The Politics of Definition, Part II

Editor's Note: This is the second installment of this important paper, which describes the precise nature of the dilemma faced by progressives and Democrats today and offers new solutions for the way forward.

In Part I, posted last week, the authors described the basic problem facing progressives and the Democratic Party -- the “identity gap” faced within many specific voter groups -- and then surveyed electoral areas of Democratic and progressive strength. Here, in Part II, they discuss and analyze areas of Democratic and progressive weakness. Parts III and IV will lay out the authors' recommendations for the future, and will appear later this week.

-- Michael Tomasky

Building Blocks of a Progressive Majority: Weaknesses

The White Working Class

The key weakness of the progressive coalition can be summarized easily: very weak support among white working class voters (defined here as whites without a four-year college degree). These voters, who are overwhelmingly of moderate to low income and, by definition, of modest credentials, should see their aspirations linked tightly to the political fate of the progressive movement. But they don't.

Data from the last two presidential elections vividly demonstrate this problem and underscore its significance for progressives. In 2000, Al Gore lost white working-class voters by 17 percentage points; in 2004, John Kerry lost them by 23 points, a swing of 6 points against the Democrats. In contrast, Gore lost college-educated whites by 9 points and Kerry lost them by 10 points -- not much change. 1

Therefore, white working-class voters were responsible for almost all of George W. Bush's increased margin among whites as a whole in the 2004 election (which went from 12 to 17 points). And Bush's increased margin among whites was primarily responsible for his re-election.

Almost all of the white working-class movement toward Bush was among women rather than men. Bush won white working-class men by almost identical margins in the two elections (by 29 points in 2000 and by 30 points in 2004). But he substantially widened his margin among white working-class women, going from a 7-point edge in 2000 to an 18-point lead in 2004. That 11-point swing against the Democrats among white working-class women was arguably the most important single fact about the 2004 election.

The basic reasons for this stunningly poor Democratic performance among the white working class can also be easily summarized. Among white working-class voters, 66 percent said they trusted Bush to handle terrorism, compared to just 35 percent who said the same about Kerry. That's very bad, but perhaps not all that surprising. What is more surprising is this: 55 percent of these voters said they trusted Bush to handle the economy, and only 39 percent said the same about Kerry.

It's also interesting to note that there wasn't much of a difference in these sentiments between men and women in the white working class: 55 percent of white working-class women said they trusted Bush to handle the economy and 40 percent said they trusted Kerry, while 56 percent of white working-class men said they trusted Bush on the economy and 37 percent said they trusted Kerry.

That helps explain the big shift among white working-class women described above. Not only were these women alarmed about terrorism -- which pushed them toward the GOP -- but they were also, in contrast to previous elections, no more likely to find the Democratic economic message compelling than their male counterparts. In neither area --the economy or terrorism -- did the Democratic program speak clearly to these voters' concerns and earn their trust.

It is also important to stress that Democrats did especially badly among white working-class voters who weren't poor, but rather had moderate incomes and some hold on a middle-class lifestyle. Among working class whites with $30,000 to $50,000 in household income, Bush beat Kerry by 24 points (62 percent to 38 percent). And, among working-class whites with $50,000 to $75,000 in household income, Bush beat Kerry by a shocking 41 points (70 percent to 29 percent). Clearly, these voters do not see progressives as representing their aspirations for a prosperous, stable, middle-class life.

Progressives' difficulties here are underscored by the large size of this group. According to the 2004 Current Population Survey (CPS) Voter Supplement data, white working-class voters are a larger portion of the electorate than indicated by the exit polls -- 52 percent, rather than 43 percent. Based on educational attainment trends and population trends by race, a reasonable guess is that the size of the white working class in another 10 years, even though it is shrinking, will still be around 46 percent to 47 percent -- a very large group among which to be doing very poorly. 2 In fact, a progressive majority coalition is simply not possible if that poor performance continues, despite the many ways in which demographic change and growth favor progressives, including the increasing proportion of single women within the white working-class population.

But is it really feasible for progressives to significantly improve their performance among white working-class voters? That would appear to depend on the extent to which they can they can clarify their views and principles to these voters and begin earning their trust again. Right now, the Democrats are 23 points down to the Republicans among these voters on knowing what they stand for. 3 Narrowing that gap is key to improving performance among this critical group.

