Skip to main content
ABC News
Aug. 15: To Bounce, or Not to Bounce

Wednesday brought more data that measured Representative Paul D. Ryan’s impact on the horse race. There was a new set of polls in swing states from the firm Purple Strategies, and there was a weekly national tracking poll from YouGov.

This set of polling was variable. Mitt Romney made clear gains in some of the polls (in Ohio and Virginia, among the states that Purple Strategies tracked), while his numbers receded in others (he lost two points relative to President Obama in the YouGov weekly tracker).

But if you simply average all the data that post-dates the unveiling of Mr. Ryan as Mr. Romney’s running mate, we’re getting largely the same story that we did on Tuesday. Mr. Romney has gained a net of one point, on average, in the eleven polls conducted wholly or partially after his announcement of Mr. Ryan, compared to the prior renditions of the same surveys in the same states. This is a below-average “bounce” for the selection of a vice presidential candidate; in past elections, the bounce has averaged in the neighborhood of 4 percentage points instead.

We’ve been giving a lot of attention to this bounce phenomenon in our last couple of polling updates. That’s partly because it helps us to take the polls in context. For instance, the set of polls that Purple Strategies released today had some pretty favorable numbers for Mr. Romney — he was ahead in three of the four swing states they polled, including some in which he has generally trailed in polls released by other organizations. However, considering that the polls coincided with the announcement of Mr. Ryan, and that Purple Strategies polls have been a bit Republican-leaning so far this cycle (by an average of 2 points relative to the consensus), they aren’t quite as interesting.

At the same time, as Jonathan Bernstein reminds us, whether Mr. Romney gets a significant bounce in the polls right now is not really what matters. If Mr. Romney makes a one-point advancement in the polls and holds it permanently from his selection of Mr. Ryan, then this would count as a successful vice presidential selection: a one-point shift in the polls is actually fairly meaningful given how close this race is.

Conversely, if Mr. Romney initially picked up 6 points in the polls from his choice of Mr. Ryan, but the bounce quickly evaporated and the choice wound up costing him a point or two in the end, it would count as a poor selection — despite the gaudy headlines that he would get in the meantime.

If you had a theory predicated on the idea that Mr. Ryan would produce a huge near-term swing in the polls — that he’d just sweep people off their feet, producing huge momentum for the Republican ticket as they headed into their convention — then the evidence so far is damaging. But I did not hear very much of that when Mr. Ryan was announced last weekend, even from conservatives who were pleased with the pick. Instead, the arguments both for and against Mr. Ryan were about the long-term balance of risk and reward that he might present to Mr. Romney.

By definition, we don’t really know anything about those long-term effects yet. Perhaps you could make an inference and put forth a theory that a below-average initial bounce is a leading indicator that Mr. Ryan will eventually prove to be a damaging choice. But that inference would need to be made with a lot of modesty, since the correlation between how a vice presidential nominee is initially perceived and what voters eventually come to think of him has only been moderate in past elections.

Better Economic Numbers in July

In non-polling news, the government released reports on inflation and industrial production on Wednesday, both of which were reasonably favorable. These are two of the seven reports that our forecast uses to calibrate its economic index, and they have usually been among the better-looking ones for the economy. Inflation has been low throughout Mr. Obama’s term in office, and industrial production, a measure of manufacturing activity, has picked up much faster than most other major economic indicators.

As a side note, the four economic indicators that economists use to track recessions — which parallel those that the forecast model uses — have been reasonably bullish lately. Industrial production was up considerably in July, as were retail sales. The nonfarm payrolls report was better than in past months. The fourth data series, real income, is the slowest of the four to be updated — the latest print of that data covered June rather than July. But it has been strong so far in 2012 after being very poor in 2011.

Although the recovery has been touch-and-go — the solid July data makes up for some weakness in the late spring — it is very unlikely that we are officially in a recession right now, and almost as unlikely that one will be declared prior to the November election.

Voters have their own ways of thinking about the economy, and these may be more sensible in many respects than the semantics that economists use to describe it. At the same time, the way the political news media covers the economic numbers also leaves a lot to be desired. Perhaps the most persistent problem is the failure to recognize how noisy individual economic data points can be, and therefore how it is best to look for consensus evidence rather than focusing on one or two reports. (Exactly the same mistake is made, of course, with the coverage of political polls.) In any event, the balance of economic data has been a little better so far in July and August than during May or June.

Nate Silver founded and was the editor in chief of FiveThirtyEight.