Exploring the democracy-climate nexus: a review of correlations between democracy and climate policy performance

ABSTRACT In order to explore the strengths and weaknesses of democracies in mitigating climate change, this article presents a review of more than two decades of research on the democracy-climate nexus. It studies 72 identified articles and book chapters in which correlation analyses between indicators of democracy and climate policy performances have been conducted. The review confirms that democracies tend to generate better climate policy outputs than autocracies, in terms of adoption of policies, laws and regulations. However, there is weak empirical evidence for an association between democratic development and CO2 emission reductions. While empirical evidence shows that democracy can promote decarbonization, aspects such as economic growth, income distribution, energy mix, state capacity and corruption can influence the outcome of decarbonization policy or even counteract it. The research indicates that with deployment of renewable energy, economic activities are increasingly disconnected from fossil fuel dependence, and the political influence of the fossil fuel industry reduced. This process could also enhance the capacity of democracies to accelerate the energy transition and reduce emission levels. Investments in renewable energy, together with policies aiming at combating corruption and accomplishing a fairer wealth distribution, could help to unleash the transformative capacity of democracy towards a low-carbon future. Key policy insights Studies show that democracies tend to generate better climate policy outputs than autocracies, but the empirical evidence that democracy promotes climate policy impact, such as decrease of CO2 emissions, is weak. Economic growth, income distribution, energy mix, and corruption influence climate policy performance significantly more than the level of democracy. In developing countries with rapid growth, democratic qualities do not have any noteworthy alleviating effect on emissions generated by growth; hence combating poverty and in parallel advancing and sustaining human freedom, without carbon-intensive economic development, can be challenging. Investments in renewable energy can enhance the capacity of democracies to decrease CO2 emissions, as economic activities are increasingly disconnected from fossil fuel dependency and the related political influence of the fossil fuel industry. None of the reviewed studies provide any evidence suggesting that autocratic regimes perform better on climate policy than democracies.


Introduction
In the years after the fall of the Berlin wall and the disintegration of the Soviet Union, the question of environmental credentials of democracies began to receive increasing attention.The experience of ecological disasters under socialist rule were seen as evidence of the pre-eminence of democracy in environmental terms (Schultz & Crockett, 1990;Waller & Millard, 1992).In a widely cited article, Payne (1995) articulated arguments for the environmental merits of democracy.Democracies are beneficial for the environment, he argued, since they allow a freer flow of information, have a capacity for political learning, draw lessons from environmental successes and failures, and are more likely to respect the human rights of environmental lobby groups.
At the time though, there was 'little direct empirical research on the possible affinities between democracy and ecology' (Payne, 1995).Only a few articles were published in the 1990s, presenting evidence that democracies performed better on environmental policies (Congleton, 1992;Kamieniecki, 1991;Waller & Millard, 1992).In subsequent years, several studies were conducted, verifying a correlation between the level of democracy and environmental performance, particularly related to local air pollutants (Barrett & Graddy, 2000;Bernauer & Koubi, 2006;Midlarsky, 1998;Scruggs, 1998).Such emissions have a more obvious impact on human health and the environment and are prone to upset the electorate.In this sense, democracies have proven more responsive to environmental degradation than autocratic regimes.Other studies confirm an association between democracy and water quality (Barrett & Graddy, 2000;Mak Arvin & Lew, 2011), biodiversity (Shandra et al., 2005) and soil erosion (Midlarsky, 1998) respectively.
Research on these aspects is, however, not conclusive.Several studies have found that the level of democracy has insignificant or even negative environmental effects.Scruggs (1998) suggested that democratic progress is associated with reduced water quality and Povitkina et al. (2015) found evidence for an association between democracy and damages on marine ecosystems.This ambiguity was confirmed in a comprehensive literature review of democracy and biodiversity (Rydén et al., 2020).
When it comes to climate policy performance, the influence of democracy is possibly even more uncertain.Regardless of the consensus on the relationship between anthropogenic emissions of greenhouse gases and climate change (IPCC 2021), governments worldwide have failed to reduce emissions in line with the Paris Agreement (UN 2021).Within the climate governance literature, democratic decision-making has accordingly been criticized for being unfit to deal with the urgency of climate change.While global warming requires farsightedness, the election cycles and public debates tend to make democracies short-term focused (Andrić, 2019;Compston & Bailey, 2013;Giddens, 2009).Other critical aspects raised in the literature are the influence of 'veto players' (Madden, 2014;Tsebelis, 2002), geographic constraints of democratic decisionmaking (Lawrence, 2015), the lack of scientific literacy of the electorate (Shearman & Smith, 2007), and social and psychological barriers demotivating voters from climate action (Lorenzoni et al., 2007).A few climate scientists have even argued for putting democracy on hold in order to properly handle the climate crisis (see Adam, 2009;Hickman, 2010), and for the need for authoritarian approaches (Beeson, 2010;Shearman & Smith, 2007).
Other scholars argue that the problem is not democracy as such, but rather the incapacity of existing liberal democratic institutions to channel the interests of citizens, claiming that there is a need to deepen and advance democracy, rather than restricting or abandoning it (Willis, 2020).Deese (2019) argues for the need to establish supranational democracy and Fischer (2017) calls for global eco-localism, while proponents of 'critical environmental theory' (Eckersley, 2020), claim that existing institutions need to be reformed to adequately deal with the complex and global challenges of climate change.Motivated by theories of environmental and ecological democracy, such scholars advocate for deliberative democratic innovation (Dryzek & Pickering, 2017;Niemeyer, 2011), such as mini-publics or citizens juries or assemblies (Pickering et al., 2022), through which environmental protection may be expanded beyond national boundaries, and social learning developed between laypeople and experts (Eckersley, 2020).Others have argued for the need to stimulate communication practices to generate political engagement on climate change (Carvalho et al., 2017).
The call for democratic reforms may also be seen as a response to the crisis of democracy, manifested by declining trust in public institutions, rising populism and spread of disinformation (Lockwood, 2018).Moreover, social mobilization on climate-related issues, both the youth-led climate movement, organizations such as Extinction Rebellion and protest movements such as the yellow-vests (Gilet jaunes), have demanded democratic reforms.
This overview shows that various scholars and studies interpret the democracy-climate nexus quite differently.The understanding of the basic correlations between democracy and climate governance seems to differ significantly.So far however, a recent comprehensive review of the literature is missing, and it is not known why conclusions diverge and what factors may best explain the failure to achieve climate goals in democracies.
The aim of this article is to investigate and explore what the literature says about the strengths and weaknesses of democracies in mitigating climate change, by undertaking a comprehensive literature review, with a focus on studies applying correlation analysis between indicators of democracy and climate policy performance.Systematically exploring the literature on the climate-democracy nexus, and the capacities of various governing systems, may provide insights into which governance models may speed up climate transformation, while concurrently preventing social disputes that may arise in this process.
The research questions this article brings clarity to are ( 1) what conclusions can be drawn from the accumulated peer reviewed quantitative research on the relationship between democracy and climate policy performance (2), why do several of these studies provide different conclusions, and ( 3) what factors can explain the weak climate policy performance of democratic governance so far?

