On the Distinct Effects of Left-Wing and Right-Wing Populism on Democratic Quality

This study examines the differences and commonalities of how populist parties of the left and right relate to democracy. The focus is narrowed to the relationship between these parties and two aspects of democratic quality, minority rights and mutual constraints. Our argument is twofold: first, we contend that populist parties can exert distinct influences on minority rights, depending on whether they are left-wing or right-wing populist parties. Second, by contrast, we propose that the association between populist parties andmutual constraints is a consequence of the populist element and thus, we expect no differences between the left-wing and right-wing parties. We test our expectations against data from 30 European countries between 1990 and 2012. Our empirical findings support the argument for the proposed differences regarding minority rights and, to a lesser extent, the proposed similarities regarding mutual constraints. Therefore we conclude that, when examining the relationship between populism and democracy, populism should not be considered in isolation from its host ideology.


Introduction
Populist actors around the world have gradually evolved into influential political forces in various countries and regions.Independent of the region, they share the ideas of anti-elitism and people centrism.On this basis, they can challenge common democratic rules, including those of liberal democracy (Plattner, 2010), according to which power must be restrained and individual rights protected.Through the populist lens, features of liberal democracy, such as systems of checks and balances, undermine the proper implementation of the general will, which they claim to be the only true representative of.Thus, their presence can have a negative impact on the quality of democracy if populist parties challenge these institutions, particularly when they are in government (Albertazzi & Mueller, 2013).
Populist parties, however, are not only characterized by their populist element but also by their host ideology (Mudde, 2004).Thus, they can take the form of rightwing populist parties (Mudde, 2007), left-wing populist parties (March, 2011), or centrist populist parties (Havlík & Stanley, 2015).In other words, populist parties differ on a wide-ranging set of issues such as the promotion of exclusive (right-wing populist parties) or inclusive (left-wing populist parties) societies (Mudde & Rovira Kaltwasser, 2013).These differences have been shown to manifest themselves in the behavior of populist par-ties, for instance with regards to parliamentary voting where the populist element plays little to no role (Otjes & Louwerse, 2015).Despite these well-known differences, however, there is little debate in the literature about whether the postulated relationship between populist parties and democracy is a function of their host ideology, their populist element, or both.This article seeks to fill this void, taking as its starting point the discussion between populism and liberal democracy (Mudde & Rovira Kaltwasser, 2012).
We adopt the proposition that the host ideology, like the populist element, is central to the actions that parties take.We argue that the role of ideology is essential to understanding why populist parties of varying host ideologies relate differently to subdimensions of liberal democracy, namely political inclusion (minority rights) and mutual constraints.Focusing on left-wing and rightwing populist parties, we anticipate the host ideology to be the deciding factor for how these parties relate to the dimension of political inclusion.In comparison to rightwing populist parties, we expect left-wing populist parties to be associated with more positive effects on minority rights.For the second dimension, mutual constraints, we expect the populist element to play the central role and in consequence, expect no differences in associations between populist parties of different host ideology.
Empirically, we test our propositions against data from 30 European countries from 1990 to 2012.This dataset, although limited to one region, allows us to test our arguments for a diverse set of populist parties.The results lend support to our argument that host ideologies matter for how certain populist parties relate to democracy and liberal democracy in particular as differences in effects of left and right-wing populists occur for minority rights.At the same time, the results do not suggest a strong association between populist parties and mutual constraints.Therefore, this study highlights the need to investigate subdimensions of (liberal) democracy (Houle & Kenny, 2016;Immerzeel & Pickup, 2015) to generate a better understanding of the complex relationship between populist parties and democracy.