And there is a lot of room for that improved performance. Keep in mind that Bill Clinton actually carried white working-class voters in both his successful presidential campaigns (by a single percentage point in both instances). 4

But Democrats need not replicate that performance. If Democrats can simply keep the Republican margin among white working-class voters to the low double digits (say 11 to 12 points), and maintain their margins from 2004 among college-educated whites and among minority groups (note that we assume no improvement from 2004 in the Democratic performance among Hispanics, though we strongly believe that is likely to happen), our estimates indicate that the Democrats would win the popular vote in the next presidential election by 3 points. That would be an exact reversal of the 2004 popular vote, which Bush won by around 3 points.

And if the Democrats can keep the Republican margin among working-class whites to single digits? Then it should be possible to start building a solid majority coalition for progressives in very short order.

White Catholics

There are several other important characteristics of white voters that intersect the white working-class category -- either reinforcing or mitigating Democrats' basic problem with those voters -- but are worth considering independently. One such characteristic is Catholicism. White Catholics have historically been a relatively good group for progressives among white voters but have also been quite volatile. Here are the margins among white Catholic voters in the past five presidential elections:

1988: plus-14 Republican

1992: plus-5 Democratic

1996: plus-7 Democratic

2000: plus-7 Republican

2004: plus-13 Republican

As the data show, that volatility has lately sent them away from the Democrats and toward the GOP. That has been a development of real consequence, since they are most certainly a large enough group (21 percent of voters in the 2004 election) to have a serious impact on election outcomes. 5

Yet, as a March 2005 Democracy Corps survey report on white Catholics pointed out, white Catholic voters are considerably more Democratic than other white voters and more moderate on a whole range of issues, including tolerance on homosexuality and openness to stem-cell research. 6 So what explains their surge away from the Democrats? Part of the reason lies in the fact that, according to that survey, the GOP had a 33 point advantage among this group on “know what they stand for.” (In January 2006 Democracy Corps polling, the GOP advantage was still a very healthy 26 points). Moreover, the top reason cited by white Catholics on why Kerry lost the 2004 election was “not clear on what he stood for” (48 percent selected this reason as one of the two top reasons Kerry lost, twice as many as selected “permissive views on issues like abortion and gay marriage” as one of the reasons).

White Married Women

The “marriage gap” -- where married voters lean toward the GOP and single voters toward the Democrats -- is famously an important part of today's political landscape. A particular problem for progressives lies among a subset of married voters, white married women. Specifically, married white working-class women (62 percent of white working-class women) gave Bush a 15-point margin in 2000 and more than doubled that margin, to 31 points, in 2004. These women are responsible for most of the shift toward Bush among white working-class women, which, as mentioned above, was probably the key electoral shift against the Democrats in the 2004 election. 7

The reasons for the shift track pretty closely with the data cited above on white working-class women in general: these married women trusted Bush not only on security issues but also on handling the economy. The failure of Democrats to convince these women that their families' aspirations for economic advancement would be best served by a progressive agenda indicates a serious weakness -- as does the continued failure to convince these voters progressives know what they stand for. Right now, the GOP has a whopping 32-point advantage among these voters in this area. 8

White Evangelicals

Perhaps no feature of the 2004 election received more attention than the allegedly central role of white evangelical Christians and their high turnout in Bush's victory.

But the evidence that white evangelicals were so very, very important (as opposed to merely important, which seems reasonable) is shockingly thin. Perhaps the main piece of evidence for this claim is that 23 percent of voters in the NEP exit poll were white “born-again or evangelical” Christians, who supported George Bush, 78 percent to 21 percent. 9

Unfortunately, we have no idea how that compares to 2000, since the exit polls didn't ask the same question in 2000. Instead they asked a very different question about being part of the “religious right,” which categorized 14 percent of voters as part of the white religious right. Clearly, to conclude from these two different questions that white evangelical turnout increased from 14 percent to 23 percent from 2000 to 2004 is inappropriate.

Better purchase on the question of whether evangelical turnout increased may be obtained from the exit poll question on the frequency of religious service attendance. And this question shows that whites who said they attended services more than weekly were rock steady at 11 percent of voters in 2000 and 2004. Moreover, whites who said they attended weekly actually declined across the two elections from 23 percent to 21 percent. This hardly seems consistent with a wave of white evangelical turnout.

