Electoral volatility, competition and third-party candidacies in US gubernatorial elections

Third-party participation in plurality elections should be rare, given the low probability of electoral success. In the United States, the entrenched two-party system makes third-party candidacies especially puzzling. We develop a general theory of these candidacies based on the electoral context, focusing on electoral competition and volatility. When electoral competition is either low or high we expect the number of third-party candidates to be high, due to the opportunities to raise attention to policy issues or affect the election outcome. Moderate levels of competition will produce low levels of third-party candidate participation, as there are fewer prospects of drawing votes or attention. Volatility is expected to have a positive effect, since high volatility signals a de-aligned electorate. We evaluate our claims using US gubernatorial elections, 1977-2005. The results support our claims, suggesting that third-party candidacies are shaped by the degree of electoral volatility and competition.