The Tocqueville Review/La Revue Tocqueville, Vol. XXIV, n

The Tocqueville Review/La Revue Tocqueville, Vol. XXIV, n 2 2003 THE LEGITIMACY OF INEQUALITY ON BOTH SIDES OF THE ATLANTIC A Comparative Analysis of Attitudes in Canada and Germany Heinz-Herbert NOLL and Lance W. ROBERTS All societies contain organized inequalities, although to a different degree. The amount of equality possible and beneficial for societal wellbeing and the magnitude of inequality tolerable, acceptable and required as a prerequisite for competition and growth are questions continuing at the centre of sociological as well as political debates. 1 Inequality concerns the differential allocation of resources and rewards between social locations. Since resources and rewards associated with social statuses are major determinants of opportunities available to status occupants, social inequalities reflect also differences in power, which affects both people s welfare and the respect they receive (Sennett 2002). Persons in inferior social locations are more likely to experience both deprivation and insolence; those in privileged positions relative comfort and admiration. The central importance of inequalities to life chances requires that such differences be legitimated. It creates problems for privileged and deprived groups, including tensions and conflicts within the society as a whole, if evident inequalities are perceived as illegitimate, unfair or unjust. Thus, social inequalities need continuous legitimation, which, however, can only come from abstract cultural scripts that need to be transformed into social norms and institutional mechanisms. Beliefs about which kinds of inequalities, under which conditions, to which degree are considered legitimate change across time, and vary among social groups within a society, as well as between nations and

2 Heinz-Herbert Noll & Lance W. Roberts across cultures. Thus, the degree to which a certain distribution of income, wealth and life chances the inequality regime of a specific nation is perceived as legitimate, how this differs between societies and whether these perceptions and evaluations change across time needs to be studied empirically. This article examines the notions and beliefs underlying the legitimation of inequality as reflected in individual perceptions and attitudes of the population in two federal states on opposite sides of the Atlantic, Canada and Germany. Several reasons justify comparing the subjective perceptions of and attitudes toward inequality in these two nations. First, orientations toward inequality indicate how the society as a whole and people s positions within a society are perceived, understood, and interpreted by the population. Second, such examinations are of sociological relevance since perceptions of justice, of actual and just distributions of rewards, of power and inequality, are not formed as individual idiosyncrasies, but are themselves results of past and present power struggles (Swallfors 1993:87). Finally, value orientations and attitudes reflect the degree to which inequality is tolerated and justified within a society or culture. Thus, these orientations and attitudes at the micro level ultimately form the basis of legitimization for national inequality regimes, a topic which is not only a prominent subject of scientific research but is also of central importance to politics and the policy making processes. Since tolerance of inequality and legitimation beliefs do not only correlate with social positions within societies but also seem to depend on the macro characteristics of a society, there is good reason to study attitudes toward inequality from a cross-national point of view. But why choose to compare Canada and Germany? Canada and Germany present interesting structural and cultural similarities and differences, as shall be elaborated later. One represents a European nation; the other a North American society, with all the trans-atlantic historical and institutional division that implies. Both are modern federations, with Quebec and East- Germany containing distinctive regional variants to the predominant culture and institutional structures. The latter characteristic bears the particular opportunity to study differences and similarities in attitudes toward inequality not only at the national, but also at the sub-national level. Although our analysis will be focussed on the comparison of the two nations Canada and Germany as well as the divisions within these

The Legitimacy of Inequality 3 two nations, we will occasionally refer to the United States and include respective data as a reference in our comparative study. ATTITUDES TOWARDS INEQUALITY DIMENSIONS AND DETERMINANTS What do we precisely mean by talking about attitudes toward inequality? Attitudes toward inequality are as multidimensional and complex as inequality itself. Survey programmes and analyses on such attitudes usually address different kinds of perceptions, preferences, images and beliefs related to social and economic inequality. Nonetheless, previous studies (e.g. Sandberger 1994) have repeatedly demonstrated that attitudes toward inequality are typically structured along an axis reaching from egalitarian-critical disapproval to affirmative-legitimative approval of the inequality structure given in a specific society. With this structuration various attitude dimensions may be distinguished including, for example, equality-related value orientations and ideals, perceptions and evaluations of distributions and inequality structures, attitudes toward the functionality of inequality as an incentive for achievement and economic growth, attitudes toward principles of legitimation as well as mechanisms and norms of distribution. 2 Diagram 1 provides an overview of various attitude dimensions and illustrates how these relate to the two opposite poles distinguished. Attitudes close to the egalitarian-critical pole of the axis, for example, frequently go together with a high priority on egalitarian values, a critical view of the functionality of inequality for individual achievement, wealth and economic growth, and a supportive view of welfare state interventions. Thus, this syndrome of critical perceptions and evaluations of inequality may also be considered an egalitarian-statist point of view. Attitudes close to the affirmative-legitimative pole of the axis, on the other hand, are usually characterised by a low priority of egalitarian values, the belief that inequality is functional for individual achievement and economic growth, the emphasis on skill and achievement as major distribution criteria, and an anti-welfare state attitude. This syndrome of beliefs may thus be labeled as a meritocratic-liberal point of view. An extensive literature exists as far as the character, causes, and consequences of inequality in modern societies are concerned (e.g. Lemel/Noll, 2002). This literature demonstrates that both structural and cultural factors help account for evident inequalities. As far as attitudes toward inequality and their determinants are concerned, previous studies suggest a distinction between micro- and macro-level explanations.

