Quick and dirty: how populist parties in government affect greenhouse gas emissions in EU member states

ABSTRACT

Populist parties are on the rise. But what happens when they are in government? In order to grasp the effect of populist parties in government systematically, the paper includes all 28 EU member states in an analysis which estimates the effect of populist parties in power on the increase of greenhouse gas (GHG) emissions. The results show that depending on their ideological orientation populist government participation is clearly associated with increasing GHG emissions. Furthermore, the analysis shows that the effect is quick. This is above all true for right-wing populist parties in government. Furthermore, the analysis shows that populist parties have different impacts in various regions of the EU. While right-wing populist governments are associated with an increase of GHG emissions in North Western and Eastern Europe, left-wing populist governments in Southern Europe have the opposite effect.

KEYWORDS:

• Climate change
• European Union
• partisan theory
• populist parties
• post-factual communication
• socio-political divisions

Introduction

As the second decade of the new millennium ends, the European Union (EU) is challenged on several fronts. One such challenge is the increasing importance of populist parties, which contest the authority of the EU in various respects (Albertazzi & Mueller, 2013). Another challenge for the EU is to find a coherent way of mitigating climate change (Dupont, 2019; Jordan et al., 2010). This paper examines the interplay between populism and the achievements towards reducing greenhouse gas (GHG) emissions in EU member states.

In contrast to most established parties, which have all lost votes over the last two decades, populist parties have been able to increase their vote share considerably and, with that, their representation in parliament (Timbro, 2019). In the EU member states, the share of parliamentary seats for populist parties has increased substantially over the last decades (Biard et al., 2019; Schaller & Carius, 2019; Timbro, 2019). This in turn has consequences for government building: populist parties participated in governments in more than half of the EU member states in the period from 1990 to 2018, and in 2018 more than one-third of the governments in EU member states had populist parties in government.

Studies looking at the effect of populist parties show that their impact increased over the years. Findings published just ten years ago found that the impact of populism on policy or the party system is rather modest (Minkenberg, 2013; Mudde, 2014). However, this has changed fundamentally. Newer investigations attest that populist parties have a significant impact on migration policy (Akkerman, 2018), on socio-economic policies (Röth et al., 2018), as well as on party systems (Wolinetz & Zaslove, 2017) and even the quality of democracy (Huber & Schimpf, 2016). However, even if our knowledge about the impact of populist parties has improved over the last decade, the field is still underexplored (Akkerman, 2018; Biard et al., 2019). This is especially true when it comes to studying the impact of populist parties in power on policy outcomes. The few studies which exist conclude that the impact seems to be weak. Minkenberg (2001) speaks of a ‘taming effect' when populist parties participate in coalition governments (see also Akkerman, 2018; Huber et al., 2021; Van Kessel, 2015). Instead, the major impact of populist parties seems to be indirect in that they create a political climate which supports their issues of xenophobia, denial of climate change and national anti-European values.

Studies looking at the connection between populism and climate change focus on attitudes and world views of populist parties and even here the causal link is complex and underexplored (Brown, 2014; Forchtner & Kølvraa, 2015; Lockwood, 2018; McCright et al., 2015). Studies covering a large number of countries show that populist parties have nuanced standpoints depending on ideological positions and socio-political contexts (Forchtner, 2019; Lockwood, 2018; Schaller & Carius, 2019).

For the analysis, I take the findings of these studies into account and consider the effect of populist parties in government with different ideological leanings, operating in various regions. To do this, I analyze all 28 EU member states from 1990 to 2018 in a macro-comparative study. Using the full sample of EU member states is not a trivial aspect as most EU evidence relies only on a handful of member states, which frequently implies a selection bias.

The analysis in this paper shows that populist parties in government have a substantial and significant impact on GHG emissions and that this impact is dependent on the ideological leaning of the populist parties and its socio-geographical context. Furthermore, the analysis shows that the effect of populist parties is faster than that of established party families, which indicates that previous studies’ conclusion of a taming and fizzle out effect might underestimate their impact.

