11/02/2010 11:56 am ET | Updated May 25, 2011

Finding a Better Way to Talk About the US Electorate in 2010

The Whole Idea of Base/Swing Needs to Be Tossed Out

Here is a mental image of the electorate that is often used, which goes something like Dem Base/Left -- Middle/Moderate/Independent -- Right/GOP Base. One hears commentary as if the size of the two party's bases are static, and the only change that happens in elections is that swing group of "moderate, independents in the middle."

The problem with this mental image, or idea of the American electorate, is that it simply does not adequately capture the complexity of what has or will happen in an electorate that is going through profound demographic change every two years, where independents are not monolithic or even moderate, and the bases of the two parties are not at all static. Lets drill down a bit deeper on all this:

Demographic Change -- America is undergoing one of its most profound demographic transformations in all of its history. Central to this change is the vast waves of immigration from Asian and Latin Americans to the US over the last 45 years, which is transforming America from a 90/10 white/black society to one which will be majority minority in about 30 years. The second change is the entry into the electorate of the Millennial Generation, the largest generation in American history. Every two years the percentage of the US population made up of Asian and Latinos and Millennials continue to grow, meaning that the power of models of previous voting behavior in fast-changing communities will be of much less predictive use, and that a "new politics" is being born in communities across the country as these populations alter the composition of local, statewide and national electorates.

As an illustration of the rapidity of these changes, from 2004 to 2008 the share of the electorate that is Hispanic grew in CO from eight to 13 percent, in NM from 32 to 41 percent and in NV 10 to 15 percent. These are huge changes in such a short period of time.

Independents and Party Base -- One hears extraordinary things on television -- "the country is a third Democrat, a third independent, and a third Republican"; "given that there are more conservatives than liberals in the electorate, the Republican base is larger than the Democratic base"; "if you win independents voters and moderates you win," and so on. The problem with all these assessments is that they require the people who are self-described independents in each election to be a static bunch, a sort of moderating guardianship over the passions of partisans in both parties. But independents are anything but a static group. Those who are independent change in each election cycle as the electorate changes, and the relative strength of the two parties change. In 2006 and 2008 the percentage of the electorate that was independent was 26 percent. Those who were Democrat or Republican were 74 percent, or three times the number. Which means, by the way, that the percentage of each party's vote made up of partisans was 2.5:1 or three times the number of independents, which is why any party that plays to the independents over the needs of its own partisans is doomed to failure. Three quarters of each party's electoral coalition is made of partisans, and they need to be assuaged, spoken to, provided for as much as the indies in a coalition. Perhaps even more given how many of them that there.

When I got into politics in the 1980s there was this constant refrain that all voter contact money needed to be spent on the undecideds in each election, for "where else would the base voters go?" Well what we have seen in this past decade is that voter-contact programs can alter the make up of the partisan base of each party. Rove spent a great deal of money making the electorate much more Republican in his reelect period, which left 2004 an electorate that was 37D/37R and 26I. The percentage of R voters from 2000 to 2004 soared, and the number of independents dropped. This allowed Bush to win in 2004 despite losing independents and moderates, something that many believed was a statistical impossibility. This is because he identified and turnout out latest Rs, which in turn altered the composition of his party's base and the electorate as a whole.

As I showed in this analysis, more votes shifted from 2004 to 2006 inside partisans than among independents. Which makes sense as partisans have outnumbered independents by 3:1 in recent elections. There are just a lot more of them than indies, and much more analysis needs to look inside the bases of the two parties to understand what is really going on in modern elections.

All of these observations led NDN/NPI to do two massive studies of the state of the coalitions of both political parties earlier this year. You can find the first and most important one here. What we found in our first highly influential study in March was in fact that Democrats were not as interested in voting as in previous years, and we argued that if unaddressed this relative lack of interest could end up spelling big trouble for the Democrats this fall.

As Mike Hais has argued on these pages, the Democrats needed to use this new coalition they built over the past few elections or risk losing it. There was no going back to a pre-Obama Democratic coalition. Democrats had won in 2006 and 2008 through the passion and votes of a very different coalition than any Democratic coalition that had come before, and it needed to be replicated in 2010 for Democrats to win.