Populism and (Liberal) Democracy
Following the ideational approach (Hawkins, 2009), populism constitutes a set of ideas.Despite varying definitions that can be subsumed under the ideational approach, most studies consider at least four attributes central to populism: people centrism, the perception of the people as a homogenous entity with a general will, antielitism, and the depiction of a permanent crisis (Rooduijn, 2014).Mudde (2004, p. 543; italics original) summarizes the central attributes in his widely referenced minimal definition of populism, which we draw on in this article, calling populism a thin-centered "ideology that considers society to be ultimately separated into two homogeneous and antagonistic groups, 'the pure people' and 'the corrupt elite', and which argues that politics should be an expression of volonté générale (general will) of the people".The view of populism as an ideological construct, however, remains at the center of many debates.Another large branch in the literature, for instance, discusses populism as a discourse (or frame) (e.g., Aslanidis, 2016Aslanidis, , 2017;;Jagers & Walgrave, 2007).However, even among those that speak of populism as a discourse, some acknowledge that populists ultimately may implement their idea of politics as an expression of the general will (Müller, 2016).Furthermore, both branches in the literature share the view that populism can appear across the ideological spectrum giving populism its chameleon characteristic (Taggart, 2000).In the ideological approach, for instance, populism as a thin-ideology is said to attach itself to different host ideologies (Stanley, 2008).
Because "populism indirectly questions the procedural minimum that lies at the heart of our current definitions of democracy" (Hawkins, 2010, p. 37), scholars using different conceptualizations discuss the relationship between populism and (certain forms of) democracy.Most notably, they focus on how populism relates to democracy in general (Müller, 2016), representative democracy (Canovan, 1999), and liberal democracy (Kriesi, 2014;Mudde & Rovira Kaltwasser, 2012;Plattner, 2010).These studies share a focus on populism's homogenous view of society in which the common will of the people is to be articulated in an unmediated way and implemented without any restrictions (Caramani, 2017).1 Perhaps for this reason, there has been a recent focus on populism and the quality of liberal democracy (cf.Huber & Schimpf, 2016a, 2017;Pappas, 2014Pappas, , 2016)).After all, the essence of liberal democracy is that power can never be absolute as it is characterized by "the intrinsic importance of transparency, civil liberty, the rule of law, horizontal accountability (effective checks on rulers), and minority rights" (Coppedge et al., 2011, p. 253). 2 Minority rights and horizontal accountability in particular are two features of liberal democracy that run counter to the populist understanding of how democracy ought to function.Populist actors depict a homogenous society (the people) and highlight the necessity for politics to follow the general will without any unnecessary restrictions, implemented by the populists themselves as the only true representatives of the people. 3Therefore, some scholars argue that populist actors can have a negative impact on democracy, and in particular, on liberal democracy (e.g.Mudde & Rovira Kaltwasser, 2012;Ruth, 2017).For as long as populists are not in power, they are rarely in a position to implement their ideas.Instead, they may even have a corrective function as they highlight institutional shortcomings (e.g.Müller, 2002), mobilize otherwise unrepresented groups of voters (e.g.Hanley, 2012), and articulate issues or protest (e.g. de Lange & Akkerman, 2012). 4rom this debate, we can derive the expectation that populist actors may have positive side effects but generally, relate negatively to (liberal) democratic quality, in particular when they are in government (Albertazzi & Mueller, 2013).Here, their presence can result in changes or even the erosion of important components such as the system of checks and balances, as can be seen in countries governed by populist parties, such as contemporary Poland (Markowski, 2016) and Hungary (Batory, 2016).However, the argument hinges on the assumption that all populist actors must share a similar understanding of a homogenous society whose general will functions as the guiding principle for political decisions and shall not be infringed by unnecessary institutional boundaries.In this scenario, the people constitute the sovereign.Yet Mény and Surel (2002) identify two further conceptions of the people, namely the people as a nation (cultural) and the people as a class (economic).Both are linked to specific forms of populism, the former to right-wing populism and the latter to left-wing populism (Kriesi, 2014, p. 362).Studies that compare these two types of populist parties find that their behavior, such as their parliamentary voting behavior (Otjes & Louwerse, 2015) for instance, differs as a consequence of the host ideology.The questions that arise then are: which of the two elements, host ideology or populism, determines the relationship between populism and liberal-democracy?And, does this relationship play out differently depending on the subdimensions of liberal democracy?In what follows, we discuss the differences between left-and right-wing populism in detail.We focus on two aspects of liberal democracy, minority rights and mutual constraints, to highlight possible commonalities and differences in the relationships between populist parties and democracy.

