Municipal Service Delivery, Identity Politics and Islamist Parties in Turkey

Melissa Marschall and Marwa Shalaby, Rice University; Saadet Konak Unal, University of Houston 

The rise of Islamist politics over the past few decades has yielded distinctively diverse outcomes across the Middle East and North Africa (MENA) region. Despite the short-lived ebb of the Islamist parties in the aftermath of the Arab uprisings, more recent electoral victories have proved these parties to be an enduring phenomenon showing little signs of abating. Turkey’s Justice and Development Party (Adalet ve Kalkinma Partisi, AKP) is a particularly strong case in point. It remains the only Islamist party in the region to have won four consecutive national elections, maintaining a solid support base for almost two decades. Though much has been written about the party’s electoral and governing strategies on the national level, few studies have examined its efforts on the local level. This is a striking oversight given the importance of urban processes and outcomes for the both the AKP and the Islamist parties that preceded it.

In this research note, we refocus attention to the local level and investigate a set of research questions related to municipal politics and governance under the AKP and Turkey’s other Islamist parties. We briefly describe the rise and consolidation of Islamist parties in Turkish municipal politics and then consider how these parties have fared when it comes to governance. Focusing first on service delivery, we present results from related research that suggests Islamist parties do tend to prioritize redistributive policy more than their secular counterparts, but that under the AKP this has been achieved largely through housing policy, that are more neoliberal in nature. From here, we ask whether and how municipalities address issues related to identity and morality. In particular, do municipalities governed by Islamist mayors promote Islamic political identity? Using qualitative methods of analysis, we content analyzed municipal websites to evaluate whether municipalities that strongly support Islamist parties in mayoral elections feature web content promoting religion, morality, and Islamic identity, while those strongly supporting secular parties shy away from such content and instead highlight activities, events and images consistent with Western lifestyles and secular identities.

The rise and evolution of Islamist parties in Turkish local politics

Religious parties are a relatively recent phenomenon in Turkey, in part because the Turkish constitution prohibits the use of religious symbols for political purposes. It was not until 1970 that the first Turkish party with clear Islamic credentials, the National Order Party (MNP), emerged. Although the MNP and its most immediate successor, the National Salvation Front (MSP), were both banned in relatively short order by Turkey’s Constitutional Court, the Welfare Party (WP), proved to be not only more durable, but also much more effective at winning elections. Though the WP won less than 5 and 10 percent of the vote respectively in the 1984 and 1989 municipal elections (Eligur 2010), by 1994 it had established itself as formidable party, winning nearly 20 percent of the vote and more than 100 mayoralties. As Figure 1 shows, the WP had elected mayors in more than half of the provinces, and more than 40 percent had two or more municipalities governed by this Islamist party.

Figure 1: Vote Share and Percent of WP Municipal Mayoralties, by Province, in 1994

While the WP was truly distinctive in its emphasis on the provision of social services, its strident pro-Islamic, pro-Ottoman (and thus anti-secular and anti-Western) message and culture led it to also be banned by the Constitutional Court in 1997 (Akinci 1999). By 2001, the National Outlook Movement had split into two parties: the AKP and the Felicity (Saadet) Party. While the former, led by (current president) Recep Tayyip Erdoğan, adopted a reformist, pro-western stance, tempered its Islamist identity and aggressively sought to broaden the party’s base, the latter remained loyal to Necmettin Erbakan and the core ideas of the National Outlook (Gumuscu & Sert 2009; Taspinar 2012). In the 2004 local elections, the AKP registered an extremely strong showing capturing 512 of the 914 (56 percent) district municipal mayoralties.[1] The Saadet Party had much less impressive results, winning only 4 percent of the vote and 13 district municipal mayoralties. Similar results obtained in the 2009 and 2014 municipal elections – with the AKP continuing to dominate and the Saadet Party averaging less than 5 percent of the vote.

Service delivery, developmental policy, and municipal governance

The question of how Islamist parties govern once in office is an important one. Unfortunately, much of what we know about service delivery in Turkish local politics comes from research on early Islamist parties, particularly the WP. The WP was essentially an urban party-machine that capitalized on the social, economic, and existential problems brought on by the processes of urbanization and rural-urban migration that had accelerated in the 1970s and 1980s (Eligür 2012). Using machine-style politics, the party provided newly arrived urban residents with material benefits such as patronage, jobs and social services in exchange for their votes. The WP also featured an extremely loyal cadre of foot soldiers who went door-to-door to provide material support in the form of food, financial assistance, and to spread and the party’s message of “Just Order” (Atacan 2005; Eligür 2010; White 2002).