Methods and material
This review followed conventional praxis (Denyer & Tranfield, 2009) and is based on a search on Scopus, conducted on 11 September 2022.The search was limited to peer-reviewed research articles and reviews in peerreviewed journals, studying the correlation between indicators for democracy and climate policy performance.The aim is thus not to cover the entire academic discussion on the topic, but to explore analyses of correlation between these variables.The focus of the search was on various aspects of democratic governance and climate included in the title, abstract or among key words of articles. 1Since climate policy indicators are denoted by different terms in different studies, several aspects related to environment and sustainability were included.The search was limited to articles in the English language published after 1992, when the UN Framework Convention on Climate Change was established.Conference papers, books and book chapters were initially omitted.
The search identified 1,656 articles.All abstracts were scanned aiming to select articles that include correlation analysis between democracy and climate policy performance.Studies using only qualitative data were excluded.Articles focusing on less than five countries were also exempted since such limited samples would not allow a sufficiently strong analysis.The initial number of identified articles was fairly high, but a strict selection of only articles applying correlation analysis brought down the number and enabled in-depth analysis.In order to enable comparisons, the review is restricted to studies on the national-level.
After abstract scanning, a total of 68 articles remained (see appendix 1).Through citation tracking we identified another 3 relevant articles.In order to identify additional significant studies, book chapters were included in a second search, with the same terms, on Scopus, on 5 January 2023.However, out of 182 resulting hits, only 1 fulfilled the criteria outlined above regarding research method used and sample size.Consequently, 72 articles were reviewed in detail, and they were divided between those that supported or rejected a positive statistical correlation between democracy and climate policy performance, and those that came to indecisive results.An additional analysis was made of dependent variables applied and which were identified to substantially influence the democracy-climate nexus.These were economic growth, income equality, energy mix, institutional capacity (level of corruption) and variations in the democratic system.The articles were coded manually 1 TITLE-ABS-KEY ((democ* OR "governance quality" OR "governance indicator" OR "governance performance")) AND TITLE-ABS-KEY (("carbon emission" OR "carbon dioxide emission" OR "CO* emission" OR "greenhouse gas" OR "pollution" OR "environment quality" OR "climate change mitigation" OR "climate policy" OR "environmental policy*" OR "environmental agreement" OR "sustainability indicator" OR "sustainability performance"))) AND PUBYEAR > 1991 AND (LIMIT-TO (DOCTYPE, "ar") OR LIMIT-TO (DOCTYPE, "re")) AND (LIMIT-TO (LANGUAGE, "English")) AND (LIMIT-TO (SRCTYPE, "j")).
by the authors according to these variables, which enabled in-depth analysis of comparable studies.We also analysed methodological approaches such as time-period covered and sample size.

