Clientelism,partisanship,and ideology

Abstract

This paper aims to highlight the role of clientelism in the Justice and Development Party's (Adalet ve Kalkınma Partisi) electoral dominance in Turkish politics. Based on intensive fieldwork in Istanbul's one of the poorest and conservative districts of Bağcılar, it argues that the expansion of clientelist networks under the Adalet ve Kalkınma Partisi government has twofold ramifications, which in turn have reproduced the cycle of dominance. First, it strengthened clients’ partisan identification. Second, it changed clients’ ideology such that they became less resistant to or even supportive of neoliberal reforms that extensively undermined their well-being.

Keywords

Adalet ve Kalkınma Partisi clientelism Islam neoliberalism party identification

Introduction

The Justice and Development Party (Adalet ve Kalkınma Partisi or AKP) in Turkey is now enjoying the longest period of one-party government in modern Turkey. The literature well established that the electoral system (D’Hondt) with the highest threshold in the world (10%—since 2019, 7%) the charismatic-populist leadership of Recep Tayyip Erdoğan, and the manipulation of ethnic and religious cleavages largely contributed to that success (See, e.g., Arslantaş & Arslantaş, 2020b; Arslantaş & Kaiser, 2022; Arslantaş et al., 2020). These elements, however, do not account for the ups and downs in the AKP's vote share. To elaborate on that, the AKP's changing economic performance and its use of clientelism must be emphasized. This paper specifically elaborates on the latter within the context of a deepening economic crisis since 2018.

There is a growing interest in different aspects of clientelist exchange in Turkey thanks to methodological enrichment (See, for instance, Ark-Yıldırım, 2017; Bulut, 2020; Kemahlıoğlu, 2005; Kılıçdaroğlu, 2020; Marschall et al. 2016; Ocaklı, 2016). This paper adds to the existing literature in three ways. First, by combining an ethnographic approach with semi-structured interviews, this paper brings original data to the study of clientelism. Second, this research goes beyond the reductionist views that the AKP supporters simply vote for pasta. In particular, it introduces intervening variables, that is, partisanship and ideology, that bridge clientelism and electoral mobilization. Finally, from a more comparative perspective, this study shows how clientelism helps to reproduce the electoral dominance of long-ruling parties such as the AKP.

Based on fieldwork in one of the most densely populated and poorest districts of Istanbul, Bağcılar, between November 2017 and April 2018, we demonstrate that clientelism strengthens partisan identification. Second, clientelism makes clients, especially the poor, less resistant or even supportive of the neoliberal reforms that economically works against their interest. Also, we posit that Islamic ideology, which constitutes a non-material dimension of the AKP's clientelism, has turned into a political instrument for rent extraction and networking for utility maximizers. Overall, we argue that the deepening economic crisis in Turkey since 2018 as the result of “state capture” during the neoliberal era (Trantidis & Tsagkroni, 2017) destabilizes the AKP rule by putting its maintenance in the short run in jeopardy.

This paper is organized as follows. The “Clientelism, partisanship, and ideology” section delves into the concept of clientelism in relation to partisanship and ideology. The “The changing nature of clientelism in Turkey” section briefly traces party clientelism in Turkey. The “Ideological change” section how clientelism interacts with partisanship and neoliberal and Islamic ideology. The “Conclusion” section concludes and wraps up the main findings of this research.

Clientelism, partisanship, and ideology

Clientelism, can be defined as “the proffering of material goods in return for electoral support, where the criterion of distribution that the patron uses is simply: did you (will you) support me?.” (Stokes, 2007, p. 605), is more resilient than ever. Currently, it is well established that clientelism operates effectively in both traditional and modern societies, and democratic and non-democratic regimes, as well as at local, regional, national, and supranational levels (Hopkin, 2006a; Kitschelt & Wilkinson, 2007; Mouzelis, 1985), although its form and shape may vary substantially (Landé, 1983). This stands contrary to the claims that clientelism would disappear with economic and political modernization (Hicken, 2011; Kopecký & Mair, 2006; Piattoni, 2001; Roniger, 2004).

The importance of clientelism as a collective mobilization strategy does not only stem from its robustness. Clientelism can also be conceptualized as a strong political instrument that affects the level of partisanship and ideology which are the main pillars of most political party strategy. To start with the former, empirical evidence reveals that the clientelist parties are keen to target voters who have strong partisan identification and can be easily mobilized in the elections (See, e.g., Gans-Morseet al. 2009; Kitschelt & Wilkinson, 2007; Stokes et al., 2013). Drawing on fieldwork in Buenos Aires and Cordoba, for instance, Szwarcberg (2015) finds a strong relationship between having Peronist party identification and the targeting of voters with clientelism.

Such an interaction is not surprising given that the incumbent parties tend to reward their committed supporters when they assume office with their aid. Therefore, for the parties that build a long-term strategy to remain in power—be it predominant or hegemonic—clientelism is an effective instrument to mobilize voters at the polls (Altamirano, 2019; Greene, 2007; Magaloni, 2006). As Trantidis (2016) insightfully formulated in his “clientelist game” model, clientelism may offer a strong mechanism to attract supporters, discipline party members, and secure party cohesion. Besides, if the relevant party system is overtly clientelist, clientelism becomes the Nash equilibrium. In this case, the clientelist party gets aware of the fact that it will be worse off electorally if it quits this strategy halfway. This indicates that “path dependency” and “increasing returns” narrows down the possibility of adopting a non-clientelist political program along the way (Arthur, 1994).

The ideological change is both prior and posterior to the clientelist exchange. It is prior because the decline of working-class ideology set the basis for the clientelist parties to expand (Kitschelt & Wilkinson, 2007; Roniger, 2004). It is posterior, on the other hand, because clientelism as a materially induced strategy may well affect the hearts and the minds of the targeted people. In this paper, ideological change is investigated with regard to the neoliberal and Islamic transformation of the actors which are involved in the clientelist exchange. While there is abundant literature on neoliberal transformation under the AKP governments, the interaction between clientelism and the Islamic ideology—both at the level of discourse and practice—remained virtually un(der)explored. Broadly speaking, this is the case because the Islamist movements and parties are either not allowed to exist (e.g., Egypt) or are not popular enough to come to power (e.g., Tunisia) in Islamic geography. In this light, studying Turkey as a case study has the potential to improve our understanding of that interaction.