4 Heinz-Herbert Noll & Lance W. Roberts Diagram 1: Dimensions of Attitudes Toward Inequality Attitude Poles egalitarian-critical affirmative-legitimative Value Orientation and Evaluation of Distribution Outcomes high priority of equality magnitude of inequality inappropriate / unjust low priority of equality magnitude of inequality appropriate / just Attitude Dimensions inequality not functional for individual achievement, wealth and economic growth Functionality of Inequality inequality functional for individual achievement, wealth and economic growth Norms of Distribution / Principles of Legitimation distribution according to demand distribution according to skill and achievement Distribution Mechanisms (re)distribution by governmental action pro-welfare state distribution as result of market activities anti-welfare state Concerning the impact of structure, macro properties of societies, such as class structure or the distribution of power resources, leave their imprints at the micro level not only as different life chances but also in different explanations and interpretations of social conditions (Svallfors 1993:87). Beyond the effects of structural variables, cultural influences like national institutions and policies are supposed to be particularly relevant from this point of view. In this sense, international comparisons may be regarded as experiments that allow the investigation of cultural differences in institutional configurations and national policies (Svallfors 1993:88). Explanations of cross-national differences in attitudes toward inequality at the macro level are based on the assumption that these attitudes and beliefs at least partially reflect societal characteristics. Attitudes as manifestations of societal or cultural characteristics have

The Legitimacy of Inequality 5 been labeled by some scholars as dominant (Kluegel/Smith 1986) or primary (Wegener 1992) ideologies, being internalised by individual citizens via socialisation processes. Macro-level characteristics supposed to shape attitudes toward inequality include the general level of living and the degree of inequality, the institutional framework and political environment, the predominance of specific religious denominations, as well as history and cultural patterns. Explanations at the micro level start from the premise that attitudes toward inequality also vary considerably within societies. Differences in attitudes toward inequality among individuals and socio-economic groups sometimes referred to as secondary ideologies (Wegener 1992) are supposed to reflect differences in socio-economic position and economic interests, but also differences in value orientations or political beliefs: People in different positions (defined by status, race, gender, or other social distinctions) will be expected to react differently to social inequalities that affect them (Kluegel/Smith 1986:11). Respective hypotheses claim that attitudes towards inequality are more egalitariancritical the lower the social-economic status of a person, or that women are on average more in favour of egalitarian values than men. Other hypotheses claim that critical views of inequality go together with preferences for left-wing political parties as well as post-material value orientations. Although our cross-national study will be focussed primarily on macro-level factors, micro-level variables need to be taken into account as well. Since micro-level variables which are supposed to have an impact on attitudes toward inequality may be distributed differently within the societies under study, they must at least be controlled for in order to avoid composition effects. COMPARING CANADA AND GERMANY Given the fact that both nations, Canada and Germany are modern, pluralistic, democratic, federal, market-based societies, we may expect considerable similarities between them in terms of current structures but also trends of social change. But obviously there are also evident differences derived from the historical, cultural, and structural differences that distinguish these two societies. Moreover, the differences may to some extent be due to a principle of singularity (Caplow 1998), which suggests that although modern societies may share several dominant trends, the expression of these trends is conditioned by social and cultural differences.

6 Heinz-Herbert Noll & Lance W. Roberts Based on this insight, it is reasonable to expect that a comparison of Canada and Germany might reveal both interesting similarities and differences regarding the populations attitudes and orientations toward inequality. From this point of view our discussion is guided by the crossnational research, from which we selected five macro level hypotheses and factors that seemed pertinent to examining comparative differences in perceptions and evaluations of social and economic inequalities in Canada and Germany (Haller 1989; Klingemann 1989; Alwin et al. 1990; Noll 1998; Svallfors 1993; Haller/Mach/Zwicky 1995; Frizzell/Pammett, 1996; Delhey 1998; 1999). These factors include the level of development and standard of living, the existing level of inequality, the dominant religious orientation, the character of the welfare state, and the division between Europe and the New World. The anticipated differences these factors make, and the location of Canada and Germany on each dimension, are considered in turn. Level of development and income inequality: Haller (1989) and Klingemann (1989), among others, have argued that attitudes toward inequality are affected by the general level of socio-economic development and standard of living in a society. From the assumption that for people who are better off, inequalities will be relatively less important than for people worse off, it is hypothesized that the populations of societies with higher development levels and living standards will be less critical of inequalities. The level of income inequality is anticipated to have an opposite effect, with sensitivity and resistance toward inequality being greater where the distribution of resources and rewards is relatively more unequal. Of particular importance on this account is the experience of relative deprivation, in which one group compares itself to some significant others (usually those better off) who define standards of expectation.

