Assessing The Obama Coalition, Part II

Assessing The Obama Coalition, Part II

By Sean Trende - November 19, 2010

In my last column, I spoke broadly about the difficulties Democrats faced in 2010. In particular, I focused on the weaknesses in the Democratic voting coalition. That coalition, which made possible the Democratic gains in 2006 and elected Obama as President in 2008, frayed badly over the course of the past two years. Rural voters, suburban voters, and white working class voters all moved substantially against the Democratic Party. The only Democratic constituencies to remain fully with the Party were racial minorities and progressives.

In this article, we'll explore how that frayed Democratic coalition has affected the President's electoral coalition. I'll re-emphasize upfront that this isn't meant as a prediction - it is way too early for that. We're merely taking a look at the President's electoral coalition to see where things stand and to identify any particular causes for concern for 2012.

First, a bit of background. For all of the talk of 2008 as a realigning election, the truth of the matter is that Obama's electoral coalition wasn't really all that different from earlier Democratic electoral coalitions. The overall map just shifted leftward, along with the mood of the country. Unlike the 2010 midterms, 2008 was a surprisingly even wave.

Consider the following map. It demonstrates the relative change in the parties' standing from 2004 to 2008. In other words: Barack Obama received 53 percent of the vote, five points higher than John Kerry's 48 percent of the vote. In the map below, a state that moved a little more than five points toward the Democrats -- 7-8 points -- is colored light blue, while a 8-9 point shift is coded a shade darker blue, and so forth. Likewise, a state that only moved 1-2 points toward Obama is shaded light red, and so forth:

Note how little change there is. Democrats performed a little better, relatively speaking, in the Mountain West and northern Great Plains, plus Indiana and Virginia, while the Republicans performed a little bit better in the "Jacksonian diaspora" from West Virginia over to Oklahoma. But overall, there was very little fundamental remaking of the electoral map.

So we're really in the same basic red/blue divide we've been in for a decade. If things get bad enough for one party, the other party can put substantially bluer or redder states into play (like Indiana or Montana for Democrats, or Washington or Maine for Republicans). But the list of states that are likely to get either party to 270 electoral votes has remained roughly the same.

So let's look at the results from 2010 in races that have finished within four points of the national average in 2004 and 2008 (in other words, those states with PVIs that are between R+4 and D+4). The columns in the following table are fairly self-explanatory. The first two are the state and its PVI, the two to the right denote the Democratic percent of the two-party vote in any senate or gubernatorial contest in 2010 (I've omitted the quirky 3-way races in Colorado and Florida). The next column shows the average Democratic performance in downticket statewide races, such as secretary of state, attorney general, etc

The next column shows the percentage of the seats in the state legislatures that the Democrats lost in 2010, while the second column from the right shows the total share of the statewide congressional vote won by Democratic candidates. Finally, the rightmost column shows Obama's average approval rating in recent polls. If there was no appropriate data, I simply put an asterisk in the column.

As you can see, the Democrats' position in the five Republican-leaning swing states is very shaky right now. Roy Blunt won by double digits, over a Carnahan no less, in Missouri. Republicans picked up 3 House seats in Virginia, and missed a 4th by a whisker (or a traffic jam on I-95). Marco Rubio won a near-majority in Florida, while a highly problematic Republican won the governorship. A tsunami hit in Ohio, where only one Democrat at the statewide or federal level cleared 60 percent of the vote. North Carolina gave Richard Burr - once considered vulnerable - the highest percentage for any Republican Senate candidate in its history.

Among these five Republican-leaning states, the President is deeply underwater in his approval rating, and the Democrats topped 45 percent of the statewide congressional vote only in Virginia. Statehouse losses were massive (except in Florida, where the GOP already dominated the legislature), and downticket Democrats were obliterated.

In fact, Democrats didn't win a single statewide race any of these five states.

After reapportionment, losing these states, Indiana, and the electoral vote Obama collected out of Nebraska would leave the President with a slim 273-to-265 lead over the Republican candidate. Flipping any of the remaining states would give Republicans at worst a 269-269 tie.*

And flipping one of the remaining states doesn't look all that difficult. Democrats look especially weak in New Hampshire (which now has more Republicans in its legislature than at any point since 1900), Pennsylvania, Wisconsin and Michigan. Recent Public Policy Polling results showing the President trailing "generic Republican" in almost all of these states: North Carolina (by fifteen), Florida (by fourteen), Ohio (by fifteen), Colorado (by five), Nevada (by nine), Minnesota (by two), New Hampshire (by fourteen), Pennsylvania (by ten), and Wisconsin (by nine).