The changing nature of clientelism in Turkey

Clientelism in the early years of the Republic—to a large extent—was inherited from the clientelist practices during the late Ottoman period. To extend its influence in the periphery, the central administration built clientelist relations with local notables. In return for this, the ruling elites overlooked the expanding influence of these mediators, be it, small-town merchants (eşraf), or large landowners (ağa) (Ayata, 1994). The symbiotic structure of governance, which is maintained in the multi-party period as well, contrasts markedly with class-based horizontal mobilization in Western Europe (Sayarı, 2011). From the 1960s onwards, clientelism expanded rapidly in metropolitan areas. In this context, clientelism acquired a problem-solving function. Sherwood (1967, p. 57) illustrates this transformation with the following example:

A typical villager arriving in Ankara or Istanbul goes immediately to that district populated by people from his home village. The local Justice Party [AP] man helps him settle, aids him with his problems with the authorities, and functions as an employment agency, or a marriage bureau, as the case may be.

Clientelist networks further expanded in the post-1980 period. Unlike the major parties of the 1970s, the Motherland Party (Anavatan Partisi) achieved to nurture some cronies who were very well connected to Prime Minister Turgut Özal and his family (Boratav, 2016; Öniş, 2004). Similarly, Özal granted title deeds to squatter settlements; institutionalized preferential electrification; and frequently turned sub-provinces into provinces if they vote otherwise (Ayata, 1994; White, 2002). Cronies continued to expand their influence during subsequent coalition governments. As corruption scandals were widely manifested in center media (e.g., the Civangate scandal), the Welfare Party's (Refah Partisi or RP) Islamic justice discourse found a large audience, especially among the poor. When the Islamic discourse was combined with the door-to-door electoral strategy of the highly committed party workers, the RP rapidly extended its voter base. With this momentum, it first captured Istanbul and Ankara in the 1994 local elections and then became the largest party in the 1995 general elections. Then, it was able to form an 11-month-old government with the True Path Party (Doğru Yol Partisi or DYP).

Although the AKP was established by the splinters from the RP, their practice of clientelism differed sharply. First, these parties contrasted in terms of their relationship with the business. Although the RP could not cultivate strong loyalty in the business sector due to the pressures of the military and secular social sectors (Gülalp, 2001; Öniş, 2006), the AKP has been able to nurture such loyalty with partisan bourgeoise (Demiralp, 2009; Öniş, 2019; Yeldan & Ünüvar, 2016). Second, the AKP commanded much more resources than the RP, emanating from its control of the central government as well as the municipalities. Third, the AKP's clientelism has been more efficient in deterring exit from the clientelist network because of the complex division of labor among relevant actors. To maximize the efficiency of the clientelist network, the AKP set up a complex monitoring mechanism, as can be seen in other clientelist party examples (Kitschelt & Wilkinson, 2007). In the densely populated district of Bağcılar, for instance, the monitoring mechanism included the AKP's neighborhood organizations as well as religious and charity organizations. The main result of this strategy was “perverse accountability” (Stokes, 2005) which distorted good governance in the country.

Table 1 derived from the Democratic Accountability and Linkage Project¹ data shows the extent of the AKP's clientelism. Statistically speaking, the AKP scored 4.0 in gift and payments and regulatory favors; 3.8 in social policy benefits; 3.9 in patronage jobs and procurement contracts; and 3.7 in the effectiveness of clientelist targeting, on a scale that goes from 1= a negligible [clientelist] effort or none at all to 4= a major effect. Although the opposition parties also offered some degree of clientelist benefits to their supporters, they are by far overwhelmed by the AKP. Similarly, the cross-country comparisons presented in Table 2 reveal that the AKP's use of clientelism by far overwhelms other long-ruling parties such as the Institutional Revolutionary Party (PRI, Mexico), Kuomintang (KMT, Taiwan), Indian National Congress (INC, India) and Liberal Democratic Party (LDP, Japan).

Table 1.

Clientelism and mainstream parties in Turkey*.

Party	b1 (gift and payment)	b2 (social policy benefits)	b3 (patronage jobs)	b4 (procurement contracts)	b5 (regulatory favors)	b11 (effectiveness of clientelist targeting)
HDP	2.0	3.3	3.7	3.4	3.4	2.9
MHP	1.5	2.9	3.7	3.2	3.3	2.6
AKP	4.0	3.8	3.9	3.9	4.0	3.7
CHP	2.1	3.0	3.5	3.2	2.9	2.4

Table 2.

Clientelism in a comparative perspective.

Country	Party	b1 (gift and payment)	b2 (social policy benefits)	b3 (patronage jobs)	b4 (procurement contracts)	b5 (regulatory favors)	b11 (effectiveness of clientelist targeting)
Mexico	Institutional Revolutionary Party (PRI)	3.7	3.5	3.4	3.5	3.4	2.2
Mexico	National Action Party (PAN)	2.7	2.7	2.7	3.1	3.3	1.7
Mexico	PRD	3.3	3.2	3.1	3.1	3.0	2.0
Taiwan	KMT	3.1	3.5	2.6	2.6	3.6	3.5
Taiwan	DPP	2.7	3.2	2.5	2.5	3.4	3.1
India	INC	3.3	3.4	3.1	3.1	3.2	3.2
India	BJP	3.1	3.3	2.9	3.0	3.1	3.1
India	CPI	2.2	3.0	2.8	2.1	2.2	3.0
Japan	LDP	2.3	3.3	2.4	3.5	3.1	3.2
Japan	DPJ	2.1	2.3	2.1	2.1	2.1	2.6
Japan	JCP	1.6	1.6	1.7	1.1	1.2	2.5
Japan	JSP	1.4	1.5	1.6	1.0	1.2	2.1

Note: *Scores: 1 to 4–1= a negligible [clientelist] effort or none at all; 2= a minor effort; 3= a moderate effort; 4= a major effort. PRD: Party of Democratic Revolution; KMT: Kuomintang; DPP: Democratic Progressive Party; INC: Indian National Congress; BJP: Bharatiya Janata Party; CPI: Communist Party of India; LDP: Liberal Democratic Party Japan; DPJ: Democratic Party of Japan; JCP: Japanese Communist Party; JSP: Japanese Social Democratic Party;DALP: Democratic Accountability and Linkage Project.

