Assessing Regional Variation in Support for the Radical Right-Wing Party ‘Alternative for Germany’ (AfD)—A Novel Application of Institutional Anomie Theory across German Districts

Abstract

The paper at hand aims to address a research gap by examining the spatial impact of economic and non-economic institutions on regional variation in the support for the German populist far-right party ‘Alternative for Germany (AfD)’. Adopting an interdisciplinary approach and drawing on institutional anomie theory, the study explores the relationship between economic dominance—economic inequality and economic strength—and regional AfD vote share, as well as the role of non-economic institutions in shaping support for the party. Using various regional indicators, the study assesses the strength of different institutions and their impact on the AfD’s electoral results in the year 2021 across 401 German districts (NUTS 3-level). The results point to significant differences between eastern and western Germany in the analysis of regional populist voting patterns. Economic dominance emerges as a stronger predictor of the AfD’s vote share in eastern regions. However, the findings on the protective effect of strong non-economic institutions against AfD voting are ambiguous and partly contrary to theoretical expectations. Overall, the study’s findings show that the newly introduced indicators deduced from institutional anomie theory contribute to the existing literature to help better understand regional variances in AfD support. Furthermore, the study contributes to the existing literature on the rise of contemporary radical right movements by employing Karl Polanyi’s framework of ‘disembedded’ markets. In doing so, it sheds light on the complex interplay between social change, anomic tendencies, and the growth of radical right movements in capitalist societies.

1. Introduction

We are now ‘riding the fourth wave’ of right-wing politics in post-war western Europe, which Mudde (2020, p. 297) characterized as the ‘mainstreaming and normalization’ of far-right actors and ideas. After a period of isolation and insignificance of small neo-fascist groups after World War II, political parties described as ‘extremist’, ‘radical’, ‘populist’, or ‘new right’ have been gaining electoral success across Europe since the 1980s (Arzheimer 2008; Von Beyme 1988). Following Mudde (2007, p. 2), these parties, despite their differences belong to the same ‘populist radical right party family’, sharing the core ideological features of nativism, authoritarianism, and populism (Mudde 2007, p. 22).

The fourth wave of right-wing politics began at the turn of the century, following the terrorist attacks of 9/11, and gained renewed momentum due to the so-called refugee crisis in 2015. Such events brought the favored topics of the political right—immigration and security—to the forefront of political debate, mobilized right-wing voters, and contributed to a wider ideological shift towards the extreme right within the general political discourse. We now see that far-right politicians can win presidential elections and that coalitions between far-right and moderate parties are being formed (Mudde 2016, 2020).

While, for a long time, Germany was an exception in Europe, with only marginally successful (local) right-wing parties, it has finally joined the fourth wave of far-right politics (Berbuir et al. 2015). The turning point that marked this shift was the 2017 federal election, when the ‘Alternative for Germany’ (AfD) party was elected, with 13 percent of the vote, as the largest opposition party (Eiermann 2017). Starting as a ‘softly’ Eurosceptic party, the AfD radicalized quickly in 2015. It is now generally accepted that the AfD fits Mudde’s (2007) concept of a populist radical right party (Arzheimer and Berning 2019). Only a few years after its formation in 2013, the AfD became the first radical right-wing party in German parliament since the end of World War II. Having now been in existence for over 10 years, contrary to initial forecasts, the party is politically established, with 78 members in the German federal parliament and representation in 15 of the 16 German state parliaments.

In the last federal election in 2021, the AfD lost around two percent of support nationally but received more than twenty percent of the vote in some parts of Germany and emerged as the strongest party in Saxony. The recent election results reflect a stable trend in which the AfD receives more support in eastern Germany, the states of the former German Democratic Republic, than in the west. Notably, in the last election, the AfD’s vote share in eastern Germany was twice as high as in western Germany (Weisskircher 2022a). However, this trend is not uniform for either ‘the east’ or ‘the west.’ There are strong regional differences, such as in Gelsenkirchen, where the AfD received 13% of the vote, while in the federal state of North Rhine-Westphalia, the party barely managed to cross the 5% threshold.

These patterns of regional variation in right-wing party success are not only apparent in Germany. A growing body of research has demonstrated the existence of geographic heterogeneity within countries in the support for populist movements, e.g., in Britain’s Brexit referendum or Donald Trump’s election in the US (Autor et al. 2016; Essletzbichler et al. 2018; Los et al. 2017; Rodríguez-Pose 2018). Therefore, it is important to look at smaller subnational units, such as neighborhoods (Bowyer 2008) and electoral districts (Rydgren 2018), to gain a more nuanced understanding of the underlying spatial drivers of populism. National-level analysis may oversimplify and mask this intra-country variance (Essletzbichler et al. 2018).

Publications studying the rise of populism in Germany have explored various factors, from individual determinants such as anti-immigrant or anti-establishment sentiments (e.g., Goerres et al. 2018; Hansen and Olsen 2019) and the east–west (Weisskircher 2020) or rural–urban divide (Förtner et al. 2021) to historical contexts, such as the spatial proximity to concentration camps (Hoerner et al. 2019; Homola et al. 2020). However, most (quantitative) studies have focused on concepts like relative deprivation or economic inequality and explored mainly single contextual factors, such as the share of immigrants, unemployment rate, and their interaction with immigration levels (Arzheimer 2018; Dülmer and Ohr 2008; Eiermann 2017; Kurtenbach 2019; Pokorny 2012).

The objective of the present study is to fill a research gap by investigating spatial imprints of basic economic and non-economic institutions and their impact on regional variations on right-wing electoral outcomes at the district level in Germany. This aspect has received limited attention thus far, with the existing literature on institutional characteristics primarily focused on the electoral system (e.g., Arzheimer and Carter 2006), decentralization, and the decline in welfare protection (Swank and Betz 2003).

To bridge this gap, we adopted an interdisciplinary approach by incorporating institutional anomie theory (IAT) into our study. Originally developed to explain crime, we applied IAT to investigate whether geographic variations in support for AfD within Germany reflect variations of areas with anomic tendencies in their institutional structure. These anomic tendencies, according to the main argument of IAT, are an inherent feature of modern capitalist societies, stemming from the dominance of economic interests over other sectors of society and the weakening of non-economic institutions, such as education, polity, religion, and family. According to the tenets of the theory, on one hand, we expected economic dominance—operationalized as high economic inequality combined with economic strength—to be positively associated with AfD vote shares. On the other hand, we expected regions with enfeebled non-economic institutions to pave the way for higher AfD vote shares.

