Stable Support for Democracy in East and Southeast Asia? Examining Citizens’ Trust in Democratic Institutions

Recent democratic regressions and crises suggest democracy is at risk across East and Southeast Asia. One of the factors that can determine democratic stability are citizens’ attitudes. While previous research has concentrated on support for democracy-in-principle, this contribution argues that it is political trust, i.e. support for democracy-in-practice, which is crucial for democratic stability. For democracies to be stable, political trust should be high as well as rooted in long-term factors like liberal democratic value orientations or social trust to protect it from short-term fluctuations following economic crises or political scandals. This contribution therefore examines not only the current levels and development of political trust but also whether it is influenced more by long-term factors (liberal democratic value orientations, social trust) or short-term factors (economic performance evaluations, incumbent support). The empirical analysis shows political trust in five East and Southeast Asian democracies (Indonesia, Japan, Mongolia, South Korea, Taiwan) to be mostly mediocre and primarily dependent on economic performance evaluations and incumbent support. Among the five democracies, citizens in Japan appear most resilient to democratic regressions; on the other hand, Taiwanese democracy seems least equipped to master future crises.


Introduction
The regression of democracies around the globe has worried observers everywhere [32]. is at stake across all of East and Southeast Asia. For instance, Mongolia faced "yet again another constitutional crisis" [23] in 2019, when parliament passed a bill allowing the National Security Council to dismiss judges and the head of the anti-corruption department. In Indonesia, authoritarian populist Prabowo Subianto came close to being elected president in both 2014 and 2019 [52], and Taiwan had to face targeted disinformation campaigns by authoritarian neighbour China during the runup to its 2020 presidential elections [4]. While the literature on democratic regression identifies a multitude of factors that determine the chances of democratic survival, most scholars agree that citizen attitudes play a key role for the stability of democracy [51]. After all, democracy is built on the rule of the people and heavily relies on at least some level of goodwill and compliance from its citizens to function. If citizens decline to participate in democratic politics or to accept its outcomes, democracy may become easy prey for anti-democrats. On the other hand, when citizens' support for the democratic political regime is strong and stable, antidemocratic actors will have a harder time pushing their agenda and dismantling core democratic principles.
Previous literature interested in the role of citizen attitudes for democratic stability has predominantly examined citizens' support for democracy, democratic principles, and democratic value orientations. In this context, Claasen [11] has shown that public support for democracy helps democracy survive. Digging deeper into the roots of democratic resilience, Brunkert et al. [8] argue and demonstrate empirically that emancipative values, i.e. values emphasizing core liberal democratic values such as autonomy and freedom, are crucial for democratic survival and that democratic backsliding and autocratization only occur in countries with underdeveloped emancipative values. These results echo previous findings by Welzel [54], who finds liberty aspirations to be central for both democratization and the prevention of democratic regressions. These contributions emphasize the relevance of citizen attitudes for democratic stability; yet studies on support for democracy-in-principle in East and Southeast Asia show that support for the values underpinning liberal democracy is neither wide nor deep in the region [10,52]. Does that mean that East and Southeast Asian democracies are all at the brink of collapse? I argue that, while support for democracy-in-principle is clearly important for the long-term stability of democracy, support for democracy-in-practice has even more immediate consequences for democratic stability. In particular, political trust, i.e. citizens' attitudes towards the concrete democratic regime and its institutions, has direct behavioural consequences that can be key for the stability of a democratic political regime. Among others, prior research has shown political distrust to relate to voting for populist and antisystem parties [5,46], to demands for institutional reform [12], and to reduced compliance with the law [35,49] -behaviours that can potentially destabilize democracy and make it vulnerable to authoritarian attacks from both within and without. If we are interested in how stable East and Southeast Asian democracies are, investigating political trust among their citizens can help us assess how well equipped they are to face the challenges ahead of them.