Source: DALP, https://sites.duke.edu/democracylinkage/

Case selection and method

This paper is based on fieldwork in Istanbul's district of Bağcılar. The fieldwork lasted from November 2017 to April 2018. Although Bağcılar has a surface area of only 22 km², it hosts 748.483 inhabitants (2017) which makes Bağcılar the third most populated district in Istanbul.² Within the scope of fieldwork, the semi-structured interviews, guided by the Interview Guide presented in Appendix 1, were conducted with partisans, brokers, municipal employees, and a mukhtar of one of the neighborhoods. The selection of Bağcılar for the field study is justified on the following grounds. First, in the June and November 2015 elections, the AKP won the highest number of votes there of any district in Istanbul (See, Table 3). Second, Bağcılar is one of the poorest districts of Istanbul. Research shows that the average monthly income in the Beşiktaş district (10.560 Turkish Lira (TL)) is more than triple-doubles that of Bağcılar (3.197 TL) (Posta, 2017).³ Third, the representativeness of Bağcılar is high given that its demographic structure largely overlaps with that of other poor districts such as Esenler, Sultangazi, and Sultanbeyli.

Table 3.

Vote shares of the Adalet ve Kalkınma Partisi (AKP) in the legislative elections (%).

	Legislative					Local
	2002	2007	2011	2015/1	2015/2	2004	2009	2014
Nationwide	34.2	46.5	49.8	40.8	49.5	41.6	38.3	42.8
Bağcılar	48.9	56.9	60.4	50.0	59.5	43.9	49.2	57.2

To reveal the extent of clientelism, different methods have been adopted. Since clientelism involves informal exchange based on unwritten rules, a quantitative approach would potentially reveal only a small portion of reality (Muno, 2010). Similarly, expert surveys are costly to conduct (Wanthchekon, 2003). Taken these, this study combines interview data and ethnographic fieldwork. Unlike the mentioned methods, using of ethnographic approach has a number of merits: it provides deep insights into the daily practice of clientelism (Muno, 2010; Auyero, 2001); enhances rapport; and makes it possible to observe non-verbal behavior, the context, and relationships (Scupin, 2015).

The strengthening of the partisan identification

Partisan identification is essential to political behavior. While one stream of research takes it as a psychological trait or predisposition, another takes it as active behavior, evidenced in party membership or going to the polls (Converse & Pierce, 1985). Above all, party identification is a matter of degree. While strong identifiers “always vote for the same candidate or party regardless of what it says or does,” weak identifiers are prone to defection (Sartori, 2005, p. 293). This does not mean that weak identifiers are always unstable voters. In fact, weak identifiers can be very stable if their vote is motivated by “negative voting,” that is, parties they dislike (Sartori, 2005).

Historically speaking, partisan identification remained weak in the context of Turkey due to the changes in party names (Sayarı, 2008), the military interventions (Çarkoğlu, 1998), and the disappointment with the performance of the mainstream parties (Özbudun, 2001). These elements hindered party institutionalization and the building of a long-term base for the parties. However, since 2002, partisan identification has strengthened remarkably—to the extent that three of the four identified themselves with a party in 2018—as the party system has become more stable and the possibility of military intervention has become less likely (Kalaycıoğlu, 2008). In particular, an analysis of the pre-election field survey of the Turkish Elections Study demonstrates that party identification increased from 20.6% in 2010 to 40.3% in 2018.⁴

Normally, strong partisan identification is taken as a stabilizing force for the political system, whereas weak identification is taken to be alarming for the quality of democratic representation (Campbell et al., 1960; Dalton & Norris, 1996; Rose & Mishler, 1998). In contrast to that strong partisan identification in Turkey fostered polarization and thereby destabilized the political system. This is because partisan identification has been strengthened due to “personalistic clientelism” (Kitschelt & Wilkinson, 2007). What we observed in the field was that the more clients benefit from clientelist resources, the stronger their identification with the party. In this framework, Erdoğan as the party leader carries out two main functions within clientelist linkage (Arslantaş & Arslantaş, 2020a). First, as a charismatic leader, Erdoğan emerged as a “guarantor” of the clientelist machine who is tasked with the unification of normally antagonistic classes under the same party flag. This especially matters in the context of an economic downturn. Second, he reduces the burden on the party organization by energizing its supporters through direct communication during non-election periods.

Strong partisan identification first and foremost requires the existence of encompassing party organization. During the interviews, the party officials emphasized that the AKP has more than 120,000 members in Bağcılar. Although at first glance the number of members sounds quite high, amounting to half the AKP votes cast in Bağcılar in the 2015 November elections, it is not surprising given that the typical party membership in Turkey remains only on paper and does not impose any responsibilities on members, such as paying membership fees or regularly taking part in party activities. Reflecting on this, the first thing to realize was the party officials’ self-confidence. The words of a party official [55] illustrate this:

There is no party if there is no strong organization. We are very strong in Bağcılar. The municipality cannot do its job if we do not work. We are the backbone of the party.⁵

Strong partisan identification has severe implications for party competition in Turkey. First, clients started to read the political, social, and economic developments through the prism of the party, or rather the party leader. This endorses the view that parties not only interpret ongoing cleavages (cf. Lipset & Rokkan, 1967) but also have the capacity to change voters’ perceptions and identities (cf. Campbell et al., 1960), especially in the context of high polarization (Druckman et al., 2013). This argument is more valid when potential intervening variables such as economic crises are not at stake, especially before the summer of 2018.