To test our hypothesis, we examined the structural indicators of spatial and urban development and their impact on AfD’s 2021 election results across 401 German administrative districts (NUTS3 level).

2. Institutional Anomie Theory and Its Theoretical Origins

In order to trace the theoretical roots of this study, it is necessary to outline institutional anomie theory in terms of its origins, which were shaped by Karl Polanyi’s ([1944] 2001) idea of disembedded markets and Émile Durkheim’s ([1897] 1987; [1893] 1992) idea of solidarity and social integration. While those theoretical perspectives vary in scope and historical setting, at their core, they share the same notion of the ‘pathologic relation between market and society’ (Holmes 2018, p. 13).

In his renowned work ‘The Great Transformation’, Polanyi ([1944] 2001) described the change in social organization from pre-industrial societies to industrialized capitalist societies took place in the first half of the 20th century. This shift was driven by the ideology of laissez-faire free markets, with the economy becoming more and more ‘disembedded’ from society and independent from non-economic social relationships and institutional control.

Émile Durkheim ([1897] 1987; [1893] 1992) investigated the social consequences that result from such rapid social change brought about by extreme industrial growth and introduced his concept of anomie. He argued that a disembedded economy causes social disruption because the society is no longer able to normatively regulate social behavior, leading to anomie—a chronic state of normative deregulation (Durkheim [1897] 1987).

It is a specific setting in which Karl Polanyi’s work is rediscovered, tracing the development of financialized capitalism since the 1990s, the deepening of neoliberal reforms, the 2008 financial crisis, and the austerity policies that followed (Atzmüller and Décieux 2020). Messner and Rosenfeld (2000, 2001), along with many others, including a report by the German government (WBGU 2011), have revisited Karl Polanyi’s ideas, claiming a “new Polanyian ‘movement’ of market expansion” (Atzmüller and Décieux 2020, p. 2) in contemporary Western societies (for an overview, see Aulenbacher et al. 2019; relevant to the present study, e.g., Atzmüller and Décieux 2020; Dörre 2019; Fraser 2011, 2017; Lim 2021; Sandbrook 2018).

Following Teeple (1995); Messner (2018); and Levchak (2015), we refer to the current changes in market economies as the precursors of a ‘Great Retrogression’, characterized by a persistent rollback of market regulation, unleashing of financial institutions, and dismantling of social welfare programs.1

By building bridges between Polanyi’s thoughts on disembedded markets and Durkheim’s thoughts on anomie and social organizations, Messner and Rosenfeld elaborated their concept of institutional anomie as a consequence of societal changes associated with the ‘Great Retrogression’.

2.1. Disembedded Economy and Institutional Anomie

When scholars in the field of critical political economy (Aulenbacher et al. 2019; Wigger 2022) refer to Polanyi to gain insight into how contemporary radical right movements emerge in the context of free market capitalism, they often turn to his concept of a ‘double movement’ (Dörre 2019; Fraser 2011, 2017; Holmes 2018; Lim 2021). Polanyi’s (Polanyi [1944] 2001, p. 138) double-movement refers to the dialectical counterbalance between the two forces that have continually shaped the history of capitalist development: economic liberalism, which attempts to extend free markets and free trade, and social protectionism, which aims to safeguard people and the society from the destabilizing forces of the former.

However, as Lim (2021) argues, there are two divergent interpretations when scholars invoke Polanyi’s double-movement concept.

On the one hand, scholars interpret radical right politics as a counter-movement for social protection against the negative effects of free market liberalism. They assert that neoliberal politics and social welfare reforms have worsened economic and social conditions, leading to a sense of insecurity among certain members of society. This has created a political space for the radical right to offer its own vision of ‘social protection’ (Eichengreen 2020; Fraser 2016; Hopkin 2017).

On the other hand, scholars argue that, despite their protective rhetoric, radical right politics are part of the self-regulating market movement, which supports disciplinary attitudes towards welfare, particularly targeting groups deemed as ‘undeserving’ (Chueri 2022), while, at the same time, prioritizing capital accumulation over protecting human well-being (Atzmüller and Décieux 2020; Lim 2021, p. 528; Scheiring and Szombati 2020).2

While authors disagree on the importance of the role of economics in the agenda of populist parties, there is now a common understanding that contemporary populist radical right-wing parties have shifted their market-liberal stance. Instead of advocating for a minimal state, these parties now tend to support a strong but exclusionary welfare state (Chueri 2022) and embrace economic nationalism (Clift and Woll 2013), departing from global market liberalism (Betz 2003, p. 76; Mudde 2007). However, empirical evidence from countries such as Hungary (Scheiring and Szombati 2020), Austria (Ausserladscheider 2022), Turkey (Güven 2016), Britain (Hopkin 2017), and the US (Pierson 2017) suggests that when right-wing parties come to power, they tend to implement policies that benefit the wealthy and powerful rather than implementing policies that promote social and economic equality (Lim 2021).

Alongside these different interpretations of the ‘double-movement character’ of right-wing politics, numerous scholars have offered insights into how the political right has been able to benefit from what Polanyi called ‘disembedded market economies’ (Atzmüller and Décieux 2020; Fraser 2016, 2017; Kalb 2018; Scheiring 2018; Scheiring and Szombati 2020).

In his work ‘The Great Transformation’ Polanyi ([1944] 2001) posits that the shift in social organization from pre-industrial societies to industrialized capitalist societies in the first half of the 20th century was characterized by a process of ‘disembedding’ markets from society, social relationships, and institutional control. A key aspect is ‘the conceptualization of labor, land and money as commodities’ (Buğra and Ağartan 2007, p. 3).

In revisiting Polanyi’s ideas, in the last five decades, researchers have observed parallel changes to what Polanyi described for the 19th and 20th century (Aulenbacher et al. 2019). After long periods of ‘re-embedding’ and decommodification (Esping-Andersen 1990), the recent movement of market expansion and the financialization of capitalism has led to the ‘fragmentation and polarization of welfare systems and social policies’ (Atzmüller and Décieux 2020, p. 7).

We argue that Messner and Rosenfeld’s ([1994] 2007) institutional anomie theory provides an analytical toolkit to empirically measure Polanyi’s concept of disembedded markets in the institutional setting and how this reflects current social change and anomic tendencies (Bernburg 2002).

