Existing research has demonstrated that citizens' political trust increases when their preferred party is an electoral winner. Supporters of electoral winners perceive the political system as more responsive and believe that their interests are better represented. In contrast, supporters of parties that have lost an election show a decrease in political trust (e.g. Anderson and Guillory Reference Anderson and Guillory1997; Anderson and LoTempio Reference Anderson and LoTempio2002; Banducci and Karp Reference Banducci and Karp2003; Singh et al. Reference Singh, Lago and Blais2011). While we know that citizens that support mainstream parties have particularly high levels of political trust (Citrin and Stoker Reference Citrin and Stoker2018; Hooghe Reference Hooghe, Oscarsson and Holmberg2020; Miller and Listhaug Reference Miller and Listhaug1990), it is also well known that this does not apply to supporters of populist parties. Populist party supporters strongly distrust the main democratic institutions and political actors, which is not only due to their strong anti-elite sentiments but also because of their perception that their concerns and demands are not adequately represented by established parties' elites (Hooghe Reference Hooghe, Oscarsson and Holmberg2020; Mauk Reference Mauk2020; Mudde Reference Mudde2004; Rooduijn Reference Rooduijn2018; Söderlund and Kestilä-Kekkonen Reference Söderlund and Kestilä-Kekkonen2009).
However, what has been less frequently addressed so far is whether winning an election can also increase political trust among generally distrustful populist party supporters. Recent studies shed light on the impact of electoral gains of populist parties on non-populist voters' level of democratic satisfaction (Fahey et al. Reference Fahey, Allen and Alarian2022) and show that non-populist voters turn more dissatisfied when right-wing populist parties perform well in elections. Similarly, Hajdinjak (Reference Hajdinjak2022) demonstrates that populist party supporters in post-communist Central European countries are more trusting of their government when a populist party is part of the coalition, whereas non-populist voters' trust decreases. Government participation is of particular importance in PR systems, as winning an election usually subsequently leads to participation in a government coalition (Hooghe Reference Hooghe, Oscarsson and Holmberg2020). Hence, in this study, I do not focus on the impact of electoral outcomes on political trust but rather on the impact of government participation of populist parties on their voters' political trust.
Following previous research, I argue that while populist party supporters usually show low levels of political trust, they become more trustful when their preferred party joins the coalition. Populist party supporters are expected to react more strongly to electoral outcomes compared to non-populist party supporters, who show rather stable levels of political trust in general. This is mainly explained by populists' belief that the established elite fails to represent their interests. Once a populist party joins the government, populist party supporters are satisfied with the outcome of the election and coalition formation process, which results in higher levels of political trust (Mauk Reference Mauk2020) and trust in government among populist voters (Hajdinjak Reference Hajdinjak2022). In order to test this argument, this study draws on the case of Austria by using European Social Survey (ESS) data from 2002 to 2020.
Austria serves as a good case to test the impact of government participation on the political trust of populist party supporters for several reasons. It has a quite well-established populist party – the radical-right populist Austrian Freedom Party (FPÖ) – that has been both part of the opposition and part of governmental coalitions (Heinisch and Hauser Reference Heinisch, Hauser, Decker, Henningsen and Jakobsen2015). Moreover, the FPÖ was part of the government more than once as it participated in the governing coalition in the early 2000s and then later in 2017. Analysing a time span of 18 years allows the inclusion of two governmental participations, and therefore enables us to shed light on potential changes in political trust over a long time.
The empirical findings provide support for the theoretical argument, as they demonstrate that political trust varies among non-populist and populist voters. While populist voters in general show lower levels of political trust than other party supporters, they tend to trust more when their preferred party is part of the governing coalition. This increase in political trust is rather strong compared to non-populist party supporters, whose political trust is less affected by government participation. Moreover, the level of political trust among populist party supporters almost equals the political trust of non-populist voters when a populist party is in government.
