PropertyValue
?:linkContext
  • Concerns about the damaging effects of a deep economic crisis on the viability of democracy are particularly pertinent in the light of the European experience during the Great Depression of the early 1930s when democracy collapsed in several European countries and was replaced by authoritarian or totalitarian regimes. It is therefore not surprising that the global economic and financial crisis that started in 2008, the Great Recession, has revived those memories, especially since some of the repercussions of the current crisis are similar: Rising levels of unemployment, widespread public protests, increasing voter volatility, dwindling support for traditional parties and a rise of populist ones. The comparative analysis of democratic stability and breakdown in interwar Europe by Berg-Schlosser and colleagues confirms these similarities. After the crash of the stock-market on October 4, 1929, in New York, "per capita income declined, industrial production fell and unemployment rose sharply" . "The sober economic figures meant severe suffering and outright misery for families and individuals affected by the crisis" . Even though there is no hard empirical evidence from public opinion surveys for that time, historical studies indicate the existence of widespread political dissatisfaction. The number of violent demonstrations increased dramatically, while electoral data show a rise of protest parties and a high degree of voter volatility (for Germany . However, although all 15 European democracies included in Berg-Schlosser's study suffered from economic distress and political conflicts, democracy survived in eight of the 15. The analysis identified seven explanatory factors that were responsible for the survival or breakdown of democracy: Historical background, the impact of the post-World War I crisis, the effects of the intermediate period , the economic impact of the Great , the social and electoral reactions after 1929 as well as the behavior of individual political actors at critical turning points. Among the survivor cases, the Netherlands, Sweden, and the United Kingdom stand out as longstanding stable democracies where the economic crisis did not produce any serious destabilization of democracy. Albeit an at the time recent democracy in a rather poor country, Ireland was not much affected by the economic crisis and also had a staunchly prodemocratic prime minister who was able to hold antidemocratic movements at bay. Belgium and France both experienced a rise of sizeable antidemocratic political movements, which were not strong enough, however, to endanger the survival of democracy. In Czechoslovakia and Finland the situation was even more critical, but democracy was saved through vigorous prodemocratic interventions by their presidents. Democracy broke down in the other seven democracies. Hungary is the only one of them in which the economic crisis did not contribute to the breakdown. Germany was the most notable breakdown case where the soaring unemployment following the economic shock of 1929 enabled Hitler's ascent to power in January 1933. Likewise, the social and electoral reactions to the Great Depression played a role in the democratic breakdowns of Austria, Estonia, Greece, Romania, and Spain (Berg-Schlosser 2012, 54). The situation in today's democracies differs in many important respects, however. First, and most important, the Great Recession was much less severe and could be overcome must faster than the Great Depression. This was primarily due to the fact that the available economic policy tools for dealing with economic crises have become more elaborate. Moreover, since World War II global institutions have been created that allow for swift international responses to such crises. Most democracies also have established welfare policies that contribute to alleviating the economic hardships associated with unemployment, inflation, or other crisis-related consequences for the livelihood of the less well-to-do. Additionally, support for democracy is more deeply rooted in the political culture of most European democracies than it was in the 1930s. Last but not least, most of today's political elites are convinced that democratic institutions are the only acceptable way of governing. These factors have not only contributed to easing the economic downturn associated with the recession and bringing most European countries quickly back onto the path of economic recovery, but also have prevented economic dissatisfaction from evolving into anti-democratic mass protests. However, the political risks associated with a continued economic downturn are far from over yet. This is particularly true for Greece, 2 but also for some other major countries around the globe (e.g., Brazil and Russia). Moreover, the current refugee crisis adds to the concerns of citizens about the impact of the rapid social and economic changes on their societies and their personal well-being. Elites are crucial actors both for democratic transition and for the consolidation of new democracies . The survival of democracy in the interwar period confirmed that the stability of democracy during economic and international crises involved a mixture of favorable socio-economic conditions, a supportive political culture, and accommodative elite behavior. The following analysis will focus primarily on the latter aspect. It will analyze the economic perceptions and democratic value orientations of members of parliament (MPs) in five young and two established democracies 3 to find out whether the economic and financial crisis has produced conflicting elite perceptions of its causes and its impact. It will start out by studying the assessments of how the recession has influenced the economic situation of their country, how they perceive the effectiveness of their country's government in coping with the crisis, their own and their party's positions on economic policy issues as well as other political orientations. Next, the relationships between these perceptions, evaluations, and policy preferences will be analyzed to determine the factors that influence them: Objective facts, membership in governing vs. opposition parties, or party ideology. Finally, it will be asked whether 2 A recent report published by the World Bank shows that the recession is still ongoing in a number of countries. The report warns for instance that the economic and financial stress in Greece "presents a risk to the regional outlook." It also mentions the risk that spillovers from the European periphery could raise sovereign and corporate risk premiums and force countries to tighten fiscal policy even further. "Persistently high unemployment and anemic investment could undermine potential longer-term growth" (2015, 38-9). 3 The five new democracies (re)democratized around 1990 while the consolidation of democracy in Sweden and Germany was achieved much earlier. Background information on the project and the survey data used for analysis are provided in Klingemann and Hoffmann-Lange in this issue. the crisis has had an impact on the legitimacy of democracy among MPs and citizens of these countries. The analysis is primarily based on data of a comparative survey of parliamentarians in the seven democracies, conducted in 2013 (CMP 2013). The advantage of the data is that many of the MPs' political orientations can be compared to the results of a previous MP survey conducted in 2007, before the onset of the crisis. The changes between 2007 and 2013 can show if the crisis has had substantial effects on these orientations. Finally, the political orientations of the MPs can also be compared to the results of the 5th and 6th waves of the World Values Survey (WVS), thus allowing additional comparisons of the changes in the political orientations of MPs and voters. The changes between the years before the onset of the crisis and the data collected between 2010 and 2013 do not allow a direct assessment whether changes in political orientations can be considered as being caused by the economic crisis, however, because the earlier surveys included only a few questions on economic policy attitudes. Nevertheless, they shed light on changes in politically relevant value orientations and support for democracy over the last decade. Therefore, they can show if consensus on democratic values has declined. They also provide information on the degree of polarization between the party-political camps and if it has increased since 2007. 5 The questionnaire included four items asking for the respondents' evaluations of the present economic situation in their country and their expectations for its future economic prospects. Another five items dealt with the perceived impact the crisis had on the country in general, the economy, the banking system, the standard of living, and democracy. A principal components analysis of these perceptions was performed to determine if they could be combined into one or two indices. The analysis revealed a two-dimensional structure confirming that the perceptions of the economic situation and those regarding the impact of the crisis loaded on two statistically independent factors. Only the item asking for the perceived impact of the crisis on the country's democracy showed very low correlations with the other impact items and was therefore excluded from fur-ther analysis. 6 Without this item, the two factors accounted for nearly 73% of the total variance (34.9% plus 37.7%). The reliability coefficient for the two sets of items was rather high with a Cronbach's α of .84 for the economic situation and α = .89 for the impact of the crisis. Separate analyses for the seven countries confirmed the overall pattern. Only one of the country-specific reliability coefficients was slightly below α = .70. Based on these results, two indices were constructed and will be used for the analysis. 7 The scores for both indices vary considerably across the seven countries and do not show any uniform pattern. However, at the aggregate level there is an obvious connection between the evaluations of the current economic situation in the MPs' countries and the perceptions of the impact of the crisis. tries that received the highest ratings for the current economic situation (Chile, Germany, and Turkey), also have the lowest averages for the perceived impact of the crisis. provides macro data by the World Bank on GDP per capita and the growth impact of the recession in the seven countries that allow to compare how well the subjective perceptions match the objective data. 8 It shows that the perceptions of the MPs and the economic facts are more or less unrelated. While South Korea (henceforth simply denominated as Korea) is the third-richest country in our group, the Korean MP's evaluations of the current Korean economy is the lowest. Likewise, the scores of the Swedish MPs for the economic situation in their country are lower than those of their counterparts in Chile and Turkey, although Sweden is the eight-richest country in the world and the richest among the seven countries in our study. The same is true for the perceived impact of the crisis. Although Poland was hardly hit by the crisis, 9 Polish MPs perceived the crisis as a great setback for the country, while Turkish MPs were less pessimistic about the impact of the crisis. At the same time, Turkey was the country most affected by the crisis among our seven countries. These results confirm that political perceptions are not only influenced by facts, but that they also reflect the situation in relevant reference countries, information by the media, other sources of information and last but not least 'wishful thinking' . Likewise, analyses of the influence of economic perceptions on voting have shown such perceptions to be strongly influenced by preexisting ideological orientations and the party preference of the voters . Therefore, one should not expect that the MPs base their ratings on a global comparison of economic well-being, but rather by comparing the situation in their own country with that in neighboring or economically similar countries, with the past performance of the economy and their personal expectations. includes information on the relationships among the different economic perceptions. As was mentioned before, the perceptions of the economic situation and the perceptions of the impact of the crisis loaded on two independent factors. This implies that they were not related to each other. The low correlation coefficients in five of the seven countries confirm this expectation. However, the very high negative correlation coefficient for Turkey and the fairly high coefficient for Korea show that respondents who evaluated the economic situation more positively were at the same time inclined to perceive a smaller impact on the economy in their country. A number of additional questions allow to determine the degree to which the perceptions of the economic situation of one's country are related to other economic perceptions. Neoliberal economic policies are frequently blamed for having been responsible for the crisis. Likewise, the increase in economic inequality is assumed to be a result of the crisis. The latter assumption is based on the fact that a large number of financial institutions -primarily failing investment banks and insurance companies -had to be bailed out with taxpayers' money. This strained public budgets and required governments to introduce austerity policies that involved tax hikes and cut-backs in welfare programs. Such policies have fueled protest movements directed against neoliberalism and the irresponsible actions of financial institutions as well as demands for the introduction of government programs to ease economic hardships and to stimulate the economy. 