Social Welfare: Interdisciplinary Approach eISSN 2424-3876
2025, vol. 15, pp. 82–105 DOI: https://doi.org/10.15388/SW.2025.15.5

Mapping Welfare Diversity: A Regional Analysis of Türkiye’s Welfare Regime through Esping-Andersen’s Welfare Regimes Typology

Nurten Ebru Özdemir
Bitlis Eren University, Turkey
E-mail: neozdemir@beu.edu.tr
https://orcid.org/0000-0002-3113-2378
https://ror.org/00mm4ys28

Doğa Başar Sarıipek
Kocaeli University, Turkey
E-mail: sariipek@kocaeli.edu.tr
https://orcid.org/0000-0002-3525-5199
https://ror.org/0411seq30

Gökçe Cerev
Kocaeli University, Turkey
E-mail: cerevgokce4@gmail.com
https://orcid.org/0000-0001-9908-343X
https://ror.org/0411seq30

Abdülkadir Akturan
Piri Reis University, Turkey
E-mail: aakturan@pirireis.edu.tr
https://orcid.org/0009-0008-9107-0333
https://ror.org/02eq60031

----------------------------------------------------------

This study was supported by the Scientific and Technological Research Council of Turkey under the TÜBİTAK 1002B project and produced within the scope of a doctoral thesis.

----------------------------------------------------------

Abstract. This study develops a comprehensive regional welfare map of Türkiye by using Esping-Andersen’s welfare typologies with the objective to analyze regional variations in welfare policies. By focusing on poverty and unemployment as key indicators, the research categorizes Türkiye’s welfare systems at both macro and micro levels, offering valuable insights for shaping future social welfare policies. A total of 243 semi-structured interviews were conducted across all 81 provinces with Non-Governmental Organizations (NGOs) selected based on their involvement in welfare-related activities. The data, analyzed by using Maxqda software, reveals that Turkey exhibits a hybrid welfare system, incorporating liberal, conservative, and social democratic elements, with significant regional disparities. NGOs play a critical role in bridging gaps in the State welfare provision, particularly in regions with limited State intervention. Supported by the Scientific and Technological Research Council of Turkey (TÜBİTAK), this study makes a significant contribution to the welfare typology literature and provides practical insights for reducing regional inequalities.

Keywords: Regional welfare map, poverty, unemployment, welfare typologies, welfare regime in Türkiye.

Introduction

Welfare policies form a central pillar of social policy, by virtue of shaping the structure and functioning of societies by addressing key issues such as poverty, unemployment, and social inequality. These policies are not only mechanisms of economic redistribution, but they also reflect the ideological underpinnings of State governance. Welfare is thus more than just an economic issue; it is a socio-political framework which influences every aspect of a country’s development. Since the outbreak of Industrial Revolution, welfare systems have become integral to ensuring social stability and addressing the disruptions caused by economic modernization, including the rise of income inequality and the shifting dynamics of labor markets.

The Esping-Andersen’s welfare regimes typology – liberal, conservative, and social democratic – provides a robust framework for analyzing how different countries are approaching welfare provision. This typology is traditionally applied at the national level, categorizing states based on the degree to which the market, state, NGOs, or family plays a role in welfare distribution. However, such a national focus often overlooks the significant regional variations within countries, especially those with diverse social, economic, and political landscapes. Türkiye, with its distinct regional disparities and unique socio-political history, represents an important case for studying welfare at the regional level. Table 1, which provides a classification of regional units in Türkiye, serves as the foundation for this study by delineating the territorial divisions that shape the welfare dynamics across the country. Given its geographic position straddling Europe and Asia, Türkiye’s welfare system is influenced by a combination of Eastern and Western welfare traditions, thereby making it a compelling subject for examining its regional welfare typologies.

The objective of this study is to create a comprehensive regional welfare map of Türkiye by applying the Esping-Andersen’s welfare regimes typology to different regions, as detailed in Tables 2 through 13. A particular focus is placed on the indicators of poverty and unemployment, as these two measures offer tangible insights into the effectiveness of the social welfare policies and provide a basis for comparing welfare outcomes across different regions. While poverty and unemployment are prevalent across Türkiye, they manifest in distinct ways across different welfare regime regions, as detailed in Tables 2 through 13, each of which presents the welfare structure of a specific region.