Previous studies suggest that the WP largely followed through on its promises to provide social and economic support for the working-class voters who represented the core supporters of the party. Not only did WP mayors tend to be less corrupt and more effective at providing municipal services to working-class neighborhoods than mayors representing other parties (Akinci 1999; Sayari 1996), but in WP municipalities, buses ran, the garbage was collected, and the quality of social services had generally improved (White 1997: 26). When it comes to service delivery under AKP led municipalities, however, evidence suggests that the party’s commitment to social welfare may not be as strong. From the beginning, the party adopted a pro-business stance and promoted market forces and neoliberal policies (Karaman 2013). One concrete example of this is housing policy, where urban transformation projects and a massive expansion of social housing under Turkey’s Mass Housing Development Authority (TOKİ) have stimulated the construction sector. However, recent work by Marschall et al. (2016) found that housing contracts awarded by the central government to local municipalities also fueled distributive politics in the form of jobs, contracts, and subsidized housing, which, in turn, played a key role in consolidating and expanding the AKP’s electoral base.

In related research, we analyzed municipal expenditures on housing and social welfare to see if municipalities with AKP or Saadet Party mayors spend more on these policies than municipalities with non-Islamist mayors. Our multivariate models also tested to see whether municipalities formerly governed by WP mayors spend more on social welfare today. We found that whereas municipalities with AKP or Saadet mayors do not allocate more to social welfare than non-Islamist municipalities, those with the experience of WP mayors (in 1994) have significantly higher social welfare program expenditures. On the other hand, municipalities with AKP mayors have significantly higher expenditures on housing than all other municipalities. Together these findings suggest that, ceteris paribus, the presence of Islamist parties in municipal government matters for local service delivery outcomes in Turkey. At the same time, not all Islamist parties in Turkey are the same. When it comes to social welfare spending, what matters most is whether the municipality has historical ties to the WP, not which party occupies the mayor’s office today. On the other hand, the mayor’s party plays a substantial role in housing policy: Controlling for migration and other factors that might influence housing demand, AKP-led municipalities spend significantly more on housing than municipalities with secular mayors.

Identity politics and municipal governance

Beyond service delivery, our final research question focuses on identity politics and builds on work by urban politics scholars like Sharp (2002), who argue that identity issues involve values and moral concerns and are not only distinct from “politics as usual” – fixing potholes, picking up garbage, and getting the buses to run on time – but also quite salient in people’s everyday lives. Because these issues typically involve competing claimants who can be extraordinarily passionate and strident, the way in which local governments involve themselves in identity politics has important implications for their ability to govern, manage conflict, and win elections. How does this work in Turkish local government? Do Islamist mayors promote Islamic political identity and the Islamization of Turkish society, while those governed by secular parties promote a secular and/or Turkish political identity?

As a first step in investigating identity politics in municipal governance, we selected a set of municipalities with strong support for either Islamist or secular parties and content analyzed their websites (see Appendix for details about the sampling frame and coding). Our goal was to systematically measure the extent to which municipal websites communicated Islamist versus secular frames and identities. We expected municipalities that strongly support Islamist parties would feature web content promoting religion, morality, and Islamic identity, while those strongly supporting secular parties would highlight activities, events and images consistent with Western lifestyles and secular identities.

Though the conflict between secularists and Islamists is longstanding, identity issues came to play an especially important role with the rise of the WP (Yavuz 1997:74; see also Evren Celik Wiltse’s essay). At the local level, WP mayors undertook actions that ranged from the purely symbolic (changing street names and removing statues), to the more substantive (closing down restaurants and nightclubs that served alcohol) (Akinci 1999). The promotion of Islamic identity by AKP mayors has also become more apparent in recent years. The Istanbul Metropolitan Municipality’s 1.2 million dollar Panorama 1453 History Museum, which gives visitors the opportunity to see (in 3-D) Constantinople’s fall to the Turks, is but one example of this trend.

The content analysis of municipal websites targeted issues, activities, and events related to religion, morality, and Islamic versus secular/western identity. One of the most prominent dimensions of identity politics relates to veiling. Here we find a sharp contrast in the images featured on websites for municipalities with Islamist and secular mayors. As the summary statistics in Table 1 reveal, 87 percent of AKP municipalities and 89 percent of core Islamist municipalities (those with WP mayors in 1994 and consecutive Islamist party mayors since 1994) feature pictures of covered women on their websites, whereas only 44 percent of CHP municipalities do. Images of veiled children are also present on 50 percent of websites for municipalities classified as strong Saadet party. In contrast, none of the websites in our sample of CHP and BDP municipalities (Turkey’s main secular parties) include any images of veiled children.