1 Perceptions of the current economic situation of the country ('very bad' to 'very good'), of the impact of the crisis on the country ('no negative impact' to 'severe negative impact'), and perception that the crisis was a consequence of neo-liberal economic policies ('did not play any role' to 'were primarily responsible') were measured on a 10-point scale. Perception of income differences due to the crisis was measured on a 3-point scale ('remained the same,' 'somewhat larger than before,' 'much larger than before'). The correlation coefficients for the perception that neoliberalism was responsible for the crisis indicate an inverse, albeit only weak relationship with the evaluation of the economic situation in the country. The better the evaluation of the economy, the less MPs tend to blame the crisis on neoliberalism. Likewise, the perception that income differences have increased is negatively related to the perception that the economy is doing well in six of the seven countries, with the exception of South Africa (r = -.04). The correlation coefficients in the other countries vary between a low r = -.22 in Korea and a high r = -.71 in Turkey. The results show that both assumptions are not universally shared by the MPs in our seven countries. It can be expected that parliamentarians whose party participates in the sitting government hold more favorable perceptions of the economic situation than parliamentarians belonging to opposition parties. This is a matter of cognitive consonance because otherwise they would have to concede their own party's inability in handling economic policy. They also need to assure citizens that things are not so bad as long as they remain in charge of the government. At the same time, MPs of center-left parties can be assumed to hold more pessimistic perceptions of the economic situation because these parties tend to be more critical of capitalism and free markets. 10 This should be even more pronounced after an economic recession that has depleted public budgets due to costly bail-out measures for saving failing financial institutions. Therefore, participation in government of the MP's party and the ideological position of the MPs' party are the two main factors that can be expected to explain the economic perceptions of the MPs. provides the correlation coefficients for both explanatory factors. Government participation is a straightforward dummy variable based on factual information, 11 while the ideological position of the political parties in the seven countries is based on the parliamentarians' average ratings of the position of their party on a 10-point left-right scale. 12 confirms that government participation has the expected effect in all seven countries. However, the correlations coefficients vary considerably and range from r = .51 in Germany to r = .89 in Turkey. illustrates this effect. MPs belonging to the governing parties indeed provided more favorable ratings of the economic situation of their country than those of the opposition parties. They obviously tended to evaluate the actual economic situation on the basis of what the government could achieve under adverse global conditions, while the opposition parties tended to demand more. The governmentopposition difference is particularly pronounced in Poland, South Africa, and Turkey. 3 Index perception of the economic situation measured on a 10-point scale from very bad to very good. The correlation between party ideology and the evaluation of the economic situation is only a modest r = .23 instead. shows the expected effect for five countries, while Poland as well as South Africa are deviating cases where the sign of the correlation coefficient is reversed. This is due to the fact that the MPs of the main right-of-center party (the PiS and the DA) that were in the opposition evaluated the economic situation more pessimistically. This confirms that participation in government is definitely the more important explanatory factor. The South African case also indicates that the two explanatory variables can create a cross-pressure situation for leftist parties in government (the ANC in South Africa) and rightist opposition parties. The Polish case is more complicated since Poland does not have a strong party of the left. Both the PO and the PiS are centrist parties with respect to their positions on economic policies. To determine the relative influence of government participation and party ideology on the evaluations of the economic situation, a multivariate analysis is required. This is impossible for three of our seven countries, however, because of the near perfect intercorrelation of the government-opposition and the party ideology variables in these countries. In 2013, Chile, Germany, and Korea were governed by center-right parties, while the leftist parties were in the opposition. Therefore, it is impossible to determine the relative importance of the two predictors in these three countries. In the four other countries, the opposition was split in 2013. In Sweden, the government was made up of four center-right parties (Moderates, Liberals, Christian Democrats, and the Center Party), while three center-left parties (Social Democrats, the Left Party, and the Greens) as well as the right-wing Sweden Democrats constituted the opposition. The figures in confirm that government participation is a slightly stronger predictor of economic perceptions than party ideology (r = .64 vs. r = .61), primarily because the evalua- Opposition tions of the Sweden Democrats were as low as those of the other opposition parties. In Poland, a center-right government made up of the market liberal PO and the centrist agrarian PSL was in power, while the opposition was made up of the nationalist PiS as well as the conservative Solidarna Polska and two centerleft parties (the social democratic SLD and the left liberal Ruch Palikota). The ratings by the PiS parliamentarians were even more negative than those by the leftist opposition parties. The AKP government in Turkey faced an opposition made up of the leftist CHP, the Kurdish BDP as well as the nationalist MHP. The governmentopposition divide has a much stronger influence than the left-right dimension, too, and the ratings of the economic situation by the MPs of both the leftist CHP and the rightist MHP were equally low. In South Africa, finally, the leftist ANC governed alone, while the opposition included the liberal DA, the ethnic Inkatha Freedom Party, and COPE, a splitoff from the ANC. Here, the correlation coefficient between party ideology and perception of the economic situation is even reversed with r = -.27. The figures in provide information for a better understanding of the South African results. First, the means for the four South African parties on the left-right continuum show that all four of them are left-of-center parties. Therefore, the distinction between Left and Right is less relevant in the South African party system. This is also confirmed by the much higher standard deviations in the ratings of their party's ideological position within the South African parliamentary parties. The results reveal considerable differences between the MPs of the ANC and those of the other three parties with respect to their perceptions of the economic crisis. The three smaller parties are much less optimistic in this respect. This indicates that wishful thinking predominates among the MPs of the ANC and overrides the normal pattern of left-wing skepticism vis-à-vis a free market economy. Multiple regression analyses could be performed only for the four countries in which government participation and party ideology were sufficiently distinct variables. In the joint analysis of all four countries, only the influence of gov-. . . . .65 . . . . . . .82 ernment participation has a significant effect on the perceived situation of the national economy. The model explains 55% of the total variance and increases the likelihood for a positive perception by .38. The country-by-country results are similar. Party ideology does not significantly influence the perceptions of the economic situation of the country in any of the four countries. confirms that the opposition parties differ only marginally in their evaluations of the economic situation, regardless of their own economic policy preferences. The question of how much the legitimacy of political regimes depends on economic success has been a central concern of political philosophers and social scientists for a long time. It is intuitively plausible that their economic wellbeing is of utmost importance to citizens and determines their evaluations of government performance more than most other policies. Even though governments have only limited powers to stimulate economic growth and prosperity, they may severely damage the long-run economic prospects of nations by stifling economic productivity . Because of their complexity and their only indirect effects on the economic behavior of individuals, however, the effects of economic and social policies on the overall performance of a national economy are difficult to predict and therefore controversial. This is especially true in times of economic crisis when it is unclear whether decisions to save failing banks, to provide subsidies to ailing industries, or to stimulate customer demand will only prolong existing structural problems rather than restore economic growth, while they adversely affect different social groups and may fuel political protest. The major advantage of a democratic constitution is that it allows to differentiate between the performance of the current government and the democratic political system, thereby shielding the democratic institutions from being blamed for poor governmental policies. If the interplay among government and opposition works well, we should expect a considerable differential between the performance ratings of MPs of both party camps. confirms this expectation. With the exception of Chile and Korea, the great majority of the MPs belonging to the parties in government rated the performance of their country's government in fighting the crisis as competent, while the ratings of the MPs of the opposition parties were much lower. In Korea, not only the oppositional MPs were dissatisfied with the government's handling of the crisis, however, but even less than half of the MPs of the governing party believed that the government had performed well. Chile stands out with only minor differences between government and opposition parties. What do our data say about other economic policy attitudes of the MPs in the seven countries? The assumption that the crisis was caused by neoliberal policies is fairly widespread. The overall mean is .63 and the country scores range from .49 (Sweden) to .73 (Chile). At the same time, conventional wisdom holds that parties of the right are more inclined to favor neoliberal economic policies than those of the left, even though some left-of-center parties such as the in the UK and the Schröder government in Germany pursued neoliberal policies after they came to power. This has to be understood as a concession to the neoliberal Zeitgeist prevailing during the two decades before the beginning of the recession rather than being caused by their participation in government. 14 A relationship between party ideology and the perception that neoliberalism was responsible for the crisis can be found only in the three countries where the traditional class cleavage is strongly an-chored in the party system, in Chile, Germany, and Sweden. In the other countries, the differences between the MPs belonging to the two ideological camps are much smaller. In Korea and South Africa they are even negligible. shows that more than 60% of the MPs in each of the countries -in South Africa even close to 80% and in Chile more than 90% -supported the statement that income differences had increased during the crisis. Overall, less than a fourth of the MPs claimed that the differences had not increased, while one third even believed that they had become much larger. This raises the question whether such a violation of the principle of distributive justice has produced a decline in the satisfaction with the government and in the way it handled the crisis. This question is especially pertinent since the perception of rising income differences coincides with the traditional class cleavage. Once again, the relative influence of government participation and party ideology on the perception of rising income disparities can only be determined for four of the seven countries in which the two variables do not coincide. Regression analyses for all four countries as well as country-specific analyses show that both factors do indeed have some impact. While one could suspect that party ideology is the more important predictor of this perception, government participation again turns out to be more important. The latter decreases the belief that income differences have increased due to the crisis by .31, while a rightist ideological position decreases it by only .05. This can be explained by the fact that some far-right parties such as the PiS in Poland, the Sweden Democrats, and the MHP in Turkey answered the question affirmatively. Therefore, it can be concluded that the relationship between concerns about rising income inequality and party ideology is curvilinear. The MPs of centrist parties tended to take more moderate positions, while the MPs of the more radical parties of the left and right tended to claim that economic inequality is on the rise. The data thus support the claim that economic crises provide a fertile ground for populists of both persuasions to project themselves as fighters against the disastrous effects of globalization and neoliberalism and as protectors of citizens who -so they claim -have been neglected by the centrist political parties. Changes in economic policy attitudes of the MPs can be studied by comparing the results of the CMP 2013 survey with those of the previous survey of 2007. The latter had included a number of general questions regarding economic policies that were again asked in 2013. While it is of course not possible to interpret such changes as direct effects of the economic crisis, they provide at least indirect evidence. shows the results for a question asking whether the respondents preferred more income equality over larger income differences as incentives for individual effort. This question taps the socio-political class cleavage between left and right which is quite pronounced in Chile, Germany, Korea, Sweden, and Turkey, but much less so in Poland and South Africa. This does not imply that there are no labor conflicts in the latter two countries, but rather that they are not rooted in their party systems. The typical pattern of a socio-political cleavage would require that the two major parties have close affiliations with either organized business or organized labor. In Poland, class is less relevant for inter-party conflicts than religiosity and the emphasis on Poland's national interest. Neither of the two major parties, PiS and PLO, have a distinctive economic pro-labor or pro-business profile. In South Africa, socio-economic disparities and race used to be closely interrelated since there was no black middle class under Apartheid. Therefore, race was the central socio-political cleavage. Even though intra-racial differences have increased