The present contribution therefore examines current levels of citizen confidence in five core democratic institutions: government, parliament, courts, civil service, and police across the five countries in East and Southeast Asia that can presently be classified as democratic 1 : Indonesia, Japan, Mongolia, South Korea, and Taiwan. 2 High levels of trust alone, however, are not sufficient for democratic stability. Political trust should also be reasonably stable, regardless of economic crises or major political scandals. For that to be the case, it needs to be rooted in a firm belief in the political and social values underpinning democracy, i.e. in liberal democratic value orientations and social trust, rather than depend primarily on short-term factors like satisfaction with the economy or the popularity of the incumbent government. This is especially important for institutions of the state like the judiciary, police, and civil service, where -unlike for the representative institutions government and parliament -citizens have little impact on who holds office and can thus not vent any dissatisfaction through electoral means. Broad and steady citizen support for these institutions therefore forms the backbone of democratic stability and should not be affected by day-to-day politics. Consequently, this contribution not only tracks the development of political trust over time but also investigates into the sources of political trust for each country, focusing on trust in the institutions of the state (courts, civil service, and police). It examines two long-term factors -liberal democratic value orientations and social trust -and two short-term factors -economic performance evaluations and incumbent support -as potential sources of trust. Where trust even in the institutions of state is mainly rooted in economic performance evaluations or the popularity of the incumbent government, we can expect it to be much more susceptible to short-term fluctuations than where it is predominantly rooted in relatively stable characteristics like liberal democratic value orientations or social trust. Using both historical and recent survey data from the Asian Barometer Survey (2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2014)(2015)(2016), it helps gauge how resilient the five East and Southeast Asian democracies can be against future crises of democracy. It finds that levels of political trust are generally mediocre and fluctuate over time, with government and parliament receiving less trust than courts, civil service, and especially the police. Political trust also depends primarily on short-term factors like economic performance evaluations and incumbent support across the entire region. There are, however, considerable differences between individual countries, with Japanese society appearing to be best prepared to weather future crises of democracy and Taiwan being most at risk of democratic regression.

The Relevance of Political Trust
Political trust, defined as citizens' confidence that the political system, its institutions, or actors will "do what is right even in the absence of constant scrutiny" [38], is a central concept in political-culture research. Scholars have examined the relevance of political culture mainly in the context of democratic consolidation, where citizen support has long been attributed a central role in determining the success of democratic consolidation processes [14,22,31]. In this tradition, scholars have previously focused primarily on the role of support for democratic principles, arguing that democracy can only be sustained in the long run if a majority of citizens uphold values like freedom or equality [8,11,54]. While support for democratic principles is surely important for what type of political regime will eventually emerge in a given country, political trust has more immediate consequences for the stability of any given democratic political regime. With the value orientations underlying support for democratic principles being very broad and abstract concepts, they are unlikely to guide human behaviour; the more concrete attitudes underlying political trust are more likely to have behavioural consequences. Prior research supports this proposition: Among other phenomena, low levels of trust are associated with the rise of unconventional and elite-challenging actions [25,53], demands for institutional reforms [12], and a lower willingness to comply with the law [35,49]. In addition, disenchantment with and lacking trust in the political system have been identified as a major driver of the recent rise of populist parties, with citizens less trusting of the political system more likely to vote for not only populist but also extremist and anti-democratic parties [5,46]. Democratic value orientations and support for democracy in the abstract, on the other hand, seem to have little to none behavioural consequences [16,44]. 3 Despite these undesirable behavioural consequences, some scholars, especially those in the critical-citizens tradition, have argued that political distrust is not necessarily worrisome for the stability of democracy and might even be desirable at least to some extent [41]. They ground this assertion in the argument that low levels of trust in established democracies are simply an expression of rising expectations and a soundly sceptical stance towards authority. While the latter part of the argument may be true -prior research has demonstrated that more demanding understandings of democracy, for instance, relate to reduced support for and satisfaction with the existing democratic institutions [21] -we should never consider low levels of trust as unproblematic, especially when they concern not only incumbent officeholders but the very essence of democratic institutions. Even though a certain amount of distrust against officeholders may help keep authorities in check, a general lack of trust in core institutions of the democratic state like the courts or civil service is hardly beneficial for the functioning and stability of democracy as it makes citizens disobey the law, leads to protest behaviour, and enhances the chances of antidemocratic populists to succeed in elections -all of which can trigger or intensify crises of democracy. Accordingly, in her empirical analysis, Dorenspleet [16] concludes that "a growing number of dissatisfied democrats are a sign of democracy in decline" rather than an indicator of a healthy democracy, and Yap [56] argues that political trust "buffers" new and emergent democracies against public pressure for reform. For East and Southeast Asian democracies to be resilient to anti-democratic challenges, we would thus want political trust among their citizens to be high.

