Renewable Energy Transition and the Paris Agreement: How Governance Quality Makes a Difference?

Abstract

This paper investigates whether the Paris Agreement affects renewable energy deployment and how institutional quality moderates this relationship. According to a generalized method of moments estimation for panel data for both developed and developing countries over the period 2000–2022, the Paris Agreement positively influences renewable energy deployment, suggesting that countries are promoting renewable energy to align with institutional expectations to maintain their reputations. The results further show that governance quality is the main determinant of renewable energy deployment. However, the moderating role of governance underscores the less-pronounced impact of the Paris Agreement on countries with high governance indicators, suggesting that these countries may have shifted their focus toward other avenues of climate management beyond the deployment of renewable energy. Furthermore, there is strong evidence of the relationship between forest area, CO₂ emission, trade openness, domestic credit, and renewable energy deployment. The results are robust with the use of a dynamic panel threshold model.

1. Introduction

The need for a global shift to renewable energy sources is more important than ever, considering the worsening environmental conditions and the necessity to combat climate change [1]. This momentum highlights the global recognition of the need to reduce carbon emissions, which is a key step toward sustainable development [2]. Thus, renewable energy development has become a key component in the transformation toward a low-carbon economy [3,4]. According to Tolliver et al. [5], the share of renewable energy in total energy is expected to increase by 60% by 2040 to accomplish the United Nations’ energy-related Sustainability Development Goals (SDG). Given that renewable energy is a critical solution in combating climate change, studying the determinants of renewable energy consumption based on the quality of a country’s institutions is essential for businesses, policymakers, and communities.

The existing literature reviews the influence of macroeconomic, environmental, and technological factors on renewable energy deployment, as evidenced by studies conducted by Gozgor et al. [6], Akintande et al. [7], Lawal [8], and Gogoi and Hussain [9], neglecting the impact of countries’ engagement in international agreements on renewable energy deployment. This gap leads us to understand how international agreements such as the Paris Agreement affect renewable energy transition.

International commitments are working on responding to climate change and reducing greenhouse emissions [10]. The Kyoto Protocol, adopted in 1997 and entered into force in 2005, marked a significant milestone in global climate governance, exclusively for developed countries, by establishing binding emission-reduction targets. Empirically, Liu et al. [11] show that implementing the Kyoto Protocol had a strong positive impact on renewable energy development for a sample of 29 developed countries from 2000 to 2015.

International agreements combatting climate change are insufficient to control investments and actions toward achieving sustainability goals [12]. Despite its goals, the Kyoto Protocol has several limitations. For example, Badrinarayana [13] notes that major emitters reject binding obligations unless developing countries participate, which, in turn, cannot comply with the Kyoto Protocol’s objective. Compared to the Kyoto Protocol, the Paris Agreement was a new international commitment, adopted in 2015, characterized by Nationally Determined Contributions (NDCs), which is a framework that allows both developed and emerging countries to set targets for reducing carbon emissions. This framework aims to limit global warming to well below 2 degrees Celsius above pre-industrial levels, with efforts to limit the temperature increase to 1.5 degrees Celsius [14]. According to institutional theory [15,16], which suggests that institutions line their behavior within a social system, governments are subject to institutional pressures that influence their decisions and policies. These pressures may include citizens’ expectations and international standards [17]. Organizations tend to conform to the dominant norms of their institutional environment through mimicry or normalization [16]. This leads countries to increase their interest in renewable energy deployment in alignment with global expectations and standard norms for sustainable development.

Theoretically and regarding governance indicators, Belaid et al. [18] document that investment in the renewable energy sector is important to the country’s institutions’ quality. Genus and Mafakheri [12] define institutions as organizational entities, rules, and economics. Gutermuth [19] acknowledges that the legal and institutional framework can be barriers to the transition to renewable energy, as they can be a way to have an efficient and quick transition. Empirically, Uzar [20] examines the effect of institutional quality on renewable energy consumption in 38 countries and finds a positive relationship between institutional quality and renewable energy consumption. Saba and Biyase [21] examine the determinants of renewable electricity development for 35 European countries over the period 2000–2018. Their findings reveal that institutional quality has a positive impact on renewable electricity development. Rahman and Sultana [22] find that institutional effectiveness has significant effects on renewable energy. Notwithstanding the importance of the international community to renewable energy transition, it is important to understand how governance factors influence the effectiveness of the Paris Agreement in promoting renewable energy deployment. In this vein, the interactive governance theory initially developed by Kooiman [23], suggests that governance is a collective effort involving different societal actors closely linked to national and international policies.