Moreover, if one looks at intensity of support, the exit polls do not indicate that more observant white voters dramatically increased their support of Bush in 2004. Indeed, Bush received a greater increase in support (a 6-point gain in margin) from less observant white voters -- those who are moderately observant to completely unobservant -- than he did from more observant white voters (a 3-point shift). Moreover, this unimpressive 3-point shift was driven entirely by those who attend weekly, since his support actually dropped slightly among the most observant white voters, those who attend more than weekly.

None of this seems consistent with the idea that surging evangelical turnout and support put Bush over the top in 2004. Neither does evidence from other surveys. For example, the leading academic survey of religion and politics, the National Survey of Religion and Politics (NSRP), conducted by the University of Akron's Bliss Institute, found that white evangelical protestants (measured by a sophisticated series of questions on religious affiliation, beliefs and practices) were 26 percent of voters in 2004, identical to their level in 2000. The NSRP also found only a modest shift in support toward Bush among these voters; less, for example, than the shift against Bush among mainline Protestants. (Interestingly, the 2004 NSRP survey also identified a group of white evangelicals -- “modernist evangelicals,” about 11 percent of the overall group -- who are far more liberal than typical white evangelicals and who actually supported Kerry, 52-to-48.)

However, even though their turnout and support levels for the GOP do not appear to be surging, it cannot be denied that white evangelicals overall are still a very strong group for the GOP and a problem for progressives. If this group was growing over the long term, the task of building a progressive majority would be far more difficult.

But there is little evidence this is happening. The NSRP, for example, has found essentially no change in the level of white evangelicals in the population since its first survey in 1992. Recent Gallup surveys are consistent with the level of white evangelicals measured by the NSRP and also show little sign of an increase in that level.

Nor do indicators of religious observance provide indirect evidence that evangelicals' share of the population is increasing. Gallup data show no change in the share of the population attending church weekly or almost once a week since the early 1990s .10 And in the University of Chicago's GSS, which has asked a consistent question on frequency of church attendance since the early 1970s, there has actually been an 8-point drop over time in the share of the population who say they attend every week or nearly every week (from 41 percent to 33 percent). Note that the same survey shows a 9-point increase since the early 1970s in the share of the population attending only once a year or less (from 29 percent to 38 percent).

“Red” States and Regions

In the last four elections, the Republicans have carried 16 states (Alaska, Arkansas, Idaho, Indiana, Kansas, Mississippi, Nebraska, North Carolina, North Dakota, Oklahoma, South Carolina, South Dakota, Texas, Utah, Virginia, Wyoming) for a total of 135 electoral votes. This is far fewer than the 18 states (plus the District of Columbia) and 248 electoral votes carried by the Democrats in all four elections.

But this GOP base should be adjusted to reflect their exceptionally strong performance in some states in the last two elections. One way to do this is to add states (besides those included in the above list) that the Republicans have carried by an average of 10 points or more in 2000 and 2004. Using this procedure, we add Georgia, Kentucky, Louisiana, and Montana, making a total of 20 states with 170 electoral votes in the GOP base.

We divide the remaining states (that is, outside of the GOP base and outside of the Democratic base and purple-leaning blue categories described earlier) into three categories. First, there is a small category of three “pure purple” states that have split their support between the two parties in the last two elections: Iowa, New Hampshire and New Mexico. These states have a total of 16 electoral votes.

Next, there is a very significant group of states -- Florida, Missouri, Nevada, and Ohio -- where the average GOP margin in the last two elections has been 5 points or less. This purple-leaning-red category has a total of 63 electoral votes.

Finally, there is a very interesting group of five states–Arizona, Arkansas, Colorado, Tennessee, and West Virginia–that might be characterized as “red vulnerable.” In these states, the average GOP margin in the last two elections has been less than 10 points (though more than five). And, by definition, they are also states that have been carried by the Democrats as least once in the last four elections. They have a total of 41 electoral votes.

That these suggestions of GOP vulnerability are not completely fanciful is indicated by the latest Gallup party identification figures by state. 11 Democrats have the party ID advantage in every state in the pure purple, purple-leaning-red, and red-vulnerable categories, save Tennessee (which is dead even) and Arizona (where they have a 5-point deficit). In the pure purple states, they lead by 6 points in Iowa, 14 points in New Hampshire and 8 points in New Mexico. In the purple-leaning-red category, they lead by 1 point in Florida, 8 points in Missouri, 12 points in Nevada, and 7 points in Ohio. And in the red-vulnerable states, they lead by 11 points in Arkansas, 3 points in Colorado, and 13 points in West Virginia.