The underlying premise is that, in order for a society to function, it is essential that there is a balance of power between its major institutions—economy, polity, family, education, and religion (Parsons 1951). According to IAT, in the United States in particular but also in other modern capitalist societies, this balance is skewed because the economy dominates other non-economic institutions and thus weakens their power to socially control the members of their society (Messner and Rosenfeld [1994] 2007, p. 76 ff.). A lack of non-economic institutional control and institutional trust, in turn, leads to an increase in crime and deviant behavior, according to the theory. The dominance of the economy manifests itself in three ways: first, the devaluation of non-economic institutional functions and roles; second, the accommodation of non-economic institutions to economic requirements; and third, the penetration of economic norms into other institutional domains (Messner and Rosenfeld [1994] 2007, pp. 76–78).

Such a decoupling of markets from social embeddedness leads to the prioritization of economic interests, resulting in a disruption of social order and a weakening of non-economic institutions. Following IAT, this fosters a state of institutional anomie, which, drawing on Polanyi, is associated not only with criminal outcomes but also with electoral support for right-wing parties, as demonstrated in this paper.

2.2. Shift Back to Durkheim’s Mechanical Solidarity in Terms of ‘Regressive Collectivism’

We suggest that this is not only evident in structural processes but also (in line with the conceptual logic of IAT) in its relation between structure and culture. Modern societies are experiencing a cultural shift from unlimited individualism to a regressive form of collectivism. This shift is seen as a reaction to the disintegrative effects of individualism and a return to mechanical forms of solidarity (Dörre 2019).

As posited by Durkheim, the transition from pre-industrial to modern, functionally differentiated societies marks a shift from mechanical to organic solidarity. In pre-industrial societies, social integration was based on mechanical solidarity, a group-collective conscience based on similarities; however, this idea has morphed in modern societies into an organic solidarity, a solidarity based on diversity and differences. In order to detach individuals from their collective groups and integrate them into a complex society structured by the division of labor, traditional collectivism has been replaced by individualism (Thijssen 2012).

However, Durkheim ([1897] 1987, p. 86) argued that unregulated individualism can become pathologic in modern capitalist societies, leading to excessive self-interest and a lack of social integration, which can ultimately culminate in ‘egoistic suicide’. Messner et al. (2008, p. 172) described this pathologic individualism as ‘disintegrative individualism’, which they see as the cultural foundation of economic dominance in the institutional balance of a society and widespread anomie.

We argue that, in order to link IAT to right-wing populism, the cultural foundation of disintegrative individualism must be supplemented by the concept of ‘regressive collectivism’, which was proposed by Helmut Thome (2007, 2022)3. This concept defines a cultural rollback in modern societies, reinforcing collective elements of social integration as a consequence of ‘the lack of regulation or expanding disintegrative individualism’ (Thome 2007, p. 196). This concept encapsulates a desire for a return to collective forms of consciousness based on Durkheim’s mechanical solidarity (Thijssen 2012).

The concept of state nationality has transferred the traditional collectivism of primitive societies into modern societies, resulting in a strong form of exclusion of anyone deemed an ‘Other’, making it easier for right-wing extremism, nationalism and xenophobia to take hold (Thijssen 2012; Thome 2007, p. 196).

Although they are distinct cultural forces, disintegrative individualism and regressive collectivism are not mutually exclusive. They both coexist in modern societies and contribute to the rise of right-wing extremism, nationalism, and xenophobia in the way that they share the same prioritization of particularistic interests over universalist principles, neglecting the universal equality of all individuals and its institutional incorporation (Thome 2007, p. 49).4

It is beyond the scope of the present study to empirically test for cultural aspects. Nevertheless, considering forms of regressive collectivism is an important theoretical element to link IAT and right-wing populism. We assume that they both serve as a cultural foundation of current radical right movements, which differ from 20th century fascism in that they do not seek to establish an authoritarian dictatorship, but instead use authoritarian methods to promote free market ideology (see also libertarian authoritarianism by Amlinger and Nachtwey 2022). Capitalist and right-wing ideologies ‘resemble each other in the sense that both sought to deprive people of their personality as human persons along with their rights to their livelihood and push them into the survival of the fittest’ (Atzmüller and Décieux 2020; Lim 2021, p. 5).

3. The Present Study

Turning to the empirical scope of the present study, we focus on the ‘Alternative for Germany’ party for two reasons: a practical reason due to specific data sources and a theoretical reason due to the AfD’s specific form of ordoliberal5 competitive populism.

The AfD’s economic agenda is characterized by an authoritarian turn in neo-resp. ordoliberalism, which emphasizes the role of a strong state in preserving competition principles and establishing a hierarchical order based on economic performance (Bebnowski 2015, 2016; Bebnowski and Förster 2014).

However, over the course of the last ten years, the party AfD has undergone a shift in its programmatic foundations. It started as an anti-Euro party, strongly opposing Angela Merkel’s support for the euro zone bailouts and promoting a national liberal market radicalism (Bebnowski 2016). This has changed with the rise of issues related to migration and the refugee crisis in 2015. Since then, the ‘völkisch-national’ wing, led by Björn Höcke and politically secured by Alexander Gauland, has become more and more power within the AfD (Joswig 2022; Mullis and Zschocke 2019).

The party is now experiencing difficulties in determining its political stance, balancing between its original market-liberal positions and the radical right positions of the völkisch-national wing (Eberhardt et al. 2019). In alignment with the former, some AfD politicians still continue to pursue market-liberal policies, e.g., rejecting the minimum wage and inheritance tax (Nocun 2016; Orde 2018; Salzborn 2016). Höcke, as the dominant figure of the völkisch-national course, is currently pushing AfD’s social policies, e.g., promoting and advocating for an expansive reform of the pension system that would be partly financed by the state but only apply to those who are German (Orde 2018).

An important part of their foreign and domestic policy is to ‘ethnisize’ and ‘culturalize’ economic performance; for example, the inability of countries like Greece or Italy to manage the European Debt Crisis between 2009 and 2010 is ‘reflecting differences in mentality’, as Henkel (2012), a former AfD member of the European Parliament, argued (Bebnowski and Förster 2014, p. 9). They justify the exclusion of nations and individuals due to their allegedly low economic contributions (see racial neoliberalism by Goldberg 2011).

In multiple studies, Groß, Hövermann, and colleagues explored this specific form of discrimination, which devalues and excludes people due to their assumed economic ‘unprofitability.’ This widespread practice is prevalent not only in the AfD party but also in society at large, as demonstrated by their research findings (Hövermann et al. 2015b, 2015c). They found that this form of hostility, which is economically neoliberal in nature, is particularly prevalent among people who are strongly oriented towards economic ideals and values. These economic values reflect a broader trend of economization in society, which has been driven by a shift towards neoliberal government politics since the mid-1980s (Bourdieu 1998; Currie 1997; Neckel 2005; Sandel 2012).