Populism Left, Right, and Center: Differences and Commonalities
Populist parties, independent of their host ideology, are united in their critique against the political establishment (Rooduijn & Akkerman, 2017).Yet, it is the host ideology that determines against whom the people should rally (Katsambekis, 2017, p. 205).Left-wing populist parties define the people on a class basis, referring mostly to the poor.In contrast, right-wing populist parties define the people on a cultural, nativist base (March, 2011;Mudde, 2004).In other words, whereas left-wing populist parties frame their criticisms economically and seek to protect the proletariat from exploitation by capitalists, rightwing populist parties champion nativism (Mudde, 2007, p. 19) and seek to protect "the nation from dangerous others" (Rooduijn & Akkerman, 2017, p. 196), stressing cultural issues above the rest.Thus, left-wing populist parties differ from right-wing populist parties in that they embrace an inclusive as opposed to an exclusive view of society (Katsambekis, 2017;Mudde & Rovira Kaltwasser, 2013). 5More importantly though, despite left-wing populist being inclusive on the society level, this does not necessarily imply that they are not anti-pluralistic on the political level.Essentially, the question of how democracy is organized is political and differences between leftwing and right-wing populism can be illustrated focusing on central aspects of democracy, namely political inclusion and political contestation (Dahl, 1971).While left-wing populist parties generally neither discredit minority groups nor object to granting these groups political rights, they do not accept political competition for that they, and only they, are the true representatives of the people.Consequently, they consider political control through effective opposition and institutional power check mechanisms as obstacles that prevent them from implementing the people's will.In this sense, left-wing populists are inclusive on the society level and the dimension of political participation.Yet, they are exclusive and essentially anti-pluralistic with regards to public contestation and the control of power (Mudde & Rovira Kaltwasser, 2013, p. 162).In contrast, right-wing populist parties are generally exclusive with regards to all of these aspects for that they object the extension of political participation rights to minorities in addition to claiming to be the only true representatives of what they consider to be the people.In short, left-and right-wing populism differ with regards to political inclusion but share similarities in their ideas of political contestation and control of power.It is for this reason that we choose two subdimensions of liberal democracy to highlight differences and commonalities between left-and right-wing populism, minority rights and mutual constraints.

Minority Rights
We define minority rights as descriptive representation of minorities in the political system, that is, the absence of systemic exclusion of groups considered minorities from exercising central power or other political rights (Merkel et al., 2016).
Here, we expect the populist element to interact with the host ideology.As written above, left-wing populist parties typically consider ethnic minorities as part of the people and hence, demand equal rights as part of their socially egalitarian tradition.In Europe as in Latin America, left-wing populist parties tend to be inclusive.In contrast, right-wing populist parties tend to be exclusive (Mudde & Rovira Kaltwasser, 2013), particularly in government where they have a lower incentive to reach out to minorities as part of a vote-maximizing campaign strategy.March (2011, p. 134), for example, shows that the left-wing Scottish Socialist Party (SSP) and the allied Respect coalition raised awareness about the Muslim population in the United Kingdom and initiated a pro-Muslim discourse.In contrast, the Slovak National Party (SNS) routinely targeted Hungarian minority parties and went as far as to propose a ban of all ethnic parties in Slovakia (Koev, 2015, p. 652).
H1.The presence of left-wing populist parties is positively associated with minority rights, whereas the presence of right-wing populist parties relates negatively to minority rights.

Mutual Constraints
Mutual constraints inhibit absolute power in a democracy by balancing the power of the executive vis-à-vis the judiciary and the legislature (Plattner, 2010).
In contrast to minority rights, where the host ideology is central, we argue that in the case of mutual constraints, the populist element determines the direction of the relationship.The strong focus of populist parties on the people can delegitimize the indirect aggregation of the volonté générale via the representative system.On the one hand, populist parties demand either stronger or more frequently employed measures of direct democracy to ease the implementation of the general will.On the other hand, if populist parties enter government, they see no need for a check on power, as they represent the peoples' will (Mudde & Rovira Kaltwasser, 2012).One such example is the attempt by the Polish Law and Justice Party (PiS), which assumed governmental power in 2015, to reform the Polish Supreme Court.Among the proposed changes, was the suggestion that judges would be appointed by the National Council of the Judiciary in which half the members would also be members of the parliament, effectively weakening the court's power to oversee political decisions (Walsh, 2017).And although the law was partially rejected by Poland's President Andrzej Duda at first, Poland's Premier Beata Szydlo (PiS) in reaction to the rejection said that, just like the parliament, the court should be under the control of the people (Waldoch, Krajweski, & Bartyzel, 2017).
This association does not depend upon the host ideology.Most populist parties regularly call for changes to the constitution to empower the executive.Parties on the left, such as Fico's Smer in Slovakia (Malová & Učeň, 2010), have voiced similar demands to those on the right, for example, Haider's Austrian Freedom Party (Austrian Freedom Party, 2011;Fallend, 2012) and the Czech Rally for the Republic/Republican Party of Czechoslovakia (Hanley, 2012).Each of them has demanded more power for the ruling executive to shift power away from parliaments and courts.Since the volonté générale is the rationale for decisions made and actions were taken by all populist parties here, we expect no differences among them regarding their influence on mutual constraints.
H2.The presence of populist parties is negatively associated with mutual constraints, regardless of the parties' host ideology.