Some reports indicate that Democrats may match 2006 turnout numbers, but it is both the loss of independents and a very high midterm turnout of Rs -- perhaps depressed in recent years due to the Bush presidency -- that will explain the Rs strong showing tonight. Perhaps. But is critical now is that we are seeing in some reporting this much more nuanced understanding of what it takes to win these days -- an excited base, and winning the swing.

The Necessity of Excited Partisans -- At times this year one heard noises coming from the Obama White House complaining about the "professional left" or other such terms to describe unhappy Democratic partisans. I never really understood this for in the modern post-broadcast age of politics, an excited base becomes a much greater necessity than in the broadcast age of politics. First, as we covered earlier, partisans represent about 3/4s of the coalitions of the two parties, and if you factor in the independents who always vote one way or the other it is actually much higher Second, in the 20th century broadcast model of politics, all the money a campaign spent was designed to speak to a very narrow group -- the swing. There was no method in this old model of having separate conversations with different audiences, or more narrow conversations with the overwhelming majority of each party's coalition which were already with you.

The modern conservative/Republican machine moved beyond this broadcast model much earlier than the Democratics/center-left. This is perhaps best exemplified by the two political advisers to the last two presidents, Karl Rove and David Axelrod. Rove came up through the direct mail business, which was a one to one, narrow cast business. It involved databases, modeling, interactivity, and helped him and the Rs understand that they were in effect building a coalition made up of individuals or groups. The broadcast model of politics, which Axelrod has come up in, sees the electorate differently. It is not interactive, and communications is two broad swatches of the electorate, like women 18-34 or Latinos. This is no dig on Axelrod, but Democrats remained a broadcast, non-data driven party for much longer than the Republicans, and it was really only been in the last five years or so that Democrats have begun to make the transition to a much more nuanced understanding of their coalition than base/swing. If you are a Democratic political operative of Axelrod's generation, you grew up in a politics where Democrats did not spend money or time talking to partisans. Urban-turnout machines did that. Your job was to win the swing with a barrage of television ads in the final weeks of a campaign. So there is still lingering in folks my age in Democratic politics a sense that the big game is those indies, and that there is little knowledge or experience -- and thus interest or facility -- in working with or talking to those already with you. It is an "old dogs, new tricks" problem (Given Axelrod's leadership of Obama's team these last few years it is clear that he is part of a new vanguard of thinkers on these matters).

The reason this matters so much is that in the still young, post-broadcast, post Dean/Trippi, bottom-up, people-led era of American politics the entire theory of Democratic campaigns requires the passionate involvement of everyday people, everyday people who are partisan Democrats of course. It is their grassroots contributions, their canvassing and phone banking, their spamming all their friends on Facebook, email and Twitter that is the model of a Democratic post broadcast age campaign. Without this passion of partisans this new internet/social networking model collapses, and then creates a vicious cycle of less well funded and robust campaigns reaching fewer people, etc. There is now a much greater cost to not keeping your base happy, while there is also a much greater return in doing so. According to this recent Pew study, the excited Republican base this cycle and the money and infrastructure it generated allowed the Republicans to match or beat the Democrats in all forms of voter contact, something of course that has very real political consequences in an election where the House may be decided by 1-2 percentage points.

So this is also a lesson of 2010 for Democrats -- the "new politics" of the 21st century requires an engaged and excited base, as much as it does require the winning over of those not with you. Commentators over the next few days who try to argue that the key to 2010 was what happened with the independents are just not in touch with the reality of modern American politics. From where I sit, the Republicans will do well tonight because they won the independents, fired up their base, executed the blocking and tackling of campaigns well, while Democrats have managed to both lose the independents and been unable to adequately engage their own partisans. The tragedy for the Democrats this year is that in the last few years there have been more Democrats in the electorate than there has been in perhaps 40 or 70 years, which means the upside of engaging their own partisans is much greater today than it has been in the adult lifetime of just about everyone involved in American politics today.

This is long enough now. What's the point here? Our pundit class needs to work hard these next few days to be precise and accurate about these complex elections in a volatile and unpredictable era. There are lingering mental images of our politics in common use which reduce this complexity to a level of simplicity no longer descriptive or accurate, and it is critical we all work to reject those frames for ones that capture the true dynamism of American politics today.

For reference also see:

-- My work on what is causing the central volatility in American politics today.

-- Some recent thoughts on how the changing demographics of the two parties' coalitions will explain some of the regional results we will see tonight.

-- This great essay by Mike Hais on how many people are many true independents there are in the electorate.

Crossposted from