Science: Moderation Did Not Lose McCain The Election, And Is An Asset For Electability – And That’s Why The “Establishment” Acts Like It Is
There is a belief on the right that John McCain lost because he was too liberal and conservatives chose to stay home on election day. Some who hold this belief threaten to go Scozzafava on the GOP and refuse to work to stop Obama if Romney (or in fewer cases, even Gingrich, Perry, or assorted others) gets the nomination.
This is a comforting view for many conservatives – it means we didn’t actually lose the 2008 election. It is the cousin of the liberals’ belief that conservatives won the 2004 election only because we were better wordsmiths, a view that was equally unproductive for any other purpose than making George Lakoff famous.
The evidence does not support the claim that McCain lost because conservatives did not show up at the polls. Considering the economic conditions, the unpopularity of the incumbent Republican president, and Obama’s well-funded and well-run campaign, it would have been extremely difficult for a Republican to win no matter they were.
However, even putting that aside, the evidence also quite specifically is not consistent with the claim that conservative discontent could have made the difference. Depressed conservative turnout did not happen.
A Washington Post poll taken the day of the election of registered voters found 34% conservatives, 21% liberals, and 43% moderates. The last 15 daily polls the Washington Post conducted before the election averaged 21.4% liberal, 41.2% moderate, and 34.2% conservative.
The national exit poll showed that those who voted were 34% conservative, 44% moderate, and 22% liberal – pretty much mirroring the registered voters. This is actually remarkable, considering the advantages that Obama had over McCain in funds and underlying economic and political conditions. If anything, we should have expected turnout to skew against conservatives.
It is possible that the Obama campaign may have done a better job of registering voters, although that cannot account for that large a percentage of the electorate. Also, the difference between registered voters and adults is tiny, and surely the number of disgruntled conservatives who refused to vote for McCain make up a fairly minuscule proportion of the not-registered-to-vote population.
Other factors militated against conservative turnout as well. The fact that the Obama campaign had a vastly better funded operation than the McCain campaign (mostly due to the desire to give to winners and the support bonus he garnered from Bush’s unpopularity and the bad economy), and the differential effects of economic impacts on the parties (i.e. the effect-via-registration-and-turnout of the economic conditions that mostly determine presidential elections) are ordinarily worth quite a lot. These issues go hand-in-hand, and while differential campaign effects can swing especially close elections , they probably are mostly included (indirectly) in economics-based forecasting models since predictors of support in turn predict campaign resources. This is also why McCain’s moderation losing him the race by suppressing volunteering is also implausible, along with the fact that there should be some evidence of depressed conservative turnout under such circumstances as well.
While 20% of conservatives voted for Obama, it is unlikely that these included many conservative purists. It is well established that far more people call themselves conservative than actually are – the vast majority of these were real CINO’s, not protest voters, and even in 2004, a with a vastly weaker (electorally) Democratic candidate , 15% of “conservatives” voted for John Kerry.
In short, the ideological breakdown of the 2008 electorate shows no depressed conservative turnout and is overdetermined as it is – there is no known unexplained difference that could be plausibly attributed to McCain’s moderation, and in fact conservative turnout may even be better than expected under the circumstances. This is not to say that no conservative stayed home, but this analysis has also not considered the non-zero quantity of moderates that turned out for McCain but might not have voted for a True Conservative.
But enough about that one race. The current state of science is that looking at the full historical picture, moderation is an electoral asset.
The Republican field is now probably narrowed down to the two candidates with the most heterodox records (other than John “We Are The 1%, Literally” Huntsman and Magneto). True, Romney owns his record of moderation more than Gingrich. Gingrich has been in politics so long that commentators or opponents can seek to portray him as the far-right mean ol’ “Gingrinch Who Stole Christmas” or Scozzafava and Pelosi’s boy-toy, depending on the biases and motivations of the accusing rhetorician.
Considering the current landscape, it is important to emphasize that the scientific consensus on presidential elections is that moderation is an asset – not a liability – for electability.
There are numerous statistical models of that predict presidential election outcomes, mostly relying on economic indicators. When the challenger is considered at all, the relevant variable is generally their moderation. Other studies find an overall benefit to moderation for members of Congress as well. Similarly, most rational-choice theory and game theory work, particularly common in the analysis of congressional voting, assumes that voters will usually support the closest option to their ideal, meaning that those with fairly extreme views will frequently compromise rather than render themselves irrelevant and make no gains whatsoever. Much to many activists’ chagrin, this does explain congressional behavior fairly well.
Now it is possible that someone establishes as the new king of the hill a better model for predicting or explaining presidential elections that demonstrates that overly-moderate but plausible candidates will lose more base votes than they will gain from the center – as always, science marches on.
But until then, “establishment” politicians and consultants, immersed in data and the election sciences, are going to treat the current state of science as fact, as they should. And that means assuming that more extreme candidates are probably harder to elect than more moderate ones, or at least that, else equal, going more extreme in a tough election is not a good strategy.
This does not mean the most moderate possible candidate is always the best option. Instead, it recommends the Buckley Rule, which, based on the idea that conservatives can’t win elections without moderates, says to vote for the most conservative candidate in the primary who can beat the Democrat. This rule was famously suspended by Rush Limbaugh in 2010 in order to support Christine O’Donnell over Mike Castle.
This brings up another important point. While Republicans hurl steady streams of RINO accusations at each other, the main strategic point of contention is whether we prefer the advice of William F. Buckley or Rush Limbaugh. Most people in differing GOP factions are not actually that far apart. When looking at it that way, these fights and accusations about who is a legitimate conservative look a bit silly.