First off, let me be clear: this post isn’t solely about the current horse race. It’s just some generic observations regarding economic indicators in elections. Second, I defer to the folks who live and breathe this stuff, like Nate Silver and Ray Fair. Third, there will be some math.
Well, not so much math as a refresher re three aspects of a variable from calculus 1: the level, the first derivative, and the second derivative. The level is the level… like the fact that the unemployment rate is 8.1%.
The first derivative is the rate of change in the level. Is unemployment going up or down?
And the second derivative tells you whether the first is accelerating or decelerating. Is job growth (a first derivative) faster this month than last? That’s a second derivative question.
My point here is that I’m pretty sure that while all of these different measures matter, we’re largely a first derivative nation when it comes to forming our impressions about the economy. Yes, you’d rather run for office as the incumbent with low rather than elevated levels of unemployment, but if the trend is your friend, that’s a good friend to have (putting aside issues about what politicians actually have to do with job growth).
And yes, you’d like your favorable trends to be accelerating (positive second derivative)—you’d like to be able to say, “we’re not only adding jobs, but we’re adding them faster than we were X months ago!” But a favorable first derivative matters a lot. In fact, a good level that’s moving in the wrong direction—say, a low unemployment rate that’s notably rising—may be worse for a candidate than the moderately high one that’s falling.
This comes to mind whenever I hear pundits suggest that you can get a meaningful handle on the election from one number: the rate of unemployment. I don’t think that’s right.
I was also thinking about this today in the context of a post by Greg Sargent, posted here. By “momentum,” I’m talking about the first derivative. And here’s the relevant picture re unemployment in Ohio vs. the nation, where you see both the level and the momentum story combined.
None of this should be taken to mean I think we’re out of the economic woods. There are still far too many foreclosures in the pipeline and neither GDP nor jobs are growing fast enough—both are growing; neither is accelerating. But until one of the experts tells me otherwise, and while I’m sure levels matter a lot, I’m going to keep thinking that when it comes to people forming opinions re the indicators, it’s the first derivative that matters most.
UNEMPLOYMENT RATES IN US AND OHIO