Smear Campaigns: Odious but Ineffective?

Brendan Nyhan, commenting on a New York Times in Review piece, postulates on the supposed effectiveness of famous campaign ads:


"While I obviously have normative concerns about misleading campaign attacks, it's much less clear that the LBJ ad had "crushing smear power," that the Swift Boat ads "severely undercut" John Kerry, or that Michael Dukakis lost his 17-point lead in the polls as a result of the Willie Horton ad. The leading models of presidential elections predicted that Goldwater, Kerry, and Dukakis would lose. Journalists tend to construct post hoc narratives based on dramatic visuals from debates and campaign ads, ignoring the fundamentals that actually drive elections (the state of the economy, presidential approval, war casualties, etc.)"

This is fair enough, but I think the election of 2004 is somewhat different, being such a dead-heat for so long that it was inordinately responsive to the actions of 527's like the Swift Boat Vets (which, it seems, did have rather an effect on the polls, when any effect would have been decisive). With the last two elections being so close and the next seemingly not deviating from this emerging pattern, the 'smear' campaigns, their attempts to define candidates and the small effects these will have on the polls are all going to become more important, and of course, the more important they become, the more they will be integrated into a media narrative.

Whats interesting to me is whether, in times of national depression or distress, times when the "fundamentals that drive elections" are more urgent and felt more keenly by the electorate, such tactics are more or less effective.