What Shapes Satisfaction with Democracy? Interests, Morals, and the German East–West Divide

Thirty years after reunification, East and West Germany are still characterized by a considerable difference in satisfaction with democracy (SWD). This paper proposes and tests a model which assumes SWD to be shaped by the interests (economic and cultural) and moral values individuals demand to see fostered by the democratic system. Empirical application of the model reveals substantial differences between the East and West German SWD function in that the satisfaction of economic interests is much more important in the East than the West whereas the opposite applies to moral concerns. Demands on redistribution and immigration policies—conceptualized as proximate drivers of SWD—also shape SWD differently in the East and the West, in addition to being shaped by interests and morals in different ways. East–West differences in the relationship between economic demands and SWD are more important than differences in the levels of demand.


Introduction
The three decades since German reunification on October 3, 1990, have seen a considerable convergence of material living conditions, but important differences in values and attitudes persist. As stated in the 2019 edition of the German Government's "Annual Report on the State of German Unity", a major case in point is satisfaction with democracy: "The rate of approval of democracy in East Germany is alarming: Almost one half of the individuals in the East are rather dissatisfied with the way democracy works". (Bundesregierung, 2019, p.13; translated by the author). The 2020 edition of the Report reiterates this point and states the persistence of a substantially lower level of satisfaction with democracy in the East than in the West.