These findings have important implications for the study of the support for political systems and democracy in general. First of all, they indicate that populist party supporters react more strongly to favourable outcomes and government participation than other party supporters. The findings corroborate the evidence from previous studies on the success of populist as well as authoritarian politicians (Cohen et al. Reference Cohen, Smith, Moseley and Layton2023; Hajdinjak Reference Hajdinjak2022; Mauk Reference Mauk2020) and show that these can also fuel political trust among their electorate when they are in government. However, while Hajdinjak (Reference Hajdinjak2022) focuses on the effect of trust in government, this study demonstrates that trust in politics more broadly is affected by government participation of populist parties.
Second, the findings support empirical evidence that populist party supporters in general hold more unstable attitudes towards democratic institutions. The volatile political trust among populist party supporters can have implications for democratic support in general. As they seem to react more strongly to electoral outcomes as well as coalition formation outcomes that are favourable to them, they potentially become more distrustful when their party experiences harsh electoral losses, or in the worst case, is not elected to parliament at all. This has important consequences for democracy, as citizens' trust in politics, as an important indicator for political support, should not be affected by electoral outcomes at all. These findings also speak to recent research that demonstrates that perceived electoral success only increase the winners' satisfaction with democracy in countries with lower democratic quality. Large gaps in satisfaction with democracy between winners and losers thus can imply that the views of winners and losers are not represented equally in a democracy (Merkley et al. Reference Merkley, Cutler, Quirk and Nyblade2019; Nadeau et al. Reference Nadeau, Daoust and Dassonneville2021). With regard to this study, this would also imply that in particular among populist voters, where the gap in political trust is large between times with and without government participation, political support in general is not stable. Moreover, such instability might also lead to increasing distrust not only in democracy or politics but also in elections and their outcomes (see also Nadeau et al. Reference Nadeau, Daoust and Dassonneville2021).
Political trust, populism and government participation
Trust in politics is considered an essential feature of solid and well-functioning democratic systems (Easton Reference Easton1975; Hetherington Reference Hetherington1998; Levi and Stoker Reference Levi and Stoker2000). As Hooghe (Reference Hooghe, Oscarsson and Holmberg2020: 360) points out, political trust ‘can be considered as a crucial reservoir of legitimacy and diffuse support for the political system’. The concept of political trust includes trust not only in the main democratic institutions, such as the parliament, but also in parties and politicians. In general, the level of political trust among citizens is highly dependent on individuals' political support, which can be identified as specific support or diffuse support.
Specific support, which is defined as support for the current government, parties and politicians, is greatly exposed to short-term influences. These short-term influences can be government outputs, the performance of certain politicians or also governmental changes (Easton Reference Easton1975; Hetherington Reference Hetherington2005). Diffuse support, in contrast, refers to citizens' long-term support for democratic institutions (Easton Reference Easton1975; Norris Reference Norris, Zmerli and van der Meer2017). Political support thus is, as Anderson et al. (Reference Anderson, Blais, Bowler, Donovan and Listhaug2005: 19) state, ‘the belief that the political system (or some part of it) will generally produce “good” outcomes’. While some scholars stress the importance of diffuse support for a solid democracy, others argue that even a loss in specific support can cause legitimacy problems for the current government and elected officials (Hetherington Reference Hetherington1998). Hooghe (Reference Hooghe, Oscarsson and Holmberg2020) further adds that decreasing levels of political trust in particular can lead to a destabilization of the political system in the long run. Thus, low levels of either specific or diffuse support can have far-reaching consequences for democracy (Hooghe and Zmerli Reference Hooghe, Zmerli, Zmerli and Hooghe2011).
