Opening hours of polling stations and voter turnout: Evidence from a natural experiment

Many other studies examine determinants of voter turnout. Geys (2006) and Cancela and Geys (2016) review studies that use voter turnout at the aggregate level (district, municipality, country) as a dependent variable. Geys (2006) concludes that population size and electoral closeness are significant explanatory variables: “turnout is higher when the population is smaller and the election closer” (p. 653); and: “the institutional procedures governing the course of the elections strongly affect turnout. Compulsory voting, easier registration procedures, concurrent elections and the use of proportional representation all significantly stimulate turnout” (p. 653). Institutions, however, are often endogenous.

US: The Times-Picayune, 29.05.2012, http://​www.​nola.​com/​politics/​index.​ssf/​2012/​05/​post_​469.​html; Germany: EurActiv, 19.09.2014, http://​www.​euractiv.​com/​section/​justice-home-affairs/​news/​spd-mulls-supermarket-voting-booths-to-boost-election-turnout/​; Austria: VOL, 22.01.2015, http://​www.​vol.​at/​laengere-wahlzeit-soll-beteiligung-heben/​vol-news-traffl-20041111-093017; Ireland: Houses of the Oireachtas, 04.07.2014, http://​oireachtasdebate​s.​oireachtas.​ie/​debates%20​authoring/​debateswebpack.​nsf/​takes/​dail201407040002​1?​opendocument.

Clearly, the major difference in extending opening hours of polling stations and shops is that voters have much less choice in polling stations (in some countries citizens can choose which polling stations they use) than consumers in choosing shops. Shopping hours are a strategic tool for shops to attract consumers. Studies on extending opening hours of shops therefore deal with issues of competition and regulation (e.g., Clemenz 1994; Gradus 1996; Thum and Weichenrieder 1997; Rouwendal and Rietveld 1998).

Language issues also did not receive any attention in the 2015 state election campaigns, which were dominated by labor and economic issues.

The capital city of Eisenstadt and the city of Rust are districts on their own. Both cities are located in the urban north of Burgenland.

In the debate in the state parliament, the ÖVP MP Rudolf Strommer pointed out that his party is not happy with the reform, but will agree to it for coalition reasons. His statement also implies that the social democratic SPÖ and the rightwing populist FPÖ were highly in favor of a second election day, i.e., longer opening hours of polling stations. Original in German language: “Ich sage auch klar, dass meine Partei nicht euphorisch für die Einführung eines zweiten Wahltages war, weil wir der Meinung sind, dass wir mit einem guten System der Briefwahl, mit einem Wahltag und mit der Sonderwahlbehörde eigentlich das Auslangen hätten finden können. Aber in einer Verhandlungsrunde kann sich niemand zu 100 Prozent durchsetzen. Wir haben hier den Willen anderer Parteien, auch die Freiheitliche Partei und die SPÖ haben vehement diesen zweiten Wahltag gefordert, wir haben dem auch schlussendlich zugestimmt, weil sich niemand zu 100 Prozent durchsetzen kann.” (Stenographisches Protokoll der 57. Sitzung der XX. Gesetzgebungsperiode des Burgenländischen Landtages (Thursday, 11. December 2014, 10.07 a.m.- 4.10 p.m.), p. 6927).

Local politicians may have strategically chosen where in a locality they open the additional ballot box on the second voting day (the location of polling stations may well influence voter turnout and party vote shares – see Brady and McNulty 2011, Bhatti 2012 and Gibson et al. 2013). For example, municipalities with leftwing majorities may open the additional ballot box in a neighborhood in which citizens are especially inclined to vote for the leftwing party. There was no strategic placement of ballot boxes. Two thirds of the municipalities have just one locality and thus offered one additional ballot box in the city hall. The other third of the municipalities (with more than one locality), however, also offered just one additional ballot box in the city hall. Only two municipalities (Mattersburg and Rotenturm an der Pinka) offered more than one additional ballot box.

The within coefficient of variation amounts to only 0.06. The coefficient of variation is computed as the ratio of the within standard deviation of opening hours (0.36) and the mean of opening hours (5.67).

Card (1992) shows the extent to which employment changed when the federal minimum wage increased in the United States. The effects differed across states depending on how many workers initially earned less than the new federal minimum wage. Acemoglu et al. (2004) show that the higher WWII mobilization rate of men in US states gave rise to a greater female labor supply. Finkelstein (2007) identifies the effect of introducing Medicare in 1965 on hospital spending. Health insurance coverage of the elderly differed substantially between states before Medicare was introduced in the United States. After the health care reform of 1965, insurance coverage was fairly equal across states.

We also use a fractional logit model (Papke and Wooldridge 1996) because our dependent variable voter turnout is conceptually censored to a minimum of 0 and a maximum of 1 (or rather 100%). In our sample, voter turnout takes on values from 52% to 93% (see Table 2). Inferences do not change when we use a fractional logit instead of Ordinary Least Squares (OLS).

There is no time-varying data available on religion of the population that may well also influence voter turnout (Hillman et al. 2015).

We use the Stata command provided by Hsiang (2010) and test spatial cutoffs of 5 km, 10 km, and 20 km. In any event, standard errors clustered at the municipal level appear to be more conservative. Therefore, we stick to clustering standard errors at the municipal level. We also use standard errors robust to heteroscedasticity (Huber-White sandwich standard errors – see Huber 1967, White 1980) for robustness tests. Inferences do not change.

The municipalities of Mattersburg and Rotenturm an der Pinka were exceptions: the second election day was held in two localities each.

See also Figure A.3 in the Appendix.

We assume the distance to the next polling station to be 0 in localities with one ballot box on the second election day.

The state elections in Burgenland and Upper Austria took place on 31 May 2015 and 27 September 2015, and the circumstances in both states differed a great deal, which we cannot address in a difference-in-differences model. In particular, the European refugee crisis was in full swing on 27 September 2015 (it was much less of an issue on 31 May 2015), Upper Austria had state and local elections taking place on the same day (Burgenland just had state elections). Trends in voter turnout in Upper Austria and Burgenland differed prior to 2015 (and continue to do so). Estimating a difference-in-differences model would not help to disentangle whether voter turnout increased in Burgenland because of a second election day or the pure increase in opening hours either. To examine the causal effect of increasing opening hours on voter turnout, it is therefore more useful to exploit variation within Burgenland over time and elaborate on treatment intensity (as done by Finkelstein 2007 and others).

See also footnote 1.

Invalid vote shares in the 2010 state election do not significantly differ between SPÖ mayors (1.7%), ÖVP mayors (1.7%), and the small number of other mayors (1.9%). We implement t-tests for each pair of mayors.

See §§ 74 et seq. of the Gesetz vom 9. November 1995 über die Wahl des Burgenländischen Landtages (Landtagswahlordnung 1995 - LTWO 1995). See also the stepwise procedure for the realized 2010 state election results: http://​wahl.​bgld.​gv.​at/​wahlen/​lt20100530.​nsf/​vwMandate/​100009.