The Legitimacy of Inequality 7 Table 1: Indicators of Level of Living, Income Inequality and Welfare State Canada Germany United States Human Development Index 1 0.937 0.921 0.937 GDP Per Head (PPP US$) 2 27.130 25.350 34320 Gini-Coefficient 3 0.285 0.261 0.343 5 Percentile Ratio (90/10) 4 3.87 3.18 5.60 5 Poverty Rate (less than 50 % of average 13 9 23 equivalence-household income) 5 Public Social Expeditures in % of GDP 6 16,9 26,6 16,0 1) 2001; Human Development Report 2003; 2) 2001; Human Development Report 2003; 3) 1994; LIS-Data; 4) 1994; LIS-Data; 5) Glatzer/Hauser (2002: 196, 206); OECD 2001: 73. Table 1 provides indicators on the standard of living, income inequality and welfare state expenditures for Canada, Germany as well as the U.S. and reveals some differences of note. Although both economies belong among the richest in the world, the standard of living (as measured by GDP) seems to be somewhat higher in Canada, since Germany s average GDP per capita has been diminished moderately after reunification. Canada also scores somewhat higher on the Human Development Index, a composite measure of Human Development annually published by the United Nations Development Programme, covering GDP per capita as one of its three dimensions. Measures of income inequality indicate that the income distribution in Canada is slightly more unequal than in Germany. All three measures of income inequality, including the Gini-coefficient, percentile ratio and relative poverty, are higher for Canada than Germany. Based on these findings it is reasonable to conclude that, on average, Canadians enjoy a somewhat higher level of living than Germans, but at the same time are living in a society characterized by a more unequal distribution of rewards. For all the indicators presented in table 1 Canada falls between the United States and Gemany, except the Human Development Index, which is at the same level for the U.S. and Canada in 2001. When considered in light of the general hypotheses, these findings lead to mixed

8 Heinz-Herbert Noll & Lance W. Roberts anticipations about Canadians and Germans attitudes toward inequality. Based on their higher standard of living Canadians should be, on average, less critical of inequality than Germans, while more critical because of the higher degree of income inequality. Religious orientation: Some scholars suggest that religious orientations matter in accounting for cross-national differences in attitudes toward inequality. Based on a comparison of ten nations, Haller, Mach and Zwicky (1995:227) claim that populations of Catholic societies are more in favour of egalitarian attitudes and state intervention than populations of Protestant societies. The explanation for this finding is that Protestant doctrine assigns a more active role and greater responsibility for personal fate to the individual citizen, whereas Catholic doctrine encourages a greater emphasis on collective and state responsibilities for the welfare of the individual. Regarding religious orientation and denominations, some important differences exist between Canada and Germany. On average Canada is less Catholic and more ecumenical than Germany. The overwhelming majority (87%) of Canadians belong to three mainline denominations: the Catholic and Anglican Churches and the United Church, an ecumenical Protestant denomination with sectarian origins (Lipset 1991:88). Among these adherents, the largest proportion are Catholic (45%). Throughout Canada the processes of secularization have reduced the participation of adherents in their churches. Nationally, between 1975 and 1995, weekly attendance at church services declined from 31 to 25%; while in Quebec, weekly attendance dropped from 49 to 24% during the same period. These data suggest that, while most Canadians continue to identify themselves as religious, their support of religious organizations is declining (Bibby 1995). While in West Germany the proportions of the population belonging to either the Catholic (41%) or the Protestant denomination (42%) are with some regional differences about the same size, there is only a small Catholic minority in Eastern Germany (3%). People in the Eastern part of the country are either Protestant (22%) or, in large measure, do not belong to any denomination (73%). 3 Weekly attendance at church services among those belonging to a denomination amounts to only 11% in 1998. In West Germany weekly attendence dropped from 18% in 1978 to 12% in 1998, in East-Germany from 13% in 1993 to 8% in 1998. 4 Because of religious heterogeneity, the implications of these differences in religious orientation and denominations between the two

The Legitimacy of Inequality 9 societies are ambiguous. However, we might expect more affirmative attitudes toward inequality in Canada than in Germany because of the smaller share and influence of the Catholic Church, particularly in the Anglophone provinces of Canada. Following the same reasoning we could also expect more affirmative attitudes in the East of Germany than in West-Germany due to its more Protestant character. Ideological and Institutional Regimes: Beyond religious orientations, the predominant political environment is said to play a crucial role in forming attitudes toward inequality. Within political environments it is useful to distinguish between ideological and institutional regimes. Ideological regimes refer to the dominant political ideology in a society that most people adhere to... to a greater or lesser extent (Kluegel & Smith 1986:7). Such ideologies reside at the individual level and are incorporated through the processes of political socialization. Institutional regimes, by contrast, are expressions of the collective interest nested in the characteristic social policies and programmes supported by the state. Institutional regimes are typified by welfare state types. Of course, ideological and institutional regimes have a reciprocal relationship in which institutional arrangements both shape and express the political interests of the population. Dominant ideological environments and political cultures vary along a conservative to liberal continuum. The liberal types stressing collectivist, egalitarian values, like those predominant in socialist systems or social democratic milieus, are expected to produce more critical attitudes toward inequality than more conservative societal environments and milieus, like those found in liberal market economies, stressing individualistic values and policies. For much of the last half century, three political parties have been evident on the national stage in Canada. These include the Progressive Conservative, Liberal, and New Democratic parties, the latter capturing the social democratic position on the political spectrum. The fact that the New Democratic Party has never formed either the government or the official opposition in the Canadian Parliament indicates that the Canadian ideological environment is not on the liberal end of the ideological continuum but, in fact, is somewhere near the middle. During the decade of the 1990s, the reference point for these survey results, the Liberal Party formed the government with a reputedly centre-right mandate. The Situation in Germany is more complicated due to the two separate states existing after World War II until 1990. While the German Democratic Republic was under a non-democratic communist