Based on a sample of German respondents, the authors showed that these market-radical values are not only positively associated with prejudices but also with right-wing extremism and support for the AfD (Groß and Hövermann 2014, 2018; Hövermann et al. 2015a). Their research is both implicitly and explicitly linked to IAT. The predominance of economized cultural values is understood as the individual manifestation of anomic culture, what Messner and Rosenfeld ([1994] 2007) conceptualized as the ‘American Dream’ ethos and what Messner et al. (2008) described as disintegrative individualism.

Studies that explicitly refer to IAT to explain the devaluation of unprofitable groups have highlighted that it is important to consider not only marketized cultural values but also the protective function of non-economic institutions. Hövermann et al. (2015b, 2015c) showed that individuals who do not integrate well into non-economic institutions, such as family and friendship networks, religion, and politics, or those who perceive these institutions as enfeebled are more prone to market-based values and prejudices.

Subsequently, Hövermann and Messner (2019) and Nickel (2022) expanded upon this approach by incorporating a multi-level framework to examine IAT’s impact on anti-immigrant attitudes in a cross-country comparison. In addition to individual measurements, by utilizing a variety of macro-level indicators to assess the strength of non-economic institutions on a national level, both studies emphasized the importance of non-economic institutions as a counterbalance to economic dominance.

The studies cited above indicate that the tenets of IAT can be meaningfully applied to the devaluation of vulnerable groups and to anti-immigrant attitudes, which form a crucial part of right-wing extremist attitudes. Building on these prior findings, our research aims to determine whether geographic variations in AfD support can be explained by the presence of anomic tendencies as depicted by IAT within a region.

In line with IAT, we posit that economic dominance leads to an imbalance of power in the institutional structure within a specific region, thereby weakening regional non-economic institutions and reducing their ability to provide stability and social integration in those social spaces. On one hand, we expect economic dominance—high economic inequality combined with economic strength—to be positively associated with AfD vote shares. On the other hand, we hypothesize that the strength of non-economic institutions, as a counterbalance to economic dominance, is negatively associated with the regional distribution of AfD vote share in German districts, similar to its ‘protective’ impact on anti-immigrant and criminal outcomes.6 Moreover, we assume an interaction effect between economic dominance and the strength of non-economic institutions, meaning that the strength of non-economic institutions within a region lessens the effect of economic dominance on regional AfD vote share.

Hypothesis 1 (H1).

Regional Economic dominance (high economic inequality combined with economic strength) is positively associated with regional AfD vote share in the 2021 German federal election.

Hypothesis 2 (H2).

The regional strength of non-economic social institutions, such as education, policy, family, and health, is negatively associated with regional AfD vote share in the 2021 German federal election.7

Hypothesis 3 (H3).

The regional strength of non-economic social institutions lessens the effect of economic dominance on regional AfD vote share in the 2021 German federal election.

4. Data, Methods, and Measures

4.1. Data and Methods

To test our hypotheses, we used multiple structural indicators for spatial and urban development (INKAR—’Indikatoren und Karten zur Raum- und Stadtentwicklung’), provided by the ‘Federal Institute for Research on Building, Urban Affairs, and Spatial Development’ (Bundesinstitut für Bau-, Stadt- und Raumforschung (BBSR)). Our analysis is based on the latest available data, which has been published in February 2023 and covers a time period from 2019 to 2021.8 Our sample consists of 401 German districts classified at the NUTS 3-level (‘Kreise und kreisfreie Städte’), including 77 districts in eastern Germany and 324 in western Germany.

We conducted Ordinary Least Squares (OLS) regression models to estimate the impact of economic dominance, non-economic institutions, and their interaction and effect on the vote share for the Alternative for Germany (AfD) across these German districts. To test for multicollinearity between the independent variables, we used variance inflation factors (VIFs); VIFs greater than ten were considered as problematic (Levchak 2015, p. 395). As the Breusch–Pagan/Cook–Weisberg test detected heteroscedasticity, we calculated the standard errors using a heteroskedasticity-consistent covariance matrix (HC3), following Hughes et al. (2015). All statistical analyses were performed using the statistical program Stata (Version 14.2).

4.2. Measures

4.2.1. AfD Vote Share in 2021 Federal Election

We introduced a new research approach by applying IAT to a novel outcome variable: the regional vote share of the AfD in the 2021 federal election. Our initial analysis identified potential heteroskedasticity issues, which were addressed by using the logarithmic transformation of the AfD party’s vote share. Given the normal distribution, we utilized ordinary least squares (OLS) regression to examine the relationship between economic dominance, non-economic institutions, and AfD vote share across German districts (for descriptive statistics and further information of all measures used in the analysis, see

Table A1. Descriptive statistics and description.

	Description	M	SD	Min	Max
Dependent Variable
AfD voter share in the 2021 federal election	Percentage of second votes for the AfD of all second votes.	11.393	5.944	2.87	32.53
Economic inequality
Old-age poverty	Percentage of the population receiving basic security in old age out of the population aged 65 and older.	2.307	1.602	0.41	12.46
Long-Term unemployment	Percentage of unemployed people who have been unemployed for 1 year or more among the total unemployed population.	27.554	7.867	10.69	49.84
Housing benefits recipients	Average monthly amount of housing allowance per recipient household in euros.	14.329	5.401	1.2	34.95
Economic Strength
Volume of work	Number of work hours performed by employees.	1278.96	42.142	1174.69	1384.6
Household income	Average household income per capita in euros.	1960.43	231.449	1420.67	3514.12
Non-Economic Institutions
Polity
Voter turnout	Proportion of second votes (valid and invalid) as a percentage of the electorate in the 2021 German Federal Election.	76.243	4.16	63.4	85.5
Family
Fertility rate	Total Fertility Rate.	1.614	0.149	1.11	2.15
Number of households with children	Proportion of households with children to all households.	28.988	4.234	18.73	42.34
Single-person households	Proportion of one-person households to all households.	38.394	7.587	21.92	64.81
Education
Number of students enrolled in University	Number of students in universities per 1000 inhabitants.	29.076	53.52	0	385.1
Number of students enrolled in ‘Fachhochschule	Number of students in ‘Fachhochschule’ per ‘1000 inhabitants’.	11.4195	22.04124	0	233.42
School leavers with general university entrance qualification	Proportion of school leavers with qualifications enabling them to enter university per all school leavers in %.	29.6281	12.065	0	60.5
Health
life expectancy	Average life expectancy of a newborn in years.	81.122	0.999	78.13	83.98
premature mortality of males	Deaths of men under the age of 70 per 1000 men/women under the age of 70.	76.711	38.943	1	159
premature mortality of females		109.157	61.883	1	234
Control variables
eastern–western Germany		0.192	0.394	0	1
Share of refugees	Percentage of refugees in the population.	4.382	2.555	0	36.37
Outmigration	Outmigration per 1000 inhabitants.	44.045	15.764	20.17	165.76
Average age	Average age of the population in years.	44.804	2.021	40.19	50.71
Population size	Number of inhabitants who have a permanent residence (main residence), including foreigners registered as living there for a longer period of time.	207,369	245,008	34,001	3,664,088

Notes: N = 401, not standardized.

in Appendix A).