3
In a European context, low satisfaction with democracy is not a specifically East German phenomenon. Rather, while approval of democracy is typically high in South European countries, the post-communist societies of Central and Eastern Europe usually rank close to the bottom of the respective distribution. In comparison to other post-communist societies, approval of democracy is actually relatively strong in East Germany (Dalton, 2004;Fuchs & Klingemann, 2006). Yet, in spite of the more favorable attitude in East Germany than in other post-communist countries, the persistent East-West divide in German citizens' satisfaction with the way democracy works is salient. 1 Understanding its sources is an important issue not just from an academic perspective, but because it can inform policy makers about ways to address dissatisfaction with democracy. This is important not least in the light of the rise of populist parties particularly in the East of Germany.
Explanations of individuals' satisfaction with democracy (SWD) rely on personal valueorientations and socio-political attitudes-both intrinsic and socialization-related-on the one hand (e.g., Ceka and Magalhaes, 2016;Haerpfer & Kizilova, 2014), and the structure and performance of political institutions and the attitudes towards those on the other (e.g. Armingeon & Guthmann, 2014;Dalton, 2004;Magalhaes, 2014;Torcal & Trechsel, 2016;Wagner et al., 2009). Performance-related explanations arguably are particularly relevant when the working of democracy is concerned rather than general attitudes towards democracy. Nevertheless, it seems clear that person-related and performance-related explanations of individuals' SWD-in this sense-can hardly be separated from each other since persons likely differ with respect to what constitutes their individual notion of "performance" and how they aggregate the various dimensions of performance in their SWD function. Person-related and performance-related factors should thus be viewed as providing complementary rather than competing explanations of SWD. 2 Based on this view, the present paper proposes and tests a political-economy model of SWD that conceptualizes "performance" in terms of a set of demands individuals place on the democratic system and explains individuals' SWD by the individual-specific levels of (deprivation of) the various demands and the weights individuals attach to those demands in their SWD function. The demands on the democratic system are differentiated into two categories. The first category comprises interests, that is, items which the individual appreciates for self-regarding reasons. Consistent with the two-dimensional model of political space (Bornschier, 2010;Lachat, 2018), the model distinguishes between economic interests (employment and wealth) and cultural interests (compositional amenities in terms of cultural homogeneity). 3 The second category of demands comprises morals, that is, ideals which individuals endorse out of moral or ethical concern and which they want to see implemented by the democratic system. 4 The morals are specified as Care, Fairness, Liberty, Loyalty, Authority and Sanctity, as suggested by so-called moral foundations theory (Graham et al., 2011;Haidt, 2012), While the various interests and moral concerns are conceived of as ultimate drivers of SWD, several interests and moral concerns jointly determine two kinds of issue-specific demands-economic and cultural, respectively-which constitute proximate drivers of SWD. The first is demand for redistribution, which can be seen as reflecting both the selfinterest concern for wealth and the moral concerns for Care and Fairness. The second is a demand for restriction of immigration, which can be seen as reflecting a trade-off between the self-interest concern for cultural homogeneity and the moral concerns for Care and Fairness. 5 The focus of the empirical analysis lies on investigating potential East-West differences in the SWD function that may stem from a "communist legacy" with respect to demands on the democratic system. Considering the case of Germany is useful for this purpose because, given the common institutional framework in the East and the West, there is no need to control for differences on the "supply side" of the democratic system, as would be required in a cross-country comparison of "Western" and post-communist societies.
Having described how SWD is linked to demands on the democratic system and which specific demands may be relevant, it will now be explained what determines those demands and, in particular, how East Germany may differ from West Germany in this regard. This involves two steps. The first step relies on a distinction between (a) objective conditions prevailing in the East and West (levels of wealth, unemployment, and compositional amenities) and (b) East and West Germans' evaluation of those conditions on the basis of their respective preferences. Following Senik (2014), it will be assumed that the preference for and evaluation of given conditions reflects "mentalities", which are culture-related and installed in individuals during socialization. In this regard, it was found that-reflecting materialist socialization and a socialist work ethic-East Germans place a higher value on material aspects of life (income) and on working (rather than being unemployed) than do West Germans (van Hoorn and Maaseland ). The second step involves the question how such culturally determined preferences translate into demands on the democratic system. In this regard, it was found that East Germans ascribe a greater responsibility to the government for satisfying politico-economic preferences, in particular for providing jobs and high wages, than do West Germans (Alesina & Fuchs-Schündeln, 2007).
Overall, it can thus be assumed that the demands on the democratic system differ between the East and West for two reasons. First, East and West Germans differ in their preferences with respect to important living conditions and, second, they differ in the extent to which they expect those preferences to be satisfied by the state. This implies the key hypothesis studied in this paper: Observed differences in SWD between East and West Germany reflect not just differences in living conditions but differences in the SWD function that stem from socialization-related differences in both preferences and views about state tasks.
In addition to different preferences concerning various aspects of living conditions and views about the state's responsibility for satisfying them, East Germans' socialist education (before reunification) arguably involved a greater emphasis on the values of solidarity and equality. To the extent that these value orientations persist, this may imply that the moral values of Care and Fairness play a greater role for SWD in the East than in the West, in particular when it comes to redistribution policy.
Using representative data from the European Social Survey (ESS) I find evidence consistent with the above framework and, in particular, the idea that higher demands on the democratic system, as specified, are associated with less satisfaction. Specifically, higher demands for jobs, wealth and compositional homogeneity go with less SWD, and these influences work both directly and indirectly, that is, through the demands for redistribution and (anti-)immigration policies. The demands for redistribution and immigration policies, in turn, are shaped by self-interest concerns and moral concerns, as hypothesized.
These general patterns notwithstanding, I find important differences between the SWD functions in East and West Germany. First, in the East the demand for jobs shapes SWD much more strongly than it does in the West. Second, in the East, the demand for wealth also shapes SWD more strongly than it does in the West, but-in contrast to the West-the demand for wealth does not translate into a demand for redistribution, and the demand for redistribution is not significantly related to SWD. In the East, the demand for redistribution is largely morally motivated, by a concern for Care and Fairness, whereas in the West it stems from both self-interest and moral concern. Third, while the demand for anti-immigration policy depends on homogeneity preference (positively) and a concern for Fairness (negatively) in similar ways in the East and the West, the demand for anti-immigration policy shapes SWD much stronger in the East than in the West. The results concerning employment, wealth and redistribution are consistent with the idea that the provision of jobs (rather than unemployment insurance) and high wages (rather than redistribution through taxes and transfers) are considered to be more important state tasks in the East than in the West (Alesina & Fuchs-Schündeln, 2007).
A core insight from this analysis is that the East-West SWD gap is shaped not just by differences in the extent to which (economic) interests are satisfied or deprived in the two parts of the country, but by the strength with which this satisfaction/deprivation translates into SWD due to East-West specific weights placed on several interests and ideas about what the state is responsible for-that is, different notions of democratic performance. In contrast to the satisfaction of interests, the satisfaction of moral concerns is more important to SWD in the West than the East.
The paper contributes to the literature in several ways. First, it provides and explores a consistent conceptual framework that integrates interests and morals into a political-economy explanation of SWD. Second, it differentiates between ultimate and proximate drivers of SWD and relates the latter to the former. Third, while value orientations have been considered before, the paper incorporates into the SWD function the specific moral values identified by moral foundations theory. Fourth, it demonstrates that socialization-related "mentalities" play an important role for the SWD divide in Germany.
The paper is structured as follows. Section 2 discusses the background and literature which the analysis builds upon. Section 3 provides the conceptual framework and empirical implementation. Section 4 presents and discusses the results. Section 5 concludes.