This research is pioneering in the sense that it provides the first in-depth regional analysis of welfare in Türkiye, going beyond national categorizations to explore how welfare systems operate on a micro level. Supported by the Scientific and Technological Research Council of Turkey (TÜBİTAK), the study employs a qualitative research approach to capture the nuanced ways in which welfare policies are implemented and experienced across Türkiye’s diverse regions. Semi-structured interviews were conducted across all 81 provinces, involving 243 representatives from the relevant institutions, including Non-Governmental Organizations (NGOs), whose activities in each welfare regime region are explored in Tables 2 through 13. These organizations were selected based on their type of activity, scope, the number of beneficiaries, and their involvement in welfare-related policies, thus providing a rich source of qualitative data.

The analysis, conducted by using Maxqda software, allowed for the identification of regional welfare typologies, highlighting the variations in welfare regimes across Türkiye’s subregions. Tables 2 through 13 present a detailed examination of each region’s welfare regime, while illustrating the prevalence of State, market, and NGO involvement in welfare provision. For example, Table 2 (Istanbul Subregion) and Table 3 (Western Marmara) highlight the significant role of market-driven welfare, whereas Tables 4 and 5 (Aegean and Eastern Marmara) show a mixed welfare model with substantial State intervention. Similarly, Tables 6 through 9 (Western Anatolia, Mediterranean, Central Anatolia, and Western Black Sea) reveal variations in family-based and NGO-driven welfare approaches.

Furthermore, the study examines how poverty and unemployment, as key indicators, vary across different regions and the extent to which the currently existing welfare policies are addressing these issues. Tables 4 through 13 provide an in-depth look at regional welfare disparities, illustrating how socio-economic factors shape welfare distribution. The classification of regions according to the Esping-Andersen’s typology, outlined in these tables, allows for a structured comparison of welfare outcomes across Türkiye.

The role of different welfare providers – the market, the State, and the family – is also assessed through the comparative analysis of different regional welfare models presented in Tables 2 through 13. Additionally, detailed case studies of specific provinces, captured within these tables, further illustrate the nuances of welfare implementation at the local level. Finally, the comparative analysis summarized in Table 13 (Southeast Anatolia) highlights the overall successes and challenges of different welfare models in Türkiye, offering critical insights for both academia and policymakers.

By integrating regional analysis into welfare state research, this study provides a framework for developing targeted welfare policies which take into account the specific needs and challenges of different regions. The findings contribute not only to the academic literature on welfare state typologies but also offer practical guidance for policymakers seeking to address regional disparities in the social welfare provision.

Theoretical Background

Social policy practices play a crucial role within the framework of welfare and welfare typologies. Their main aim is to mitigate social inequalities, support disadvantaged groups, and create mechanisms which would guarantee the fulfillment of fundamental needs for all individuals. One of the key objectives for states is to develop and implement effective policies that enhance welfare, which is a central concern in the social policy (Şenkal, 2024; Küçükoba, 2022). To improve societal welfare, it is essential to first accurately assess the current level of welfare, and then measure the progress achieved after implementing these policies. Therefore, the success of welfare policies is closely tied to the precise measurement of welfare. This underscores the importance of selecting the appropriate indicators for evaluating welfare. Poverty and unemployment statistics are widely recognized as two critical indicators in this regard (Koray, 2005). These issues have historically been, and still continue to be, major challenges for both developed and developing countries, presenting significant risks that demand attention.

Poverty is a conceptually multidimensional issue widely discussed in the literature. As a critical aspect of welfare policies, poverty must be evaluated from various angles, including the differing development levels of countries, the diversity of social resources and needs, the impact of time and place, and the variations in expectations and consumption habits. The lack of a universal consensus on the definition of poverty further complicates the development of a single, all-encompassing definition. Historically, poverty was first understood through the lens of accessibility and deprivation of material resources. This perspective primarily framed poverty as the inability to meet basic needs such as nutrition, shelter, and physiological necessities, which are essential for survival (Rowntree, 1901). This economic interpretation of poverty emphasized the insufficiency of income to cover these minimum needs (Lewis, 1966).

However, as perceptions and contributing factors evolved over time, the concept of poverty expanded to include an individual’s ability to meet the standards of the society in which they live. Thus, poverty began to be assessed by using not only material but also subjective data. Today, poverty is defined in broader terms, by encompassing social, cultural, technological, and other criteria which vary across different quantitative and qualitative dimensions (Zimmermann and Graziano, 2020; Benjamin, 2006). This evolution has led to the emergence of two main approaches to defining poverty: the one-dimensional view, which focuses on per capita income and expenditures by using a single indicator, and the multidimensional view, which considers not only economic factors but also social, cultural, and political elements (Padda and Hameed, 2018).