Defining key variables: 'climate policy performance' and 'democracy'
Climate policy refers to policies aiming at reducing greenhouse gas emissions ('mitigation') and to adapt societies to climate change ('adaptation') (Huitema et al., 2011).Climate policy performance can be measured on the basis of policy 'outputs', 'impacts' or 'outcomes' (Young, 2004).Climate policy outputs encompass mitigation policies such as adopted legislation, targets, and engagement in international climate negotiations.Policy impacts are the actual results of such policies, i.e. reduction of carbon dioxide (CO 2 ) or other greenhouse gas (GHG) emissions.Progress can also be assessed by the ranking of countries in indexes, which can encompass both output and impact measurements.Policy impacts can also be measured on indirect mitigation parameters such as decreased deforestation, investment in renewable energy, or technology adoption.Climate policy outcomes, finally, are also used in some papers and commonly refer to social parameters such as changes of views or behaviours.Since only a few studies measure attitudes or perceptions toward climate polices this study focuses on outputs and impacts, thus only including articles measuring mitigation policies and policy impacts (i.e.actual reduction of CO 2 or other GHG emissions).
Democracy is a concept that varies in shapes and forms, but can be defined as a system that 'emphasizes popular control over decision-making and political equality among those exercising that control' (Beetham et al., 2008).This study focuses on the concept 'democracy' as a variable in a broad sense and therefore allows a variety of definitions in the reviewed studies, provided that recognized democracy indexes are used as variables, such as Polity, Freedom House and V-Dem.These indexes measure democracy by the quality of electoral systems, political participation, level of equal access to resources, power, and freedoms across various groups within a society, strength and independence of judiciary and institutions, freedom of media, civil society and respect of the freedom of speech and association and other relevant aspects of civil liberties and human rights.

Studies exploring the influence of democracy on climate policy performances
Over the past two decades, the relationship between democracy and climate policy performance, has been analysed in several studies.The earliest article in this review was published in 1998, while 10 articles were published in the first half of 2022.Unsurprisingly, the number of studies has been growing, probably due to the increasing urgency of climate change.Earlier articles focused on the influence of democracy on environmental policy in general rather than on climate policies.
Figure 1 characterizes the reviewed studies by their results relating to policy output and policy impacts, and whether they support, reject or are indecisive on the correlation between democracy and climate policy performance.Several studies came to indecisive results, meaning that democracy may have an effect on policy performance depending on aspects such as economic development or corruption (see below).Eight studies measured both impacts and outputs and the total number of results presented here is thus 80.
Figure 1 shows that a solid body of research verifies the proposition that democracy influences climate policy outputs.The search string identified eight studies exploring policy outputs and eight exploring both outputs and impacts.Only two of these came to an indecisive result, while the 14 others confirmed a positive correlation between democracy and climate policy outputs.Studies measuring climate policy impacts show far more inconclusive results.A slight majority failed to confirm any positive correlation between democratic governance and CO 2 emissions.Among the 64 studies measuring policy impacts, 22 rejected a correlation between democratic governance and emission levels, while 20 found a positive correlation, and 22 came to an indecisive result, impeding any firm conclusions.The studies moreover differ on research design, time-period analysed, sample size and number and type of variables applied.Quantifying the number of studies supporting or rejecting a correlation between the level of democracy and climate change performance could thus be misleading.

Democracy variables used in the studies
Democracy is a concept that can be measured by different parameters.Most studies apply well-known indexes, such as the Polity data series, Freedom House and the Varieties of Democracy (V-Dem).A few of the reviewed articles use the Vanhanen index or the Democracy Index of the Economists Intelligence Unit.Several studies use a combination of these and other datasets (see appendix 1).All datasets include different indicators of democratic performance.None of the studies have applied a comparative approach with different datasets, and it is thus difficult to say whether the choice of dataset influences the outcome.Certain variations between different democratic systems were found and presented below.

Variables measuring climate policy performance
Climate policy impacts are normally measured by production-based emissions.Data come from a few sources such as the World Resources Institute Climate Data Explorer, EDGAR (Global Atmospheric Research) and the World Bank.The choice of emission data appears not to have a decisive effect on the results of the studies.It is however likely that the results would differ if studies would be done on consumption-based emissions, but that could affect the research design since such emissions are largely beyond national policy control.
Climate policy output is a more elusive measurement than impacts since it is difficult to quantify performance success.A few studies use policy performance indexes, such as the rating on the Climate Laws, Institutions and Measures Index (CLIMI) (Fredriksson & Neumayer, 2013;Pellegrini & Gerlagh, 2006), while others use indicators not restricted to climate policy performance (Baettig & Bernauer, 2009;Ghodrati et al., 2018;Mukherjee, 2013;Neumayer, 2002).With the development of climate policies worldwide, the ability to measure policy outputs has improved.In accordance with the Paris Agreement, governments are required to submit Nationally Determined Contributions (NDCs), which constitute a fairly robust policy output indicator used by Boräng et al. (2019) and Tørstad et al. (2020).The growing body of climate policy research has also enabled more accurate measurements and databases on climate change laws have been developed by e.g. the Grantham Institute, Environment and Sabin Center for Climate Change Law and the OECD (Environmental Policies Stringency, EPS).These data sources are used by Lamb and Minx (2020) and Povitkina and Jagers (2022).Povitkina and Jagers (2022) also use data from the Transformation Index from the Bertelsmann Stiftung.Further advancement of policy research would enable more elaborated studies of the effect of different governance capacities.