Historical and Cultural Background
Before reunification in 1990, East and West Germans had been living under different political and economic regimes for more than 40 years. The political system of the Federal Republic of Germany (FRG) was liberal-democratic, with free competition of political parties, whereas the system of the German Democratic Republic (GDR) was characterized by the ruling of the "Socialist Unity Party of Germany". While the economic system of the FRG was basically private-property capitalism, the economy of the GDR was essentially a system of state-property central planning. Due to its Marxist origin, the politico-economic ideology endorsed in the GDR and propagated in schools and universities was strongly materialist, emphasizing the importance of work and the material aspects of life. The moral values of equality and solidarity were much more strongly endorsed in the GDR than the FRG, whereas the opposite applies to freedom.
The societies of the GDR and the FRG were characterized not only by different institutions and cultures but different economic outcomes. Notably, up to the end of its existence, the GDR was characterized by both lower per capita income and lower income inequality than the FRG and an absence of (overt) unemployment.
After the collapse of the Soviet Union and the fall of the Berlin Wall, FRG and GDR were united in October 1990, the new state adopting both the name and the political and economic institutions of the Federal Republic of Germany. Some aspects of the way the transition of the former GDR was handled (e.g. the privatization or unwinding of formerly state-owned firms) as well as attitudes and behaviors of West Germans were experienced by East Germans as arrogant and spurred perceptions of disempowerment, disrespect for life achievements, and unfriendly takeover (Kowalczuk, 2019;Oberender, 2020), in violation of their needs for self-determination (in particular, autonomy and competence). The increase in inequality with which East Germans were faced constitutes a violation not just of East Germans' self-interest, but of the normative value of equality characteristic of the GDR's political culture. The frustration of both East Germans' self-determination needs and normative preference for equality are factors that may have contributed to dissatisfaction with democracy after reunification. 6

Satisfaction with Democracy and the "Communist Legacy"
Political scientists conceive of "political space" as two-dimensional, involving an economic and a cultural domain (Bornschier, 2010;Lachat, 2018). The economic domain involves concerns about employment and wealth, whereas the cultural domain involves concerns about compositional homogeneity. Economic concerns are commonly viewed as primary determinants shaping attitudes toward the democratic process and have been found to have a strong impact on satisfaction with democracy (e.g., Anderson & Guillory, 1997;Clarke et al., 1993). Likewise, concern about compositional homogeneity shapes people's attitude towards important policy issues, notably immigration policy (Card et al., 2012), 1 3 and dissatisfaction with "excessive" immigration in portions of the population may spill over to dissatisfaction with democracy more generally.
While economic and cultural concerns constitute ubiquitous factors shaping satisfaction with democracy, the corresponding demands on the democratic system can be viewed as differing between East and West Germany as a legacy of having lived under different political and economic regimes. As regards economic concerns, East Germans were found to have a greater preference for working over being unemployed (Van Hoorn & Maseland, 2010) and to display a higher level of materialism, as manifested in the relationship between income and life satisfaction being significantly stronger among East Germans than West Germans (Frijters et al. 2004a, Van Hoorn & Maseland, 2010. Both the appreciation of work and material wealth can be viewed as manifestations of Marxist philosophy and education. In a similar vein, Friehe and Mechtel (2014) found conspicuous consumption to be more important in East than in West Germany and attributed the difference to the influence of different political regimes. As regards cultural concerns, intergroup contact theory (Allport, 1954) suggests that individuals living in isolation from foreign cultures tend towards stereotyping and prejudice, and have a greater preference for cultural homogeneity. Contact with foreigners in everyday life was typically rather limited in the GDR. 7 These differences in preference are potentially relevant to satisfaction with democracy because they coincide with different views about who is responsible for satisfying those preferences: private forces or "the state". With respect to this, Alesina and Fuchs-Schündeln (2007) found that East Germans have a profound view that the state is essential for individual well-being and attribute this to "45 years of anti-capitalist indoctrination".
In addition to views regarding issue-specific state tasks, another major dimension of endogenous, path-dependent differences in political preferences refers to attitudes towards democracy more generally. In this regard it was found that support of democracy in new democracies is weaker when people had spent less time under democracy, suggesting that trust in the working of liberal-democratic institutions needs familiarity with them and considerable time to develop (Fuchs-Schündeln & Schündeln, 2015). 8 The evidence on important differences between ex-communist societies and Western societies concerning materialist values, responsibilities of the state, and familiarity with democratic institutions suggests that such differences may play a role in explaining East-West differences in German citizens' satisfaction with democracy.
The discussion so far refers to avenues through which individuals' economic and cultural self-interest may shape satisfaction with democracy. Political psychology suggests that, in addition to such interests, the conformity of democratic outcomes with individuals' moral values constitutes another determinant of satisfaction with democracy.

