Full Text

1. Introduction There is a great deal of literature on the causes and conditions of successful democratization. A variety of empirical studies in the post-Soviet bloc have been conducted in order to test different propositions, nonetheless due to lack of individual level data in Mongolia,the country has been neglected as a case. In The Third Wave, Samuel Huntington introduced the idea of three waves of democratization based on historical processes.1 Mongolia stands as a success story of democratization in the region, because it is considered “one of the more remarkable outliers of the post-communist universe in regards to democratization,” because it is “the only third wave democracy east of the Balkans that avoided political erosion and successfully consolidated democracy.”2 In addition to “the peaceful manner” of the process, it is also believed to be “one of the least likely cases” to undergo a successful transition to democracy. 3 1 Samuel P. Huntington, The Third Wave: Democratization in the Late Twentieth Century.(Norman: University of Oklahoma Press, 1991). 2 Verena Fritz, “Mongolia: Dependent Democratization,” Journal of Communist Studies and Transition Politics 18 (2002), 75. 3 UNDP 1997 in Richard Pomfret, “Transition and Democracy in Mongolia,” Europe- Asia Studies 52 (2000): 27. Steven M. Fish, “Mongolia: Democracy without Prerequisites,” Journal of Democracy 9 (1998): 128.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 138 Mongolia’s geographical position between Russia and China heavily influenced most of its political developments in the 20th century and continues to affect policy- making.4 However, in comparison to other post-Soviet countries in Central Asia, Mongolia managed to preserve its cultural heritage, language, and avoid “Russification.”5 Furthermore, in general, landlocked states outside Europe face the worst problems and are “uniformly poor.”6 These contribute to the “total anomaly” status according to macro-level system analysis based on the traditions of Seymour Martin Lipset and Samuel Huntington.7 In recent years, Mongolian democratic success has been widely associated with pro- democratic political culture.8 However, for most of the 20th century, the way to independence and modernity had appeared to be through Communism. Through the efforts of Russian and Mongolian Bolsheviks, in 1921 Mongolia became the second socialist state in the world, and a: testing ground for much of the Communist policy in the Third World: methods of education, cultural work, collectivization, and anti-religious propaganda that appeared later in other countries were first introduced by Soviet advisors in Mongolia,who ran the country on behalf of its Communist rulers.9 Throughout the communist period, Mongolia maintained a status similar to Soviet satellites. It wasn’t incorporated into the territory of the Soviet Union and remained a buffer state due to the Soviet Union’s geopolitical rivalry with China.10 A single- party state with the governing Mongolian People’s Revolutionary Party (MPRP) was formed, which resembled a Soviet satellite and followed a path strongly influenced by the Soviet Union, to the extent that it collapsed in a similar fashion. The leaders of the MPRP followed Mikhail Gorbachev’s glasnost’ and perestroika, which led to open calls to end the dictatorial rule of the Party and the formation of the Mongolian Democratic Union.11 The young Mongolian reformers developed programs that led to the largest demonstration in the country, and in 1990 the 4 Narangoa Li, “Mongolia and Preventive Diplomacy: Haunted by History and Becoming Cosmopolitan,”Asian Survey 49 (2009). 5 Oxford Business Group, The Report: Mongolia 2012, 8-9. Odd A. Westad, The Global Cold War: Third World Interventions and the Making of Our Times.(Cambridge: Cambridge University Press, 2006), 40. 6 Pomfret, “Transition,” 32. 7 Paula. L. W. Sabloff, “Why Mongolia? The Political Culture of an Emerging Democracy,”Central Asian Survey 21 (2002): 19. 8 Fish “Mongolia”; Sabloff “Why Mongolia?”; Ganbat “The Mass Public and Democratic Politics in Mongolia;” OBG, “The Report: Mongolia 2012”. 9 Westad, The Global Cold War, 51. 10 Alan M. Wachman, “Mongolia’s Geopolitical Gambit: Preserving a Precarious Independence, While Resisting “Soft Colonialism,” EAI Fellows Working Paper 18 (2009). 11 OBG, “Mongolia 2012,” 13.CEU Political Science Journal. Vol. 8, No. 2 139 MPRP resigned and the first democratic elections were held.12 Multiparty elections were introduced in the 1990s, and in 2007 the government officially declared the democratic transition to be complete. The constitution of 1992 introduced a semi-presidential form of government, which resulted in constant power struggles between the office of the president and the parliament. The president is directly elected by popular vote, but his power is severely limited by the parliament, to which he is directly accountable. The prime minister is elected by the parliament and is also accountable to it. This creates a sort of system of checks-and-balances between the institutions, where there is much overlap between the offices of the president, the prime minister, and the parliament.13 This is considered one of the institutional strengths which prevented Mongolia from the drift into authoritarianism seen in former Soviet states in Asia.14 Moreover, the country has consistently been ranked as democratic and free by foreign observers.15 In summary, Mongolia’s location between China and Russia historically limited its foreign policy options and resulted in a focus on preserving sovereignty and avoiding dependence on either neighbor.16 In addition, the absence of a sufficiently strong “national father figure” in the executive who could monopolize power during transition has also contributed to Mongolia’s success.17 The efficiency of international donor contributions during the transitional period and especially during the systemic crisis in the late 1990s is also an important factor to consider.18 However, availability of data makes it possible to test whether suggestions that Mongolia is an outlier are justified. In this work, I would like to examine another significant but hitherto largely neglected aspect which contributed to the successful transition and consolidation of democracy, despite all the favorable and unfavorable developments, which can described as the Mongolian “critical mass.”19 12 Most prominent of Mongolian young elites were educated in Moscow and Eastern Europe, which consequently echoed the transition processes of those regions. See Morris Rossabi. Modern Mongolia: From Khans to Commissars to Capitalists. (Berkley: University of California Press, 2005). 13 Sumati Luvsandendev, “Mongolia.” KAS Democratic Report(2009), 81-82. 14 Fish, “Mongolia.” 15 Freedom House, Polity IV, UNDP. 16 Narangoa, “Mongolia and Preventive Diplomacy” and Wachman “Mongolia’s Geopolitical Gambit.” 17 Steven M. Fish, “The Inner Asian Anomaly: Mongolia’s Democratization in Comparative Perspective,”Communist and Post-Communist Studies 34 (2001): 329. 18 Fritz, “Mongolia: Dependent Democratization.” 19 Pippa Norris, ed. Critical Citizens: Global Support for Democratic Governance. (Oxford: Oxford University Press, 1999).Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 140 2. The Revised Modernization Theory In order to do this, a theoretical framework will be built on general modernization theory, which holds that economic aspects matter to democratization. The term modernization entails a number of concepts, generally indicating a shift from a traditional to modern society. In the context of Mongolia, the focus is on the aspect of modernization that is relevant to developing countries and the post-Communist bloc in their efforts to reach the level of developed countries. The modernization theory received empirical support through the work of Seymour Martin Lipset, who established a link between the level of development of a given country and its probability of being democratic.20 Moreover, “[i]t is considered as one of the best established correlations; however, causes of this relationship are debatable.”21 In the original study, the patterns between averages of economic development indicators in European, English-speaking and Latin American countries allowed Lipset to conclude that “the more well-to-do a nation, the more likely it will sustain democracy.”22 The more recent developments of modernization theory can be divided into two main branches: the theory of democratic culture represented by Inglehart and Welzel, and the theory of economic development represented by Przeworski and Limongi.23 The theory of democratic culture holds that a key prerequisite of democracy is support for democratic norms and the associated behavior among citizenry.24 For Inglehart and Welzel, the general public’s democratic values are the appropriate method of indicating the prospects of consolidating democratic governments. In contrast, in the theory of economic development, modernization theory was tested on time-series analysis of cross-sections and it was concluded that modernization doesn’t necessarily bring democracy. In their methods the authors suggested that one of the main indicators of economic development should be per capita GDP as a good predictor of the stability of democracies.25 And most notably the findings demonstrated that economics play a crucial role in democratic survival.26 However, in a later work, it was clarified that although the level of 20 Julian Wucherpfennig and Franziska Deutsch, “Modernization and Democracy: Theories and Evidence Revisited,” Living Reviews in Democracy 1 (2009). 21 BarbaraGeddes, cited in Inglehart and Welzel, Modernization, 169. 22 Lipset 1959, 30 inWucherpfennig and Deutsch 2009. 23 Inglehart and Welzel, Modernization; Adam Przeworski and Fernando Limongi, “Modernization: Theories and Facts,”World Politics 49 (1997): 155-183. 24 Wucherpfennig and Deutsch, “Modernization.” 25 Przeworski and Limongi, “Modernization,” 165. 