the affluent or wealthy.

A leading and perhaps most well-known proponent of the “poor conservatives

versus rich liberals” thesis is Thomas Frank, who outlined his views in the popularized

work What’s the Matter with Kansas?: How Conservatives Won the Heart of America.

Frank provides a straightforward summary of his views:

If you earn over $300,000 a year, you owe a great deal to this derangement.

Raise a glass sometime to those indigent High Plains Republicans as you contemplate your good fortune: It is thanks to their self-denying votes that

you are no longer burdened by the estate tax, or troublesome labor unions,

or meddling banking regulators. Thanks to the allegiance of these sons and daughters of toil, you have escaped what your affluent forebears used to call “confiscatory” income tax levels. It is thanks to them that you were able to

buy two Rolexes this year instead of one and get that Segway with the special gold trim. (Frank, 2004, p. 2)

According to Frank, Republicans have been able to successfully appeal to the social

conservatism of blue collar workers and the rural poor on cultural controversies like

abortion, gay rights, immigration, the role of religion in public life, gun control and

affirmative action. Frank sees this as a “bait and switch” tactic on the part of the

Republican Party, whereby working class voters are pushed to vote according to their

cultural values, and are then given economic policies that are harmful to their own

interests. Frank describes what he regards as the consequences of this arrangement:

Vote to stop abortion; receive a rollback in capital gains taxes. Vote to

make our country strong again; receive deindustrialization. Vote to screw

those politically correct college professors; receive electricity deregulation.

Vote to get government off our backs; receive conglomeration and monopoly everywhere from media to meatpacking. Vote to stand tall against terrorists; receive Social Security privatization. Vote to strike a blow against elitism;

receive a social order in which wealth is more concentrated than ever before

in our lifetimes, in which workers have been stripped of power and CEOs are rewarded in a manner beyond imagining.(Frank, 2004, p. 7)

Liberals who agree with Frank’s analysis will argue that working class Republican voters

are under the grip of what the Marxists call “false consciousness,” meaning such voters

are distracted by what the Left would consider to be religious superstition, irrational

prejudices like racism or homophobia or conservative economic propaganda generated by

corporate-funded think tanks and media outlets. Allegedly, such distractions prevent

working people from perceiving and voting for their rational economic self-interest.

Even some conservatives will agree with Frank’s general thesis, but from a polar

opposite perspective. These conservatives will argue working class Republicans really do

perceive their economic interests accurately, and that it is perfectly legitimate for workers

to desire tax cuts in order to increase their take-home pay and deregulatory policies that

ostensibly accelerate economic growth and therefore job creation and rising living

standards. (Gelman, Park, Shor, Bafumi, Cortina, 2008, p. 16) An even more extreme

argument is offered by the neoconservative commentator David Brooks, who suggests

that because the red state/blue divide appears to be driven more by cultural and social

issues than by class or economic ones, that perhaps the idea of “class,” which he derides

as “Marxist” in nature, is not applicable to American society at all.  Brooks sees

Americans divided on the basis of cliques rather than classes, with these cliques being

comparable to the various teenage subcultures one might find at a high school, such as

“nerds, jocks, punks, bikers, techies, druggies, God Squadders,” etc. (Brooks, 2001)

The methodology utilized by commentators like Frank and Brooks is

problematical. Frank relies very heavily on anecdotal evidence gathered from his

experiences with Republican-leaning, working-class Kansas communities of the kind that

he grew up around. He provides examples like a friend’s father, a man with liberal

economic views but whose Catholic religious beliefs led him to the pro-life Republicans. (Frank, 2001, p. 4) Much of Frank’s work includes sweeping political, cultural and historical analysis with very little in raw statistical data provided as supporting evidence.  Likewise, many of Brooks’ arguments are anecdotal in nature, relying on his personal experiences of living in an upper class liberal community and his ventures into conservative working class towns and conversing with the locals.

What Does the Data Show?

The most comprehensive and up to date analysis of the available data concerning

voting patterns in relation to class position, income, occupation and cultural background

is provided by Andrew Gelman, David Park, Boris Shar, Joseph Bafumi and Jeronimo

Cortina. This group of scholars published their research in 2008 under the title Red State,

Blue State, Rich State, Poor State: Why Americans Vote The Way They Do. Contra Frank,

these researchers found that the image of “working class conservatives versus affluent

liberals” is a false one, arguing instead that “lower-income Americans don’t, in general,

vote Republican-and, where they do, richer voters go Republican even more so.” With

regards to Kansas, for instance, that particular state has leaned Republican by ten percent

greater than the national average for sixty years, and the real source of Republican

strength in Kansas is the middle to upper classes. (Gelman, 2008, pp. 14-15)

Political scientist Larry Bartels argues that it is only in the South that the trend of

whites without college education voting Republican has emerged.(Bartels, 2006) Even

so, Gelman, Park, et.al. found that in the 2004 presidential election the “poor vote” went

to Democratic candidate John Kerry in all of the Southern states except Texas!(Gelman, 2008) Bartels maintains that there is no identifiable pattern of white working class voters

favoring cultural issues over economic ones. Jeffrey Stonecash argues that “the last 40

years shows a growing class division in American politics, with less affluent whites more

supportive of Democrats now than 20-30 years ago. Indeed, even in Kansas less affluent

legislative districts are much more supportive of Democrats than affluent

districts.”(Stonecrash, 2005)

The evidence indicates that the rich are overwhelmingly Republican in their

voting preferences. Republican candidate George W. Bush only won thirty-six percent of

the vote from those earning less than $15,000 annually in the 2004 election. Among those

earning over $200,00 Bush obtained sixty-two percent of the vote. (Gelman, 2008, p. 9)