Long-term and Short-term Sources of Political Trust
It is not only, however, the sheer levels of political trust that are crucial for the long-term stability of democratic political regimes. In addition to being on a fairly high level, political trust should also be resilient against short-term fluctuations to ensure the stability of any given democratic regime. This brings us to the sources of political trust: if political trust is mainly determined by factors that are prone to short-term fluctuations, political trust itself is likely to vary considerably over time, making support for democracy-in-practice unstable. If, in contrast, political trust is based on factors that are typically stable, political trust is likely not to vary considerably over time, making support for democracy-in-practice stable over a longer period of time.
Previous literature has discussed a myriad of determinants of political trust (for a cursory overview, see [36]). Two of the most central long-term determinants are liberal democratic value orientations and social or interpersonal trust. Recurring to Easton's idea of "diffuse political support", i.e. support for the current regime that is rooted in a perceived alignment of the regime's structure with citizens' individual value orientations, scholars have argued that liberal democratic value orientations such as the belief in political pluralism or vertical and horizontal accountability can act as reference points for what values should be realized within and through the political regime [17,19]. As long as there is congruence between citizens' value orientations and the political system's institutions, political trust should be high. Accordingly, empirical studies find that citizens who hold more liberal democratic value orientations extend more support to their democratic political regime [9,47]. Even though these values are not widespread in East and Southeast Asian societies [10,52], political trust may still relate to them or their absence. In addition, scholars have ascribed a central role to social trust as a prerequisite of political trust since it indicates a "trusting personality" [20]. While effect sizes are usually small, empirical studies consistently find a positive effect of interpersonal trust on political trust [25,57]. Both liberal democratic value orientations and social trust pertain to the so-called culturalist tradition of research on political trust. This culturalist tradition emphasizes characteristics of the individual citizen, i.e. factors exogenous to the political system, as sources of trust, and assumes that these are predominantly shaped through socialization experiences and thus not subject to rapid change. This is certainly the case with respect to liberal democratic value orientations and social trust: both have been shown to remain relatively stable over the life course, resulting in only gradual change over time on the aggregate level through generational replacement [13,24].
As far as short-term determinants are concerned, the literature has prominently discussed economic performance evaluations and incumbent support as sources of political trust. Echoing the Eastonian idea of "specific support", i.e. support for the current regime that is based on citizens' satisfaction with the regime's outputs [18], scholars in the so-called institutionalist tradition have argued that citizens form their attitudes towards the political system based on their experiences with the regime's actors and outputs, i.e. factors endogenous to the political system. Taking a rational-choice perspective, they hypothesize that citizens will extend more trust to the political system and its institutions if they perceive them and the incumbent authorities as delivering the goods they desire [26]. Accordingly, support for the incumbent government has been shown to be a significant determinant of political trust: citizens who voted for the winning party and those who are satisfied with the current government express more political trust than those who voted for the opposition and those who disapprove of the current government [2]. With regard to the goods that a political system should provide in the eyes of citizens, prior research often emphasized the role of economic performance, i.e. the provision of economic goods such as economic growth, employment, and stable currency, and more positive evaluations of the current and prospective macroeconomic situation are persistently found to increase political trust [50]. As factors endogenous to the political system, economic performance evaluations and incumbent support are subject to rather short-term fluctuations according to changes in the macroeconomic situation or composition of government [3]. This is not necessarily the case for the third major determinant discussed in the institutionalist tradition of research: democratic performance evaluations. Democratic performance is high if a political system provides political rights and freedoms, e.g. civil liberties, horizontal accountability, or rule of law, and political trust increases when citizens evaluate their regime's democratic performance more positively [39]. However, a political system's democratic performance is unlikely to change rapidly and, consequently, neither are citizens' democratic performance evaluations. Owing to this hybrid nature of democratic performance evaluations, the following analysis will focus on liberal democratic value orientations and social trust as long-term sources and economic performance evaluations and incumbent support as short-term sources of political trust. 4 For political trust to be stable, we would thus want it to be based primarily in liberal democratic value orientations and social trust rather than in economic performance evaluations and incumbent support.