Political trust and populism
Low levels of political trust have become particularly relevant with the rise of populist parties in many countries. As populist actors criticize the political establishment for its lack of political representation, they not only attract voters with low levels of trust but also reinforce decreasing trust among their electorate and the public (Mudde Reference Mudde2004; Rooduijn Reference Rooduijn2018). Populists' distrust in politics is related to the ‘thin ideology’ of populism. Populists consider society as ‘ultimately separated into two homogeneous and antagonistic groups, “the pure people” versus “the corrupt elite”’ (Mudde Reference Mudde2004: 543). Populism, as defined in the ideational approach, thus consists of two main dimensions: people-centrism and anti-elitism. While the populists' claim that the people have a general will that should be represented in politics is a crucial dimension of populism, anti-elitism is the dimension that is more closely related to political distrust (Rooduijn Reference Rooduijn2018). As Geurkink et al. (Reference Geurkink, Zaslove, Sluiter and Jacobs2020: 250) point out, ‘political trust taps into a core component of populism, namely, the opposition to the “corrupt elite”’, and therefore is strongly connected to anti-elitism.
The impact of populism, and also anti-elitism, on political trust has already been analysed in previous studies that find that voters of populist parties tend to distrust politics (Akkerman et al. Reference Akkerman, Zaslove and Spruyt2017; Belanger and Aarts Reference Belanger and Aarts2006; Swyngedouw Reference Swyngedouw2001). While high levels of distrust among the public have fuelled the rise of populist parties (Hooghe and Dassonneville Reference Hooghe and Dassonneville2018), the strong anti-elite rhetoric of populist actors – in which the established parties and politicians are accused of not being able to represent the interests of the people – further fuels distrust among populist voters (Rooduijn et al. Reference Rooduijn, Van Der Brug and De Lange2016). By claiming that the political elite is corrupt and also unable to represent the people in politics, these parties are able to mobilize voters who are dissatisfied with the political system. Criticizing the main democratic institutions and political actors thus leads to declining political trust, particularly among the mobilized voters. The political elite, with parties and politicians as the most important parts of it, is for many people the most visible actor in politics and thus is often equated with politics as a whole. Low levels of political trust among populist party supporters hence mainly derive from their strong distrust in the political elite (Curini Reference Curini2020; Geurkink et al. Reference Geurkink, Zaslove, Sluiter and Jacobs2020; Mudde Reference Mudde2004).
The impact of populist party government participation on political trust
While populism is an important predictor of citizens' levels of political trust, electoral outcomes also affect political trust. A large number of studies suggest that political trust strongly depends on electoral outcomes (Anderson and LoTempio Reference Anderson and LoTempio2002; Curini et al. Reference Curini, Jou and Memoli2012; Hooghe and Stiers Reference Hooghe and Stiers2016). As democratic systems and their institutions are legitimized through decisions made by voters in elections, it is inevitable that not all voters are satisfied with electoral outcomes. Elections and their outcomes therefore potentially increase or decrease the level of political trust, depending on whom the voters supported in the election (Anderson et al. Reference Anderson, Blais, Bowler, Donovan and Listhaug2005; Hooghe Reference Hooghe, Oscarsson and Holmberg2020).
This is also demonstrated by previous studies on the effect of winning an election on voters' political trust that show that trust increases when the preferred party performed well in the election and that it decreases when the preferred party is among the electoral losers. The same effect applies to satisfaction with democracy, as voters are more satisfied with the way democracy works when their preferred party increases its vote share or even wins an election (Anderson et al. Reference Anderson, Blais, Bowler, Donovan and Listhaug2005; Anderson and LoTempio Reference Anderson and LoTempio2002; Banducci and Karp Reference Banducci and Karp2003; Blais and Gélineau Reference Blais and Gélineau2007; Citrin and Stoker Reference Citrin and Stoker2018; Dahlberg and Linde Reference Dahlberg and Linde2017; Harteveld et al. Reference Harteveld, Kokkonen, Linde and Dahlberg2021). In addition to the impact of the electoral performance of the voters' preferred party, their perception of the electoral process is also important. Based on the argument that free and fair elections are essential for high-quality democracies, perceptions of the integrity of the electoral process shape citizens' satisfaction with democracy after elections. In general, it has been shown that in democracies of higher quality, the impact of electoral outcomes on satisfaction with democracy is hence lower than in countries with lower democratic quality. Additionally, winning or losing an election is less important when voters feel that their views are represented in any case. In these countries, the winner–loser gap is smaller (Merkley et al. Reference Merkley, Cutler, Quirk and Nyblade2019; Nadeau et al. Reference Nadeau, Daoust and Dassonneville2021). This is not only explained by losers feeling less dissatisfied but also because the increase in democratic satisfaction among winners is smaller, as they ‘are less likely to be exposed to information and cues that will prompt them to interpret the victory as the most decisive indication of the fairness of the electoral process’ (Nadeau et al. Reference Nadeau, Daoust and Dassonneville2021: 5). Depending on the voters' commitment to democracy, even the electoral success of authoritarian candidates does not necessarily lead to less support among the losers but, in contrast, can foster support for democratic values among these (Cohen et al. Reference Cohen, Smith, Moseley and Layton2023).