Moral Values and Political Preferences
Individuals' moral identity, that is, the configuration of the moral principles they endorse, constitutes an important explanation for "Why Good People Are Divided by Politics" (Haidt, 2012) and is thus potentially relevant to what different people expect from the democratic system.
Social psychologists have described individuals' moral identity in terms of a distinct set of "moral foundations", that is, universally available moral intuitions that people across many cultures endorse: Care, Fairness, Liberty, Loyalty, Authority, and Sanctity (Graham et al., 2011;Haidt, 2012). 9 The moral foundations correspond to the virtues of caring and kindness (Care), equality and justice (Fairness), sovereignty (Liberty), loyalty, patriotism and self-sacrifice (Loyalty), obedience and deterrence (Authority), and temperance, chastity and piety (Sanctity). Considering the virtue words is helpful for understanding the contents of the respective moral foundations. For instance, the virtues of chastity or piety associated with the Sanctity foundation suggest that endorsement of Sanctity involves an esteem for traditions and customs. 10 Endorsement of the moral foundations can be found across various cultures, societies, and socio-economic groups, but to different degrees. In particular, cultural differences exist with respect to endorsement of Care, Fairness and Liberty, which are individual-focused in the sense of applying to all individuals independent of their membership to one's group (universalist morality), and the group-focused foundations Loyalty, Authority and Sanctity (parochial morality), the former being endorsed more in Western than in other societies (Graham et al., 2011;Haidt, 2012).
The development of moral identity involves a genetic predisposition, which later undergoes so-called characteristic adaptations during adolescence in response to the specific environments and challenges individuals happen to face, followed by the construction of "life narratives", that is, simplified and selective reconstructions of their past, connected to an idealized vision of the future, and saturated with morality (McAdams & Pals, 2006). 11 Life narratives provide a bridge between an individual's developing adolescent self and her adult moral identity as individual-focused-with a taste for Care, Fairness and Liberty-or group-focused-with a taste for Loyalty, Authority and Sanctity and a heightened sense for the interests of the group. 12 Once acquired during adolescence and early adulthood, moral identity is stable thereafter (Haidt, 2012).
Given the dependence of moral identity on the environments and challenges faced when growing up and the life narratives constructed thereupon, it can be conjectured that individuals socialized under different political regimes may differ not only in their self-regarding, but in their moral preferences with respect to the democratic system and, in particular, the demands for morally motivated policies. Preferences for Care and Fairness are potentially 9 The terminology differs somewhat between sources. Graham et al. (2011) refer to Care as Harm, to Loyalty as Ingroup and to Sanctity as Purity. 10 In finding proxies for the moral foundations I will rely on their content as described by the virtue words. 11 The dispositional roots of moral identity are closely related to the personality trait of "openness to experience": a tendency to cherishing novelty, variety, and diversity, while simultaneously being less sensitive to signs of threat (Haidt, 2012). 12 Endorsement of individual-focused as opposed to group-focused moral foundations is different from the ideological position as liberal-progressive (left-leaning) and conservative (right-leaning), though some similarities exist. While adherents to both political camps endorse Care, Fairness and Liberty in similar ways, liberal-progressives endorse Loyalty, Authority and Sanctity less than do conservatives (Haidt, 2012). relevant in this respect, as solidarity and equality were key moral values emphasized in socialist education.

Broad Hypotheses
In sum, it can be stated that the socialization in different political regimes and social contexts implies not only a difference in East and West Germans' preference for wealth, jobs and cultural homogeneity, but different views on the state's responsibility for satisfying those preferences. Together, these differences suggest that economic and cultural concerns may shape East and West Germans' SWD functions differently, as may differences in key moral values endorsed.

Conceptual Framework
The main assumption to be explored in this paper is that individuals are more satisfied with democracy the more their demands on the democratic system are satisfied. Assuming there is a given set of demand items and denoting by D ki individual i's demand for item k and by S k the corresponding supply, the generic SWD function can be stated as While the level of supply of the various items is actually the result of policies pertaining to them, this formulation explicitly accounts for the possibility that individuals view the resulting levels (e.g., the degree of redistribution or the amount of immigration allowed, see below) as indicating the performance of the democratic system. Since, in a withincountry context, the levels of supply are invariant with respect to individuals, the above specification stipulates that people with higher demands for the various items are less satisfied.
Tying in with the literature reviewed, I differentiate the demand items into interests, that is, demands which individuals want to be satisfied by the state for self-regarding reasons, and morals, that is, ideals which individuals want to see fostered by the democratic system out of moral or ethical concern. Consistent with the two-dimensional model of political space, interests are differentiated into economic interests, specified as employment and wealth, and cultural interests, specified as compositional amenities. The latter can conceptually be understood as a demand for compositional homogeneity or for its converse, compositional diversity (multi-culturalism). Thus, while it is theoretically plausible that higher demands for employment and wealth go with less SWD (provided that individuals regard their supply as a state task), the sign of the relationship is theoretically ambiguous in the case of compositional homogeneity. The morals are specified as Care, Fairness, Liberty, Loyalty, Authority and Sanctity, as suggested by Moral Foundations Theory. Similar to compositional homogeneity, the signs with which the morals enter the SWD function may also be ambiguous (for instance, people may demand less rather than more authority).
While the various interests and moral concerns are conceived of as general and ultimate drivers of SWD, I also consider two issue-specific and proximate drivers of SWD. I refer to them as proximate drivers because they are assumed to be themselves driven by interests and morals and may thus mediate the relationship between the ultimate drivers and SWD. The first is demand for redistribution (e.g. Alesina & La Ferrara, 2005;Corneo & Grüner, 2002), which can be seen as reflecting both the self-interest concern for wealth and the moral concerns for Care and Fairness. The second is a demand for restriction of immigration, which can be seen as reflecting a trade-off between the self-interest concern for compositional homogeneity and the moral concerns for Care and Fairness.