Comparing the effects of culturalist and institutionalist determinants, prior research typically finds institutionalist determinants like economic performance evaluations and incumbent support to exert a stronger effect on political trust than culturalist determinants like liberal democratic value orientations and social trust [20,37,40,55]. We thus should not expect liberal democratic value orientations and social trust to be dominant in determining political trust in the East and Southeast Asian democracies. However, the extent to which economic performance evaluations and incumbent support outweigh the effects of these long-term factors in the region and within each individual country can still give us valuable information on how stable political trust is likely to be.

Data and Measurement of Key Variables
To examine how widespread political trust in the democratic institutions is among Asian citizens and whether this trust is based on long-term factors like value orientations and social trust or rather on short-term factors like economic performance evaluations and incumbent support, this study uses individual-level survey data from the Asian Barometer Survey (Waves 1-4, 2001-2016). 5 These data cover a total of five East and Southeast Asian democracies: Indonesia, Japan, Mongolia, South Korea, and Taiwan. Other countries in the region are not currently classified as democratic (e.g. China, the Philippines, Thailand, Vietnam) or were not included in the survey data (e.g. Timor-Leste). With respect to the central variable political trust, the empirical analyses look at confidence in five of modern democracies' core institutions: government, parliament, courts, civil service, and police. Particular emphasis is placed on trust in courts, civil service, and police: Other than representative institutions like government and parliament, these three institutions of the state are not typically subject to direct political competition and cannot be held directly accountable for policymaking outputs. From a normative point of view, unlike support for government or parliament, support for these institutions should thus not depend on day-to-day politics. Gauging citizens' confidence in the courts, civil service, and police should therefore give us a good indication of how much they support the broader democratic political system they live in.
For the independent variables, liberal democratic value orientations are measured by a factor composed of four items asking respondents about core liberal democratic values: vertical accountability, electoral democracy, and media freedom (for the measurement model, cf. online appendix). The standard item asking respondents whether they think that most people are trustworthy assesses social trust. A question inquiring about the country's present economic situation gauges economic performance evaluations. Finally, I take into account respondents' satisfaction with the current government to probe incumbent support. For the exact item wordings, see the online appendix.

Levels and Trends in Political Trust Across East and Southeast Asia's Democracies
So how much can East and Southeast Asian democracies rely on popular support? A glance at current levels of institutional trust shows that of the currently five democracies in the region, Indonesia has the strongest backing from its citizens. 5 The analysis is based on survey data collected by the Asian Barometer Project between 2001 and 2016. The Asian Barometer Project was co-directed by Professors Fu Hu and Yun-han Chu and received major funding support from Taiwan's Ministry of Education, Academia Sinica and National Taiwan University. The Asian Barometer Project Office (www. asian barom eter. org) is solely responsible for the data distribution. The author appreciates the assistance in providing data by the institutes and individuals aforementioned. The views expressed herein are the author's own.
In contrast, a majority of citizens in South Korea, Taiwan, Mongolia, and Japan do not express particular trust in their core democratic institutions (Fig. 1).