While electoral winners and losers are easy to identify in a two-party system with a majoritarian electoral system, in multiparty systems and under proportional representation it is not so simple. Stiers et al. (Reference Stiers, Daoust and Blais2018) show that for voters in such electoral systems, winning an election indeed means that their party ends the election as the number one party with the majority of seats. In this regard, Hooghe and Stiers (Reference Hooghe and Stiers2016) demonstrate that in PR systems participating in elections has a positive effect on all voters, as they have the chance to cast their ballot. Whether their preferred party was among the winners or losers of the election did not affect the political trust of the voters. Identifying electoral winners and losers through their seat gains or losses in parliament might seem to be the answer, but such changes do not necessarily mean that the party has gained or lost influence in the political system. Particularly in multiparty systems, it is more important whether a party becomes part of the governing coalition after an election (Daoust et al. Reference Daoust, Plescia and Blais2023; Hooghe Reference Hooghe, Oscarsson and Holmberg2020).
This study thus focuses on the impact of government participation, as a consequence of being an electoral winner in the last election, on the level of political trust among populist party supporters. I argue that voters of populist parties have higher levels of political trust when their preferred populist party is part of the government. Although political trust among populist party supporters is generally low due to their anti-elite sentiments, populist party supporters are expected to show higher levels of political support if they perceive the system as more responsive. This is the case if their preferred party is able to join the governmental coalition. Thus, the level of political trust is expected to increase among populist party supporters when their party is part of the government, as ‘winning votes and parliamentary seats may make the political system appear more responsive to citizens’ demands’ (Mauk Reference Mauk2020: 47). Put simply, trust in politics among populist party supporters increases if they perceive the system as responsive to the will of the people. Research, however, also shows that there is no effect when a populist party joins the political system as a new party that potentially could represent the demands of parts of the electorate, but there is an effect on political trust when populist parties gain seats in the national parliament (Ivarsflaten Reference Ivarsflaten2008; Mauk Reference Mauk2020) or join the government (Hajdinjak Reference Hajdinjak2022). Reflecting on these findings, populist party supporters are expected to have higher levels of political trust if the populist party is able to be part of the government compared to times when the party is in opposition.
However, one could also argue that, because of their strong anti-elite sentiments, political trust among populist voters should not be higher when their party is in government. Research suggests that the anti-elite sentiments of populist parties become less salient when the party is in power (Polk et al. Reference Polk2017). Moreover, populist parties in power also try to distance themselves from the ‘political elite’ by shifting their anti-elite discourse to other ‘corrupt and bad’ elites in the country (Jungkunz et al. Reference Jungkunz, Fahey and Hino2021; Schwörer Reference Schwörer2022). Populist parties in power also still claim to represent the will of the people, while still accusing other politicians of not being able to represent the people. Thus, although populist parties become part of the often criticized political elite, their voters might not perceive this as problematic and instead gain more trust in the political system. The hypothesis of this study is thus:
Hypothesis: The impact of government participation by the preferred party on political trust is larger among populist party voters than among non-populist party voters.