Populist government participation and climate change

Populist parties differ from other parties substantially as well as strategically and therefore it is reasonable to categorize them as a party family type of their own. Populist parties use an expressive rhetoric, and they challenge established knowledge by engaging in post-factual communication and even refer to conspiracy theory (van Prooijen et al., 2015; see also Huber et al., 2021). The purpose of this strategy or style of communication is to make their post-factual interpretation more widely accepted in order to come into or remain in power. The strategy of defining reality has already forcefully been analyzed by Mannheim (1952, p. 198) in his lecture on ‘Competition as a Cultural Phenomenon' at the Conference of German Sociology in 1928: ‘ … every historical, ideological, sociological piece of knowledge (even should it prove to be Absolute Truth itself), is clearly rooted in and carried by the desire for power and recognition of particular social groups who want to make their interpretation of the world the universal one.' More recently, Ulrich Beck (1999) uses the concept of the ‘relations of definition' (referring to Marx’s ‘relations of production') in order to describe the cultural struggle in risk society in which hazards and insecurity are induced by the processes of modernization. Environmental risks and in particular climate change as a risk manufactured by human activity is a typical example of this struggle. The consequence of post-factual communication is that it creates and reinforces societal divisions within and between countries and regions in the EU.

In addition to this strategic and rhetorically often aggressive style, populist parties possess an ideological core (Mudde & Kaltwasser, 2013). The basic distinction between populism and other political ideologies is that populist parties question the established elite and that they claim to speak on behalf of ‘the people' (Mudde, 2007). However, the ideology of populist parties is not a holistic view on society. Some call it a ‘thin’ or a ‘thin-centered ideology’ (Freeden, 2006; Stanley, 2008). That means that ‘populism always severs itself from wider ideational contexts' (Mudde & Kaltwasser, 2013, p. 498). Often, populist parties associate with right or left radical ideologies but there are also cases in which they align with more middle-of-the-road ideologies such as conservatism or they favor free market principles. This aspect makes hypothesis building complex. While the populist core ideology would suggest that populist parties are skeptical towards the assumption of anthropocentric climate change, right, left and more center leaning populist parties may modify or even contradict this position. In addition, even within the same ideological group, the socio-political context in which populist parties operate may lead to different positions on policies and political issues (Mudde & Kaltwasser, 2013; Roberts, 2006). This is because post-factual communication needs to resonate within national and regional contexts.

Putting the additional aspects of populist ideology aside for a moment, the first hypothesis is that populism in itself is bad for the environment (Huber, 2020; Stronberg, 2018) and that populist parties in government have a negative impact on GHG emissions. From the core ideology of populism, the major reason for this is that populist parties doubt established processes and responses because they believe they are corrupt and inefficient:

Hypothesis 1: When populist parties come into power, greenhouse gas emissions significantly increase.

Most previous studies refer to far right-wing populist parties and this literature supports this hypothesis (Forchtner, 2019; Lockwood, 2018). However, there seems to be some ambiguity between environmental positions and populism (Hansen, 2020; Machin & Wagener, 2019). In general, even far right-wing populist parties support environmental issues in the form of nature conservation. In this perspective, they take a romanticized nationalist position on nature conservation similar to the first wave of (nationalist) nature conservatism movements at the turn of the twentieth century (Dalton, 1994). However, when it comes to more global, science-based or politicized issues such as climate change, populist parties take a more hostile position or deny the very fact of human-caused climate change. Right-wing populist parties often frame this position in the context of conspiracy theory (Brown, 2014; Gemenis et al., 2012; Hobson & Niemeyer, 2011). An empirical study of European right-wing populist parties shows that most of them are skeptical towards climate change policy (Gemenis et al., 2012; Schaller & Carius, 2019).

Concerning the effect of right-wing populist government, studies agree that there is an effect: ‘In terms of effect in the policy-making arena, the radical right in power clearly makes a difference in all cases' (Minkenberg, 2013, p. 19). Most of the few studies looking at the direct government effect of populist parties testify that the impact is modest. Others reach different conclusions, however, Pappas (2019; see also IDEA, 2020) discovers in his study fundamental effects of populist governments. Once in power, populists seek to establish an order that is anti-liberal. They establish a strategic pursuit of political polarization with a systematic use of patronage to reward supporters and crowd out the opposition. Furthermore, they develop a drive to seize control of the state, emasculate liberal institutions, and finally aim to impose an illiberal constitution. Concerning the influence of populist parties in governments on GHG emissions the impact would be fundamental. This leads us to the second hypothesis:

Hypothesis 2: When far right populist parties come into power, greenhouse gas emissions increase significantly.