26 Ibid., 177.CEU Political Science Journal. Vol. 8, No. 2 141 economic development is among the best predictors of political regimes, there are persistent dictatorships or flourishing democracies against the odds.27 From the beginning of the transition, and for most of the 1990s, Mongolia was in a state of continuous economic decline. Przeworski and Limongi’s emphasis on the positive role of economic performance for newly established democracies’ survival can only offer an explanation for the steady process of democratization in Mongolia without reverting back or leading to another alternative mostly for the period after 2000.28 The GDP per capita survival threshold was reached only in 2006.29 In addition, the lingering economic crisis with few advances and worsening conditions throughout the first decade led to an eventual systemic crisis, which at the time had a high chance of undermining the established regime. Hence, modernization theory’s economic development branch, focused on the macro level, offers an insufficient explanation of democratization, especially due to the linearity it requires for democratic survival in new regimes. This leads to the notion that by itself, it does not provide for the underlying reasons for the start of democratization or an explanation for regime survival in the face of long economic downfalls.30 Nevertheless, the main theoretical statement of modernization theory should still correspond to the notion that economic development is a positive factor in democratization. Economic development is also favorable for further development of efficient democracy by being a driving force for social change. This implies approaching the mature liberal democracy standard in a consolidated regime. Alternatively, persistent economic decline and crisis will lead to reversal of democratization during the transitional period. In a consolidated regime this will entail departing further from mature democracy. However, it should be noted that being social science phenomena, these claims are probabilistic and not deterministic.31 That is to say, making predictive statements is subject to great uncertainty. In the end, this leads to the puzzle the present research seeks to resolve: if economic development is a sufficient and necessary condition for a political democracy to develop or be sustainable, why is Mongolia a democracy? 27 Adam Przeworski, Michael Alvarez, Jose Antonio Cheibub, and Fernando Limongi. Democracy and development Political Institutions and Well-Being in the World, 1950-1990. (Cambridge: Cambridge University Press, 2000), 88. 28 Przeworski and Limongi’s main finding was that that in countries with GDP per capita under $1000, the probability that a democracy would regress in a particular year was 0.125, leading to an expected life of eight years (1997, 165). Nonetheless, it is the potentially explanation for the mechanism behind the systemic crisis of the late 1990s in Mongolia. 29 EBRD, NSOM, World Bank. 30 Inglehart and Welzel, Modernization, 167-169. 31 Ibid., 157.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 142 Considering Mongolia’s historic low levels of economic development and economic downturns throughout the transition process, Inglehart and Welzel’s revised claim of modernization theory with political culture as a mediator between economic development and democratization is applicable. This will require reassessment of how economic and social phenomena relate and the methodology will be discussed in the next section. 3.Methodology To begin with, the closest counterparts to Mongolia in Central Asia of the Third wave democracies are Kazakhstan and Kyrgyzstan. Their historical nomadic roots and Soviet legacies make them the most similar cases for comparison. Nevertheless, the current religious and social structures of these societies are different. Kazakhstan and Kyrgyzstan no longer have significant nomadic populations. Moreover, the persistence of strong clan networks in politics and an Islamic tradition are among potential explanations for their failure in successful democratization.32 In contrast, Mongolia has a third of the population living as traditional nomads and clan influence is considered very weak, as two thirds of the population claim to be Chinggizids. Additionally, the dominant religion is Lamaist Buddhism, which possibly presents less of a cultural barrier to democratization.33 A third factor is that Mongolia is one of the least densely populated countries in the world, with a little less than half of the population residing in the capital, Ulaanbaatar.34 These features combined make it a unique case not suitable for a cross-country small n comparative analysis. Furthermore, the research question is associational with the general purpose of finding the strength of associations between the constructs in the pro-democratic culture claim. Performing empirical (quantitative) analysis with the micro unit of analysis as the individual and transitioning to the macro level of political culture potentially resolves some of the methodological implications. Most studies of political culture rely on survey analysis, reinforced by the argument that only aggregated attitudes of individuals can influence macro-political institutions, in turn representing the "connection between individual values and what governments do."35 In view of the fact that the key to Mongolian success in democratization lies in its political culture, aggregating individual-level data without considering standard 32 Richard Rose, “How Muslims View Democracy: Evidence from Central Asia,” Journal of Democracy 13 (2002): 103. 33 Fritz, “Mongolia,” 77. 34 Estimated at 2.78 mln resides on the territory of 1.5 mln km2 (NSOM 2010, 47). 35 Richard Rose, “Political Behavior in Time and Space,” Studies in Public Policy 414 (2006a), 20.CEU Political Science Journal. Vol. 8, No. 2 143 demographics will entail an assumption of homogeneity of subclasses. On the one hand, this benefits the analysis by producing inferences at a rather general level.36 On the other hand, it also raises the potential of committing an ecological fallacy if individual-level inferences follow from analysis of macro level or aggregated indicators.37 Nevertheless, if interpreted and used correctly, “survey data avoids the ecological fallacy of drawing inferences about individuals from aggregate data, such as election results, or from such reified terms as national history and traditions.”38 Theoretically, the notion of political culture is considered a macro level construct, and also considering that “cultural and historical approaches predict common opinions among individuals within a country, and differences between countries,” adds to support the assumption of homogeneity. 39 Finally, Krishna et al. claim that most analysts’ findings have been based on aggregate level data or indexes, and conclusions of individual behavior were also derived that way.40 This produced a number of outlier cases where democracy was successful despite high levels of poverty. Their studies in Asia, Africa, and Latin America, based on analysis of individual-level data, demonstrated that poor people in those regions do not value democracy any less than their richer counterparts. In order to assess public pro-democratic values empirically most of the following analysis will be based on data drawn from opinion polls covering the period from 1995 to 2012. This will involve analysis of social and economic aspects that influence the population in their support for and willingness to participate in the democratic system. In this work, Pippa Norris’ dimensions of mass level political support consisting of evaluation of current political regime and “support for democracy per se” will be addressed.41 4. Evaluation of Current Regime Ganbat, Tusalem, and Yang’s former studies have underlined that, although Mongolians view political institutions with skepticism and are rather negative when evaluating efficiency of political participation, they are confident in their own ability to participate in politics. This phenomenon has been understood as a 36 Humbert M. Blalock Jr., Theory Construction: From Verbal to Mathematical Foundations. Eaglewood Cliffs: Prentice-Hall, 1969), 149-50. 37 William S. Robinson, “Ecological Correlations and the Behavior of Individuals.” (Reprint) International Journal of Epidemiology 38 (1950): 337-341. 38 Rose, “Political Behavior,” 3. 39 Richard Rose, “Diverging Paths of Post-Communist Countries: New Europe Barometer Trends since 1991,” Studies in Public Policy 418 (2006b): 18, 20. 40 Krishna, Poverty. 41 Norris,Critical Citizens, 37.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 144 “frustrated desire for political influence.”42 Prohl and Luvsandendev also highlighted a negative evaluation of institutions, with an exception of the president.43 Furthermore, they established a correlation between belief in voter’s influence and satisfaction with the political system, which corresponded with election cycles. This satisfaction was particularly high when people believed that casting a vote was worthwhile.44 Consequently, this suggested weak political support for institutions, but not a disapproval of the political system as a whole. This allows us to infer that political self-confidence is at the core of these links. Thus, it will be necessary to look at the assessment of personal ability to influence politics, which generally in a democracy is manifested in belief in voters’ influence, or in other words, belief in political efficacy. As a standard, it holds that if people are going to be affected by political decisions they should have a say in making them. Following from this, Mishler and Rose indicate that in studies of popular support in post-Communist regimes economic factors dominate, with the main disagreement being on the principal sources of economic effects.45 They stress the reciprocal effects of economic and political evaluations conditioned by countries’ communist legacies. In those societies support for a political regime is significantly shaped by economic factors, as in command economies citizens were used to holding the government responsible for both macroeconomic and individual welfare.46 In the case of Mongolia, Pomfret describes the presence of two major economic stabilizers contributing to post-Communist development.47 One can be attributed to the “traditional pastoral lifestyle” and the other to the informal economy.48 The nomadic household is largely outside of the formal monetary economy and is subject to seasonal earnings. Such circumstances would mean that “household income is a poor proxy for poverty” and as a result relying on household earnings would present a limited picture of the micro-level well-being. 