10 Heinz-Herbert Noll & Lance W. Roberts government during the whole period of its existence, all the governments in the Federal Republic of Germany were established by coalitions of several democratic parties, led by either the Christian Democratic Party or the Social Democratic Party. The latter formed the federal government in the periods from 1969 to 1982 (chancellors W. Brand and H. Schmidt) and from 1998 to the present (chancellor G. Schröder). Although German politics at the federal level were most of the time predominated by conservative-liberal governments, they still seem to be influenced more by social democratic ideologies and beliefs than Canadian politics. Recent studies indicate that an important correlate of attitudes and values at the political-institutional level is the type of welfare state regime (Svallfors 1997; Andreß/Heien 2001). Referring to the wellknown typology developed by Esping-Andersen (1990), three major welfare state or policy regimes are distinguishable: the liberal, the conservative and the social-democratic types. Based on Esping- Andersen s typology, Canada is classified as a liberal policy regime, while Germany represents the conservative type. Liberal welfare state regimes are characterised by a strong reliance on the market, where the role of the state as a provider of welfare is limited to the supply of a basic security net to those worst off. In contrast, conservative welfare state regimes are characterised by a more active role of the state as a provider of welfare. Welfare state benefits are mainly based on insurance systems which to a large degree are financed by contributions of the economically active population. Thus, compared to liberal welfare states, conservative welfare states are more in favour of economic regulation and state intervention, resulting in more even average distributions of income and wealth. These institutional differences between the two types of welfare state also result in pronounced differences in the amount of public social expenditures as can be seen from table 1. Accordingly, we might expect conservative welfare state regimes like Germany to be characterised by stronger egalitarian attitudes and more critical views on inequality than liberal welfare state regimes like that of Canada. American Exceptionaltism? In the last decade, Seymour Lipset (1996), among others, has re-explored de Tocqueville s notion of an American exceptionalism. The proposal is that, in relation to perceptions and evaluations of inequality, the division between Europe and the New World is almost as important as the division between post-communism and western capitalism (Delhey 1999:9). Consequently, it is plausible to examine the continental roots that distinguish Canadian and German

The Legitimacy of Inequality 11 society. On this account, there is an extensive research tradition in Canadian studies exploring the nature and extent to which Canadians beliefs, attitudes, and values are part of a North American State (e.g. Lipset 1991). An important point about this discussion is that it sees Canada in reference to the United States. This point is important for it underscores the fact that the Canadian experience is a North American experience, with all historical and cultural meaning this implies. From this perspective the core questions for our current considerations concern, first, the meanings and implications of an American exceptionalism and, second, the similarity or distinctiveness of Canada as compared to the United States. What makes America exceptional? According to Lipset the term exceptional has primarily been used to describe the United States the first new nation as a qualitatively different society and a nation not to have ideologies, but to be one (Lipset 1996:18). The core of this ideology, the American Creed, can be described by five terms: liberty, egalitarianism, individualism, populism, and laissez-faire. Egalitarianism, in its American meaning, as Tocqueville emphasized, involves equality of opportunity and respect, not of result or condition (Lipset 1996:19). This latter point is relevant to our current considerations, since this notion of equality goes together with an emphasis on personal achievement, individual mobility, and meritocracy, rather than a concern for statism and welfare policies designed to level out inequalities in living conditions and to improve the situation of less privileged groups within the society. With this specific American sense of egalitarianism is a set of individual values and attitudes corresponding, anticipating Americans to be much more individualistic, meritocratic-oriented, and anti-statist than people elsewhere (Lipset 1996:22). For our comparison, it is important to inquire about the extent to which such American exceptionalism applies to Canadians as well. Referring again to Lipset, in terms of national ideology as well as individual attitudes, Canada apparently falls somewhere in between the United States and European societies, notably Britain, but, overall, is more similar to the U.S.: Figuratively, on a scale of 0 to 100, with the United States close to 0 on a given trait and Britain at 100, Canada would fall around 30 (Lipset 1996:33). Some traits distinguishing Canada in this respect from the U.S. are a higher union density, a stronger social democratic movement, a more pronounced collectivity orientation, and a stronger support of state policies to narrow income differences. But