Some argue that not all forms of populism are necessarily hostile to the environment and that some populist parties accept the thesis of anthropogenic climate change. In this context (Beeson, 2019) even framed the term of environmental populism. Some claim that left populist parties are more climate friendly – even more so than many non-populist parties (see Buzogany & Cetkovic, 2021; Huber et al., 2021). However, left populist parties have been much less analyzed than right-wing populist parties (but see Damiani, 2020; March, 2012). Even if these parties are also skeptical of the political elite, they may be less hostile toward supporting initiatives against climate change or may even support such efforts. Considering this, the general hypothesis needs to be modified for left-wing populist parties:

Hypothesis 3: When far left populist parties come into power, greenhouse gas emissions do not increase.

Populist parties which are neither far right nor far left may have various positions towards climate change. These centrist populist parties are mainly conservative or market liberal, which suggests that they may have the effect of increasing GHG emissions but much less so than far right populist parties:

Hypothesis 4: When centerist populist parties come into power, greenhouse gas emissions increase but not to the same extent as for far right populist parties in government.

The final issue I address in the context of conditional impacts of populist parties in government concerns regional differences. As mentioned above, populist parties act differently in different socio-political contexts. Mudde and Kaltwasser (2013), for example, show that populist parties in the North of Europe demonstrate less ‘macho behavior’ than in the South of Europe. The same is true concerning the attitude of populist parties towards climate change. Neglecting or opposing the idea of human-made climate change is popular in the populist parties in Austria, France, and the Netherlands, but the populist parties in Sweden and Finland are more open to combatting climate change (Schaller & Carius, 2019). These regional differences water down the direct impact of the above-mentioned hypotheses. Without going into an ideographical analysis, it is difficult to postulate precise hypotheses of the extent to which the socio-political context has an impact. However, as a proxy, I compare the impact of populist governments in three regions of the EU which often show different results in environmental studies: the North Western, Southern and Eastern European regions.¹

The North Western region is the region where the fundamental social and cultural changes have occurred most strongly. There has been a value change leading to a new politics cleavage (Inglehart, 1990). Environmental issues have high priority and countries from this region are often considered as environmental pioneers. However, in these countries, there has also been a strong counter trend which has been labeled as a ‘silent counter-revolution' which has been accompanied by the emergence of extreme right-wing populist parties (Ignazi, 1992). That would mean that right-wing populist parties are important in this region. Southern Europe has been said to have a distinctive Southern aspect of environmental politics (Börzel, 2000; Pridham, 1996). Southern European EU member states are considered as laggards in their environmental performance. In many Southern European EU member states, there have been governments with populist parties. However, populist parties have often a left populist appeal and therefore may have a weaker impact on GHG emissions (AKEL in Cyprus, SYRIZA in Greece; PRC and PdCI in Italy). Finally, the Eastern European member states are also lagging behind in their environmental standards. In particular on climate change, some speculate about a new East–West divide in the EU (Lessenki, 2019). Furthermore, many of these states have a long-lasting tradition of right-wing populist governments (Minkenberg, 2017), which makes it very likely that those governments have a substantial impact on a poor climate change performance.

Analyzing the impact of populist parties in government on GHG emissions in a macro-comparative perspective requires a few words about the causal mechanism. The causal chain between a government composition and outcome is long and complex and it is difficult to determine in a quantitative study like this. The ‘naïve' assumption of the impact of parties on policy outcomes suggests that once parties are in power, they create and modify laws, regulations and policies, which in turn have an impact on policy outcomes. However, it is well acknowledged that the link between policies and outcomes is far from perfect (Pressman & Wildavsky, 1973; Pülzl & Treib, 2007). In the context of environmental policies, Knill et al. (2012, p. 442) – among many other studies – reach the conclusion ‘that regulatory outputs do not automatically lead to desired impacts' (Biesenbender & Tosun, 2014). Another important aspect is that outcomes do not only change through policies. Parties in government often have the possibility to change important officials of key institutions or judges that decide on the implementation of regulations. An extreme example of this is the Trump presidency, but similar trends can be seen in other countries as well.

An important but difficult to measure aspect of populist parties’ impact and success is post-factual communication. With that, populist parties create an atmosphere which affects the public discourse. Populist parties use this to erode knowledge based on scientific findings and question established procedures. Since climate change is not perceived directly, but rather its danger is captured through relatively abstract scientific knowledge, it is susceptible to post-factual communication that denies the effects of global warming. Post-factual communication may seem to influence political discourse in non-government times, but it is much more effective once a populist party is in power. In this situation, more resources are available to disseminate post-factual ideas (Escartin, 2020; Pappas, 2019). Thus, the effect of changes may be more quickly visible when populist parties come into government than for other parties with more expertise-based messages.

