2010 Could Easily Be Disastrous For Democrats

2010 Could Easily Be Disastrous For Democrats

By Sean Trende - October 1, 2009

In this week’s New Republic, Ed Kilgore writes a piece entitled "Some Revolution: Why The 2010 Election Will Not Be A Repeat Of 1994." The piece is a pretty good summary of arguments I've read concluding that 2010 won't be bad for Democrats.  These arguments are worth a closer analysis.

Kilgore concludes that the Democrats’ most likely fate for the 2010 midterm elections is a loss of about ten seats. Bear in mind that this would be well below the post-World War II average loss of 24 seats in a midterm. This is well below what non-partisan analysts like Charlie Cook are predicting. In fact, if Republicans only gained ten seats, it would be the eighth worst midterm election for an out-party since the Civil War.

I think that such minor losses are highly unlikely. But before getting into what Kilgore’s analysis misses, let me say that I really don’t purport to know what’s going to happen in 2010. It’s a bit like trying to predict who will win the 2011 Super Bowl. We can probably say with some certainty it won’t be the Detroit Lions, but once you get past perennial bottom dwellers there are too many variables to assign realistic odds (and even then, who knows; witness the turnaround of the Miami Dolphins from 2007 to 2008).

So beyond saying that Republicans will almost certainly not lose another 20 seats, we can’t do much in the way of precise predicting right now. But while the optimists may ultimately be correct, they’ll be correct in spite of his analysis, not because of it, much like a math student who gets the correct answer, but has numerous errors in his work. So let’s go through the main arguments, and see what holds up, and what doesn’t.

1. Open Seats Kilgore notes that “[t]wenty-two of the 54 [sic -- Republicans picked up 56 Democratic seats and lost four of their own for a net gain of 52 seats on election night] seats the GOP picked up [in 1994] were open. By comparison, the authoritative (and subscription-only) Cook Political Report counts only four open, Democrat-held House seats in territory that is even vaguely competitive.”

If we are asking whether Republicans can net 52 seats like they did in 1994, then this is an appropriate analysis (though we still have to note that it is relatively early in the cycle, and more seats are likely to come open if the situation doesn’t improve for Democrats, as Cook has also noted).

But if we’re just asking whether or not the Republicans can back take the 40 House seats they need in order to take back the House there’s a real problem here with this analysis (and if we’re just asking whether Republicans can pick up the 30 or so seats they’d need to bring the Democrats’ agenda to a grinding halt, it’s irrelevant).

You see, open seats are a cause of midterm loss.  But they're also an effect of a bad election shaping up.  Running a losing campaign is no fun, and an awful lot of these open seats were open in 1994 because the incumbents saw that they were likely going to lose. Rather than fight it out, these incumbents picked up their toys and went home.

Almost all of these open seats won by Republicans were in Republican-leaning territory, eight of these vacating incumbents had won with 55% of the two-party vote or less in the relatively good Democratic year of 1992, and all but two of these incumbents had voted for either the 1993 Clinton budget or the assault weapons ban. Thirteen had voted for both controversial pieces of legislation.

As I’ve noted before, similarly-situated incumbents in 1994 suffered close to a 50% mortality rate. It doesn’t require a stretch of the imagination to conclude that, even if all of these Democrats had run for re-election, Republicans would have defeated enough of them to retake the House. While the potential lack of retirements in 2010 certainly won’t help Republicans any, it won’t hold them to single digits, either.

Remember what tsunami midterms are like.  In 1994, Mike Synar's seat was open because he was defeated in a primary election by a retired high school principal who spent less than $20,000 on his campaign.  Synar would not have survived the general election even if he had made it that far.  Democrats like Earl Hutto in Pensacola, Marilyn Lloyd of Chattanooga, and William Hughes of South Jersey weren't coming back to Congress, regardless of whether they ran or not (much like Jim Nussle and Bob Beauprez in 2006).

2. “The Great Ideological Sort Out”

Kilgore argues:

The 1994 election was the high-water mark of the great ideological sorting that occurred between the two parties. That made the environment particularly harsh for southern Democrats, as well as those in the Midwest and Rocky Mountain West, where many ancestral attachments to the Donkey Party came unmoored. . . . Nothing similar to those handicaps exists today. The ideological filtering of the parties is long over; any genuine conservative Democrats or liberal Republicans left in the electorate clearly have reasons for retaining their loyalties, which will be difficult to erode.

The problem with this was identified by Kilgore earlier in his piece, where he explained that “if you apply the Partisan Voting Index, (PVI), which compares a district’s prior presidential results to national averages, you find that there are 66 Democrats in districts with a Republican PVI and only 15 Republicans in districts with a Democratic PVI--a similar situation to the 79 Democrats in Republican districts in 1994.”

To the extent there was a “Great Ideological Sort Out” in the 1990s, it largely reversed itself in 2006 and 2008, when the Democrats decided to back pro-life, pro-gun Democrats in places like NC-11, AL-02, and other conservative districts. In fact, most Southern Congressional delegations are almost as heavily Democratic today as they were right before the 1994 elections; Democrats presently have one fewer seat than in 1994 in Alabama, Georgia, Louisiana, South Carolina, Tennessee, and Virginia, have broken even in North Carolina and Florida, and gained a seat in Arkansas (Republicans have picked up a few more seats in these states due to reapportionment).

The end result is that, as Kilgore notes, there are almost as many Democrats in Republican-leaning districts today as there were in 1994. And there’s no great trick to eroding the voters' loyalties to these Democrats here; many are newly elected and were elected because of incompetent or corrupt Republican incumbents, because they weren’t being pressured by their leadership to vote for highly unpopular (in conservative districts) measures like cap and trade or health care reform, or because of unusually high African American and/or youth turnout in 2008 (on which more later).

