In looking at any midterm election, observers should take into account several important considerations. First, will it be a “normal” election? That is, will the type described by the late Tip O’Neill as one in which “all politics is local”—the kind of election in which the natural voting patterns of that state or district, as well as the relative strengths of the candidates, campaigns, resources—determine the outcome? This way of approaching elections is what I call “micro-political.”
Or will it be “macro-political”? That is, will it be a wave election in which an invisible force seems to be pushing the candidates of one party forward while holding the candidates of the other back—one in which the natural, unique circumstances in each race are still important, but are then shifted at least several notches in favor of one party at the expense of the other? Democrats benefited from elections like this during midterms in 1958, 1974, 1982, and 2006, as well as in presidential years 1964 and 2008. Republicans were the beneficiary in the 1966, 1994, and 2010 midterms, plus the 1980 presidential election. In a macro-political election, all politics is decidedly not local. And it’s important to note that four such midterms were held halfway through a party’s second term in the White House—1958, 1966, 1974, and 2006.
When thinking about the 2014 midterm, the obvious question is whether this election will be a continuation of the dynamics of 2012, in which Republicans continue to have problems with minority, female, younger, and moderate voters, or whether the historic dynamic of voters showing displeasure with presidents halfway through their second term reoccurs. When the latter happens, it usually brings about substantial losses in the House or Senate, or both. Should Democrats face a backlash during the next phase of implementation of the Affordable Care Act, those sentiments would feed into this dynamic as well.
At The Cook Political Report, we have identified five metrics to help ascertain whether either party is facing an undertow, although we could obviously note dozens more as well. These diagnostic indicators are the president’s job-approval ratings; public attitudes toward the economy as measured by consumer-confidence ratings; attitudes toward the Affordable Care Act; the favorable/unfavorable ratings for each party; and the generic congressional ballot test (taking into account that, for some inexplicable reason, it has about a 3-point bias toward Democrats).
Using the president’s approval rating, even in a midterm election, is useful because, for better or worse, a president is the face of his party. Although voters rarely reward a party for having a popular president, they are quick to register their displeasure with a chief executive by voting against the candidates of that party in a midterm election.
How the economy is faring and, more important, how the public feels the economy is doing make a huge difference in whether voters are happy with a president. While voters are more likely to punish when unhappy than reward when happy, they quite extraordinarily didn’t punish Democrats in 1998 during the midterm election of President Clinton’s second term. Although they didn’t approve of Clinton’s behavior in the Monica Lewinsky affair, the economy was strong and growing, and they disapproved of Republicans’ impeachment efforts.
In terms of the ACA, public attitudes have been pretty stable for the past two years. More Americans disapprove rather than approve of the health care law, though not by a particularly wide margin. If those proportions start to significantly change in one direction or another—for example, if a backlash develops against rising health care premiums, deductibles, and co-pays; or if employees are asked to shoulder even more of their health care costs than before—this could become a major factor in 2014.
Currently, the Democratic Party’s favorable/unfavorable ratings are not impressive, but at least its favorables are somewhat higher than its unfavorables. On the other hand, the Republican Party’s unfavorable ratings are, in some polls, at record levels, reaching 60 percent unfavorable. Whether Republicans have repaired enough of the damage to their brand since 2012 is clearly important if they are to take advantage of any problems that Democrats may have, and the party’s favorable/unfavorable ratings will reflect that. If any of President Obama’s problems are starting to hurt Democrats, we will see this effect in their favorability ratings.
Finally, there is the generic congressional ballot test. To be sure, the generic test is not good at helping estimate how many seats a party may gain or lose; however, it is useful in signaling which way the wind is blowing. To a certain extent, it can help determine whether it is blowing mildly, moderately, or heavily in that particular direction. Historically, it has often been the case that one has to knock about 3 points off the Democratic number to get a true, yet very rough, reading of how the popular vote for the House may go.
This is not to say that other factors aren’t important. My Cook Political Report colleague David Wasserman has written extensively about how the makeup of the current Democratic Party’s base is uniquely vulnerable in midterm elections, because many highly variable 2012 voters—those who are least likely to show up in the 2014 midterm election—are disproportionately Democratic.
But for determining where this election is headed, these five indicators are probably the best to watch over the next 17 months.