The time-periods of emission data
The time-period of the emission data applied does not seem to have any decisive influence on results.Most studies base results on time-periods prior to the Paris Agreement, omitting more recent potential policy effects.Only 15 studies include emission-data from the period 2016-2020, out of which five reject a positive correlation between democratic development and emission levels, while five support such a correlation and five are indecisive.Studies applying data from earlier time-periods show similar variations, and out of the 50 analyses conducted on data up to 2015, 15 verified a positive correlation between democracy and climate policy impacts, while 17 came to indecisive results and 18 rejected an association. 2 Kammerlander and Schulze (2021) ran analyses with different time-periods in order to control potential variations, but found no evidence for the influence of democracy on emission levels regardless of the time-period of the emission data.

Sample size
The sample size in terms of the number of countries involved in the correlation analysis affects the results.Exploring a limited number of countries might be beneficial since they enable comparisons between countries within the same geographic area or countries with similar economic development.However, in order to draw general conclusions on the democracy-climate nexus, a sufficiently large sample size might be required.Studies involving a large sample size (in the order of 100 countries), tend to give less support to democracy as a factor influencing climate policy impacts, than studies with a smaller sample size or studies focusing on countries of a specific geographic area.Out of the 20 studies exploring a sample size of around 100 countries, only two confirmed any positive correlation between democratic development and climate policy impact, while seven rejected a correlation.Most studies came to inconclusive results.

Economic growth
Exploring the reasons behind the unconvincing empirical evidence, economic growth seems to be an outstanding factor.The correlation between economic development and democratization is well established (Lipset, 1959).Economic growth does not necessarily lead to democratization, but there is empirical evidence for a certain level of economic welfare to be a prerequisite for the development of liberal democracy (Przeworski, 2000).The correlation between economic growth and CO 2 emissions is consistent in the reviewed studies, and several of the studies that fail to show any positive correlation between democracy and climate policy performance single out economic growth as a determinant factor.In the earliest study identified in this review, Midlarsky (1998, p. 353) concludes that 'as expected economic development is most strongly associated with CO 2 emissions', and this factor can explain why the association between democracy and climate policy impacts is non-existent in his analysis.
Several studies suggest however, that the relationship between income growth and emissions has an inverted-U shape, a theory referred to as the Environment Kuznets Curve (EKC).Grossman and Krueger (1995) argued that with a rise in income levels, the emissions of pollutants will increase, yet beyond a certain level of per capita income the trend reverses.According to this theory, economic growth leads to a development of new and cleaner technologies and an electorate that demand environmental measures.Several studies have also confirmed the relevance of the EKC when it comes to certain local pollutants, such as sulphur dioxide, lead and mercury (Gassebner et al., 2011), while correlations for CO 2 emissions are less robust (Farzin & Bond, 2006;Farzanegan & Markwardt, 2018;Jahanger et al. 2022a).An argument for the lack of statistical evidence of the EKC on CO 2 emissions is the global aspects of climate change.
It could also be argued that the level of democracy is influential in defining the tipping point in the EKC, since political freedom in combination with growing income would enable and inspire people to demand pollution-reducing policies.According to Laegreid and Povitkina (2018) climatic pressures of economic growth can be alleviated by strong civil society activity, along with democracy.They found that an increase of GDP per capita of 1 percent generates increases of CO 2 emissions of 0.75%, but the climate impact of growth is 2 A few studies explored emissions of just one single year.
Research on the correlations between democracy, income growth and emissions is however inconclusive and several studies show mixed or insignificant support for the assumption that democracy could influence the EKC for CO 2 emissions.Selseng et al. (2022) claim that democracy can mitigate emissions generated by GDP per capita growth, yet this influence subsides at higher level of growth, and the authors consequently do not provide any convincing evidence for a non-linear GDP-CO 2 relationship.Kammerlander and Schulze (2021) show that GDP growth per capita leads to higher emissions, regardless of the level of democracy, and this relationship is only insignificantly weaker at high income levels.Mak Arvin and Lew (2011), who examine the relationship between democracy and environmental quality in a sample of 141 countries, conclude that while democracy gives positive climate policy impacts in upper and middle-income countries, it has no effect in the poorest and richest countries.
Several other reviewed studies reject the effect of democracy in combination with the EKC on CO 2 emissions altogether.Sulemana et al. (2017), examining 47 African and 31 high-income OECD countries, found support for the EKC in both samples, but no evidence for democracy to have any influence on emissions.Joshi and Beck (2018), studying 22 OECD countries, Jain and Kaur (2022), studying 5 South East Asian counties, and Satrovic et al. (2021), studying 6 Arab states, found weak evidence for an EKC effect on CO 2 emissions and results did not support that democracy influence CO 2 emissions.Gill et al. (2019) argue, on a basis of their study of the influence of civil liberties on CO 2 emissions in ASEAN countries, that democracy can even have an opposite effect, delaying the turning point in the EKC.Akalin and Erdogan (2021) examined the democracy-environmental degradation nexus in 26 OECD-countries, and came to a similar conclusion, arguing that democracy may 'promote entrepreneurial freedom and encourage economic growth, but it may not help with reducing environmental pollution or provide clean and sustainable ecology'.Eren (2022), explored the effect of democracy and income level on CO 2 emissions in a sample of 72 countries from 1993 to 2019, and found that voter demand for welfare is stronger than the desire for environmental protection, a hypothesis sometimes referred to as the 'Treadmill of Production Theory' (TPT) (Mayer, 2017).Acheampong et al. (2022) explored data on 46 sub-Saharan African countries (SSA), demonstrating that regardless of an improvement in democracy, economic growth leads to higher carbon emissions in several SSA countries.Democratically elected leaders in these countries seem to be unable to 'overcome the resistance of stakeholders who see environmental regulation as inimical to their short-term economic interest' (2022, p. 12).
Most of these studies are, however, focusing on developing countries, and the results of several studies applying a larger sample than 90 countries reveal differences between high-and low-income countries.These conclusions are also reached by Mao (2019) and Aller et al. (2021), showing that economic growth in poor countries results in increasing emissions regardless of the level of democracy.Studies based on samples of high income countries also present more favourable results for the prospects of democracy to influence the EKC.Kim et al. (2019) argue, based on an analysis of 78 higher income countries and 53 lower income countries, that it is only in high-income countries that democracy positively contributes to the environment.Lu et al. (2022), studying growth rates in 35 OECD economies, suggest that per capita income and democracy reduces carbon emission growth rates in the developed economies.Laegreid (2014) verifies the EKC hypothesis in a sample of 36 countries in western Europe and the US, demonstrating that a robust 'green' civil society influences emissions in countries where the democratic system is functioning well, but also that the turning point in the EKC occurs at exceptionally high-income levels.
These findings point out that it is primarily in developed and wealthy countries where democracy has an alleviating effect on growth-related emissions, while the level of democracy has less influence in rapidly growing and low-income countries.Developing and poor countries aiming to reach higher levels of national wealth might thus have difficulties in combining such efforts with both democratization and stringent climate polices.Soysa (2022), exploring a sample of 160 countries, concludes that economic freedom shows positive environmental effects compared with increased democratic egalitarianism.This conclusion is not supported by Sheraz et al. (2021), analysing a panel data sample consisting of 179 countries and finding evidence that democracy combined with economic freedom reduces emissions.Evidently, the influence of democracy on the EKC and CO 2 emissions depends on national contexts, and aspects such as technological development, energy mix and carbon intensity, institutional capacity and income distribution, as discussed below.