considerably since democratization and occupation has become more politically salient, race continues to be the most important determinant of voting behavior for the time being ). The changes between 2007 and 2013 reveal a differentiated pattern. In Chile, the distance between parties of the left and right has somewhat narrowed because support for more income equality decreased by .11 among the center-left parties, while it remained stable among the center-right parties. In Germany, support increased in both party groups. At the same time, the distance between them remained about the same. In Korea, support increased significantly among the MPs of the liberal and somewhat among the MPs of the conservative party, thereby increasing the distance between the two political camps from .20 to .29. The results for Poland indicate stability. In South Africa, the support increased among the MPs of the center-left parties, while support among those of the center-right parties remained stable. Polarization is highest and at the same time rather stable in Sweden, where the left-of-center parties show very high support for more income equality and the right-of-center parties very little. In Turkey, support for more equal incomes increased considerably among the MPs of both party groups. At the same time, the two party camps are somewhat less polarized in Turkey than in Germany or Sweden. Thus, the pattern does not indicate a uniform trend toward more support for income equality, but rather a high degree of stability in six of the seven countries. Both surveys also included an item battery asking respondents to rate the degree to which they considered a number of fundamental political principles and economic policies to be essential characteristics of democracy on a tenpoint scale, among them respect for human dignity, protection of minorities, and unemployment benefits. A principal components analysis of these items revealed a two-dimensional structure. Five items dealing with an active eco- nomic role of government showed high factor loadings on one factor and a good scalability (Cronbach's α = .83). The index is based on an affirmative answer for the following items: -Governments tax the rich and subsidize the poor -People receive state aid for unemployment -The economy is prospering -Basic necessities like shelter, food, and water for everyone -Jobs for everyone These items were combined to an index "Active economic role of government" with a range from 0 to 1. The index scores indicate broad support for social welfare policies in all countries with the exception of Poland (only in 2007) and Sweden. Since it can be assumed that support for welfare policies is higher among parties of the left, the means in are again broken down by the dichotomized party ideology variable. While the differences go into the expected direction, the polarization between the MPs of the left and the right is considerably lower than for the question on income equality. The scores show a high degree of stability over time, indicating that these items tap a basic political orientation. Legitimacy theories distinguish between different objects of legitimacy as well as different types of political support, specific or diffuse . The distinction between support for authorities, for the existing institutional structure, and for the political regime, provides a theoretical basis to study the impact of the current economic crisis . This distinction involves a hierarchy of objects of political support. At the lowest level, the electoral mechanism allows voters to withdraw electoral support if they are dissatisfied with the performance of their government. This mechanism is designed to shield democracies from a loss of legitimacy in situations of perceived deficits in government performance. Economic crises that involve a decline in standards of living may have adverse effects on political support, especially if a crisis persists and government policies to cope with it turn out to be ineffective. While dissatisfaction with the economic situation can be expected to first affect specific support for the current government, in the long run such economic dissatisfaction may spill over and lead to an erosion of diffuse support. This is especially true for new democracies where these democratic institutions are not consolidated yet. Therefore, argued that long-standing democracies with a history of well-functioning democratic institutions and a satisfactory level of economic security are better able to withstand economic crises than young democracies that have not had a chance to prove their political and economic effectiveness. This raises the question whether the third-wave democracies have already achieved a sufficiently high degree of consolidation and a broad enough support for their democratic institutions to prevent anti-democratic movements from successfully mobilizing dissatisfaction with hardships resulting from an economic crisis by blaming them on the malfunctioning of the democratic institutions. Therefore, it can be assumed that the new democracies in our group of countries are more vulnerable than the two established democracies Sweden and Germany, where the mobilization efforts of explicitly antidemocratic movements do not find similarly fertile ground. Four indicators of legitimacy were included in the MP surveys of 2007 and 2013. The question asking for the national government's performance in dealing with the economic crisis can be interpreted as indicator of specific support for the present government. A second question asked respondents to rate how democratically their country was presently governed. A third indicator is confi-dence in the political institutions and political parties of the country. Both of the latter indicators capture evaluations of the institutional structure of the country. The fourth indicator, finally, is a dummy variable measuring support for democracy. Among the MPs surveyed in 2013, the correlation coefficients between the perceptions of the economic situation and indicators of political legitimacy provide a first indication if the data show the assumed effects of economic distress also in the present crisis. While a positive perception of the economic situation is associated with a positive evaluation of government performance (r = .67), with trust in political institutions (r = .52) and satisfaction with the current performance of democracy in the country (r = .52), it is not related to support for democracy (r = -.01). With the exception of the question asking for the evaluation of government performance, these indicators were also included in the 5th and the 6th wave of the World Values Survey. shows that -with the exception of Poland, South Africa, and Turkey -confidence in political institutions is very high among the MPs, while it is between .20 to .30 lower among the citizens. This is not really surprising since the MPs are involved in running these institutions. The results also indicate a high degree of temporal stability. It can be concluded that confidence in political institutions has not suffered much in the wake of the economic recession. The lower scores of the citizens do not necessarily imply a lower level of democratic legitimacy. They rather should be seen as resulting from the different political role of citizens most of whom are not actively involved in policy-making, but who have the constitutional right to vote governments out of office if they have failed to deliver satisfactory political outputs. This conclusion is supported by the fact that citizen confidence in political institutions is only slightly higher in the two benchmark democracies Sweden and Germany. Furthermore, the evaluations of the MPs of opposition parties are markedly lower and much closer to those of the citizens than those of the MPs of governing parties (not shown). Therefore, low scores for confidence in political institutions do not necessarily indicate a lack in legitimacy, but should be interpreted as a certain political vigilance on the side of the opposition parties, the citizens, and also the media, which is after all a precondition for the proper functioning of democracy. The results for the evaluation of the degree of democraticness of one's own country are rather similar to those for confidence in political institutions, although the means are somewhat higher for the latter question. All four surveys included a question asking respondents to rate three types of political systems on a four-point scale as very good, fairly good, fairly bad, or very bad: -"Having a strong leader who does not have to bother with parliament and elections" -"Having the army rule" -"Having a democratic political system." The average ratings for democracy by the MPs were above .90 in five of the seven countries with the exception of Poland (.85) and Korea (.87). Those of the citizens were slightly lower in Chile, German, Sweden, and Turkey with averages between .80 and .90, but considerably lower and only slightly above .65 in the other three countries. has questioned the validity of this item as an indicator of support for democracy. He argued that these ratings simply show that the term democracy tends to evoke positive associations even among respondents who do not understand its institutional implications. Diamond's verdict is supported by the fact that the ratings for an autocratic regime were fairly high as well in two of the seven countries. They reached especially high values in Turkey (.54) and South Africa (.56) while they were lowest among the citizens of the two established democracies Sweden (.30) and Germany (.24). In order to avoid this response bias, it makes sense to create a more demanding indicator of support for democracy. Therefore, a dichotomous index was constructed by subtracting the higher score for either an autocratic or a military regime from the score for democracy. Positive values indicate that democracy received a higher value than its two nondemocratic alternatives, negative or zero values indicate that this is not the case. shows that support for democracy was near universal and also stable over both waves of the survey among the MPs of all seven countries. At the same time, however, democracy achieved similarly high levels of support only among citizens in Sweden and Germany, with a slight decline between 2007 and 2013. Citizen support for democracy was considerably lower in Chile and Poland, where it ranged from 60 to 70 percent. It reached finally only 50% in Turkey and was even lower in Korea and South Africa. The low levels of support among citizens in the latter three countries indicate that many citizens of these countries do not really care if they live under a democratic or an authoritarian regime as long as the government produces satisfactory outputs. This makes these democracies more vulnerable to economic downturns and political crises. Citizen support for democracy has not declined since the mid-2000s in six of the seven countries, however. South Africa is the great exception. Here, support for democracy plummeted from 58% in 2006 to 27% in 2013. In 2013, only 72% of the South African respondents still believed that democracy was a good thing, while at the same time 60% found the idea of having an autocratic ruler appealing. This was the deepest drop in support for democracy among nine fragile democracies included in these two waves of the World Values Survey (Hoffmann-Lange 2015, 74). While it may of course indicate an only temporary dissatisfaction due to the dismal performance of the South African government, it confirms that democratic consolidation has suffered a considerable setback and that the stability of today's South African democracy depends primarily on the presence of a strong elite consensus. The fact that citizen support for democracy has remained more or less stable in the other four new democracies of our sample, confirms that democratic consolidation of these countries has not suffered in the wake of the recession, but that it has also not made any significant progress in recent years. The main result of the foregoing analysis is that the MPs' economic perceptions have had little impact on their political orientations. The perceptions of the current economic situation differ considerably between members of the parties in government and of the opposition parties. Members of the governing parties perceived the economic situation of their country more positively than those of the opposition parties. Compared to government participation, party ideology (right vs. left) shows an only minor effect on the perceptions of the country's economic situation. Party ideology plays a more important role for the perception of rising income inequalities, though, and contributes to intensifying the traditional cleavage over economic policies. The data on political legitimacy show high levels of temporal stability and considerable differences between MPs and voters. The lower scores for confidence in political institutions among citizens can be explained by the different roles of politicians and ordinary voters. The fact that support for democracy among citizens is significantly lower, finally, confirms the existence of an elite-17 It should be noted that Lekalage (2016) reports a much higher rejection rate for autocratic regime types in 2015 (80%). However the Afrobarometer survey used a different question wording. It asked if respondents were in favor of abolishing elections and parliament and letting the President decide everything instead. The broad rejection of this statement was probably caused by the widespread dissatisfaction with President Zuma rather than measuring a general rejection of autocratic rule. Another Afrobarometer question asked if respondents were willing to forego elections in favor of a non-elected government that would guarantee basic services such as safety, rule of law, housing, and jobs. The latter question was supported by 61% of the respondents. These results confirm that the wording of survey questions is of utmost importance. The second question is probably a better indicator of regime preference than the first question that primarily measures confidence in the sitting president. mass gap in political value orientations and attitudes that has regularly been found by political culture research (Hoffmann-Lange 2008). The MP-citizen differential found in support for democracy is much more pronounced in the five new democracies. While Chile and Poland have meanwhile reached levels of democratic support approaching those of consolidated democracies, citizen support for