Why I think the economic voting hypothesis is not quite right

If you read political science blogs discussing the American presidential election then by now you will have seen many versions of the following post: some journalist makes a claim that something Barack Obama or Mitt Romney has said or done will influence the outcome of the election, and the political scientist responds that this is unlikely because presidential elections are decided by the economy.  Campaigns, we are told, do not matter because voters essentially decide how to vote based on the state of the economy in the 3rd quarter of every election year.

I do not doubt the pile of research that shows that the state of the economy is the best predictor of electoral success.  However, I have a problem with the causal mechanism proposed by the stronger, almost deterministic, versions of the economic voting hypothesis.  The reason is pretty simple: how do voters know the condition of the national economy?

Let me elaborate on why this is a problem.  We know that voters decide who to vote for based on the state of the national economy, not based on the state of voters’ local economies.  But if this is true, how can they possibly make that initial appraisal into the factual state of the economy without resort to the messages found in the media, which are inevitably influenced by the framing produced by the campaigns (and their allies)?  Where else could they get that info?  Almost no individual voters have sufficient experience or information to make determinations regarding the health of the national economy.  E.g. I have a pretty good sense of how many businesses have opened and closed in central Lincoln, but I don’t really have a sense of things in Omaha, much less how well things are going in Dallas and San Diego and Detroit, etc.  Even professional economists do not know what the state of the economy is in real time.  They rely on reports issued by the Department of Labor, the Fed, etc. months after those economic conditions were in existence.

The only way to get information about the national economy is from reading the news media and from listening to campaign messages.  The news mostly reports what the candidates and advocacy groups say.  And campaigns are obviously engaged in the sort of competitive discourse that the economic voting hypothesis seeks to refute.  Especially when you combine this with the fact that most of us do not evenhandedly appraise new information, but rather seek out info that confirms our pre-existing biases, it is hard to see how the material conditions of the economy could influence vote choice.

What this would mean is that elections are more about discourse than about material economic conditions, but it is a discourse about the empirical state of the world.  And when you think about the types of “debates” we have over the course of a campaign, you see that most of them are about what is, not what ought to be, suggesting perhaps the practitioners have already figured this out.