This chapter proceeds in three sections. The first evaluates the extent to which the findings from the preceding chapters are generalizable beyond two “treatment” and two “control” elections in Mexico and the United States. Can a campaign-centered model explain patterns of economic voting across different types of elections and in different institutional and political contexts? Analysis of the effect of campaign communication in the 2007 South Korean, 2006 Canadian, and 1972 West German national elections supports the generalizability of a priming-based theory of economic voting. In the second section, I extend my theoretical model and consider whether campaign messages shape voting behavior beyond the economic vote. Focusing largely on the 2000 Brazilian presidential election, I find that candidates can change votes by conditioning the salience of issues other than economic retrospections.

Together these analyses support the broad generalizability of the campaign-centered approach to economic voting. However, this leads back to an important question: so what? What have we learned about electoral politics from a study of the psychological mechanism that conditions the economic vote? The third section of this chapter, then, returns to the economic voting puzzle that I posed in Chapter 1 and presents aggregate-level tests of my campaign-centered theory. Can a model of election outcomes that explicitly incorporates the conditioning influence of campaign dynamics outperform the conventional model?

HOW DOES THE MODEL PERFORM IN OTHER TYPES OF ELECTIONS?

Do candidates in democracies around the world have the capacity, via political communication, to activate or deactivate the economic vote? Contrary to the predictions of conventional economic voting theory, I have shown that candidates' strategic dissemination of campaign messages in two US and two Mexican presidential elections conditioned voters' willingness to hold governments accountable for economic performance.