The multidimensional approach to poverty is crucial for the development of social policies which address not just economic growth but also social development. This broader perspective defines poverty more generally as the lack of sufficient income to maintain a sustainable standard of living, or the deprivation of essential physiological needs such as food, shelter, water, and more.

Unemployment is widely recognized as the second crucial indicator for measuring welfare. Historically and in the present day, the most significant impact of unemployment remains the individual’s inability to meet their needs due to a lack of income. The industrial revolution brought about profound changes, transforming unemployment into a chronic issue with global implications, and leading to its entrenchment as a persistent problem. Unemployment is not only an economic challenge but also a social and psychological one, affecting both individuals and society at large due to various factors. As such, unemployment is intrinsically linked to welfare. The adverse effects of unemployment are now felt across entire societies, making it a global issue which concerns all nations (Uyar, 2008).

Poverty and unemployment are central concerns in welfare policies, particularly within the realm of social policy, and are crucial to the overall well-being of society. Welfare policies address these issues while also influencing the reshaping of other domains such as economic policies, tradition, culture, politics, and broader social policies. The scope and implementation of welfare policies vary based on a country’s level of development, as well as on its economic, political, and cultural structures. Despite these variations, the common objective of welfare policies across states is to enhance the welfare of both individuals and society. In this context, welfare policies are often closely associated with the concept of the social state, and they aim to meet individuals’ basic needs, reduce poverty and social exclusion, and promote equality and justice. To achieve these goals, states employ a wide range of social policy instruments, including social insurance, social security, cash and in-kind benefits, and subsidies (Briggs, 1961).

Within the realm of social policy, states have taken on various roles in their pursuit to enhance the societal welfare. Initially, states acted as passive observers in welfare policies, but, over time, they adopted a more interventionist stance, and ended up becoming central to the implementation of these policies (Roosma and Van Oorschot, 2020; Yay, 2020). The welfare state model that many states have embraced integrates a combination of social policies, social expenditures, institutional arrangements, and regulatory frameworks, all aimed at improving welfare outcomes (Esping-Andersen, 1991).

Those states which adopt welfare policies have been classified in various ways based on the methods they employ (Mishra, 1997; Flora, 1986). Among these classifications, Esping-Andersen’s “Three Worlds of Welfare Capitalism” (1990) stands out as a seminal framework in the study of the welfare state typologies. According to Esping-Andersen (1990; 1999), welfare state regimes are characterized by legal and organizational complexities. He defines them as systematic, legal, and organizational functions that regulate the interactions within the state-economy-household triangle, shaped by historical development (Kaariainen and Lehtonen, 2006).

Esping-Andersen’s ideal welfare state typology possess a holistic structure that enables welfare states to develop policies addressing the challenges and risks individuals may encounter throughout their lives. He categorizes welfare states into three distinct typologies: Liberal, Social Democratic, and Corporatist/Conservative, each reflecting different welfare policies practiced globally to varying degrees (Esping-Andersen, 1990). The diversity of these typologies is rooted in the roles of the market, state, and family. In Esping-Andersen’s framework, the market is central to the Liberal model, the family to the Conservative model, and the state to the Social Democratic model.

Esping-Andersen’s objective in developing this welfare regimes typology was to create a mid-level theory of welfare systems which would avoid both theological and functionalist approaches, which would emphasize commonalities and convergence, as well as post-modern perspectives, and which would focus on national and sub-national uniqueness (Gough, 2001; Frericks, Gurín, and Höppner, 2023). In the light of these approaches, the cultural, structural, and institutional characteristics of states are crucial in shaping welfare regimes. The overarching goals of welfare policies are to maximize the standards of equality and social citizenship, to provide basic income protection to citizens, and to minimize social exclusion. Consequently, welfare policies are significant both at the national level and within regions of a country, where distinct welfare regimes may emerge based on regional characteristics.