49 In addition, household contributions of migrant workers are not captured. These aspects of the informal economy were very crucial during the transitional period and the systemic 42 Ganbat Damba, RolinTusalem and David Yang.”The Mass Public and Democratic Politics in Mongolia,” in How East Asians view democracy, ed. Yun-han Chu, Larry Diamond, Andrew J. Nathan, and DohChull Shin. (New York: Columbia University Press, 2008). 43 Werner Prohl and Sumati Luvsandendev, Voters Voices: People’s Perception of Mongolia’s Political and Economic Transition as Reflected in Opinion Polls from 1995 to 2007 (Ulaanbaatar: MunkhiinUseg, 2008), 117-27. 44 Ibid., 109-112. 45 William Mishler and Richard Rose, “Learning and Re-learning Regime Support: The Dynamics of Post-Communist Regimes,” 41, Issue 1, January 2002. 46 Ibid., 5-6. 47 Pomfret, “Transition.” 48 Ibid., 152. 49 Michael Bratton in Krishna, Poverty, 31.CEU Political Science Journal. Vol. 8, No. 2 145 crisis of the late 1990s and still are significant contributors to the main economy. In the end, suggesting that economic effects depicted in official statistics provide only “a partial picture of how individuals cope with the challenges of transformation through activities in multiple economies.”50 Such gaps in economic data affect both household incomes and GDP per capita measurements. Consequently, the main economic considerations should be at a non-monetary level, yet depict material well-being, which can be captured by the objectivity of the standard of living. Finally, Mishler and Rose assert that people have a certain degree of patience in deficient regimes as long as there’s a belief that circumstances are likely to improve in the nearest future.51 In the context of post-Communist regimes these future expectations would also most likely be economic in nature due to the impact of social and economic transformations. 5. Regime Support Hypotheses Following on from the above and from Limongi and Przeworski’s main claim that GDP per capita depicts macro-level economic development favorable to democratic survival, it is possible to put forward the following propositions. In particular, the main consequence of economic development should be increasing citizens’ standard of living (H1). This is the first main step in developing the desired support for democracy on the micro level. In the context of Mongolia, however, due to very low levels of material security as a base, increasing the standard of living will lead to the belief that circumstances will improve in the foreseeable future (H2). This is a crucial step in a society with a high level of poverty. Next, to secure these interests, the system will have to maintain legitimacy by providing elections as the general method of citizen participation in politics. All of this is reflected in the corresponding belief in the ability to influence political decisions or feeling of political efficacy, which for the general population is mainly limited to casting an effective vote (H3). Then, increasing material well-being should be associated with improving macroeconomic conditions and consequent positive assessment (H4). After that, testing the ability to make informed political decisions and assessing political involvement will lead to investigating societal interest in politics (H5). This has been posited, despite previous research and inferences from general political interest concluding that it either depicts the role of politics in the lives of ordinary citizens, or possibly represents high level of 50 Rose, Diverging Paths, 8. 51 William Mishler and Richard Rose, “Political Support for Incomplete Democracies: Realist vs. Idealist Theories and Measures,” Studies in Public Policy 333 (2000): 11.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 146 societal development.52 And finally, increasing material well-being should lead to increasing support of the regime (H6). Table 1 below shows the indicators selected, while Table 2 summarizes the hypotheses. Table 1:Concepts and Indicators Concept Indicators: Abbr Level Economic Development GDP per capita [ED] Macro Material well-being Standard of living [L] Micro Future economic expectations53 Future outlook [F] Micro Political efficacy Belief inVoter Influence [V] Micro Macroeconomic performance Present Economic Situation [E] Micro Political involvement Interest in Politics [I] Micro Support of the Political Regime54 Satisfaction with the Political System [S] Micro/Macro Table 2: Summary of Core Hypotheses Hypotheses Statement Main Hypothesis If revised modernization theory in Mongolian context is supported then increasing economic development will encourage development towards mature liberal democracy. Null Hypothesis Decreasing economic development will lead Mongolian democracy away from becoming a mature liberal democracy. Economic Development Hypothesis 1: Increasing economic development will produce an increase in standard of living.[ED]→[L] Future Economic Hypothesis 2: An increase in standard of living will 52 Rose, “Political Behavior”; Ronald Inglehart, “Changing Values and Changing Societies,” in Challenges of Theories on Democracy, ed. Stein U. Larsen (New York: Columbia University Press, 2000). 54-90. 53 It should be noted that this is Personal Future Economic Expectations, questionnaires structure makes it economic in nature (not shown here). Tested and shown different from future macroeconomic assessment and expectations of economic development in 5 years’ time (not addressed). 54 1995-2007 “How much are you satisfied with the present political system?” and 2008-2012 “How much are you satisfied with the Democracy and present political system?” To test confidence, comparison with satisfaction with government and opposition was done, which demonstrated that all three are highly correlated ( Appendix 4: SystemicVariables).CEU Political Science Journal. Vol. 8, No. 2 147 Expectations produce an increase in future economic expectation. [L]→[F] Political Efficacy Hypothesis 3: An increase in standard of living will produce an increase in belief in political efficacy. [L]→[V] Macroeconomic Assessment Hypothesis 4: An increase in standard of living will be associated with increase in evaluation of macroeconomic performance.[L]↔[E] Political Involvement Hypothesis 5: An increase in standard of living will produce an increase in political involvement. [L]→[I] Support of the Political Regime Hypothesis 6: An increase in standard of living will produce an increase support of the political regime. [L]→[S] Figure 1 illustrates the multilevel process of change, and Figure 2 depicts the assumed causal order for the micro level process in more detail. It can be seen that the processes depicted in Figure 1 require multilevel thinking. In order to examine them I divide the analysis of regime support into trend lines and log linear modeling for mathematical simplicity. These phenomena are influenced by the dynamics of time and as subjects to the same cause, the affected variables are most likely interrelated (depicted in Figure 2). Consequently, trend lines will cover the dynamics of change and make it possible to distinguish situational and structural factors at work. Log linear analysis will involve finding a model that can represent regime support. Figure 1: The Multilevel Process of Change55 55 Adapted from “Coleman’s bathtub” cited in Michael Oakes, “Commentary: Individual, Ecological, and Multilevel Fallacies,” International Journal of Epidemiology 38 (2009): 361-368.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 148 Figure 2: The Micro Level System 6. Support for Democracy The dimension of political support seen in supporting democracy per se, which implies supporting it as a political good, can be assessed in different ways. Among three identified methods the first would be the assessment of people’s preference of democracy over other types of regimes. This method was especially prevalent in post-Communist and transitional regimes where the citizens were believed to be “better judges of differences due to first-hand experience.”56 The second method would be Inglehart and Welzel’s analysis of primacy of “self-expression” over “survival” values, which reveals democratic support for intrinsic reasons or instrumental purposes. In other words, support of it as a political good or a source of economic gain. The third evaluative method is a definition of democracy compiled from mass opinion. This was attempted in former studies in Mongolia. The Asian Barometer survey asked respondents to provide a definition or meaning of democracy. This measurement was implemented to reflect whether “minimalist (procedural) or maximalist (substantive)” understanding of democracy prevails in the society.57 The findings led to the conclusion that “substantive interpretations of democracy among Mongolians are minimal at best” and most “identify democracy with a minimalist 56 Mishler and Rose, “Political Support,” 10. 