Second and what is more alarming for the political system in general is that clientelism makes “hostile partisanship” more evident. Hostile partisanship is one among a set of destructive effects of establishing a clientelist linkage with the voters (others include colluding into authoritarianism (Trantidis, 2013)). Given that clientelism entails unequal distribution of scarce resources, such a destructive effect is not surprising. The widespread view among the AKP supporters that “if the AKP goes so does the clientelist benefits” best illustrates this fear. This view is endorsed by the party leadership and reflects the contingent character of the clientelist linkage. The statement of one interviewee best illustrates this case:

I can swear that if the CHP comes to power, aid will be cut off. I remember that the municipalities were providing scholarships to university students. Then, they stopped after the CHP complained to the Constitutional Court.⁶ The CHP takes from the public rather than serving it. [13]

As Almond & Verba (1963) argue a long time ago, hostile partisanship has the potential to ruin or delegitimize the political system by exacerbating political and social polarization. Research on the dimensions of polarization in Turkey (Erdoğan & Semerci, 2018), for instance, found that 69.6% of respondents do not want to be neighbors with voters for the party to which they are most hostile. Similarly, 78.7% of respondents declared that they would not allow their daughter to marry a supporter of the party to which they are most hostile.

Due to the expansion of the clientelist network, finally, partisan identity has begun to overwhelm other competing identities. This is by no means to say that identification based on religion, ethnicity, or regionalism has lost its salience. On the contrary, these identities are as strong as ever. Rather, a change has occurred in a hierarchical pyramid that defines these identities. The expansion of clientelist networks subordinated competing group identities to the clientelist party.⁷ One partisan [45], for instance, stated that:

I have been a member of a religious community for many years. The difference between the party and the religious community is that the religious communities only think about their interests and they nurture regimented people. They are closed to criticisms. The members take whatever the leader says as a command. In most cases, the imams regard party leaders as rivals and oppose them. After July 15 [2016], the public thoroughly understood this point. Now, no one wants to send their children to religious schools. And this is good for our society.

This sentiment makes more sense for the voters who do not come from a religious background. The deteriorating image of the religious communities in the eyes of partisans has also been captured by public surveys. According to the “Religion and Religious Values” survey conducted by MAK in 2017, half of the respondents indicated that religious communities should be subjected to state control. More interestingly, only one-sixth of the participants declared an affiliation with religious communities or orders (Sputnik, 2017). This indicates the weakening function of Islamic ideology as a non-material component of clientelist linkage.

Ideological change

Clientelism and religion

Ideology is an effective way to build a common identity with the voters (Dixit & Londregan, 1996). Therefore, even clientelist parties seek to incorporate some degree of ideological element in maximizing their votes. In this sense, the AKP as a clientelist party incorporated Islamic ideology as part of its clientelist appeal. It did so in two ways. First, clientelism transformed some radical-Islamist voters to realign with the neoliberal AKP. More specifically, clientelism helped small-scale business owners to upgrade into big capitalists by winning public procurements or undertaking urban transformation projects (e.g., Eliçin, 2014; Emek & Acar, 2015; Soyaltın, 2017). In conforming to that almost all clients observed that most of the urban transformation projects in some neighborhoods of Bağcılar were undertaken by pro-AKP constructors.

The realignment of the radical Islamist voters with the AKP is mostly motivated by “organizational patronage” (Kopecky et al. 2012) as a subclass of clientelism (Stokes, 2007). The story of twin brothers who had radical Islamic backgrounds illustrates this. The twins studied together in the underground Quranic courses. They had to choose this path because their father was against enrolling them in “secular” public schools. After finishing high school externally in 2013, they were appointed Imams by the Directorate of Religious. The decision to enroll children in the school came after the father saw that the twins might take a position in the state, which was not seen as “feasible” before. This shows that “state capture”⁸ (Trantidis & Tasgkroni, 2017) by the AKP seems to have a transformative effect on some radical Islamic voters who have traditionally boycotted elections under the secular regime. The integration of radical Islamist voters into the clientelist machine substantially expanded the vote base of the AKP.

Second, the AKP's clientelism, similarly, led some clients who come from center-right backgrounds to become moderate Islamists. This can be observed both at the level of discourse and practice. In this regard the center-right voters are the latest addition to the AKP's voter base whose main motivation is to get as much as they can from expanding the clientelist mechanism, especially in big cities. One partisan [21] portrayed this in this way:

Religious words are now in everyone's mouth. Someone addresses you with “Selamun Aleykum.” Then, he talks about how to steal people's money. In recent times, people have begun to fast and go to the mosque to get something out of it, procurement or land.

This testimony shows that religion has turned into a way of rent extraction and networking within the AKP's clientelist machine. In this sense, the religious transformation has practical rather than symbolic implications. The fact that moderation and radicalization occur simultaneously inevitably leads to surprising (if not conflicting) outcomes, which are widely reacted to at a grassroots level. Illustrating this, one partisan [22] argued that:

What has happened in the past decade is that the left has cast off nationalism and the state, whereas the right has cast off religion. Practicing religion has been rendered merely formal. It now consists of visiting religious places with girlfriends on Fridays and tweeting verses on Fridays. Although the number of women with a headscarf has increased, religious people have become more secular. Now, women wear a headscarf but also put on makeup. This is how Islam is being distorted.

The transformation of Islamic ideology into an instrument for rent extraction and networking first and foremost served the religious communities whose activism reached a new height under the AKP governments. It is widely documented that the AKP government has co-opted a substantial portion of religious communities by allowing them to capitalize on private and public resources. In this process, some of the religious groups such as Gülenists acquired such power that they constituted a “parallel state” within a state. In return for that the religious organizations expanded the AKP's support base by appealing to new clients; establishing close contact with the clients; normalizing hierarchies, domination, and exploitation (Arslantaş & Arslantaş, 2020a); and generating the feeling of “reciprocity” among clients. Despite that during the fieldwork, the overwhelming majority of the interviewees admitted how their views have changed negatively toward the religious groups, especially in the aftermath of the abortive coup attempt in July 2016. This shows that an attempt at “state capturing” by religious groups largely backfired even among the AKP supporters.

