Insight Turkey
Insight Turkey
Challenging ideas
On Turkish politics and International affairs

Insight Turkey > Articles |

Change and Continuity in Turkey’s June 2018 Elections

In this paper, we first present a background to the watershed elections of 2018, describing the context preceding the elections. In order to explain the changes and continuities in the geographic patterns of the 2018 elections, we present ballot-box and district level data from 2018 and previous elections. We refer to long-term ideological/cultural differences as well as short-term evaluations about security and the economy in shaping the 2018 election results. Even in the face of various important developments, radical changes in the geographic voting patterns do not appear to have taken place. The incumbent AK Party and its leader, President Erdoğan, continued to have an electoral edge over competitors in regions that mostly overlap with those observed in earlier elections.

Change and Continuity in Turkey s June 2018 Elections
On June 24, 2018, Turkey held presidential and parliamentary elections and as a result, the presidential system of government officially took effect. YAKUP BAKAR / AA Photo


In arguably the most critical election in modern Turkish history, incumbent President Recep Tayyip Erdoğan received nearly 26 million (52.5 percent) of about 50 million valid votes in the elections of June 24, 2018. Following a narrowly approved referendum in April 2017, President Erdoğan’s new term starts with sweeping executive privileges that include the power to issue decrees with the force of law and to appoint the cabinet and vice-presidents as well as senior judges amongst other civil servants. However, in contrast to its repeated success in the second presidential election since 2014, Erdoğan’s party, the Justice and Development Party (AK Party) lost its parliamentary majority. Although it remains the largest party with 295 seats in a parliament with 600 seats, the AK Party’s electoral support declined from 47.8 percent in November 2015 to 42.6 percent in June 2018.

Muharrem İnce was the leading opposition candidate against Erdoğan and received about 31 percent of the votes. Similar to Erdoğan’s electoral performance compared to his party’s in the parliamentary election, İnce’s party, the Republican People’s Party (CHP) did poorly and received only 22.7 percent of the votes, securing 146 seats in the parliament. This marks a noteworthy reduction of votes from 25.3 percent in November 2015. The left wing Kurdish Peoples’ Democratic Party (HDP) once again remained above the 10 percent threshold and obtained 67 seats. HDP leader Selahattin Demirtaş campaigned from behind bars and garnered 8.4 percent of the votes, effectively being the third largest candidate in the presidential election. The Nationalist Movement Party (MHP) supported Erdoğan’s candidacy and formed an electoral alliance with the AK Party, performing unexpectedly well by gaining 11.1 percent of the votes that brought 49 seats. As such, the MHP’s performance is better compared to its rival the İyi Party (İYİP) founded by splinters from the MHP, which obtained slightly below 10 percent with 43 seats. The İYİP had been formed in 2017 after its leader Meral Akşener resigned from the MHP and ran in the June election within an alliance with the CHP.

Our objective in this article will be to present a detailed geographic analysis of the election results across districts. We will compare parliamentary election results in November 2015 and June 2018, as well as presidential election results in 2014 and June 2018. To do so, we collected election data based on the official Higher Council of Election results at the ballot-box level. Based on this data, we will explore the extent to which the 2018 elections compared with previous elections.

The next section describes the background to the 2018 elections. We present a framework to explain the reasons for holding a snap election which was initially scheduled for November 2019. We argue that short-term factors based primarily on economic concerns, combined with long-term ideological differences, were critical in the decision to hold snap elections in June 2018. We also mention several security concerns stemming from the coup attempt in 2016. The third section presents what we call mobilized ballot-boxes and shows patterns of electoral competition in the most recent elections. The fourth section presents a comparative geographical analysis of the most recent elections by referring to previous elections. In this section, we present six politically distinct clusters across districts and describe geographical party competition in Turkey, based on the 2018 elections, and the previous parliamentary elections of 2015 and the presidential elections of 2014. The fifth section concludes with expectations based on our findings and the newly established political system in Turkey.



Background to the June 2018 Elections

The June 2018 election results differed to a great extent from what had been predicted by most polls and punditry preceding the elections. According to expectations, the presidential election would go into a run-off with a small likelihood of İnce as the winner. To describe these expectations, we collected 22 publicly available results of opinion polls since January 2018 which mentioned at least one of the presidential candidates who eventually ran for election in June 2018. Before the elections, in early June, there were eight such polls and their average expected vote share for Erdoğan was 50.6 percent and for İnce, it was 26.9 percent. Figure 1 below presents public opinion results from these 22 different polls.1

Figure 1: Public Opinion Polls Before the Elections


Results from these polls indicate that in June, two polls by Sonar and Gezici public opinion companies predicted Erdoğan to be below the 50 percent threshold while six polls by MAK, Konda, Genar, Metropoll and another one done for Bloomberg predicted Erdoğan to be above the threshold. Since most of these results (under simplifying assumptions) included the 50 percent within their confidence intervals, uncertainty continued until election night. Although Erdoğan had a clear advantage in the first round, if there was a run-off between him and İnce, most pundits predicted a competitive election with İnce garnering additional support from other candidates.2

İnce’s concession may be out of line with the CHP’s partisan agenda and may be seen as an unexpected legitimation of the new system –one very much welcomed by the AK Party leadership

Looking at the development of candidate support levels across 22 publicly available opinion polls, we observe that Erdoğan’s win in the first round became a possibility only in the last few weeks before election day. Only two polling results before June predicted Erdoğan to be above 50 percent (May average polling results show Erdoğan at 45.6 percent). However, the average of all eight polls for Erdoğan in June was 50.6 percent. İnce’s level of support slowly rose and appears to have remained stable around 22-27 percent until June (May average polling results show İnce at 24.4 percent). Only two polling results showed İnce above 30 percent in June, and the rest ranged between 20 to 29 percent with an overall June average equalling 27.2 percent. Hence the average of eight polls for both Erdoğan and İnce were under-predictions for both candidates by (52.38-50.6) 1.78 and (30.79-27.2) 3.59 percentage points respectively.

For Demirtaş’s candidacy, there appears to be a consensus from early in May through the end of the campaign that he would garner about 10 percent of the votes. Only four of these 22 polling results showed Demirtaş to be below nine percent, while nine polls showed him to be above 11 percent. Akşener’s level of support appears to be between 18 to 24 percent before May in a total of five polling results (the average being 20.5 percent). From May to June we observe a steady decline in Akşener’s level of support down to about 9 to 12 percent. The average level of support in May for Akşener was 16.5 percent, and the June average was 10.5 percent. However, even in June, five of the eight polling results showed her level of support ranging from 10.2 to 12.5 percent. Considering the fact that Demirtaş obtained only 8.3 and Akşener 7.4 percent of the of the valid votes, these predictions were clearly misleading.

In short, the polling results underpredicted the performance of both Erdoğan and İnce while overpredicting the vote shares for both Demirtaş and Akşener. Felicity Party leader Karamollaoğlu’s poor showing in the election results was no surprise since he appeared only in three polling results above the margin of error.

In the upcoming local elections scheduled for March 2019, the MHP leadership is likely to follow an emboldened strategy and keep on pressing to obtain policy changes and outcomes that appeal to their electoral base

A similar perspective can be obtained about the level of electoral support for the major political parties from a total of 34 polling results that were made publicly available from January through June 2018. We again brought them all within a comparable range of predictions that lie between 0 and 100 percent. Figure 2 below shows the averages of these polls over time. We observe that while the predicted share of votes for the AK Party on average declined slightly from January to June, the share predicted for the opposition CHP rose by about three percentage points. While the decreased average AK Party share for June was still an overestimate within the typical margin of error, the overestimate for the CHP was off the mark by about 3.5 percentage points.