As mentioned, Bush’s home state of Texas was the only southern state where Bush won

the “poor people” vote in the 2004 election. Yet even in Texas there was a significant

class division in voting patterns. In Zavala County, the poorest Texas locality, Bush won

twenty-five percent of the vote. However, in the wealthiest Texas community, Collin

County, Bush won seventy-one percent of the vote. The capital city of Austin is located

in Travis County, where the mean income of $45,000 is solidly middle class, and where

Bush received fifty-three percent of the vote. (Gelman, 2008, p. 12)

Voting patterns indicate that poor voters are overwhelmingly Democratic, as are

racial minorities. This is not to say that there are no significant cultural differences

among the poor. After all, “the poor” can include everything from rural Alabama whites

who belong to the Ku Klux Klan to black street gang members in the inner city areas. Yet

there is no evidence that such differences play significant roles in American electoral

politics. Many poor people do not vote at all. Those who do are, by a wide margin,

consistently Democratic-leaning.  The growing gap between socio-economic groups

that has escalated over the past thirty years has been widely documented, but this

growing divide between rich and poor is not the source of the red state/blue state divide.

The evidence supports the conclusion that the red state/blue state divide has its

roots in cultural conflict within middle to upper-middle income groups. As Gelman

summarizes:

There is still a rich-poor divide in voting, in popular perceptions of the

Democrats and Republicans, and in the parties’ economic policies. But

voting patterns have been changing, and the red-blue map captures some of

this. The economic battles have not gone away, but they intersect with cultural issues in a new way. In low-income states such as Mississippi and Alabama, richer people were far more likely to vote (Republican)…But in richer states

such as New York and California, income is not a strong predictor of individual votes. (Gelman, 2008, p. 17)

In the poor states, the pattern of wealthy people voting Republican and poor people

voting Democratic is very reliable. In states where the mean income is more in the

middle, the pattern begins to blur somewhat, and in the wealthiest states, income is not a

determining factor in voting patterns. While the middle to upper classes in wealthier

states are just as likely to favor the Democrats as poor people, the same socio-economic

groups in the poor states are more likely to favor the Republicans. To break it down

further on a regional basis, Democrats only win the “rich vote” in the most liberal

states. For instance, in the 2004 election the Democrats won the vote of those with an

income of over $200,000 annually in only four states: California, Connecticut,

Massachusetts, and New York. Middle class support for the Democratic Party is the

strongest in the Northeast, parts of the upper Midwest/Great Lakes region, and on the

West Coast.  To break it down to the level of local communities, affluent to wealthy

urban people tend to lean towards the Democrats, even though the majority of affluent

people are Republicans. The wealthiest states are also those which are the most

urbanized. (Gelman, 2008, p. 19-20)

A key question that arises from these observations concerns the matter of why

voting patterns are more divided on the basis of income in poor states. These patterns are

relatively new. For instance, in the 1976 presidential election, the Democrat Jimmy

Carter won the South, and the Republican Gerald Ford won California, New Jersey and

parts of New England. In the 1976 election, the level of correlation between the wealth of

a state and partisan sympathies was relatively small. Why do affluent people in poor

states hold such greater differences in their political allegiances than poor people when

compared to affluent people in wealthier states? Gelman and associates offer four

primary explanations:

1. Division between races is the most evident in poor states in the South. This racial division overlaps with a class division. Because of the relationship between race and class position, economic policies such as social welfare programs that involve transfer payments from rich or affluent persons to the poor are seen as race-based entitlements for African-Americans.

2. Wealthier people in the poor states attend church more regularly or frequently than poor people, and are also more likely to belong to conservative religious denominations than persons with comparable levels of wealth in richer states.

3. Geography and history. The wealthier states have a much larger number of unionized workers, more large cities, and stronger immigrant communities, thereby creating a more liberal political and cultural atmosphere in these states. A direct correlation exists between cosmopolitanism and Democratic voting patterns.

4. Middle to upper income persons have greater freedom and ability to choose where they will live and whom they will associate with. For instance, affluent persons with liberal social or cultural views tend to migrate towards urban enclaves such as Portland, Seattle, Madison, Minneapolis, San Francisco or Montgomery County, Maryland where such views are most prevalent. (Gelman, 2008, p. 22)

Political polarization in the United States occurs on two levels, the economic and the

cultural. A divide exists not only between rich and poor, but between affluent Americans

holding different cultural values.  Analysts differ as to the causes of this polarization.

Political scientists Nolan McCarty, Keith Poole, and Howard Rosenthal attempt to

explain contemporary American political polarization as an outgrowth of growing income

inequality.  Between the 1920s and the mid-1970s, patterns of wealth distribution in the

United States were comparable to those of other nations with relatively similar levels of

economic, industrial and technological development. However, economic inequality has

grown immensely in the United States in the last thirty-five years, and at a much greater

rate than what can be found in other comparable nations. McCarty, Poole and Rosenthal

also point out that this wealth gap has appeared within the individual American states,

and not among them. The growth of wealth inequality has transpired on a class rather

than sectional basis. (McCarty, Poole, Rosenthal, 2008)

Since the mid-1970s, many of the more underdeveloped areas of

the U.S. have improved their economic standing. Wealthy people in wealthy states have

been have been getting rich at a quicker pace, while poor people in poor states have been

rising out of poverty at a quicker pace. This is no doubt attributable to a variety of causes,

including the growth of the industrial base of the so-called Sunbelt, the effects of tax cuts

and deregulation policies implemented by several administrations, and the expansion of

the welfare state as a barrier to total poverty. Economic inequality has also grown in