12 Heinz-Herbert Noll & Lance W. Roberts overall as Lipset (1996:108) concludes Canada is more like the United States than the industrialised countries across the oceans, not as Tory, not as class-conscious, and more classically liberal than these other nations. This conclusion is supported by a recent review of the literature on Canadian/American differences which concludes the consensus among Canadian sociologists is that Canadians are an awful lot like Americans in most respects (Brym 2002:4). However, as the recent work of Adams (2003) makes plain, over the last decade Canadian and American orientations are diverging in important ways. The comparison of Germany and Canada offers the specific opportunity to compare major cultural and/or institutional divisions within the two nations. For Canada, the comparison between the French speaking and culturally distinct province of Quebec and the English speaking provinces grounded in British culture is enduring. For the German case the major division is between East and West Germans, who not only lived for 40 years under contrasting economic and sociopolitical systems, but are still facing major differences in terms of living standards and social structures and thus presumably should have developed different value orientations and attitudes toward inequality. Based on our hypotheses and available results of previous studies in this field (Laczko 1996; Noll 1998), we expect the population of Quebec within Canada to express less affirmative attitudes toward inequality than the populations of the other provinces and, within Germany, we expect East Germans to be more inequality-averse than West Germans. While, for historical, political and cultural reasons Quebec is supposed to be more collectivity-oriented and statist than the English-speaking Canadian provinces, East Germany was, until the end of the 1980s, a separate socialist state under a communist government. We might therefore expect strong emphasis on equality still persisting in the value orientations and attitudes of the whole or parts of the East German population. Methodology Notes Comparing two nations as diverse as Canada and Germany requires data sets that use similar measures on representative samples of the respective populations. Our investigation relied on two data sets that meet these criteria for examining inequality issues, the World Values Survey (WVS) and the International Social Survey Programme (ISSP). The available WVS with indicators relevant to this investigation is the 1991 survey, which used a German sample of 3695 (East-Germany: 1500) and

The Legitimacy of Inequality 13 a Canadian sample of 1499 (Quebec: 369) respondents. Unfortunately the more recent WVS were not yet available for general use. The ISSP included two social inequality modules in the years 1992 and 1999. For these two years, the sample sizes for Canada were 1004 (Quebec: 275; Anglophone Canada: 769) in 1992 and 971 (Quebec: 173; Anglophone Canada: 801) in 1999, while for Germany the sample sizes were 3391 (West-Germany: 2297; East-Germany: 1094) in 1992 and 1432 (West- Germany: 921; East-Germany: 511) in 1999. 5 Table 2: Attitudes Towards Inequality Indicators Used Database Affirmative: Meritocraticliberal Critical: Egalitarianstatist Agreement / Disagreement Equality Preference and Evaluation of Distributional Outcomes Equality preference WVS + Income differences too large ISSP + Functionality of Inequality Income differences functional for ISSP + prosperity Income differences as incentive for WVS + individual achievement Norms of Distribution and Principles of Legitimation Earnings according to skill and ISSP + achievement (years of education; hard work) Earnings according to demand (family ISSP + needs; support children) Attitudes Toward Redistribution of Income through the State Responsibility of the state to reduce ISSP + income differences progressive taxation of high incomes ISSP + The WVS and ISSP surveys contain a variety of indicators related to different dimensions of attitudes toward inequality such as those discussed above. Table 2 presents an overview of the indicators to be used in our analysis allocated to the four dimensions distinguished. The first dimension, equality preference and evaluation of distributional outcomes, examines the comparative importance of equality as well as the perception of the distribution of incomes. The second dimension,

14 Heinz-Herbert Noll & Lance W. Roberts functionality of inequality, focusses on the question to which degree income differences are perceived as functional for prosperity and as incentives for individual achievement. The third dimension, norms of distribution and principles of legitimation, refers to the criteria, e.g. achievement, skill and demand, according to which individual earnings should be determined and legitimised. The final set of indicators refers to attitudes toward redistribution of income through the state. These four attitude dimensions and groups of indicators cover a broad range of topics related to inequality issues and provide sufficient scope to meaningfully compare Canada and Germany. Table 2 also provides information on the data sources for each of the indicators and moreover indicates, whether agreement (+) or disagreement (-) with a single item is being classified as meritocratic-liberal or egalitarian-statist attitudes toward inequality. Based on this classification an index is constructed to provide an overall view of how people in Canada and Germany perceive and evaluate inequality. The index includes only those indicators from the ISSP, which are available for 1992 and 1999. In our following analysis we will compare national averages for each indicator and as far as available two years of observation, 1992 and 1999. National averages for the U.S. will also be included into each chart and table for reference purposes only. Beneath national averages our study compares data for Quebec and the Anglophone Canadian provinces within Canada as well as East- and West-Germany within Germany. 6 A note of caution needs to be added concerning the comparison of regions within Canada given the nature and size of the samples used for our analysis. While for Germany the datasets used include separate subsamples for East- and West-Germany, 7 this is not the case for the Canadian subdivisons. Also the number of cases is rather small as far as the national subdivisions are concerned, particularly for Quebec. Regression analysis was applied to each of the indicators for each of the years for which data were available. Separate analyses were conducted for the Canadian and German nations, as well as for the two regions within each nation (Canada Quebec and Anglophone Canada, West- Germany and East-Germany). Within each of the analyses the effects of sex, age and subjective social status (for Germany also position on leftright scale in a separate model) were controlled for and examined. Where applicable, region (East- vs. West-Germany; Quebec vs. Anglo Canadian

