Do closeness and stake increase voter turnout? Evidence from election results in small French towns in Brittany

ABSTRACT

Do closeness and stake increase voter turnout? Evidence from election results in small French towns in Brittany. Regional Studies. This article explores the determinants of second-round voter turnout in small French towns with a focus on two hypotheses. First, an increase in the closeness of the race may lead to an increase in turnout. Second, a higher seats stake, that is, a higher proportion of the total number of seats that remain to be filled in the second round of the election, may increase turnout. Results show that increases in both closeness and seats stake lead to higher voter turnout. Thus, evidence is given in favour of the assumption of rationality in voting in local elections.

摘要

亲近性与关键席次比率会增加投票率吗？来自法国布列塔尼小镇的选举结果之证据.  区域研究。本文探讨法国小镇第二轮选举的投票率之决定因素，并聚焦两项假说。首先，种族亲近性的增加，或许会导致投票率的增加。再者，较高的关键席次比率，亦即在第二轮选举中，总席次中有较高的比例须被填补，亦有可能增加投票率。研究结果显示，亲近性与关键席次比例的增加，皆导致了较高的投票率，因此有证据支持地方选举投票的合理性之预设。

RÉSUMÉ

Le caractère serré et l’enjeu augmentent-ils la participation électorale? Des preuves provenant des résultats électoraux dans les communes de petite taille en Bretagne. Regional Studies. Cet article examine les déterminants de la participation électorale au second tour dans des communes françaises de petite taille en se concentrant sur deux hypothèses. Premièrement, une compétition plus serrée pourrait entraîner une participation plus forte. Deuxièmement, un enjeu plus élevé en termes de sièges, c'est-à-dire une proportion de sièges restant à pourvoir au second tour plus élevée, pourrait conduire à une participation plus forte. Les résultats montrent qu’à la fois le caractère serré et l'enjeu en termes de sièges entraînent une participation électorale plus élevée, confortant l'hypothèse de rationalité dans le vote lors des élections locales.

ZUSAMMENFASSUNG

Erhöht sich die Wahlbeteiligung bei einem knappen Rennen und einer hohen Sitzquote? Belege von den Wahlergebnissen in französischen Kleinstädten der Bretagne. Regional Studies. In diesem Beitrag untersuchen wir die Determinanten der Wahlbeteiligung in der zweiten Runde in französischen Kleinstädten unter besonderer Berücksichtigung von zwei Hypothesen. Erstens, dass sich die Wahlbeteiligung bei einem knapperen Rennen erhöht. Zweitens, dass sich die Wahlbeteiligung erhöht, wenn in der zweiten Wahlrunde die Sitzquote höher ausfällt, also ein höherer Anteil an der Gesamtanzahl von Sitzen noch besetzt werden muss. Aus den Ergebnissen geht hervor, dass sowohl ein knappes Rennen als auch eine hohe Sitzquote zu einer höheren Wahlbeteiligung führen. Dies dient als Beleg für die Anname von Rationalität beim Wahlverhalten in Kommunalwahlen.

RESUMEN

¿Aumenta la participación electoral en elecciones reñidas y con una mayor competición por los escaños? Evidencia de los resultados de elecciones en pequeñas ciudades francesas de Bretaña. Regional Studies. En este artículo analizamos los determinantes de la participación electoral en la segunda vuelta en pequeñas ciudades francesas centrándonos en dos hipótesis. Primero, la participación aumenta en elecciones reñidas. Segundo, la participación aumenta con una mayor competición por los escaños, es decir, un mayor porcentaje del número total de escaños que se han de cubrir en la segunda vuelta de las elecciones. Los resultados indican que tanto un estrecho margen como la competición por los escaños llevan a una mayor participación electoral. Esto nos da evidencia a favor de la suposición de la racionalidad en la votación de elecciones locales.

KEYWORDS:

• turnout
• local elections
• closeness
• seats stake
• France
• municipalities

关键词:

• 投票率
• 地方选举
• 亲近性
• 关键席次比例
• 法国
• 行政区

MOTS-CLÉS:

• participation
• élections locales
• caractère serré
• enjeu en termes de sièges
• France
• communes

SCHLÜSSELWÖRTER:

• wahlbeteiligung
• kommunalwahlen
• knappheit
• sitzquote
• Frankreich
• gemeinden

PALABRAS CLAVES:

• participación
• elecciones locales
• elecciones reñidas
• competición por los escaños
• Francia
• municipios

JEL:

• D72

ACKNOWLEDGMENTS

The authors thank the anonymous reviewers for their helpful comments and suggestions.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

SUPPLEMENTAL DATA

Supplemental data for this article can be accessed at http://10.1080/00343404.2015.1118029

Notes

1. The authors thank a referee for this suggestion in the interpretation of the SEATS STAKE effect.

2. The last municipal elections were in 2001, 2008 and 2014.

3. See the fifth section for robustness checks that provide alternative closeness indicators measured at the list level as well as at the candidate level.

4. When there is only one list in the competition, the share of seats for the list ranked two is set to zero (and therefore RELATIVE CLOSENESS = 1). This is not a crucial choice because when RELATIVE CLOSENESS = 1 it can mean that there is only one list in the competition or that there are several lists in the competition and one of them obtained all the seats; closeness is weak in both cases. Computing RELATIVE CLOSENESS using the seats obtained by the two top lists is not problematic: in the sample there is only one municipality where the list ranked third has obtained at least one seat.

5. The closeness and the stake are not necessarily linked. An election could, for example, at the same time be close and have a low stake. Table A1 in the supplemental data online mentions that the correlation between RELATIVE CLOSENESS and STAKE is –0.35.

6. Previous versions of this paper tested two specific hypotheses: (1) whether the fact that one list already won a majority of the seats in the first-round election reduced the impact of stake on turnout; and (2) whether the marginal effect of stake on turnout depends on closeness. However, in both hypotheses the results were not robust to changes in specification. Details are available from the authors upon request.

7. The same remarks as given above in note 4 apply. First, RATIO CLOSENESS and LEADER CLOSENESS are not defined when S2 = S1 = 0. In these cases, the value of RATIO2/1 is set at 1 and those of LEADER CLOSENESS at 0 because S2 = S1 = 0 implies that the ballot is close. Second, when there is only one list in the competition, S2 is set to 0. In this case, ENTROPY CLOSENESS is not defined and its value is set to 0. Finally, computing the closeness measures RATIO CLOSENESS and ABSOLUTE CLOSENESS using the seats obtained by the two top lists is not a crucial decision: there is only one municipality where the list ranked third has obtained at least one seat.

8. The authors thank an anonymous referee for his suggestion on this respect. See also Grofman and Selb (2009) on indexes of political competition in the context of multi-member election races.

9. In 26 cities, no candidate was elected in the first round. In these cases, the vote share for the last elected candidate was set at 50%. In three cities, all the candidates were elected in the first round. In these cases, the vote share for the first non-elected candidate was set at 50%.

10. In two cities, candidates were not elected even though they received a vote share greater than 50.0%. This resulted from a mistake made by the mayor (who organizes the election) when computing the number of votes necessary to be elected. This is why the maximum for the variable FIRSTNONELECTED is 50.1% in Table A2 in the supplemental data online, and not 50.0% as it should be.

 

Additional information

Funding

The authors thank the Agence nationale de la recherche (ANR) for financial support [grant number ANR-08-GOUV-054] for the project SOLITER ‘Négocier la solidarité territoriale dans les intercommunalités'.