3. Ye Olde Emerging Democratic Majority. I’m on record expressing great skepticism about the conglomeration of demographic observations that have come to be known as the “Emerging Democratic Majority Theory.” Not that I believe there’s a “Permanent Republican Majority,” emerging or otherwise.  I just don’t believe in permanent majorities. I won’t rehash everything I’ve written; you can click here to get the basic argument; the article contains lots of links to other things Jay Cost and I have written on the subject.

As for the argument that there is not a “single discernible long-term trend that favors the Republican Party,” two observations are in order. First, you can almost always make this argument when you’ve beaten down the other party. If you look at the exit polls from 1988 (or 2004), it's hard to make a case for the Democrats returning to power anytime soon. This is why highly respected political scientists penned well-received articles in top-tier scholarly journals in the late 1980s and early 1990s about what was known as the “Republican Lock On The Electoral College.” We know how that turned out (Jay and I have both written at greater length on the longer-term trends).

Second, long term trends are largely irrelevant here. We aren’t so much concerned with what the electorate is going to look like in ten years as we are what it’s going to look like in thirteen months. And in thirteen months, the electorate is almost certainly going to be less favorable to the Democrats than it was in 2008.

4. Short Term Trends. As Kilgore points out, we can be fairly certain of one thing: The 2010 electorate is going to be older than the 2008 electorate. And as Kilgore notes, these voters are not particularly crazy about the Democrats right now. Moreover, a grayed electorate robs Democrats of one of Obama’s strongest constituencies in 2008: The young. Kilgore notes that this shift alone could cost Democrats ten seats if the Democrats are as popular in 2010 as they were in 2008. They’re substantially less popular today than they were in 2008, which is why I’m perplexed by Kilgore’s prediction of a ten seat loss.

There’s another problem: The midterm electorate is likely to be whiter than in 2008. Obama brought boatloads of marginal voters to the polls, who were excited to vote for the first African American President. Will the voters in VA-02 who turned out to vote for Obama and happened to also pull the lever for Glenn Nye now turn out to vote just for Nye? Perhaps, but I think it is highly unlikely. The more likely scenario is that we can expect to see a larger-than-usual dropoff in minority turnout from a general election than we are accustomed to seeing. Again, this weakens the Democratic performance.

Finally, Kilgore points to two articles by Alan Abramowitz. The first argues that our electorate is progressively less white and that Republicans will need to win 60% of the white vote to take the House, while the second notes that Obama is holding serve with his core constituencies. There are all manner of problems with these Abramowitz articles, problems which are beyond the scope of this article. Suffice it to say that (1) in good Republican years, Republicans have typically come close to winning the 60% of the white vote that Abramowitz thinks they need to claim the House outright; (2) the dropoff in minority participation from 2008 is likely to be more substantial than Abramowitz’s model allows; and (3) the Gallup polling data in the second article is from July of 2008, when Obama’s approval rating was about seven points higher than it was today.

At any rate, the Democrats’ concern isn’t that core constituencies will turn against them; these voters are often packed into heavily Democratic districts, and they aren’t likely to abandon Obama. The real concern is that independent voters – who went Democratic by eight points in 2008 and who are more concentrated in swing districts – will abandon the Democrats. Obama’s approval rating is presently 43 percent among this group.

5. The Senate Kilgore is knocking down a bit of a straw man here; I don’t think anyone is predicting that Republicans can take back the Senate. A pickup of 11 seats hasn’t happened since the Republicans took 12 in 1980, and this cycle’s playing field is not particularly favorable to the Republicans. Republican gains on the order of 5-6 seats are possible, but the most likely scenario is something of a wash. A lot will depend on whether Rep. Mike Castle opts to run for the Senate, whether North Dakota Governor John Hoeven decides to challenge Byron Dorgan, and whether George Pataki challenges Kirsten Gillibrand. Then again, no one but the hardiest partisan thought Democrats would take back the Senate until late 2006; few thought they would just miss a filibuster-proof majority in 2008 until summertime.

6. Stabilizing Approval. It is true that Obama’s approval ratings have stabilized – he doesn’t seem ready to go below 50% in Gallup. But his approval ratings right now are about where Reagan’s and Clinton’s were in September of their terms. Both of them eventually got shellacked in their first midterm elections. Indeed, if Obama’s approval merely holds steady at 52%, it would probably still be a rough midterm for the Democrats.

7. The Economy Missing from Kilgore’s analysis entirely is, strangely, what is likely to be the most important factor in the 2010 elections: the economy. In 1994, the economy was sluggish, but had been recovering for four years, and it still proved to be a drag on Democrats. I don’t think anyone really has a clue what the public’s perception of the economy will be in 2010. Reading Realclearmarkets day-to-day, you’re just as likely to find someone predicting Morning In America II as you are someone predicting gloom-and-doom.

So let’s say this: If it is apparent to the average American by the summer of 2010 that we are in the midst of a robust recovery, then I think that the Democrats’ losses will be very limited. We could even see minor gains. But if we’re seeing double digit unemployment numbers that are only beginning to crest or come down (or worse still, are still going up), the Democrats are going to have an absolute debacle on their hands. Every Democrat in a red district that voted for the stimulus package, which is almost all of them, will have to face charges that they voted for a trillion dollars in spending with nothing to show for it. Many will also have to defend votes on cap-and-trade, a health care proposal that isn’t particularly popular in red states, and other votes yet to be determined (immigration reform?).

If that’s the playing field on which the 2010 elections are fought, then 2010 won’t look like 1994. It will look worse.

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.

Early Forecast: More Gridlock
Ruth Marcus · November 9, 2014
Progressives Don't Need Washington All That Much
Froma Harrop · November 6, 2014

Sean Trende

Author Archive

Follow Real Clear Politics

Latest On Twitter