democracy is particularly low in Korea, Turkey, and South Africa. This implies that democracy in these three countries is more vulnerable when they have to cope with economic crises or have to ward off anti-democratic political movements. It seems questionable, however, whether the decline in support of democracy in South Africa can be attributed to the economic crisis. While the crisis may have aggravated the problems besetting South Africa, the decline in support of democracy can probably be primarily attributed to political factors (De Jager and Steenekamp 2019). Political dissatisfaction has tremendously increased in the 2010s. It is based on a mixture of both political and economic concerns. Crime (95%), lack of employment opportunities (92%), government corruption (84%), and poverty (82%) were the foremost problems mentioned in a 2015 national poll (Pew Center 2016). The apparent erosion of credibility of some democratic and 'legitimate' institutions is essentially twofold. One, these institutions appear not to be run effectively -often undermined by poor governance, corruption and mismanagement scandals. […] There appears to be a feeling that protective institutions such as the police and judiciary remain as hostile as they were for blacks under apartheid. It is noteworthy that the differences found between MPs of the governing and the opposition parties with respect to most of their economic evaluations and political orientations is higher in Turkey than in the other six democracies. Such a high degree of polarization at the elite level is likely to have negative effects on support for the existing democratic institutions among citizens, especially in the absence of a strong civic culture. Based on our data, it is therefore not surprising that the constitutional changes which dramatically increased the powers of the president and simultaneously weakened countervailing powers did not trigger widespread public protest. This decline in the quality of Turkey's democracy (Berg-Schlosser and Hoffmann-Lange 2019) occurred in spite of the broad support for democracy expressed by Turkish MPs in 2013. It seems that the military coup attempt of July 2016 was responsible for a dramatic surge in support for strengthening governmental powers at the expense of civil rights among Turkish elites, even though the electoral majority in favor of the constitutional changes was rather slim. The marginal impact of the economic crisis on the political orientations in the seven democracies does not disconfirm the assumption that a prolonged economic crisis may increase the vulnerability of new democracies. First of all, none of the seven countries was severely hit by the crisis. In fact, all of them had achieved a per capita income by 2013 that was higher than before 2008. Nevertheless, the fact that this crisis was less severe in the seven countries than in many others does not seriously diminish the validity of the study. The results suggest that the perceptions of the economic situation and the impact of the crisis did not mirror the actual situation. This implies that such perceptions take on a reality of their own and influence the world views and attitudes of people regardless of the actual impact of the crisis. Why Nations Fail: The Origins of Power, Prosperity, and Poverty Beyond the Running Tally: Partisan Bias in Political Perceptions Political Behavior Conditions of Authoritarianism, Fascism and Democracy in Inter-War Europe International Journal of Comparative Sociology The impact of the Great Depression on democracy Democracy under stress forthcoming. Democratic Quality and Legitimacy in the TRU Countries Democracy Under Threat: A Crisis of Legitimacy? Authoritarianism and Democracy in Europe 1919-39. Comparative Analyses. Houndmills Europe: The Third Way/Die Neue Mitte 2019, forthcoming. Political Radicalism: Responding to the Legitimacy Gap in South Africa Democracy under Threat. A Crisis of Legitimacy? The Spirit of Democracy Knowing Versus Caring: The Role of Affect and Cognition in Political Perceptions Political Psychology Deliberate Recovery. Policy, Politics, the Economic Recovery from the International Financial Crisis Taiwan Journal of Democracy A Systems Analysis of Political Life Reversing the Causal Arrow: The Political Conditioning of Economic Perceptions in the 2000-2004 U.S. Presidential Election Cycle The Journal of Politics HSR The Political Culture Paradigm The Oxford Handbook of Political Behavior David Easton: The Theory of the Political System Donatella Campus and Gianfranco Pasquino Race, Class and Underlying Trends in Party Support in South Africa Party Politics Electoral Behaviour in Deeply Divided Societies: The Case of Contemporary South Africa Marikana: a crisis of legitimacy in the institutions that form the foundations of South Africa's 1994 post-apartheid political settlement Social Dynamics Elite Foundations of Liberal Democracy Studying Elite vs Mass Opinion The Sage Handbook of Public Opinion Research The Impact of the Great Recession on Support for Democracy Taiwan Journal of Democracy Support for democracy in South Africa declines amid rising discontent with implementation From Fragmented Party Democracy to Government by Emergency Decree and National Socialist Takeover: Germany The Breakdown of Democratic Regimes: Europe Some Social Requisites of Democracy: Economic Development and Political Legitimacy American Political Science Review South Africa's emerging black middle class: A harbinger of political change? Impersonal Influence in American Politics The Public Perspective South Africans worried about crime and corruption, prioritize education Class, Race, and Inequality in South Africa HSR Economic Perceptions and Vote Choice: Disentangling the Endogeneity Political Behavior Global Economic Prospects. The Global Economy in Transition The World Bank Challenged Elites -Elites as Challengers Challenged Elites -Elites as Challengers. Towards a Unified Theory of Representative Elites Maurizio Cotta Vulnerability, Resilience, and Responses: The European Elites System under a Prolonged Crisis Meg Rincker Blood is Thicker than Water: Family Ties to Political Power Worldwide Elena Semenova Corporate Recruitment and Networks in Germany: Change, Stability, or Both? Jérôme Heurtaux Elites and Revolution: Political Relegation and Reintegration of Former Senior Government Officials in Tunisia Trygve Gulbrandsen Continued Elite Support for the Norwegian Version of the Nordic Model? Oxana Gaman-Golutvina Political Elites in the USA under George W. Bush and Barack Obama: Structure and International Politics Klingemann & Ursula Hoffmann-Lange The Impact of the Global Economic Crisis on Support for Democracy Ursula Hoffmann-Lange Parliamentarians' Evaluations of the Global Economic Crisis Klingemann The Impact of the Global Economic Crisis on Patterns of Support for Democracy in Germany Patrik Öhberg Not all Crises are Detrimental for the Government. The Global Economic Crisis and the Swedish Case Radosław Markowski & Agnieszka Kwiatkowska The Political Impact of the Global Economic Crisis in Poland: Delayed and Indirect Effects Shim The Global Economic Crisis, Dual Polarization, and Liberal Democracy in South Korea Yilmaz Esmer & Bahar Ayça Okçuoğlu Dimensions of Political Representation: Ideological and Policy Congruence between the Representative and the Represented in Seven Countries Parliamentarians' Evaluations of the Global Economic Crisis Ursula Hoffmann-Lange ursula.hoffmann-lange@uni-bamberg.de. Ursula Hoffmann-Lange Westendstr. 60 63128 Dietzenbach Germany Parliamentarians' Evaluations of the Global Economic Crisis 10.12759/hsr.43.2018.4.175-202 GROBID - A machine learning software for extracting information from scholarly documents Economic perceptions Great Recession government participation party ideology confidence in political institutions support for democracy political legitimacy »Einschätzungen der Wirtschafts-und Finanzkrise durch Parlamentarier«. Based on two surveys of parliamentarians in five new (Chile, South Korea, Poland, South Africa, Turkey) and two established (Germany, Sweden) democracies, the paper analyzes perceptions of the global economic crisis as well as evaluations of policies to fight the crisis and their determinants. In a second step, it will be determined if these perceptions and evaluations are related to participation in government and to the ideological positions of the political parties. Finally, it will be asked if a decline in democratic legitimacy in the political orientations of MPs and citizens can be observed in the wake of the crisis. Introduction Concerns about the damaging effects of a deep economic crisis on the viability of democracy are particularly pertinent in the light of the European experience during the Great Depression of the early 1930s when democracy collapsed in several European countries and was replaced by authoritarian or totalitarian regimes. It is therefore not surprising that the global economic and financial crisis that started in 2008, the Great Recession, has revived those memories, especially since some of the repercussions of the current crisis are similar: Rising levels of unemployment, widespread public protests, increasing voter volatility, dwindling support for traditional parties and a rise of populist ones. The comparative analysis of democratic stability and breakdown in interwar Europe by Berg-Schlosser and colleagues (Berg-Schlosser and Mitchell 2002) confirms these similarities. After the crash of the stock-market on October 4, 1929, in New York, "per capita income declined, industrial production fell and unemployment rose sharply" (Berg-Schlosser 1998, 351) . "The sober economic figures […] meant severe suffering and outright misery for families and individuals affected by the crisis" (ibid., 352) . Even though there is no hard empirical evidence from public opinion surveys for that time, historical studies indicate the existence of widespread political dissatisfaction. The number of violent demonstrations increased dramatically, while electoral data show a rise of protest parties and a high degree of voter volatility (for Germany Lepsius 1978; Berg-Schlosser 1998, 355) . However, although all 15 European democracies included in Berg-Schlosser's study suffered from economic distress and political conflicts, democracy survived in eight of the 15. The analysis identified seven explanatory factors that were responsible for the survival or breakdown of democracy: Historical background, the impact of the post-World War I crisis, the effects of the intermediate period [1924] [1925] [1926] [1927] [1928] , the economic impact of the Great Depression 1929 Depression -1934 , the social and electoral reactions after 1929 as well as the behavior of individual political actors at critical turning points. Among the survivor cases, the Netherlands, Sweden, and the United Kingdom stand out as longstanding stable democracies where the economic crisis did not produce any serious destabilization of democracy. Albeit an at the time recent democracy in a rather poor country, Ireland was not much affected by the economic crisis and also had a staunchly prodemocratic prime minister who was able to hold antidemocratic movements at bay. Belgium and France both experienced a rise of sizeable antidemocratic political movements, which were not strong enough, however, to endanger the survival of democracy. In Czechoslovakia and Finland the situation was even more critical, but democracy was saved through vigorous prodemocratic interventions by their presidents. Democracy broke down in the other seven democracies. 1 Hungary is the only one of them in which the economic crisis did not contribute to the breakdown. Germany was the most notable breakdown case where the soaring unemployment following the economic shock of 1929 enabled Hitler's ascent to power in January 1933. Likewise, the social and electoral reactions to the Great Depression played a role in the democratic breakdowns of Austria, Estonia, Greece, Romania, and Spain (Berg-Schlosser 2012, 54). The situation in today's democracies differs in many important respects, however. First, and most important, the Great Recession was much less severe and could be overcome must faster than the Great Depression. This was primarily due to the fact that the available economic policy tools for dealing with economic crises have become more elaborate. Moreover, since World War II global institutions have been created that allow for swift international responses to such crises. Most democracies also have established welfare policies that contribute to alleviating the economic hardships associated with unemployment, inflation, or other crisis-related consequences for the livelihood of the less well-to-do. Additionally, support for democracy is more deeply rooted in the political culture of most European democracies than it was in the 1930s. Last but not least, most of today's political elites are convinced that democratic institutions are the only acceptable way of governing. These factors have not only contributed to easing the economic downturn associated with the recession and bringing most European countries quickly back onto the path of economic recovery, but also have prevented economic dissatisfaction from evolving into anti-democratic mass protests. However, the political risks associated with a continued economic downturn are far from over yet. This is particularly true for Greece, 2 but also for some other major countries around the globe (e.g., Brazil and Russia). Moreover, the current refugee crisis adds to the concerns of citizens about the impact of the rapid social and economic changes on their societies and their personal well-being. Elites are crucial actors both for democratic transition and for the consolidation of new democracies (Higley and Burton 2006) . The survival of democracy in the interwar period confirmed that the stability of democracy during economic and international crises involved a mixture of favorable socio-economic conditions, a supportive political culture, and accommodative elite behavior. The following analysis will focus primarily on the latter aspect. It will analyze the economic perceptions and democratic value orientations of members of parliament (MPs) in five young and two established democracies 3 to find out whether the economic and financial crisis has produced conflicting elite perceptions of its causes and its impact. It will start out by studying the assessments of how the recession has influenced the economic situation of their country, how they perceive the effectiveness of their country's government in coping with the crisis, their own and their party's positions on economic policy issues as well as other political orientations. Next, the relationships between these perceptions, evaluations, and policy preferences will be analyzed to determine the factors that influence them: Objective facts, membership in governing vs. opposition parties, or party ideology. Finally, it will be asked whether 2 A recent report published by the World Bank shows that the recession is still ongoing in a number of countries. The report warns for instance that the economic and financial stress in Greece "presents a risk to the regional outlook." It also mentions the risk that spillovers from the European periphery could raise sovereign and corporate risk premiums and force countries to tighten fiscal policy even further. "Persistently high unemployment and anemic investment could undermine potential longer-term growth" (2015, 38-9). 