We also find religious symbols are much more visible on Islamist municipalities’ websites: 40 percent of AKP and 56 percent of core Islamist municipalities feature images of mosques on the front page of their website, whereas none of the CHP municipality websites do. In addition, these municipalities are also much more likely to prominently advertise religious holidays, events, customs, and symbols on their municipal websites. For example, 70 percent of all Islamist municipalities include content about circumcision festivals, religious feasts such as kandils, the holy birth of Mohammed, or Ottoman Empire military victories (conquest of Istanbul) that the municipality organizes or sponsors. This contrasts sharply with the CHP municipalities we analyzed, where none of this type of website content was found.

When it comes to websites of municipalities with strong secular (CHP and BDP) support, we found content prominently featuring the arts, music, sports, and national holidays such as National Sovereignty Children’s Holiday (April 23), the Commemoration of Ataturk (November 10), and Youth and Sports Day (May 19) – all of which celebrate secular values. For example, every CHP municipal website in our sample featured its own centers where children and adults can participate in music, arts and sports. In addition, 89 percent of CHP and 80 percent of the BDP municipal websites have a picture of Atatürk, the founder of Turkish Republic, on the front page of their websites. In contrast, only 20 percent of the AKP municipalities and 25 percent of the Saadet municipalities have a picture of Atatürk on main page their websites. Overall, our analysis finds that the activities and events sponsored by municipal governments vary considerably and align closely with the cultural frames and values of the governing parties.


In this brief research note we have documented the evolution of Islamist parties in Turkish local politics and provided evidence to suggest that they govern in ways that distinguish them from their secular counterparts. Our quantitative analysis reveals that Islamist parties not only differ from non-Islamist parties in terms of prioritizing municipal services that promote redistributive versus development policy, but that key differences also exist among Islamist parties. Our qualitative analysis finds that overall Islamist and secular parties sharply differ from each other with respect to identity politics. Municipalities with strong electoral support for Islamist parties use their websites to advertise religious values, lifestyle, and morals, whereas municipalities with strong electoral support for secular parties feature website content that promotes Western lifestyles and secular values. While this investigation reveals primarily symbolic aspects of identity politics, the next step of this research involves more systematic analyses based on surveys of the municipalities themselves. This survey will allow us to measure mayors’ priorities and attitudes about a more comprehensive set of municipal services so that we can better evaluate the processes and outcomes of municipal governments in Turkey.

[1] AKP was a majority winner in 177 of the 512 (35 percent) mayoral elections it won.

[2] Hamamözü, Kalkandere, and Gerger won in 2014 and Korkut won in 2009.


Akinci, Ugur. 1999. “The Welfare Party’s Municipal Track Record: Evaluating Islamist Municipal Activism in Turkey.” Middle East Journal 53 (1): 75-94.

Çınar, Alev İnan. 1997. “Refah party and the city administration of Istanbul: liberal Islam, localism and hybridity.” New Perspectives on Turkey 16: 23-40.

Eligür, B. (2009). :Turkey’s March 2009 Local Elections.” Turkish Studies 10 (3): 469-496.

_____. (2010). The Mobilization of Political Islam in Turkey. NY: Cambridge University Press

Gumuscu, Sebnem and Deniz Sert. 2009. “The power of the devout bourgeoisie: the case of the Justice and Development Party in Turkey. Middle Eastern Studies 45 (6): 953-968.

Karaman, Ozan. 2013. “Urban Neoliberalism with Islamic Characteristics.” Urban Studies 50(16): 3412–3427.

Mardin, S. 1993. “Religion and Secularism in Turkey,” in The Modern Middle East ed. Hourani, A., Khoury, P. S., and Wilson, M.C. Berkley, CA: University of California Press.

Marschall, Melissa, Abdullah Aydogan and Alper Bulut. 2016. “Does housing create votes? Explaining the electoral success of the AKP in Turkey.” Electoral Studies 42: 201-212.

Sayari, Sabri. 1996. “Turkey’s Islamist Challenge.” Middle East Quarterly 3(3): 35-43.

Taspinar “Turkey: The New Model.” In Robin Wright, ed., The Islamists are Coming: Who are They Really? . Washington DC: The Wilson Center.