Case selection and data
The case of Austria
This study utilizes the case of Austria to test this assumption. Austria has been chosen as a case for this study for several reasons. First, Austria is one of the few countries in Western Europe with a long-established populist radical-right party that had its first huge electoral success as far back as the 1980s (Heinisch and Hauser Reference Heinisch, Hauser, Decker, Henningsen and Jakobsen2015). Therefore, the FPÖ is not a newly founded radical-right populist party but rather an established power in the Austrian political system, which is also reflected in its recurrent electoral success over time. This also has the advantage that the FPÖ has a core electorate that frequently votes for the party. Thus, changes in political trust over time among these voters are interesting, as these can show more pronounced patterns of political distrust. Newly founded populist parties that join the government after an election might lead to stronger reactions among their electorate as such electoral successes might not have been expected.
Second, over the last two decades, Austria has had several changes in the composition of governmental coalitions. The FPÖ was able to join the government as the second-strongest party after the elections of 1999, but the governing coalition with the Austrian People's Party (ÖVP) resigned due to problems between the governing parties. After re-elections and two more years of an ÖVP/FPÖ coalition in power, a further re-election took place in 2006. This election was followed by a governing coalition between the ÖVP and the Social Democrats (SPÖ) that lasted from 2006 until 2017. In the 2017 elections, the balance of power in parliament switched and the FPÖ became not only the third most successful party in the election but also formed a new coalition with the Liste Kurz (previous ÖVP) (Bodlos and Plescia Reference Bodlos and Plescia2018; Dolezal and Zeglovits Reference Dolezal and Zeglovits2014; Meffert and Gschwend Reference Meffert and Gschwend2010). While the FPÖ, Green Party and SPÖ, therefore, participated in government as well as in opposition during these 20 years, the ÖVP was the only party that was a constant in governmental coalitions.
Data and variables
This study draws on data from the ESS for Austria from 2002 to 2020. The data include nine rounds of the ESS, with only Round 6 (2012) missing in the data. In contrast to other election study data, the ESS does not survey respondents shortly before or after the general elections in the respective countries, but at any other time point during the electoral cycle. On the one hand, this means that the last election was longer ago, which in election studies can have the consequence that respondents are not able to recall their last electoral choice as accurately as if it was only a few weeks ago. On the other hand, the timing also has its advantages. As this study is interested in analysing the impact of government participation of the preferred party, it is necessary that the respective governments are in office. The timing of the ESS fieldwork of all included rounds ensures that the governments were already formed during these times.Footnote 1
While there is a large amount of research conducted on the impact of winning and losing in elections as well as government participation on satisfaction with democracy (e.g. Fahey et al. Reference Fahey, Allen and Alarian2022; Harteveld et al. Reference Harteveld, Kokkonen, Linde and Dahlberg2021; Merkley et al. Reference Merkley, Cutler, Quirk and Nyblade2019; Nadeau et al. Reference Nadeau, Daoust and Dassonneville2021), this study instead focuses on political trust as the dependent variable. Existing research suggests that satisfaction with democracy might not be an appropriate measure to capture basic support for democratic values. In contrast, it seems to be more likely to capture current satisfaction with the incumbent government and politicians (Hooghe and Stiers Reference Hooghe and Stiers2016) and not support for the political system more generally (Hooghe and Zmerli Reference Hooghe, Zmerli, Zmerli and Hooghe2011).
With regard to political trust, I follow existing research that aggregates the main components of political trust – trust in parliament, politicians and parties – into one measure for political trust (see also Marien Reference Marien, Zmerli and Hooghe2011).Footnote 2 The individual components are measured in the ESS by asking the respondents how much they trust the mentioned institution or actors on a 10-point scale ranging from 1 (‘no trust at all’) to 10 (‘complete trust’). The scales for trust in parliament and in politicians are summed up and then divided by 2, which leads to a political trust scale ranging from 1 (‘no trust at all’) to 10 (‘complete trust’). The main explanatory variable, vote choice, is measured through the self-reported vote choice in the last national election of the respondents. Further, I also include other independent variables in my models. These are several sociodemographic indicators, such as age, gender and education (in years) as well as political interest, ranging from 1 (‘not interested at all’) to 4 (‘very interested’).