3 The five new democracies (re)democratized around 1990 while the consolidation of democracy in Sweden and Germany was achieved much earlier. Background information on the project and the survey data used for analysis are provided in Klingemann and Hoffmann-Lange in this issue. the crisis has had an impact on the legitimacy of democracy among MPs and citizens of these countries. The analysis is primarily based on data of a comparative survey of parliamentarians in the seven democracies, conducted in 2013 (CMP 2013). 4 The advantage of the data is that many of the MPs' political orientations can be compared to the results of a previous MP survey conducted in 2007, before the onset of the crisis. The changes between 2007 and 2013 can show if the crisis has had substantial effects on these orientations. Finally, the political orientations of the MPs can also be compared to the results of the 5th and 6th waves of the World Values Survey (WVS), thus allowing additional comparisons of the changes in the political orientations of MPs and voters. The changes between the years before the onset of the crisis and the data collected between 2010 and 2013 do not allow a direct assessment whether changes in political orientations can be considered as being caused by the economic crisis, however, because the earlier surveys included only a few questions on economic policy attitudes. Nevertheless, they shed light on changes in politically relevant value orientations and support for democracy over the last decade. Therefore, they can show if consensus on democratic values has declined. They also provide information on the degree of polarization between the party-political camps and if it has increased since 2007. 5 Economic Perceptions of MPs in 2013 The questionnaire included four items asking for the respondents' evaluations of the present economic situation in their country and their expectations for its future economic prospects. Another five items dealt with the perceived impact the crisis had on the country in general, the economy, the banking system, the standard of living, and democracy. A principal components analysis of these perceptions was performed to determine if they could be combined into one or two indices. The analysis revealed a two-dimensional structure confirming that the perceptions of the economic situation and those regarding the impact of the crisis loaded on two statistically independent factors. Only the item asking for the perceived impact of the crisis on the country's democracy showed very low correlations with the other impact items and was therefore excluded from fur-ther analysis. 6 Without this item, the two factors accounted for nearly 73% of the total variance (34.9% plus 37.7%). The reliability coefficient for the two sets of items was rather high with a Cronbach's α of .84 for the economic situation and α = .89 for the impact of the crisis. Separate analyses for the seven countries confirmed the overall pattern. Only one of the country-specific reliability coefficients was slightly below α = .70. Based on these results, two indices were constructed and will be used for the analysis. 7 The scores for both indices vary considerably across the seven countries and do not show any uniform pattern. However, at the aggregate level there is an obvious connection between the evaluations of the current economic situation in the MPs' countries and the perceptions of the impact of the crisis. tries that received the highest ratings for the current economic situation (Chile, Germany, and Turkey), also have the lowest averages for the perceived impact of the crisis. Table 1 provides macro data by the World Bank on GDP per capita and the growth impact of the recession in the seven countries that allow to compare how well the subjective perceptions match the objective data. 8 It shows that the perceptions of the MPs and the economic facts are more or less unrelated. While South Korea (henceforth simply denominated as Korea) is the third-richest country in our group, the Korean MP's evaluations of the current Korean economy is the lowest. Likewise, the scores of the Swedish MPs for the economic situation in their country are lower than those of their counterparts in Chile and Turkey, although Sweden is the eight-richest country in the world and the richest among the seven countries in our study. The same is true for the perceived impact of the crisis. Although Poland was hardly hit by the crisis, 9 Polish MPs perceived the crisis as a great setback for the country, while Turkish MPs were less pessimistic about the impact of the crisis. At the same time, Turkey was the country most affected by the crisis among our seven countries. These results confirm that political perceptions are not only influenced by facts, but that they also reflect the situation in relevant reference countries, information by the media, other sources of information and last but not least 'wishful thinking' (Dolan and Holbrook 2001; Bartels 2002; Mutz 1992) . Likewise, analyses of the influence of economic perceptions on voting have shown such perceptions to be strongly influenced by preexisting ideological orientations and the party preference of the voters (Evans and Pickup 2010; Wlezien, Franklin and Twiggs 1997) . Therefore, one should not expect that the MPs base their ratings on a global comparison of economic well-being, but rather by comparing the situation in their own country with that in neighboring or economically similar countries, with the past performance of the economy and their personal expectations. Table 2 includes information on the relationships among the different economic perceptions. As was mentioned before, the perceptions of the economic situation and the perceptions of the impact of the crisis loaded on two independent factors. This implies that they were not related to each other. The low correlation coefficients in five of the seven countries confirm this expectation. However, the very high negative correlation coefficient for Turkey and the fairly high coefficient for Korea show that respondents who evaluated the economic situation more positively were at the same time inclined to perceive a smaller impact on the economy in their country. A number of additional questions allow to determine the degree to which the perceptions of the economic situation of one's country are related to other economic perceptions. Neoliberal economic policies are frequently blamed for having been responsible for the crisis. Likewise, the increase in economic inequality is assumed to be a result of the crisis. The latter assumption is based on the fact that a large number of financial institutions -primarily failing investment banks and insurance companies -had to be bailed out with taxpayers' money. This strained public budgets and required governments to introduce austerity policies that involved tax hikes and cut-backs in welfare programs. Such policies have fueled protest movements directed against neoliberalism and the irresponsible actions of financial institutions as well as demands for the introduction of government programs to ease economic hardships and to stimulate the economy. 1 Perceptions of the current economic situation of the country ('very bad' to 'very good'), of the impact of the crisis on the country ('no negative impact' to 'severe negative impact'), and perception that the crisis was a consequence of neo-liberal economic policies ('did not play any role' to 'were primarily responsible') were measured on a 10-point scale. Perception of income differences due to the crisis was measured on a 3-point scale ('remained the same,' 'somewhat larger than before,' 'much larger than before'). The correlation coefficients for the perception that neoliberalism was responsible for the crisis indicate an inverse, albeit only weak relationship with the evaluation of the economic situation in the country. The better the evaluation of the economy, the less MPs tend to blame the crisis on neoliberalism. Likewise, the perception that income differences have increased is negatively related to the perception that the economy is doing well in six of the seven countries, with the exception of South Africa (r = -.04). The correlation coefficients in the other countries vary between a low r = -.22 in Korea and a high r = -.71 in Turkey. The results show that both assumptions are not universally shared by the MPs in our seven countries. Participation in Government and Party Ideology as Explanatory Factors It can be expected that parliamentarians whose party participates in the sitting government hold more favorable perceptions of the economic situation than parliamentarians belonging to opposition parties. This is a matter of cognitive consonance because otherwise they would have to concede their own party's inability in handling economic policy. They also need to assure citizens that things are not so bad as long as they remain in charge of the government. At the same time, MPs of center-left parties can be assumed to hold more pessimistic perceptions of the economic situation because these parties tend to be more critical of capitalism and free markets. 10 This should be even more pronounced after an economic recession that has depleted public budgets due to costly bail-out measures for saving failing financial institutions. Therefore, participation in government of the MP's party and the ideological position of the MPs' party are the two main factors that can be expected to explain the economic perceptions of the MPs. Table 3 provides the correlation coefficients for both explanatory factors. Government participation is a straightforward dummy variable based on factual information, 11 while the ideological position of the political parties in the seven countries is based on the parliamentarians' average ratings of the position of their party on a 10-point left-right scale. 12 Table 3 confirms that government participation has the expected effect in all seven countries. However, the correlations coefficients vary considerably and range from r = .51 in Germany to r = .89 in Turkey. Figure 2 illustrates this effect. MPs belonging to the governing parties indeed provided more favorable ratings of the economic situation of their country than those of the opposition parties. They obviously tended to evaluate the actual economic situation on the basis of what the government could achieve under adverse global conditions, while the opposition parties tended to demand more. The governmentopposition difference is particularly pronounced in Poland, South Africa, and Turkey. 3 Index perception of the economic situation measured on a 10-point scale from very bad to very good. The correlation between party ideology and the evaluation of the economic situation is only a modest r = .23 instead. Table 3 shows the expected effect for five countries, while Poland as well as South Africa are deviating cases where the sign of the correlation coefficient is reversed. This is due to the fact that the MPs of the main right-of-center party (the PiS and the DA) that were in the opposition evaluated the economic situation more pessimistically. This confirms that participation in government is definitely the more important explanatory factor. The South African case also indicates that the two explanatory variables can create a cross-pressure situation for leftist parties in government (the ANC in South Africa) and rightist opposition parties. 13 The Polish case is more complicated since Poland does not have a strong party of the left. Both the PO and the PiS are centrist parties with respect to their positions on economic policies. To determine the relative influence of government participation and party ideology on the evaluations of the economic situation, a multivariate analysis is required. This is impossible for three of our seven countries, however, because of the near perfect intercorrelation of the government-opposition and the party ideology variables in these countries. In 2013, Chile, Germany, and Korea were governed by center-right parties, while the leftist parties were in the opposition. Therefore, it is impossible to determine the relative importance of the two predictors in these three countries. In the four other countries, the opposition was split in 2013. In Sweden, the government was made up of four center-right parties (Moderates, Liberals, Christian Democrats, and the Center Party), while three center-left parties (Social Democrats, the Left Party, and the Greens) as well as the right-wing Sweden Democrats constituted the opposition. The figures in Table 3 confirm that government participation is a slightly stronger predictor of economic perceptions than party ideology (r = .64 vs. r = .61), primarily because the evalua- Government Opposition tions of the Sweden Democrats were as low as those of the other opposition parties. In Poland, a center-right government made up of the market liberal PO and the centrist agrarian PSL was in power, while the opposition was made up of the nationalist PiS as well as the conservative Solidarna Polska and two centerleft parties (the social democratic SLD and the left liberal Ruch Palikota). The ratings by the PiS parliamentarians were even more negative than those by the leftist opposition parties. The AKP government in Turkey faced an opposition made up of the leftist CHP, the Kurdish BDP as well as the nationalist MHP. The governmentopposition divide has a much stronger influence than the left-right dimension, too, and the ratings of the economic situation by the MPs of both the leftist CHP and the rightist MHP were equally low. In South Africa, finally, the leftist ANC governed alone, while the opposition included the liberal DA, the ethnic Inkatha Freedom Party, and COPE, a splitoff from the ANC. Here, the correlation coefficient between party ideology and perception of the economic situation is even reversed with r = -.27. The figures in Table 4 provide information for a better understanding of the South African results. First, the means for the four South African parties on the left-right continuum show that all four of them are left-of-center parties. Therefore, the distinction between Left and Right is less relevant in the South African party system. This is also confirmed by the much higher standard deviations in the ratings of their party's ideological position within the South African parliamentary parties. The results reveal considerable differences between the MPs of the ANC and those of the other three parties with respect to their perceptions of the economic crisis. The three smaller parties are much less optimistic in this respect. This indicates that wishful thinking predominates among the MPs of the ANC and overrides the normal pattern of left-wing skepticism vis-à-vis a free market economy. Multiple regression analyses could be performed only for the four countries in which government participation and party ideology were sufficiently distinct variables. In the joint analysis of all four countries, only the influence of gov-. 