Research Strategy
To test our theoretical arguments, we use a twofold strategy.First, we assess the differences between populist parties in government and opposition.Second, we distinguish between left-wing, center and right-wing populism to investigate the expected associations with minority rights and mutual constraints.
Empirically, we use a pooled cross-sectional design.Since we use the role within a political system as well as a party's host ideology, we opted for cabinets as the temporal unit of analysis.This approach allows us to determine whether a party is in government or opposition with great precision, while other approaches such as country-years are considerably more imprecise.Our data includes information from 30 European countries between 1990 and 2012. 6This time span captures major events for European populist parties, from their establishment through their rise.We exclude cabinets with duration of fewer than six months as we assume that any measurable impact is only evident after some time.

Dependent Variable
To measure democratic quality based on our concept of liberal democracy, we draw on three different sources.We measure aggregated liberal democratic quality by using the liberal democracy score (v2x_libdem) of the Varieties of Democracy Project (Coppedge et al., 2011). 7 To capture the two subdimensions, mutual constraints, and minority rights, we rely on the Democracy Barometer.For mutual constraints, we use the aggregated di-mension of horizontal checks (MC_CHECKS, Merkel et al., 2016, p. 29) which measures the balance between executive and legislative (ratio of parliamentary seats controlled by government to parliamentary seats controlled by opposition), the balance of checks between executive and legislative (ratio of control instruments of legislative over executive to control instruments of executive over legislative), and the power of judicial branch to review political decisions. 8This measure taps into the checksand-balances aspect.For minority rights, we use an indicator for the effective access to power for minorities (REP_DR3, Merkel et al., 2016, p. 53) that measures the descriptive representation of minorities and the extent to which they have access to central power.

Independent Variables
For testing our hypotheses, we use dummies to i) capture the presence of populist parties in opposition and government and ii) to capture either the presence of left-wing or right-wing populist parties.We further use a middle category for ambivalent cases that would not fit either of the right-or left-wing categories based on our coding procedure, which followed a three-step approach.First, we surveyed the existing secondary literature on populist parties in Europe to categorize parties in populist and non-populist parties. 9In the next step, we determined the role of populist parties in the political system, applying the following coding scheme: populist parties in government had to hold some position in the cabinet and populist parties in opposition had to hold at least one seat in the national parliament.Thus, we excluded parties identified as populist but without a seat in parliament during the relevant cabinet from our analyses.Table A1 in Appendix A lists all parties analyzed.
Finally, to determine whether a populist party belongs to one of our three categories (left, right, center), we code populist parties in relation to their countryspecific party system.The procedure is the following.First, we calculate a weighted party system's ideology mean for each cabinet.This average takes into consideration the ideological positions of all parliamentary parties in one particular country.We use the seat share of the respective parties to weight the mean.For each cabinet, we then classify populist parties as left, right, or center according to their relative distance to this weighted mean.We code every populist party within one weighted standard deviation of this reference as centrist, while parties further to the left or right are coded respectively (also see Huber & Ruth, 2017).10

Control Variables
In addition to our central variables, we include a selected set of covariates that, in theory, may relate to both the presence of populist parties as well as the levels of our democratic measures.These variables are the level of democratic consolidation (time in years since democratization), cabinet duration (in years), cabinet composition (surplus governments, minimal winning coalitions, and minority governments), economic development (GDP per Capita in 1,000 US Dollar), and a dichotomous variable to distinguish between post-communist countries and other countries.A detailed rationale for the inclusion of these variables can be found in Appendix C.

Empirical Model
To control for country-specific effects, we apply a linear mixed-effects model with cabinets nested under each country (Gelman & Hill, 2007). 11The respective countries serve as groups.This particular model also allows us to compare both intra-and cross-country variance.Given our interest in the change in democratic quality as a consequence of the presence of populist parties, modeling intra-country variance allows us to approximate this process.12