Clientelism and neoliberalism

The second dimension of ideological change is the drift toward neoliberalism.⁹ This happened at the expense of dramatic setbacks in the economic conditions of clients residing in urban poor space.¹⁰ As widely documented, the AKP allied with the urban poor and partisan businessmen in implementing neoliberal reforms including the privatization of health and education services and the expansion of flexible work and subcontracting (for detailed analysis of these changes, See, Akçay, 2021; Bozkurt-Güngen, 2018; Dorlach, 2015; Karadağ, 2010). While the poor provided votes at the ballot box, the businessmen provided informal money in the form of bribery to finance the clientelist network. Although the dynamics of the alliance between the AKP and the partisan businessmen are extensively studied and remain out of the scope of this paper, the reaction of the AKP base to this alliance remained largely untouched within the framework of a theory on clientelism.

In this light, what we observed in the field is that the poor partisans were well aware of the fact that the AKP promoted its own cronies through informal channels. Surprisingly, this did not cultivate discontent at the party base since such enrichment of the pro-AKP businessmen—mostly by infringing the law—seems to serve as a “role model” for them. This is the case despite the enrichment of loyal businessmen occurring at the expense of dragging more and more poor partisans under the poverty line. This attitude of the clients best illustrates how a clientelist party can change the hearts and the minds of its supporters. One partisan [34], for instance, noted that:

Before the AKP government, conservative business was completely excluded from the state. Kemalists were taking all procurement contracts. This does not happen anymore. Now, Kemalists accuse the AKP government of having corrupt relations with businessmen, as if they got rich another way.

The most critical result of this apathy was the normalization of inequalities which was emanating from clientelism itself (Hopkin, 2006b; Kaufmann, 1974). This tendency stands in stark contrast to the RP's redistributive justice program which promised to safeguard the urban poor against exploitation (Gülalp, 2001; Öniş, 1997; Tuğal, 2009). By normalizing economic inequalities, the AKP largely restricted the notion of justice to political issues such as lifting the ban on headscarves. The views of the interviewees illustrate this:

Under the RP, we were thinking more radically. We were against riches and waste. I don't think it's the same now. Turkey has changed a lot. Now, there is wealth in the country. Traders and builders quickly become rich. [20]

The RP could not get the votes of the rich. Mostly, poor people residing in the squatter settlements voted for it. Of course, there were people voting for religious reasons, but mostly people were voting because of the economy. Unlike the RP, the AKP gets votes from every group. [25]

Many people have got rich under the AKP government. In this period, becoming rich is easier than in other periods. God is bestowing at least something on working people. [24]

For the urban poor, being part of the clientelist network also required strong support for neoliberal reforms including subcontracting and flexible labor. This outcome is compatible with the nature of clientelism which obstructs class-based horizontal solidarity. Accordingly, the lower classes have started to lose their ability to empathize with each other because they are part of a clientelist network. This is a manifestation of what we call “de-identification,” which aims to destroy any notable resistance to neoliberal reforms that undermine the well-being of the poor. Reflecting this, for instance, one partisan [11] noted that:

I am against unionization. Unionized workers earn money by sleeping, so to speak. The solution is subcontracting. Subcontractors don't cost too much, and they work harder because there is no job guarantee. Bağcılar municipality implements this policy successfully.

In the same vein, the testimonies below indicate that the clients as well as the patrons blamed personal failures in accounting for poverty. This means that the reproduction of inequality and poverty which is embedded in neoliberal transformation (Feldman, 2019) is left out of the question. This spots a radical break with the Welfare Party tradition, which attributed the source of poverty to the capitalist order (see, e.g., “just order” discourse). When evaluated in this context, the statements of the party officials are imperative in terms of reflecting the general opinion of the clients:

I believe poor people are lazy. Today, anyone willing to work can get a job easily. But because many institutions provide assistance, no one wants to work [59].

It is impossible for a person to stay poor in Bağcılar. You can get aid from numerous places. Our people are not poor, they are “luxury poor.” They are not poor in the sense of eating and drinking, but in relation to social activities.

The term “luxury poor” means more than it might appear. This designation takes the notion of poverty in the narrower sense of access to food and drink and considers anything beyond that (e.g., access to social life) as a luxury. This rationale normalizes poverty and is instrumental in hampering further economic demands by the clients.¹¹ In a nutshell, these testimonies show that the lower classes have started to lose their ability to empathize with each other because they are part of a clientelist network.

Conclusion

This study elaborates on the causal link between clientelism and the electoral dominance of the AKP. While doing that it questions the widely recognized “clientelism automatically brings votes” argument. More specifically, since the voters may easily vote for another party because voting is secret (Szwarcberg, 2015), the mechanism that connects clientelism to vote must be highlighted.

Based on fieldwork in one of Istanbul's poorest and most densely populated districts, Bağcılar, it is argued that the AKP's use of clientelism has two ramifications, which in turn reproduce the cycle of dominance. To start with, clientelism leads to strong partisan identification. This contrasts with the traditionally weak partisan identification in Turkey. Strong partisan identification has two dimensions in the Turkish context: (a) it entails reading political developments through the eyes of the party or rather the party leader and (b) it corresponds to hostile partisanship.

Second, clientelism has prompted ideological change, namely neo-liberalization in an Islamic setting. This accounts for three developments: (a) it explains how Islam and neoliberalism have become intertwined; (b) it expresses the support of the poor for neoliberal policies, although they undermine their well-being; and (c) it shines a light on the realignment of radical Islamists and center-right voters with the AKP.

What would be the future of the clientelist network in Turkey? We are far from arguing along with Esen & Gümüşçü (2018) that the recent shift from neoliberalism to crony capitalism may stabilize the AKP rule. On the contrary, as clientelist resources diminish due to the shrinking economy, partisan businesspeople will be less reluctant to share the additional costs which would accelerate splits from the AKP. The waning clientelist machine also has broader implications. Since the AKP lacks resources to intensively target the swing voters as it did before, the swing voters, especially in big cities, may change sides which matters most in a Presidential race where even one vote matters.

This study opens up two areas of research for further investigation. The first one is to reveal to what extent and in which respects the use of clientelism by the oppositional parties affects voting patterns among clients. The second is the gathering of data, quantitative or qualitative, on the impact of clientelism on partisan identification and ideological change in other authoritarian party settings.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Universität zu Köln, (grant number NA).