For the analyses respondents who voted for the parties Bündnis Zukunft Österreich (BZÖ) or Das Neue Österreich und Liberale Forum (NEOS) or other small parties are excluded. Both the BZÖ and NEOS only competed in some of the elections. The BZÖ ran for office in the elections from 2002 until 2013, the NEOS only in the elections from 2013 onwards. Further, in the regression models, respondents who voted for the ÖVP and non-voters are excluded. The ÖVP is the only party that was always in the governmental coalition from 2002 until 2020. Hence, there can be no effect of being in the governmental coalition, as the party was always part of it. Likewise, non-voters are excluded as they are not able to be a winner or loser of elections as they do not vote. It is thus not possible for non-voters that their preferred party joins the government or opposition, which is important for the analyses. Therefore, only voters of the Green Party, the FPÖ and the SPÖ are included in the regression models.
Results
Political trust over time
Figure 1 shows the average level of political trust among party supporters of the ÖVP, the SPÖ, the FPÖ and the Green Party over time. The blue dots indicate that the respective party was not part of the governmental coalition during that time period, whereas the orange dots indicate government participation.
Average Levels of Political Trust among Party Supporters in Austria over Time
Note: The field period of the respective ESS round was as follows: Round 1 = 2002, Round 2 = 2004, Round 3 = 2006, Round 4 = 2008, Round 5 = 2010, Round 7 = 2014, Round 8 = 2016, Round 9 = 2018, Round 10 = 2020.
For the populist FPÖ voters, we see overall lower average levels of trust across all ESS rounds. However, trust among FPÖ voters increases and is the highest when their party is part of the government. This is not only the case for their first government participation in the early 2000s, but is best visible for their government participation starting in 2017 (ESS Round 9), where the average level of trust increases from around 3 points in Round 8 to around 4.5 points when they joined the government.
The average level of trust among the other party supporters, in contrast, varies. The average level of political trust among ÖVP voters, which is always part of the government, is rather stable and located at around 5 (on a 10-point political trust scale). There are only slight changes in trust visible, with higher political trust in Round 9 (2018) and lower political trust in Round 10 (2020) of the ESS. In contrast, political trust among Green Party voters increases, beginning in Round 8 (2016) of the ESS, but drops in Round 10 (2020). This is interesting, as the Green Party became part of the governmental coalition after the election of 2019. Political trust among SPÖ voters is also volatile when the party is part of the governing coalition but is almost the lowest among its supporters in ESS Round 10, when the party was in opposition.
Further, it is important to note that for almost all party supporters political trust decreases between Rounds 9 and 10 of the ESS. This decrease in trust is likely to be explained by a large corruption scandal that led to the resignation of Heinz-Christian Strache as party leader of the FPÖ, the breakdown of the FPÖ/ÖVP coalition after only 1.5 years in power, followed by snap elections in autumn 2019 (Eberl et al. Reference Eberl, Huber and Plescia2020).
The impact of government participation
Figure 1 shows that FPÖ voters in particular have on average higher levels of political trust when their party is part of the governmental coalition. To further analyse the impact of government participation on political trust I ran OLS regression models with fixed effects for ESS rounds. The dependent variable of both models is political trust. The main independent variable of interest is vote choice with SPÖ voters as the reference category. I further include fixed effects and clustered standard errors for ESS rounds in the models. The results of the regression analyses are displayed in Table 1.
OLS Regression Models with Fixed Effects and Clustered Standard Errors for ESS Rounds
Notes: ***p < 0.001; **p < 0.01; *p < 0.05; †p < 0.1.