4.2.2. Economic Dominance

The key component to measuring institutional anomie theory is the operationalization of economic dominance in the institutional balance of power in relation to the strength and/or enfeeblement of other non-economic institutions. To measure economic dominance, we relied on studies that considered indicators of economic inequality, e.g., percentage of families below the poverty line (Chamlin and Cochran 2007; Piquero and Piquero 1998), GINI-Index (Kim and Pridemore 2005; Maume and Lee 2003), or unemployment rates (Schoepfer and Piquero 2006), as well as indicators of economic freedom or strength, e.g., economic freedom index (Hughes et al. 2015) or gross domestic product (Bjerregaard and Cochran 2008a, 2008b; Cochran and Bjerregaard 2012).

In order to assess economic dominance, we constructed a product term by multiplying the values of high economic inequality and economic strength for each district. This product term represents the joint effect of high economic inequality and economic strength. Consistent with prior research, we suggested that economic dominance captures the degree to which the state fails to protect individuals from market dynamics (Batton and Jensen 2002; Esping-Andersen 1990; Hannon and Defronzo 1998; Rosenfeld and Messner 1997). Thus, when both economic inequality and economic strength are high, it indicates that the political and social protective measures in place have not been effective in mitigating the negative consequences of a free market economy (Levchak 2015).

Economic inequality is measured as the group mean of three standardized variables: the level of poverty among individuals aged over 65 or those who are permanently incapacitated and incapable of work, the number of individuals considered long-term unemployed (unemployed for one year or more), and the number of individuals receiving welfare benefits. Regional economic inequality is higher when there is a greater demand for social welfare support.

Economic strength is measured via the group mean of two variables: the total amount of working hours within a region and the average regional household income. Economic strength is higher when both the employment rate and household incomes are higher.

Economic dominance is calculated as the product term of economic inequality and economic strength, with higher values indicating higher economic dominance.

Economic Dominance = Economic Inequality × Economic Strength

4.2.3. Non-Economic Institutions

Following Messner and Rosenfeld ([1994] 2007), we examined the following institutions: polity, family, education, and health.

We measured non-economic institutions as factor scores using the principal components method with varimax rotation. The selection of indicators was based on theoretical considerations and verified through reliability and correlation analyses. The use of factor scores has advantages in terms of simplifying interpretation and avoiding multicollinearity issues (Bjerregaard and Cochran 2008a; Hughes et al. 2015; Messner and Rosenfeld 1997). Prior to factor analysis, the variables were standardized and transformed to a positive value range.

• Polity

According to Messner and Rosenfeld ([1994] 2007), the political system functions as a social institution that is designed to achieve collective goals, unless it is undermined and weakened by the economy. In line with the existing literature, we argued that the effectiveness of the political system can be measured via a region’s voter turnout, whereas a low voter turnout indicates an enfeebled political system (Bjerregaard and Cochran 2008a, 2008b; Kim and Pridemore 2005). Thus, we operationalized the strength of the political system as the ratio of the turnout in the 2021 federal election to the total number of eligible voters expressed as percentages. Considering the complexity and multi-dimensional nature of the concept, relying on a single indicator to capture political strength is not ideal. However, within the constraints of data availability, the higher voter turnout can be considered a valid measurement for stronger trust and engagement in the democratic process within political institutions.

• Family

Even though Messner and Rosenfeld ([1994] 2007) proposed various indicators to measure the level of economic dominance as it pertains to families, the majority of studies have relied on a single indicator: the divorce rate (Bjerregaard and Cochran 2008b; Chamlin and Cochran 2007; Dolliver 2015; Maume and Lee 2003; Piquero and Piquero 1998). Only a few studies have utilized alternative indicators, such as single-parent households (Hannon and Defronzo 1998; Piquero and Piquero 1998) or women in the workforce (Bjerregaard and Cochran 2008a).

The items that best measure the strength of the family institution were chosen based on their correlation and reliability. We calculated a factor score (alpha = 80.89) to indicate the strength of the family institution, which included the fertility rate (factor loading = 0.73), the number of households with children (0.88), and the number of single-person households (reverse coded, factor loading = 0.94). It is important to note that the indicators used to operationalize the strength of the family institution not optimal. The only available structural data for this study focus on measures related to traditional family structures, neglecting other social relationships that reflect the diversity of family dynamics in contemporary societies.

• Education

Messner and Rosenfeld ([1994] 2007) emphasize the role of the educational system in relation to how it informs socialization and vocational preparation among young people. Various indicators have been used in previous studies to measure the strength of education, including the amount of individuals enrolled in college (Kim and Pridemore 2005; Piquero and Piquero 1998), the amount of individuals enrolled in primary education (Dolliver 2015), and public expenditure on education as a percentage of GDP (Bjerregaard and Cochran 2008b; Hirtenlehner et al. 2013; Hövermann et al. 2016). Based on the available regional indicators, we measured the strength of education (alpha = 0.66) by considering the number of students enrolled in universities (0.81), the number of students enrolled in ‘Fachhochschulen’ (0.89), and the percentage of school leavers with the qualifications needed to enter a university (0.62). Higher strength of the education institution is indicated by a larger number of highly qualified individuals.

• Health

In this study, we also consider the strength of the health system, given its conceptual overlap with Messner and Rosenfeld’s decommodification index and government social support (Hughes et al. 2015, p. 107; Savolainen 2000). However, the health system has been mainly used as a control variable in previous studies (Bjerregaard and Cochran 2008a). We believe that considering health as a non-economic institution within IAT can provide insights into how public policies related to healthcare can affect the balance of power among institutions. To measure the strength of the health care system (alpha = 0.93), we utilized three indicators: life expectancy (factor loading = 0.93), the premature mortality of males, and the premature mortality of females (reverse coded; factor loading = 0.93/0.93). These indicators collectively provide insights into the effectiveness, accessibility, and commitment of the region’s health system in improving and maintaining the population’s health and well-being.

