Rainbow Murray, "The Power of Sex and Incumbency: A Longitudinal Study of Electoral Performance in France," Party Politics, 14 (September, 2008),.539-554.

First paragraph:
Performing well in elections is a central goal of any mainstream political party (Downs, 1957: 11; Ware, 1996: 9). In the case of French parties, on which this article focuses, electoral success determines their financial resources, their relative strength within potential coalitions and hence their ability to influence policy agendas and outcomes. Even when parties do not win, every extra vote increases their financial and political resources. As a result, it is unsurprising that parties will seek to select the electoral candidates that will maximize the party's success at the polls (Gallagher and Marsh, 1988: 7). This is the case particularly in single-member constituency electoral systems, such as the one used in France's legislative elections. The candidate becomes the public face of the party within the constituency contested, and the fortunes of the individual and the party become inseparably linked.

Figures and Tables:

Table 1. Impact of change in candidate sex

Table 2. Impact of incumbency

Table 3. Combined impact of sex and incumbency

Table 4. Success rates of incumbents and inheritors

Table 5. Female incumbents and inheritors

Table 6. Sex and safety of seat

Last paragraph:
The question regarding the electoral weight of incumbency therefore remains partly unresolved. The question regarding women's electoral performance, on the other hand, is far more conclusive. In addition to refuting the suggestion that women perform less well at the polls, this article provides strong evidence to suggest that women are indeed placed in the most difficult seats, and, further, that this is an integral part of party electoral strategy. This leads to the conclusion that parties may not be objective in seeking the candidates that are the most electable, and/or that they may be selecting candidates for motives other than anticipated performance at the polls. Either way, I have demonstrated that parties' claims regarding the electoral cost of political renewal and feminization are unfounded and are entirely inadequate as a justification for selecting so few women. The law needs to be reformed to prevent the strategic discrimination against women evidenced by this article. First, the law needs to take into account not only the number of women selected but also the number of women elected, thus forcing parties to place a greater proportion of women in winnable seats. Second, the penalties for non-implementation need to be strengthened in order to force parties to overcome their reluctance to field women. Until the financial penalty for non-implementation of parity exceeds the perceived financial loss of fielding women candidates, the 'parity' law will not be sufficient to break the cycle of discrimination against women candidates by French parties.