Research Methodology and Application

Purpose and Importance of the Research

In Türkiye, unemployment and poverty are not treated as issues necessitating direct political intervention, largely due to the country’s adopted welfare typologies, which incorporate various mechanisms to prevent these problems from becoming chronic (Buğra and Keyder, 2003: 16). This study aims to map the regional distribution of welfare typologies, identify the welfare models prevalent in different regions, and analyze their role in shaping welfare state structures. By utilizing unemployment and poverty indicators, this research contributes to the social policy literature by offering a regional assessment of the welfare typologies. Therefore, this study aims to create a regional welfare map of Türkiye by using Esping-Andersen’s Three Worlds of Welfare Capitalism as a framework. While the currently existing research on welfare regimes generally focuses on national-level evaluations, this study uniquely examines and interprets welfare regimes at the regional level within a single country, specifically, Türkiye. This makes the study a pioneering contribution to the literature. Additionally, it holds significant value in guiding micro-level welfare policies in Türkiye, with a focus on improving welfare through unemployment and poverty indicators at the regional level, and fostering advancements in the social policy.

The fieldwork for this study was conducted in collaboration with NGOs which play a crucial role in the implementation of welfare policies. The research involved NGOs in each province of Türkiye, by employing an inductive approach serving the objective to explore the subject comprehensively across the country’s administrative structure.

Scope of the Research

The research encompasses Türkiye as a whole, with an applied field study conducted across the entire country. Specifically, the study was implemented in all 81 provinces, thus reflecting Türkiye’s administrative divisions. To delineate the scope and limitations of the research, the study utilized the three sub-levels of the Statistical Classification of Territorial Units (NUTS) provided by the Turkish Statistical Institute. This classification includes 12 regions at Level 1, 26 regions at Level 2, and 81 provinces at Level 3. The scope of this study is primarily limited to the 26 regions classified under NUTS Level 2 in Türkiye. This limitation is intended to structure Türkiye’s welfare map into 12 broader regions, considering regional and provincial development levels within the three-tiered NUTS classification. This approach enhances the effectiveness of the assessment. Under NUTS Level 1, Türkiye is divided into 26 regions, and, in order to ensure homogeneity among the subregions, factors such as the population structure and the metropolitan status, have been considered. Within this framework, welfare typologies at the subregional level have also been analyzed with the objective to provide a more comprehensive evaluation.

In this research, the data were analyzed across the 81 provinces of Türkiye, categorized into 26 regions at Level 2 according to the NUTS classification of the Turkish Statistical Institute. This classification ensured uniformity within sub-regions, and allowed for a balanced interpretation of economic, cultural, and social differences, particularly considering the population structure of each province. The research facilitated an assessment initially of the 26 regions, and, subsequently of Türkiye as a whole, by using an inductive approach. Fieldwork for the study involved conducting face-to-face interviews with 243 officials from NGOs, with three interviews conducted per province. The field study was carried out between September 2022 and April 2023.

Research Method and Question

This study employed a mixed research method, integrating both quantitative and qualitative approaches. The quantitative component utilized numerical data from official statistical institutions, specifically, the Turkish Statistical Institute, with the objective to provide a foundational framework of poverty and unemployment data. The qualitative component involved a semi-structured interview form, which had been developed in advance by using scientific methods.

To define the research problem in the qualitative segment, questions from the ‘role of the State’ scale of the International Social Research Programme (ISSP, 2006) were adapted. The semi-structured interview questions were initially reviewed by language experts, and subsequently finalized based on feedback from academic experts in the field. The research questions addressed in the semi-structured interviews are as follows:

• As a humanitarian aid organization, what are your attitudes towards providing employment opportunities for citizens seeking work?

• What is your stance on supporting health insurance for sick citizens?

• What assistance do you provide to ensure a minimum standard of living for elderly individuals?

• Do you offer a minimum level of income to unemployed citizens?

• How do you address income inequality between high-income and low-income citizens?

• Do you provide financial support for university students from low-income families? If so, what are the criteria and amounts?

• Do you assist individuals who are unable to acquire housing with their own resources?

• How do disparities in the prices of goods and services and wages in the labor market impact social assistance financially?

• What in-kind and cash aid do you provide to disabled individuals to ensure a minimum standard of living?

• Do you ensure a minimum standard of living for the children of low-income families, and if so, at what educational levels and ages?

Validity, Reliability and Ethical Approval

In scientific research, the credibility of the results largely hinges on the validity and reliability of the study. While validity and reliability are crucial, Gruba and Lincoln emphasized the importance of credibility over these traditional metrics in qualitative research (Houser, 2015; Merriam, 2013; Whittemore, Chase, and Mandle, 2001).

