The not-fast-enough feedback loop…and its problems

Posted by padams under Election 2008, Election 2008 Campaign strategy, Election 2008 Media commentary

Paul Adams

There is any idea popularized by the wonderful Mickey Kaus, which he has labelled the “Feiler Faster Thesis”, named after the guy he stole it from. Essentially, Kaus/Feiler argue that the modern news environment has radically shortened the news cycle, but that this is not necessarily a bad thing because we are adjusting to this reality. 

Here’s one formulation of the thesis by Kaus:

The news cycle is much faster these days, thanks to 24-hour cable, the Web, a metastasized pundit caste constantly searching for new angles, etc. As a result, politics is able to move much faster, too, as our democracy learns to process more information in a shorter period and to process it comfortably at this faster pace.

In general, I think there is some truth to this. However, there is a limit to our capacity to identify relevant information, disseminate it through the media, and allow the public to absorb it.

As a sometime pollster and sometime journalist, I have long observed the (relatively) lengthy feedback loop involving polls. Polls are not just snapshots: they are snapshots out the rear-view mirror. Even the quickest turn-around daily tracking polls are looking backward over three or four days. 

So when reporters pick up on trends in the polls, they are starting with information which is already a few days old, at least in part. It then takes another day or so for the reporters to explore the implications of the changes, through quizzing politicians, strategists, voters and so on. And it similarly takes the parties at least a day or two to adjust to the new reality (even if they are relying on their own internal polls). Typically (but not invariably) columnists follow in the rear. 

And then, of course, there’s the public, who actually drop the kids off at daycare, go to work, schlep to hockey practice, and don’t spend their entire lives examining the minutiae of the political campaign. They take a few more days to absorb the information they receive through the media, and then, in the case of so-called “strategic voters”, perhaps adapt their own voting choice accordingly. When they do so, they close the loop, because as their preferences change they start showing up in the polls, and we start all over again.

This all takes at the very least a week. At the very least.

Now, let’s look at this in the context of the polls here in Canada in the last week. There has been, as some of you will have noticed, a somewhat puzzling discrepancy among the polls, which is a topic for another day, But there is agreement on one thing: the Liberals rose somewhat and the Conservatives fell somewhat just after the debates and coincident with the deepening of the international credit crisis last week.

For a few days, the gap between the two leading parties closed  — in all the polls, albeit to varying extents.

But then something interesting happened: the gap started opening up again but the media did not instantaneously react. For example, CBC television was trumpeting Liberal momentum on their morning show today, and the Globe had an editorial cartoon to the same effect even though there is general agreement now among the polls that the gap between the Liberals and Conservatives has been widening in recent days. The disagreement amongst the polls is about the timing and extent of these trends, not their direction.

Who cares? Well, we all should. The EKOS tracking poll last night showed that almost a quarter of respondents think the Liberals will win the election, even though this now seems quite unlikely based on where the public has been moving this week. This growing expectation that the Liberals may win is concentrated among non-Conservatives — in other words, the voters who might potentially vote strategically to stop a Conservative victory if they thought this was likely. It may also influence some voters who would like the Tories on a leash, but can’t see Dion as prime minister.

The value of polls is that they can supply timely and relevant information to the public, which voters may (or may not) choose to consider when they cast their ballots. But in this election, this year, it may be that some voters go to the polls with old information on their minds.

The Kaus/Feiler Faster Thesis is true to an extent. But it has also been articulated in the context of the much, much longer American election campaigns. The news cycle in this Canadian election may actually be turning too slowly for some voters to have the best information available on the inclinations of their fellow citizens before going to vote.

Paul Adams is a former political reporter with the CBC and the Globe and Mail, and is now a member of Carleton’s journalism faculty, and executive director of EKOS Research Associates.