Democratic states and decreased in Republican ones. Concerning economic policies that

primarily affect individuals, Republicans will generally favor the affluent while

Democrats will favor the low-income. However, Gelman and associates point out that

there is deviation from this pattern when it comes to policies that affect regions, states or

local communities. In some instances, Democrats will favor more affluent communities

while Republicans will favor poor localities. Gelman observes that “one might see certain

policy areas where Democratic officeholders, as friends of the rich areas, become friends

of the rich people, for example, in supporting the federal tax deduction for state income

tax (which benefits taxpayers, especially upper-income taxpayers, in New York and

California).” (Gelman, 2008, pp. 61-62) Also, interstate social transfer payments are

greater from Democratic states to Republican states rather than vice versa. The richest ten

states receive only eighty cents in federal spending for every dollar paid in taxes while

the poorest ten states receive $1.60. (Gelman, 2008, p. 62) The evidence indicates that

while economic inequality is indeed growing, this expanding class divide is not expressed

in regional divisions and cannot explain the conventional “red state/blue state” political

polarization.

The Voting Patterns

It has been mentioned that in the 2004 presidential election, the “rich people vote”

(persons earning more than $200,000 a year) went overwhelmingly for the Republicans,

with the votes of this group going to the Democrats in only four states. In the same

election, the Democrats won the middle income vote (between $15,000 and $200,000) in

California, Washington, Oregon, Michigan, Minnesota, Illinois and all of the northeastern

states from Maryland upward. The Republicans won the “poor people” vote (less than

$15,000) only in Bush’s home state of Texas, Indiana, and the sparsely populated western

states of Oklahoma, Kansas, Nebraska, Utah, Wyoming, Idaho and the Dakotas.

It is much more striking to observe the voting patterns with regards to church

attendance. In the 2004 election the Republicans won the votes of those who attend

church at least once a week in forty-eight of the fifty states! The Democrats won the votes

of regular churchgoers only in Maryland and Massachusetts. Among semi-regular

churchgoers, the Democrats won fourteen states: California, Minnesota, Wisconsin,

Illinois, Arkansas, New York, Massachusetts, Connecticut, Rhode Island, New

Hampshire, Vermont, New Jersey, Delaware and Maryland. The Republicans won the

votes of non-churchgoers only in ten states: Texas, Idaho, Utah, South Dakota, Kentucky,

Oklahoma, Louisiana, Mississippi, Alabama and South Carolina.

According to the World Values Survey, the United States is unique in that it is the

only one of the world’s wealthier nations with a high level of religiosity. (Inglehart,

2005)) Some observers attribute this to the fact that many Americans are descended from

immigrants who were often from the poorest and most religious sectors of the countries

from where they came. The comparatively high level of economic inequality in the U.S.

makes the nation more likely to display characteristics more common to poor countries

like a greater amount of religious practice or belief. Still another explanation is America’s

tradition of separation of church and state. The lack of an established national church

opens up the “religion market” to competition among a wide variety of denominations

and sects that must rely on the voluntary participation and contributions of adherents in

order to remain active. (Gelman, 2008, pp. 76-77)

It would certainly appear on the surface that the “red/blue divide” simply reflects

the polarization between the religious and the non-religious and that this polarization is

played out in terms of party loyalty and voting patterns.  The reputation of the Republican

Party as the “Party of God” is a relatively new phenomenon. The identifiable pattern of

religious people voting Republican by a significant margin did not appear until the 1992

presidential election when the incumbent George H. W. Bush obtained twenty percent

greater support among those who church attendance was consistent than among those

who were not regular church goers. (Gelman, 2008, p. 84) While Ronald Reagan

received the enthusiastic support of the newly organized “religious right” in the 1980 and

1984 elections, the data shows that the impact of the religious vote in those two elections

was actually less significant that it had been in the election between Gerald Ford and

Jimmy Carter in 1976 (Gelman, 2008, p. 86)

The overall level of religiosity in the United States has decreased significantly

since the early 1960s. The number of people who say they never or rarely attend church

when responding to surveys has grown from only a few percent of Americans in 1960 to

twenty-five to forty percent, with the variation being dependent on such factors as

geography, class position and income levels. Additionally, American society has become

more liberal with regards to a wide variety of issues including race relations, gender

roles, sexuality, and abortion. This social liberalization has coincided with an increased

secularization of public educational institutions. Even some religious denominations have

followed the wider trend of liberalization by, for instance, accepting women and gays

into the ranks of the clergy. Not surprisingly, this process of greater liberalization and

secularization of society at large and greater liberalization within religious institutions

themselves has produced a conservative backlash. Religious conservatives have become

more politically active since the 1970s, and some religious people with more traditional

views have sought out more conservative denominations in response to the increased

liberalism of their former denomination. All of this is well-known.  It is also well-known

that the “red states” tend on average to possess more devoutly religious people that the

“blue states.”

However, there are problems with interpreting the “red/blue” conflict as purely

religious in nature, though it may be tempting to do so from a surface look at the data.