23 . 16 . 52 . 34 .65 . 35 . 38 . 34 . 59 . 29 . 27 .82 ernment participation has a significant effect on the perceived situation of the national economy. The model explains 55% of the total variance and increases the likelihood for a positive perception by .38. The country-by-country results are similar. Party ideology does not significantly influence the perceptions of the economic situation of the country in any of the four countries. Figure 3 confirms that the opposition parties differ only marginally in their evaluations of the economic situation, regardless of their own economic policy preferences. Effects of the Economic Perceptions on Political Orientations The question of how much the legitimacy of political regimes depends on economic success has been a central concern of political philosophers and social scientists for a long time. It is intuitively plausible that their economic wellbeing is of utmost importance to citizens and determines their evaluations of government performance more than most other policies. Even though governments have only limited powers to stimulate economic growth and prosperity, they may severely damage the long-run economic prospects of nations by stifling economic productivity (Acemoglu and Robinson 2012) . Because of their complexity and their only indirect effects on the economic behavior of individuals, however, the effects of economic and social policies on the overall performance of a national economy are difficult to predict and therefore controversial. This is especially true in times of economic crisis when it is unclear whether decisions to save failing banks, to provide subsidies to ailing industries, or to stimulate customer demand will only prolong existing structural problems rather than restore economic growth, while they adversely affect different social groups and may fuel political protest. The major advantage of a democratic constitution is that it allows to differentiate between the performance of the current government and the democratic political system, thereby shielding the democratic institutions from being blamed for poor governmental policies. If the interplay among government and opposition works well, we should expect a considerable differential between the performance ratings of MPs of both party camps. Table 5 confirms this expectation. With the exception of Chile and Korea, the great majority of the MPs belonging to the parties in government rated the performance of their country's government in fighting the crisis as competent, while the ratings of the MPs of the opposition parties were much lower. In Korea, not only the oppositional MPs were dissatisfied with the government's handling of the crisis, however, but even less than half of the MPs of the governing party believed that the government had performed well. Chile stands out with only minor differences between government and opposition parties. What do our data say about other economic policy attitudes of the MPs in the seven countries? The assumption that the crisis was caused by neoliberal policies is fairly widespread. The overall mean is .63 and the country scores range from .49 (Sweden) to .73 (Chile). At the same time, conventional wisdom holds that parties of the right are more inclined to favor neoliberal economic policies than those of the left, even though some left-of-center parties such as the Blair government (1997 Blair government ( -2007 in the UK and the Schröder government in Germany (1998) (1999) (2000) (2001) (2002) (2003) (2004) (2005) pursued neoliberal policies after they came to power. This has to be understood as a concession to the neoliberal Zeitgeist prevailing during the two decades before the beginning of the recession rather than being caused by their participation in government. 14 A relationship between party ideology and the perception that neoliberalism was responsible for the crisis can be found only in the three countries where the traditional class cleavage is strongly an-chored in the party system, in Chile, Germany, and Sweden. In the other countries, the differences between the MPs belonging to the two ideological camps are much smaller. In Korea and South Africa they are even negligible. Table 6 shows that more than 60% of the MPs in each of the countries -in South Africa even close to 80% and in Chile more than 90% -supported the statement that income differences had increased during the crisis. Overall, less than a fourth of the MPs claimed that the differences had not increased, while one third even believed that they had become much larger. This raises the question whether such a violation of the principle of distributive justice has produced a decline in the satisfaction with the government and in the way it handled the crisis. This question is especially pertinent since the perception of rising income differences coincides with the traditional class cleavage. Once again, the relative influence of government participation and party ideology on the perception of rising income disparities can only be determined for four of the seven countries in which the two variables do not coincide. Regression analyses for all four countries as well as country-specific analyses show that both factors do indeed have some impact. While one could suspect that party ideology is the more important predictor of this perception, government participation again turns out to be more important. The latter decreases the belief that income differences have increased due to the crisis by .31, while a rightist ideological position decreases it by only .05. This can be explained by the fact that some far-right parties such as the PiS in Poland, the Sweden Democrats, and the MHP in Turkey answered the question affirmatively. Therefore, it can be concluded that the relationship between concerns about rising income inequality and party ideology is curvilinear. The MPs of centrist parties tended to take more moderate positions, while the MPs of the more radical parties of the left and right tended to claim that economic inequality is on the rise. The data thus support the claim that economic crises provide a fertile ground for populists of both persuasions to project themselves as fighters against the disastrous effects of globalization and neoliberalism and as protectors of citizens who -so they claim -have been neglected by the centrist political parties. Changes in economic policy attitudes of the MPs can be studied by comparing the results of the CMP 2013 survey with those of the previous survey of 2007. The latter had included a number of general questions regarding economic policies that were again asked in 2013. While it is of course not possible to interpret such changes as direct effects of the economic crisis, they provide at least indirect evidence. Figure 4 shows the results for a question asking whether the respondents preferred more income equality over larger income differences as incentives for individual effort. This question taps the socio-political class cleavage between left and right which is quite pronounced in Chile, Germany, Korea, Sweden, and Turkey, but much less so in Poland and South Africa. This does not imply that there are no labor conflicts in the latter two countries, but rather that they are not rooted in their party systems. The typical pattern of a socio-political cleavage would require that the two major parties have close affiliations with either organized business or organized labor. In Poland, class is less relevant for inter-party conflicts than religiosity and the emphasis on Poland's national interest. Neither of the two major parties, PiS and PLO, have a distinctive economic pro-labor or pro-business profile. In South Africa, socio-economic disparities and race used to be closely interrelated since there was no black middle class under Apartheid. Therefore, race was the central socio-political cleavage. Even though intra-racial differences have increased considerably since democratization and occupation has become more politically salient, race continues to be the most important determinant of voting behavior for the time being (Seekings and Nattrass 2005; García Rivero 2006 , 2014 Mattes 2014 ). The changes between 2007 and 2013 reveal a differentiated pattern. In Chile, the distance between parties of the left and right has somewhat narrowed because support for more income equality decreased by .11 among the center-left parties, while it remained stable among the center-right parties. In Germany, support increased in both party groups. At the same time, the distance between them remained about the same. In Korea, support increased significantly among the MPs of the liberal and somewhat among the MPs of the conservative party, thereby increasing the distance between the two political camps from .20 to .29. The results for Poland indicate stability. In South Africa, the support increased among the MPs of the center-left parties, while support among those of the center-right parties remained stable. Polarization is highest and at the same time rather stable in Sweden, where the left-of-center parties show very high support for more income equality and the right-of-center parties very little. In Turkey, support for more equal incomes increased considerably among the MPs of both party groups. At the same time, the two party camps are somewhat less polarized in Turkey than in Germany or Sweden. Thus, the pattern does not indicate a uniform trend toward more support for income equality, but rather a high degree of stability in six of the seven countries. Both surveys also included an item battery asking respondents to rate the degree to which they considered a number of fundamental political principles and economic policies to be essential characteristics of democracy on a tenpoint scale, among them respect for human dignity, protection of minorities, and unemployment benefits. A principal components analysis of these items revealed a two-dimensional structure. Five items dealing with an active eco- nomic role of government showed high factor loadings on one factor and a good scalability (Cronbach's α = .83). The index is based on an affirmative answer for the following items: -Governments tax the rich and subsidize the poor -People receive state aid for unemployment -The economy is prospering -Basic necessities like shelter, food, and water for everyone -Jobs for everyone These items were combined to an index "Active economic role of government" with a range from 0 to 1. The index scores indicate broad support for social welfare policies in all countries with the exception of Poland (only in 2007) and Sweden. Since it can be assumed that support for welfare policies is higher among parties of the left, the means in Figure 5 are again broken down by the dichotomized party ideology variable. While the differences go into the expected direction, the polarization between the MPs of the left and the right is considerably lower than for the question on income equality. The scores show a high degree of stability over time, indicating that these items tap a basic political orientation. Changes in Democratic Legitimacy among MPs and Citizens Legitimacy theories distinguish between different objects of legitimacy as well as different types of political support, specific or diffuse (Easton 1965; Fuchs and Klingemann 2009) . The distinction between support for authorities, for the existing institutional structure, and for the political regime, provides a theoretical basis to study the impact of the current economic crisis (Fuchs 2007) . This distinction involves a hierarchy of objects of political support. At the lowest level, the electoral mechanism allows voters to withdraw electoral support if they are dissatisfied with the performance of their government. This mechanism is designed to shield democracies from a loss of legitimacy in situations of perceived deficits in government performance. Economic crises that involve a decline in standards of living may have adverse effects on political support, especially if a crisis persists and government policies to cope with it turn out to be ineffective. While dissatisfaction with the economic situation can be expected to first affect specific support for the current government, in the long run such economic dissatisfaction may spill over and lead to an erosion of diffuse support. This is especially true for new democracies where these democratic institutions are not consolidated yet. Therefore, Lipset (1959) argued that long-standing democracies with a history of well-functioning democratic institutions and a satisfactory level of economic security are better able to withstand economic crises than young democracies that have not had a chance to prove their political and economic effectiveness. This raises the question whether the third-wave democracies have already achieved a sufficiently high degree of consolidation and a broad enough support for their democratic institutions to prevent anti-democratic movements from successfully mobilizing dissatisfaction with hardships resulting from an economic crisis by blaming them on the malfunctioning of the democratic institutions. Therefore, it can be assumed that the new democracies in our group of countries are more vulnerable than the two established democracies Sweden and Germany, where the mobilization efforts of explicitly antidemocratic movements do not find similarly fertile ground. Four indicators of legitimacy were included in the MP surveys of 2007 and 2013. The question asking for the national government's performance in dealing with the economic crisis can be interpreted as indicator of specific support for the present government. A second question asked respondents to rate how democratically their country was presently governed. A third indicator is confi-dence in the political institutions and political parties of the country. 