While the other smaller party average vote shares reached in June for the HDP and İYİP were quite close to their realized vote shares, the average prediction for the MHP was significantly lower than the party’s realized level of support in the June election. All eight available polling results in June predicted that the MHP would remain below the 10 percent threshold and hence would get into the Turkish Grand National Assembly (TGNA) solely due to the alliance it had formed with the AK Party. The predictions of these eight polling results ranged between 4.6 to 9.3 percent.3 Our analyses of the geographical distribution of election outcomes across districts may provide at least a partial explanation as to why this large error took place. As we will underline below, the AK Party and the MHP have typically garnered their electoral support from the same geographic constituencies. In June 2018, the MHP appears to have benefitted from vote flows primarily from the AK Party. Such vote switches took place within the constituencies of the two large alliance members and hence might not have been considered as an effective party switch by the voters. Nevertheless, given the quite comfortably large sample sizes, especially toward the end of the campaign period, such large differences between the predictions and realized vote share for one of the major parties indicates a deficiency in the sampling and/or other data collection technologies being used in this sector.

Figure 2: Average Polling Results for the Political Parties, January to June 2018


The absence of a level field during the campaign and expectations of possible electoral irregularities may partly explain the surprise Erdoğan’s win caused in the first round. Given the above-summarized polling results, many among the electorate expected the presidential election to be decided in a run-off between Erdoğan and İnce. In this regard, the concessions of the opposition leaders are very significant. Their concession of defeat right after the elections on the evening of June 24 effectively gave the seal of approval to Erdoğan’s term and consolidated the transition from a parliamentary to a presidential system. From a different angle, the opinion polls we present in Figure 1 may be taken to suggest that Erdoğan’s win in the first round was not a huge surprise after all. İnce’s concession may be out of line with the CHP’s partisan agenda and may be seen as an unexpected legitimation of the new system –one very much welcomed by the AK Party leadership. Such a cooperative stance is also much needed from the perspective of Turkish democracy at large, which suffers from rising partisan polarization. However, continued attention to the technical details of election administration and scrutiny of election results analyzed by scientifically sophisticated methodologists will only add to the credibility of the Turkish electoral system at large.

The enlarged influence of the nationalist agenda reflected in the electoral support for the MHP and the new electoral center of attention the İYİP has the potential to redirect Turkey’s policy orientation both domestically as well as from a foreign policy perspective towards its relations with the outside world. What is the role and importance of sectarian and ethnic minorities (as understood from a sociological perspective) in Turkish politics? How should Turkey shape its foreign and domestic policies in response to perceived or real domestic/foreign enemies? How should Turkey’s longstanding conflict in the East and southeastern provinces with the PKK be dealt with? Should Turkey remain within the Western alliance or should it seek alternatives? Should Turkey continue to democratize itself with enlarged freedoms for minorities of all kinds (ethnic, sectarian and others) or should it stick to a strictly Turkish ethnic definition of civic, democratic freedoms? Such questions are likely to pose challenges in any deliberation with the Turkish nationalists who are likely to see this emerging electoral pattern as a sign of their rising electoral support.

For this pattern to continue in the upcoming local elections scheduled for March 2019, the MHP leadership is likely to follow an emboldened strategy and keep on pressing to obtain policy changes and outcomes that appeal to their electoral base. At critical junctures, the AK Party leadership will need the support of the MHP, which may necessitate certain concessions pushing Turkey to an even more nationalistic line. Domestically, such a move will complicate prospects of democratization on all fronts. From a foreign policy perspective, a bolstered nationalistic stance is likely to exert challenges and create bottlenecks in Turkey’s relations with the U.S., the EU, Syria, and the broader Middle East.

If the elections were not held earlier, the erosion of the economic balances could have gotten out of hand before the elections and hence negatively influenced voters’ decisions

The April 2017 referendum on constitutional amendments transformed Turkey’s parliamentary system into a presidential system. The changes were designed to create a strong executive rule by a president who continues to be affiliated with a party. The office of the prime minister is replaced by an executive president who appoints a cabinet from outside the parliament. The president also appoints members of the judiciary and the higher ranks of the bureaucracy without a confirmation process in the legislature. Bypassing legislative debate and control, the new president also has the power to issue executive decrees on topics that are not regulated by existing laws. Hence, the power of the parliament will significantly decline.

Nevertheless, the parliament still has the authority to approve the central government budget and challenge presidential decrees with new legislation. Given a strict party discipline, the partisan affiliation of the president allows direct control over the ruling party organization and the legislative process. Given the new constellation of seat distribution in the TGNA, however, critical junctures that require a legislative majority will necessitate collaboration with the MHP, possibly creating some obstacles for the Erdoğan government.

Going for snap elections in June 2018 was a risky decision. Escalating tensions in the Turkish economy and rising security concerns appear to have necessitated this decision. If the elections were not held earlier, the erosion of the economic balances could have gotten out of hand before the elections and hence negatively influenced voters’ decisions. It is likely that the AK Party and MHP leadership considered the weak state of preparedness on the part of the opposition parties as an advantage as well. Given time, Meral Akşener’s challenge to the MHP from within could have hurt the electoral chances of the nationalists as well as spreading this damage into the AK Party strongholds. Similarly, by holding the elections early, the main opposition party CHP was pushed to nominate an outsider to the party administration and hence potentially create a challenge to its own leadership. The tension between the CHP administrators and Muharrem İnce came to the agenda during the election campaign, limiting the electoral potential of his candidacy.4

Most important in the decision to schedule the elections in June 2018 rather than November 2019, was the electoral risk of the municipal elections scheduled for March 2019. The second order municipal elections have always allowed local dynamics to dominate the electoral scene, making it difficult for parties to control their organizations. Typically, the incumbent parties suffer electoral losses in mid-term municipal elections. The AK Party expanded its support base in the first municipal elections of its term in 2004 from the initial general election success of 2002. However, in both the 2009 and 2014 municipal elections, AK Party support was lower than its vote shares in the preceding general election.5 Hence, if, as scheduled, the local elections originally were to be held before the presidential and parliamentary elections, then the opposition could potentially build momentum, slowly eating into the electoral constituencies of the AK Party by riding upon the wave of discontent due to rising economic tensions. This scenario was swiftly eliminated from the agenda by the decision to hold snap elections in June 2018. The results of these elections suggest that the gamble paid off, more generously for Erdoğan’s candidacy than the AK Party’s electoral campaign.

Given the past record of electoral support for the AK Party, Erdoğan’s campaign appealed to the conservative religious branch of the cultural cleavage that characterizes a deeply polarized Turkish electorate

A critical factor shaping the election results was the absence of a level playing field that kept the opposition potential limited. Despite the disadvantage of the early timing of the elections and other limitations that the opposition parties have complained about, the election results could barely be considered an electoral victory. Erdoğan won more votes than ever, but his percentage share only marginally improved over 2014 while the AK Party lost about 2.3 million votes and suffered a loss of about seven percentage points compared to November 2015. The rise of the nationalist agenda as reflected in the surprising electoral survival of the MHP suggests a flow of support from the AK Party constituency towards the MHP and possibly towards the İYİP which is also likely to have attracted votes from the CHP.

One primary cause of concern for the opposition was the state of emergency rule that was in effect during the campaign period, having been in place since the coup attempt of July 2016. A large group of academics, journalists, and university students were still being detained under questionable charges, awaiting their indictments for exceedingly long terms. The former party co-chair of the HDP Selahattin Demirtaş was one of those in jail, and his candidacy and campaign for the presidential election had to be conducted from behind bars. Many HDP elected politicians remained detained, and some of the HDP’s mayors had been replaced by trustees appointed by the government, even if the reason for their long terms of detainment were not caused by the state of emergency.