Class and geography are also important parts of the wider picture. For instance, lower-

income people are much more likely to claim the importance of religion to their own

lives, attend church, pray or engage in other religious practices regularly, or to describe

themselves as “born-again” Christians.  The class division between the religious and the

non-religious is also greatest outside the “Bible Belt” of the southern states. These are

fairly predictable statistics.  What is more interesting is to observe the relationship

between income levels and church attendance within individual states. In the poor states,

the higher one’s income, the likelihood of regular church attendance increases. In the

richer states, the higher one’s income, the less likely one will be to attend church

regularly. In other words, in poor “red” states, more affluent people are more likely to

attend church than poor people, but in the wealthier “blue” states it is the other way

around. (Gelman, 2008, pp. 83-84)

With regards to denominational affiliation, mainline Protestants have traditionally

tended to vote Republican, but these have started to move away from consistent support

for the Republicans as the party’s conservative wing has become dominant and the older

Rockefeller-Eisenhower Republicans have been eclipsed. Catholics have traditionally

supported the Democratic Party, but the Catholic vote has been less consistently

Democratic as the party has become more liberal on social questions such as abortion and

gay rights. Prior to the 1980s, “evangelical,” conservative, or fundamentalist Protestants

were primarily a Democratic constituency. Yet the evangelical vote has shifted by a wide

margin to the Republicans since the liberalization of the Democratic Party and the advent

of the “religious right.” (Gelman, 2008, p. 86)

What Does the Data Mean?

The red state/blue state divide and the division between religious and non-

religious voters did not appear until 1992.  As Gelman, et.al. explain:

Part of the story is Bill Clinton, who repelled many religious conservatives

who saw a connection between his adulterous lifestyle and his support for

liberal social causes. (Reagan had been divorced, but that was long in the past, and he sided with the Religious Right on many issues.) There was also the growing strength of the evangelical movement as followers of Pat Robertson

and other gained influence in state Republican parties…On the other side, Democrats became more committed to liberal positions on abortion and gay rights…With the closer alignment of moral issues to the political parties, voters have sorted themselves on these attitudes. (Gelman, 2008, p. 87-88)

Within this political framework and alignment of political parties with particular social

causes and sets of cultural values, a voter who is both affluent and religious will

unsurprisingly vote for the Republicans. A voter who is poor and religious could vote

either Democratic or Republican. The data also shows that wealthy, non-religious people

are about evenly divided between the two parties. In other words, support for the

Republicans comes primarily from middle to upper class people who are also religious.

Support for the Democrats comes from the non-religious and lower-class religious

people. Contra the Marxist view of religion as the “opium of the masses” whereby the

working classes are distracted from pursuing their material interests because of religious

or cultural values or biases, the evidence indicates that it is the affluent whose politics

are most influenced by their cultural norms. Gelman, Park, Shor, Bafumi and Cortina

offer this assessment of their research:

Voters consider cultural issues to be more important as they become

more financially secure. From this perspective it makes perfect sense

that politics is more about economics in poor states  and more about

culture in rich states. And it also makes sense that, among low-income

voters, political attitudes are not much different in red or blue states,

whereas the cultural divide of the two Americas looms larger at high

incomes. For predicting your vote, we suspect that it’s not so important

whether you buy life’s necessities at Wal-Mart or the corner grocery, but

that it might be more telling if you spend your extra income on auto-racing

tickets or on a daily gourmet coffee. We can understand differences between

red and blue America in terms of cultural values of upper-middle-class and

rich voters. Religious attendance is associated with Republican vote most

strongly among high income residents of all states. This does not mean that

lower-income Americans all vote the same way-far from it-but the differences

in how they vote appear to depend less on religious values. (Gelman, 2008, pp. 89-92)

As an illustration, the data from the 2004 election demonstrates that the relationship

between income and church attendance was a predictable indicator of how one would

vote in heavily Democratic states, heavily Republican states and “battleground” states

alike. In all three types of states, high income persons who attend church were likely to

vote Republican, while in strongly Democratic states there was no demonstrable

relationship between income and voting patterns.

Why Is the South Different?

The Southern states present two distinct anomalies. The first of these is Bartels

observation that it is only in the South that the phenomenon of white voters lacking

college education voting Republican emerges. (Bartels, 2006) Even so, it has been

established that lower-income voters in the South overwhelmingly vote Democratic.

What makes the South distinct is the proportionately high number of blue-collar whites

who vote Republican, generally lower-middle class persons with annual earnings in the

$20,000-$40,000 range. Even more interesting is that prior to the civil rights revolution of

the 1960s and 1970s, the Democratic Party was so deeply entrenched and

institutionalized in the South that the Southern states essentially comprised a one-party

region. Indeed, the South was known as the “Solid South” in national electoral politics

because the region’s Democratic loyalties were so predictable. It was not until the

passage of the Civil Rights Act of 1964 and the subsequent Voting Rights Act that white

voters in the South began to drift towards the Republicans. These pieces of legislation

had been passed by a Democratic-controlled Congress and signed into law by the

Democratic President Lyndon B. Johnson. (Lamis, 2005)

This explains the shift of the South to the Republicans generally but what about

working class whites in the South? It was this class of whites that proved to be the most

resistant to civil rights in the South. Upper-income whites were more accommodating to

the institutionalization of civil rights, as it was these whites who stood to gain the most

from the economic transformation of the South during the postwar era from a

predominately agricultural society to a modern industrial society, which necessitated at

least some degree of social modernization as well. Furthermore, upper-income whites

were more able to insulate themselves from the perceived “negative” effects of civil

rights, such as racially integrated public spaces and institutions (schools, parks, pools,

golf courses, theaters, etc.) Many of these whites simply formed private schools and

recreational associations for themselves that remained de facto segregated, and often

resided in neighborhoods where the price of housing was cost prohibitive for blacks. In

other words, upper class whites could enjoy the economic and political benefits of public

desegregation while essentially retaining segregation for themselves on a private basis.