Research design and method

In order to analyze the link between populist government participation and GHG emissions, I focus on the event when a populist government is first installed or when it cedes (resigns or gets voted out). For populist government participation, I use the data of the Timbro Authoritarian Populism Index 2019 (Timbro, 2019), the PopuList data (Rooduijn et al., 2019; Zulianello, 2020) and my own research on particular parties (see supplementary material). For populist and right-wing populist parties, Mudde's (2007) definition is used. Right-wing populist parties are nativist and authoritarian parties and left-wing populist parties reject the socio-economic structure of capitalism (March, 2012). Populist parties with no far right- or far left orientation are called here centrist populist parties (Rooduijn et al., 2019). A documentation of the populist governments and their ideological leaning is found in Table A1 in the supplementary material.

Because I am interested in the effect when governments with populist parties come to power (or resign or get voted out), I apply a first difference model which specifies a change to a populist government with 1, a resignation of a populist government with −1 and is otherwise 0. All other variables have also been included as first differences. First difference models have several advantages. First of all, they fit the theoretical assumption that when populist parties come to power, outcomes change. From a methodological point of view, first difference models overcome the omitted variables problem with panel data more efficiently than fixed effects models (Wooldridge, 2000, pp. 447–448).

The dependent variable is the annual change of GHG emissions. The data for GHG emissions are taken from Eurostat and refer to total GHG emissions measured in tons per capita, excluding land use, land use change and forestry (LULUCF).²

In order to control if there is a partisan effect of populist parties at all I compare the impact of populist party participation to governments with other party families in government. For that, I use dummy variables based on the classification and data of center and left governments according to Armingeon et al. (2020). I exclude right governments because they often include many populist coalition governments. Center party governments are defined by the participation of Christian Democratic and Catholic parties and similar parties such as agrarian parties. A government is considered left if it is made up of social democratic parties and parties left of social democrats. Parties left of social democrats are socialists, communists and green parties. However, it is not clear what party families’ policy goals are concerning climate change policy (Jahn, 2021). While Green parties may follow a green agenda and strive to reduce GHG emissions, this is not so clear for other left-wing parties. It can be assumed that left-wing parties emphasize employment and material welfare for the working class over environmental concerns. However, in contrast one can also follow the argument that left-wing governments are more prone to intervene in the market and that they may also do so for the sake of improving environmental quality. Aside from these two positions, King and Borchardt (1994) assume that the left-right dimension has no substantial impact on environmental issues (Knill et al., 2010; but see Jahn, 2018). Center governments often pay lip service to protect the environment but when it comes to policies, they are heavily criticized for favoring business interests. Chancellor Merkel’s ambiguous climate involvement may be a case in point (Thalman & Wettenenegel, 2019).³

In order to estimate the impact of populist parties in government on GHG emissions, I include some control variables which may also be influential factors. These are variables concerning the energy mix of a country, on the one hand, and the economic situation, on the other. The energy mix may have a very substantial impact on GHG emissions. A high share of fossil fuels (coal, oil, gas) may increase GHG emissions profoundly, while nuclear energy and solar and wind power may have the opposite effect (Le Quéré et al., 2019). Economic aspects may also be highly relevant in explaining GHG emissions. In particular economic growth may be important. In prosperous times, countries may be more willing to combat pollution and lower GHG emissions. That growth has no harmful impact on the environment is a basic assumption of concepts of ecological modernization which builds on the environmental Kuznets curve (Jänicke, 2007; Mol et al., 2009). However, it is also reasonable to claim that growth promotes pollution and that economic growth is associated with increasing GHG emissions (Jahn, 2018, pp. 159–163). The descriptive statistics and the sources for the variables are documented in Table A2 in the supplementary material.

I employ a panel corrected standard error time-series – cross-section regression analysis (Beck & Katz, 1995). This is a standard procedure for this kind of analysis. From 1990, when the dependent variable is first available, I include all countries for the years when they are EU member states. To estimate the pace at which parties in government change the outcome, I use optimized lags. This procedure acknowledges that the political process involves several stages which differ in various political systems. For example, some countries have more veto players or veto points (Jahn, 2010) so that there are varying time lags between an initiative of a government and its effect on policies and outcomes. As mentioned above populist parties may have a rapid effect when supported by the government’s authority. But even from a methodological point of view, some suggest that standard lags are not suitable for empirical analysis, and claim: ‘Though the optimization of lags is certainly time consuming, it is absolutely essential in first difference models' (Plümper et al., 2005, p. 344).

