https://orcid.org/0000-0002-2421-9109
![Sample our Area Studies journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5922/TOC-Subject-Sample-2021-Area-Studies.jpg"})
ABSTRACT
Why do voters accept clientelism? Previous research suggests that poorer voters are more likely to accept clientelistic benefits. However, identities may moderate the effect of poverty through identity-based economic comparisons across groups. The role identity plays in partisanship, and dense ethnic identity networks may make it easier for parties to enforce clientelism among specific groups. This paper presents evidence from a survey experiment in Turkey to argue that politicized Kurdish ethnic identity, combined with heightened perceptions of relative economic deprivation, explains why certain voter groups are more likely to accept clientelism. Additionally, experimental evidence shows that support for clientelism may depend on the quality of benefits rather than quantity. Focusing only on the amount of resources or the recipients’ economic conditions may fail to explain why certain voters accept clientelism more in the Turkish context.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Stokes, ‘Perverse Accountability’; Kitschelt and Wilkinson, Patrons, Clients and Policies; and Hicken, ‘Clientelism.’
2 Gans-Morse, Mazzuca, and Nichter, ‘Varieties of Clientelism,’ and Nichter, ‘Vote Buying or Turnout Buying?’
3 Diaz-Cayeros, Estévez, and Magaloni, The Political Logic.
4 Van de Walle, ‘Presidentialism and Clientelism.’
5 Stokes et al., Brokers, Voters, and Clientelism.
6 Nichter and Peress, ‘Request Fulfilling.’
7 On the effect of the secret ballot, see Mares, From Open Secrets, and Kam, ‘The Secret Ballot’. On the role of poverty, see Bobonis et al., ‘Vulnerability and Clientelism.’
8 Chandra, ‘Counting Heads.’
9 Kenny, Populism and Patronage.
10 For recent research on clientelism in Turkey, see Sayarı, ‘Interdisciplinary Approaches,’ which indicates that recent years have ‘witnessed the emergence of a second major research wave that has provided important new insights and knowledge’ (p. 660).
11 Yörük, ‘Welfare Provision,’ and Akdag, Ethnicity and Elections.
12 On the relationship between poverty and clientelism, see introductory studies by Stokes, ‘Perverse Accountability’; Kitschelt and Wilkinson, Patrons, Clients and Policies; and Hicken, ‘Clientelism.’
13 Weitz-Shapiro, Curbing Clientelism.
14 Brusco, Nazareno, and Stokes, ‘Vote Buying.’
15 Hicken, ‘Clientelism.’
16 Tajfel, Social Identity.
17 Huddy, ‘From Group Identity.’ For different applications of social identity on vote choice and ethnic or class group membership, see Corstange, ‘Ethnicity on the Sleeve’; Huber, ‘Measuring Ethnic Voting’; and McClain et al., ‘Group Membership.’
18 Grimalda, Buchan, and Brewer, ‘Social Identity Mediates.’
19 Huddy, ‘From Group Identity,’ 539.
20 Kitschelt, ‘Linkages.’
21 Auyero, Poor People’s Politics; Calvo and Murillo, ‘When Parties Meet Voters’; Larreguy, Olea, and Querubin, ‘The Role of Labor Unions’; and Thachil, ‘Elite Parties.’
22 Finan and Schechter, ‘Vote-buying,’ and Lawson and Greene, ‘Making Clientelism Work.’
23 Mares and Young, ‘The Core Voter’s Curse.’
24 Cruz, ‘Social Networks.’
25 Oakes, ‘Psychological Groups,’ 812.
26 Kasara, ‘Tax Me If You Can.’
27 Bobonis et al., ‘Vulnerability.’
28 Stokes, ‘Perverse Accountability.’
29 Kitschelt, ‘Linkages.’
30 Kramon, Money for Votes.
31 Wang and Kolev, ‘Ethnic Group Inequality.’
32 Note that different identity groups may respond positively to portfolio diversification by political parties regardless of their preferences for accountability. See Diaz-Cayeros, Estévez, and Magaloni, The Political Logic, and Stokes, ‘Perverse Accountability.’
33 McClain et al., ‘Group Membership.’
34 Huddy, ‘From Group Identity.’
35 Wang and Kolev, ‘Ethnic Group Inequality.’
36 Baldwin and Huber, ‘Economic versus Cultural Differences.’
37 Corstange, ‘Ethnicity on the Sleeve’; Isaksson and Bigsten, ‘Clientelism’; and Wang and Kolev, ‘Ethnic Group Inequality.’
38 For a similar relationship, see McClendon, Envy in Politics, which discusses the role of status motivations and envy to explain various facets of political life, such as redistributive preferences, policy implementation, and political participation. Additionally, Gidron and Hall, ‘Populism as a Problem,’ discuss how perceptions of social status can influence support for radical right and left political parties.
39 Trump, ‘When and Why Is Economic Inequality.’
40 Heper, The State Tradition.
41 Buğra and Keyder, ‘The Turkish Welfare Regime.’
42 Membership figures do not necessarily imply due paying party activists, but they may be relevant to gauge the potential client pool whose contact details are readily available for reach at party headquarters. Membership figures are taken from:
https://www.yargitaycb.gov.tr/kategori/117?page=24, Access date: 14 June 2022.