This paper examines the determinants of renewable energy consumption and focuses on the relationship between the Paris Agreement and renewable energy deployment, by studying how governance indicators moderate this relationship. This analysis aims to answer empirically the following research questions:

-

How does the Paris Agreement affect renewable energy consumption in both developed and developing countries?

-

What are the determining factors of renewable energy consumption?

Using panel data for developed and emerging countries over the period 2000–2022, this study documents a positive relationship between the Paris Agreement and renewable energy deployment. This finding suggests that after the Paris Agreement, countries are actively promoting renewable energy to align with institutional expectations and enhance their global standing and reputation. Unexpected results show that the ratification of the Paris Agreement for countries with strong governance has not accelerated the energy transition, as observed in countries with lower governance quality. This unexpected result can be attributed to the advanced state of renewable energy deployment in high-quality institutional settings before the agreement’s ratification. We suggest that for those countries, the additional benefits of some investment in renewable energy are relatively small. Accordingly, they are asked to shift their climate strategies, such as participating in the carbon market.

This study contributes to the literature in several significant ways. Theoretically, it shifts attention to the relationship between the Paris Agreement 2015 and renewable energy consumption through institutional theory. Furthermore, this paper extends the works of [18,24,25], initially focused on the effect of institutional indicators on renewable energy transition, by studying the moderating effect of these indicators in the relationship between renewable energy deployment and international agreement. The empirical contribution concentrates on an estimation method known as the dynamic panel threshold model, which we use to confirm our main findings.

The remainder of the paper is organized as follows. Section 2 reviews relevant research concerning the determinants of renewable energy deployment and hypothesis development. Section 3 presents the data and research design. Section 4 discusses the main results. Section 5 presents the additional analysis. Section 6 documents the robustness check, and Section 7 concludes the paper.

2. Literature Review and Hypothesis Development

Despite the presence of abundant literature dealing with the determinants of the deployment of renewable energies [20,21,26,27], few studies have studied the role of ratification of international agreements on the transition to renewable energies [3,28].

2.1. The Effect of the Paris Agreement (2015) on Renewable Energy Deployment

Referring to neo-institutional theory, norms and values cause international pressures to motivate countries to internationalize their strategy. Countries must meet the expectations of all stakeholders to continue to exist [16]. Economies are affected by societal pressure, customs, institutional values, and expectations of their institutional environment, which include social standards and rules [15,16]. According to institutional theory, the signing of public policies, such as the Paris Agreement, can be described as organizations’ actions to adhere to norms shared in organizational fields, rules, and standards to gain institutional legitimacy. Furthermore, institutional pressures lead to homogeneous and mimetic behaviors [15].

As a sign of compliance with other countries, authorities imitate practices and strategies to comply with expectations imposed by the external environment. Economies, therefore, will invest in renewable energy in line with institutional expectations to maintain their position within their environment.

According to neo-institutionalism, not only economic efficiency but also social legitimacy is important to survive in a challenging environment [29]. By adhering to international agreements to combat climate change, economies will commit to transitioning to a less polluting model, particularly by using more renewable energy in line with institutional expectations to maintain their position in the environment [30]. The Kyoto Protocol is a good example of institutional and external pressure on organizations. In line with the protocol’s objectives, countries are asked to adopt their strategies to reduce carbon emissions, which, in turn, has created significant pressure on private and public institutions to achieve these goals [31]. The Kyoto Protocol was adopted on 11 December 1997 and entered into force on 16 February 2005. Based on the norms of Kyoto, developed countries are asked to limit and reduce greenhouse gas (GHG) emissions under agreed individual targets.

Empirical findings document a positive relationship between the Kyoto Protocol and the development of renewable energy. Przychodzen and Przychodzen [28] show that the implication of the Kyoto Protocol led to an increase in renewable energy production for Central and Eastern Europe and the Caucasus and Central Asia. Liu et al. [17] confirm this finding in a sample of 29 countries from 2000 to 2015. Dogan et al. [3] investigate the impact of the Kyoto Protocol on energy transition by applying the econometric method of Sigmund and Frestl [32]. The authors show that the Kyoto Protocol has a positive impact on energy transition, suggesting that the protocol encourages developed countries to improve environmental quality.