ORCID iD

Düzgün Arslantaş

Notes

Appendix 1: Interview guide

Introduction	■ How long have you been living in Bağcılar?
AKP	■ How many times have you voted for the Adalet ve Kalkınma Partisi (AKP)? ■ Have you ever voted for another party? ■ Which party and why?
Form of integration in clientelistic networks	■ Are you aware of the material benefits distributed by the government and the pro-AKP actors? ■ Have you received any benefits so far? ■ If yes, what, when and how much? ■ Is there any mediator that secures your access to goods? ■ When deciding on which party to vote for, how important is the distribution of material benefits for you?
Partisan Identification	■ What is your type of relationship to the AKP? (voter, member or partisan) ■ How has it changed in recent years? ■ How regularly do you take part in party activities? ■ How close do you feel to the AKP? (weak, moderate, strong) ■ Are you a sympathizer/member of any other religious or social group? ■ If the AKP's interest clashes with that of your mentioned group (religious group, regional-communal organization or civil society groups) which side would you take? ■ Why?
Ideology	■ How do you define yourself politically? (e.g., left, right, conservative, nationalist) ■ Have you observed any change in your political position during the AKP period? ■ If yes, on which issues?

Appendix 2: The list of interviews

Number	Gender	Date	Islamic tradition?	C or S?
Voters
1	M	17.11.2017	Y	C
2	M	19.11.2017	Y	C
3	M	21.11.2017	N	C
4	M	23.11.2017	Y	C
5	F	28.11.2017	N	S
6	F	28.11.2017	N	S
7	F	1.12.2017	Y	C
8	F	1.12.2017	N	S
9	F	5.12.2017	Y	C
10	F	5.12.2017	Y	C
11	M	9.12.2017	N	C
12	M	9.12.2017	Y	S
13	M	9.12.2017	Y	S
14	M	14.12.2017	N	C
15	M	14.12.2017	Y	C
16	F	19.12.2017	Y	C
17	F	19.12.2017	Y	S
18	F	19.12.2017	Y	C
19	M	3.01.2018	Y	S
20	M	3.01.2018	Y	C
21	M	3.01.2018	N	S
22	M	9.01.2018	Y	S
23	M	9.01.2018	N	C
24	M	11.01.2018	N	S
25	M	11.01.2018	Y	C
26	M	11.01.2018	Y	C
27	M	16.01.2018	N	C
28	F	16.01.2018	N	S
29	F	20.01.2018	Y	C
30	F	20.01.2018	Y	S
31	F	20.01.2018	Y	C
32	M	24.01.2018	N	S
33	M	24.01.2018	N	S
34	M	27.01.2018	Y	C
35	M	27.01.2018	Y	C
36	F	14.03.2018	Y	S
37	F	14.03.2018	Y	S
38	F	14.03.2018	Y	C
39	M	20.03.2018	Y	C
40	F	20.03.2018	N	S
41	F	20.03.2018	N	C
42	F	20.03.2018	N	C
43	M	3.04.2018	N	S
44	M	3.04.2018	N	S
45	M	3.04.2018	Y	C
Municipality
46	F	17.11.2017	-	-
47	F	17.11.2017	-	-
48	M	17.11.2017	-	-
49	M	17.11.2017	-	-
50	M	17.11.2017	-	-
51	M	17.11.2017	-	-
52	M	17.11.2017	-	-
53	M	17.11.2017	-	-
54	M	17.11.2017	-	-
Party
55	M	6.03.2018	-	-
56	M	6.03.2018	-	-
57	M	10.03.2018	-	-
58	M	10.03.2018	-	-
59	M	10.03.2018	-	-
Mukhtar
60	M	20.11.2017	-	-

M: male; F: female; Y: yes; N: no; C: core; S: swing.

References

Akçay

(2021) Authoritarian consolidation dynamics in Turkey. Contemporary Politics 27(1): 79–104.

Almond

Verba

(1963) The Civic Culture: Political Attitudes and Democracy in Five Nations, An Analytic Study. Princeton: Princeton University Press.

Altamirano

(2019) Economic vulnerability and partisanship in Latin America. Latin American Politics and Society 61(3): 80–103.

Ark-Yıldırım

(2017) Political parties and grassroots clientelist strategies in urban Turkey: One neighbourhood at a time. South European Society and Politics 22(4): 473–490.

Arslantaş

(2020a) How does clientelism foster electoral dominance? Evidence from Turkey. Asian Journal of Comparative Politics 7(3): 559–575.

Arslantaş

and Arslantaş Ş (2020b) Keeping power through opposition: Party system change in Turkey. New Perspectives on Turkey 62: 27–50.

Arslantaş

, Arslantaş Ş and Kaiser A (2020) Does the electoral system foster a predominant party system? Evidence from Turkey. Swiss Political Science Review 26(1): 125–143.

Arslantaş

Kaiser

(2022) The ‘competitive authoritarian’turn in Turkey: Bandwagoning versus reality. Third World Quarterly 1–17. Online first.

Arthur

(1994) Increasing Returns and Path Dependence in the Economy. Michigan: University of Michigan Press.

10.

Auyero

(2001) Poor People’s Politics: Peronist Survival Networks and the Legacy of Evita. Durham: Duke University Press.

11.

Ayata

(1994) Roots and trends of clientelism in Turkey. In: Roniger

Ayata

(eds) Democracy, Clientelism and Civil Society. London: Lynne Rienner, pp. 51–63.

12.

Boratav

(2016) The turkish bourgeoisie under neoliberalism. Research and Policy on Turkey 1(1): 1–10.

13.

Bozkurt-Güngen

(2018) Labour and authoritarian neoliberalism: Changes and continuities under the AKP governments in Turkey. South European Society and Politics 23(2): 219–238.

14.

Bulut

(2020) How clientelistic parties go programmatic: The strategic logic of responsiveness in a least likely case (the AKP of Turkey). Comparative Politics 52(2): 333–356.

15.

Campbell

Converse

Miller

, et al. (1960) The American Voter. New York: J. Wiley & Sons.

16.

