Economic Voting in Turkey: Perceptions, Expectations, and the Party Choice

Abstract

Following a brief summary of macroeconomic developments since 2001 and a review of the economic voting literature in the Turkish context, we examine the effect of economic evaluations on the likelihood of Turkish voters to opt for the currently-ruling Justice and Development Party (Adalet ve Kalkınma Partisi, AKP). In addition to several socio-demographic characteristics and religiosity, our empirical model – estimated using micro data from a 2014 electoral tendency survey – includes both retrospective and prospective economic voting variables. Binary logit estimates provide strong evidence in favour of the idea that perceptions and expectations regarding the economy have a considerable impact on the party choice, especially when the voter holds the government responsible for the recent changes in economic conditions. Our findings also suggest that as long as the proportion of voters who are satisfied with their economic conditions is reasonably large, the AKP will remain at the centre stage of Turkish politics for at least another term.

Keywords:

• Economic voting
• Turkey
• Justice and Development Party
• party choice
• binary logit

Notes

1. Papers focusing on the various aspects of the rise and the ideology of the AKP include Çarkoğlu (2002a, 2010); Gülalp (2004); Atacan (2005); Tepe (2005); Öniş (2006); Özbudun (2006); Sayarı (2007); and Yıldırım et al. (2007).

2. See Lewis-Beck and Stegmaier (2007, 2013) for reviews of the existing international literature on economic voting.

3. In the absence of reliable economic indicators at the province level, Başlevent (2013) has shown that the level of education, mean age, and internal migrant share can be utilized to great success both at the province and district levels to predict the AKP’s vote share.

4. Rudolph (2003) and Tilley and Hobolt (2011) find that responsibility attributions are also shaped – to varying degrees – by economic ideology and partisanship, implying that the above-mentioned concerns about the endogeneity of subjective evaluations apply here as well.

5. At the time the survey was conducted, the BDP – which can be thought of as the predecessor party to the HDP – was in existence.

6. The ‘Alevi’ dummy drops out of the model in the smaller sample due to perfect multicollinearity.

7. Çarkoğlu and Avcı (2002), Dulupçu (2005), West (2005), Güvenç and Kirmanoğlu (2009), Akarca and Başlevent (2011), and Tezcür (2012) are some of the studies that deal with the geographical patterns observed in election outcomes in Turkey.