I optimize all government variables at the same time. For that, I keep all but one country with no time lags. For each country, all possible combinations of analytically set ranges of time lags are tested. For the empirical analysis, I test time lags between zero and four years, which normally is a legislative term. As an effectiveness criterion, R-square is used as suggested by Plümper et al. (2005). In the next step, the same procedure is applied for all other countries.

When testing for the regional impact, I compare the three regions in one fully interactive model (Kam & Franzese, 2007, pp. 103–111). Such a model, in which all independent variables interact with the moderator, has the advantage that the estimated coefficients are comparable and that we maintain the number of observations (Kam & Franzese, 2007, p. 109). As the moderator, I use dummy variables for the three EU regions.

In a final model, I drop the countries from the analysis which never had a government with populist parties. In fact, such a sample may show the within-country effects of populist parties more clearly than a sample also including countries without a change to populist governments, which in turn focuses more strongly on between-country difference. If the hypotheses are confirmed in this set of countries, it would provide strong evidence that populist parties actually have an effect on changes in GHG emissions when in government. All models include period effect in order to control for unspecified effects of time.

Findings

The models in Table 1 show the effects of governments on all, right, left, and center populist parties in government. However, caution should be exercised as the number of observations for left and center populist governments is small. In the sample of countries/years there are around 21 per cent of governments with populist party participation, 14 per cent of which are right-wing, but only 4.3 and 3.7 per cent with left-wing or centrist populist party participation, respectively. In contrast, left-wing governments participate to 66 per cent in the period of investigation and center parties to 48 per cent.

Table 1. Partisan effects on greenhouse gas emissions in the EU member states from 1990 to 2018.

	No optimized lags				Optimized lags
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	All	Right	Left	Center	All	Right	Left	Center
ΔPopulist	0.0694*	0.0808*	−0.0172	0.0187	0.165***	0.160**	−0.0595	0.0755*
Governments	(1.96)	(2.18)	(−0.37)	(0.55)	(4.85)	(3.20)	(−1.46)	(2.17)
ΔLeft	−0.0400	−0.0382	−0.0364	−0.0367	−0.196***	−0.193***	−0.194***	−0.197***
Governments	(−1.13)	(−1.08)	(−1.03)	(−1.05)	(−5.79)	(−5.74)	(−5.80)	(−5.92)
ΔCenter	0.0242	0.0230	0.0280	0.0282	0.0856*	0.0856*	0.0831^+	0.0823^+
Governments	(0.55)	(0.52)	(0.63)	(0.63)	(1.97)	(1.97)	(1.89)	(1.86)
ΔFossil	3.258***	3.261***	3.258***	3.258***	3.306***	3.305***	3.304***	3.305***
Per capita	(34.55)	(34.43)	(34.58)	(34.58)	(34.90)	(34.65)	(34.70)	(34.74)
ΔSun and wind	0.199	0.238	0.191	0.191	0.292	0.296	0.148	0.189
Energy per capita	(0.40)	(0.47)	(0.38)	(0.38)	(0.59)	(0.60)	(0.30)	(0.38)
ΔNuclear energy	−0.292**	−0.292**	−0.303**	−0.302**	−0.358**	−0.369**	−0.392***	−0.394**
Per capita	(−2.81)	(−2.80)	(−2.93)	(−2.91)	(−2.87)	(−2.95)	(−3.30)	(−3.29)
Real economic	0.00665	0.00657	0.00645	0.00648	0.00480	0.00513	0.00509	0.00493
Growth	(1.33)	(1.31)	(1.27)	(1.28)	(1.02)	(1.09)	(1.00)	(0.96)
_cons	−0.0462	−0.0462	−0.0453	−0.0454	−0.0427	−0.0435	−0.0430	−0.0427
	(−1.11)	(−1.07)	(−1.03)	(−1.05)	(−1.15)	(−1.22)	(−0.96)	(−0.98)
N	588	588	588	588	588	588	588	588
r2	0.829	0.829	0.828	0.828	0.846	0.846	0.842	0.842
adjusted_r2	0.819	0.819	0.818	0.818	0.837	0.836	0.833	0.833

Δ = first differences; t statistics in parentheses; ⁺p < 0.10, *p < 0.05, **p < 0.01, ***p < 0.001

In the first step, I compare the impact of party governments on GHG emissions with and without optimized time lags. It shows that the partisan effect is rather modest when we do not use optimized time lags. Actually, populist parties have the strongest and most significant effects in the model. This is in particular true when analyzing all populist governments or when looking at far right populist governments. For all other party families, there is only a weak and insignificant relationship.