Citizens' views are especially critical when it comes to the representative institutions, which are most subject to political contention: government and parliament. This is hardly surprising given the partisan nature of these two institutions. Democratic political competition and elections always divide the population into winners and losers, with government and opposition camps often depicting each other as adversaries rather than partners. In addition, parliament is the one democratic institution where political conflicts are fought out in the open on a daily basis. Different factions engage in often heated debates about policies, and some conflicts even turn violent. For example, parliamentary debate on the rather mundane topic of school meals ended in members of the South Korean parliament pushing and shoving each other in a fight over the speaker's podium in December 2010. Only a week later, the budgetary debate descended into a mass brawl, leaving one MP hospitalized with a head wound. In Taiwan's Legislative Yuan, perhaps the world's most infamous parliament in this respect, parliamentary debates regularly turn violent, with legislators becoming physical on numerous occasions, including pulling each other's hair and biting one another. With media coverage typically focusing on these adversarial and hostile aspects of parliamentary politics-rarely covering what in most cases is a rather cooperative way of work in the day-to-day operations of parliament -it is little wonder citizens do not see parliament as being a very trustworthy institution. However, citizen trust is also not particularly high when it comes to two of the three institutions of the state, courts and civil service. Across the region, of the five core institutions of democracy, only the police receive support from a majority of citizens. While the police seem to receive a roughly equal amount of confidence across all five countries, cross-country differences are considerable when it comes to trust in the courts and, to a somewhat lesser extent, the civil service. Trust in the courts is especially low in Taiwan and Mongolia, whereas the Japanese and Indonesians exhibit remarkably high levels of confidence in their judicial system. With respect to trust in civil service, Indonesia again stands out as the country with the highest levels of citizen trust; at the other end of the spectrum, Japanese, Taiwanese, and South Koreans appear least trusting of their countries' civil service.
Going beyond citizens' current views of their democratic institutions, data from previous years show notable fluctuations in citizens' backing for democracy since the early 2000s, with severe drops in confidence often following devastating events or major political scandals (Fig. 2). For instance, citizens' trust in government, parliament, civil service, and judiciary alike plummeted in Mongolia after the dzud of 2009-2010, a natural disaster of unprecedented scale, had killed millions of animals and destroyed the livelihoods of tens of thousands of people. Similarly, South Koreans lost faith in pretty much all their core democratic institutions after the so-called X file scandal had revealed illegal wiretapping and large-scale bribery in 2005. Whereas the South Korean institutions have largely managed to recover from this sudden loss in trust-perhaps at least in part thanks to the subsequent change in government -the Mongolian political system continues to suffer from the plunge in confidence until today, with levels of trust not nearly having returned to the precrisis years. And in Japan, citizens' trust in government dropped to a record low in the aftermath of the 2011 Fukushima disaster. Unlike in Mongolia and South Korea, however, Japan's other institutions remained virtually unaffected, indicating that ordinary people assigned responsibility for what happened primarily to the government.
In addition to these patterns of volatility, the longitudinal data reveal a steady decline in citizens' trust in some of their core institutions. This is the case for example in Taiwan, where citizens have become less and less confident in their democratically elected government as well as the country's courts since the early 2000s. Similarly, Indonesians' views of their national parliament have continuously decreased over the past 15 years. On the other hand, we can also observe some more promising trends: following major crises, citizens' trust in their democratic institutions has at least begun to recover in both South Korea and Mongolia.
Only one institution seems to defy these general patterns of volatility: Citizens' views of the police have remained remarkably stable over the past 20 years in South Korea, Japan, Indonesia, Mongolia, and Taiwan. In fact, the police is the only institution which citizens across East and Southeast Asia express more trust in today than they did in the early 2000s. For all other core democratic institutions, the overall trend points to a decline in citizen confidence rather than an increase. In sum, these results do not bode particularly well for the stability of East and Southeast Asian democracies. Nevertheless, the findings look more promising if we concentrate on the institutions of police and civil service as well as, albeit to a lesser extent, courts: citizen trust in these three institutions of state is not nearly as low as for government and parliament, fluctuates less over time, and does not decline over time as steadily and to the same extent.

