ABSTRACT
Do campaigns really matter? There is a good bit of academic literature suggesting that they do not—or at least not as much as politicians and the media tend to think they do. It has now been more than half a century since President Franklin Roosevelt’s campaign manager, James Farley, offered what is known as Farley’s Law: that most elections are decided before the campaign even begins. More recently, in an analysis of presidential elections, Thomas Holbrook concluded that the general level of support for candidates during a campaign season is a function of national conditions (which vary primarily from one election to the next), while fluctuations in candidate support over the course of a single election year occur mostly in response to campaign-specific events. While Holbrook acknowledged that these events (including the so-called “convention bump,” debates, blunders by one of the candidates, and so on) do have an impact, national conditions (measured in terms of consumer sentiment and presidential job approval, both factors that are in place before the general election campaign begins in earnest) ultimately matter a great deal more in determining who wins and who loses. 1