The root of the problem for the President is that in most of these swing states there aren't enough minority voters and progressives to allow a Democrat to win without substantial assistance from white working class or suburban voters, especially since rural voters have been pushed solidly into the Republican camp (at least for now). The more the President alienates working class and suburban voters, the harder it becomes for him to win.

The following table illustrates this by showing the percentage of the electorate in swing states that was non-white, liberal, and from suburban or rural centers, according to the 2010 exit polls (several states didn't have exit polling data, so not all swing states are included). It also includes a pair of solidly "blue" states (CA and NY) for comparison. There were no exit polls that asked a good question about whether voters were "blue collar" or not so I included the proportion of the state classified as blue collar according to the census bureau.

The table is sorted by comparing the ratio of non-whites and liberals in the state to rural/suburban voters and blue collar voters. In other words, at the top of the table, we find states with a very high proportion of demographic groups among whom the President is struggling, compared to those groups where he's performing well. Toward the bottom of the table, we find states where there are almost as many voters in pro-Obama demographic groups as in the groups where the President performs poorly. Obviously there is considerable overlap here: It is, after all, possible to be a non-white, liberal, blue collar voter living in suburbia. It just gives us a rough metric for measuring the President's coalition in swing states.

This rough metric actually works pretty well. As you can see, in the bluer states like California and New York, there are almost enough non-white and liberal voters to overcome even a complete Democratic wipeout among rural/suburban voters and blue collar voters. Note also that the demographics of swing states where Democrats fared pretty well this year, such as Oregon and Nevada, are similar to the demographics in these two blue states.

On the other hand, the states where non-whites and liberal make up a smaller portion of the electorate, and where rural/suburban voters and blue collar voters are more plentiful, such as New Hampshire, Iowa, Pennsylvania, and Ohio, are the states where Democrats got their worst shellacking. The only real outlier here is Wisconsin, where Democrats performed quite a bit worse than its demographics would suggest.

Even if the President were to recreate the 2008 electorate - and as I mentioned last column, it is by no means certain that he'll be able to - it wouldn't be enough for Democrats to fix this problem. A CNN poll released immediately after the midterms showed Obama trailing Mitt Romney among registered voters. Even after re-weighting the state polls to a 2008 electorate, Obama was tied in Wisconsin and Minnesota, and trailed in Colorado, New Hampshire, Nevada, Ohio, North Carolina, Pennsylvania and Florida. Clearly this isn't just a matter of getting 2008 voters to turn out. He has to convince some 2010 voters to re-evaluate their voting selections.

The President still has two important things going for him, however. The first is time. There is ample time for perceptions of the economy to turn around, and for his job approval to increase. Second, he won't be running against "generic Republican." All of the major Republican candidates have flaws, and the results from 2010 in Colorado and Nevada show that a Republican candidate with serious enough flaws can still lose in swing states, even to badly damaged incumbents.

But the President's electoral coalition is showing serious signs of weakness right now. If the election were held today, he would almost certainly lose, probably by a margin in the Electoral College similar to that by which he won in 2008. If he wants to be re-elected, he has some serious work to do reaching out to the suburban and working class voters that Democrats shed in 2010. Simply maxing out the vote among minority and progressive voters won't be enough to win.


*In that unlikely scenario the Republican would probably win; if no candidate gets a majority then each state delegation gets a single vote. With Republicans presently having a 33-16 lead in state delegations, (with one tie), as well as control of redistricting, it seems unlikely that Democrats will control a majority of delegations after the 2012 elections.

Sean Trende is senior elections analyst for RealClearPolitics. He is a co-author of the 2014 Almanac of American Politics and author of The Lost Majority. He can be reached at Follow him on Twitter @SeanTrende.

A President Who Is Hearing Things
Richard Benedetto · November 12, 2014
Obama Is No Clinton
Larry Elder · November 13, 2014
Bret Stephens' Call for Robust U.S. Foreign Policy
Peter Berkowitz · November 16, 2014

Sean Trende

Author Archive

Follow Real Clear Politics

Latest On Twitter