15 Both of the latter indicators capture evaluations of the institutional structure of the country. The fourth indicator, finally, is a dummy variable measuring support for democracy. Among the MPs surveyed in 2013, the correlation coefficients between the perceptions of the economic situation and indicators of political legitimacy provide a first indication if the data show the assumed effects of economic distress also in the present crisis. While a positive perception of the economic situation is associated with a positive evaluation of government performance (r = .67), with trust in political institutions (r = .52) and satisfaction with the current performance of democracy in the country (r = .52), it is not related to support for democracy (r = -.01). With the exception of the question asking for the evaluation of government performance, these indicators were also included in the 5th and the 6th wave of the World Values Survey. Figure 6 shows that -with the exception of Poland, South Africa, and Turkey -confidence in political institutions is very high among the MPs, while it is between .20 to .30 lower among the citizens. 16 This is not really surprising since the MPs are involved in running these institutions. The results also indicate a high degree of temporal stability. It can be concluded that confidence in political institutions has not suffered much in the wake of the economic recession. The lower scores of the citizens do not necessarily imply a lower level of democratic legitimacy. They rather should be seen as resulting from the different political role of citizens most of whom are not actively involved in policy-making, but who have the constitutional right to vote governments out of office if they have failed to deliver satisfactory political outputs. This conclusion is supported by the fact that citizen confidence in political institutions is only slightly higher in the two benchmark democracies Sweden and Germany. Furthermore, the evaluations of the MPs of opposition parties are markedly lower and much closer to those of the citizens than those of the MPs of governing parties (not shown). Therefore, low scores for confidence in political institutions do not necessarily indicate a lack in legitimacy, but should be interpreted as a certain political vigilance on the side of the opposition parties, the citizens, and also the media, which is after all a precondition for the proper functioning of democracy. The results for the evaluation of the degree of democraticness of one's own country are rather similar to those for confidence in political institutions, although the means are somewhat higher for the latter question. All four surveys included a question asking respondents to rate three types of political systems on a four-point scale as very good, fairly good, fairly bad, or very bad: -"Having a strong leader who does not have to bother with parliament and elections" -"Having the army rule" -"Having a democratic political system." The average ratings for democracy by the MPs were above .90 in five of the seven countries with the exception of Poland (.85) and Korea (.87). Those of the citizens were slightly lower in Chile, German, Sweden, and Turkey with averages between .80 and .90, but considerably lower and only slightly above .65 in the other three countries. Diamond (2008, 31-4) has questioned the validity of this item as an indicator of support for democracy. He argued that these ratings simply show that the term democracy tends to evoke positive associations even among respondents who do not understand its institutional implications. Diamond's verdict is supported by the fact that the ratings for an autocratic regime were fairly high as well in two of the seven countries. They reached especially high values in Turkey (.54) and South Africa (.56) while they were lowest among the citizens of the two established democracies Sweden (.30) and Germany (.24). In order to avoid this response bias, it makes sense to create a more demanding indicator of support for democracy. Therefore, a dichotomous index was constructed by subtracting the higher score for either an autocratic or a military regime from the score for democracy. Positive values indicate that democracy received a higher value than its two nondemocratic alternatives, negative or zero values indicate that this is not the case. Figure 7 shows that support for democracy was near universal and also stable over both waves of the survey among the MPs of all seven countries. At the same time, however, democracy achieved similarly high levels of support only among citizens in Sweden and Germany, with a slight decline between 2007 and 2013. Citizen support for democracy was considerably lower in Chile and Poland, where it ranged from 60 to 70 percent. It reached finally only 50% in Turkey and was even lower in Korea and South Africa. The low levels of support among citizens in the latter three countries indicate that many citizens of these countries do not really care if they live under a democratic or an authoritarian regime as long as the government produces satisfactory outputs. This makes these democracies more vulnerable to economic downturns and political crises. Citizen support for democracy has not declined since the mid-2000s in six of the seven countries, however. South Africa is the great exception. Here, support for democracy plummeted from 58% in 2006 to 27% in 2013. In 2013, only 72% of the South African respondents still believed that democracy was a good thing, while at the same time 60% found the idea of having an autocratic ruler appealing. 17 This was the deepest drop in support for democracy among nine fragile democracies included in these two waves of the World Values Survey (Hoffmann-Lange 2015, 74). While it may of course indicate an only temporary dissatisfaction due to the dismal performance of the South African government, it confirms that democratic consolidation has suffered a considerable setback and that the stability of today's South African democracy depends primarily on the presence of a strong elite consensus. The fact that citizen support for democracy has remained more or less stable in the other four new democracies of our sample, confirms that democratic consolidation of these countries has not suffered in the wake of the recession, but that it has also not made any significant progress in recent years. Conclusions The main result of the foregoing analysis is that the MPs' economic perceptions have had little impact on their political orientations. The perceptions of the current economic situation differ considerably between members of the parties in government and of the opposition parties. Members of the governing parties perceived the economic situation of their country more positively than those of the opposition parties. Compared to government participation, party ideology (right vs. left) shows an only minor effect on the perceptions of the country's economic situation. Party ideology plays a more important role for the perception of rising income inequalities, though, and contributes to intensifying the traditional cleavage over economic policies. The data on political legitimacy show high levels of temporal stability and considerable differences between MPs and voters. The lower scores for confidence in political institutions among citizens can be explained by the different roles of politicians and ordinary voters. The fact that support for democracy among citizens is significantly lower, finally, confirms the existence of an elite-17 It should be noted that Lekalage (2016) reports a much higher rejection rate for autocratic regime types in 2015 (80%). However the Afrobarometer survey used a different question wording. It asked if respondents were in favor of abolishing elections and parliament and letting the President decide everything instead. The broad rejection of this statement was probably caused by the widespread dissatisfaction with President Zuma rather than measuring a general rejection of autocratic rule. Another Afrobarometer question asked if respondents were willing to forego elections in favor of a non-elected government that would guarantee basic services such as safety, rule of law, housing, and jobs. The latter question was supported by 61% of the respondents. These results confirm that the wording of survey questions is of utmost importance. The second question is probably a better indicator of regime preference than the first question that primarily measures confidence in the sitting president. mass gap in political value orientations and attitudes that has regularly been found by political culture research (Hoffmann-Lange 2008). The MP-citizen differential found in support for democracy is much more pronounced in the five new democracies. While Chile and Poland have meanwhile reached levels of democratic support approaching those of consolidated democracies, citizen support for democracy is particularly low in Korea, Turkey, and South Africa. This implies that democracy in these three countries is more vulnerable when they have to cope with economic crises or have to ward off anti-democratic political movements. It seems questionable, however, whether the decline in support of democracy in South Africa can be attributed to the economic crisis. While the crisis may have aggravated the problems besetting South Africa, the decline in support of democracy can probably be primarily attributed to political factors (De Jager and Steenekamp 2019). Political dissatisfaction has tremendously increased in the 2010s. It is based on a mixture of both political and economic concerns. Crime (95%), lack of employment opportunities (92%), government corruption (84%), and poverty (82%) were the foremost problems mentioned in a 2015 national poll (Pew Center 2016). The apparent erosion of credibility of some democratic and 'legitimate' institutions is essentially twofold. One, these institutions appear not to be run effectively -often undermined by poor governance, corruption and mismanagement scandals. […] There appears to be a feeling that protective institutions such as the police and judiciary remain as hostile as they were for blacks under apartheid. (Gumede 2015, 334) It is noteworthy that the differences found between MPs of the governing and the opposition parties with respect to most of their economic evaluations and political orientations is higher in Turkey than in the other six democracies. Such a high degree of polarization at the elite level is likely to have negative effects on support for the existing democratic institutions among citizens, especially in the absence of a strong civic culture. Based on our data, it is therefore not surprising that the constitutional changes which dramatically increased the powers of the president and simultaneously weakened countervailing powers did not trigger widespread public protest. This decline in the quality of Turkey's democracy (Berg-Schlosser and Hoffmann-Lange 2019) occurred in spite of the broad support for democracy expressed by Turkish MPs in 2013. It seems that the military coup attempt of July 2016 was responsible for a dramatic surge in support for strengthening governmental powers at the expense of civil rights among Turkish elites, even though the electoral majority in favor of the constitutional changes was rather slim. The marginal impact of the economic crisis on the political orientations in the seven democracies does not disconfirm the assumption that a prolonged economic crisis may increase the vulnerability of new democracies. First of all, none of the seven countries was severely hit by the crisis. In fact, all of them had achieved a per capita income by 2013 that was higher than before 2008. Nevertheless, the fact that this crisis was less severe in the seven countries than in many others does not seriously diminish the validity of the study. The results suggest that the perceptions of the economic situation and the impact of the crisis did not mirror the actual situation. This implies that such perceptions take on a reality of their own and influence the world views and attitudes of people regardless of the actual impact of the crisis. Figure 1 : 1 MPs' Perceptions of the Economic Situation and of the Impact of the Economic Crisis in their Country Source: CMP Survey 2013. Mean index scores, range 0-1. Figure 1 1 provides a first impression regarding the MPs' economic perceptions. Figure 2 : 2 MPs' Perceptions of the Economic Situation in Their Country by Government vs. Opposition Parties Source: CMP 2013 Survey. Mean index scores, range 0-1. . Figure 3 : 3 MPs' Perceptions of the Economic Situation in the Four Countries with Bilateral Opposition Source: CMP Survey 2013. Mean index scores, range 0-1. Figure 4 : 4 MPs' Support for 'Incomes Should Be More Equal' 2007 and 2013 Source: TRI 2007 and CMP 2013 Surveys. Mean score, range 0-1. Figure 5 : 5 MPs' Support for an Economically Active Role of Government 2007 and 2013 Source: TRI 2007 and CMP 2013 Surveys. Mean index scores, range 0-1. . Figure 6 : 6 Confidence in Political Institutions among MPs and CitizensSource: TRI 2007 and CMP 2013 Surveys; World Values Survey waves 5 and 6. Index: Mean confidence ratings for parliament, government, public service, police, and political parties on a 4-point scale, range 0-1. . Figure 7 : 7 Support for Democracy among MPs and Citizens Source: TRI 2007 and CMP 2013 Surveys; World Values Survey waves 5 and 6. Dichotomous index of evaluation of democracy minus evaluation of autocratic or military regime. Table 1 : 1 Economic Indicators and Perceptions of MPs in 2013 Chile Germany Korea Poland South Africa Sweden Turkey GDP p.c. 2007 in US $ 10,379 42,761 23,101 11,252 5,851 53,325 9,312 GDP p.c. 2013 in US $ 15,732 46,269 25,977 13,648 6,618 60,430 10,972 Rank among the 7 countries 4 2 3 5 7 1 6 Rank among 187 independent coun- 44 20 32 55 74 8 62 tries 1 Growth impact of recession 2 2.03 1.76 2.22 0.55 2.58 3.06 3.57 Rank among the 7 countries 5 6 4 7 3 2 1 Rank among 187 independent coun- 84 93 82 123 69 57 48 tries 1 Perception of the current economic situation of the .71 .73 .38 .43 .51 .56 .59 country (range 0-1) Rank among the 7 countries 2 1 7 6 5 4 3 Perception of the impact of the reces-sion on the country .38 .47 .70 .64 .55 .60 .44 (range 0-1) Rank among the 7 countries 7 5 1 2 4 3 6 Source: World Bank economic indicators and CMP survey 2013. 