Sources of Political Trust in East and Southeast Asian Democracies
Turning to the sources of political trust, I now focus on citizens' confidence in courts, civil service, and police. As argued above, citizens' views of these institutions of state should depend to a much lesser extent on day-to-day political outputs than citizens' views of the representative institutions government and parliament. Stable as well as ample citizen trust in these three institutions can therefore be considered the backbone of democratic stability. While the preceding analyses have shown that especially the police and civil service indeed fare considerably better in the eyes of citizens than government and parliament, they have also shown at least some fluctuation in levels of trust in all democratic institutions except the police over the past 20 years. The following analyses will consequently investigate more deeply into the sources of trust in the three institutions of state. If these are primarily rooted in long-term factors like liberal democratic value orientations and social trust, we may be optimistic when looking into the future of East and Southeast Asia's democracies; if, however, they depend heavily on short-term factors like economic performance evaluations and incumbent support, democratic stability may be at stake across the region.
The following analyses will proceed in two steps. First, they will regress political trust on both long-term and short-term factors to identify which predictors exert a significant influence on political trust in each of the five East and Southeast Asian democracies. Second, they will examine the relative explanatory power of longterm and short-term factors by regressing political trust on only long-term factors and only short-term factors, respectively, and compare how much of the variance of political trust these long-term and short-term factors, respectively, explain. Following the finding from the previous section that trends in institutional trust run mostly parallel for all three institutions and as trust in courts, civil service, and police indeed seems to form a unidimensional construct (cf. online appendix), the main analyses will use the factor score of political trust in institutions of state as the dependent variable. All analyses regarding sources of political trust control for political interest, respondents' financial situation, level of education, gender, and age. Robustness checks will additionally look at how long-term and short-term factors affect trust in each political institution individually, as well as include democratic performance evaluations as a control variable. All analyses of sources of political trust rely on the most recent survey data, i.e. data from the fourth wave of the Asian Barometer Survey (2014)(2015)(2016). This allows us to include more sophisticated measures for both long-and short-term sources of political trust, which are unfortunately not available for all waves of the Asian Barometer Survey. 6 If we look, first, at which predictors exert a significant influence in East and Southeast Asia (Fig. 3), we find long-term social trust and short-term economic performance evaluations as well as incumbent support to increase political trust in institutions of state in each of the region's five democracies. Liberal democratic value orientations, in contrast, do not emerge as a significant predictor of political trust in most countries. Whether people are committed to liberal democratic values like vertical accountability and media freedom does not seem to make much of a difference for how they view the core democratic institutions of state in their country. On the other hand, those who are more trusting of their fellow citizens do express more confidence in their country's courts, civil service, and police, as do those who view the current national economic situation as more favourably and those who are more satisfied with the incumbent government.
Taking a closer look at individual countries, South Korea stands out as the country in which liberal democratic value orientations play by far the largest role for political trust-and in fact as the only country where they play a substantial role at all. With regard to social trust, we find the strongest effects on political 6 It is possible to conduct a somewhat similar analysis for data from the third wave of the Asian Barometer Survey, and the results are included in the online appendix (Fig. 1, Table 4). They mostly corroborate the findings for the fourth wave: Social trust, economic performance evaluations, and incumbent support have a significant effect on political trust in all countries (with the exception of social trust in Mongolia), while liberal democratic value orientations do not emerge as a significant predictor of political trust in most countries. Comparing the relative explanatory strength of different types of determinants, we again find long-term and short-term factors to have similar explanatory power in Japan and Mongolia, whereas short-term factors clearly outweigh long-term factors in the other countries. The online appendix contains a more detailed discussion of the third wave results. trust in Japan and, to a somewhat lesser extent, in Taiwan and South Korea. For Indonesia and in particular Mongolia, social trust takes only a comparatively minor role in shaping citizens' trust in institutions of state. Perhaps unsurprisingly, then, short-term economic performance evaluations exert a slightly greater effect in these three countries than they do in Japan and South Korea. When it comes to incumbent support, Indonesians give by far the greatest weight to how satisfied they are with the current government, followed by the Taiwanese. Incumbent support appears of comparatively minor importance in Mongolia.
