Work in progress

Cultural issues have become the dominant battleground in European elections, yet mainstream parties have struggled to respond effectively. While voters increasingly demand stricter positions on immigration, acting on that demand carries a cost - a norm against legitimizing the far right and what it stands for. Voting either for the extreme party itself or for a mainstream party that has converged toward its positions bears the cost of violating this norm, whose strength and prevalence in the electorate are, however, uncertain. We develop a two-period model of electoral competition to address two objectives. First, we show why representation gaps persist: convergence toward the far right is not always optimal, as the decision depends on both the strength of the norm and the party's valence advantage. Second, we characterize why some mainstream parties waver. A mainstream party may sacrifice voter trust to experiment and learn about the norm before the election, with the incentive depending on norm uncertainty and the size of the valence base. When the extreme party becomes a serious competitor, the calculus changes: parties with a stronger valence base, not weaker ones, are drawn into experimentation. Finally, we show that experimentation can reshape the norm distribution. Revealing that the norm is weaker than expected triggers conformity-based norm erosion, making convergence less costly. When the extreme party is the main competitor, revealing that the norm is stronger than expected creates an additional motive of reinforcing the norm to contain the extreme party.