Empirical Implementation
The data used in this study are taken from the European Social Survey (ESS); see www. europ eanso cials urvey. org. The ESS is a cross-sectional, multi-country survey. ESS data are obtained using random (probability) samples, where the sampling strategies are designed to ensure representativeness and comparability across European countries. For topicality, I use data from Round 9 (2018) for Germany, differentiated by the East and West German Federal States (n = 386 for East and n = 2026 for West). The ESS is useful for the present study because it offers not only data on satisfaction with democracy but on demands for redistribution and immigration policies and proxies for the various interests and moral concerns discussed in the preceding subsection. Satisfaction with democracy is elicited with the question "How satisfied are you with the way democracy works in Germany?" ("extremely dissatisfied" = 0 to "extremely satisfied" = 10). The formulation indicates that the item refers to the performance ("working") of the specific institutional system in Germany in terms of what individuals demand it to deliver (not general approval of democracy as a regime type).
As a proxy for the demand for jobs (to be supplied by the state) I use a dummy variable that indicates whether an individual is unemployed and actively looks for a job. The idea behind this operationalization is the following: (i) People are assumed to have a preference for work over being unemployed (possibly differing in strength between East and West), that is, unemployed individuals are less satisfied than employed ones. (ii) To the extent that they hold the state responsible for providing jobs, dissatisfaction with being unemployed translates into dissatisfaction with democracy. The unemployed are thus hypothesized to be less satisfied with democracy than individuals with other employment status.
As a proxy for the demand for wealth (to be supplied by the state) I use household income. The idea behind this operationalization is the following: (i) People are assumed to have a preference for wealth (possibly differing in strength between East and West), that is, less wealthy individuals are less satisfied with their level of wealth. (ii) To the extent that they hold the state responsible for providing wealth, dissatisfaction with wealth translates into dissatisfaction with democracy. Less wealthy individuals are thus hypothesized to be less satisfied with democracy than wealthier ones.
Finally, as a proxy for the demand for compositional homogeneity (to be supplied by the state) I use the perception that immigrants make the country a worse, rather than better place to live. 13 The proxies for the moral demands are items that measure the degree to which individuals consider ethical principles that correspond to Moral Foundations Theory to be important. While the survey items do not explicitly refer to those principles, the wording of the survey questions (see Table 1) indicates a correspondence in substance.
The demands for redistribution and (anti-)immigration policies are measured by survey items that elicit, respectively, to what extent individuals disagree/agree to the proposition that the government should reduce differences in income levels, and how many individuals of a different race or ethnic group than the majority should be allowed to come to Germany. 14 Table 1 provides information on the definition and characteristics of the main variables used, whereas Table 5 in the Appendix reports the summary statistics of the socio-demographic variables (age, gender and level of education) that will serve as control variables. 15 The table reveals a considerable difference in satisfaction with democracy in East and West Germany: mean SWD is 4.87 in the East and 5.95 in the West (on the 0-10 scale). With regards to interests, the demands for employment and wealth as well as for compositional homogeneity are larger in the East than in the West, as indicated by the higher level of unemployment, lower level of income and the more pronounced view that immigrants make life worse. It should be noted, however, that the standard deviations of these variables are quite large, so that the differences in mean levels are not statistically significant. With respect to the morals, there are some East-West differences, but they are not large (and far from being significant). East Germans have a stronger concern for Authority and a slightly stronger concern for Care and Sanctity than West Germans, while the opposite applies to Fairness and Liberty. The concern for Loyalty is the strongest moral concern in both parts of the country and almost equally strong in East and West. The concern for Sanctity is the weakest moral concern in both parts of the country.

Econometric Approach
The main identifying assumption is that individuals regard satisfaction of the various demands discussed so far as tasks of the democratic system and evaluate its working accordingly. With this assumption in mind, the main estimating equation is an augmented version of Eq. (1) stated above: where the vectors D and P refer to the ultimate demands (interests and morals) and proximate demands (for redistribution and immigration policies), respectively, X is a vector of socio-demographic controls (age, gender, level of education) and is an error term. 16 (2) SWD i = const + � D i + � P i + � X i + i , 14 Alternatively I used indicators of how many individuals of the same race or ethnic group or from poorer countries outside of Europe should be allowed to come. This did not affect the main results. 15 The mean age and share of women in the East (53.16 years and 53 percent) are higher than in the West (49.04 years and 48 percent) whereas the levels of education are the same (slightly higher than upper tier secondary). 16 Age, gender and the level of education are the standard socio-demographic variables, along with income and unemployed status, routinely used in research on satisfaction with democracy (e.g., Hansen and Goenaga, 2021 for a recent discussion). They are included here to minimize the risk of omitted variable bias with respect to the variables of main interest.