Empirical Results
In Figure 1, we plot the coefficients from the results. 13o start with the aggregate measure of liberal democracy (left panel in Figure 1), we observe that in Europe, there is no general association between populist parties and democracy, independent of their status (government or opposition).While we find the anticipated direction of correlation, that is negative for populists in government and positive for those actors in opposition, they are not statistically significant.However, a different picture emerges when distinguishing populist parties according to their host ideology.First, we see that rightwing populist parties are associated with lower levels of quality of liberal democracy.Second, both center and left-wing populist parties are associated with higher levels of quality of liberal democracy.Third, while the center and left-wing populist parties are statistically different from right-wing populist parties, there is no significant difference between those two categories.These findings provide a first idea about potential differences in the association between populist parties and liberal democracy as a consequence of varying host ideologies.For a detailed look, we now turn to the results from the two subdimensions, minority rights, and mutual constraints.
From the middle panel in Figure 1, we can first observe that populist parties in government are not associated with any particular direction regarding minority rights.In contrast, populist parties in opposition are associated with a positive development.Upon taking into consideration host ideology, however, we find that this does not apply to all populists equally.For minority rights, we expected a more negative effect of rightwing populist parties and a positive influence of left-wing populist parties, both in comparison to the absence of populist parties.The empirical results lend support to our argument.First, on average we observe a substantial positive relationship between left-wing populist parties and minority rights, whereas we find negative effects for right-wing populist parties.The presence of centrist populist parties is neither negatively nor positively correlated with minority rights.
For mutual constraints, we expected no differences between the different type of populist parties but on average, small negative associations for all of them.Figure 1 confirms these expectations.We observe no systematic effect of a populist parties' host ideology.However, populist parties in government and opposition are negatively associated with mutual constraints compared to instances where no populist parties are in government or opposition.This effect is in line with our expectation that populists undermine the separation of power. 14e conducted two types of robustness checks.First, we reran our analysis using a continuous variable that measures the logged seat share of populist parties instead of dummies (See Tables B9 and B10 in the Appendix). 15Second, we used alternative model specifications (lagged dependent variable models-See Tables B5, B6 and B7 in the Appendix). 16The most consistent finding across these additional checks is the positive association between left-wing populist parties and minority rights in comparison to right-wing populist parties, particularly in opposition. 17Other findings, such as populist parties' relationship with mutual constraints, are less consistent as they are met with greater uncertainty in our statistical models.Overall, these findings further support our idea that substantive differences in the relationship between populist parties can arise from host ideologies.

Discussion and Conclusion
Populist parties, because of their central ideas of antielitism, the belief in a general will, and their people centrism, challenge some of the commonly accepted rules of democracy, especially those of liberal democracy (Plattner, 2010).However, a series of studies have identified not only negative but also positive effects of populist parties on democratic quality (Canovan, 1999;Mudde & Rovira Kaltwasser, 2012).Largely absent from the debate surrounding populism and its relationship to democracy, however, has been the role of host ideologies.This article sought to initiate such discussion.We proposed that the host ideology, focusing on left-and right-wing populism, has consequences for how these parties relate to the dimension of political inclusion and minority rights in particular.However, we expected the host ideology to be irrelevant for a populist parties' association with mutual constraints.The empirical findings lend support in particular to the first of our propositions.The main takeaway from our article, therefore, is that populism should not be examined in isolation from its host ideology when considering the relationship between populist actors and democracy.This finding, of course, should not diminish the role populism plays in this relationship, particularly in the wake of temporary developments in cases such as Poland and Hungary.In some cases, however, populism may matter less or even only constitute as an additional qualifier of radical right parties rather than being a steady feature (cf.Rydgren, 2017).Future studies thus could explore under which conditions ideology and populism may play a greater role for populist parties and how they relate to specific aspects of democracy, an issue in which fundamental differences in historical legacies between East-and West-Europe may well play into (Gherghina & Soare, 2013).Furthermore, right-wing populist parties have been shown to mobilize certain voter groups which have been neglected by other political parties, such as citizens who are lower educated or poor (e.g., Huber & Ruth, 2017;Rooduijn, 2017).At the same time, they may also discourage certain voters from turning out in elections (Immerzeel & Pickup, 2015).Future research, by focusing on the dimension of political participation, may therefore also explore whether left-wing populist parties exert similar effects or, whether mobilization and de-mobilization effects depend on a populist party's host ideology.Note: Sources for these parties are Mudde (2007), Arter (2010), March (2011) andVan Kessel (2015).We should note that the Progress Party (FPd) in Demark and the Swedish party New Democracy (NyD) are borderline cases for that they may fit the category of neoliberal populist parties better.This is the case, as Mudde (2007, p. 48) writes, because "their xenophobic rhetoric is primarily informed by their liberalism".We kept these parties in our analyses nonetheless as a) they still fit the category of populist parties and b) would only be relevant for the analyses of minority rights in which case our results are more conservative given the inclusion of two cases that do not share the strong focus on nativism with other cases included here.The table below provides an overview of all abbreviations used for the various populist party dummies in the Appendix included in the models, their meaning (when coded as "1"), and the reference category (when coded as "0").B2 and B3).Note:  B4).Note: Figure B2 disentangles combinations of both government status and host ideology.B5 and B6).