4.2.4. Control Variables

Lastly, we accounted for several indicators drawn from studies that investigated the regional characteristics of populist voting, including the region’s share of refugees, the differences between eastern and western Germany, the average age in the region, and the region’s population size (Dülmer and Klein 2005; Dülmer and Ohr 2008; Lubbers and Scheepers 2000).

5. Results

Table 1. OLS regression estimates of economic dominance, non-economic institutions, and their interaction and effect on predicting AfD vote share across German districts (N = 401).

	Model 1			Model 2			Model 3
	b (S.E.)	p	β	b (S.E.)	p	β	b (S.E.)	p	β
Economic Dominance	0.081 (0.015)	0.000	0.182	0.011 (0.012)	0.397	0.023
Non-Economic Institutions
Polity	−0.159 (0.021)	0.000	−0.353	−0.0834 (0.018)	0.000	−0.188
Family	0.01 (0.028)	0.001	0.225	0.078 (0.02)	0.000	0.177
Education	0.095 (0.031)	0.003	0.215	−0.043 (0.02)	0.034	−0.097
Health	0.055 (0.027)	0.037	0.124	−0.019 (0.02)	0.348	−0.043
east (1)–west (0)				0.771 (0.045)	0.000	0.684
Interaction Effects 1
Polity × Economic Dominance							−0.001 (0.00)	0.1	−0.078
Family × Economic Dominance							0.037 (0.22)	0.1	0.064
Education × Economic Dominance							−0.021 (0.015)	0.155	−0.044
Health × Economic Dominance							−0.014 (016)	0.357	−0.038
Constant	−16.6 (2.83)	0.000	-	1.18 (2.2)	0.593	-
R2	0.57			0.77
V.I.F.	<3			<3

¹ Separate models were computed for each interaction term, accounting for the main effects, all individuals, control variables, and the east–west distinction. See Table A4 (Appendix A) for details on all models. Note: All models included control variables.

,

Table A2. OLS regression estimates of economic dominance predicting AfD vote share across German districts (n = 401).

	Model 1			Model 2
	b (S.E.)	p	β	b (S.E.)	p	β
Economic Dominance	0.099 (0.016)	0.000	0.224	0.012 (0.014)	0.391	0.027
east (1)–west (0)				0.733 (0.05)	0.000	0.651
Constant	−14.17 (2.16)	0.00	-	2.47 (2.18)	0.257	-
R2	0.503			0.7213

Notes: All models included control variables.

,

Table A3. OLS regression estimates of economic dominance and non-economic institutions predicting AfD vote share in eastern (n = 77) and western (n = 324) regions.

	East			West
	Model 1			Model 2
	b (S.E.)	p	β	b (S.E.)	p	β
Economic Dominance	0.139 (0.059)	0.021	0.25	0.007 (0.013)	0.584	0.025
Non-Economic Social Institutions
Polity	0.148 (0.036)	0.000	0.479	−0.12 (0.02)	0.000	−0.369
Family	−0.021 (0.06)	0.71	0.061	0.081 (0.022)	0.000	0.296
Education	−0.055 (0.062)	0.387	−0.178	−0.0418 (0.021)	0.044	−0.153
Health	0.008 (0.0627)	0.897	0.021	0.02 (0.021)	0.327	0.068
Constant	−13.1 (3.77)	0.007		4.86 (2.82)	0.086
R2	0.711			0.434
V.I.F.	<3.5			<3

Notes: All models included control variables.

and

Table A4. OLS regression estimates of economic dominance, non-economic institutions, and their interaction predicting AfD vote share across German districts (n = 401); separate models for each interaction term.

	Model 1			Model 2			Model 3			Model 4
	b (S.E.)	p	β	b (S.E.)	p	β	b (S.E.)	p	β	b (S.E.)	p	β
Economic Dominance	0.035 (0.021)	0.096	0.078	0.013 (0.013)	0.291	0.03	0.012 (0.012)	0.312	0.028	0.011 (0.012)	0.373	0.024
Non-Economic Social Institutions
Polity	−0.084 (0.018)	0.000	−19	−0.088 (0.018)	0.000	−0.2	−0.084 (0.02)	0.000	−0.188	−0.079 (0.02)	0.000	−0.18
Family	0.079 (0.021)	0.000	0.18	0.06 (0.026)	0.02	0.133	0.08 (0.02)	0.000	0.177	−0.77 (0.02)	0.000	0.174
Education	−0.041 (0.02)	0.041	−0.092	−0.042 (0.020)	0.037	−0.095	−0.03 (0.024)	0.227	−0.0644	−0.043 (0.021)	0.039	−0.097
Health	−0.021 (0.019)	0.29	−0.048	0.019 (0.02)	0.34	−0.043	−0.021 (0.02)	0.308	−0.046	0.03 (0.023)	0.194	0.067
east (1)–west (0)	0.769 (0.045)	0.000	0.682	0.77 (0.05)	0.000	0.68	0.762 (0.045)	0.000	0.677	0.763 (0.05)	0.000	0.68
Interaction Effects
Polity × Economic Dominance	−0.001 (0.00)	0.01	−0.078
Family × Economic Dominance				0.037 (0.22)	0.1	0.064
Education × Economic Dominance							−0.021 (0.015)	0.155	−0.044
Health × Economic Dominance										−0.014 (016)	0.357	−0.038
Constant	−787 (2.2)	0.719	-	1.11 (2.2)	0.613	-	0.928 (2.18)	0.67	-	0.97 (2.2)	0.656	-
R2	0.766			0.766			0.766			0.766

Notes: In all models V.I.F. < 6.

(Appendix A) and present OLS regression models for the effects of economic dominance and the four non-economic social institutions on (logged) cross-regional AfD vote share in the 2021 German federal election.

Initially, we applied Table A2 Model 1, which only includes economic dominance as an independent variable, while controlling for cofounding variables, such as emigration rate, rate of asylum seekers, and population size (see Appendix A). The results demonstrate a significant positive association between economic dominance and the AfD party’s vote share, indicating that regions with higher economic dominance are more likely to vote for AfD (H1; b = 0.099, p = 0.000, β = 0.224).

In the second step, we applied Model 2, Table A2, which also controlled for whether the regions are located in the eastern or western part of Germany (see Appendix A). After adding the east–west distinction, the explained variance in regional AfD vote share increased from 50% to 70%, suggesting that the east–west variable accounted for an additional 20% of the variance. The east–west dummy (west = 0, east = 1) displays the highest regression coefficient and is highly significant (b = 0.733, p = 0.000, β = 0.65). The predicted vote share increases by 0.733 when moving from western to eastern Germany, holding all other variables constant. Moreover, when controlling for east–west, the regression coefficient of economic dominance decreases and is no longer significant (b = 0.012, p = 0.391, β = 0.027), suggesting that caution should be exercised when interpreting our findings regarding the differences between eastern and western Germany.