The Kyoto Protocol imposed specific reduction targets on developed countries while initially excluding developing countries from binding commitments (Kyoto Protocol, 1997). By contrast, the Paris Agreement of 2015 represents a significant step forward from the Kyoto Protocol. The Paris Agreement recognizes the need for global climate action by committing all developed and developing countries to contribute to emissions reduction. It also introduces greater flexibility by allowing countries to set their emission-reduction targets, called Nationally Determined Contributions (NDCs), which promote an approach that is more responsive to national realities and individual country capacities [11]. In 2015, the United Nations Framework Convention on Climate Change (UNFCCC) organized in Paris, held the 21st Conference of Parties (COP 21) [33]. The Paris Agreement was founded in 2015 and took effect on 4 November 2016. It aims to limit global temperature rise to well below 2 degrees Celsius, with a preference for a more ambitious target of 1.5 degrees Celsius relative to pre-industrial levels.

The existing literature has extensively examined the effect of the Kyoto Protocol on renewable energy. However, the Paris Agreement sets new goals to guide all nations to reduce their global carbon emissions. In this paper, we are particularly interested in the impact of the Paris Agreement and how it influences the energy transition. Based on the preceding discussion, the effect of the Paris Agreement on renewable energy deployment is expected to be positive. Therefore, we formulate the following hypothesis:

Hypothesis 1 (H1).

The Paris Agreement has a positive impact on renewable energy deployment.

2.2. The Effect of Governance Indicators on Renewable Energy Deployment

The quality of institutions plays an essential role in the adoption of strategies allowing a transition to renewable energies [20], given that their use is, above all, a political decision [14]. Indeed, renewable energies need more funds than fossil fuels, and thus, their deployment requires rigid institutional policies [4,34]. This has been confirmed empirically by Uzar [20], who found a positive relationship between institutional quality positively and renewable energy consumption. Fredriksson and Svensson [35] argue that in cases of high levels of political instability and corruption, environmental policies are not the priority of policymakers.

Saba and Biyase [21] found that the control of corruption, the rule of law, voice and accountability, and institutional quality positively impact renewable electricity development. Rahman and Sultana [22] showed that institutional effectiveness significantly affects renewable energy. Belaid et al. [18] support that political stability and governance effectiveness are essential determinants for promoting renewable energy production in nine selected Middle East and North Africa (MENA) countries. Based on the above literature, the second hypothesis is formulated as follows:

Hypothesis 2 (H2).

Governance indicators positively impact renewable energy deployment.

2.3. The Moderating Role of Governance Indicators between the Paris Agreement and Renewable Energy Production

Although the literature review largely focuses on the moderating role of quality governance in financial development and renewable energy development [18,25], the interaction between international agreements, including the 2015 Paris Agreement, and the quality of governance on the impact of the transition to renewable energy remains unexplored.

Using generalized method of moments (GMM) and two-stage least squares (2SLS) estimation methods for 123 countries from 1990–2017, Kassi et al. [25] suggest that improvements in the governance quality in low-income and higher-income countries lead to enhanced marginal impacts of financial development and renewable energy consumption. Furthermore, Belaid et al. [18] examine the determinants of renewable energy production in nine selected MENA countries during the period 1984–2014. The authors find that the effectiveness of governance moderates the relationship between financial development and the deployment of renewable energies positively and negatively, depending on the quantiles of the latter.

The Paris Agreement sets ambitious targets to reduce greenhouse gas emissions and accelerate the transition to renewable energy. The success of the international commitment’s goals depends not only on international efforts but also on the governance framework. Referring to interactive governance theory, initially developed by Kooiman [23], which underlines that governance refers to collective governing activities involving various societal actors in response to public needs and visions, we suggest that this interconnectivity leads us to understand how governance mechanisms play an interaction role between all stakeholders. We can consider that international and national policies are linked to each other, and the decisions taken at one level can have an impact at the other level.

