I’ve been on deadline with a conference paper, so am commenting somewhat belatedly on last Tuesday’s election results, focusing on the highly publicized Senate primary races in Colorado and Connecticut. As you know, in Colorado, incumbent Democrat Michael (“One T”) Bennet defeated challenger Andrew Romanoff by a relatively comfortable 54%-46%, a margin somewhat larger than what many pre-election forecasts suggested. You will recall that Bennet was appointed to the Senate in 2009 to fill the seat of Ken Salazar, who became Obama’s Secretary of the Interior. In the Republican primary, local District Attorney Ken (“One Tea Party”) Buck topped former Lieutenant Governor Jane Norton, 52-48. Buck has Tea Party backing, while Norton was viewed more favorably by the Republican “establishment”.
In Connecticut, meanwhile, former WWE wrestling executive Linda McMahon won the Republican Senate primary over Representative Rob Simmons and the Tea Party candidate Peter Schiff. McMahon now faces Democratic Attorney General Dick “I might have fought in Vietnam” Blumenthal in November to fill the seat of retiring Senator Chris Dodd. Despite some biography-related gaffes, Blumenthal is up by double digits over McMahon in early polls, although his lead is shrinking.
For the most part, the mainstream media painted this as a good night for Democrats, and for President Obama – indeed that was the exact Politico headline in their coverage of the results: “Primary Night Yields Good News for President Obama and Democrats”. A bit more cautiously, the New York Times headline reads, “Incumbent Backed by Obama Wins Colorado Primary” thus associating Obama with the outcome without necessarily crediting him for it.
Each story went on to suggest that the results indicated that the prevailing narrative, which shows Republicans making great gains come November, might need to be reconsidered. In Politico, John Harris writes, “Republicans, meanwhile, were left with several new reasons to wonder whether all the favorable national trends showing in the polls are enough to overcome local candidates who are inspiring little confidence about their readiness for the general election 12 weeks from now.” In the Times’ coverage of the results, Kirk Johnson writes, “The predictions of doom for incumbents and establishment candidates this campaign season are proving to be more complex in the real world.”
Well, yes and no. It is true that leading Democrats, most notably President Obama, as well as Vice President Joe Biden and others, campaigned heavily on Bennet’s behalf. Had Bennet lost, the prevailing media narrative almost certainly would have been that still another candidate backed by Obama went down to defeat. So, given the alternative, the Colorado outcome certainly was good news for Obama. However, do the results “complicate” our understanding of midterm elections? Do they potentially herald a strengthening of Democrat support heading into November?
No. In fact, the Colorado and Connecticut Senate primary outcomes do nothing to change my belief that the underlying electoral fundamentals are still pointing to strong Republican gains come November. Let’s start with Colorado. Keep in mind that it highly unusual for candidate challenging an incumbent in a primary to win 46% of the vote. This despite the fact that Bennet spent about $1.9 million on advertising to Romanoff’s roughly $757,000, much of it in a ten-day advertising blitz near the end of the campaign when polls indicated Romanoff might pull off the upset. All told, Bennet spent almost $6 million on the primary campaign, compared to less than $2 million for Romanoff. Bennet officials admitted that they had not anticipated the need to spend this much on advertising but they took no chances when polling suggested they were in a tight race. In fact, total spending in these Colorado primaries broke the state record.
Note also that Bennet is not running like a typical incumbent; despite his backing from the Democrat establishment in the primary, in his victory speech he immediately pivoted to portray himself as a political outsider not yet tainted by Washington politics. When asked, he made no commitment to having the President come to Colorado to campaign for him in the general election. This despite that fact that polling data has Bennet in a dead heat with the Republican Buck.
Note also that although turnout was up in both party primaries, due to the contested races, it was up more in the Republican primary, where 45% of registered Republicans voted compared to about 40% of registered Democrats. According to state records, there are 802,000 registered Democrats in Colorado, 830,000 Republicans, and 730,000 independents. Bottom line: it’s not clear to me that Bennet has a markedly better chance than did Romanoff to beat Buck. Polls showed both Democrats in a dead heat with the Republican.
In Connecticut, meanwhile, analysts pointed to the fact that McMahon won less than 50% of the vote in a three-way race as evidence that she’s not a strong candidate. While Blumenthal must certainly be considered the front-runner in this Democrat-leaning state, I would be cautious about adopting the interpretation that Tuesday’s results signify McMahon’s weakness. First, in contrast to the Colorado races, turnout in the Connecticut Republican primary was quite low, in part because it was vacation time and also because McMahon already had the backing of the party establishment coming out of the Republican convention earlier this year. Second, keep in mind that independents don’t vote in the Connecticut primary, and they constitute about 43% of Connecticut voters. How many of them will support McMahon over Blumenthal? Third, how likely is it that Simmons’ and Schiff’s supporters are going to vote for Blumenthal over McMahon? Fourth, Democrats are pouncing on McMahon’s wrestling background as a sign that she’s not electable. Jesse Ventura anyone? Finally, keep in mind that McMahon has very deep pockets. Given all these factors, I expect this race to tighten come November.
In short, you should not let the headlines from Tuesday’s races fool you into thinking the underlying electoral fundamentals suddenly shifted. I’ve just begun crunching some of the forecast numbers and will have much more to say about them in the coming months. But the bottom line is that, unless these midterm Senate races are driven by an entirely different set of dynamics than previous midterms, the situation is still not good for Democrats.
As evidence, some of you may remember my warning about the recent results of Gallup’s generic ballot question which asks voters whether they will support the Republican or Democrat congressional candidate come November. After weeks of favoring the Republicans, the results flipped during a two-week period in mid-July to put Democrats in the lead. Progressive bloggers hoped it heralded a reversal of political fortunes. In contrast, I cautioned (along with others like Mark Blumenthal) that the change might simply reflect the typical random variation associated with public opinion sampling. Here’s the most recent Gallup results:
As you can see, Republicans are back in the lead, and by the largest margin (tied) held by either party since Gallup began tracking the generic ballot in March. This does not mean that voters have suddenly switched back to supporting Republicans. Instead, it is more likely that the underlying voting dynamics have remained remarkably stable and that July’s results were, as I suggested, simply random variation due to sampling. Since March Republicans have been tied or ahead of Democrats in almost every poll except for that two-week period in July, and they continue to maintain an “enthusiasm” advantage in terms of their likelihood to vote in November.
Bottom line? Tuesday’s “good news” headlines notwithstanding, the election results in Colorado and Connecticut do not complicate or even change the political science narrative at all: the fundamentals are no more favorable to Democrats or Obama today than they were last week.