Looked at by region, the GOP base is entirely in the South (defined here as the 11 states of the Old Confederacy plus Kentucky and Oklahoma) and in plains and mountain states. It does not include a single Midwestern (except Indiana), Northeastern, Southwestern, or Western (except Alaska) state. With the two exceptions noted, all states in those areas are either in the progressive base/purple-leaning-blue categories or in purple or red categories that suggest GOP vulnerability. In addition, there are two Southern states -- Arkansas and above all Florida -- that also seem vulnerable.

The argument is made, however, that population shifts over time will alter the electoral calculus profoundly in favor of the GOP. This is wildly overstated. For example, looking 10 years in the future, when the 2010 Census results will have reapportioned today's electoral vote, the GOP base is projected to grow by only four electoral votes, according to demographer William Frey. 12 In addition, the highly contested states of Florida and Nevada will grow by a total of three electoral votes. Together, these changes would not appear to fundamentally alter the picture sketched above. 13

Emerging Suburbs, True Exurbs, and Rural Areas

As mentioned earlier, progressives are strongest in the cities and inner suburbs, while the GOP gets stronger the farther away from the urban core you get -- into emerging suburbs, true exurbs, and rural areas. To properly analyze where GOP domination of these areas is most important and where progressives might challenge that domination, it is important to distinguish between true exurban and emerging suburban areas (which together constitute what people usually think of as “exurbia”). Today's true exurbs contain only 2 percent of the nation's population. Emerging suburbs on the other hand contain 13 percent of the nation's population and, on average, are growing faster than any other type of county in the United States, including true exurbs. 14 Emerging suburbs include such well-known counties as Loudoun County, Virginia, outside Washington, D.C.; Anoka County, Minnesota, outside Minneapolis; Warren County, Ohio, outside Cincinnati; and Douglas County, Colorado, outside Denver.

Virginia Tech's Metropolitan Institute (MI) describes the true exurbs as:

[T]he most far-flung [metropolitan] counties with the lowest -- essentially rural -- population densities. Large-scale suburbanization is just about to take hold in these places, as they offer even better bargains, and more land (but longer commutes) than emerging counties. Exurban counties are included in metropolitan areas by the census because they share a functional relationship with neighboring counties via commuting. But by appearance, these places are barely touched by urbanization.

These true exurban counties voted for Bush over Kerry by 62 percent to 37 percent, a lop-sided result, to be sure, and a 10-point gain in GOP margin over 2000. But these counties only contributed 9 percent of Bush's net vote gains between 2000 and 2004, mostly due to their relatively modest population sizes.

The emerging suburban counties were more consequential, though the actual numbers of exurban and emerging suburban counties are roughly equal in the MI typology. They are described as:

The new “it” county of today. They are mostly the fastest growing counties in the region, and are often found in even slow growing regions such as St. Louis (e.g., St. Charles County, Missouri) and Cincinnati (e.g., Boone County, Kentucky). Emerging suburbs are almost wholly products of the past two decades and are booming with both people and the beginnings of commerce (although they remain mostly commuter zones). Emerging suburbs are both upscale and downscale and may feature everything from McMansions to trailer parks. Residents in emerging suburbs typically see these places as bargains compared to mature suburbs. That is true for households that buy a McMansion over an older and smaller tract home in a mature suburb, or a first-time home buyer that “drives to qualify” by finding a modest attached dwelling at the edge of the region.

The Bush-Kerry split here was less lopsided (56 percent to 43 percent) and represented only a 5-point gain in margin over 2000. But since these emerging suburban counties are much larger than exurban counties, they contributed 26 percent of Bush's net vote gains between 2000 and 2004, dwarfing the true exurban contribution.