Income inequality
An aspect of the EKC hypothesis is income equality, which together with democratic development could affect climate policy performance.A few of the reviewed studies show that income equality influences the capacity of democracies to deliver climate policy impacts.According to Policardo (2016), exploring the effect of democracy on CO 2 emissions in 47 transition countries, income inequality appears to influence climate policy impact more profoundly than the level of democracy does.In democracies that suffer from severe income inequalities, ' … the decisive citizen is poorer and so less willing to pay for environmental protection', she concludes (2016, p. 21).You et al. (2020) found evidence for democracy levels to promote a nonlinear correlation between income inequality and CO 2 emissions, and suggest that high inequality together with poor democratic quality generate higher emissions.Bakaki et al. (2022) conclude that democracy has a positive effect on climate policy impacts; however, the effect disappears in more unequal societies.They conclude that ' … when inequality is high and the rich oppose tighter environmental regulation, those characteristics of a democratic system that are linked to more progress in fighting carbon emissions are unlikely to be effective' (2022, p. 171).Povitkina and Jagers (2022) studied democracies with more social-liberal features, implementing policies to achieve equal access to public services, such as healthcare and education.They conclude that when such democratic countries spend higher amounts on public rather than particularistic goods, they tend to adopt more stringent environmental policies.This conclusion is valid for not only high-income countries, but also for developing countries.
Even though inequality influences both emissions and the ability of democracies to deliver climate policies, it is nevertheless unclear whether policies aiming at a fairer distribution of wealth would result in positive climate policy impacts.Selseng et al. (2022) explored the effect of democracy on 127 countries from 1992 to 2014.They suggest that policies for equal wealth distribution can increase CO 2 emissions.Democratic qualities of a country had no significant impact on emission levels, while countries that over time provided more equal access to universal goods, such as food, water, housing and income, had significantly increased per capita CO 2 emissions.Such policies are, just as economic growth, carbon intensive.

