In lieu of an abstract, here is a brief excerpt of the content:

CHAPTER 6 The Initial Success of New Parties The implications derived from the theoretical model mostly relate a series of theoretical variables with the likelihood ofnew political parties. But some implications also give hints at the relative frequency ofweak and strong new parties . Consequently, some limited information is available on the initial strength of the newcomers on the electoral scene. The implications that yield this information appear in table 6.1. They correspond to implications 1, 3, and 4 discussed in chapter 3.1 The first of them suggests that the increasing importance of new issues should lead to more new parties, both of the strong and weak type, as long as the latter are credible. If the weak parties are not a credible challenge, the increasing importance ofnew issues only leads to more strong new parties. Consequently, on average, new parties should be stronger when new issues become more important. The third implication discussed in chapter 3 suggests that weak new parties , provided that they are credible, arise more frequently if the benefits of high demands increase. In parallel, strong new parties emerge less frequently under such circumstances. Hence, as the benefits of high demands increase, the average strength of new parties should go down. If the weak challengers are not credible threats, however, the relationship is reversed. Then increasing benefits do not change the frequency of weak new parties, but do so for the strong ones. Hence the average strength should increase with higher benefits. Finally, the fourth implication presented in chapter 3 predicts that weak new parties arise more frequently if the costs of fighting a newcomer in the electoral arena increases. But this relationship only holds if weak new parties are credible challengers. If they are not, changes in the costs of fighting an electoral battle do not alter the frequency of weak and strong new parties. Consequently, the average strength ofnew parties should not change. These implications rely, as mentioned above, on the relative strength ofthe potential new parties. The initial success at the ballot, while certainly related to this relative strength, depends also on another set of actors, namely voters. 1. Implications 2 and 5 of chapter 3 fail to give indications on the strength of new parties. Consequently, they do not appear in table 6.1. 125 126 Altering Party Systems TABLE 6.1. Implications for initial strength of new parties implication direction of relationship if credible if non-credible ne\v Issues weak + weak 0 strong + strong + strength 0 strength + benefits of high demand weak + weak a strong - strong + strength - strength + costs of electoral fight weak + weak 0 strong 0 strong 0 strength - strength 0 However, these do not appear in the theoretical model. Consequently, the following tests of the implications have to be taken with a grain of salt. They reveal certainly one aspect of the initial strength of new parties. But strategic behavior, like tactical voting and coordination among voters (Cox 1997), might reduce considerably the link between the relative strength ofa potential new party and its first electoral result. Following the discussion in chapter 4, it is obvious that compared to the empirical analyses on the emergence of new parties in chapter 5, I have to switch my observational unit to test these implications. From now on I will focus on the new political parties that presented candidates at an election. As table 5.2 in chapter 5 shows, overall 361 new parties emerged in the 261 elections under consideration. For each of these parties I attempted to determine the vote share that they received in the first national election where they fielded candidates. Despite considerable effort it proved impossible to obtain exact electoral returns for all of these new parties. Most likely, a considerable number ofthese parties obtained so few votes that they appeared in official electoral statistics under the heading of "others." Overall, I was able to determine the exact first electoral result for 260 new parties. This implies that for a considerable number of new parties I fail to have information on the dependent variable that I wish to explain, namely their initial electoral success. Despite the fact that these parties with missing electoral returns are likely to have had only marginal electoral success, omitting them from the analyses that follow might lead to biases. This especially, since the accuracy of electoral statistics is likely to vary from country to country. And since some of my explanatory variables are country-specific, simply omitting the parties with...

Share