These practices created a climate of fear and self-censorship in media outlets that kept their coverage of the election campaign primarily closed to opposition perspectives. An exemplary case of bias can be found in the coverage of the campaign by the public broadcast company TRT. It is reported that during the May 14-30 period, TRT broadcast a total of 67 hours of coverage for President Erdoğan and the AK Party. In contrast, the main opposition CHP and its presidential candidate Muharrem İnce had only 6 hours of coverage. While the HDP and its presidential candidate Selahattin Demirtaş had no coverage, the newly founded İYİP and its candidate Meral Akşener received only 12 minutes of airtime.6

Biased media coverage is nothing new in Turkey, and opposition newspapers also have certain biases against the incumbent party. We have already diagnosed rising bias for an earlier election in a previous study.7 The grouping of the pro-government and opposition media outlets typically have favorable coverage for the parties of their choosing, mainly ignoring or negatively covering the other side. For the June 2018 campaign period, for example, impressive rallies organized by the opposition candidate Muharrem İnce in Izmir and Ankara involving significant numbers of supporters were largely ignored by many media outlets. As the OSCE report emphasizes: “… the media landscape comprises numerous broadcast and print outlets, most of which are considered to be associated with the ruling political party.”8 This acquired a particularly dire dimension when one of the last mainstream media groups, the Doğan Media Company, was sold to a conglomerate widely considered to be associated with the governing party ahead of the snap elections.9

Commentators and opposition officials have also complained about the security measures imposed under the state of emergency, undermining not only the possibility of campaigning freely but also potentially causing voter intimidation.10 The government claimed that the motivation behind amending the electoral laws was based mostly on security concerns in the southeastern region. Based on these concerns, in recent years, various modifications to election regulations had been introduced, including the adoption of the new electoral law in March 2018.11 The new amendments “allocated authority to municipal and election councils to register voters in polling stations different from their place of residence and to change the location or merge polling stations; modified the composition of ballot box committees (BBCs) by providing that only civil servants can be appointed as chairpersons; provided the right to every voter to request police presence in polling stations (previously, only BBC members could do so); introduced mobile ballot boxes for specific groups of voters; and amended technical aspects of election day procedures, including by mandating that the ballots that had not been stamped by the BBCs would be counted as valid and to regulate tabulation of results for coalitions…”12 During the counting of the votes cast in the April 2017 referendum, the Supreme Board of Elections adopted a decision to consider ballots improperly stamped by BBCs and those without a stamp as valid, contrary to the law, which undermined confidence in the process.”13

The OSCE report following the elections appeared to confirm earlier warnings and observed that candidates lacked equal opportunities, freedom of expression and assembly were limited, media was skewed, and changes to the electoral law (e.g., relocating polling stations on security grounds, counting ballots without official seals) were problematic.14

Besides these criticisms concerning the free and fair nature of the election campaign, there are also accusations of ballot stuffing and voting fraud.15 CHP presidential candidate İnce called on his supporters to remain attentive at their designated ballot stations and at the headquarters of the Supreme Board of Elections where votes were counted.16 However, he eventually conceded that fraud did not entirely explain his loss of the election: “Did they steal votes? Yes, they did. But did they steal 10 million votes? No.”17 He did recognize that the voting data reported by the Board and monitored by his party were similar and that the victory margin was so wide that it “cannot be explained merely by election irregularities.”18 These alleged irregularities remain to be explored by academic experts. In this study, we provide a preliminary analysis of geographic patterns in the election returns that are suggestive of some stability across elections. However, before engaging with the data analysis, we briefly note a conceptual framework for explaining voting patterns in Turkey.19

Given the past record of electoral support for the AK Party, Erdoğan’s campaign appealed to the conservative religious branch of the cultural cleavage that characterizes a deeply polarized Turkish electorate. The cultural divide that shapes this cleavage was initially argued by Mardin and Yalman in 1973.20 A sizeable minority group that historically controlled the ruling circles and that valued a vision of the good society built around positive science and a secular understanding of society and nature, opposed another, more dominant group or social coalition composed of the religious, traditional, and conservative masses and elites peripheral to the ruling circles. Historically, the former smaller group was comprised of military and civilian officials. Relying on secularist education during the Republican era, this group grew to include segments of Turkish society that thrived on the forces of globalization and remained in control of the ruling circles until quite recently. The AK Party’s electoral base remains centered in the more dominant religious, traditional, and conservative masses. The political preferences of the masses in Turkey are molded by political socialization shaped around the central tenets of this admittedly irreconcilable vision for Turkish society, one strand based on science and secularism, and the other on tradition and religion.

The political preferences of the masses in Turkey are molded by political socialization shaped around the central tenets of this admittedly irreconcilable vision for Turkish society, one strand based on science and secularism, and the other on tradition and religion

The AK Party’s rise to prominence clearly brought the numerically dominant side of this cleavage to power and slowly established control over the centrist establishment. As such, the centrist establishment that once controlled the state and its bureaucracy is no longer valid. However, despite turbulent developments in Turkish society and electoral politics for over more than half a century of elections, the Turkish polity and its party system continue to mirror the central tenets of this cultural cleavage.21 The AK Party is rooted in the pro-Islamist movement of the late 1960s which remained electorally marginal until the mid-1990s. Erdoğan was part of the younger generation that first captured the metropolitan municipalities and then became the largest party in a highly fractionalized party system. The AK Party was founded by this younger generation of politicians who capitalized on the electoral opportunity created by the financial crisis of 2001, which demolished the credibility of the existing political parties. The AK Party’s original electoral appeal was shaped not only by respect for conservative pro-Islamist ideals but also by a promise of economic growth and prosperity, commitment to EU membership and democratic reforms.

The electoral success story of the AK Party can be summarized as an effective consolidation of all branches of the peripheral elements in the Turkish party system under its banner by using a conservative social policy agenda. The lifting of the long-standing employment ban on women wearing religiously meaningful headscarves constitutes an exemplary success story in this regard.22 A critical element that complemented this social policy agenda was continuous economic growth with economic policies that remained respectful of the open market economy and fiscal discipline. The AK Party was successful in bringing about favorable economic conditions for the electorate, which led to reward at the polls in 2007 as well as 2011. However, going into the electoral campaign before June 2015, the economy remained stagnant, and the AK Party suffered a significant loss of electoral support that it cost the majority in parliament.

The AK Party’s loss of its parliamentary majority in 2015 coincided with the HDP’s first attempt to run at the national level and its successful surpassing of the 10 percent threshold for gaining representation in parliament. The collapse of the Kurdish opening process and the HDP securing 80 seats by obtaining nearly 13 percent of the votes in June 2015 necessitated a change in the AK Party’s electoral strategy and the building of new coalitions.23 Given the collapse of the Kurdish opening and the rising ethnic conflict in the east and southeastern regions of the country, the nationalist agenda and heavy emphasis on security policies remained the only viable option. The collapse of coalition negotiations following the June election provided the opportunity to go for a snap election. The change in the campaign agenda from an emphasis on the economy to an overwhelming focus on security concerns due to rising terror attacks over the summer and fall months before the November 2015 election shifted the issue advantage in favor of the AK Party. As a result, the AK Party recovered from its losses and again secured 49.5 percent of the votes, restoring its comfortable majority in the parliament.24

The collapse of the Kurdish opening process and the HDP securing 80 seats by obtaining nearly 13 percent of the votes in June 2015 necessitated a change in the AK Party’s electoral strategy and the building of new coalitions

On top of these political developments, Turkey experienced a coup attempt on July 15, 2016. This undemocratic action was thwarted thanks to the resolve of the elected government, an alliance against such an attempt by all relevant political parties, opposition to such an intervention by the public, as well as efforts by certain anti-coup cliques within the armed forces. However, the coup attempt stirred up vast uncertainties about the functioning of the Turkish political system, namely via the remnants of pro-coup, clandestine groups within the Turkish bureaucracy and judiciary. A state of emergency was declared on July 20, 2016, five days after the failed attempt. This state of emergency continued for two years, during which time Turkey held a referendum for constitutional amendments towards a new presidential system. The most recent presidential and parliamentary elections on June 24, 2018, also took place under this state of emergency. While the state of emergency may have helped Turkey deal with its security challenges and reduce tensions stemming from uncertainties in the post-coup period, it has also contributed to increasing opposition to the government as it has been mostly able to bypass the legislature through decrees. Opposition to the state of emergency also brought forward accusations of democratic backsliding in Turkey, as the rising global trend of populist backlash against democratic pluralism/globalism became a controversial point of debate among international and domestic pundits.25 All in all, the coup attempt and the government’s heavy-handed response became a topic of contention among the opposition as well as the AK Party’s supporters, creating a new foundation for electoral volatility.