In qualitative research, ensuring credibility involves several factors beyond just validity and reliability. These include the appropriateness of the research conditions, meticulous pre-planning of the interview questions, thorough documentation of the research process, detailed description of the data, and the selection of a sufficiently large sample (Başkale, 2016). The primary focus in qualitative research is to explore and interpret the existence and meaning of the phenomena under consideration. Validity in qualitative research is assessed through the comprehensive and detailed reporting of data and the clarity with which the researcher explains how the conclusions were reached (Yıldırım and Şimşek, 2011).

In qualitative research, validity is defined as the researcher’s ability to observe and interpret the phenomenon as it exists, with minimal bias (Kirk and Miller, 1986). Reliability, on the other hand, is characterized by the researcher’s diligence in obtaining accurate information, which is essential for the scientific integrity of the research, including data collection and analysis.

To ensure validity and reliability in this study, input was sought from a panel of experts: four in social work, three in social policy, and three in labor economics and industrial relations. This expert feedback was used to evaluate the research questions based on their relevance, significance, and methodological alignment. Initially, a pool of 15 questions was reviewed by experts, resulting in the removal of five questions that fell below a 50% approval threshold. This process led to the creation of a refined 10-question semi-structured interview form.

Ethical approval for the research was granted by the Kocaeli University Institute of Social Sciences Ethics Committee under permission number 155728, dated 2021. Additionally, the study received support from the TÜBİTAK 1002 program.

Population and Sample of the Research

The primary objective of qualitative research is not to generalize findings across a specific population, but rather to achieve a comprehensive understanding of a phenomenon by capturing its diversity, richness, contradictions, and variations (Goetz and LeCompte, 1984). Qualitative research is inherently flexible, and this flexibility is evident throughout all stages of the research process. This characteristic allows for the simultaneous use of multiple sampling methods when making sampling decisions (Yıldırım and Şimşek, 2011). In this study, various purposive sampling techniques were employed to leverage the benefits of this flexibility and to enhance the depth of the research findings.

This research aims to conduct an in-depth examination of the phenomena of unemployment, poverty, and welfare, which are crucial aspects of social policy practices. To achieve this, it is essential that the study should not only cover a broad sample but also include participants who are well-versed and knowledgeable about these issues, thereby allowing for a thorough investigation (Islamoğlu and Alnıaçık, 2016). The study encompasses all 81 provinces within Türkiye’s administrative structure. The sample, intended to represent this entire population, includes 243 NGOs, with three organizations selected from each of the 81 provinces, within the framework of the NUTS 2 level (26 sub-regions) as per the scope and limitations of the research.

Evaluation of Research Findings

In the initial stage of coding the research data, the notes taken during face-to-face interviews were transcribed into Microsoft Word and underwent a preliminary evaluation. Following this evaluation, the data were imported into the Maxqda qualitative data analysis software according to the framework established by the researchers. Variables were assigned within the Maxqda program, and the coding process commenced. Upon the completion of the coding, the results were analyzed and interpreted within the defined scope of the study.

In this study, the data were analyzed across all 81 provinces within their respective sub-regions. During the evaluation phase, NGOs were coded with the letter ‘S’, and those interviewed in each province were identified as ‘S1’, ‘S2’, ‘S3’, in the coding system. The analysis was conducted by using the criteria established by Esping-Andersen’s Three Worlds of Welfare Capitalism. The criteria for Esping-Andersen’s welfare regimes’ classification are as follows:

• Liberal Welfare Regime: In this model, the market is the primary provider of welfare, with the State playing a compensatory role in addressing social issues. The State intervenes only as a last resort and provides need-based assistance financed through taxation. Access to aid is often bureaucratically intensive and procedurally complex.

• Conservative Welfare Regime: This model emphasizes the role of the State and prioritizes the family unit in the welfare system, while rejecting the market’s primary role. Social assistance is rooted in solidarity, and the preservation of traditional family structures is encouraged. The State complements the family institution when it falls short.

• Social Democratic Welfare Regime: This model features universal welfare policies with a focus on achieving full employment. The State is central to welfare provision, delivering social assistance through tax revenues. It promotes high social solidarity, egalitarianism, and comprehensive support services, particularly for disadvantaged groups.

Table 1.
Welfare regime of TR1 Istanbul Subregion

Theme	Participants (TR10 İstanbul Sub-region)	f
Theme	TR100 İstanbul	f
Liberal Welfare Regime	S1, S2, S3, S4, S5, S7, S9	7
Social Democratic Welfare Regime	S1, S2, S3, S5, S6, S9, S10	7
Conservative Welfare Regime		0
Total		14