57 Ganbat, “The Mass Public,” 8.CEU Political Science Journal. Vol. 8, No. 2 149 definition of basic freedoms.”58 In other words, that most people don’t know exactly what they want in terms of democracy. However, a further available subjective measurement of valuing democracy as a political good can be drawn from the general value theory elaborated by Schwartz, which describes values as “desirable, trans-situational goals, varying in importance, that serve as guiding principles in people’s lives.”59 These goals create belief systems that trigger action according to circumstances and are interrelated with many other competing values. Moreover, if values are “conceptions of the desirable, used in moral discourse, with a particular relevance for behavior” the degree to which particular values are prevalent can suggest the underlying belief system.60 In terms of supporting democracy as a political good, the values concerned are political, which involve only a segment of the individual’s life.61 Furthermore, even if certain political values are held, most people are not actively engaged in politics to advance them, suggesting a limited role of politics in people’s lives.62 Nevertheless, examining different levels of importance that people assign to values will offer general directions of their expectations. In general, the three main principles of democracy are freedom, equality, and justice. However, the content of specific democratic values can be very extensive. The broad definition of liberal democracy includes valuing individual freedom, rights, justice, equality, and divergent views. The block of statements included in Politbarometer surveys covers a range of democratic principles and issues, which are measured by degrees of importance assigned to each value. This makes it possible to indirectly assess what democracy involves or represents to the masses, and will be selected for the analysis.63 As a preferred analytic method, factor analysis will aid in reducing the information on democratic values and issues in order to evaluate support of democracy per se. In particular, it will represent a large number of relationships in a simpler way. To conclude, the main theoretical claim of the revised modernization theory is stated as economic development and is associated with democratization and political culture; they go together and are subject to the dynamics of time.64 Ideally, the research question requires comprehensiveness and a multilevel answer. 58 Ibid., 10. 59 Shalom H. Shwartz, “Basic Human Values: Theory, Methods, and Applications,” unpublished manuscript (2006). 60 Ibid., 28. 61 Ibid. 62 Rose, “Political Behavior.” 63 See Appendix 3: Liberal Democratic Principles and Issues 2008-2012 for list of values. 64 Inglehart and Welzel, Modernization.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 150 However, due to impossibility of the former, and the practical limitations of achieving the latter, simplicity will be introduced to provide mathematical thinking and clarity. Moreover, despite the scope, the availability of empirical data restricted inferences to be based only on recent development, thus whether the main claim holds will be a subject of time. As a result, many of the assumptions made might suffer from oversimplification. Nonetheless, with certain limitations on interpretations, testing of the proposed hypotheses will be carried out in the following analyses. 6.1 Trend lines 1995-2012 If the hypothesized systemic performance is “best” evaluated by the population, trend lines will depict those long-term social changes. They also offer an evaluation of the impact of time, and potentially estimate the dynamics of multilevel changes. To begin with, one of the first hypothesized transitions was from macro-level economic development to micro-level individual well-being. It can be seen from Figure 3 that macro-level economic development experienced a sharp drop after 1989 with the start of the transition, and hitting its lowest point in 1993. It also depicts that the low continued throughout the rest of the 1990s, only bottoming out in the early 2000s. It thus corresponds to initial economic breakdown, long-term stagnation of 1990s, and eventual improvements. The subsequent sharp rise began from 2005, matching the start of growth due to mining developments. Figure 3:Macro and Micro Economic Development Indicators 65 Sources: World Bank 1981-2011; Politbarometer 1995-201266 . 65 Standard of living presented is a result of collapsing categories “very good,” “good”, and “not good - not bad”. As the dominant category is “not good - not bad” and “very good” is practically non-existent, this is the beyond “survival” level (complete figure in Appendix 5: Present Standard of Living). 66 Politbarometers from 1995 to 2000 covered only Ulaanbaatar (UB), due to unique demographic structure of the society it can be argued that the samples are still representative of the whole population.CEU Political Science Journal. Vol. 8, No. 2 151 Next, some substantial trends in well-being can be observed during this period, even if available data only covers the period after 1995. There are some fluctuations, with the first visible sharp drop matching the systemic crisis of the late 1990s, caused by stagnant economic conditions for most of the 1990s. Furthermore, the first actual victory of the Democratic Union led to two dismissals in the government adding to mass disillusionment with the competence of the established system.67 Most probably, as Rose asserts with the passage of about a decade, people’s patience deteriorates as they no longer evaluate current regimes by comparison to previous regimes or potential improvements, but based solely on their present performance.68 As a result, the systemic survival at this stage is mostly attributed to the role of international donors and economic stabilizers.69 It can be speculated that the second sharp drop seen from 2004 to 2006 matches the decline of social welfare coverage as a result of failure of government coalition at the time, and the third sharp drop reflects the impacts of the global financial crisis.70 Otherwise, overall there is a gradual increase, which suggests that the material conditions of citizens did improve, but with some drawbacks. After that, as shown in Figure 4 satisfaction with the political system, belief in political efficacy and assessment of macroeconomic conditions reveal considerable fluctuations related to changes in government. However, macroeconomic evaluations are more negative, which probably reflects the underdeveloped institutions and persistence of poverty. Satisfaction with the political system showed a sharp rise in 2007, and reached a steady high point in the period between the fifth and sixth parliamentary elections. This coincides with the money distributed as an election campaign promise; nevertheless, this “incentive” was only partially fulfilled and thus the influence started to drop steadily. Also, in Figure 4 it can be seen that political interest remained rather moderate and stable for most of the period. However, during the fifth election cycle it began a steady drop, which has two potential explanations. One is the influence of the new generation that did not know the previous regime and, to put it simply, assigns less value to political aspects. The other is that with the passage of time, unrealistic expectations of democracy started weakening.In comparison, the future economic expectations were mostly positive, but showed some fluctuations in the period between 1995 and 2000, another effect of the political crisis. Afterwards they rose gradually and remained steady. 67 Prohl and Luvsandendev, Voters Voices, 110. 68 Rose, Diverging Paths, 8. 69 Fritz, “Mongolia”; Pomfret, “Transition.” 70 2008 financial crisis.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 152 Figure 4: Major Trends71 Note: 2.0 Very; 1.0 Rather; 0 Neutral; -1.0 Rather Not; -2.0 Totally Not Overall, these results show that material well-being improved together with general macroeconomic developments. This supports the claim that macroeconomic developments over time also improved micro-level living conditions, supporting Hypothesis 1. Moreover, future economic expectations reached their peak with the start of sustained economic development, supporting Hypothesis 2. However, even though throughout the examined period political involvement was stable and rather moderate, it has goneinto a decline in the last few years. This does not support the notion that it will rise with material well-being, thus Hypothesis 4 did not receive sufficient support. The most obvious possibility is that the positive change should entail a much larger time span. In addition, it is subject to a plurality of people’s interests and also, despite some improvements, the general material level of well- being has not reached the favorable level. Then again, if the decline continues, it is potentially an indicator of mass disillusionment. This could suggest no substantive 71 The variables measurement consisted of different Likert-scales of two, four, and five levels of measurement. Thus in order to create comparable trend lines I've standardized them using scaling from -2 to +2. Two level variable SUM(maxV*1:minV*-1)/Total, four level variable SUM(maxV*2:maxV*1:minV*-1:minV*-2)/Total, five level variable SUM(maxV*2:maxV*1:V*0:min:V*-1:minV*-2)/Total. For originals look at Appendix 5: Present Standard of Living; Appendix 6: Belief in Voters Influence – Political Efficacy; Appendix 7: Present Economic Situation – Assessment of Macroeconomic Situation; Appendix 8: Interest in Politics – Political Involvement; Appendix 9: Future Economic Expectations.CEU Political Science Journal. Vol. 8, No. 2 153 improvement in the quality of the regime, which in the long run could prove harmful for the democratic system by undermining its support. Belief in voter influence or political efficacy goes through considerable fluctuations with changes in government, reaching its peak in election years, which suggests the relevance of election campaigns. Moreover, this trend is most closely correlated with satisfaction with the political system. Nevertheless, satisfaction with the political system had gradually improved to a slightly more positive evaluation, but with considerable fluctuations. Similarly, macroeconomic performance evaluations fluctuated with changes in government, but were rather negative. Even though there is a slight improvement over time, one might speculate that weakness of institutions and poor “rule-of-law” are related to this negative assessment. Thus, support ofHypotheses 3, 5, and 6will further require testing by log linear analysis in the next section. 