The Paris Agreement sets global targets to limit climate change by encouraging investment in renewable energy. However, the effectiveness of these objectives depends on the countries’ ability to translate international commitment into national action.

For countries with good governance indicators, such as low corruption and high political stability, the implementation of policies resulting from the Paris Agreement is generally more effective. These countries have institutional strategies to attract investment to regulate the market of energy and to facilitate the transition to renewable energy.

By contrast, for countries with weaker governance indicators, where institutions are less developed, transparency is limited, and the control of corruption is weaker, the implementation of the Paris Agreement is more difficult. These countries may have difficulties elaborating effective regulations and ensuring continuity and coherence of climate policies. As a result, the impact of the Paris Agreement on renewable energy projects in these countries is often mitigated, thus slowing down the energy transition.

Under the above discussion, we hypothesize that the effect of the Paris Agreement is more pronounced in the presence of good governance indicators.

Hypothesis 3 (H3).

The positive impact of the Paris Agreement is accentuated in the presence of good governance indicators.

2.4. Other Factors Affecting Renewable Energy Deployment

The main determinants of the deployment of renewable energies identified by the literature are economic, financial, and environmental. Several studies have found that economic development positively affects renewable energy consumption [10,20]. Previous literature suggests that trade openness is a proxy for a country’s level of engagement in renewable energy [36]. However, a few studies have found that trade openness can reduce the deployment of renewable energy. Based on a panel data set of 25 OECD countries, Alam and Murad [37] show that trade openness positively influences renewable energy use over the long term. Omri and Nguyen [38] and Jebli and Youssef [34] find that trade openness increases renewable energy consumption, except for high-income countries. Brini et al. [39] conclude that increased trade promoted renewable energy in Tunisia, a notion corroborated by the findings of Zeren and Akkus [40] for emerging countries, and Vural [41] for selected Latin American countries. Using panel data for 16 Asian countries for the period 1990–2019, Akintande et al. [7] show that trade activities can significantly reduce the consumption of renewable energy. Chen et al. [36] demonstrate that increased trade openness leads to a lower growth rate of renewable energy consumption in less democratic countries.

Banking credits are the main source of external financing for energy investment in most countries [28]. Brunnschweiler [42] finds a positive relationship between commercial banks and renewable energy production. In addition, Ang et al. [43] confirm that the availability of domestic credit to the private sector is an important factor affecting financing decisions in the context of renewable energy power generation for OECD and G20 countries. However, Przychodzen and Przychodzen [28] find no significant relationship between renewable energy production and domestic credit availability.

The deployment of renewable energy requires considerable investment in clean energy technologies. These technologies can be effectively supported by foreign direct investment (FDI) [21,44]. Paramati et al. [45] document that FDI enables hosting countries to easily overcome the shortage of capital for renewable energy generation projects. Using panel data of 21 African countries from 1990 to 2012, Ergun et al. [46] show a positive effect of FDI on renewable energy transition, which Akintande et al. [7] confirm for Asian countries. However, Saba and Biayse [21] show a negative relationship between renewable energy consumption and FDI. The study by Kutan et al. [47] on the impact of FDI on renewable energies in the case of the BRICS economies (Brazil, Russia, India, China, and South Africa) confirms that FDI plays a vital role in promoting renewable energy consumption.

Other important factors also influence the renewable energy transition, such as energy market competitiveness. Painuly [48] indicates that market failures arising from government monopolies in the energy sector have significant barriers to renewable energy production. These findings are corroborated by Marinot [49], who suggests that the implantation of competitive markets and privatization impede renewable energy production. Furthermore, Lin and Omjou [50] suggest that competitiveness within the energy market, measured by the natural resources rents as a percentage of GDP, encourages investors to engage in markets that are not dominated by other actors.