Fossil fuel dependency and renewable energy deployment
The share of fossil fuel energy in a specific country obviously impacts on emission levels, but the energy mix can also influence democracy and its capacity to deliver climate policy outputs and impacts.A few of the reviewed studies explore the relevance of the energy mix and identify a positive correlation between deployment of renewable energy and reduced CO 2 emissions (Adams & Nsiah, 2019;Akalin & Erdogan, 2021;Acheampong et al., 2022;Jahanger et al. 2022a,b).While this is expected, domestic fossil fuel production does not seem to result in an opposite effect.Kammerlander and Schulze (2021) and Selseng et al. (2022) did not find any significant evidence for the influence of domestic oil production on emission levels.Selseng et al. (2022) show on the other hand that the carbon intensity and technology advances of individual countries have significant effects on CO 2 emissions per capita.In countries heavily dependent of fossil fuels, economic growth will lead to higher emission levels per capita, while in countries having succeeded in deploying renewable energy and green technology, the negative impact of growth is lower.
With falling costs for renewable energy and other types of environmental technology, the detrimental effects of economic growth could be eased, and the capacity of democracy to influence emissions enhanced.Güngör et al. (2021), who found evidence for the EKC in their study, demonstrate that democracy could increase both energy consumption and CO 2 emissions in the short run.However, in the long run, when a more advanced technological stage is reached, democracy starts to have a strong emission-reducing effect.The capacity for democratic governance to influence CO 2 emissions are thereby likely to increase once the carbon intensity in the energy mix is reduced.Lv (2017) shows accordingly that high income countries with lower emission levels can benefit from democracy.Eren (2022) demonstrates that democracy can affect emission levels in lowand medium-emission countries, but not in countries with higher emissions.
An issue of relevance here is to what extent democracy as such could influence the transformation away from fossil fuels.A few of the studies focusing on climate policy outputs apply measurements that include deployment of renewable energy as an indicator (Böhmelt et al., 2016;Lamb & Minx, 2020;Povitkina & Jagers, 2022), and democracies tend to perform better in this respect.Lu et al. (2022) argue based upon their study of 35 OECD countries, that democracy and trade liberalization are potential drivers for adopting emission-reducing technologies.On the other hand, von Stein (2022, p. 349) shows that countries with free and fair elections consume more energy and rely more heavily on non-renewables, but also that 'countries with robust civil societies fare significantly better on all of those outcomes'.Studies also indicate that energy system transformations are promoted by market forces rather than voter demand.Ahmed et al. (2022), studying the G7 countries, demonstrate that renewable energy technology investments are spurred by economic growth, which thus may reduce ecological footprints, but they saw no effect of democracy significantly influencing ecological footprints.When it comes to energy consumption, Adams and Nsiah (2019) and Acheampong et al. (2022) show that democracy can lead to higher energy consumption, while Chou et al. (2020), exploring 26 countries in America, find that democracy positively influences energy efficiency.It is obviously difficult to draw definite conclusions on these correlations upon the few heterogeneous studies reviewed.
A few studies also analyse the influence of fossil fuel dependency and fossil fuel extraction on climate policy performances.Tørstad et al. (2020) explore correlations between climate commitments under the Paris Agreement and democracy performance, democratic values and oil, coal and gas rent.According to them, democracy is associated with higher climate ambition, while fossil fuel and especially coal dependency had a negative influence.Lamb and Minx (2020) observe that the group of countries strongly dependent on oil and gas resource extraction have the lowest levels of climate policy adoption and stringency, the highest fossil fuel subsidies, and the highest emissions.The high emission levels could be explained by the extraction processes, they argue, and from poor incentives to pursue climate policies.These countries also present serious deficiencies in controlling corruption and advancing democratic norms.This is not necessarily the case for coal dependent states, which constitute a group of countries showing greater diversity in both democratic norms and climate policy outputs.

Institutional capacity and corruption
Besides economic conditions, climate and environmental policy performances are strongly associated with state and institutional capacity.State capacity is closely related to functioning democratic governance.This aspect is indirectly covered by the Polity data set, while corruption is explicitly included in V-Dem datasets and in the Freedom House index.Several of the reviewed studies have also highlighted the negative climate impact of corruption.Halkos and Tzeremes (2013) studied governance and CO 2 emissions by using data from the economies in the G20 group.They present evidence for the level of freedom of expression, freedom of association and freedom of media to influence CO 2 -emission levels, yet the respect of the rule of law and level of corruption have stronger impact.Pellegrini and Gerlagh (2006) also confirm that the current level of democracy has a minor and statistically insignificant effect once corruption is included in the correlation analysis.One of the more comprehensive studies on the subject was conducted by Povitkina (2018), using data on corruption, democracy and CO 2 emissions from a sample of 144 countries over the time-period between 1990 and 2012.The study shows that democratic development is associated with decreased CO 2 emissions only in countries with a relatively low level of corruption.If the level of corruption is high, the effect of democracy is insignificant.Similar results were found by Laegreid and Povitkina (2018).
Studies on a variety of nations also point out that corruption and institutional quality influence the level of CO 2 emission or climate policy impacts, such as Rey and Ozymy (2019), studying 20 Latin American countries, Bae et al. (2017), exploring 15 post-Soviet countries, Zandi et al. (2019), examining a panel of six ASEAN countries, Azam et al. (2021), studying 66 developing countries, Aller et al. (2021), studying 92 developed and developing countries, and Hamid et al. (2022) exploring BRICS-countries.Adams and Klobodu (2017) studying 38 African countries, and Abid (2016), exploring 25 Sub-Saharan Africa countries, also identified corruption and institutional quality as factors influencing climate policy impacts.Sommer (2020) uses a two-way fixed effects regression analysis for 150 nations from 1971 to 2014 to understand how clientelistic relationships and democracy impact CO 2 emissions.He argues that 'clientelism is a missing piece of the puzzle in explaining how democracy impacts CO 2 emissions ' (2020, p. 416).When controlling for clientelist relationships in nations with high and low incomes, democracies are more effective in reducing CO 2 emissions.
In a study of 99 countries, Lamb and Minx (2020) explore constraints that actively hinder climate policy progress.They identify that lack of democracy and corruption are aspects that matter for both climate policy outputs and impacts.However, they also conclude that 'strong democratic norms and institutions free from corruption are a rarity', demonstrating that emissions remain high in wealthy countries for other reasons (2020, p. 12).Povitkina and Jagers (2022) analysed climate policy performances of 137 countries and found, on the other hand, a positive correlation between corruption and climate policy output, suggesting that countries with high level of corruption adopt more climate-related laws and policies.Despite this, corrupt countries have lower environmental policy stringency, revealing that the adopted policies are not translated into emission reductions.Such countries may adopt more ambitious policies, but when corruption is widespread, public authorities seem incapacitated to properly execute and enforce the adopted environmental laws and regulations.