Sharp, Elaine B. 1997. 2002. “Culture, Institutions, and Urban Officials’ Responses to Morality Issues.” Political Research Quarterly 55 (4): 861-883

White, Jenny B. 2002. Islamist mobilization in Turkey: A study in vernacular politics. University of Washington Press.

_____. 1997. “Pragmatists or Ideologues? Turkey’s Welfare Party in Power.” Current History 96 (January): 25-30.

Yilmaz, Muzaffer Ercan. 2002. “The rise of political Islam in Turkey: the Case of the Welfare Party, Turkish Studies 13 (2): 363-378

Yavuz, M. Hakan. 1997. “Political Islam and the Welfare (Refah) Party in Turkey.” Comparative Politics. 30(1): 63-82.

Yavuz. H. 2003. Islamic Political Identity in Turkey. New York: Oxford University Press.


Case Selection and Methodology for Municipal Website Content Analysis

Empirical data for our content analysis were collected from the websites of 48 district municipalities in Turkey (April-May 2017). The cases were selected with two criteria in mind: (1) the electoral strength of Islamist versus other parties, (2) broader contextual features of the municipality (population, region). District municipalities were selected based on parties’ vote share over the past three municipal elections (2004, 2009, 2014). We included cases where Islamist and secular parties were uniformly in power since 2004 and won with the largest vote shares in municipal elections. Among these cases, we selected municipalities where Islamist parties compete both with each other (e.g., AKP and Saadet are the two top vote getters) and where they compete with secular parties (e.g., AKP and CHP are the two top vote getters). In selecting these cases, we also took our second criterion into account to ensure that our sample captured variation in the size and location of municipalities.

For the present analysis, we focus only on the municipalities where single parties are dominant (for CHP, AKP, and BDP, this is defined as winning all of the last three municipal elections, for Saadet this is defined as winning at least one election out of three municipal elections). We define Core Islamist municipalities as those electing WP mayors in 1994 and having consecutive Islamist party mayors since 1994. Table A1 reports cases included in our sample for each of our municipality types.

Table A1: Sample Cases Analyzed

Type Municipalities (by Region & Province) N
Strong CHP Aegean: Aydin (Didim), Izmir (Karşıyaka), Muğla (Datça); Marmara: Balıkesir (Ayvalık), Istanbul (Beşiktaş, Kadıköy), Edirne (Keşan); Mediterranean: Hatay (Defne, Samandağ) 9
Strong BDP Southeast Anatolia: Diyarbakır (Lice, Dicle), Şırnak (İdil) ; Eastern Anatolia : Hakkari (Yüksekova), Van (Başkale) 5
Strong AKP Black Sea: Ordu (Kabadüz), Rize (Güneysu, İyidere), Trabzon (Hayrat), Central Anatolia: Ankara (Altındağ, Çubuk), Karaman (Başyayla), Kayseri (Kocasinan, Melikgazi) Nevşehir (Acıgöl); Eastern Anatolia: Malatya (Pütürge); Marmara: Istanbul (Bağcılar, Bayrampaşa, Güngören); Southeastern Anatolia: Şanlıurfa (Haliliye) 15
Strong Saadet Black Sea: Amasya (Hamamözü), Rize (Kalkandere); Eastern Anatolia: Muş (Korkut)[2]; Southeastern Anatolia: Adıyaman (Gerger) 4
Core Islamist Central Anatolia: Konya (Karatay, Meram, Selçuklu), Eastern Anatolia: Erzincan (Kemaliye); Southeastern Anatolia: Adıyaman (Kahta); Marmara: Bursa (Inegöl), Istanbul (Esenler), Kocaeli (Gölcük), Sakarya (Akyazı) 9
Total Cases 42

The content analysis was based on a coding scheme developed to measure website content promoting or explaining: (1) issues, activities and events related to religion, morality and identity, (2) social service provision. Coders examined the home/front page of each municipality’s website based on the following set of questions:

Q1- Is there any picture of Atatürk on the front page of the municipality’s website?

Q2- Is there any picture of covered women on the front page of the municipality’s website?

Q3- Is there any picture of non-covered women on the front page of the municipality’s website?

Q4-Is there any picture of covered girls on the front page of the municipality’s website?

Q5- Does the municipality organize any secular events? Does it celebrate the national holidays and deliver messages on these days?

Q6- Does the municipality organize any religious events? Does it celebrate the religious holidays and deliver any religious messages on these days?

Q7- Does the municipality have activities related to arts and music?

Q8- Does the municipality have activities related to sports?