Strategic voting and effective representation thresholds: Evidence from three Spanish general elections

Abstract

The extent of strategic voting in the Spanish general elections of 2000, 2004 and 2008 is estimated using a new measure of strategic incentives suitable for proportional representation systems that avoids some of the problems associated with lagged variables. Strategic behaviour increased from 12 to 33 per cent of the electoral base of the United Left Party, the major victim of strategic defection. This estimate is a conservative one as elite mobilisation is controlled for in the constituencies, which is unusual in the literature on strategic voting that uses regression‐based methods.

Does increasing district magnitude increase the number of parties? Evidence from Spain, 1982–2011

“…Each of the country's 50 provinces serves as an electoral district and is guaranteed at least 2 seats; additional seats are allocated by population. Spanish voters have been shown to strategically abandon small parties in small districts and to concentrate their votes on the main two parties or prominent regional parties (Gunther, 1989;Lago, 2009;Lago and Montero, 2009;García Viñuela and Art es, 2012). District magnitudes thus range between 1 (Ceuta, Melilla) and 36 (Madrid in 2011) with a median magnitude of 5 and with 96 percent of the districts having less than 16 seats.…”

Tactical Voting in Norway: Context, Motives and Occurrence

“…; Bäck & Rosema ; Cox ; Cox & Shugart ; Duch et al. ; Gschwend & Stoiber ; Hobolt & Karp ; Holmberg ; Fredén ; Irwin & Van Holsteyn ; Jenssen ; Lago ; Lago & Martínez i Coma ; Meffert & Gschwend ; Oscarsson & Holmberg ; Viñuela & Artés ).…”

“…In contrast, the last district seat in the capital Oslo was won with 5.6 percent of the votes. According to Viñuela and Artés () voters are more sensitive to the ‘effective threshold’ than the number of seats as such.…”

Strategic voting and non-voting in Spanish elections

“…We define the most-preferred option as the one that the voter would choose when a party is on the verge of attaining parliamentary representation. We can capture this with the threshold gap variable, tg j , used by García Viñuela and Artés (2012). The threshold gap is the difference between the share of the vote that allows a small party j to achieve its first seat in a district (the effective threshold of representation) and the expected vote for party j in the district: tg j ¼ Effective threshold-Expected vote for party j This measure takes positive values when it is expected that party j will not achieve sufficient votes to obtain representation, while it takes negative values when the expectation is that it will garner enough votes to win a seat.…”

“…We compute the prevalence of each type of behaviour using a simulation procedure similar to that applied by Alvarez and Nagler (2000), Fieldhouse et al (2007), Herrmann and Pappi (2008) and García Viñuela and Artés (2012). The novelty of our approach compared to these authors is that we include abstention as another possible outcome in the model.…”