This was not true of the white working class. Urban working class whites

whose resistance to desegregation failed would then relocate to racially homogenous

white neighborhoods in suburban areas outside of cities. Hence, the well-known pattern

of “white flight.” These patterns of a shift from public segregation to private segregation

by upper-income whites and white flight by working class whites tended to push

Southern whites in general towards fiscal conservatism. Simply put, these whites

did not want to pay taxes to support public institutions and facilities that they regarded as

having been “handed over” to blacks. (Kruse, 2005) Consequently, fiscal and economic

conservatives associated with the Republican Party in the Northern states began to regard

de jure or de facto “racial conservatives” in the South as their natural allies and the two

forces began to bend towards one another. (Lewis, 2006) Over time, the openly racial

dimension of this phenomenon would fade into a middle-class oriented fiscal

conservatism that emphasized “color blindness.” It would be an overstatement to claim

that contemporary working class Southern whites who vote Republican in the name of

fiscal and economic conservatism are simply closet racists who hide their real views

behind something more socially acceptable. Indeed, many of them may well be unaware

of the origins of this particular brand of conservatism, and some of these contemporary

Southern white conservative Republicans are transplanted Northerners (or their

descendents) who had little or no personal exposure to the old system of segregation, but

the roots of contemporary Southern white working class political conservatism in

resistance to civil rights is a demonstrable fact. (Lassiter, 2004; Hall, 2005)

The other anomaly to be found in the South is the greater attachment of upper-

income persons to organized religion over lower-income persons. This phenomenon

defies the usual pattern not only in the United States, but world wide. In most societies,

the higher one’s class position, the less likely one will be to practice formal religion. The

American South reverses this pattern. Thus far, it does not appear that enough research

has been done on this situation to make a thorough understanding of its origins or causes

available. One possibility may be the fact that the South was for all practical purposes a

feudal society with a rigid racial caste system and a primarily agrarian economy until the

post-World War Two era. The use of religion as a means of social control by the

traditional Southern white ruling class is well-known. For instance, each of the major

U.S. Protestant denominations split into northern and southern factions over the issue of

slavery prior to the Civil War. Hence, the existence of such contemporary denominations

as the Southern Baptists and Southern Methodists. White fundamentalist preachers were

often defenders of the segregationist status quo during the civil rights era as well.

If indeed religion was used as a force for social control, it is understandable that a

tradition of greater than usual attachment to religious institutions would develop among

privileged Southern whites. Likewise, it would certainly be understandable that lower-

class persons would experience greater alienation from religious institutions in such a

situation, leading to an inversion of the usual norm where it is the lower classes that are

more religiously devout than the upper classes. Similar situations have emerged in other

nations. For instance, the radical labor and peasant movements in Spain during the pre-

Franco years included many otherwise culturally conservative persons who developed a

militant anti-clericalism in response to the role of the Catholic Church in Spain as

accomplices to a highly oppressive ruling class. (Bookchin, 2001)

The American South displays characteristics concerning the relationship between

personal religiosity, class position and political affiliation that are in some ways similar to

what is often found in Latin American countries. The American South is also more

similar in its history to Latin America than other regions of North America. Both the

South and most of Latin America have a feudal or quasi-feudal past as agrarian societies

with a rigid class structure with organized religious institutions being very much on the

side of the ruling class. In Latin America, the lower-classes tend to be very religious on a

personal level, while formal displays of religious piety through such things as regular

church attendance are more common to the middle classes. The upper layers of the

Church hierarchy in Latin America tend to be very conservative. Voting patterns

in Latin American countries are such that the lower classes typically vote for the Left,

while the middle classes will vote for the center-right Christian Democratic parties, and

the upper classes will vote for the “hard Right.” (Yglesias, 2007) This fairly closely

mirrors class voting patterns in the southern states in the U.S.  It is also true that

evangelical religion in Latin America takes on different forms depending on the class

position of the participants. Middle to upper class Latin American evangelicals will often

espouse social or political views similar to those of the U.S. “Religious Right.” The

Guatemalan dictator Rios Montt was an example of this. On the other hand, lower class

evangelicalism in Latin America tends to take on a “social gospel” flavor much like

African-American religion in America or past expressions of left-wing evangelicalism

that emerged in American populism during the late nineteenth and early twentieth

century. (Freston, 2008) The American South and Latin America are similar to one

another in unique ways in that both regions have both a fairly recent quasi-feudal,

agrarian past and democratic governments. This would set both regions apart from the

rest of North America, Europe, Asia, Africa or the Middle East. There appears to be

unique and similar dynamics working in both regions that give these two regions

characteristics that are difficult to find elsewhere.

The Big Sort

Still another factor affecting voting patterns in American elections is what author

Bill Bishop has called “The Big Sort.” This is a phenomenon where persons with the

financial means of doing so will relocate to a neighborhood, community or even a state

that is more compatible with their cultural interests. This creates a system of cultural self-

segregation among middle to upper income Americans.(Bishop, 2008) To demonstrate

his argument, Bishop acknowledges that in the 1976 Ford-Carter election, the number of

counties in the United States where either candidate won by a landslide (a margin of

twenty percentage points or greater) was significantly fewer in number than the number

of counties where victory was determined by a landslide in the Bush-Kerry election of

2004. Bishop also describes his experience of living in a liberal enclave in the Austin,

Texas area:

My wife and I…didn’t intend to move into a community filled with

Democrats, but that’s what we did-effortlessly and without a trace of understanding about what we were doing…In 2000, George W. Bush…

took sixty percent of the state’s vote. But in our patch of Austin, Bush came

in third, behind both Al Gore and Ralph Nader. Four years later, eight out of

ten of our neighbors voted for John Kerry. (Bishop, 2008, p. 1)