In short, the outcomes of the June 2018 elections were shaped by three linked factors. One of these is the long-term ideological/cultural divide (conservative pro-Islamism vs. secularist left-leaning progressive).26 The other two factors are short-term and directly linked to the government’s performance evaluations about the state of the Turkish economy and security concerns among the electorate. The cultural divide clearly favors a conservative agenda in any Turkish election. However, rising tension on the nationalist branch of this divide was also noticeable. Many uncertainties were created by splinters from the MHP forming a party of their own under İYİP. Additionally, the refusal of the Felicity Party to cooperate and join the electoral alliance with the AK Party and MHP created additional uncertainty from the cultural divide perspective. In a similar fashion, Altun also identifies the main reasons behind the snap elections as uncertainties stemming from the coup attempt, concerns about economic stability and the expectation of early elections in both the public and private sectors, the opposition’s electoral unpreparedness, and the mismatch in the political system since the 2017 presidential system referendum.27

Most pressing perhaps were Turkey’s rapidly deteriorating economic conditions. Erdoğan appears to have followed two populist strategies in response to these economic challenges. Focusing primarily on the monetary policy, the Central Bank was put under pressure to keep interest rates low and credit cheap, fueling consumption, which in turn generated impressive growth figures. However, the credibility of the autonomy of the Central Bank deteriorated and attracted reactions from domestic as well as international market players who continuously argued that these policies will worsen Turkey’s economy and are unsustainable in the long-term. Much in line with a populist reactionism, Erdoğan and the AK Party leadership then adopted the argument that “external players and their domestic collaborators” are acting against Turkey’s interests and are responsible for the country’s financial problems.28 As such, the sustainability of budgetary discipline and maintenance of a prudent current account deficit were mostly ignored until after the election. However, with local elections approaching, such concerns are bound to resurface, and worsening economic conditions could affect electoral outcomes.

During the June 2018 election campaign, Erdoğan was successful in downplaying Turkey’s economic problems. His agenda remained focused on the AK Party’s successful performances in its earlier years in power. He was also successful in shifting public attention to promises of future grand plans such as the Canal İstanbul project. He remained attentive to the fight against terrorism, apparently to appeal to the nationalist constituencies.29 Erdoğan’s populist rhetoric appears to have created a strengthened vision that the outside world is conspiring against Turkey. For instance, when discussing the economy, he contextualized the debate within a heightened sense of security and external threats. Such a discourse of securitized economic challenges is likely to conveniently help shift the blame for future economic crises toward external financial circles and their collaborators within. Surprisingly, the opposition candidates’ efforts to blame the single-party government for the country’s rising economic difficulties did not pay off. Both opposition candidates, İnce as well as Akşener were unable to focus their attention on the economy while Erdoğan maintained a security discourse even when discussing economic issues. More critically, both opposition candidates were unable to reach out to the masses via their media campaigns. As stated earlier, their coverage in the printed media, as well as the TV channels, remained very limited. Instead, both candidates held large numbers of provincial rallies and engaged in extensive online campaigning. Not surprisingly, both kinds of campaigning reached out mainly to their own core constituencies instead of voters who might have been likely to switch their votes away from the ruling party or the MHP. The opposition leaders organized numerous provincial rallies while Erdoğan remained focused on selective TV appearances and favorable coverage in the media.

Looking back, it is not possible to pin down a specific economic policy agenda advocated by the opposition that could have appealed to large masses of voters. The unfocused electoral campaigns of the opposition, which were largely ignored by the media, appear to have forced them into an old-fashioned 1970s approach to campaigning built around provincial rallies. Such a strategy could have worked if it had relied on effective party organization that could expand its reach into constituencies of voters likely to switch away from the ruling party and its leadership. Such a constituency did indeed exist, as shown by declining support for the AK Party. However, when it came to discrediting the leadership of Erdoğan, this program simply did not gain enough credibility and traction among the voters. Both opposition candidates only had about a month to establish their campaign policy agenda, which was not long enough to build a momentum of following amongst the voters. Given the unpreparedness of the opposition, the snap election decision worked once again in favor of Erdoğan, who successfully deflected any policy-specific momentum against his incumbency.



Mobilized Vote in 2018

To visualize geographic patterns in the 2018 elections, we analyzed ballot-box level data from all 972 districts in the country, excluding ballot boxes in prisons, custom borders and those in foreign countries. In total, there were 180,065 ballot boxes across all of the districts.30 To discuss the geography of the elections, we first consider what we call “mobilized ballot boxes.”31 These are the ones where all valid votes were cast for one single party or candidate. Several reasons come to mind that can account for this type of behavior. From the perspective of the communities that cast these votes, one could conjecture that these voters seek a certain recognition of their commitment to the party or candidate in question. In return, these communities may be expecting favors of different types, such as access to, or priority in, some service provision. Security concerns and the need for protection in certain areas may motivate such behavior, while in others signaling an ideological commitment to the party or candidate in question may underlie such behavior. Lack of different choices can only be seen as an unusual circumstance in democratic settings where party competition typically creates divided communities. However small a community these ballot boxes may represent, they still are a rare occurrence in the context of recent Turkish elections. Table 1 below presents mobilized ballot boxes for parties (and affiliated presidential candidates) in comparison with the two elections held in 2015.

Table 1: Mobilized Ballot Boxes


Table 1 suggests that the presidential candidates of the AK Party (Erdoğan) and CHP (İnce) were able to mobilize more ballot boxes than their parties. In fact, President Erdoğan was able to get all the votes in an impressive total grand of 500 ballot boxes across the country. Although these ballot boxes comprise only a small fraction of the total votes in the country, they nevertheless represent a significant jump compared to the earlier November 2015 election. Closest to such figures was the HDP’s success across 369 ballot boxes in November 2015. Additionally, these 500 mobilized ballot boxes voting entirely for Erdoğan were not concentrated in a specific region. On the contrary, Erdoğan was able to mobilize voters across many different regions of the country. Figure 3 below plots districts according to four quartiles of Erdoğan’s votes and the total number of mobilized ballot boxes in that district.

Figure 3: Erdoğan’s Vote Percentage and Number of Mobilized Ballot Boxes



As expected, Figure 3 shows a high number of mobilized ballot boxes where Erdoğan was relatively more successful. The geographic pattern of Erdoğan’s success is variegated: there is a belt in the inner Black Sea region with a high number of mobilized ballot boxes, ranging from the districts of Bolu in the west to Erzurum in the east. Also, there are pockets of high support in Inner Anatolia and the Inner Aegean region in provinces such as Kütahya, Afyon, and Konya, but the number of mobilized ballot boxes there was relatively low. Lastly, a third region where Erdoğan received high support was in the southeastern Anatolian provinces southeast and southwest of Tunceli. In certain areas, namely Malatya, Elazığ, Şanlıurfa, Bingöl, Bitlis, and Siirt, support for Erdoğan was relatively high. More interestingly, in three districts of Şanlıurfa that are close to the Syrian border, mobilized ballot boxes were extremely high with 36 such instances in Akçakale, 39 in Harran and 34 in Haliliye district.

A similar visual analysis of mobilized ballot boxes does not reveal a systematic pattern, as the number of mobilized ballot boxes for other presidential candidates is very low, the highest being 31 such ballot boxes for the CHP’s candidate İnce. Interestingly, more than half of these mobilized ballot boxes (17 out of 31) were from the Sivas Divriği district. In fact, four of the CHP’s nine mobilized ballot boxes in the parliamentary elections were also from Divriği.

500 mobilized ballot boxes voting entirely for Erdoğan were not concentrated in a specific region. On the contrary, Erdoğan was able to mobilize voters across many different regions of the country

The HDP was able to mobilize more ballot boxes in 2015, but in the 2018 elections, the party and its candidate Demirtaş could not successfully mobilize the same number of ballot boxes. While there were 229 ballot boxes where the HDP received all valid votes in 2015, the party and its candidate failed to mobilize the same number of ballot boxes in 2018. Both the HDP and Demirtaş mobilized six boxes in the districts of Van, Diyarbakır, and Şanlıurfa while the HDP additionally mobilized two ballot boxes in Mazgirt Tunceli.