It is important to note that the GOP does by far the best in emerging suburban counties in their solid red base states. In those states, they had a crushing 34-point margin (67-to-33 percent) in the emerging suburbs in 2004. But everywhere else, the Democrats were much closer. In the solid blue, Democratic base states, Kerry lost emerging suburban counties by only 51-to-48 percent. In purple-leaning-blue states and purple-leaning-red states, he did just a bit worse in the emerging suburbs, losing 53-to-46 percent (and there are states in these categories, of course, where the Democrats did far better than this average, like Florida, where Kerry lost the emerging suburbs by only a single point, 50-to-49). 15 And even in red-vulnerable states, Kerry was still within a 58-to-41 percent margin in these counties.

Note also that the GOP margin in the emerging suburbs overall in 2000 was only 52 percent to 44 percent and in 1996 a mere 45 percent to 44 percent. Together, these data make clear that emerging suburban counties are not only far more important to Bush's coalition than true exurban counties, but also far more contestable by progressives.

This assessment is supported by results from the 2005 elections. In the 2005 Virginia gubernatorial race, Republican Jerry Kilgore, after running a bruising, culture wars-driven campaign against Democrat Timothy Kaine, lost the emerging suburban county of Loudoun–the second fastest-growing county in the entire nation since 2000 -- to Kaine by 3,400 votes, 51 percent to 46 percent.16 In contrast, John Kerry lost this county in 2004 by 13,000 votes, 56 percent to 44 percent. And even Mark Warner, Kaine's Democratic predecessor, lost Loudoun by 53 percent to 46 percent in his successful 2001 gubernatorial bid.

How did this happen? After all, it was supposed to the Republicans who really “got” voters in these kinds of areas. But it now appears that Republicans have misinterpreted their past success in these areas as evidence that these voters endorsed and wanted a stridently anti-government, socially conservative agenda. But that was never a warranted assumption, either then or now.

In reality, emerging suburban voters are tax-sensitive and concerned about government waste, but not ideologically anti-government. They tend to be religious and family-oriented, but socially moderate in comparison to rural residents. They are not anti-business, but they do hold populist attitudes toward corporate abuse and people who game the system. And they worry as much or more about public education as they do about moral values. 17

No wonder Kilgore couldn't connect in Virginia's emerging suburbs. He ran a campaign on cultural wedge issues like the death penalty and illegal immigration when emerging suburban voters were looking for solutions on education, transportation, and health care. Kaine, in contrast, spoke clearly to these voters about such solutions and famously did not hide his views on values issues, even when some of them (capital punishment) were not popular. He was open about who he was and what he truly believed in and voters rewarded his candor, since they knew who he was and what he stood for.

John Halpin is a senior fellow at the Center for American Progress Action Fund. Ruy Teixeira is a senior fellow at the Center for American Progress and The Century Foundation, as well as a Fellow of the New Politics Institute.

Coming in Part III: The authors begin to lay out their case for the future.

*** References ***

1 Unless otherwise identified, all data in this section from authors' analysis of 2000 and 2004 exit poll data.

2 Authors' analysis of Census educational attainment, population projections and CPS Voter Supplement data

3 Greenberg Quinlan Rosner Research, “Democracy Corps National Survey,” January 22-25, 2006.

4 Authors'analysis of 1992 and 1996 exit polls.

5 Authors' analysis of 1988-2004 exit polls.

6 Stan Greenberg and Matt Hogan, “Reclaiming the White Catholic Vote”, based on Democracy Corps survey of white Catholics, February, 2005.

7 Authors' analysis of 2000 and 2004 exit poll data.

8 Greenberg Quinlan Rosner Research, “Democracy Corps National Survey,” January 22-25, 2006.

9 All data in this and next three paragraphs from authors' analysis of 2000-2004 exit poll data.

10 See data on the Gallup website

11 Jeffrey M. Jones, “Many States Shift Democratic in 2005”, Gallup Organization, January, 2006.

12 “The Electoral College Moves to the Sunbelt”, Brookings Institution, May, 2005.

13 Even by the year 2032, which is as far as Frey's study extends, the changes are less than cataclysmic. See Ruy Teixeira's analysis.

14 All data in this section from Teixeira, “The Next Frontier: A New Study of Exurbia”.

15 In pure purple states Kerry actually beat Bush by a point in the emerging suburbs, 50-49--but there were only two such counties, so not much should be made of this.

16 For much more analysis of Virginia's emerging suburbs in the 2005 gubernatorial race, see Teixeira, “The Next Frontier: A New Study of Exurbia”.

17 Ibid.