Concerning environmental factors, two opposing views can explain how carbon emissions affect renewable energy. The first view emphasizes a positive relationship between CO₂ emissions per capita and renewable energy deployment. Bamati and Raoofi [10] suggest that an increase in CO₂ emissions per capita contributes to climate change and air pollution, highlighting the need for more sustainable energy options. Consequently, external pressure leads authorities to prioritize the production of renewable energy. An empirical analysis from Omri and Nguyen [38] and Gozgor et al. [6] supports this positive relationship. The second view highlights a negative relationship between CO₂ emissions and renewable energy production. Lucas et al. [51] attribute the negative relationship between CO₂ emissions and renewable energy development to the dependency on fossil fuels and the power of lobbies. Baigorri et al. [52] demonstrate that the promotion of nuclear energy has gained significant momentum in recent years, achieving widespread social acceptance by emphasizing environmental benefits, particularly the reduction of CO₂ emissions. This nuclear lobbying strategy has enabled nuclear energy to effectively compete with renewable energy sources. Geels [53] concludes that the energy transition faces considerable challenges due to opposition from established central station facility owners, particularly those in the nuclear sector. This opposition presents a major obstacle to the shift towards renewable energy sources.

Furthermore, Bayale et al. [9] assert that the negative relationship between CO₂ emissions on renewables may be owing to the complementarity between renewable energy production and CO₂ emissions in smaller quantities. The authors suggest that renewable energies may be perceived as less necessary when CO₂ emissions are relatively low. Empirical findings from Przychodzen and Przychodzen [28] report a negative relationship between CO₂ emissions per capita and renewable energy production in Central and Eastern Europe, the Caucasus, and Central Asia. This is confirmed by Bayale et al. [9] for the West African Economy and Monetary Union (WAEMU).

Forest areas were neglected as an environmental factor affecting renewable energies deployment. We can expect a positive relationship between forest areas and renewable energy consumption based on the wood-derived biomass hypothesis, which states that countries with large forests will exploit wood for production and energy consumption. In the same vein, Laureti et al. [54] document a positive relationship between the value of forest area and the investment in renewable energy production.

More recently, researchers associated the energy transition with energy justice to ensure that the energy transition is equitable and inclusive [55]. A more equitable and sustainable energy future can be achieved through the strategic use of renewable energy and supportive policies to increase access to clean energy, create jobs, and address societal inequalities [56].

3. Research Design

3.1. Sample Selection and Data

We used panel data with annual observations of 24 countries. This panel covers the period 2000–2022. Two data sources are combined to create our sample: the World Bank database and the ICRG database. The final sample comprises 552 observations. The variables are then derived from theoretical understanding and existing literature, considering data availability as presented in

Table 1. Variables of the empirical model.

Variables	Measures	Previous Studies	Data Sources	Notation
Dependant variable
Renewable energy consumption	Renewable energy consumption (% of total energy)	[6,24]	WBD	REC
Exploratory variables
Paris Agreement	Takes 1 since 2016, 0 Otherwise	Author’s measure		Paris
Control variables
CO2 emission	CO2 metric tons per capita	[20]	WBD	CO2
Forest area	Forest area (% of land area)	[54]	WBD	Forest
Trade openness	Exports of goods and services (% of GDP)	[3]	WBD	Open
Domestic credit	Domestic credit to private sector (% of GDP)	[28]	WBD	Credit
Foreign direct investment	Foreign direct investment, net inflows (BoP, current US$)	[7]	WBD	FDI
Natural resource rents	Total natural resources rents (% of GDP)	[50]	WBD	Natural
Moderating variables
Control of corruption	−2.5 (weak) to 2.5 (strong) governance performance	[21]	WGI	Crr
Political stability		[21]	WGI	PS
Governance effectiveness		[21]	WGI	GE
Institutional Quality	Composite index using PCA	Author’s measure

.

3.2. Variables Definitions

The dependent variable, that is, renewable energy production, is represented by renewable energy consumption (share in % of total final energy consumption). The selection of the control variables was motivated by prior literature addressing the determinants of renewable energy deployment (Bayale et al. [9], Tu et al. [24], Dogan et al. [3], and Przychodzen and Przychodzen [28]).

3.3. Descriptive Statistics

Table 2. Descriptive statistics.