Most of these patterns are reproduced when examining trust in individual institutions instead of the composite measure of trust in state institutions as well as when controlling for democratic performance evaluations (cf. online appendix). The most notable exception concerns trust in the police and the role of economic performance evaluations. Unlike for the other institutions, how favourably citizens view the current national economic situation does not significantly affect their confidence in the police in all East and Southeast Asian democracies: the effect only reaches conventional levels of statistical significance for Taiwan and Indonesia, and generally remains much weaker than for the other institutions or the composite measure. This could point to the police being perceived as farther removed from day-to-day politics than the other institutions and/or to citizens evaluating this particular institution based on different criteria, for example on the level of security they feel is provided. Turning to the second part of the analysis, we find the relative explanatory power of long-term factors to be weaker than that of short-term factors in all five East and Southeast Asian democracies (Table 1).
Yet, the extent to which short-term factors outperform long-term factors varies considerably across countries. On the one end of the spectrum, long-term and shortterm determinants are relatively balanced in both Japan -where both long-term and short-term factors have a rather strong effect -and Mongolia -where both long-term and short-term factors have only weak effects. On the other end of the spectrum, the institutions of state of Taiwan and Indonesia seem to rely primarily on short-term factors like economic performance evaluations and incumbent support.
Again, these general patterns are mainly reproduced when examining trust in individual institutions instead of the composite measure of trust in courts, civil service, and police (cf. online appendix). The most notable difference, once again, concerns trust in the police. For citizens' confidence in this particular institution, long-term and short-term factors appear much more balanced than for the other two institutions or the composite measure across all of East and Southeast Asia. For South Korea, long-term factors like liberal democratic value orientations and social trust even show more explanatory power than short-term factors like economic performance evaluations and incumbent support, even though both types of predictors only explain trust in the police to a very limited extent. Additionally controlling for democratic performance evaluations mostly obliterates the differences between longterm and short-term factors as democratic performance evaluations alone explain a considerable share of the variance of political trust.

Conclusion
This contribution set out to explore how stable East and Southeast Asian democracies are based on their citizens' attitudes. Building on previous literature that links low levels of political trust to a whole range of system-threatening behaviours, it argued that citizens' trust in core democratic institutions, rather than support for democracy-in-principle, is the most consequential attitude for the immediate stability of any given democratic political system. Empirically, it first examined levels of political trust in each of five East and Southeast Asian democracies (Japan, Indonesia, Mongolia, South Korea, and Taiwan), and found that levels of political trust are generally mediocre. Only in Indonesia, a majority of citizens express confidence in their core democratic institutions. Trust is generally lower for government and parliament, i.e. those institutions which are most subject to political contention and can be held accountable most directly for dayto-day political outputs. Among all five core democratic institutions, the police by far receive the highest amount of trust from citizens across the entire region; the courts and civil service are typically being viewed more critically than the police but more favourably than government and parliament. These findings are in line with results from other democracies, where trust in representative institutions like government and parliament is generally also lower than trust in institutions of state like courts and police [9,39,48]. Data from previous waves of the Asian Barometer Survey ranging back to the early 2000s revealed considerable fluctuations in political trust across East and Southeast Asia. Most notably, severe drops in citizen confidence occurred after natural disasters or major political scandals like the 2009-2010 "dzud" in Mongolia or the 2005 "X file scandal" in South Korea. While levels of trust mostly recovered from these crises during the following years, the general trend in all East and Southeast Asian democracies still points to an overall decline in citizen confidence over the years. Going beyond the sheer levels of political trust, I further argued that for a citizenry to be resilient to crises of democracy, its trust in core democratic institutions needs to be rooted in factors that are likely to remain stable at least in the medium run, like liberal democratic value orientations and social trust, rather than determined by factors likely to change on short notice and in times of crisis, like economic performance evaluations and incumbent support. Yet, the empirical analysis showed short-term factors to be stronger predictors of political trust in institutions of state than longterm factors in every East and Southeast Asian democracy. Again, these results are largely in line with prior research from other world regions [20,40]; however, the fact that in most countries short-term factors vastly outperformed long-term factors and that they did so even though the analysis deliberately excluded government and parliament, the two institutions most directly responsible for day-today political outputs and most clearly linked to present incumbents, makes citizen support for East and Southeast Asian democracies appear remarkably dependent on their economic performance and government popularity nevertheless. Of course, citizen support is not the only factor influencing democratic stability. Next to citizens' attitudes, the literature on democratic resilience and regression discusses a broad range of factors that can contribute to the survival or breakdown of democratic rule. For instance, agency-based theories focus on core political actors and the decisions these elites make [29,34,42]. Institutional theories argue that its specific institutional makeup and the resultant levels of vertical and horizontal accountability as well as government efficaciousness determine how susceptible a given democratic regime is to autocratizing tendencies either amongst its political elite or amongst its citizenry [30,43,45]. Others view structural-economic variables such as economic growth or income distribution as decisive for democratic stability [1,6,7], or examine international factors like interactions with other democracies or foreign aid [15,27,28].