Like other observers of these issues, Bishop traces the beginnings of the “big sort” to the

cultural revolution of the 1960s and 1970s and the subsequent backlash from social

conservatives and religious traditionalists. However, Bishop maintains that the sorting

process really did not begin to manifest itself until the 1990s. During that decade, the

baby boom generation, the first to be heavily influenced by the 1960s-era “cultural

revolution,” entered middle age. The economic expansion of the 1990s and the growth of

the educated population converged to create a situation where large numbers of persons

existed who possessed a combination of affluence, education and a relatively liberal

social outlook. Consequently, both middle aged baby boomers and their younger,

“Generation X” cohorts began to congregate in urban centers “where they would not be

bound by old ideas or tight social ties.” (Bishop, 2008, p. 144)

It is also important to recognize that the “big sort” occurs primarily at the level of

local communities, and sometimes individual neighborhoods, rather than at the state

level.  John Tierney observes that in the 2000 presidential election, George W. Bush

received the smallest numbers of votes in the states of Maryland, Connecticut, Rhode

Island, Massachusetts, New York and Hawaii. However, all of these states had

Republican governors at the time. Tierney believes such patterns indicate that the “red

state/blue state” divide is a myth, and that most Americans are centrists. (Tierney, 2005)

Jonathan Kandel observes that in the 2000 election, there were only five red states

(Wyoming, North Dakota, Utah, Nebraska and Idaho) and one blue state (Rhode Island)

where the candidate of either party won by more than sixty percent. Kandel also observes

that of the eleven states that passed initiatives prohibiting same-sex marriage in 2004, two

of these states (Oregon and Michigan) went for the Democrats in the presidential

election, and many others were competitive in that neither party won the presidency by

more than sixty percent. (Kandel, 2006)

Bruce Oppenheimer argues that the division between red and blue states

represents divisions between Congressional districts rather than states, and he attributes

this to partisan redistricting, which groups together voters with similar views and partisan

sympathies and has the effect of creating “safe” districts for incumbents or their parties.

(Oppenheimer, 2005) Yet the most compelling evidence is that offered by Bishop.

According to Bishop, in 1976 only twenty-six percent of Americans lived in what he calls

“landslide counties” where the presidential vote is determined by more than a sixty

percent total for the winner. By 1992, the year that Gelman and associates consider to be

the starting point for the “red/blue” divide, thirty-eight percent of voters resided in

landslide counties. That percentage increased with each subsequent presidential election,

and by 2004, forty-eight percent of Americans were living in landslide counties. (Bishop, 2008, pp. 9-10)

The 2008 Presidential Election

Bishop has updated his research to include the 2008 presidential election.  In

2008, the number of Americans living in landslide counties was the same as in 2004:

forty-eight percent. This division has tilted strongly towards the Democrats. In 2004, 94

million lived in Democratic landslide counties, while in 2008 it was only 64 million. In

2008, 53 million Americans were in Republican landslide counties, while in 2004 it had

been 83 million. Among states, the average winning margin was seventeen percent, as

opposed to sixteen percent in 2004, fifteen percent in 2000, and ten percent in 1976. The

number of landslide states increased to thirty-six from twenty-nine in 2004. The number

of states where the election was decided by five or less percentage points was down to

seven, from eleven in 2004. Barack Obama won forty-three percent of the rural vote, up

from Kerry’s forty percent in 2004, and fifty-seven percent of the urban vote, up from

Kerry’s fifty-one percent.  Bishop attributes Obama’s greater vote totals in rural America

over Kerry to the success of his strategy of targeting college towns within rural areas.

Also, the 2008 election demonstrated strong divisions among racial and ethnic groups. In

those counties where Obama won by a landslide, only 1.3 whites can be found for every

minority. Yet in McCain-landslide counties, there are five whites for every minority. (Bishop, 2008)

The Future

The most striking feature of the 2008 election is the fact that while the number of

landslide counties remained the same, on a partisan basis the number of persons living in

a landslide county increased by a third for Democrats and decreased by about the

same amount for Republicans. Bishop attributes this to a higher out-migration rate among

Democrats, who relocate to traditionally “red” areas but bring “blue” values with them,

and consequently influence voting patterns in their new localities accordingly. (Bishop, 2008) However, such a shift in a four year period might also be attributed to much more far reaching demographic, cultural and generational change. In 1997, the conservative writer Peter Brimelow made this prediction:

The Republican hour is rapidly drawing to a close. Not because the (Republican base) of the West and the South, of the middle class and urban blue-collar voters, is breaking up in the traditional manner. Instead, it is being drowned—as a direct result of the 1965 Immigration Act…Nine-tenths of the immigrant influx is from groups with significant—sometimes overwhelming—Democratic propensities. After thirty years, their numbers are reaching critical mass. And there is no end in sight.

To estimate the future impact of Immigration, we took the 1988 presidential race, in which George Bush beat Michael Dukakis with 53 per cent of the vote. This figure happens also to be the average vote received by the Republicans in presidential elections since 1968—the largest advantage won by any party over any six elections in American history. And it is the vote received by Republicans in 1994, when they took control of the Senate and House. It can reasonably be regarded as the Republican high-water mark.

Then we lowered this high-water mark by accounting for the shifting ethnic balance that the Census projects will result from immigration, assuming that the ethnic groups continued to vote as they did in 1988. The results are startling…Even if the Republicans can again win their 1988 level of support in each ethnic group—which they have miserably failed to do against Bill Clinton—they have at most two presidential cycles left. Then they go inexorably into minority status, beginning in 2008. (Brimelow, 1997)

Subsequent events since the publication of Brimelow’s article in 1997 would seem to

vindicate his prognosis. Another work making a similar prediction was published by two

writers associated with The New Republic in 2002. In their The Emerging Democratic

Majority, authors John P. Judis and Ruy Teixeira predicted the rise of a new electoral

majority rooted in educated urban professionals, racial and ethnic minorities, feminists

and educated working women, college students, environmentalists, secularists, gays and

lesbians. Judis and Teixeira refer to this phenomenon as “George McGovern’s Revenge”

as these were largely the groups that comprised the 1972 McGovern coalition that lost in

a landslide to President Nixon.