The Legitimacy of Inequality 15 provinces) and nation were also included into the models in order to estimate the statistical effects of these variables. ATTITUDES TOWARD INEQUALITY: EMPIRICAL RESULTS Value Orientations and Evaluation of Income Distribution Values like equality play an important role in providing an ideal standard against which real circumstances can be assessed. Like other values, the importance of equality is neither self-evident nor independent of other considerations. To appreciate the centrality of equality as a value in a national population it is necessary to compare it to some reference point. A question from the WVS allows to measure the relative importance of and preference for equality compared to the value of freedom as a reference point. 8 The upper part of figure 1 portrays the percentage of the respective populations who placed a higher value on the option of equality over freedom. The striking feature of Figure 1 is the generally low levels of preference for the value of equality. In most cases between a quarter and a third of the population select the value of equality over that of freedom. Even among East Germans, this preference did not exist among a majority of the population. Despite their minority opinion, East Germans still express much greater preference for equality than is found among the other regions, which express substantial variation. On this account, the West Germans show the least preference for equality over freedom, which comes very close to the percentage for the U.S.. The Canadian scores are between the German scores, with Quebec respondents being more favourable toward equality than other Canadians. Closely related to the equality preference or relative importance of the value of equality is the question of how the current distribution of income is evaluated by the population. A question from the ISSP indicates whether people agree or disagree with the notion that differences in income are too large. 9 The lower part of figure 1 shows that both the Canadian and German publics are quite concerned about the current levels of income inequality in their society, a view shared by over 70% of Canadians and over 80% of Germans. Despite this overall similarity, remarkable differences are becoming evident in the regional decompositions of the national averages.

16 Heinz-Herbert Noll & Lance W. Roberts Figure 1: Value Orientations and Evaluation of Income Distribution Preference equality vs. freedom: which of the two comes closest to own opinion Canada 34 CAN - Quebec 38 CAN - Anglophone Germany - West 25 32 WVS 1991 Germany - East 44 United States 24 0 10 20 30 40 50 % consider equality more important Differences in income in r's country are too large Canada CAN - Quebec CAN - Anglophone Germany Germany - West Germany - East United States 71 71 76 79 69 69 79 1992 1999 76 84 94 98 66 77 0 20 40 60 80 100 Share "strongly agree" and "agree" in % Database: WVS 1991 and ISSP 1992 and 1999. Within Germany, there is extensive disparity between the views of East and West Germans, with almost all (94%) of East Germans reporting that income differences are too large. In Canada, regional differences also exist, with Quebec residents expressing more concern than other Canadians. It is interesting to observe, however, that concern with the level of income inequality is lessening over time. Throughout the 1990s,

The Legitimacy of Inequality 17 all regions showed declines in public concern about this issue, with the exception of English Canadians, whose attitudes remained constant. The decline is most pronounced in the U.S., where the notion that differences in income are too large in 1998 comes close to the Anglo Canadian average. Functionality of Inequality A dimension crucial for the legitimation of inequality in market economies is the functionality of inequality. This attitude dimension focusses on the question to which degree income differences are perceived as functional for prosperity and as incentives for individual achievement and thus considered as legitimised. While the first of these two aspects of functionality is addressed by a question from the ISSP, the latter aspect is addressed by a WVS-question. 10 The findings indicate that income disparities are not generally considered to be functional and thus necessary for national prosperity (figure 2). Among the Canadian and German as well as the U.S. general populations there is strong consensus on this point; only a minority of the respondents agreeing with the proposition that large differences in income are necessary for national prosperity. Nonetheless, within and between the Canadian and German populations, there are substantial differences of note about the functionality of income inequality. Surprisingly Germans are more convinced about the functionality of income differences to national prosperity than Canadians. Moreover, between regions, West Germans express considerably more sympathy for the importance of income inequality than either East Germans or Quebecois. On this account, within Canada, Quebecois are the least sympathetic for the importance of income differences to prosperity. It is worth noting, however, that the levels of support for the imperative of income inequality may be growing over time. Over the decade of the 1990s most regions showed substantial growth in support for this orientation, with the sharpest increases registered among West Germans. Only among Quebecois was there a substantial erosion in support for the view that income inequality is functional for economic prosperity.