CHP (2018) 15 Yılın Faturası AKP'nin Yıkım Tablosu. Retrieved from http://dijitalmecmua.chp.org.tr/PageMecmua.aspx?Mecmua=66#p=1.

17.

Converse

Pierce

(1985) Partisanship and the party system. Political Methodology 11(3/4): 143–166.

18.

Çarkoğlu

(1998) The Turkish party system in transition: Party performance and agenda change. Political Studies 46(3): 544–571.

19.

Dalton

Norris

(1996) Political cleavages, issues, and electoral change. In: Leduc

Niemi

(eds) Comparing Democracies: Elections and Voting in Global Perspective. Thousand Oaks, CA: Sage Publications, pp. 319–342.

20.

Demiralp

(2009) The rise of Islamic capital and the decline of islamic radicalism in Turkey. Comparative Politics 41(3): 315–335.

21.

Dixit

Londregan

(1996) The determinants of success of special interests in redistributive politics. The Journal of Politics 58(4): 1132–1155.

22.

Dorlach

(2015) The prospects of egalitarian capitalism in the global south: Turkish social neoliberalism in comparative perspective. Economy and Society 44(4): 519–544.

23.

Druckman

Peterson

Slothuus

(2013) How elite partisan polarization affects public opinion formation. American Political Science Review 107(1): 57–79.

24.

Eliçin

(2014) Neoliberal transformation of the Turkish city through the urban transformation act. Habitat International 41: 150–155.

25.

Emek

Acar

(2015) Public procurement in infrastructure: The case of Turkey. The Anticorruption Report 3: 84–96.

26.

Erdoğan

Semerci

(2018) Fanus’ta Diyaloglar: Türkiye’de Kutuplaşmanın Boyutları. Istanbul: Bilgi Üniversitesi Yayınları.

27.

Esen

Gümüşçü

(2018) Building a competitive authoritarian regime: State–business relations in the AKP's Turkey. Journal of Balkan and Near Eastern Studies 20(4): 349–372.

28.

Feldman

(2019) Neoliberalism and poverty: An unbreakable relationship. In: Greve

(ed) Routledge International Handbook of Poverty. London: Routledge, pp. 340–350.

29.

Gans-Morse

Mazzuca

Nichter

(2009) Who Gets Bought? Vote Buying, Turnout Buying, and Other Strategies. Harvard University Weatherhead Center for International Affairs Working Paper, 09-0006.

30.

Greene

(2007) Why Dominant Parties Lose: Mexico’s Democratization in Comparative Perspective. Cambridge: Cambridge University Press.

31.

Gülalp

(2001) Globalization and political Islam: The social bases of Turkey’s welfare party. International Journal of Middle East Studies 33(3): 433–448.

32.

Harvey

(2005) A Brief History of Neoliberalism. Oxford: Oxford University Press.

33.

Hicken

(2011) Clientelism. Annual Review of Political Science 14(1): 289–310.

34.

Hopkin

(2006a) Conceptualizing Political Clientelism: Political Exchange and Democratic Theory. APSA Meeting, pp. 1–19.

35.

Hopkin

(2006b) Clientelism and party politics. In: Katz

Crotty

(eds) Handbook of Party Politics. London: Sage Publications, pp. 406–412.

36.

Hürriyet (2008, November 21) Belediye Bursları İptal. Retrieved from https://www.hurriyet.com.tr/gundem/belediye-burslari-iptal-10410064.

37.

Kalaycıoğlu

(2008) Attitudinal orientation to party organizations in Turkey in the 2000s. Turkish Studies 9(2): 297–316.

38.

Kalaycıoğlu

(2021) Party identification and vote choice in Turkey. In: Çarkoğlu

Kalaycıoğlu

(eds) Elections and Public Opinion in Turkey. London: Routledge, pp. 169–192.

39.

Karadağ

(2010) Neoliberal Restructuring in Turkey: From State to Oligarchic Capitalism (No. 10/7). MPIfG Discussion Paper.

40.

Kaufman

(1974) The patron-client concept and macro-politics: Prospects and problems. Comparative Studies in Society and History 16(3): 284–308.

41.

Kemahlıoğlu

(2005) Intra-party Competition and Clientelism: An Analysis of Public Sector Employment in Argentina and Turkey. APSA Annual Meeting, pp. 1–54.

42.

Kitschelt

Wilkinson

(2007) Citizen–politician linkages: An Introduction. In: Kitschelt

Wilkinson

(eds) Patrons, Clients, and Policies Patterns of Democratic Accountability and Political Competition. Cambridge: Cambridge University Press, pp. 1–49.

43.

Kılıçdaroğlu

(2020) Change in political party strategy and organization in Turkey: The Republican People's Party in government and in opposition. Southeast European and Black Sea Studies 20(4): 593–615.

44.

Kopecký

Mair

(2006) Political Parties and Patronage in Contemporary Democracies: An Introduction. ECPR Joint Sessions of Workshops, pp. 25–30.

45.

Kopecký

Mair

Spirova

. (eds.) (2012) Party Patronage and Party Government in European Democracies. Oxford: Oxford University Press.

46.

Landé

(1983) Political clientelism in political studies. International Political Science Review 4(4): 435–454.

47.

Lipset

Rokkan

(1967) Cleavage Structures, Party Systems, and Voter Alignments: An Introduction. New York: The Free Press.

48.

Magaloni

(2006) Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico. Cambridge: Cambridge University Press.

49.

Marschall

Aydoğan

Bulut

(2016) Does housing create votes? Explaining the electoral success of the AKP in Turkey. Electoral Studies 42: 201–212.

50.

Mouzelis

(1985) On the concept of populism: Populist and clientelist modes of incorporation in semiperipheral polities. Politics & Society 14(3): 329–348.

51.

Muno

(2010) Conceptualizing and Measuring Clientelism. Workshop on Neopatrimonialism in Various World Regions, pp. 1–26.

52.

Ocaklı

(2016) Political entrepreneurs, clientelism, and civil society: Supply-side politics in Turkey. Democratization 23(4): 723–746.

53.