We also checked mobilized neighborhoods (and villages) where the whole community voted for a candidate or a party. These are the smallest meaningful administrative units in the Turkish context. There were 49,301 neighborhoods and villages in the 2018 elections. On average, there were 1,142 registered voters in these units and 984 valid votes. The AK Party was able to mobilize 63 neighborhoods, and more than half of these neighborhoods were in Şanlıurfa (27) and Adıyaman (7) provinces. In comparison, the CHP mobilized only nine neighborhoods, and six of them were in Sivas province. Table 2 below presents the number of mobilized neighborhoods by parties and their presidential candidates.

Table 2: Mobilized Neighborhoods and Villages by Parties and their Candidates


Table 2 shows that Erdoğan and İnce were more competent in mobilizing voters across neighborhoods compared to their parties’ performance. These figures show that the most successful politician who was able to mobilize voters was Erdoğan, with 413 neighborhoods fully supporting his candidacy.

Mobilized ballot boxes and neighborhoods, defined as getting all of the votes in a given unit, is a very stringent measure of (lack of) competition. Given that there were 312 registered voters and 275 valid votes per ballot box on average and 1,142 registered and 984 valid votes per neighborhood, it may be far-fetched to expect all voters to support the same party or candidate. Therefore, a more nuanced measure of competition can be the effective number of parties (ENP).32 ENP measures the effective electoral competition in a given electoral district. ENP depends on the number of candidates (parties) competing in a given district and the distribution of votes across candidates (parties). For instance, in a hypothetical election where three candidates run, if Candidate A gets 96 percent of all votes while Candidates B and C share two percent each, ENP is expected to be very low, close to 1, indicating that effective competition is meager. If these same candidates were to each get 33.3 percent of votes, then ENP would be approximately three, meaning that three effective candidates are competing in the district.

In the most recent elections, the number of parties and alliances running for parliament (8 parties, 2 alliances, and a various number of independents), and the number of presidential candidates running for the presidency (6 candidates) was not the same. The two elections were qualitatively different regarding their system of interest aggregation (majoritarian with a run-off vs. proportional) and in terms of the potential motives of vote choice. Therefore, we present two maps where we depict ENP levels across quintiles. The ENP value across 972 districts was 2.92 on average for the parliamentary elections, and it was 2.09 for the presidential elections, indicating that the level of competition in the presidential election was lower than that of the parliamentary election. ENP for parliamentary and presidential elections are highly correlated (0.83). As expected, we see the competition level being higher in the parliamentary than in the presidential elections.

The highest level of competition in both elections appears to be in the western coastal provinces and in the Marmara region

While the lowest competition was categorically in the districts of southeastern Anatolia (in districts such as Adıyaman Sincik, Diyarbakır Lice and Muş Varto), the highest competition was in the Mediterranean region in districts such as Mersin Tarsus, Hatay Erzin, and Adana Ceyhan with an additional district from Inner Anatolia (Kayseri Sarız). Although ENP does not show which parties compete or which parties dominate, the additional analyses we provide below will show how (lack of) electoral competition took place in various districts.

Figure 4 visualizes the geographic patterns of electoral competition across districts. The highest level of competition in both elections appears to be in the western coastal provinces and in the Marmara region. Most districts in these provinces are in the top twenty percentile of competition as reflected in the ENP figures. In numerous coastal districts from Hatay to Bursa, the competition was also relatively high. Additionally, a pocket of high competition is visible in the northeast in the districts of Kars, Ardahan, and Artvin. On the other side of the scale, we observe some regions lacking competition. Specifically, southeastern Anatolia had a shallow level of competition, especially in the parliamentary elections. In various districts, competition seems to have been limited by the domination of the HDP, while the AK Party was a close competitor and none of the other parties garnered tangible results in the region.

Figure 4: Effective Number of Parties across Districts




Geography of the 2018 Elections

To understand how the most recent elections fit the recent historical patterns of voting in Turkey, we traced differences in the percentage of the vote across parliamentary elections from November 2015 to June 2018. The results are presented in Figure 5 below.

Figure 5: Vote Percentage Differences from November 2015 to June 2018 Parliamentary Elections

Figure 5 shows where each party gained or lost across districts. The AK Party, for instance, appears to have suffered most in its core geographic constituency from the inner Aegean region to central Anatolia, and the Black Sea to the Mediterranean region, while gaining significantly in the southeastern Anatolian districts. One surprising result we mentioned above was the survival of the MHP despite expectations to the contrary. The MHP appears to have capitalized upon the losses of the AK Party in the central Anatolian districts. In the districts where the CHP and MHP lost in eastern Anatolia, the İYİP gained some ground. The HDP lost considerable support in southeastern Anatolia, considered to be the party’s stronghold. Admittedly, Figure 5 lacks some crucial information we mentioned in earlier sections, such as the creation of a new politically relevant party (İYİP) as well as differences in the electoral system (electoral alliances between parties).

To better elucidate the direction of party competition, we combined the 2018 results with a previous parliamentary election (November 2015), and the first popular presidential election (August 2014). We conducted a cluster analysis with which we can observe clusters of districts that acted similarly across these elections. Our cluster analysis attempts to expose natural groupings in the district level election results across the four elections in our study. We use this exploratory technique to help us visualize and understand how district groupings (clusters) occurred in the four critical elections since 2014. We use the k-means partition clustering method, breaking the total number of districts33 into non-overlapping clusters. With this method, each district is eventually assigned to the cluster that has the closest average to the district in question. This is an iterative process in which groupings continue until no district changes its cluster.34 Using this method, we decided to conduct the analysis with 6 clusters to understand the extent to which Turkish districts vote similarly on average. Of course, increasing (reducing) the number of clusters could yield more (less) fine-grained results, but in our various trials, we see that 6 clusters are optimal in explaining the multiple groupings and geographic patterns in these four elections.35 In order to visualize these clusters, we grouped them into two larger classifications where clusters 1 and 2 are districts where the AK Party and its candidate won the district, clusters 5 and 6 consist of districts where the pro-Kurdish HDP and its candidate had considerable support, while clusters 3 and 4 are districts where the CHP competed with the AK Party (cluster 3) or the party dominated the district in elections. Figure 6 plots these clusters on a district-level map and Figure 7 plots the average vote percentages of politically relevant parties across the four elections and six clusters.

The western coastal regions together with the east and southeastern provinces vote against the conservative nationalist and pro-Islamist agenda and its candidates. The rest of the country predominantly supports the conservative and nationalist parties and their candidates

First of all, there is a high variation in the number of districts across clusters, from 59 districts in Cluster 5 to 282 districts in Cluster 2. In other words, there are relatively more insulated geographic pockets such as clusters 5 and 6 with lower numbers of districts, as well as clusters that can be very comprehensive such as clusters 2 and 3 with respectively 282 and 271 districts. Relatively more insular clusters are those that have a pattern of party competition that does not fit the rest of the country. For instance, cluster 6 consists of 60 districts where the HDP and its candidate Demirtaş dominated the electoral scene. This cluster includes 13 districts from Diyarbakır (out of 17) as well as 1 district from Kars and Şanlıurfa. Similarly, cluster 5 has 59 districts where the AK Party and HDP competed in the parliamentary elections. On the other side of the spectrum, clusters 2 and 3 are those where we observe similar patterns of electoral competition across a considerable portion of districts (57 percent of all districts are in these two clusters). For instance, cluster 2 has 282 districts where Erdoğan dominated the presidential elections (with 65.22 and 67.43 percent in 2014 and 2018 respectively), and his party had a considerable margin against competitors (62.8 percent in 2015 and 52.7 percent in 2018).