18 Heinz-Herbert Noll & Lance W. Roberts Figure 2: Functionality of Income Differences Large differences in income necessary for r's country's prosperity Canada CAN - Quebec 17 17 10 15 CAN - Anglophone Germany 19 18 27 1992 1999 Germany - West Germany - East 21 18 14 30 United States 27 27 0 20 40 60 80 100 Share "strongly agree" and "agree" in % Incomes should be more equal (=1) vs. greater incentives (=10) Canada CAN - Quebec CAN - Anglophone Germany - West Germany - East United States 6,8 6,1 7 6,2 6,8 WVS 1991 7,6 1 2 3 4 5 6 7 8 9 10 Mean score (1-10) Database: WVS 1991 and ISSP 1992 and 1999. How do Canadians and Germans compare concerning their views whether incomes should be more equal or whether there should be greater incentives for individual efforts? The respective data from the WVS were collected in the early 1990s. What the results reveal is considerable consistency in attitude both between the countries and among the regions

The Legitimacy of Inequality 19 (figure 2). And this consistency centres on a moderate but somewhat biased attitude toward the need for norms that are likely to result in greater income inequality namely, the prevalence of greater incentives for individual effort. At the national level this prevalence is somewhat more pronounced in Canada than in Germany and within Canada more pronounced in Anglophone Canada compared to Quebec. The surprising result of a much stronger emphasis on greater incentives in East- Germany than in West-Germany and the fact that East-Germans on this account score even higher than U.S. Americans may be due to the timing of the data collection soon after the collapse of the German Democratic Republic and the by that time strong desire of the East-German population for more distinction and incentives for achievement. Norms of Distribution and Principles of Legitimation Legitimating mechanisms typically centre on allocation norms that specify who deserves which amount of ressources and rewards. Referring to certain legitimation principles, these allocation rules can be classified into those that rely on meritocratic, skill and achievement-based criteria and those that utilize other criteria. The surveys asked Canadians and Germans to rate the importance of a number of specific factors that could be used to determine peoples earnings. 11 These factors included meritocratic, skill and job related criteria such as job preparation and effort as well as nonmeritocratic criteria that would distribute earnings based on needs other than those directly related to job requirement, such as whether differential earnings should be based on whether a person has children to support or what is needed to support a family. For our purposes, we selected two meritocratic and two non-meritocratic criteria each (Tables 3).

20 Heinz-Herbert Noll & Lance W. Roberts Table 3: Importance of Meritocratic and Non-Meritocratic Criteria as Determinants of Earnings...how much people ought to earn, how important should be... years spent in education how hard the person works at the job what is needed to support a family % essential / very important whether person has children to support Germany East 1999 59 84 55 61 1992 61 79 63 65 Germany West 1999 62 77 57 67 1992 59 68 65 71 Germany Total 1999 61 78 57 66 1992 - - - - Canada 1999 59 82 52 29 Anglophone 1992 59 75 46 38 Canada Quebec 1999 40 73 65 51 1992 51 63 66 49 Canada 1999 54 80 55 35 1992 57 72 51 41 United States 1999 69 89 61 44 1992 72 90 63 50 Database: ISSP 1992 and 1999. Among the meritocratic criteria, the number of years spent in education and training is one of those most detached from specific job performance. As can be seen from table 3, between 50 and 60% of both Canadians and Germans believe that it is very important or essential that earnings reflect such investments in job preparation. These percentages are well below those observed for the U.S. On this account, however, Germans give this criterion somewhat more emphasis than Canadians do. Within Germany, there is no significant variation among regions in this perception, which have remained very stable throughout the decade. Among Canadians outside of Quebec, the importance of this criterion has remained constant over time, while respondents in Quebec have reduced the importance they give to job preparation substantially. A criterion relevant to determining earnings, which is more directly related to the actual work situation and personal achievement is how hard a person works at the job. As table 3 displays, among both populations

The Legitimacy of Inequality 21 Canadians and Germans there is a very strong consensus that this ought to be a very important determinant of earnings, with 80% of both groups rating this job feature as very important or essential. Again, the U.S. percentage is even somewhat higher. Moreover, it deserves noting that among all regional groups the importance of this criterion has risen substantially over the decade of the 1990s. Within Canada, those outside of Quebec rate this criterion more important than those inside Quebec do; while among Germans, those in the East rate job effort more highly than those in the West. Besides these meritocratic criteria for legitimising differential earnings, data on two non-meritocratic criteria stressing aspects of demand rather than achievement were gathered too. One of these examines whether differential earnings ought to be related to what is needed to support a family. On this issue slightly over half of both Canadians and Germans believe that this issue is particularly important. Among Canadians, those in Quebec with around two third of respondents reliably rate this issue as more important than those outside of Quebec, although the opinions of Anglo Canadians have been strengthening on this account over the last decade. Within Germany, this criterion is rated less strongly than a decade ago but equally so among the West and East populations. The final earning legitimation issue concerns the presence of children. Overall, it is clear that Germans consider this issue to be almost twice as important as Canadians do. Only about a third of Canadians rate this issue highly as a criterion for differential earnings, while two thirds of Germans do. Although the importance of this criterion has slackened somewhat among all Germans over the last decade, West Germans have rated this criterion slightly more highly than East Germans have. Among Canadians, those in Quebec have always considered this criterion more important than those in other regions, and their evaluation has not decayed the way it has outside of Quebec over time. To summarise, the international comparison of the importance of various criteria to legitimise inequalities in earnings present some interesting patterns. First, in general, both Canadians and Germans consider meritocratic criteria to be of greater importance to differential earnings than non-meritocratic criteria. Canadians in the Anglophone provinces obviously rate non-meritocratic criteria lower than all other Canadian and German regions included into the analysis. Moreover, the trends over time typically tend to hold for the two German regions and for