Figure B4
. RBC-effect of populism on liberal democracy subdimensions by combinations of host ideology and government status using a lagged dependent variable (regression results: Table B7).

Appendix C. Description of Control Variables and Rationale for Inclusion
In our empirical model, we include a set of covariates that may relate to the presence of populist parties as well as to our democratic measures.This section in the Appendix explains the detailed rationale for the inclusion of each of the control variables.First, we include democratic consolidation as a control variable.We anticipate that both the existence of populism as well as the level of democracy might depend on how long a country is democratic.Furthermore, it has been argued elsewhere that the effect of populists is stronger when democratic institutions are less established (see Huber & Schimpf, 2016b;Mudde & Rovira Kaltwasser, 2012).To capture this logic, we use the time that has passed since democratization in each country was reached (in years) and use data from Polity IV (Marshall, Gurr, & Jaggers, 2017).Second, we control for cabinet duration.The shorter a cabinet lasts, the less likely it is that a cabinet can realize its agenda.Third, we include a measure for cabinet composition.We distinguish between surplus governments and other governments such as minority or minimal winning coalitions.Albertazzi and Mueller (2013) as well as Huber and Schimpf (2016a) highlight that this might moderate the effect of populist parties in government.At the same time, long time surplus governments such as in Austria have been argued to provide fertile grounds for populist's anti-elite rhetoric.The ParlGov dataset contains information on, both, the cabinet duration and composition (Doering & Manow, 2015).Fourth, we control for economic development.Some scholars have argued that populists are especially successful in garnering support from "losers of globalization" (Kitschelt, 1995, Kriesi et al., 2012).Thus, we anticipate that populist parties might be more successful in less developed countries.On the other hand, a long-lasting discussion has emerged on whether democracy and economic circumstances are connected or not (e.g.Boix & Stokes, 2003, Przeworski, Alvarez, Cheibub, & Limongi, 2000).To capture this, we measure economic development with Worldbank data on the gross domestic product per capita (GDPpc) in $1000 to ease interpretation.Finally, we include a dummy for post-communist countries as we anticipate that this distinction might play a role, both in the existence of different populist parties and levels of democratic quality (Gherghina & Soare, 2013, p. 7).

Figure 1 .
Figure 1.Effect of Populism on Liberal Democracy and two Subdimensions (Minority Rights and Mutual Constraints).Notes: Gov = Populist Party in Government (Reference: No Populist Party in Government); Opp = Populist Party in Opposition (Reference: No Populist Party in Opposition); R = Right-Wing Populist Party present (Reference: No Right-Wing Populist Party present); C = Centrist Populist Party present (Reference: No Centrist Populist Party present); L = Left-Wing Populist Party present (Reference: No Left-Wing Populist Party present); Coefficients are plotted with 90% Confidence Intervals

Figure
Figure Effect of populism on liberal democracy subdimensions (regression results: TablesB2 and B3).Note: FigureB1additionally includes two more measures of liberal democracy by the UDS (Pemstein, 2010) and Democracy Barometer(Merkel et al., 2016)

Table A3 .
Comparison of empty and null model using VDem.

Table B1 .
Abbreviations of populist party dummies, meaning and reference categories.

Table B2 .
Populism and liberal democracy by government status.

Table B3 .
Populism and liberal democracy by host ideology.

Table B4 .
Populism and liberal democracy by combinations of host ideology and government status.

Table B5 .
RBC-association of populist parties with democratic quality by government status (lagged dependent variable model).

Table B6 .
RBC-association of populist parties with democratic quality by host ideology (lagged dependent variable model).

Table B7 .
RBC-association of populist parties with democratic quality by combinations of host ideology and government status (lagged dependent variable model).

Table B8 .
RBC-populism and liberal democracy by government status using alternative coding (see notes below the table).this robustness check, we recoded two cases where populist parties supported minority governments.These two cases were the Danish Peoples Party from 2001 to 2011 and the Dutch Freedom Party (PVV) from 2010 to 2012.

Table B9 .
Populism and liberal democracy by government status using seat share.

Table B10 .
Populism and liberal democracy by host ideology using seat share.