In Table 1 we incorporated non-economic institutions—polity, family, education, and health—to test whether strong non-economic social institutions act as a protecting factor against voting for AfD. We applied two separate models: Table 1 Model 1 (without controlling for east and west distinction) and Table 1 Model 2 (controlling for east and west distinction).

Table 1 Model 1, which does not account for the east–west distinction, exhibits a significant positive relationship between economic dominance and AfD vote share (b = 0.081, p = 0.000, β = 0.182). However, adding non-economic institutions only slightly increases the explained variance to 57% compared to Table A2 Model 1 (R² = 0.503). According to hypothesis 2, we assume that stronger social institutions lead to a decrease in AfD voting support. However, this is only the case for polity, wherein lower voter turnout corresponds to higher AfD vote share (b = −0.137, p = 0.000, β = −0.308). Contrary to our expectations, we found significant positive effects for the following non-economic institutions: family (b = 0.01, p = 0.001, β = 0.225), education (b = 0.095, p = 0.003, β = 0.215), and health (b = 0.055, p = 0.037, β = 0.124). This indicates that higher levels of strength in the family, education, and health institutions are positively associated with an increased regional share of AfD votes.

When accounting for east–west distinction in Table 1 Model 2, the explained variance increases to 77%. Also, the effect of non-economic dominance becomes insignificant (b = 0.011, p = 0.397, β = 0.023). Moreover, the negative impact of voter turnout on AfD vote share remains significantly negative (b = −0.0834, p = 0.000, β = −0.188). The effect of the family institution remains significantly positive (b = 0.078, p = 0.000, β = 0.177), suggesting that regions with more stable family structures are more likely to support the AfD party (controlling for east–west regional differentiation). In contrast, education is significantly negatively associated with AfD vote share (b = −0.043, p = 0.034, β = 0.097), suggesting that, when considering whether the regions are located in eastern or western Germany, the strength of the educational system has a protective effect on support for the AfD party. In this case, health is no longer significant (b = −0.019, p = 0.348, β = 0.043).

Subsequently, separate models were applied for the eastern and western regions of Germany to obtain a more detailed understanding of the findings, which may shed light on the unique and historically rooted disparities.

Two separate Ordinary Least Squares (OLS) regression models for regions in each area are presented in Table A3 Appendix A. Our analyses revealed that economic dominance is a significant predictor of support for the AfD party within eastern regions (b = 139, p = 0.021, β = 0.25). Controlling for other relevant factors, this model accounts for 71% of the regional variance in AfD vote share in the eastern regions of Germany. However, in western Germany, economic dominance is not significant, and the model can only explain 43% of the variance in AfD vote share (b = 007, p = 0.584, β = 0.025). Regarding the non-economic institutions, we identified three notable points. First, we observed a significant positive association between the polity institution, as measured by voter turnout, and the AfD party’s vote share in the eastern regions of Germany (b = 0.148, p = 0.000, β = 0.479). However, in Germany’s western regions, the effect was observed to be significantly negative (b = −0.12, p = 0.000, β = −0.369). These results suggest that, in the east, the AfD has more support in regions with higher voter turnout, whereas in the west, the AfD is more successful in regions with lower voter turnout. Second, while education is not a significant factor in eastern regions (b = −0.055, p = 0.387, β = −0.178), it is negatively associated with the AfD party’s vote share in the west (b = −0.041, p = 0.044, β = −0.153). Thus, the more robust a region’s education system is, the lower the vote share for the AfD party. Third, contrary to our expectations, we found a surprisingly positive and significant association between strong family institutions and AfD vote share in the west (b = 0.081, p = 0.000, β = 0.296) and a negative but insignificant one in the east (b = −0.021, p = 0.710, β = 0.061). Health was not observed to be significant for both the eastern (b = 0.008, p = 0.897, β = 0.021) and western (b = 0.02, p = 0.327, β = 0.068) districts.

Table 1 Model 3 includes interaction terms regarding economic dominance and each non-economic institutions to test whether the regional strength of non-economic social institutions lessens the effect of economic dominance on AfD support (H3). We expected the interaction effect to be negative, indicating that higher levels of non-economic institutional strength would reduce the effect of economic dominance on the AfD party’s regional vote share (further details and results in Table A4, Appendix A). When holding all independent and cofounding variables constant and taking the east–west distinction into account, we found one significant interaction term in the hypothesized direction for polity. The findings indicated that, in regions with a higher voter turnout, the positive effect of economic dominance on AfD voting is reduced (b = −0.001, p = 0.01, β = −0.078). However, the other product terms did not achieve statistical significance at a conventional level.

Furthermore, we also investigated which particular economic conditions predict AfD’s vote share

Table 2. Predicted rates of AfD vote share under various economic conditions.

Economic Inequality	Economic Strength	Predicted AfD Vote Share
Average	Average	0.301
High	High	0.315
High	Low	0.403
Low	High	0.199
Low	Low	0.288

Notes: Predicted rates were computed at ‘average’ (mean), ‘high’ (1.5 standard deviations above the mean), and ‘low’ (1.5 standard deviations below the mean) values of economic inequality and economic strength. All control variables, including the east–west distinction, were taken into account (following Bjerregaard and Cochran 2008b).

. The results revealed that regions with high economic inequality and low economic strength have the highest predicted rate, followed by regions with high economic inequality and high economic strength.

Contrary to our initial assumptions, these findings indicate that regions characterized by high economic inequality and a weak economic state are the most influential in predicting regional AfD vote share. We expected economic inequality to have an even greater impact on AfD vote share in economically strong regions, as it creates an environment where certain individuals or groups may feel marginalized despite overall economic prosperity.

However, our findings reveal that it is actually the unequal and economically weak regions that play a more significant role in predicting AfD vote share. This is in line with the existing literature, which showed how economic deprivation within a region predicts radical right voting (Bayerlein 2022; Greve et al. 2023). In alignment with Rodríguez-Pose’s (2018) concept of ‘left behind regions’, radical right populist parties perform particularly well in regions with a high level of inequality and those in long-term economic decline.

6. Conclusions

The purpose of the present study was to investigate the spatial imprints of basic economic and non-economic institutions and their impact on regional variance in right-wing electoral outcomes at the district level in Germany (NUTS 3-level).