1 187 independent countries for which economic data are available. 2 Average growth 2000-2007 minus average growth 2008-2010. Table 2 : 2 Correlations among Economic Perceptions 1(Pearson's r) Perception of the current economic situation of the country and perception of ... Economic impact of crisis Neoliberalism responsible for the crisis Income differences have increased Chile .17 -.30 -.25 Germany -.26 -.35 -.33 Korea -.36 -.08 -.22 Poland .06 -.36 -.46 South Africa -.17 .34 -.04 Sweden -.18 -.46 -.34 Turkey -.66 -.46 -.71 Source: CMP Survey 2013. Table 3 : 3 Correlations among Government Participation, 1 Party Ideology, 2 and Perceptions of the Current Economic Situation in the Country 3 (Pearson's r) Perception of the Economic Situation of the Country and ... Based on MPs' ratings of the ideological position of their party on a 10-point left-right scale. Government Participation Party Ideology Chile .74 .71 Germany .51 .54 Korea .57 .57 Poland .74 -.29 South Africa .69 -.27 Sweden .64 .61 Turkey .89 .72 Source: CMP Survey 2013. 1 0=opposition, 1=government. 2 Table 4 : 4 Economic Perceptions of South African MPs by Party African National Congress (ANC) Demo-cratic Alliance (DA) Inkatha Freedom Party Congress (COPE) of the People Total 1 Current economic situation of the Mean .65 .33 .43 .34 .56 country (0-1) Neoliberalism responsible for Mean .71 .47 .62 .51 .65 crisis (0-1) Impact of crisis has reached its % 80.7 53.8 33.3 42.9 71.1 peak Income distribu-tion is unfair % 61.9 84.3 85.7 85.8 68.6 Income differences have become % 34.5 36.2 42.9 42.9 35.6 much larger Left-right score of party Mean 4.5 5.1 5.0 3.7 4.6 Number of Re-spondents n 63 58 7 14 142 Source: CMP Survey 2013. 1 Weighted for the parliamentary strength of the different parties. Table 5 : 5 Evaluations of the National Government's Performance in Fighting the Economic Crisis Country Evaluation Government Opposition Total poor 18.8 15.8 17.1 Chile neither 33.3 29.8 31.4 competent 47.9 54.4 51.4 poor 3.4 65.4 32.4 Germany neither 5.1 21.2 12.6 competent 91.5 13.5 55.0 poor 6.9 83.0 41.0 Korea neither 48.3 17.0 34.3 competent 44.8 0.0 24.8 poor 1.4 80.3 42.2 Poland neither 7.0 9.2 8.2 competent 91.5 10.5 49.7 poor 2.0 43.2 14.7 South Africa neither 14.1 34.1 20.3 competent 83.8 22.7 65.0 poor 0.0 41.8 21.3 Sweden neither 0.0 41.8 21.3 competent 100.0 16.4 57.4 poor 1.1 61.3 25.8 Turkey neither 0.0 32.3 13.2 competent 98.9 6.5 60.9 Source: CMP Survey 2013. Column percentages per country. Table 6 : 6 Perceptions of How the Economic Crisis Has Affected the Income Distribution in the Country by Party Ideology Party Ideology Income Differences Center-left Center-right Total unchanged 0.0 4.2 2.1 Chile somewhat larger 8.7 62.5 36.2 much larger 91.3 33.3 61.7 unchanged 13.5 51.7 33.6 Germany somewhat larger 44.2 39.7 41.8 much larger 42.3 8.6 24.5 unchanged 0.0 8.9 4.9 Korea somewhat larger 54.3 55.4 54.9 much larger 45.7 35.7 40.2 unchanged 18.2 28.4 26.8 Poland somewhat larger 36.4 50.9 48.6 much larger 45.5 20.7 24.6 unchanged 23.5 12.5 20.9 South Africa somewhat larger 41.2 50.0 43.3 much larger 35.3 37.5 35.8 unchanged 12.5 47.5 31.8 Sweden somewhat larger 33.3 50.8 43.0 much larger 54.2 1.7 25.2 unchanged 2.2 53.5 37.7 Turkey somewhat larger 28.9 33.7 32.2 much larger 68.9 12.9 30.1 Source: CMP Survey 2013. Column percentages per country and party group. Three countries in which the democracy broke down already before the beginning of the Great Depression, Italy in 1922, Poland and Portugal in 1926, were not included in the study. The data are part of a larger comparative study of democracy in these countries, which is described in Klingemann and Hoffmann-Lange (2018, in this issue). Information on the surveys, including the numbers of respondents, is provided there and will not be repeated here.5 In order to make scores comparable for scales using different ranges (e.g., 1-10 or 1-4), all scores were rescaled to a range from 0 to 1. This allows a more intuitive interpretation of the means. This confirms that democracy is perceived as not directly influenced by economic factors.7 Unless otherwise indicated all indices are mean scores of the variables included in the index. The economic data were compiled by Du Plessis, Freytag and Boshoff (2015) who also developed several measures to determine the sequence and the impact of the recession. The growth impact of the recession was measured by deducting the average GDP growth rate 2008-2010 from the average GDP growth2000-2007. The impact of the recession was only .55 in Poland which was the lowest in our seven countries. Despite its short-comings, the traditional economic left-right dimension will be used for analysis. While it is obvious that multi-party systems usually represent several conflict dimensions, the economic conflict dimension is still the most important one and the simple left-right scale used in surveys reflects this dimension rather well. 11 For Germany, MPs of the coalition by Christian Democrats and Free Democrats from October 2009 and December 2013 were classified as belonging to the governing parties, while MPs belonging to the Social Democratic, the Green, and the Left party were classified as opposition. It should be noted that a change in government to a grand coalition of Christian Democrats and Social Democrats took place in Germany on December 17, 2013, while the field work for the survey was still going on. Only a few interviews were conducted after that date, however, and statistical controls showed that including these respondents in the analysis had no influence on the results.12 Respondents were asked to rate their own ideological position, the ideological position of their party as well as that of the voters of their party on a 10-point left-right scale. The party ideology variable was constructed from the second of these scores for all parties that had been rated by at least five MPs. Only three very small leftist Chilean parties could not be classified because less than five MPs of these parties were represented in the Chilean sample. Another 12 respondents were independent deputies, five in Chile and seven in Poland. They could not be classified either. The decision for party ideology rather than personal ideological orientation was based on the assumption that the parliamentary party is an important reference group for MPs and ensures a high rate of internal information exchange about economic perceptions. Both variables are closely intercorrelated (r = .85). The differences in the means between government and opposition parties as well as between left-of-center and right-of-center parties are much smaller for the perceptions of the impact of the crisis which seem to be based primarily on idiosyncratic rather than systematic factors. Blair came to power after a sweeping reform of Labour's economic policy program and explicitly called the party "New Labour." Likewise, Schröder was frequently called "the comrade of the bosses." Both published the common memorandum "The Third Way" in 1999, emphasizing the necessity of pursuing a middle way between traditional pro welfare-statist economic policies and neoliberalism. Cf.Blair and Schröder 1999. This index is based on the mean evaluations of four political institutions (government, parliament, civil service, and police) as well as political parties. The evaluations were measured on a four-point scale and the index scores were rescaled to a range from 0 to 1.16 One could suspect that these low confidence ratings are mainly due to the inclusion of political parties in the index. But while confidence in parliament is indeed about 10 percent higher than confidence in political parties, the exclusion of the latter would not have changed the picture substantially. Why Nations Fail: The Origins of Power, Prosperity, and Poverty Daron Acemoglu James A Robinson Crown Publishers New York Beyond the Running Tally: Partisan Bias in Political Perceptions Larry M Bartels Political Behavior 24 Conditions of Authoritarianism, Fascism and Democracy in Inter-War Europe Dirk Berg-Schlosser International Journal of Comparative Sociology 39 The impact of the Great Depression on democracy Dirk Berg-Schlosser Democracy under stress Ursula van Beek; Opladen Barbara Budrich Publishers forthcoming. Democratic Quality and Legitimacy in the TRU Countries Dirk Berg-Schlosser Ursula Hoffmann-Lange Democracy Under Threat: A Crisis of Legitimacy? Ursula van Beek; Basingstoke Palgrave Macmillan Authoritarianism and Democracy in Europe 1919-39. Comparative Analyses. Houndmills Berg-Schlosser, Dirk, and Jeremy Mitchell Palgrave Macmillan Europe: The Third Way/Die Neue Mitte Tony Blair Gerhard Schröder 2019, forthcoming. Political Radicalism: Responding to the Legitimacy Gap in South Africa De Jager Nicola Cindy Steenekamp Democracy under Threat. A Crisis of Legitimacy? Ursula van Beek; Houndmills Palgrave Macmillan The Spirit of Democracy Larry Diamond Henry Holt and Company New York Knowing Versus Caring: The Role of Affect and Cognition in Political Perceptions Kathleen A Dolan Thomas M Holbrook Political Psychology 22 Deliberate Recovery. Policy, Politics, the Economic Recovery from the International Financial Crisis Du Plessis Andreas Stan Willem Freytag Boshoff Taiwan Journal of Democracy 11 A Systems Analysis of Political Life David Easton Wiley New York Reversing the Causal Arrow: The Political Conditioning of Economic Perceptions in the 2000-2004 U.S. Presidential Election Cycle Geoffrey Evans Mark Pickup The Journal of Politics 72 HSR 43 The Political Culture Paradigm Dieter Fuchs The Oxford Handbook of Political Behavior Russell J. Dalton and Hans-Dieter Klingemann Oxford Oxford University Press David Easton: The Theory of the Political System Dieter Fuchs Hans-Dieter Klingemann Donatella Campus and Gianfranco Pasquino Colchester ECPR Press masters of political science Race, Class and Underlying Trends in Party Support in South Africa García Rivero Carlos Party Politics 12 Electoral Behaviour in Deeply Divided Societies: The Case of Contemporary South Africa García Rivero Carlos Madrid: Autonoma University Working Papers Online Series 162/ Marikana: a crisis of legitimacy in the institutions that form the foundations of South Africa's 1994 post-apartheid political settlement William Gumede Social Dynamics 41 Elite Foundations of Liberal Democracy John Higley Michael Burton Rowman & Littlefield Lanham Studying Elite vs Mass Opinion Hoffmann-Lange The Sage Handbook of Public Opinion Research Wolfgang Donsbach and Michael W Ursula Traugott Sage Publications Los Angeles The Impact of the Great Recession on Support for Democracy Hoffmann-Lange Taiwan Journal of Democracy 11 Support for democracy in South Africa declines amid rising discontent with implementation Rorisang Lekalage From Fragmented Party Democracy to Government by Emergency Decree and National Socialist Takeover: Germany M Lepsius Rainer The Breakdown of Democratic Regimes: Europe Juan J. Linz and Alfred Stepan Baltimore Johns Hopkins University Press Some Social Requisites of Democracy: Economic Development and Political Legitimacy Seymour Lipset Martin American Political Science Review 53 South Africa's emerging black middle class: A harbinger of political change? Robert Mattes No. 151 Afrobarometer Working Paper Impersonal Influence in American Politics Diana Mutz The Public Perspective 4 South Africans worried about crime and corruption, prioritize education Pew Research Center Class, Race, and Inequality in South Africa Jeremy Seekings Nicoli Nattras HSR 43 202 Yale University Press Economic Perceptions and Vote Choice: Disentangling the Endogeneity Christopher Wlezien Mark Franklin Daniel Twiggs Political Behavior 19 Global Economic Prospects. The Global Economy in Transition The World Bank Washington (DC World Bank Group Challenged Elites -Elites as Challengers 43 Challenged Elites -Elites as Challengers. Towards a Unified Theory of Representative Elites Heinrich Best & Ursula Hoffmann-Lange 10.12759/hsr.43.2018.4.7-32 Maurizio Cotta Vulnerability, Resilience, and Responses: The European Elites System under a Prolonged Crisis 10.12759/hsr.43.2018.4.33-53 Meg Rincker Blood is Thicker than Water: Family Ties to Political Power Worldwide Farida Jalalzai & 10.12759/hsr.43.2018.4.54-72 Elena Semenova Corporate Recruitment and Networks in Germany: Change, Stability, or Both? 10.12759/hsr.43.2018.4.73-97 Jérôme Heurtaux Elites and Revolution: Political Relegation and Reintegration of Former Senior Government Officials in Tunisia 10.12759/hsr.43.2018.4.98-112 Trygve Gulbrandsen Continued Elite Support for the Norwegian Version of the Nordic Model? 10.12759/hsr.43.2018.4.113-140 10.12759/hsr.43.2018.4.141-163 Oxana Gaman-Golutvina Political Elites in the USA under George W. Bush and Barack Obama: Structure and International Politics Klingemann & Ursula Hoffmann-Lange The Impact of the Global Economic Crisis on Support for Democracy Hans-Dieter 10.12759/hsr.43.2018.4.164-174 Ursula Hoffmann-Lange Parliamentarians' Evaluations of the Global Economic Crisis 10.12759/hsr.43.2018.4.175-202 Klingemann The Impact of the Global Economic Crisis on Patterns of Support for Democracy in Germany Hans-Dieter 10.12759/hsr.43.2018.4.203-234 Patrik Öhberg Not all Crises are Detrimental for the Government. The Global Economic Crisis and the Swedish Case 10.12759/hsr.43.2018.4.235-249 Radosław Markowski & Agnieszka Kwiatkowska The Political Impact of the Global Economic Crisis in Poland: Delayed and Indirect Effects 10.12759/hsr.43.2018.4.250-273 Shim The Global Economic Crisis, Dual Polarization, and Liberal Democracy in South Korea Sang-Jin Han & Young-Hee 10.12759/hsr.43.2018.4.274-299 Yilmaz Esmer & Bahar Ayça Okçuoğlu Dimensions of Political Representation: Ideological and Policy Congruence between the Representative and the Represented in Seven Countries 10.12759/hsr.43.2018.4.300-340 (xsd:string)
?:linkScore
  • 0.0 (xsd:float)
?:linkSource
  • GEFUREX pipeline (xsd:string)
?:linkingMethod
  • search (xsd:string)
?:mainEntity
?:name
  • WVS (xsd:string)
is ?:referenceMetadata of
?:toSource
  • gesis (xsd:string)
rdf:type