Table 1
Main variables ( Source:

ESS Round 9)
Immigration-Bad was recoded such that higher values correspond to higher demand for homogeneity. The morals were recoded such that higher values indicate stronger endorsement. Redistribution was recoded such that higher values reflect higher demand for redistribution. The expected signs refer to the relationship between SWD and higher values of the respective variables. The expectation of a positive sign for Income is based on the assumption that richer individuals have a lower demand for wealth to be provided by the state and are, therefore, expected to be more satisfied with democracy (see discussion in the text)

Proximate drivers
Demand for redistribution policy Redistribution (Government should reduce differences in income levels. "Disagree strongly" = 1 to "Agree strongly" = 5.) As already noted, the signs of some of the coefficients are ambiguous, depending on whether the (representative) individual considers the supply of an item to be lower or higher than what she demands (e.g. compositional homogeneity). Insignificance of a coefficient may indicate that the level of supply is just right or that the item is irrelevant to the individual's SWD.
In addition to Eq.
(2), a reduced-form analogue will be estimated in which the proximate demands are omitted, as well as equations which explain the proximate demands by the same set of ultimate demands and controls as in Eq. (2). In robustness checks, additional controls (position on the left-right scale, demand for environmental policy, and immigrant status) will be included. 17 Following, e.g., Angrist and Pischke (2009) ordinary least squares will be used throughout.

Main Results
Table 2 displays the estimation results for Eq. (2) differentiated by East and West Germany. Regressions A and B refer to the specifications with and without the demands for redistribution and anti-immigration policies in the East whereas regressions C and D refer to those specifications in the West.
To begin with, the demographic controls (gender, age, level of education) are insignificant in the East (regressions A and B). In the West, SWD is significantly positively related to the level of education (regressions C and D). When the demands for redistribution and immigration policies are omitted (regression D), SWD in the West is, in addition, significantly negatively related to age.
With respect to the variables of main interest, Income is always significantly positively associated with SWD, consistent with the idea that higher/lower demand for wealth implies less/more satisfaction. Likewise, Immigration-Bad is always significantly negatively associated with SWD, consistent with the idea that a higher/lower demand for compositional homogeneity implies less/more satisfaction. Unemployed status is significantly negatively associated with SWD in the East, but insignificantly so in the West. These results indicate that economic and cultural interests play an important role for SWD. As concerns the morals, a taste for Authority always attracts a significantly positive coefficient. The other morals are mostly insignificant except for Care in regression D, in which it attracts a significantly negative coefficient: West Germans demanding more Care are significantly less satisfied with democracy when the demands for redistribution and anti-immigration policies are omitted-but not so when those demands are controlled for. This provides a first indication that the moral demand for Care may play a role in these latter demands (see below). In the East, the Care variable is insignificant. 17 In the perspective of the two-dimensional model of political space, the left is characterized by a high demand for redistribution combined with a moderate or indifferent position towards homogeneity whereas the right is characterized by a high demand for homogeneity combined with a moderate or indifferent position towards redistribution (Bornschier, 2010, Lachat, 2018. From a social-psychology perspective, the position on the left-right scale partly reflects the individual's moral profile, the latter being a more fundamental notion than the former (Haidt, 2012). For these reasons, the placement on the left-right scale does not explicitly appear in the main specification. Ecological or environmental concern can be thought of as defining a third dimension of political space.
Considering the demands for redistribution and anti-immigration policies, the former attracts a significantly negative coefficient in the West: Individuals demanding more redistribution are less satisfied with democracy. In the East, redistribution demand is insignificant. Demand for anti-immigration policy attracts a significantly negative coefficient both in East and West Germany: Individuals demanding a stricter immigration policy are less satisfied with democracy.
Comparing the magnitudes of coefficients that are significant in both East and West, a striking result is that the coefficient on Income is 4 to 5 times larger in the East than in the West (and the differences are significant): East Germans' demand for wealth plays a much bigger role for SWD than West Germans' demand for wealth. But, as noted above, East Germans' SWD is not significantly related to a demand for redistribution. For East Germans wealth is thus not only more relevant to SWD, but redistribution through taxes and transfers does not seem to be appreciated as a substitute for wage income in East Germans' SWD function. In contrast to the demand for wealth, the importance of the demand for compositional homogeneity (variable Immigration-Bad) does not strongly differ between East and West. In spite of this, the demand for anti-immigration policy is a much stronger driver for dissatisfaction with democracy in the East than in the West (the coefficients on Anti-Immigration being significantly different). This point will be discussed more thoroughly below.
In sum, the demand for jobs is a significant driver of SWD in the East but insignificant in the West, and demand for wealth is a significantly stronger driver of SWD in the East than in the West. In spite of the latter, however, the demand for redistribution policy is insignificant in the Eastern SWD function, while being significant in the West.
Comparing coefficients between specifications that include the preference for redistribution and anti-immigration policies (regressions A and C) with those that exclude them (regressions B and D) shows that the coefficients on Income and Immigration-Bad are larger in magnitude when the policy preference indicators are omitted. This suggests that the demands for wealth and compositional homogeneity play a role in shaping these policy preferences. The next subsection addresses these demands in more detail. Table 3 presents estimation results for equations which explain the demands for redistribution and restrictions on immigration in terms of the various interests and moral concerns. Regressions A and C refer to the demand for redistribution in the East and West, respectively, and regressions B and D to the demand for anti-immigration policy.