Different types of democracies and influences on climate policy performances
There is limited evidence for the variations of different democratic systems to have any significant influence on climate policy performance.A few studies conclude that presidential systems or democracies where the political power is vested in one centralist institution are better placed to outmanoeuvre veto players and present efficient climate policies.Wurster (2013) and Clulow (2019) find indications for majoritarian-presidential electoral systems to perform better than proportional-parliamentary systems, partly because of the influence they can afford to environmental NGOs.von Stein (2022) measured the effects of political constraints on environmental policies, including CO 2 emissions, and concluded that a great degree of institutional constraints could slow down both favourable and damaging policy change.Ward (2008), Fredriksson and Millimet (2004) and Fredriksson and Wollscheid (2007) demonstrate on the other hand that countries with a centralistic power structure allow less room for environment lobby groups and social movements to assert political pressure, arguing than consensus-orientated parliamentary systems produce better climate policy outputs and impacts than presidential and congressional systems.A similar conclusion is reached by Laegreid (2014), whose results suggest that political systems with restrictions in decisionmaking and with more consensual and corporative systems generate more effective climate measures and lower emissions, than countries with a more centralized political system.Kammerlander and Schulze (2021) were however unable to find evidence for any differences between these two electoral systems in terms of climate policy impacts.Rey and Ozymy (2019), exploring environmental performance in 20 Latin American democracies, argue that electoral rules that prescribe higher levels of representativeness and foster party discipline, achieve better performance.
Another dividing line in the democracy policy debate concerns representative versus participatory democracy.Some scholars argue that political systems allowing deliberative approaches could offer citizens opportunities to counteract the power of veto players, such as fossil fuel interests.There are however few empirical studies exploring these theories.An attempt to measure the effect of deliberative democratic initiatives was done by Povitkina and Jagers (2022), applying indicators within the V-Dem dataset that capture the extent to which policymakers deliberate and consult with relevant actors.They found evidence for a higher level of deliberative features to positively influence the adoption of climate laws and policies, even though such democracies pursue less stringent environment polices.The authors' explanation is that 'countries that open up channels for deliberation not only enable pro-environmental interests to join forces, but in parallel also help giving voice to more anti-environmental interests … ' (2022, p. 10).Selseng et al. (2022) have also explored the impact of inclusion and deliberative measures on climate policies without finding any effect of these approaches on CO 2 emissions.Their study did not measure deliberation in climate policy processes explicitly and the authors also confessed that the effect of deliberations is difficult to study since it has not been applied at the scale necessary to influence national CO 2 emissions.
Conversely, a few studies indicate that inclusive democracies stimulate citizen participation in climate policy formation.Böhmelt et al. (2016) suggest that inclusiveness, measured by indicators of liberal freedoms, a competitive political process and the strength of civil society, is associated with increased climate policy outputs, yet finding no evidence for these aspects to influence emission levels.Escher and Walter-Rogg (2018) and von Stein (2022) also emphasize the role of civil society, although von Stein (2022) shows that countries with free and fair elections have significantly higher emissions than others.Boräng et al. (2019) explore effects of inclusiveness on climate policy outputs and impacts in 149 countries, measured by for instance equal opportunity to express views and the strength of civil society.They conclude that inclusiveness leads to better climate policy outputs measured by the ambition of NDCs under the Paris Agreement, although they fail to present evidence for a correlation with climate policy impacts.Pacheco-Vega and Murdie (2022) argue that democratic development can reduce CO 2 emissions, demonstrating that political civil liberties, the ability of citizens to participate in advocacy, and the sensitivity to international criticism, can influence policy performance.Ergas and York (2012) do not present evidence for democracy to have any significant effects on CO 2 emissions, however they suggest that the level of female political representation matters, arguing that gender equality would improve climate policy impacts.It can nevertheless be concluded that variations between different types of democracies have seemingly marginal influence on policy impacts.However, within the family of developed democracies this could be a factor of importance, although the research so far is too limited to draw any definite conclusions on the matter.