However, there is another constituent group among Judis and Teixeira’s predicted

Democratic majority: the white working class. Observing how the Democratic Party lost

substantial numbers of blue collar white voters during the post-civil rights era over race

issues, foreign policy, crime, the rise of the counterculture and the conservative religious

backlash, gun control and the economic downturn of the 1970s, Judis and Teixeira argued

that these voters began to return to the Democrats because of the recession that occurred

in the early 1990s during the administration of President George H.W. Bush. In other

words, blue collar whites were returning to the Democrats at precisely the same time as

the emergence of the red state/blue state electoral divide. President Reagan won the votes

of unionized white workers in 1980 and 1984. George H. W. Bush lost these voters by

four percentage points in 1988. Clinton won the white unionized worker vote by an

average of twenty-three percentage points in 1992 and 1996. Yet, it is during these years

that the current electoral divide emerges, so clearly the conventional view offered by

Thomas Frank and others of “working class Republicans versus upper class Democrats”

is false and likely rooted in outdated stereotypes left over from the Nixon and Reagan

eras.  Indeed, Judis and Teixeira point out that the composition of the “white working

class” has changed significantly, with nearly fifty percent of white workers being women

by 2000, and a significant number of younger, urban white workers with relatively liberal

views on social issues like abortion, the environment or gay rights. Like Brimelow, Judis

and Teixeira predicted that 2008 would be the year that the new Democratic majority

eventually became dominant. (Judis and Teixeira, 2002, p. 14, 37-66)

Gelman and associates demonstrate rather clearly that the primary driving force

in the red state/blue state “culture war” is religion. The primary indicator of whether a

middle class person will vote Democratic or Republican is whether they attend church

regularly or not. According to the American Religious Identification Survey, nearly all

American religious denominations have lost members over the last twenty years.

Catholics and Baptists, the two largest denominations, lost one and four percent of their

membership, respectively. The number of people claiming the generic label of

“Christian” has dropped by half a percentage point. Mainline Protestant denominations

have lost nearly a third of their membership since 1990. Persons claiming no religion at

all and persons with agnostic views of religion have both doubled in the past twenty

years, and collectively, skeptics, atheists, agnostics and other unbelievers are the single

largest religious group in the U.S. at twenty percent, except for Catholics with twenty-

five percent.

Adherents of the Jewish religion have decreased by one third. Fringe

Protestant denominations like the Pentecostals, Mormons, Jehovah’s Witnesses or

Seventh Day Adventists have either remained the same numerically or increased slightly,

but these are still very small when compared to American society as a whole. The only

religions that have experienced real growth in the past twenty years have been those from

outside traditional American culture. The number of U.S. Muslims and adherents of

“Eastern” religions like Buddhism or Hinduism have doubled, largely due to

immigration, and adherents of so-called “new age” spiritualities, neo-paganism, and

Wicca have grown by one third. (Grossman, 2009)

Summary and Conclusion

It has been demonstrated that the popular view of the red-state/blue-state “culture

war” divide as one pitting working class conservatives against affluent liberals is false.

This view is rooted in archaic stereotypes that have not been especially relevant to U.S.

electoral politics since the “red-state/blue-state” dichotomy has emerged. Specifically, the

defection of white working class voters to the Republicans in the 1970s and 1980s has

since reversed itself. The only region of the United States where the blue collar class

votes Republican in any significant numbers is in the South, and this is due to that

region’s unique history in matters of race, religion and economics. The present-day red-

state/blue-state divide first begins to appear on the electoral map in the 1992 presidential

election, precisely the time that blue collar whites were returning to the Democrats.

Nor is this divide a matter of “rich versus poor.” The United States is indeed

polarized along class lines, but this economic polarization takes places on a national

rather than sectional basis. As the overall pattern of wealth and income distribution in the

U.S. has become more uneven in recent decades, support for the Democratic Party among

working class voters has actually increased. Instead, the “red/blue” conflict represents an

intra-class conflict within the middle class, primarily the upper middle class, with middle

class voters in wealthy states being more culturally liberal than their counterparts in

poorer states. The driving force behind this middle class culture war is religion, with

church attendance being the primary indication of how a middle class person will vote.

Geographically, this cultural polarization transpires more at the local community level

rather than at the state level, pitting rural versus urban areas and conservative

neighborhoods against liberal ones, though differences among states are not insignificant.

The most compelling piece of evidence to support the argument that the

“red/blue” conflict represents an intra-class divide within the affluent middle-class is the

fact that electoral maps show that the “poor vote” overwhelmingly goes to Democrats

while the “rich vote” overwhelmingly goes to Republicans, and the middle-class vote

breaks down geographically on the standard “red/blue” pattern. This divide plays out on a

geographical basis to the degree that it does because of the effects of Bill Bishop’s “Big

Sort” whereby middle class persons possess the means of self-segregation along cultural,

religious and ideological lines, and this system of self-segregation occurs primarily on a

local rather than state level. The evidence to support this localized geographical divide

consists primarily of the wide margins by which a political party will often win in a

specific locality. In each of the last two presidential elections, one of the parties beat the

other by a margin of more than twenty percentage points in forty-eight percent of all

American counties. The gaps at the state level tend to be smaller. In the 2008 election, the

overall pattern of “red/blue” division among middle and upper-middle income voters

continued. The number of “blue” states increased, while the number of counties

exhibiting an electoral polarization wider than twenty percentage points remained the

same. This is apparently due to two principal factors: a greater out-migration rate from

blue areas to red areas rather than vice versa, and demographic, cultural and generational

change that indicates the population groups that are inclined to vote Republican are

shrinking, while those inclined to vote Democratic are increasing.