In the models with optimized lags the results change quite substantially; not so much for populist governments but rather for left and center party governments. In all models left party government participation has a negative and significant coefficient which means that the impact of these governments is to reduce GHG emissions. The opposite is true for center party governments. The effect of lag optimization is less strong for populist governments, although also here the coefficient is twice as large as in the no optimized lag models and the significance level increases. For centrist populist parties, there is also a significant result but the impact is only half as strong as for right-wing populist governments. For left populist governments, the impact remains insignificant. From the results, we get a slight hint that left populist parties in government have a different impact than other populist parties in government because the sign is negative.

A closer look at the optimized time lags shows that a change in GHG emissions occurs quickly after a populist party comes to power. This effect occurs in less than a year (on average about nine months). By contrast, it is roughly two years for left and center party governments. We achieve very similar results for right-wing populist governments, where the delay averages eight and a half months.

All of these results indicate that populist parties and above all right-wing populist parties in government have a significant and rapid impact on GHG emissions. Left and centrist populist parties do not have this effect. The significant impact of (right-wing) populist parties is astonishing because the included control variables have, by and large, a strong impact as well. Above all, the use of fossil fuel is highly significant and substantial. This variable alone explains around 80 per cent of the variance. But also the use of nuclear energy reduces GHG emissions quite strongly. In contrast, economic growth has no significant impact on any of the models. The results so far confirm the postulated hypotheses.

Looking at regional differences in the impact of populist government participation across EU member states shows that the overall impact is statically significant and indicates that populist governments increase GHG emissions. The magnitude of the impact, however, is twice as high in the countries in the Northwest of the EU than in Southern Europe, leaving Eastern Europe in the middle. In Eastern Europe and the Northwest, right-wing populist parties have a strong impact on increased GHG emissions, although in the latter case the level of significance is only on the p < 0.1 level. Southern Europe seems to be different. Here, left populist governments go along with a reduction of GHG emissions. This broadly confirms the assumptions made above about regional differences.

Timing over the regions also reveals interesting patterns. Populist parties in government have the fastest impact on GHG emissions in Eastern Europe. This effect materializes in less than half a year. It is slowest in Northwest Europe where it takes over one year. The regional analysis also shows that center party governments have only a negative impact in the Northwest. In the other regions, I cannot identify any significant effect of center party governments. In contrast to this, left party governments have a significant effect of reducing GHG emissions in all regions.

The significant impact of populist parties’ participation in government is also confirmed when looking at countries which have experienced a populist government at least once (Model 4 in Table 2). Right-wing populist parties have the strongest and most significant impact in that they are associated with increasing GHG emissions. Left-wing populist parties in government have the opposite effect and centrist populist parties have no significant impact. However, in general, populist parties in government have a negative effect on GHG emissions – that means emissions increase when they come to power. These results again confirm the hypotheses except that the impact of centrist populist parties is not significant.

Table 2. The Partisan effect of populist parties on greenhouse gas emissions in various regions in Europe.

	(1)	(2)	(3)	(4)
	North West	East	South	Populist Countries
All populist parties	0.210*	0.190**	0.127**	0.120***
	(2.46)	(2.95)	(3.01)	(3.61)
Right populist parties	0.185^+	0.201*	0.0545	0.192***
	(1.65)	(2.43)	(1.40)	(4.05)
Left populist parties	−0.0178	0.0783	−0.155**	−0.120^+
	(−0.16)	(1.07)	(−3.14)	(−1.93)
Center populist parties		0.0601	0.0476	−0.0567
		(0.68)	(1.07)	(−1.46)
Left parties	−0.156***	−0.255*	−0.186***	0.0213
	(−4.40)	(−2.48)	(−5.86)	(0.62)
Center parties	0.183***	0.0230	0.0328	−0.0820*
	(4.60)	(0.29)	(0.83)	(−2.49)

Explanation: t statistics in parentheses; ⁺p < 0.10, *p < 0.05, **p < 0.01, ***p < 0.001. The table is composed of a fully interactive model for the geographical regions North West, East and South (models 1–3), on the one hand, and several models for countries with populist governments (model 4). The results for Left and Center Parties are models with all governments with populist parties. That means that the coefficients are not strictly comparable. For the original regression analyses, see supplementary material (Tables A3 and A4).