6.2 The Regime Support Model Now we move to next stage of the trend analysis: current standard of living [L], future economic expectations [F], and feelings of political efficacy [V] were seen as the main positive factors in the hypothesized system.72 Political interest [I] was rather moderate and seemed to go into decline. Macroeconomic conditions assessment [E] was mainly negative throughout the period. Consequently, this section will examine the strength of their links through log linear modeling.73 First of all, explanatory and response variable associations were tested. Political interest was statistically tested on the relationship with other indicators, but in the end did not satisfy the criteria of p-value below 0.05 for the χ 2 Test of Independence. Macroeconomic performance assessment [E] passed the test, but the resulting log linear models had rather poor explanatory power.74 Substantively, this 72 See Figure 2 Micro Level System in Regime Support Hypotheses section. 73 The system includes multiple associations among categorical social science variables, which can be handled by flexibility of log linear technique. In particular, instead of fitting data to a model, it suggests finding a model to fit the data. Moreover, it permits us to express categorical data in the form of a linear model by using log values. The software used for this part of the analysis was the “psych” package in R. For practical reasons and for clarity, categories were collapsed to create lower levels of measurement (see Appendix 10). 74 The models did not seem impressive: the best fitting model included multiple high order interactions and had a p-value of 0.10. Also the five-way models were not ran on previous surveys, only on April 2012, since it had a large enough sample size (n=5020). Other samples had an insufficient size (n=~1000) to provide reliable fit measurements. Two-way and three-way models, including macroeconomic assessment, also did not fit the data well. I did not find it necessary to exhaust all possible combinations.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 154 leads to the conclusion that egocentric evaluations dominate and contain the most explanatory power. Therefore, satisfaction with democracy [S],75 standard of living [L], political efficacy [V], and future economic expectations [F] were tested, and theysatisfied the criteria of p-value below 0.05 for the χ 2 Test of Independence. This implies that the variables are not independent, thus rejecting the null hypothesis of overall variable independence and allowing for log linear analysis. The χ 2was 318.2 with df = 11 and p-value<.001(1.243e-61), indicating a statistically significant association between these variables. Moreover, Table 3 (n=3962) demonstrates that there are no zero cells, implying no reduction in test power. It also shows that these data will not fit a regular additive model, which requires the difference to be approximately equal. Table 3:Cross tabulations of Political Efficacy, Standard of Living, Future Economic Expectations, and Satisfaction with Democracy Political Efficacy Standard of Living Satisfaction with Democracy Future Expectations 0 (Dissatisfied) 1 (Satisfied) 0 (Weak) 0 (Bad) 0 (Pessimistic) 52 19 1 (Optimistic) 143 185 1 (Average) 0 (Pessimistic) 40 27 1 (Optimistic) 447 642 1 (Strong) 0 (Bad) 0 (Pessimistic) 47 29 1 (Optimistic) 154 338 1 (Average) 0 (Pessimistic) 45 36 1 (Optimistic) 556 1202 75 To increase confidence in satisfaction with the democracy variable, it was compared to satisfaction with government and opposition variables. The trend lines in Appendix 4: Systemic Variables demonstrate that the three variables are highly correlated, which allows one to consider satisfaction with the political system as a “satisfactory” measurement for the evaluation of the established regime. Though the correlation of opposition is a side effect leading to a conclusion that people do not distinguish between the government and opposition. The implication is it depicts a rotation without improvement (See Richard Rose, “Democratic and Undemocratic States,” Studies in Public Policy 444 (2008).CEU Political Science Journal. Vol. 8, No. 2 155 Finally, in the process of searching for the “best fitting model”, Knoke and Burke’s work was used for reference.76 First, a baseline model [all explanatory][the response] or [VLF][S] was selected. 77 Second, interaction terms were added to improve the fit. Third, the resulting models were evaluated for most substantive and statistical significance. In addition, running models at different periods of time and with different interaction terms makes it possible to evaluate the significance of different associations, which can aid in approximating the underlying causal structure. A further consideration is that larger samples require more complex models to pass goodness-of-fit tests. Thus, looking at previous comparable surveys with smaller sample size aided the search. Table 4: Log Linear Models78 Model Fitted Marginals April 2012 Fit (p) April 2011 Fit (p) October 2010 Fit (p) April 2010 Fit (p) 1 [VLF][S] 0 - - - 2 [VLF][VS] 1.998401e-15 - - - 3 [VLF][LS] 0 - - - 4 [VLF][FS] 5.675825e-08 - - - 5 [VLF][VS][LS] 7.993606e-15 - - - 6 [VLF][VS][FS] 0.5053405 0.727612 0.0592234 0.2473873 7 [VLF][LS][FS] 3.000488e-08 - - - 8 [VLF][VS][LS][FS] 0.4961563 0.5911393 0.5173547 0.2590765 9 [VLF][VLS][FS] 0.4488285 0.431616 0.6537358 0.1539966 10 [VLF][VFS][LS] 0.3548742 0.4968378 0.6396748 0.7843583 11 [VLF][LFS][VS] 0.756737 0.5921486 0.3573312 0.1894376 12 [VLF][VLS][LFS] 0.7862492 0.3939074 0.4729964 0.09464824 In the end I selected Model 11 for further analysis. The results in Table 4 under the April 2012 column show that this model has a very good fit (p=0.75) in comparison to less parsimonious Models 6, 8, 9, and10. In addition, compared to 6, 9, and 12, it has stability at other points in time. Comparatively, it satisfied the condition of “best fitting” model (statistically significant and substantively meaningful) and is represented in Figure 5. This model contains two three-factor associations and one two-factor association. It can be interpreted as showing that standard of living is mutually related to political efficacy and future expectations [VLF] and mutually related to future expectations and satisfaction with democracy [LFS], and that political efficacy is related to satisfaction with democracy [VS]. 76 David Knoke and Peter J. Burke, Log-linear Models. Volume 20. (Newbury Park: Sage Publications, 1980). 77 [S] conceptualized as the response variable, whose odds are a function of [L],[F], and [V]. 78 I did not rerun the models that did not fit the data in April 2012 on previous years.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 156 Figure 5: Log Linear Model [VLF][LFS][VS]79 Moreover, the hypothesized response variable [S] is allowed to interact with the explanatory variables [V], [L], and [F]. In this case it has a significant relationship with political efficacy, and a significant joint relationship with standard of living and future expectations. From Table 4 it can also be seen that statistically including interactions of standard of living and future economic expectations considerably improves the fit of the model. Substantively, this suggests the core influence produced by the simultaneous presence of the two. Hence, if the model is correct the following two tables summarize the results: Table 5: Fitted Values for Model [VLF][LFS][VS] Political Efficacy [V] Standard of Living [L] Satisfaction with democracy [S] Future Expectations [F] 0 (Dissatisfied) 1 (Satisfied) 0 (Weak) 0 (Bad) 0 (Pessimistic) 51.24857 19.75138 1 (Optimistic) 138.4143 189.5857 1 (Average) 0 (Pessimistic) 42.29351 24.70645 1 (Optimistic) 450.0437 638.9565 1 (Strong) 0 (Bad) 0 (Pessimistic) 47.75142 28.24862 1 (Optimistic) 158.5857 333.4143 79 “•” marks higher order associations. In this case three-way associations.CEU Political Science Journal. Vol. 8, No. 2 157 1 (Average) 0 (Pessimistic) 42.70648 38.29355 1 (Optimistic) 552.9564 1205.044 Note: Model fit: χ 2 = 1.18729, df=3, p=0.7567)80 Table 6: Estimated Odds and Odds Ratio Calculations for Model [VLF][LFS][VS] Political Efficacy [V] Standard of Living [L] Satisfaction with democracy [S] Future Expectations [F] Odds Odds ratio 0 (Weak) 0 (Bad) 0 (Pessimistic) 0.385404 3.553932 1 (Optimistic) 1.369698 1 (Average) 0 (Pessimistic) 0.584166 2.430412 1 (Optimistic) 1.419765 1 (Strong) 0 (Bad) 0 (Pessimistic) 0.591577 3.553932 1 (Optimistic) 2.102423 1 (Average) 0 (Pessimistic) 0.896668 2.430413 1 (Optimistic) 2.179274 The results in Table 5 suggest that the fitted values do not deviate much from the observed values presented in Table 3. Therefore, the model has good estimative power. In addition, from the results depicted in Table 6, it can be said that the odds of satisfaction with democracy improve with the presence of each factor, but much more significantly for those who are optimistic about the nearest future. Similarly, the odds improve in the presence of other factors, but not as significantly for those who are pessimistic. For instance, for those who don’t believe in political efficacy and are pessimistic about the nearest future, the odds of being satisfied with democracy are 0.38 and 0.58 (bad and average standards of living respectively). In comparison, regardless of standard of living, for those who don’t believe in political efficacy but are optimistic about the nearest future, the odds of being satisfied with democracy are about 1.4. Alternatively, for those who believe in political efficacy and are pessimistic about the nearest future, the odds of being satisfied with democracy are 0.59 and 0.89 (for bad and average standard of living respectively). Likewise, regardless of standard of living, for those who believe in political efficacy but are optimistic about the nearest future, the odds of being satisfied with democracy are slightly greater than 2.1. 