Demands for Redistribution and Anti-immigration Policies
In the East, the demand for redistribution is significantly positively related to concerns for Care and Fairness, but not to Income (regression A). In the West, the demand for redistribution is not only significantly positively related to concerns for Care and Fairness, with smaller coefficients than in the East, but significantly negatively related to Income (regression C). 18 Redistribution preference in the East is thus entirely a moral issue, unrelated to self-interest, whereas in the West it reflects both moral concerns and self-interest (in wealth). The dominance of concerns for Care and Fairness in East Germans' demand for   redistribution is consistent with the strong endorsement of solidarity and equality as normative values in the former GDR. The demands for anti-immigration policy in both East and West are significantly positively related to Immigration-Bad and negatively related to concern for Fairness, the respective coefficients being of the same magnitude in both parts of the country. In addition, the demand for anti-immigration policy in the West is significantly negatively related to concern for Liberty and significantly positively related to concern for Care, Authority and Sanctity. West Germans' stance on immigration policy is thus more morally founded than that of East Germans. However, stronger endorsement of the moral foundations does not exhibit a unidirectional relationship to anti-immigration policy. Rather, endorsement of the individual-focused (universalist) morals (in particular Fairness and Liberty) go with a preference for less restrictive immigration policy, whereas stronger endorsement of the group-focused morals (in particular Authority and Sanctity) go with a preference for more restrictive immigration policy. 19

Robustness Checks
I conducted robustness checks with respect to including additional explanatory variables in the SWD equations: placement on the left-right scale, environmental concern and being an immigrant As seen in Table 6 in the Appendix, inclusion of these additional correlates has practically no effect on the results discussed so far. A more right-leaning position, per se, is positively and marginally significantly associated with SWD in the West and not significantly related to SWD in the East. 20 Environmental concern-which represents a third dimension of political space, in addition to the economic and cultural dimensions-is insignificant both in the East and the West. Finally, immigrants (that is, people not born in Germany) are significantly more satisfied with democracy in both the East and the West, the coefficient in the East being about 4 times as large as that in the West. Assuming that immigrants in the East and the West are not different in relevant ways and can thus be taken as a benchmark, the latter finding provides a salient illustration of the East-West gap in satisfaction with democracy.