Model 1 shows an effect on political trust for vote choice for the FPÖ as well as government participation. FPÖ voters have less political trust compared to Green Party and SPÖ voters, whereas government participation of the preferred party has a positive effect on political trust among voters. To test the hypothesis of this article and to analyse the impact of government participation by populist parties on the political trust of their voters, an interaction term is included in Model 2. The interaction effect for voting for the FPÖ and being part of the governing coalition is positive and statistically significant at p < 0.1. These results indicate some support for the hypothesis that increases in political trust are particularly strong for voters of populist parties. For Green Party voters, in contrast, no statistically significant effect is visible – that is, the increase in political trust for Green Party voters is not statistically different from the increase observed for SPÖ voters. It should be noted that the ‘main effect’ of government participation in Model 2 is quite small and not significantly different from zero. This estimate reflects the effect of government participation on political trust for SPÖ voters. In contrast, government participation has a positive and significant effect on Green Party and FPÖ voters.Footnote 3
This can also be seen in Figure 2, which displays the predicted levels of trust (Panel A) and average marginal effects (Panel B) conditional on vote choice and government participation. Like Figure 1, Panel A demonstrates that political trust among FPÖ voters is very low when their party is not part of the government. However, political trust among FPÖ voters is higher when the FPÖ is in the governmental coalition. FPÖ voters' level of political trust in these times then nearly equals the level of trust among non-populist voters. This is also demonstrated in Panel B of Figure 2, in which the effect of government participation is demonstrated. The level of political trust among FPÖ voters is about one point higher when the FPÖ is in government compared to when it is not part of the government. For the non-populist Green Party and SPÖ voters, we can see that their political trust is less affected by government participation. SPÖ voters in general have rather high levels of political trust, independent of SPÖ government participation. For Green Party voters we can see that these voters also have slight changes in their level of trust depending on the government participation of the Green Party. Compared to FPÖ voters, however, these changes in political trust are smaller and their overall level of trust is rather high.
The Impact of Government Participation on the Level of Political Trust among Party Supporters
Note: Figure 2 displays the average marginal effects from Model 2 in Table 1. Panel A shows the predicted level of political trust for party supporters. Panel B shows the change in political trust for times with government participation vs. no government participation for party supporters.
These findings support previous research on the political trust of populist party supporters and also the theoretical expectations of this study. They demonstrate that otherwise very distrustful populist voters become more trusting of politics when their preferred party is part of the governmental coalition. As expected, populist party supporters show distrust in politics when their preferred party is not part of the government. This distrust is theoretically explained by their anti-establishment attitudes, which lead to low political trust (Curini Reference Curini2020; Geurkink et al. Reference Geurkink, Zaslove, Sluiter and Jacobs2020; Hajdinjak Reference Hajdinjak2022). However, political trust increases when the populist party is part of the governmental coalition. Populists then most likely perceive the political system as more responsive to their demands and feel better represented by their own elite than by other politicians (Mauk Reference Mauk2020). The findings of this study hence demonstrate that populist party voters' political trust is not stable but depends on the performance of their preferred party and their impact on political decision-making as part of the government.
Robustness check
Linear models with clustered standard errors are usually applied to data with a large number of clusters. As clustering with small numbers of clusters can lead to biased p-values and thus less accurate parameters with regard to statistical significance and confidence intervals (Esarey and Menger Reference Esarey and Andrew Menger2019), I have run robustness tests using restricted maximum-likelihood multilevel estimation. Although multilevel models with small numbers of upper-level cases can be biased (Stegmueller Reference Stegmueller2013), I follow suggestions by Elff et al. (Reference Elff, Heisig, Schaeffer and Shikano2021), who state that biased parameters are not the case for models with low numbers of upper-level units when using restricted maximum-likelihood estimation. Hence, using this estimation is an appropriate way to control for nested observations in a smaller number of upper-level units. The results of these analyses are reported in the Supplementary Material (Table A2). They demonstrate that the interaction effect of voting for the FPÖ and government participation is statistically significant at p < 0.01, and thus shows stronger statistical significance than the model reported in the results section of this article.