80 Satisfies that in log linear modeling χ2 goodness-of-fit should be small relative to degrees of freedom. Alternatively L2 =1.18447, df=3.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 158 Furthermore, the consequences of the fitted model are the symmetry of odds ratios. Thus, for respondents who have a bad standard of living, regardless of whether they believe or not in political efficacy, as long as they believe that in the near future life circumstances will be better,they are 3.55 times more likely to be satisfied with the present political system. Alternatively, for respondents who have an average standard of living, regardless of whether they believe or not in political efficacy, as long as they believe that in the nearest future life circumstances will improve, they are 2.43 times more likely to be satisfied with the present political system. This corresponds to the notion that with rising material security, people become more critical of their political system. Trend analysis and statistical results from log linear analysis disconfirm the complete hypothesized causal order on the micro level at the current stage of development.81 The suggested substantive reason for this is the importance of self-centered economic assessments due to low levels of material well-being. Nevertheless, the implications of the model are that respondents with a bad standard of living, regardless of whether or not they believe in political efficacy, as long as there is belief that in the nearest future circumstances will improve, are three and a half times more likely to be satisfied with the present political system. Alternatively, respondents that have an average standard of living, regardless of whether or not they believe in political efficacy as long as they believe that in the near future life circumstances will become better, are two and a half times more likely to be satisfied with the present political system. For the respondents with a bad standard of living, future economic optimism plays a much more prominent role in regime support and they are less critical. Nevertheless in this case, systemic support and evaluation is dependent on the respondent’s present living conditions, belief in political efficacy, and most importantly, on future economic optimism (support H1, H3, H6). The following graphical representation of the micro-level system support was made from the results of log linear analysis: 81 See Figure 2 Micro Level System in Regime Support Hypotheses section.CEU Political Science Journal. Vol. 8, No. 2 159 Figure 6: Regime Support82 From Figure 6 it can be seen that the original micro-level system had to be altered. This suggested system disproves some components of the original micro level causal system and reveals that standard of living and future economic expectations are the core influences. The model represents the basis of political equilibrium, which was achieved by the democratic system in Mongolia. Following from here the next, section will examine what democracyas a political system potentially represents to the general public. 7.Model of Democracy as a Political Good In the assessment of support for democracy per se various values and issues are included in the concept. In an ideal world, one’s theory would suggest hypotheses for a confirmatory factor analysis model, they would be tested, and the appropriate conclusions would be drawn. However in practice, due to the ‘insurmountable uncertainties’ of social science research, the choice was in favor of exploratory factor analysis to determine the structure of democratic value scale orientations. The analysis was made with SPSS and interpreted according to methods by Kim and Mueller and the results are displayed in the following table: 83 82 Combined theoretical causal structure and statistical model.[Uyt] dynamic system subject to effects of outside and unmeasured variables. 83 Jae-on Kim and Charles Mueller. W. Introduction to Factor Analysis: What It Is and How To Do It. Volume 13. (Newbury Park: Sage Publications, 1978); Jae-on Kimand Charles Mueller. W. Factor Analysis: Statistical Methods and Practical Issues. Volume 14. (Newbury Park: Sage Publications, 1978).Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 160 Table 7: Three-Factor Model of Liberal Democratic Values Dimension Variables Factor 1 2 3 Libertarian Everybody can believe in what he/she wants 0.616 I can travel wherever I want 0.586 Everybody can express his/her opinion freely 0.565 0.357 Media and research are uncensored in Mongolia 0.537 Everybody has the right to enter one's desired profession 0.496 0.401 All parties have an equal chance to come into government 0.469 All people have equal educational opportunities 0.462 0.433 Everybody can participate in the activities of their choice during one's free time 0.446 Egalitarian All people are equally treated by the law 0.63 Men and women have equal rights 0.61 There is a free, democratic market 0.546 Social Liberalism Income differences are kept as small as possible* 0.537 0.364 Social differences are kept as small as possible 0.402 0.704 The state provides for social justice in a market economy 0.407 0.622 Everybody has the freedom to decide about his property 0.606 The state provides as many social security services as possible 0.377 Eigenvalues 5.864 1.613 1.074 Percent of Variance explained 33.056 6.655 3.146 Cumulative Percent of Variance Explained 33.056 39.711 42.857 χ2 with 75 degrees of freedom = 1261.495 *in 2012 this variable was affected by variability, otherwise presents an indicator of Social Liberalism Dimension. Extraction Method: Maximum Likelihood. Rotation Method: Varimax with Kaiser Normalization. Rotation Converged in 7 iterations. Note. Loadings <0.35 suppressed. Table 7 summarizes the results of the three-factor model, which depicts the orientation in support of democracy per se. The factor names are based on the traditions of liberal political philosophy. The first factor, categorized as theCEU Political Science Journal. Vol. 8, No. 2 161 Libertarian Dimension, can be identified by high factor loadings on statements that can be grouped by their adherence to the fundamental value of self-ownership in the theory of Robert Nozick. The second factor, categorized as the Egalitarian Dimension, is identified by higher priority of equal outcomes and market competition, grouped under economic egalitarianism linked with the traditions of John Maynard Keynes. The last factor, the Social Liberalism Dimension, can be identified by high factor loadings on statements valuing combinations of state’s role in ensuring social justice and general equality, grouped by the theory of Karl Marx. However, thespecificity of this dimension is in the inclusion of a high factor loading on the freedom to decide about one’s property.84 Overall this suggests that people differentiate among the different values, and distinct value systems can be formed from them. The results of factor analysis revealed the presence of distinct value orientations. Considering that the larger the sample size in relation to the number of variables, the more reliable the resulting factors, we can accept the last factorial model as the most stable and proceed to the interpretations. Statistically, the first factor accounts for as much variance as possible, the second accounts for as much variance left unexplained by the first, while the third accounts for variance left unexplained by the first two. If Schwartz’s general hierarchy of values is applied to democratic values, it can be argued that substantively, the Libertarian Dimension carries the most information. The values of freedom of belief, expression, and travel carry high factor loadings and contribute most to the description of this dimension.85 Moreover, they are consistently linked together and can be considered the main indicators of this orientation. In general, it is suggested that this value dimension constitutes a belief system reflecting a sense of personal liberty or inalienable rights. In the Egalitarian Dimension, the values of equal treatment by law, gender equality, and democratic market carry high factor loadings and are the main indicators.86 This dimension can be considered Keynesian for underlying substantive reasons, which does not necessarily imply the importance of general equality, but rather of equality of opportunities.87 In the context of Mongolia, the low levels of material well-being will lead to an emphasis on the importance of freedom of opportunity, which is necessary to improve life circumstances. 84 It is suggested that this is not coincidental as it persisted in this dimension in other analyses. The possibility is that the nature of this value dimension implies that people will want redistribution, but will not want it to affect their property. 85 Schwartz, “Basic Human Values.” 86 It should be noted that this dimension’s indicators are less stable in comparison to the indicators of other dimensions when tested on smaller samples. However, they are consistently grouped together. 87 Another potential name is Economic Egalitarianism.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 162 The values of the state ensuring social justice in market economy, small income and social differences contribute to the Social Liberalism Dimension. However, this also includes a high loading on freedom of property. This dimension is most clearly defined and consistently present, reflecting the belief system valuing social justice in the society. It has consistent high factor loadings, reflecting the high priority of these values, which are likely to be emphasized due to the feelings of injustice caused by the unequally distributed economic growth in society. There are very few winners in the new system and a large impoverished mass, which is proportionally morevisible in a small population. 