43 In a cross-national study of parties in 88 electoral democracies, Turkey is grouped with other ‘dominant clientelism’ systems. For details, see Kitschelt, ‘Clientelistic Linkage Strategies.’
44 Toprak, ‘Islam and Democracy.’
45 Arat, Rethinking Islam, 112.
46 Kitschelt and Kselman, ‘Do Institutions Matter.’
47 Aytaç and Çarkoğlu, ‘Ethnicity and Religiosity-Based Prejudice.’
48 Bulut, ‘How Clientelistic Parties.’
49 For a review of alternative benefits, see Sayarı, ‘Interdisciplinary Approaches.’
50 Yıldırım, ‘Clientelism and Dominant Incumbent Parties.’
51 According to IDEA database, vote buying is illegal in 164 countries including Turkey. Turkish law no. 298, art. 152 indicates that offers of vote buying and acceptance of such benefits is punishable by law from 1 to 3 years of imprisonment. Art. 57 of the same law also specifically bans distribution of any gifts or benefits besides from informative material (i.e. brochures, booklets, posters, CD and DVDs are excluded from this ban).
52 Gürakar and Bircan, ‘Redistributive Politics.’
53 Akdag, Ethnicity and Elections.
54 Yörük, ‘Welfare Provision.’
55 Aktürk, Regimes of Ethnicity.
56 Yörük, ‘Welfare Provision.’
57 Aytaç, ‘Distributive Politics.’
58 Akdag, Ethnicity and Elections in Turkey.
59 Sayarı, ‘Interdisciplinary Approaches.’
60 Kudat, ‘Patron-Client Relations.’
61 Akdag, Ethnicity and Elections.
62 Yoltar and Yörük, ‘Contentious Welfare,’ and Tutkal, ‘Trustees.’
63 Aytaç and Çarkoğlu, ‘Ethnicity and Religiosity-Based Prejudice,’ and Sarigil, ‘Ethnic and Religious Prejudices.’
64 Gourlay, ‘Oppression.’
65 Laebens, and Öztürk. ‘Partisanship and Autocratization.’
66 Karakoç, ‘Ethnicity and Trust.’
67 Tezcur, ‘Electoral Behavior.’
68 Sarigil and Fazlioglu, ‘Exploring.’
69 Sarigil, ‘Curbing Kurdish Ethno-Nationalism.’
70 Tezcür and Gurses, ‘Ethnic Exclusion.’
71 Experimental balance tests with the full and the effective samples (N = 1097) are presented in the Appendix.
72 For further information on the survey methodology, questionnaire and sampling frame, see the relevant documentation on the Turkish fieldwork at GESIS, Data catalog identifier: ZA5950 (International Social Survey Programme: National Identity III – ISSP 2013) https://search.gesis.org/research_data/ZA5950
73 The answer categories are: I … 1 – would definitely … vote for this candidate, 2 – would likely vote, 3 – probably would not vote, 4 – definitely would not vote. For ease of interpretation, I created a binary variable (support for clientelism) which takes a value of 1 if the respondent would definitely (likely) vote for the candidate and 0 if else. Replicating binary models presented in the following section with ordinal choice models does not change results statistically or substantively. 157 respondents (4.5 percent) did not respond to the experimental question. This sample is left out of the analysis.
74 Readers can also contact the author for the Appendix.
75 Nichter and Peress, ‘Request Fulfilling.’
76 Calvo and Murillo, ‘When Parties Meet Voters.’
77 However, additional models in the supplementary information replicate probit models with OLS, which may be considered an alternative for such analyses (e.g., Gomila, ‘Logistic or Linear?’)
78 Çarkoğlu and Aytaç, ‘Who Gets Targeted.’
79 Sayarı, ‘Interdisciplinary Approaches,’ and Yıldırım, ‘Clientelism and Dominant Incumbent Parties.’
80 The full table with all of the coefficients is reported in the Appendix.
81 Along this line, both Kudat and Çarkoğlu and Kalaycıoğlu have argued that ‘mobilized votes’ by authority figures in the household (e.g., husbands and fathers) has been a prominent part of Turkish election experience, even if such mobilized voting has been decreasing in recent periods due to urbanization and higher educational attainment, especially among women. For an ethnographic account of how patronage is related to such vote mobilization in the Kurdish-majority areas, see Kudat, ‘Patron-Client Relations:’. For a more recent account on this point, see Çarkoğlu and Kalaycıoğlu, Turkish Democracy Today.
82 Alesina, Glaeser, and Glaeser, Fighting Poverty, and Wang and Kolev, ‘Ethnic Group Inequality.’
83 Baldwin and Huber, ‘Economic versus Cultural Differences.’
84 Kitschelt and Kselman, ‘Do Institutions Matter.’
Additional information
Notes on contributors
Kerem Yıldırım
Kerem Yıldırım is an Assistant Professor in the Department of Political Science and Public Administration at Kadir Has University. He is currently working as a research associate at Democratic Accountability and Linkages Project (DALP2) and his research interests include party-voter linkages, comparative political behavior, and political communication. For further details, you can visit www.keremyildirim.com.