Figure 6: District Clusters Depicting the Political Geography of Four Elections, 2014-2018


Figures 6 and 7 add a temporal dimension to the maps presented so far. We not only see how districts of similar election results come together in different clusters, but we can also trace how, from the presidential election in 2014 to November 2015, and the two elections in 2018 correspond across different district groupings. For instance, in the pro-AK Party grouping of clusters 1 and 2 where the districts are colored so that the darkest orange corresponds to the highest support for Erdoğan and the AK Party in cluster 1, we observe an already quite high support for the AK Party in 2015 (77.4 percent) which is also around 79.7 percent on average in 2014 for Erdoğan. However, we see that these averages are typically up by about one to two percentage points in the 2018 presidential election. While Erdoğan’s support rose from 2014 to 2018 in these clusters, we observe that his party lost considerable support. The AK Party’s support ranged from 77.4 to 62.8 percent in these two clusters in 2015. However, the same districts showed on average lower levels of support for the AK Party in 2018 that range from 64.2 to 52.7 percent. Where did this support go from 2015 to 2018? It appears that the CHP average vote shares remained stable in these two clusters. However, the average level of support for the MHP appears to have risen (from 9.9 to 15.8 percent in cluster 1, and 14.7 to 15.3 percent in cluster 2), while the İYİP also won considerable support, ranging from about 6.7 to 10.4 percent in clusters 1 and 2 respectively. Hence, in this region comprised in total 470 out of 970 districts, Erdoğan’s personal support rose in the presidential elections while the AK Party’s support lost some ground to the İYİP, and, because of the electoral alliance, the MHP garnered some additional votes, especially in districts where the AK Party dominates (cluster 1).

Figure 7: Six Clusters of Voting Support across Turkish Districts


The pro-CHP cluster 4 corresponds to the highest support for İnce and the CHP, while cluster 3 is where the party competes with the AK Party. In the larger cluster 3, we observe that on average İhsanoğlu’s support levels could not be obtained by İnce in 2018. In cluster 3 districts, we observe a sizeable level of support for İhsanoğlu in 2014 (46.9 percent on average) that seem to have gone down in most districts (İnce’s average support was 33.7 percent in 2018). Similarly, for İnce’s party the CHP, we see that on average its level of support declined from November 2015 to June 2018 from 30.5 percent to 27 percent. Also, a small number of highly insular districts in the east are observed for cluster 4 (2 districts in Tunceli and Malatya each). However, this is a lightly populated part of cluster 4 and, as expected, most of the districts in this cluster are concentrated in the western parts of Turkey. Additionally, except for the 9 districts in cluster 4 where MHP support rose slightly, we observe that the MHP lost considerable support (from 11.9 percent to 6.1 percent), which appears to have gone primarily to the İYİP. Cluster 4 is also where the AK Party lost some support compared to November 2015, declining from 27.7 percent to 25.4 percent.

We observe another pattern in southeastern Anatolia for clusters 5 and 6. Cluster 6 consists of those districts dominated by the HDP. Cluster 5 has districts where the HDP and AK Party compete. Additionally, the region has four distinctive districts where the AK Party dominates the competition (Mardin Yeşilli, Siirt Tillo, Bitlis Mutki, and Muş Hasköy). In both clusters 5 and 6, Demirtaş’s support declined from 2014 to 2018. In cluster 5, where Demirtaş competed with Erdoğan, the former’s support declined from 41.5 percent to 36.4 percent; in cluster 6, where the HDP and its candidate dominated the elections, Demirtaş’s support declined from 73.3 percent to 65.5 percent.

The MHP’s pressure on Erdoğan and the AK Party may eventually end up causing the demise of the AK Party-MHP alliance, especially if the AK Party requires new allies to legislate and execute specific economic/domestic policies

All in all, our cluster analysis reveals a pattern that has long been observed in Turkey. The western coastal regions together with the east and southeastern provinces vote against the conservative nationalist and pro-Islamist agenda and its candidates. The rest of the country predominantly supports the conservative and nationalist parties and their candidates. In 2018 we observe the same pattern. Even though AK Party support has eroded in the inner Anatolian districts, primarily in favor of the MHP, the emergence of the İYİP appears to have failed to change this pattern. İnce’s candidacy was not able to make a dent in the core conservative and nationalist constituencies of the central Anatolian and Black Sea regions. İnce’s support had to share İhsanoğlu’s earlier support bases, which were already on the losing side, with the İYİP as well as the MHP. As such, the coastal and inner Aegean districts together with the Marmara districts were not enough to change the election outcome despite the population advantage that these districts bring with them. This is in part due to Erdoğan’s strong showing in larger metropolitan centers such as Istanbul. In 126 districts of Turkey’s five large metropolitan areas (İstanbul, Ankara, İzmir, Bursa, and Adana), Erdoğan categorically garnered more votes than his party. On average, he had 9.8 percent more than his party in these 126 districts.

Given the recent tumultuous period Turkey has experienced since the coup attempt in 2016, as well as recent discussions of democratic backsliding and economic downturn, we have no expectation of observing a sudden shift in this overall depiction in the near future. Even after all of these political and economic challenges, the available official election results suggest that the dominant AK Party and its leader Erdoğan continue to be competitive in Turkish elections.



Concluding Expectations

Turkey faces a long list of challenges in the aftermath of the June 2018 elections. The weak state of the economy dominated the agenda immediately after the elections. The opposition was quick to argue that the snap election decision was motivated by the fragile state of the economy and insinuated that serious electoral losses were bound to take place if only the voters had foreseen the post-election state of the economy. This is a naturally appealing argument from the opposition perspective. However, the economic voting hypothesis critically depends on holding the incumbent responsible for the development of the economy. It is not enough that the state of the economy is seen as weak by the masses. What is necessary is that the responsibility for this unsatisfactory state of the economy is attributed to the incumbent.

Additionally, for the opposition’s arguments to take hold of the public imagination, the alternative players need to be seen as potentially performing better if they were to come to power. In other words, it is not enough that retrospective evaluations are negative for the incumbent, necessitating an electoral punishment. From a prospective perspective, it is also necessary that an alternative party or candidate is seen as potentially performing better.

Complicating this picture is the partisan nature of both retrospective as well as prospective evaluations. AK Party supporters are expected to see the economy with more rose-colored glasses than those of the opposition. The geography of the election results summarized above hints at possible developments in the future that are likely to shape economic evaluations. For instance, clusters 1 and 2 in Figure 6 above, where the conservative and nationalist constituencies reside, is likely to have a biased perspective in their evaluations of the economy that favors the AK Party government and President Erdoğan. This region is relatively more closed to global influences on the Turkish economy. When the economic shock hits the Turkish economy, its impact is most likely to be most directly and immediately felt in the western coastal regions (clusters 3 and 4). Cluster 5 and 6 where the ethnic vote behind the HDP is shaping election results are more likely to be mobilized by foreign policy developments in Syria and Iraq and democratization efforts in the future.

Democratization measures, such as ensuring regulatory and institutional autonomy and freedoms, the amelioration, and sustenance of media pluralism, and tolerance toward ethnic and sectarian minorities, are still necessary steps required to improve Turkish democracy. An unfortunate development before the June 2018 election was the loss of consensus about the principles of administering elections in the country.36 That consensus can be traced back to the initial years of multi-party competitive elections in the country, and it is absolutely necessary for the legitimacy of democratic elections. Maintenance of a level playing field with open and free access to all legal means of campaigning for all political parties has become increasingly problematic in recent elections. These problems form a foundation for rising polarization as well. Hence a new effort must be made to build consensus around the administrative principles of elections and to create a campaign environment of free competition in the country.

The decades-old Kurdish problem and the challenges it poses for Turkish society are increasingly coming under the influence of developments on the Syrian and Iraqi front, and they still await political attention. Developments on this front are most likely to be felt by the electoral preferences of voters residing in cluster 6 districts in Figure 6 above before it trickles down to the rest of the country. The rise of the nationalist agenda and its electoral support is likely to impose constraints upon any development on this front. Since local elections will take place over the next few months, the implications of these issues for local elections are critical to observe. The HDP as a party still appears to be powerful in the east and southeastern provinces and is likely to compete with the AK Party for the local elections.