8.Conclusions In conclusion, the main objective of the present work was to test a general theory in a national setting to uncover reasons behind successful democratization in a suggested anomaly state. Overall, the empirical analysis and findings do not contradict each other and are favorable to the general theoretical claim of modernization theory with regard to cultural conditions. However, testing the main theoretical claim entails a very ambitious scope, since tracing social changes covers an extensive time period, and should reflect society’s entire historical process.88 This created multiple limitations on how far this study could be taken with only aquantitative approach, especially when population statistics and surveys were introduced relatively late. Moreover, some of the simplifying assumptions made do not reflect the complexity of the world and at some points mathematical clarity came at a cost of descriptive depth. Nonetheless, when separating societies generally into hunting and gathering and agrarian empires, scholars infer that the former are "relatively liberal, egalitarian, and democratic" in comparison to the latter, which predominantly emphasize collective values and conformity.89 The traditional culture of Mongolia is pastoral nomadism, which predisposes the people to high values of individual autonomy. This in turn suggests that Rose’s realist notion of liberal democracy and a choice of "lesser evil"90 (by judgment that no other system does better to protect individual rights) would make it the preferred trajectory of development. The results of the analyses of value dimensions and log linear modeling are in favor of support of democracy motivated by mainly by economic gain as a form of governance in Mongolia. Examining trend lines revealed cycles of conditioned political support. In addition, with Mishler and Rose it can be argued that in transitional regimes, people can better assess their regimes against other 88 Inglehart and Welzel, Modernization. 89 Inglehart and Welzel, Modernization, 35. 90 Rose, “Democratic and Undemocratic States.”CEU Political Science Journal. Vol. 8, No. 2 163 alternative, altogether adding to the suggestion that they are mainly “rational democrats.”91 This also implies that self-centered economic assessments are more important for systemic evaluations. Moreover, in general, interests have to be secure and become favorable in the foreseeable future. This potentially accounts for the tolerance of systemic deficiencies in the society. Most importantly, even if there was some improvement in material well-being over time, society has not reached the level of material security necessary to overcome basic needs. Consequently, leading to a conclusion that if there is a major influence of economic development in relation to support of the regime, then it is most probably high hopes for a better future. Even if society is predominantly poor, the findings suggested that people in general are not against democracy.92 The most probable difference is that the proportion of those valuing democracy for economic gains (instrumental support) is more prevalent. According to Inglehart and Welzel, the main problem with this instrumental support is that it generally entails less tolerance in society. 93 In addition, the communist system’s equally distributed poverty was replaced with a system of market competition, which resulted in winners and losers of the new system. However, as Prohl and Luvsandendev revealed, the issue is that there are very few winners and too many who consider themselves losers.94 The view that “winner takes all” is leading a feeling of social injustice in society. In particular, the small population of the country, mainly concentrated in the capital, makes inequality very visible. Even if the government has little transparency and accountability, people can easily observe errors and injustices. Therefore, all of this adds to multiple risks and uncertainties in a developing society. In sum, this work examined the claim of pro-democratic political culture in Mongolia, which at the moment is widely accepted by the public as a common sense argument. In the process, empirical support was built on the basis of inferences from revised modernization theory represented by Inglehart and Welzel and findings in the New Democracies Barometers by Rose et al. in order to explore the scientific basis. Additionally, the generated case-specific theoretical propositions and methodology have made a small contribution to the existing literature by demonstrating how unique features can be managed. This work offers insights into how liberal democracy was sustained in a society culturally distant from the West. The findings empirically supplement previous studies, which used the political culture argument, but were restricted by availability of data or their approach. In the end, considering the complexity of analyzing changing societies, this study can 91 Mishler and Rose, “Learning and Re-learning Regime Support” “Political Support for Incomplete Democracies.” 92 According to NSOM 2010 almost 40% living under the poverty line. 93 Inglehart and Welzel, Modernization, 115-123. 94 Prohl and Luvsandendev,Voters Voices.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 164 be used as a starting point for further research on Mongolia’s democratic development in particular, and on uncovering different issues of democratization in general.CEU Political Science Journal. Vol. 8, No. 2 165 APPENDICES Appendix 1: Politbarometers and Elections ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 2004 2005 2006 2007 2008 2009 2010 2011 2012 2003 Politbarometers conducted by the Sant Maral Foundation (each survey is marked "●") 1995 1996 1997 1998 1999 2000 2001 2002 2000 Parliamentary elections 1995 1996 1997 1998 1999 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 MPRP MNDP/MSDP coalition MPRP MPRP/MDC coalition MPP 2005 2006 Presidential elections and names of incumbent presidents 1995 1996 1997 1998 1999 2000 2013 Ochirbat Bagabandi Bagabandi Enkh-bayar Elbegdorj 2001 2002 2003 2004 2007 2008 2009 2010 2011 2012Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 166 Appendix 2: Politbarometer Variables – Systemic Analysis Labels Value Labels 1995- 2007 How much are you satisfied with the present political system? Satisfied – 1 Rather satisfied – 2 Rather not satisfied – 3 Not satisfied – 4 No Answer – 8 Don’t Know – 9 2008- 2012 How much are you satisfied with the Democracy and present political system? Satisfied – 1 Rather satisfied – 2 Rather not satisfied – 3 Not satisfied – 4 No Answer – 8 Don’t Know – 9 1995- 2012 In general, how do you evaluate the present economic situation in Mongolia? Very good– 1 Good– 2 Not good, not bad– 3 Bad – 4 Very Bad– 5 No Answer – 8 Don’t Know – 9 1995- 2012 How much are you interested in politics? Very strongly interested - 1 Rather interested - 2 Slightly interested - 3 Rather not interested - 4 Totally not interested – 5 No Answer – 8 Don’t Know – 9 1997- 2012 In general, how satisfied are you with the government? Satisfied – 1 Rather satisfied – 2 Rather not satisfied – 3 Not satisfied – 4 No Answer – 8 Don’t Know – 9 1997- 2012 How satisfied are you with the opposition? Satisfied – 1 Rather satisfied – 2 Rather not satisfied – 3 Not satisfied – 4 NoAnswer – 8 Don’t Know – 9CEU Political Science Journal. Vol. 8, No. 2 167 1995- 2012 How is your present personal and family's standard of living? Very good - 1 Good - 2 Not good, nor bad - 3 Bad - 4 Very bad - 5 No Answer – 8 Don’t Know – 9 1995- 2012 In your opinion, how strong is voters' influence on political decision making? Very strong -1 Rather strong - 2 Rather little - 3 None - 4 No Answer – 8 Don’t Know – 9 1995- 2012 How do you evaluate your nearest future? Rather Optimistic – 1 Rather Pessimistic – 2 No Answer – 8 Don’t Know – 9Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 168 Appendix 3: Liberal Democratic Principles and Issues 2008-2012 Label: The following statements describe democratic principles and issues. Please rate the importance of each statement listed below: Label Value Label I can travel wherever I want Very Important – 1 Rather Important – 2 Rather not Important – 3 Totally Unimportant – 4 No Answer – 8 Don’t Know – 9 Everybody can believe in what he/she wants Everybody can express his/her opinion freely Media and research are uncensored in Mongolia Everybody has the right to enter one's desired profession Everybody can participate in the activities of their choice during one's free time There is a free, democratic market Men and women have equal rights All parties have an equal chance to come into government All people have equal educational opportunities Income differences are kept as small as possible All people are equally treated by the law The state provides as many social security services as possible Everybody has the freedom to decide about his property Social differences are kept as small as possible The state provides for social justice in a market economyCEU Political Science Journal. Vol. 8, No. 2 169 Appendix 4: Systemic Variables95 Appendix 5: Present Standard of Living 95 Trend lines are the result of collapsing categories “satisfied” and “rather satisfied.” It can be seen that the three are strongly correlated, which led to a conclusion that respondents do not distinguish between the government and opposition.