Conclusion

In a complete survey of all 28 EU member states, the paper clearly shows that once in power, populist parties have a profound impact on GHG emissions. This is not only true for the magnitude of the effect but also for the time lag between the installation of a populist government and its effect. No other party family in power causes such a rapid effect on GHG emissions. By focusing on policy outcome, the paper breaks new ground in that it not just looks at intentions and declarations but at actual effects.

However, the direction of change is dependent on the ideology a populist party aligns with. There is substantive proof that populist parties in government with a right-wing ideology have a negative impact on GHG emissions. When putting this finding in context, climate policy may be another example of the new divide in Europe between left-libertarian green positions versus traditional, authoritarian nationalism which has been identified in other policy areas (Hooghe & Marks, 2018; Kriesi et al., 2012; von Homeyer et al., 2021). Other colors of populism have a much less clear impact.

By and large, these findings hold across all EU member states, although there are also some distinct regional differences. In the Northwest and above all Eastern Europe, right-wing populist parties have a decisive impact on increasing GHG emissions while left populist parties have no significant impact. In this perspective, Southern Europe is an exception. Southern Europe is the region where left populist parties have the effect that emissions decrease when left populist parties participate in coalitions.

While the empirical findings are intriguing and robust, the underlying causal mechanism is open for speculation. The rapid impact of populist parties suggests that the causal mechanism is not simply that populist parties come to power and change policies, which then have an impact on outcomes. As mentioned above, there are many other ways in which parties in government can change outcomes. One important aspect is redefining reality. Once populists participate in government they have the authority and resources to bolster their post-factual narrative on climate change and gain support from other political and economic actors as well as part of the population to discount climate policy (see for instance, the concept of informal coalitions of populist governments in the process of Europeanization; Escartin, 2020). These other actors do not necessarily have to believe the populist narrative themselves. It may often be sufficient that they think that others will make less climate effort, in which case they may merely try to avoid the game theoretic ‘sucker’ payoff of cooperation while others defect. The effect is a lax attitude towards implementation, enforcement, compliance and proactive engagement with climate policies which results in an increase of GHG emissions. However, this causal effect needs further investigation to better understand the power of post-factual communication and its importance for populist parties in government. This macro-comparative study can only provide indications of these causal mechanisms, and systematic qualitative process tracing and pattern matching must support this assumption.

In sum: mitigating climate change is a complex task. The conclusion of this paper is that populist parties create a distinct party family which certainly has an effect on combating climate change and whose impact is much more effective than so far estimated.

Acknowledgements

I would like to thank the participants of the GOVTRAN International Workshop on ‘Climate and Energy Governance and the Future of Europe' in Rome, Italy, in September 2019 for their helpful comments on the first draft of this article. Particularly helpful were the very encouraging and constructive comments from the editors of this special issue and the anonymous reviewers of the journal on earlier versions of this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Jean Monnet Network GOVTRAN – Governing the EU’s Climate and Energy Transition in Turbulent Times, with the support of the Erasmus+ programme of the European Union.

Notes on contributors

Detlef Jahn

Detlef Jahn is a Professor of comparative politics at the University of Greifswald (Germany). His latest book is on ‘The Politics of Environmental Performance' with Cambridge University Press (2018). For more information see https://ipk.uni-greifswald.de/politikwissenschaft/professor-dr-detlef-jahn/

Notes

1 For a classification of the countries to the individual regions see Table A1 in the supplementary material.

2 Total GHG emissions include carbon dioxide, methane nitrous oxide, unspecified mix of hydrofluorocarbons and perfluorocarbons, sulfur hexafluoride, and nitrogen trifluoride.

3 The notion of ‘center parties’ is a broad term. A more detailed examination of the party families revealed that the Christian center parties had no impact on GHGE. In contrast, secular center parties like the agrarian parties in the Nordic countries or the Democrats in the United States have a negative impact on reducing GHG emissions (Jahn, 2021).