Although it would be an exaggeration to claim that the new era will be shaped by coalition politics, since there are no institutional formalities to govern such a regime, it is still impossible to ignore the implications of coalition logic in the ensuing period shaping many of the upcoming critical decisions

From a foreign policy perspective, Turkey faces significant security threats and challenges, and these are especially relevant for different vicinities in advance of the local elections. We already witnessed that Turkey’s relations with the U.S. are directly linked to developments in the Turkish economy, and hence are directly relevant for domestic political developments. By directly targeting the U.S. president and keeping his policies responsible for the way Turkish economic conditions have changed, President Erdoğan succeeded in shifting responsibility to external actors. This provides yet another example as to how developments in the economy need not directly have the expected deteriorating influences upon the incumbent politicians. The pattern of using foreign policy developments for domestic political gain may continue in the foreseeable future. The fact that the nationalist constituencies and their policy agenda is arguably more relevant than ever for the maintenance of the domestic political balance will impose restrictions on the Erdoğan government and its ability to respond to shifting foreign policy developments.


It is likely that the MHP leadership will try to put a stamp of approval on critical issues concerning pork barrel distribution and patronage, the sensitive issue of amnesty, domestic policies targeting the Kurdish minority, as well as foreign policy issues. As such, the new term of President Erdoğan and his party is likely to lean toward nationalist sensitivities. These sensitivities may in all likelihood remain closed to efforts of further societal openings and democratization in various issues concerning ethnic and sectarian minorities, independence of institutions, the inclusion of opposing perspectives in decision making, and improved collaborative relations with Western powers. Given the dire straits that the Turkish economy will have to navigate, these nationalist impositions are likely to create serious challenges. The MHP may seem like a kingmaker in this alliance for now, but economic problems may potentially cause a rupture in these relations. So, the MHP’s pressure on Erdoğan and the AK Party may eventually end up causing the demise of the AK Party-MHP alliance, especially if the AK Party requires new allies to legislate and execute specific economic/domestic policies. The MHP’s influence on current Turkish politics depends on the delicate balance of this alliance and the availability of alternative political actors that can help the Erdoğan government.

It is somewhat ironic that in this era of centralized and emboldened presidential powers, we again find ourselves discussing the necessity of collaborative/coalition politics. The formation of electoral alliances was the first formal step, and the election results appear to underline a seamless continuity in the same direction. Although Erdoğan secured a second term in office, his win came only through a coalition with support from the nationalists. The AK Party suffered a significant loss, relinquishing its single-handed control of the parliament. As such, the monolithic rule of the AK Party appears to have come to an end. The AK Party will no longer be able to pass any legislation on its own, nor will it be able to have Erdoğan’s budget approved by the parliament without the support of the nationalists. The minimal winning coalition is now comprised of the AK Party plus other political parties, unlike previous AK Party tenures. Any new initiative toward resolving the Kurdish conflict is likely to be shaped in accordance with the sensitivities of nationalist political actors. In this new parliamentary configuration, the MHP not only proved that it remains above the nation-wide electoral threshold of 10 percent, but it also became the key supporter for the smooth functioning of the presidential system. Even with the AK Party’s electoral alliance with other parties remained as a possibility, MHP’s strong showing in the June election raised its bargaining power considerably.

Although it would be an exaggeration to claim that the new era will be shaped by coalition politics, since there are no institutional formalities to govern such a regime, it is still impossible to ignore the implications of coalition logic in the ensuing period shaping many of the upcoming critical decisions. As long as the implications and necessities of a collaboration between the nationalist agenda of the MHP and the ruling AK Party and its leader Erdoğan’s preferences are observed and obeyed, we should expect smooth sailing in the executive office. Otherwise, it is likely that high seas will necessitate new precautions even under the new presidential system, which we took almost for granted to bring about stability and unified control over the executive office.  