Discussion
Having reviewed the research on the democracy-climate nexus conducted over the previous two decades, this article concludes that democracies are generating better climate policy outputs than autocracies, in terms of adopted mitigation measures, laws and institutions, as well as commitments to international treaties and engagements in international negotiations.However, the empirical evidence does not substantially support a relationship between democratic development and climate policy impact in terms of overall emission reductions.Variables such as economic growth, income distribution, energy mix, state capacity and corruption were shown to influence CO 2 emissions more than the democratic capacity of a country.
As discussed in this article, economic growth may provide the foundation for democratization, but this is also a process that historically has generated higher CO 2 emissions.In accordance with the EKC, emissions decline beyond a certain level of growth, when the economy advances to a higher stage of technological development.While far from conclusive, most studies suggest that citizens can use democracy to alleviate the carbon impact of growth in high-income countries with low-corruption.In developing countries with rapid growth, on the other hand, democratic qualities do not seem to have any noteworthy effect on reducing growth-generated emissions.Finding solutions that combat poverty and in parallel advance and sustain human freedom, without carbon-intensive economic development, is consequently a major quest for humanity.
Another issue explored in several reviewed studies is income inequality, which could generate both higher emissions and undermine the capacity of democracy to deliver effective climate polices.Citizens in countries with high levels of income inequality may oppose emission reduction policies, believing that costs for such measures will be unfairly distributed, while high income earners have extremely carbon-intensive lifestyles and may be unwilling to support policies that would restrain it.It might therefore be of relevance to further explore the influence of income inequality in the context of the democracy-climate nexus, particularly as the growing income inequality over the last three decades is claimed to be a factor behind the decline of global democracy (Han & Chang, 2016).Between 1980 and 2020 the wealthiest 1 percent captured 23 percent of global growth (Chancel & Piketty, 2021), while global democracy has declined in the years between 2006 and 2021 (Freedom House 2022).It is thus tempting to suggest that climate change and the decline of global democracy are symptoms of the same disorder, and that they could be alleviated by narrowing the gap between the rich and the poor.
Another variable central to understanding the climate policy underperformance of democracies is institutional capacity and corruption.Democracies suffering from corruption and weak state institutions might be able to present ambitious climate policies, however they tend to implement such policies poorly.
Judging from the reviewed studies, combating corruption and building up institutional capacity should be a priority for speeding up the low-carbon transformation.Here, it is relevant to highlight that corruption and policy capture are often caused by fossil fuel interests, symptoms associated with rentier effects (Ross, 2001).Several reviewed studies demonstrate accordingly that oil and gas extraction companies are negatively influencing both democratic qualities and climate policies performances.A transformation away from fossil fuels could thus result in both stronger democracy and lower emissions, although further research on these aspects would be needed before definite conclusions can be drawn.
Reducing emissions in line with the Paris Agreement is evidently challenging, and so far, not a single democracy has responded adequately.However, the poor performance of democracies cannot be used as an argument for authoritarian responses.None of the studies identified in the review provide any evidence that would suggest that autocratic regimes perform better than democracies.The insufficient climate policy performance of present democracies should rather be seen as an argument to vitalize and strengthen democracy.As demonstrated in this review, political systems that allow civil society organizations and individuals to influence the formation of climate policies, appear to present more ambitious climate policy outputs.Other research has also shown that democracies advance, enable and encourage knowledge and innovation, needed to deal with the complexity of climate change (Campbell, 2019;Carayannis & Campbell, 2021).
It has moreover been suggested in recent publications that more innovative democratic models or deliberative approaches could help to accelerate a climate transformation (Dryzek & Pickering, 2017;Pickering et al., 2022), however the studies reviewed in this paper do not present any evidence for such claims.It could thus be valuable to conduct additional research exploring different innovative democratic models as well as to compare democracies with different constitutional set up and electoral and parliamentary systems.Moreover, it is important to analyse different forms of political engagement and what kind of institutions can facilitate both democratic inclusiveness and effective science-policy interactions.We should not forget that some of the greatest advancement of social justice in modern history, such as women's emancipation and civil rights movements, were achieved through mass mobilization within liberal democracies.With this in mind, we should keep exploring what it takes to achieve a renewed and wider mass mobilization for climate action within traditional liberal democracies.What we do know for sure is that such a mobilization is more likely to occur in open and democratic societies, respecting freedom of speech and freedom of association.In this sense, democracies offer the most hopeful prospects for a sustainable future.

Conclusions
This paper demonstrates that democracies tend to generate better climate policy outputs, while the empirical evidence for democratic development to result in lower CO 2 emissions is weak.The findings on climate policy impacts varies, however, and the inconsistency can to a great extent be explained by variations in research design and data used.The reasons why the climate policy performance of democratic governance has been fairly weak can also be explained by different variables influencing the democracy-climate nexus, particularly economic growth, income equality, fossil fuel dependency, renewable energy deployment, fossil fuel extraction and corruption.The strengths of these correlations differ between different studies, and in a few cases conflicting results were found, while additional relationships between these variables may exist, but are not discussed in the reviewed papers .
Assessing the influence of democracy on climate policy performance is difficult due to the profoundly different challenges that various democratic countries are facing in their efforts to decarbonize.Some countries are in an early stage of climate transformation and are heavily dependent on fossil fuels, while others lack sufficient institutional capacity to develop and enforce climate policies.Some democracies are industrialized, while others are undergoing rapid economic growth and are in the process of, or not yet, industrializing.These variables can obscure the effects of democracy on climate policy impacts, and may explain the variation of results.Nevertheless, this review demonstrates that democracy, together with the development of renewable energy, improvement of governance capacity and reduction of income inequality, can contribute to a transformation towards a low-carbon future.

Figure 1 .
Figure1.Study results on democracy and climate policy correlations.