In the pivotal year of 2023, Turkey stood at a crossroads, entering a significant electoral event encompassing both parliament and the presidency. This posed a formidable challenge to the established 21-year rule of the AKP, led by Recep Tayyip Erdoğan.
Within the opposition, there was optimism, fueled by underlying issues that had eroded the AKP’s credibility in recent years and the success of the coalition during the previous local elections. The Turkish presidential election marked a turning point for the ruling Justice and Development Party and Erdoğan, facing criticism over their handling of the economic crisis. The Nation Alliance, formed by six parties before the electoral contest, aimed to challenge the incumbent. Unlike past post-election coalitions, this alliance comprised parties spanning the ideological spectrum.
A lingering issue centered around selecting a single joint candidate for the presidency. This challenging process led to the nomination of Kemal Kılıçdaroğlu, the leader of the CHP, the alliance’s largest party. On the other hand, the IYIP, the second-largest alliance party, suggested either Ekrem İmamoğlu or Mansur Yavaş as candidates. Despite their CHP affiliation, the IYIP argued İmamoğlu or Yavaş could better resonate with the right-leaning voter base within the alliance. The decision involved a strategic balance, considering candidates’ appeal to the substantial voter base of the Kurdish HDP.
Our study (Lemi Baruh & Ali Çarkoğlu) delves into the divided opposition’s difficulty in nominating a candidate, contributing to President Erdoğan’s continued rule. Utilizing a Response Surface Analysis (RSA) approach, we explore potential outcomes of opposition parties’ considerations regarding their presidential candidate.
Drawing on previous research by Schönbrodt et al. (2018) on actor-partner interdependence patterns in dyadic relationships provides valuable insights into expected shapes. The first extreme, the X-oriented pattern, is exclusively influenced by scores on the X-axis (refer to panel A, X-oriented in Figure 1). In this scenario, holding the X level constant shows no change in the response surface height as Y values vary. For instance, at a certain level of trust in Party X voters (like CHP, the largest coalition partner advocating for Kılıçdaroğlu’s candidacy), the candidate’s likability remains constant with different levels of trust in Party Y voters (e.g., IYIP, the second-largest coalition party opposing Kılıçdaroğlu’s candidacy). The estimated surface only responds to changes in trust among typical Party X (CHP) supporters. The second extreme is the Y-oriented pattern, where the outcome measure is solely affected by scores on the Y-axis (refer to panel B, Y-oriented in Figure 1). In contrast to these extremes, an XY-oriented pattern emerges when the outcome is influenced by a combination of the X and Y axes (see panel C, XY-oriented in Figure 1).
Our expectation was that, for the alliance to benefit a candidate, the response surface pattern should approach the XY-oriented pattern. This would have implied that when trust in voters of one political party (think CHP, the largest coalition partner, who pushed for the candidacy of Kılıçdaroğlu) is low, trust in voters of the other political party (think IYIP, the second largest party in the coalition; opposed the candidacy of Kılıçdaroğlu) can compensate for the lack of trust in CHP voters. Similarly, when trust is high for voters of both parties, likability for the candidates should also be high. In essence, we were investigating whether any of the candidates exhibited an XY-oriented pattern that could lead to potential electoral appeal even among ideologically distant voters from CHP.
Our analysis showed that Kemal Kılıçdaroğlu, the leader of the Republican People’s Party, faced challenges in appealing to the center-right wing voters of the alliance partner, the Good Party. However, he demonstrated greater success in garnering support from the Kurdish voter base of the Peoples’ Democratic Party. Despite this, our findings highlight a sense of unease among the constituencies of the Good Party (İyi Parti – IYIP) regarding Kılıçdaroğlu’s candidacy for the Nation Alliance.
In these constituencies, Imamoğlu and Yavaş were perceived more favorably than Kılıçdaroğlu. Notably, Yavaş appeared to benefit from the XY-oriented pattern, as indicated in Figure 2, Panel C. Interestingly, our results also indicate that when facing a typical voter from the Kurdish Peoples’ Democratic Party (Halkların Demokratik Partisi – HDP), Kılıçdaroğlu enjoyed a likability advantage over Yavaş.
Hence, the decision against Kılıçdaroğlu involved a difficult choice between alternatives whose vote potential differed across constituencies. The coalition was stuck between a rock and a hard place. Despite post-election war of words that blamed CHP for pushing Kılıçdaroğlu, our analysis underscores that none of the three options would have emerged as the ideal candidate.
Baruh, L., & Çarkoğlu, A. (2024). Difficult choices: Choosing the candidate of the nation alliance in the 2023 Turkish presidential election. Southeast European and Black Sea Studies. https://doi.org/10.1080/14683857.2024.2324523
This work was conducted as a part of a larger research project supported by The Scientific and Technological Research Council of Turkey (TÜBİTAK) [Project number: 121K206]
lemibaruh.me 