  1. All results are transformed into a comparable 0-100 percent range for both Figures 1 and 2.
  2.  For some stories before the election on competitiveness see, Louise Miner, “Turkey: Erdogan Facing a Tough Election,” Euronews, (June 15, 2018), retrieved from; Carlotta Gall, “Turkish Opposition Hopes 4 Parties Are Mightier Than 1 Against Erdogan,” The New York Times, (June 9, 2018), retrieved from
  3.  After the elections, the MHP’s leader Bahçeli published an advertisement, criticizing nine different polling companies as well as journalists for manipulating perceptions about his party. Also, before the elections he issued various statements criticizing these companies for “black propaganda” against the MHP. See, for instance, “Bahçeli’den Anket Şirketlerine Tepki,” Dünya, (June 8, 2018), retrieved from
  4.  Before İnce’s candidacy, the CHP appeared to be considering the previous president and one of the founders of the AK Party, Abdullah Gül, as a candidate. This consideration by higher party circles also created further tensions within the party. See, Ayla Jean Yackley, “Former Teacher Stands against Turkey’s President Erdogan,” Financial Times, (May 4, 2018), retrieved from
  5.  Here, we are comparing the general election results with provincial council election results.
  6.  For details about candidates’ coverage in media, see, “İlkesi ‘Tarafsız’ Yayıncılık: Erdoğan’a 67 Saat Ayıran TRT’de Demirtaş’a Saniyelik Bile Yer Verilmedi!” T24, (June 24, 2018), retrieved from,641792.
  7.  For details of how the media can polarize the electorate even within a short campaign period, see, Ali Çarkoğlu, Lemi Baruh, and Kerem Yıldırım, “Press-party Parallelism and Polarization of News Media during an Election Campaign: The Case of the 2011 Turkish Elections,” The International Journal of Press/Politics, Vol. 19, No. 3 (2014), pp. 295-317.
  8.  Office for Democratic Institutions and Human Rights, “Republic of Turkey: Early Presidential and Parliamentary Elections 24 June 2018 and ODIHR Needs Assessment Mission Report,” OSCE, (2018), retrieved from, p. 2.
  9.  See, details of this account are given at Nick Holdsworth, “Turkish Pro-Government Company Buys CNN Local Turkish Channel,” The Hollywood Reporter, (March 26, 2018), retrieved from; “Doğan Grubu’nun Satışı Economist’te: ‘Son Bağımsız Medya, Rejim Yanlısı Şirketçe Kapıldı,’” BBC, (March 30, 2018), retrieved from; Dorian Jones, “Turkey’s Ruling Party Extends Control Over Media,” VOA News, (March 23, 2018), retrieved from
  10.  “Turkey Moves to Extend State of Emergency Despite Calls for ‘Immediate’ Lifting,” Sputnik, (April 18,
    2018), retrieved from
  11.  Ayla Jean Yackley, “Turkey Changes Electoral Law in Boost for Ruling Party,” Al-Monitor, (March 15, 2018), retrieved from
  12.  Office for Democratic Institutions and Human Rights, “Republic of Turkey,” p. 5.
  13. Office for Democratic Institutions and Human Rights, “Republic of Turkey,” p. 5.
  14. “Opposition Denied Equal Conditions in Turkey Elections: OSCE Observers,” Reuters, (June 25, 2018), retrieved from
  15. Discussions of election forensics for recent Turkish elections increasingly appear in the international literature. See Peter Klimek al, “Forensic Analysis of Turkish Elections in 2017–2018,” PloS One, Vol. 13, No. 10 (2018), retrieved from; Peter Klimek, “Election Forensic Analysis of the Turkish Constitutional Referendum 2017,” Cornell University Library, (June 29, 2017), retrieved from; Walter Mebane Jr, “Brief Election Forensics Report on the 2015 Legislative Elections in Turkey,” University of Michigan, (2016), retrieved from
  16.  “İnce: Sandıkları Terk Etmeyin,” NTV, (January 24, 2018), retrieved from,RB6bJjTo6kG3mAq51CCHiQ.
  17.  “CHP’s Presidential Candidate İnce Accepts Defeat, Vows to Keep Fighting,” Hürriyet Daily News, (June 25, 2018), retrieved from
  18.  For an account on İnce’s acceptance of the results, see, Kemal Kirisci and Ali Çarkoğlu, “How to Make Sense of the Turkish Election Results,” The National Interest, (June 27, 2018), retrieved from
  19.  For more extensive discussion on voting behavior in Turkey see, Ali Çarkoğlu, “Voting Behavior in Turkey,” in Metin Heper and Sabri Sayari (eds.), Handbook of Modern Turkey, (London: Routledge, 2012), pp.160-170.
  20.  On the center-periphery divide see, Şerif Arif Mardin, “Center-periphery Relations: A Key to Turkish Politics?” Daedalus, Vol. 102, No. 1 (1973), pp. 169-190. On the concept of kulturkampf in Turkish politics see, Nur Yalman, “Some Observations on Secularism in Islam: The Cultural Revolution in Turkey,” Daedalus, Vol. 102, No. 1 (1973), pp. 139-168; Ersin Kalaycıoğlu, “Kulturkampf in Turkey: The Constitutional Referendum of 12 September 2010,” South European Society and Politics, Vol. 17, No. 1 (2011), pp. 1-22.
  21.  For macro-level studies that argue for electoral stability and continuity over time see, İlhan Can Özen and Kerem Ozan Kalkan, “Spatial Analysis of Contemporary Turkish Elections: A Comprehensive Approach,” Turkish Studies, Vol. 18, No. 2 (2016), pp. 358-377; Arda Can Kumbaracıbaşı, “An Aggregate Analysis of the AKP Vote and Electoral Continuities in the Turkish Party System,” Turkish Studies, Vol. 17, No. 2 (2016), pp. 223-250.
  22.  For the issue of the turban ban see, Ali Çarkoğlu, “Women’s Choices of Head Cover in Turkey: An Empirical Assessment,” Comparative Studies of South Asia, Africa and the Middle East, Vol. 29, No. 3 (2009), pp. 450-461; Ali Çarkoğlu, “Public Attitudes towards the Türban Ban in Turkey,” Utrecht Law Review, Vol. 6, No. 2 (2010); Ersin Kalaycioğlu, “The Mystery of the Türban: Participation or Revolt?” Turkish Studies, Vol. 6, No. 28 (2005), pp. 233-251.
  23.  See, Kemal Kirişçi, “The Kurdish Issue in Turkey: Limits of European Union Reform,” South European Society and Politics, Vol. 16, No. 2 (2011), pp. 335-349; Tuncay Kardaş and Ali Balci, “Inter-societal Security Trilemma in Turkey: Understanding the Failure of the 2009 Kurdish Opening,” Turkish Studies, Vol. 17, No. 1 (2016), pp. 155-180.
  24.  See, Ali Çarkoğlu and Kerem Yıldırım, “Election Storm in Turkey: What Do the Results of June and November 2015 Elections Tell Us?” Insight Turkey, Vol. 17, No. 4 (2015); Selim Erdem Aytaç and Ali Çarkoğlu, “The Impact of Terror Attacks on Voting: Evidence from a Panel Study during the Turkish Elections of 2015,” (2018) working paper; and Ersin Kalaycıoğlu, “Two Elections and a Political Regime in Crisis: Turkish Politics at the Crossroads,” Southeast European and Black Sea Studies, Vol. 18, No. 1 (2018), pp. 21-51, on the shifting votes from June to November 2015 elections.
  25.  For some recent discussions on this topic, see studies by Ziya Öniş, “Turkey’s Two Elections: The AKP Comes Back,” Journal of Democracy, Vol. 27, No. 2 (2016), pp. 141-154; Kürşat Çınar, “Local Determinants of an Emerging Electoral Hegemony: The Case of Justice and Development Party (AKP) in Turkey,” Democratization, Vol. 23, No. 7 ( 2016), pp. 1216-1235; Murat Somer, “Understanding Turkey’s Democratic Breakdown: Old vs. New and Indigenous vs. Global Authoritarianism,” Journal of Southeast European and Black Sea, Vol. 16, No. 4 (2016), pp. 481-503; Kerem Öktem and Karabekir Akkoyunlu, “Exit from Democracy: Illiberal Governance in Turkey and Beyond,” Journal of Southeast European and Black Sea,
    16, No. 4 (2016), pp. 469-480; Berk Esen and Şebnem Gümüşçü, “Rising Competitive Authoritarianism in Turkey,” Third World Quarterly, Vol. 37, No. 9 (2016), pp. 1581-1606; Kalaycıoğlu, “Two Elections and a Political Regime in Crisis”; İhsan Yılmaz and Galib Bashirov, “The AKP after 15 Years: Emergence of Erdoganism in Turkey,” Third World Quarterly, (2018), pp. 1-19.
  26.  For an application of this framework to the 2015 general elections, see Aytaç and Çarkoğlu, “The Impact of Terror Attacks on Voting: Evidence from a Panel Study during the Turkish Elections of 2015”; Kalaycıoğlu, “Two Elections and a Political Regime in Crisis.”
  27.  Fahrettin Altun, “Laying the Cornerstone for a New Turkey: The June 24 Elections,” Insight Turkey, Vol. 20, No. 3 (2018), pp. 89-103.
  28.  For Erdoğan’s discourse on the economy during the campaign period see, Laura Pitel, “Recep Tayyip Erdogan: Turkey’s Strongman Grapples with the Markets,” Financial Times, (May 25, 2018), retrieved from
  29.  Umar Farooq, “Will Istanbul’s Massive New Canal Be an Environmental Disaster?” National Geographic, (March 28, 2018), retrieved from
  30.  Note that in the İncekavak Neighborhood of Diyarbakır’s Hazro district, ballot box number 1017 was completely empty according to the official results. So, any calculation of mobilized voting shows all parties getting all votes (0 votes). In the analysis presented below, we corrected for this minor error.
  31.  See, Cenap Nuhrat, “Türkiye Köylerinde Olağandışı Oy Verme [Exceptional Voting in Turkish Villages],” Ankara Üniversitesi SBF Dergisi, Vol. 26, No. 1 (1971), pp. 219-244; Murat Erdoğan, “Olağandışı Oy Verme Davranışı ve Mobilize Katılım: Güneydoğu Anadolu Projesi Bölgesinde Toplu Oy Veren Köyler Üzerine Bir Araştırma,” Ankara Üniversitesi SBF Dergisi, 47, No. 1, (1992), pp. 277-310, on earlier analyses of mobilized voting in Turkey.
  32.  ENP was calculated using Laakso and Taagepera’s (1979) measurement. The formula for calculating ENP is:


where n is the number of parties competing in the district and p is the vote share of each party. We used ENP for a previous study where we analyzed the change in competitiveness from June to November 2015 elections. For details of measurement see, Çarkoğlu and Yıldırım, “Election Storm in Turkey: What Do the Results of June and November 2015 Elections Tell Us?”; Özen and Kalkan, “Spatial Analysis of Contemporary Turkish Elections: A Comprehensive Approach,” also uses ENP to trace geographic patterns of electoral competition since 2002 across “more than 900” districts (p. 361). However, their effective number of observations remains unclear as there have been multiple instances of administrative redistricting since 2002.

  1.  We used 970 districts for which we have available data from 2014 to 2018. In 2017, a decree was issued to create two new districts (Artvin Kemalpaşa and Aksaray Sultanhanı). We treat them as missing in the cluster analysis, and focus on the districts that existed back in 2014 rather than imputing results for these two districts that did not exist.
  2.  See, Ali Çarkoğlu, “The Geography of the April 1999 Turkish Elections,” Turkish Studies, Vol. 1, No. 1 (2000), pp. 149-171; Jefferson West II, “Regional Cleavages in Turkish Politics: An Electoral Geography of the 1999 and 2002 National Elections,” Political Geography, Vol. 24, 4 (2005), pp. 499-523, for the use of cluster analysis in geographical analysis of election results.
  3. Tests were undertaken using the Calinski-Harabasz pseudo-F index stopping rules where we tried from 2 to 20 cluster solutions also suggest 6 clusters to be optimal for increasing across-cluster variation and reducing within-cluster variation (i.e., the distinctiveness of clusters). Additional tests with Duda-Hart with hierarchical clusters based on 2 to 20 cluster solutions also show 6 clusters to be the optimal number. Based on these tests, we created 6 clusters of Turkish political geography.
  4. See OSCE report for a non-partisan evaluation of the issues involved in the present state of election administration in Turkey.


Labels »  

We use cookies in a limited and restricted manner for specific purposes. For more details, you can see "our data policy". More...