22 Heinz-Herbert Noll & Lance W. Roberts Canadians outside of Quebec. More often than not, however, the Quebec trend runs counter to the other regions. Distribution Mechanisms: Attitudes Towards Governmental Redistribution The dimension of distribution mechanisms concerns the question of how or by which constituency e.g. the market or the state - resources and rewards ought to be distributed among the population. The central issue on this account raises questions about what, if anything, ought to be done to reduce existing inequalities. In recent years consideration of such issues has regained attention in public debates and has included discussions about the role the welfare state ought to play in generating greater equality and respective policies of income redistribution. Departing from the results of previous studies we may expect rather pronounced differences between Canadians and Germans on this issues, since European countries seem to be quite different from the Anglo American democracies of the United States, Australia, New Zealand and Canada, where traditionally government intervention has been frowned upon to a much greater degree than in the heartland of the EU (Evans 1996:186). The first chart in Figure 3 considers the issue of government responsibility for reducing income inequalities 12 and shows that Germans generally have higher expectations toward the State in this regard than Canadians do. Respective expectations of U.S. Americans are even considerably lower than those of Canadians. However, these aggregate figures conceal some important variations within each society. For instance, East Germans have much higher expectations in this regard than West Germans have, even though the trend in both German regions is toward more limited expectations. In Canada, Quebec had greater expectations of government than those outside of the province on this account in the early 1990s, and this disparity has increased over time. In other words, the expectations on government to do something about income inequalities has grown in Quebec over time, while the expectations outside of Quebec have been falling during the same period.

The Legitimacy of Inequality 23 Figure 3: Expectations toward Government It is in the responsibility of the government to reduce the differences in income Canada 48 48 CAN - Quebec CAN - Anglophone Germany 42 46 55 57 65 Germany - West 52 66 Germany - East United States 35 39 76 89 0 20 40 60 80 100 Share "strongly agree" and "agree" in % People with high incomes should pay a larger share of income in taxes Canada CAN - Quebec CAN - Anglophone Germany Germany - West Germany - East United States 69 76 80 78 66 75 80 79 88 83 65 74 94 0 20 40 60 80 100 Share "larger share" and "much larger share" in % 1992 1999 Database: ISSP 1992 and 1999. Attitudes toward the issue of progressive taxation as a way to manage income inequalities are reported in the second chart of figure 3. 13 On this issue strong majorities of both Canadians and Germans believe that the wealthy should be paying either a larger or much larger share in taxes to achieve this result. Overall, however, a stronger majority of Germans support this view than is the case for Canadians. Within Germany, East Germans show somewhat greater support for this policy than West

24 Heinz-Herbert Noll & Lance W. Roberts Germans, while Canadians in Quebec are more supportive of this approach than those in other provinces. The latter are quite similar with U.S. Americans in this respect. Over time, however, it appears that the substantial levels of support for progressive taxation are eroding somewhat in all regions, with the exception of Quebec. To summarise, the significance of the state and government interventions to manage income inequalities is substantial in both nations but is stronger in Germany than in Canada. Qualifying this conclusion is the observation that all regions except Quebec are beginning to show a reduction in demands on government on these accounts. A Summary View: Egalitarian-Statist vs. Meritocratic-Liberal Attitudes The measures used in our analysis, covering different dimensions of attitudes toward inequality, indicate considerable agreement, but also consistent differences in the orientations between the Canadian and German populations. Although it seems as if there was a tendency of Germans being more sensitive to and critical of inequality than Canadians, East-Germans more egalitarian than West-Germans, and Quebecois more egalitarian than Anglo-Canadians, the results vary across the specific indicators used. Thus, in order to come to a comprehensive conclusion based on our empirical data, we will synthesize the information derived from the single indicators by calculating a composite index. The index is based on the eight survey-questions/items from the ISSP used in 1992 and 1999. Calculated as the mean of the single indicator values each running from 2 to +2 index values may vary between 2 and +2 as well, where 2 denotes the egalitarian-statist extreme and +2 the meritocratic-liberal extreme. Figure 4 displays the index values for the national averages as well as the regional subdivisions and includes as a reference also the index value for the U.S. First of all it is worth mentioning that all index values fall into the egalitarian-statist range of the scale, which means that attitudes in favour of such a view are generally more prevalent than meritocraticliberal attitudes. The index values moreover clearly confirm the hypothesis, that egalitarian-statist attitudes are more pronounced in Germany than in Canada. As expected the index value for Canada falls somewhere between the German and U.S. value, which forms the extreme case in this respect. The data are also congruent with our hypothesis that East Germans are more egalitarian-statist than West-Germans and Quebecois more than Anglo Canadians. Surprisingly the regional