Yet, most scholars still agree that citizen attitudes play a key role for the stability of democracy [51], and in this respect, the present findings paint a rather pessimistic picture: citizen trust in democracy's core institutions is far from ample, can fluctuate considerably over time, and depends primarily on short-term factors like economic performance evaluations and incumbent support. There are, however, considerable variations between countries when it comes to both levels of political trust and its potential for long-term stability. For one, Indonesians express much more support for their core democratic institutions than citizens in any East Asian democracy. Nevertheless, their trust mainly depends on short-term factors, especially incumbent support, and is therefore highly susceptible to drops in the popularity of the incumbent government. This means that, for example, a political scandal involving reigning president Joko Widodo could make levels of trust plummet and the stability of Indonesian democracy could suddenly be at stake. On the other hand, while the Japanese are much more sceptical than the Indonesians when it comes to their government and parliament, they also express comparatively high levels of confidence in their country's courts and police. In addition, Japanese trust in these institutions relies to a considerable extent on long-term social trust and should thus not be at risk even in times of crisis. The remarkable stability of trust in all institutions except the government after the 2011 Fukushima disaster exemplifies this relative resilience to short-term crises. Results for other countries do not bode equally well. In Taiwan, comparatively low levels of trust coincide with a strong dependence on short-term factors, leaving Taiwanese democracy vulnerable to economic crises or dips in government popularity. Finally, while political trust seems more stable in South Korea and Mongolia, the already low levels of citizen confidence in core democratic institutions in these countries may foreshadow future problems of democratic stability.
In sum, then, this study's findings do not make East and Southeast Asian citizenries seem like bulwarks against democratic regressions in times of crisis, and the region as a whole does not appear to be particularly well equipped to withstand future challenges to democracy from a political-culture point of view. Even though democratic survival is ultimately dependent on more than a single factor, citizenbased democratic stability seems only weakly developed in East and Southeast Asia. If democrats cannot find ways to enhance and stabilize citizens' trust in core democratic institutions or develop other means of safeguarding the democratic system against both internal and external attacks from authoritarian forces, the prospects thus look grim for democracy in the region. With the Philippines already sliding down the authoritarian road, it might only be a matter of time until the next East or Southeast Asian democracy starts to fall. Given its already below-average levels of citizen trust, the overall downward trend, and its extraordinarily high dependence on short-term factors like economic performance and incumbent popularity, Taiwan appears especially at risk. Even though the most recent attempts of election meddling and misinformation campaigns orchestrated by China proved unsuccessful in the presidential elections of 11 January 2020, observers agree that incumbent president Tsai Ing-wen's bid profited immensely from the current situation in Hong Kong and that the political landscape is far from as pro-democratic as Tsai's landslide 1 3 victory makes it look. Should her opponents become more successful in damaging Tsai's reputation or should a major economic crisis hit the country, the tide could turn very quickly.