Furthermore, it can be predicted with relative safety that, barring completely

unforeseen circumstances, the “liberal” side will be the winning side in the “culture war”

and the Democratic Party will likely be the dominant party in U.S. politics for the

foreseeable future. This is due to a combination of the aforementioned generational,

cultural and demographic changes, large scale immigration, economic downturn, an

increased number of educated urban professionals, changing gender roles that include

expanding roles for women, and declining interest in traditional religious beliefs,

practices or denominational affiliation.  This does not mean that “social conservatives” or

the Republican Party will disappear, far from it, but it does mean that the political Right

is less likely to be as influential in the foreseeable future as it has been in the recent past.

Bibliography:

Abramowitz, Alan and Kyle L. Saunders (2005). Why We Can’t We All Just Get

Along?: The Reality of a Polarized America. The Forum, Berkeley Electronic Press.

Bartels, Larry M. (2008). Unequal Democracy: The Political Economy of the New Gilded

     Age. Princeton University Press.

Bartels, Larry M. (2006). “What’s the Matter with What’s the Matter with Kansas?”

Journal of Political Science Quarterly, 2006, 1, 201-226.

Bill Bishop, (2008). No, We Didn’t: America Hasn’t Changed As Much as Tuesday’s

Results Would Indicate. Salon, November 10, 2008.

Bishop, Bill and Robert G. Cushing (2008). The Big Sort: Why the Clustering of Like-

     Minded America is Tearing Us Apart. Boston, New York: Houghton Mifflin

Company.

Bookchin, Murray (2001). The Spanish Anarchists: The Heroic Years, 1868-1936.

London: AK Press.

Brimelow, Peter and Edward S. Rubenstein (1997). Electing a New People. National

     Review, June 16, 1997.

Brooks, David (2001). One Nation, Slightly Divisible. The Atlantic Monthly, December

2001.

Fiorina, Morris P. with Samuel J. Abrams and Jeremy C. Pope. (2004). Culture War? The

     Myth of a Polarized America. Longman.

Florida, Richard (2002). The Rise of the Creative Class, And How It’s Transforming

     Work, Leisure and Everyday Life. Basic Books.

Frank, Thomas (2004). What’s the Matter with Kansas: How Conservatives Won the

     Heart of America. New York: Metropolitan Books.

Freston, Paul (2008). Evangelical Christianity and Democracy in Latin America. Oxford

University Press.

Gelman, Andrew and David Park, Boris Shor, Joseph Bafumi, Jeronimo Cortina (2008).

Red State, Blue State, Rich State, Poor State. Princeton and Oxford: Princeton

University Press.

Grossman, Cathy Lynn (2009). Most Religious Groups in USA Have Lost Ground,

Survey Finds. USA Today, March 17, 2009.

Hall, Jacquelyn Dowd (2005). The Long Civil Rights Movement and the Political Uses of

the Past. Journal of American History 91: 1233-1263

Hunter, James Davison (2005). Culture Wars: The Struggle to Define America. Making

     Sense of the Battles Over the Family, Art, Education, Law and Politics. Second

Edition. Basic Books.

Inglehart, Ronald and Pippa Norris (2005). Sacred and Secular: Religion and Politics

     Worldwide. Cambridge University Press.

Judis, John and Ruy Teixeira (2002). The Emerging Democratic Majority. New York:

Scribner.

Kandel, Jonathan (2006). The Myth of the Red State/Blue State Divide. Archived at http://www.politicsandgovernment.ilstu.edu/downloads/icsps_papers/2006/JonathanKandel1.pdf.

Kimball, David C. and Cassie A. Gross (2005). “The Growing Polarization of American

Voters,” Presented at The State of the Parties: 2004 and Beyond conference, Akron,

OH, October 6, 2005.

Kruse, Kevin M. (2005, July). The Politics of Race and Public Space: Desegregation,

Privatization, and the Tax Revolt in Atlanta. Journal of Urban History: 610-633

Lamis, Alexander (2005). The Emergence of a Two-Party System: Southern Politics in

the Twentieth Century. The American South in the Twentieth Century. Athens:

University of Georgia Press.

Lassiter, Matthew D. (2004). The Suburban Origins of “Color-Blind” Conservatism:

      Middle-Class Consciousness in the Charlotte Busing Crisis. Journal of Urban History

      30: 549-582

Lewis, George (2006, February). Virginia’s Northern Strategy: Southern Segregationists

and the Route to National Conservatism.  Journal of Southern History, 72:111-146.

McCarty, Nolan with Keith Poole and Howard Rosenthal (2006). Polarized America: The

      Dance of Ideology and Unequal Riches. Boston: MIT Press.

Oppenheimer, Bruce (2005). Deep Red and Blue Congressional Districts: The Causes

and Consequences of Declining Party Competitiveness. In Larry Dodd (Ed.), Congress

      Reconsidered, 8^th edition. Washington, D.C.: CQ Press.

Stonecash, Jeffrey (2005). Scaring the Democrats: What’s the Matter with Thomas

Frank’s Argument? The Forum: A Journal of Applied Research in Contemporary

     Politics, Vol. 3, No. 3, 2005.

Tierney, John. (2004). A Nation Divided? Who Says?. The Nation: On Message, June 13,

2004. Sec. 4, Col. 1.

Yglesias, Matthew (2007). Religion and Income. The Atlantic. November 11, 2007.