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 170 Appendix 6: Belief in Voters Influence – Political Efficacy Appendix 7: Present Economic Situation –Assessment of Macroeconomic SituationCEU Political Science Journal. Vol. 8, No. 2 171 Appendix 8: Interest in Politics – Political Involvement Appendix 9: Future Economic ExpectationsMina Sumaadii: Mongolian Values and Attitudes toward Democracy 172 Appendix 10: Recoding of Variables for Log Linear Model RESPONSEVARIABLE Original Recoded stfdemo – Satisfaction with democracy and present system (1) Satisfied, (2) Rather Satisfied, (3) Rather not satisfied, (4) Not satisfied stfdem (S) (0) Dissatisfied (including categories 3 and 4) (1) Satisfied (including categories 1 and 2) → valid n: (0) 1883, (1) 2877; total 4760 EXPLANATORYVARIABLES Original Recoded voteinfl – In your opinion, how strong is voters' influence on political decision making? (1) Very Strong, (2) Rather strong, (3) Rather Little, (4) None voteinfl (V) (0) Weak (incl. cat. 3 and 4) (1) Strong (incl. cat. 1 and 2) → valid n: (0) 1844, (1) 2642; total 4486 llevel –How is your personal and family’s life level situation? (1) Very good, (2) Good, (3) Not good, not bad, (4) Bad, (5) Very Bad llevel (L) (0) Bad (incl. cat. 4 and 5) (1) Average (incl. cat. 1, 2, 3) → valid n: (0) 1198, (1) 3795; total 4992 future –How do you evaluate your nearest future? (1) Rather Optimistic, (2) Rather Pessimistic future (F) (0) Pessimistic (incl. cat. 2) (1) Optimistic (incl. cat. 1) → valid n: (0) 344, (1) 4268; total 4612 intpols-How much are you interested in politics? (1)Very strongly interested, (2 ) Rather interested, (3) Slightly interested, (4) Rather not interested, (5) Totally not interested intpols (I) (0) Not interested (incl. cat. 4 and 5) (1) Interested (incl. cat. 1, 2, 3) → valid n: (0) 2213, (1) 2733; total 4946 macroecon -In general, how do you evaluate the present economic situation in Mongolia? (1)Very good, (2) Good, (3) Not good, not bad, (4) Bad, (5) Very Bad macroecon (E) (0) Bad ((incl. cat. 4 and 5) (1) Average (incl. cat. 1, 2, 3) → valid n: (0) 2126, (1) 2661; total 4787CEU Political Science Journal. Vol. 8, No. 2 173 Bibliography: Alsted, Jacob. “Models of Human Motivation in Sociology,” Studies in Sociology. Denmark: Roskilde University (2001). Batbayar, Tsedendamba. “Foreign policy and domestic reform in Mongolia,” Central Asian Survey 22 (2003): 45-59. Blalock, Humbert. M. Jr. Theory Construction: From Verbal to Mathematical Foundations. Eaglewood Cliffs: Prentice-Hall, 1969. Boix, Carles. “Democracy, Development, and International System.” American Political Science Review105 (2011): 809-828. Carothers, Thomas. “The End of the Transition Paradigm.” Journal of Democracy 13 (2002): 5-21. Diamond, Larry. “Is the third wave over?” Journal of Democracy 7 (1996): 23–24. _________. Developing Democracy: Towards Consolidation. Maryland: The Johns Hopkins University Press. E-book, 1999. Firebaugh, Glenn. “Empirics of World Income Inequality.” American Journal of Sociology 104 (2000): 1597-1630. Fish, Steven M. “Mongolia: Democracy without Prerequisites,” Journal of Democracy 9 (1998): 127–140. Fish, Steven M. “The Inner Asian Anomaly: Mongolia’s Democratization in Comparative Perspective,” Communist and Post-Communist Studies 34 (2001): 323–338. Fritz, Verena. 2002. Mongolia: Dependent Democratization, Journal of Communist Studies and Transition Politics.18 (2002): 75-100. Fukuyama, Franscis. “Toward a Theory of Political Development” in The Origins of Political Order: From Prehuman Times to the French Revolution. London: Profile Books. E-book, 2011, 442-619. Ganbat, Damba. “The Mass Public and Democratic Politics in Mongolia.” Asian Barometer Project Office, National Taiwan University and Academia Sinica Working Paper 29 (2004). Ganbat, Damba, RolinTusalem and David Yang. “The Mass Public and Democratic Politics in Mongolia” in How East Asians view democracy, ed. Yun-han Chu, Larry Diamond, Andrew J. Nathan, and DohChull Shin. New York: Columbia University Press, 2008. Gottard, Anna, Giovanni Maria Marchetti, and Alan Agresti. “Quasi-Symmetric Graphical Log-Linear Models.” Scandinavian Journal of Statistics 38 (2011): 447-465. Gunther, Richard and Jose Montero R. “Attitudes toward Democracy in Seven Countries: Dimentional Structure and Behavioral Correlates,” Studies in Public Policy 385 (2004). Højer, Lars. 2009. “Absent powers: magic and loss in post-socialist Mongolia” Journal of the Royal Anthropological Institute 15 (2009): 575–591. Huntington, Samuel P.The Third Wave: Democratization in the Late TwentiethCentury. Norman: University of Oklahoma Press, 1991.Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 174 _________. ”A World of Civilizations” in The Clash of Civilizations and the Remaking of the World Order. New York: Samuel and Schuster, 1996, 1-85. _________. ”After twenty years: the future of the third wave” Journal of Democracy 8 (1997): 3–12. Inglehart, Ronald, “Changing Values and Changing Societies,” in Larsen, Stein U., ed. Challenges of Theories on Democracy. New York: Columbia University Press, 2000, 54-90. Inglehart, Ronald and Christian Welzel. Modernization, Cultural Change, and Democracy. Cambridge: Cambridge University Press, 2005. Kim, Jae-on and Charles Mueller. W.Introduction to Factor Analysis: What It Is and How To Do It. Volume 13.Newbury Park: Sage Publications, 1978. _________. W.Factor Analysis: Statistical Methods and Practical Issues. Volume 14.Newbury Park: Sage Publications, 1978. Knoke, David and PeterJ. Burke.Log-linear Models. Volume 20. Newbury Park: Sage Publications, 1980. Krishna, Arinudh, ed.Poverty, Participation, and Democracy: A Global Perspective. Cambridge: University of Cambridge Press, 2008. Landman, Todd.Issues and Methods in Comparative Politics: An Introduction.3 rded. New York: Routledge, 2008. Lhamsuren, Munkh-Erdene. “The Transformation of Mongolia's Political System: From Semi-parliamentary to Parliamentary?” Asian Survey 50 (2010): 211- 334. Li, Narangoa. “Mongolia and Preventive Diplomacy: Haunted by History and Becoming Cosmopolitan.”Asian Survey 49 (2009): 358-379. Lipset, Seymour M. Political Man: The Social Bases of Politics. New York: Doubleday & Company, 1960. _________. “On the General Conditions of Democracy” in Larsen, Stein U., ed. Challenges of Theories on Democracy. New York: Columbia University Press, 2000, 11-30. Luvsandendev, Sumati. “Mongolia.”KAS Democratic Report(2009). Mishler, William and Richard Rose. “Learning and Re-learning Regime Support: The Dynamics of Post-Communist Regimes,” Studies in Public Policy 326 (2000a). _________. “Political Support for Incomplete Democracies: Realist vs. Idealist Theories and Measures,” Studies in Public Policy 333 (2000b). Munck, Gerardo L. Measuring Democracy: A Bridge between Scholarship and Politics. Baltimore:The Johns Hopkins University Press, 2009. National Statistical Office of Mongolia. Statistical Yearbook 2010.Ulaanbaatar. Norris, Pippa, ed. Critical Citizens: Global Support for Democratic Governance.Oxford: Oxford University Press, 1999. Offe, Claus. “East European Exceptionalism,” in Larsen, Stein U., ed. Challenges of Theories on Democracy. New York: Columbia University Press, 2000, 90- 99.CEU Political Science Journal. Vol. 8, No. 2 175 Pomfret, Richard. 2000. “Transition and Democracy in Mongolia,” Europe-Asia Studies 52 (2000): 149-6 Prohl, Werner and SumatiLuvsandendev.Voters Voices: People’s Perception of Mongolia’s Political and Economic Transition as Reflected in Opinion Polls from 1995 to 2007. Ulaanbaatar: MunkhiinUseg, 2008. Przeworski, Adam and Fernando Limongi. “Modernization: Theories and Facts.” World Politics. 49 (1997): 155-183. Przeworski, Adam, Michael Alvarez, Jose Antonio Cheibub, and Fernando Limongi. 2000. Democracy and development Political Institutions and Well-Being in the World, 1950-1990. Cambridge: Cambridge University Press. E-book, 2000. Robinson, William.S. 1950. “Ecological Correlations and the Behavior of Individuals.” (Reprint) International Journal of Epidemiology 38 (1950): 337-341. Rose, Richard and DohChull Shin. “Democratization Backwards: The Problem of Third Wave Democracies,” Studies in Public Policy 314 (1999). Rose, Richard and William Mishler. “Regime Support in Non-Democratic and Democratic Contexts,” Studies in Public Policy 336 (2000). Rose, Richard. “How Muslims View Democracy: Evidence from Central Asia.” Journal of Democracy, 13(2002): 102-111. _________.”Political Behavior in Time and Space,” Studies in Public Policy 414 (2006a). _________. “Diverging Paths of Post-Communist Countries: New Europe Barometer Trends since 1991,”Studies in Public Policy 418 (2006b). _________.”Democratic and Undemocratic States,” Studies in Public Policy 444 (2008). Rossabi, Morris. Modern Mongolia: From Khans to Commissars to Capitalists. Berkley: University of California Press, 2005. Oakes, Michael. “Commentary: Individual, Ecological, and Multilevel Fallacies,” InternationalJournal of Epidemiology.38 (2009): 361-368. Sabloff, Paula. L. W. “Why Mongolia? The Political Culture of an Emerging Democracy,”Central Asian Survey 21 (2002): 19-36. Sanders, Alan. J. K. “Mongolia's New Constitution: Blueprint for Democracy.” Asian Survey. 32 (1992): 506-520. Schwartz, Shalom. “Basic Human Values: Theory, Methods, and Applications,” unpublished manuscript (2006). _________. “Basic Human Values,” seminar paper (2009). Shapiro, Ian. “Democratic Justice: A Reply to Critics,” in The Real World of Democratic Theory. Princeton: Princeton University Press. E-book, 2011, 297-339. Wachman Alan. M. “Mongolia’s Geopolitical Gambit: Preserving a Precarious Independence, While Resisting “Soft Colonialism””. EAI Fellows Working Paper 18 (2009).Mina Sumaadii: Mongolian Values and Attitudes toward Democracy 176 Waldrauch, Harald. “Incommensurability? On the comparison of Eastern and Southern regime changes,” in Larsen, Stein U., ed. Challenges of Theories on Democracy. New York: Columbia University Press, 2000, 99-147. Westad, Odd, A.The Global Cold War: Third World Interventions and the Making of Our Times. Cambridge: Cambridge University Press, 2006. Wucherpfennig, Julian and Franziska Deutsch. “Modernization and Democracy: Theories and Evidence Revisited.” Living Reviews in Democracy 1 (2009). Empirical Data Sant Maral Foundation. Ulaanbaatar. Mongolia. “Politbarometers 1995-2012.” World Bank. GDP per capita (current US$) 1982-2011.