Öniş

(1997) The political economy of islamic resurgence in Turkey: The rise of the welfare party in perspective. Third World Quarterly 18(4): 743–766.

54.

Öniş

(2004) Turgut Özal and his economic legacy: Turkish neo-liberalism in critical perspective. Middle Eastern Studies 40(4): 113–134.

55.

Öniş

(2006) The political economy of Islam and democracy in Turkey: From the welfare party to the AKP. In: Jungs

(ed) Democratization and Development. New York: Palgrave Macmillan, pp. 103–128.

56.

Öniş

(2019) Turkey Under the challenge of state capitalism: The political economy of the late AKP era. Southeast European and Black Sea Studies 19(2): 201–225.

57.

Özbudun

(2001) The institutional decline of parties in Turkey. In: Diamond

Gunther

(eds) Political Parties and Democracy. Baltimore and London: The John Hopkins University Press, pp. 238–265.

58.

Piattoni

(2001) Clientelism, Interests, and Democratic Representation: The European Experience in Historical and Comparative Perspective. Cambridge: Cambridge University Press.

59.

Posta (2017) İşte İstanbul’un En Zengin İlçeleri (İstanbul’da İlçelere Göre Gelir Dağılımı). https://www.posta.com.tr/iste-istanbul-un-en-zengin-ilceleri-istanbul-da-ilcelere-gore-gelir-dagilimi-haber-fotograf-1359815.

60.

Roniger

(2004) Political clientelism, democracy, and market economy. Comparative Politics 36(3): 353–375.

61.

Rose

Mishler

(1998) Negative and positive party identification in post-communist countries. Electoral Studies 17(2): 217–234.

62.

Sartori

(2005[1976]) Parties and Party Systems: A Framework for Analysis. Colhester: ECPR Press.

63.

Sayarı

(2008) Non-electoral sources of party system change in Turkey. In: Yazıcı

Gözler

Keyman

(eds) Prof. Dr. Ergun Özbudun'a Armağan. Ankara: Yetkin, pp. 399–417.

64.

Sayarı

(2011) Clientelism and Patronage in Turkish Politics and Society. Unpublished Manuscript.

65.

Scupin

(2015) Cultural Anthropology: A Global Perspective. Hoboken, NJ: Pearson Education.

66.

Sherwood

(1967) The rise of the Justice Party in Turkey. World Politics 20(1): 54–65.

67.

Soyaltın

(2017) Public sector reforms to fight corruption in Turkey: A case of failed europeanization? Turkish Studies 18(3): 439–458.

68.

Sözcü (2020, February 26) Istanbul’da En Çok Suriyeli Hangi İlçede? Işte Sayılar. Retrieved from https://www.sozcu.com.tr/2020/gundem/istanbulda-en-cok-suriyeli-hangi-ilcede-iste-sayilar-5646012/.

69.

Sputnik (2017, June 23) Halk, Tarikat ve Cemaatlerden Uzaklaştı. Retrieved from https://tr.sputniknews.com/analiz/201706231029020267-halk-tarikat-ve-cemaatlerden-uzaklasti/.

70.

Stokes

(2005) Perverse accountability: A formal model of machine politics with evidence from Argentina. American Political Science Review 99(3): 315–325.

71.

Stokes

(2007) Political clientelism. In: Boix

Stokes

(eds) The Oxford Handbook of Comparative Politics. Oxford: Oxford University Press, pp. 604–627.

72.

Stokes

Dunning

Nazareno

, et al. (2013) Brokers, Voters, and Clientelism: The Puzzle of Distributive Politics. New York: Cambridge University Press.

73.

Szwarcberg

(2015) Mobilizing Poor Voters: Machine Politics, Clientelism, and Social Networks in Argentina. New York: Cambridge University Press.

74.

Trantidis

(2013) Clientelism and the classification of dominant party systems. Democratization 22(1): 113–133.

75.

Trantidis

(2016) Clientelism and Economic Policy, Greece and the Crisis. London: Routledge.

76.

Trantidis

Tsagkroni

(2017) Clientelism and corruption: Institutional adaptation of state capture strategies in view of resource scarcity in Greece. The British Journal of Politics and International Relations 19(2): 263–281.

77.

Tuğal

(2009) Passive Revolution: Absorbing the Islamic Challenge to Capitalism. Stanford: Stanford University Press.

78.

Wantchekon

(2003) Clientelism and voting behavior: Evidence from a field experiment in Benin. World Politics 55(3): 399–422.

79.

White

(2002) Islamist Mobilization in Turkey: A Study in Vernacular Politics. Seattle, WA: University of Washington Press.

80.

Yeldan

Ünüvar

(2016) An assessment of the Turkish economy in the AKP era. Research and Policy on Turkey 1(1): 11–28.

Party	b1 (gift and payment)	b2 (social policy benefits)	b3 (patronage jobs)	b4 (procurement contracts)	b5 (regulatory favors)	b11 (effectiveness of clientelist targeting)
HDP	2.0	3.3	3.7	3.4	3.4	2.9
MHP	1.5	2.9	3.7	3.2	3.3	2.6
AKP	4.0	3.8	3.9	3.9	4.0	3.7
CHP	2.1	3.0	3.5	3.2	2.9	2.4

Party	b1 (gift and payment)	b2 (social policy benefits)	b3 (patronage jobs)	b4 (procurement contracts)	b5 (regulatory favors)	b11 (effectiveness of clientelist targeting)
HDP	2.0	3.3	3.7	3.4	3.4	2.9
MHP	1.5	2.9	3.7	3.2	3.3	2.6
AKP	4.0	3.8	3.9	3.9	4.0	3.7
CHP	2.1	3.0	3.5	3.2	2.9	2.4

Party	b1 (gift and payment)	b2 (social policy benefits)	b3 (patronage jobs)	b4 (procurement contracts)	b5 (regulatory favors)	b11 (effectiveness of clientelist targeting)
HDP	2.0	3.3	3.7	3.4	3.4	2.9
MHP	1.5	2.9	3.7	3.2	3.3	2.6
AKP	4.0	3.8	3.9	3.9	4.0	3.7
CHP	2.1	3.0	3.5	3.2	2.9	2.4