Discussion
The results reveal some striking differences in the SWD functions in East and West Germany. Table 4 presents the absolute values of the significant coefficients in regressions B (East) and D (West), that is, the reduced form SWD functions. The entries are the sums across several categories of interests and moral concerns. Economic interests constitute a tremendously more important factor for SWD in the East than in the West 19 The positive relationship between endorsement of Care and the demand for anti-immigration policy is somewhat.
at odds with this pattern since the Care foundation is usually considered to be a universalist moral concern (Graham et al., 2011). 20 The notion that an individual's satisfaction with democracy depends on ideological congruence with the government is of limited relevance in the current case, as the German government at the time of observation rested on a coalition between the conservative Christian Democratic Party and the Social-Democratic Party. In a multi-country perspective, Dahlberg and Holmberg (2012) find government effectiveness to be more important for satisfaction with democracy than ideological congruence.
whereas moral concerns are less important SWD drivers in the East than the West. 21 The latter applies to the universalist (individual-focused) value of Care: it is insignificant in the East and significant in the West (whereas the group-focused morals, Authority in particular, are actually slightly more important in the East than in the West). Similar to economic interests, cultural interests (that is compositional homogeneity) are also more important in the East than the West, but the difference is smaller than in the case of economic concerns.
To put the results in perspective, it can be noted that fewer jobs and lower incomes are standard explanations of low SWD in East Germany (e.g. Bundesregierung, 2019Bundesregierung, , 2020. What the above analysis has shown, however, is that not only the levels of those variables differ between the East and the West, but the strength with which they are linked to SWD. With respect to employment, the association with SWD is large and significant in the East, while being insignificant in the West. With respect to income the association with SWD is at least 4 times stronger in the East than in the West (0.16 versus 0.04, Table 2, columns B and D) whereas the East-West income gap itself amounts to 12.8 percent (Table 1). 22 With respect to compositional homogeneity, the association with SWD is 10.5 percent stronger in the East and the demand itself differs by 16.6 percent (Table 1). Differences in the strength with which the various demands on the democratic system are linked to SWD are thus an important contributor to the SWD gap in addition to differences in the demands themselves, and with respect to the economic demands the former far outweigh the latter.
The great importance of economic factors for East Germans' satisfaction with democracy is consistent with earlier findings of a materialist and work-focused value orientation (van Hoorn & Maseland, 2010) as well as stronger ideas that the state is responsible for the provision of wealth and jobs (Alesina & Fuchs-Schündeln, 2007). Both of these types of attitude are a distant echo of the political ideology and cultural practice of the GDR. It is remarkable, however, that these differences in preference still prevail, three decades after reunification and one decade after they were documented in those studies.
The finding that East-West differences in the SWD function are more important for the SWD gap than are differences in the levels of the relevant factors accords well with findings by Biermann and Welsch (2021) on the East-West gap in Germans' life satisfaction. Their results indicate that the satisfaction gap is attributable to "mentality" to almost the same degree as to objective circumstances and that satisfaction with democracy plays an important part in explaining the life satisfaction gap.
From a policy point of view, the findings suggest that improving the living conditions may be necessary, but not sufficient to increase the satisfaction with democracy in East Germany. Educating people about the working of liberal-democratic institutions and what can realistically be demanded from them seems equally important.

Conclusions
This paper has explored satisfaction with democracy in Germany with a special focus on differences between East and West Germany three decades after reunification. The framework employed was a political economy model, that is, one in which satisfaction with democracy is supposed to depend on the extent to which individuals perceive various demands on the democratic system to be satisfied. A major novelty of the approach is that not only demands based on self-interest, but demands with respect to fundamental moral values were included among the potential determinants of SWD.
Empirical application of the model has revealed substantial differences between the East and West German SWD functions in that the satisfaction of economic interests is much more important in the East than the West whereas the opposite applies to moral concerns. Demands on redistribution and immigration policies-conceptualized as proximate drivers of SWD-also shape SWD differently in the East and the West, in addition to being shaped by interests and morals in different ways. Differences in the relationship between economic demands and SWD were found to be more important than differences in the levels of demand.
Most salient results are that the demand for jobs is a strong factor for SWD in the East while being insignificant in the West and that the demand for wealth is a much stronger factor in the East than the West, whereas the demand for redistribution is insignificant in the East but significant in the West. This suggests that not only the satisfaction of the various demands shapes SWD differently in the two parts of the country, but also the way people expect those demands to be satisfied: In the Eastern view, wealth should be provided not indirectly, through the tax and transfer system, but directly in terms of primary incomethat is, through wages. Noting that wages were set by the state under communism, this is consistent with the idea that East Germans' preferences concerning the performance of the democratic system are still shaped by the politico-economic regime that disappeared three decades ago.
Given the data used, the analysis described in this paper is correlational in nature. Yet, inclusion of the moral identity variables among the explanatory variables goes some way towards controlling for unobserved traits since moral identity is not only fairly stable from the beginning of adulthood but closely related to personality traits, in particular "openness to experience". While not being a perfect substitute for individual fixed effects, inclusion To put the paper in perspective, the case of Germany is special in the context of studies on SWD in post-communist societies: While post-communist countries differ from Western countries (as well as from each other) with respect to the institutional details of the respective political systems, East and West Germany have a common institutional framework. This has permitted to focus the analysis on the political preferences and the respective East-West differences as factors shaping SWD. Thus, while an obvious avenue for future research is to apply a similar preference-based analysis to a comparison between other capitalist and post-communist countries, such an analysis will have to take institutional variables into account along with personal-level demands on the democratic system. Such an extended framework will allow to investigate whether the East-West differences of the SWD function identified with respect to Germany apply to East and West European countries more generally.

Appendix
See Tables 5 and 6. The variable highest level of education is coded as 1 = "less than lower secondary" to 7 = "MA or higher". The mean values reported are slightly higher than 4 = "upper tier secondary" Acknowledgements I am grateful to two anonymous reviewers for providing useful comments and recommendations.
Funding Open Access funding enabled and organized by Projekt DEAL.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Dependent variable: satisfaction with democracy. Controls are age, female and level of education; t-statistics in parentheses *p < 0.1; **p < 0.05; ***p < 0.01

East West
A B C D