Conclusion
The aim of this study was to analyse the question of whether political trust among populist party supporters increases when their preferred party is part of the government. While previous research demonstrated that party supporters in general show higher levels of trust, it has also highlighted that the level of political trust differs between party families (Hooghe Reference Hooghe, Oscarsson and Holmberg2020; Söderlund and Kestilä-Kekkonen Reference Söderlund and Kestilä-Kekkonen2009). In this regard, populist party supporters especially are less likely to trust in politics, mainly due to their strong anti-elite sentiments and their critique of political representation. Thus, populist voters have lower levels of political trust compared to non-populist party supporters, who show generally higher levels of political trust. In addition, and most importantly for this study, winning an election, and more specifically being part of the governmental coalition, can increase political trust among voters (Hooghe Reference Hooghe, Oscarsson and Holmberg2020). This study argues that this increase in political trust can also be observed for populist party supporters, as they perceive the system as more responsive and they feel better represented by populist politicians in the government (Hajdinjak Reference Hajdinjak2022; Mauk Reference Mauk2020).
The empirical findings provide strong support for this expectation. In the case of Austria, populist voters show the lowest level of political trust across all party supporters. However, political trust is higher among populist voters when the preferred populist party is part of the government. In contrast, non-populist voters are more trustful of politics in general and their level of trust only varies slightly when the composition of the government changes. Although populist supporters hold strong anti-elite sentiments, these might be less prevalent when their preferred party is part of the political elite as part of the governmental coalition.
The results of this study contribute to the growing literature on the impact of populism on citizens' political trust. They highlight that being part of the governmental coalition can increase political trust among otherwise distrustful populist party supporters. Thus, these findings have important implications for democratic support and democracy more generally. The findings imply that political trust, which also captures trust in both democratic institutions and political authorities, is not stable among populist party supporters but rather volatile. Depending on the favourability of the outcome of an election, their level of trust is higher or lower. This is particularly dangerous for democratic systems, as their stability depends on a certain level of political support among their citizens. The effect of election outcomes on the political trust of populist voters is concerning, as their support of the political system relies on winning or losing an election. Large winner–loser gaps after elections can indicate decreasing democratic stability, as distrust in politics based on electoral outcomes can lead to doubts about the electoral process as well. While in well-functioning, high-quality democracies losers of elections also know that the governing parties will do their best for all citizens and that there is less reason to be dissatisfied with the election outcome, this is less the case for voters in lower-quality democracies (Dahlberg and Linde Reference Dahlberg and Linde2016; Nadeau et al. Reference Nadeau, Daoust and Dassonneville2021). With regard to this study – where a large gap in trust by populist voters depending on whether the FPÖ was in government or opposition is demonstrated – it might be assumed that among these voters particularly, less trust might be observed not only in politics but potentially also in democratic processes.
Of course, there are also limitations of this study that have to be mentioned. This study analyses political trust in the case of Austria and, thus, focuses on a case where a rather strong populist party is part of the political system. Moreover, this study focuses on Austria solely. While this has the advantage that we gain more in-depth knowledge on this particular case, the impact of populist government participation on political trust might look different in other contexts. Future research thus should also focus on other countries to enhance our understanding of political trust in combination with populist party government participation. Moreover, future studies should elaborate on what drives populist party supporters' political trust more broadly. As anti-elitism and people-centrism are the main dimensions of populism in the ideational approach, it would be interesting to disentangle the effect of both dimensions on political trust not just by using vote choice as an indicator for populism but also, for example, by utilizing the populist attitudes of citizens. This could help us to understand which components of populism drive political trust among citizens. This seems to be particularly interesting as populist parties become part of the otherwise criticized political elite when they join a governmental coalition. So far, however, it is not entirely clear how populist voters with strong anti-elite sentiments perceived their party when it is part of the political elite, and whether this affects their support for the party.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/gov.2023.32.
Acknowledgements
I would like to thank the reviewers and the editors of Government and Opposition for their constructive and helpful feedback on previous versions of this manuscript. I am thankful to the participants of the General Conference of the Austrian Political Science Association 2020 for their valuable feedback. I am grateful to Robert A. Huber, Michael Jankowski and Christian Stecker for their helpful comments.