Our empirical results suggest a positive correlation between economic dominance and AfD vote share, providing support for H1. However, our findings concerning regional differences in the effects investigated between eastern and western Germany should be interpreted carefully. Accounting for the east–west distinction, the regression coefficient of economic dominance decreases and becomes non-significant. Thus, the effect of economic dominance, as operationalized in the present study, on AfD vote is not stable across the eastern and western districts considered here. When separating the models for the eastern and western regions of Germany, our analyses indicate that economic dominance has a much stronger explanatory power in the east compared to the west, as well as for Germany overall.

Specific historical and political developments may partially explain these observed regional differences between the eastern and western areas of Germany (Weisskircher 2020). East Germany, as a former part of socialist Europe, underwent massive economic transformation in the 1990s and onwards, which has been described by Ther (2019) as the ‘neoliberal great transformation’. Integrating post-socialist Europe into the global capitalist economy without implementing corresponding institutional transitions such as strong state protection and anti-corruption and transparency measures resulted in ‘neoliberal disembedding’ (Scheiring and Szombati 2020). Scholars within the field of critical political economy emphasize the emergence of authoritarianism and right-wing populism in response to this massive change in the economic system driven by liberalization, deregulation, and privatization (so-called ‘shock therapy’) in Eastern Europe. They demonstrated how the political right has benefited from this process of ‘neoliberal disembedding’ in terms of an ‘authoritarian re-embedding’ (see also Hann 2019; Kalb 2018; Scheiring and Szombati 2020).

In the case of East Germany, the ‘disembedded’ transformation and economic dominance may have led to a political backlash that strengthens the AfD party. However, in western regions, where the process of neoliberal disembedding was not as rapid and accompanied by economic prosperity, there is no significant link between economic dominance and AfD vote share. This can also be considered in terms of culture and group identity. Following Durkheim ([1897] 1987; [1893] 1992), Thijssen (2012) highlights the desire for a return to collective consciousness, which right-wing actors successfully fulfill by emphasizing and strengthening collective group identities, particularly those based on nation. In Eastern Europe, the decline of socialist working-class consciousness has created fertile ground for right-wing actors to capitalize on the disintegration of traditional class-based identities and instead mobilize them around national, ethnic, religious, or cultural group identities. For Germany, Weisskircher (2022b, p. 97) describes an exclusionary eastern identity ‘which far-right players […] use for political mobilization, mainly to delegitimize governing political actors and institutions.’ This identity is based on a strong attachment to the ‘heartland’ (‘Heimat’) and a sense of being ‘second-class citizens’ who have become disadvantaged as a result of the transformation process.

Regarding the hypothesized protective effect of non-economic social institutions against voting for the AfD, our results are mixed. The only robust effect that is consistent with our hypotheses, even after controlling for the east–west-effect, is that of the polity institution: Higher voter turnout is associated with a lower AfD vote share (H2). We also found a significant interaction effect, indicating that the strength of the polity institution weakens the effect of economic dominance on the party’s success (H3). Thus, strong polity seems to have a rather robust counteractive effect on economic dominance. However, the effects of all other non-economic institutions go in the opposite direction, suggesting that the stronger non-economic social institutions are, the stronger AfD’s vote share is.

Notably, we observed significant regional differences between the eastern and western areas of Germany when investigating the voter turnout effects in both regions separately. Our results show that higher voter turnout in eastern regions is positively associated with increased AfD support, while in the western regions and in Germany overall, it is significantly negatively associated with AfD support. This seemingly unexpected finding in light of IAT may be explained by the distinctive (political) cultures in eastern and western Germany. In the eastern regions of Germany, the AfD receives significantly stronger support and has even become the dominant political party in some federal states, indicating the normalization of the right-wing party. Thus, it is plausible that individuals who sympathize with the AfD in the east may be more likely to vote for it, while those in the west may be more likely to vote for alternative parties due to social desirability concerns as the voting for the AfD is less normalized. However, in western regions with low voter turnout, there may be a more favorable political climate for the AfD.

Moreover, education only shows the expected negative effect in the west, where stronger education is associated with lower support for AfD (H2). It has no significant effect in the east. In western Germany, higher regional levels of education are associated with lower levels of support for the AfD, suggesting that education may act as a protective factor against AfD support in western but not eastern regions. Contrary to our expectations, strong family institutions are positively associated with AfD vote share in the west but are not significant in eastern regions. This suggests that there may be differences in the way family and educational structures and values influence political attitudes in different regional settings.

Overall, our empirical findings point to significant differences between the eastern and western parts of Germany in the analysis of populist voting behavior. Basic tenets of IAT do not seem to apply to both parts of Germany equally, indicating the need for more nuanced and differentiated theorizing when applying IAT to such sociospatial units, as we have in the present study.

Given this and the limitations of the study, our results must be interpreted with caution. While, theoretically, the study refers to broader developments, its empirical focus is on the intra-country variance within Germany. This may limit its generalization due to specifics such as the ordoliberal system or the rather late breakthrough of a right-wing populist party compared to other European countries. The data used in this study are limited to publicly available information at the district level, which may not fully capture the complexity of the underlying social and economic factors that influence voting behavior. The present study was conducted using cross-sectional data, meaning that it does not allow for causal inferences or the analysis of temporal changes. Therefore, the study cannot determine whether the observed relationships are stable over time or if they change in response to external factors. Moreover, the operationalization of economic dominance and the strength of the non-economic institutions examined in the present study was a first attempt at the level of German administrative counties (NUTS 3 level). Further empirical investigations are necessary to examine various approaches for operationalizing both economic dominance and the weakening or strengthening of contemporary non-economic institutions such as family. For example, using only fertility and divorce rates does not accurately reflect contemporary, modern perspectives on the family.

In conclusion, despite the limitations of our study and the partial support for IAT found in our study, we argue that institutional anomie theory nevertheless provides a valuable framework for empirically measuring Polanyi’s concept of ‘disembedded’ markets within the institutional setting of contemporary market economies. With regards to explaining intra-county variance in the support of right-wing populist parties, the interplay between and effects of different core institutions seems under-researched.

Overall, drawing inspiration from the work of Karl Polanyi, the present study contributes to the existing literature on the rise of contemporary radical right movements in the context of ‘disembedded’ market economies (Atzmüller and Décieux 2020; Fraser 2011, 2017). This framework enhances our understanding of how ongoing social changes and the emergence of anomic tendencies associated with unregulated market forces are intertwined with the rise of radical right movements in today’s capitalist societies.