07/01/2010 09:15 am ET | Updated May 25, 2011

When qualitative punditry adds value

Sean Trende at Real Clear Politics objects to my post on Peggy Noonan's mystical interpretations of presidential popularity and mounts a defense of non-quantitative punditry:

Non-quantitative punditry has a huge place in our discourse for many reasons, including one that is directly applicable here... [T]he most applicable problem here is that there is always a large portion of the data that have to be explained qualitatively.

For example, take the Presidential Approval models. There are any number of them out there, but all of them have a significant portion of the variation in Presidential approval (or variance, in geekspeak) that the model just can't account for. Even for models that make political scientists giggle with glee at the high r-square they've produced, there will still be about 10 to 20% of the data that the model won't explain. Political scientists like to call this "error," but it isn't really "error." It's just "other stuff we can't readily turn into data"...

All we know is that there is always going to be a large portion of data -- whether it be presidential approval, congressional midterm elections, or presidential election results -- that can't be easily explained quantitatively. This is where qualitative analysts like Noonan will always be valuable.

Trende points to the loss of approval Reagan suffered after Iran-Contra that was unrelated to the state of the economy, and the fact that Clinton didn't suffer a similar drop during the Monica Lewinsky scandal.*

I think Trende is largely doing battle with a straw man here -- I don't disagree in principle with any of these points. There's no question that factors other than the economy affect presidential approval (for instance, any well-specified model includes political events such as Iran-Contra), and there's no question that qualitative insights can help us understand why presidential approval deviates from what we might otherwise expect given the state of the economy.

Instead, my point in the original post was to criticize the tendency of pundits to invent elaborate rationales for presidential approval ratings or election results while neglecting or ignoring the role played by the economy. In both Noonan quotes, she briefly acknowledges the possible role played by the economy in explaining Bill Clinton's popularity and Barack Obama's political difficulties before deviating into characteristically involved accounts of why the most important explanatory factor is instead whether presidents are "snakebit" or have sufficiently clarified "The Sentence."

Since Obama took office, TNR's Jon Chait has dubbed the Republican version of this tendency the "Wehner fallacy" after former Bush administration official Peter Wehner. Here's how Chait explains it:

[A] recurring theme in Republican commentary has been to ignore the economy in assessing the public's sour mood toward the party in power, and to assert that disapproval of the Democrats is entirely a function of public revulsion at the liberal agenda. One could make a case that the Democrats have politically overreached. I disagree. But to characterize the backlash as driven entirely by concerns about policy, without mentioning the pull of an economic crisis that began before Obama took office, is not an argument that any political scientist, or even a candid pollster or political adviser, would take seriously. It's pure propaganda.

We can observe a similar version of this problem in punditry about presidential campaigns. Elections typically converge to an outcome quite close to what we would expect given the fundamentals (principally, the state of the economy), but pundits instead attribute these outcomes to campaign events, debates, etc. in a manner that is frequently inconsistent with the evidence (see, for example, here, here, and here).

I have no problem with punditry that helps try to explain deviations from expected presidential approval ratings or election outcomes given the state of the economy. But pundits who try to substitute their own made-up stories for the economy as the primary explanatory factor are peddling nonsense.

* Though Clinton's ratings did not decline, it does not follow that the Lewinsky affair had no effect on his approval ratings. Brian Newman's research (gated) concludes that the Lewinsky scandal suppressed likely gains in Clinton's approval ratings in 1998.

[Cross-posted at]