Successful mass parties need to build a stable base of partisans, citizens who feel an affinity with the party. Around the world, voters are much more likely to support the party with which they identify (e.g., Campbell et al. 1960; Green, Palmquist, and Schickler 2002; Lupu 2015a). Partisanship also stabilizes elections and protects parties from volatile shifts in public opinion. Even when a party fails to fulfill its campaign promises, or oversees a period of economic decline, partisans are willing to give their party the benefit of the doubt (Kayser and Wlezien 2011). To use Hirschman's (1970) terms, partisans will support their party out of loyalty. And loyalty "can serve the socially useful purpose of preventing deterioration from becoming cumulative, as it so often does when there is no barrier to exit" (79). Research on mass partisanship has nevertheless focused little on how partisan attachments emerge. That is partly because studies of partisanship focus largely on established democracies, where parties and partisan attachments are already widespread and stable. 2 Scholarship about partisanship in these settings typically turns on whether partisan attachments are stable predispositions or instead fleeting attitudes (Bartle and Bellucci 2009; Budge, Crewe, and Farlie 1976). Evidence of partisan stability is thought 1 For their advice and suggestions, I thank the editors, the anonymous reviewers, Fran Hagopian, Henry Hale, and participants at the conference on "Challenges of Party-Building in Latin America" at Harvard. I also thank David Samuels, Cesar Zucco, and the Center for the Study of Public Opinion at the State University of Campinas for generously sharing data. A previous version of this chapter was presented at the 2014 annual meeting of the American Political Science Association. I gratefully acknowledge the support of the Center for Advanced Study in the Social Sciences at the Juan March Institute. Sebastián Lavezzolo provided excellent research assistance. Replication material available on the author's website. 2 Extrapolating from these cases, Converse (1969) posits that mass partisanship will necessarily emerge over time in new democracies, but leaves unanswered the question of which parties will attract partisans. 3 Even the observation of partisan erosion in these countries has been contested (