	Mean	Std. Dev.	Min	Max
REC	18.75	12.39	1.44	54.99
CO2	6.81	4.65	0.88	20.46
Forest	37.34	16.94	14.02	73.73
Open	34.12	19.86	9.03	119.80
Credit	95.22	49.21	3.74	216.30
FDI	4.06 × ${10}^{10}$	7.19 × ${10}^{10}$	−4.16 × ${10}^9$	5.11 × ${10}^{11}$
Natural	2.94	3.82	0.12	28.32
Crr	0.671	0.97	−1.02	2.45
PS	0.15	0.811	−2.37	1.75
GE	0.80	0.79	−1.03	2.34
Dummy Variable Paris	Proportions
1	0.347
0	0.652

presents a summary of the descriptive statistics of the data. As expected, there is a wide variation in the dependent variable measured by renewable energy consumption (% of total energy), indicating the difference in the use of renewable energy in the sample. This result is consistent with that of Tu et al. [24], Belaid et al. [18], and Dogan et al. [3], who show a high variability across countries used in their sample.

Table 3. Descriptive statistics by countries.

Countries	RE	CO2	Forest	Open	Credit	FDI	Natural	Crr	PS	GS
Canada	21.92	15.83	38.80	33.74	145.07	4.52 × ${10}^{10}$	2.90	1.86	1.05	1.76
France	12.11	5.18	30.00	28.84	94.50	5.10 × ${10}^{10}$	0.04	1.33	0.42	1.45
United States	12.98	16.88	33.65	11.46	187.03	2.82 × ${10}^{10}$	0.90	1.38	1.52	0.28
Australia	10.70	17.04	17.16	20.93	121.06	3.64 × ${10}^{10}$	5.76	1.87	0.97	1.65
Spain	13.68	6.04	36.45	30.03	127.03	3.68 × ${10}^{10}$	0.05	0.98	1.15	0.11
Germany	11.91	9.06	32.67	42.16	91.96	7.85 × ${10}^{10}$	0.12	1.81	0.83	1.54
Belgium	6.80	9.15	22.59	78.18	64.04	4.06 × ${10}^{10}$	0.03	1.44	0.76	1.52
Denmark	24.98	7.68	15.00	52.44	165.13	5.69 × ${10}^9$	1.06	2.31	1.05	2.00
Greece	13.09	7.31	29.71	28.33	85.06	2.69 × ${10}^9$	0.14	0.12	0.20	0.48
Finland	37.68	9.69	73.53	39.22	82.27	8.03 × ${10}^9$	0.42	2.26	1.32	2.06
Mexico	10.49	3.78	34.39	30.89	24.58	2.92 × ${10}^{10}$	4.10	−0.56	−0.58	0.03
Costa-Rica	37.36	1.47	57.23	36.34	44.83	2.21 × ${10}^9$	1.55	0.52	0.68	0.26
Italy	12.64	6.44	30.60	27.79	78.72	2.11 × ${10}^{10}$	0.10	0.32	0.44	0.49
Japan	5.55	9.06	68.40	15.28	171.86	1.82 × ${10}^{10}$	0.02	1.34	1.03	1.47
Colombia	30.17	1.51	54.74	16.60	36.13	9.71 × ${10}^9$	5.64	−0.32	−1.45	−0.10
India	38.72	1.31	23.52	19.91	45.57	2.98 × ${10}^{10}$	3.18	−0.39	−1.07	0.01
Malaysia	3.81	6.83	58.63	88.39	116.72	8.24 × ${10}^{10}$	9.39	0.18	0.21	0.99
Turkey	14.03	4.17	27.57	25.68	44.10	1.12 × ${10}^{10}$	0.48	−0.15	−1.07	0.11
Thailand	21.50	3.48	38.68	65.28	126.24	7.59 × ${10}^9$	2.27	−0.39	−0.73	0.23
China	16.32	5.94	21.46	24.75	135.72	1.77 × ${10}^{11}$	3.40	−0.36	−0.45	0.18
South Africa	10.03	7.40	14.32	27.34	118.88	6.02 × ${10}^9$	5.49	0.09	−0.18	0.26
Chile	29.12	4.08	22.90	34.42	98.25	1.34 × ${10}^{10}$	9.27	1.27	0.51	0.98
Brazil	46.05	1.98	61.71	13.72	51.01	5.37 × ${10}^{10}$	3.83	−0.19	−0.20	−0.21
Ecuador	15.45	2.14	52.46	26.24	29.29	6.52 × ${10}^{10}$	10.70	−0.67	−0.49	−0.61

Note: This table shows the descriptive statistics. There are 24 countries from 2000–2022.

presents the sample composition and the mean of the characteristics of each country. The variables show strong variations, confirming the different indicators of the environment, economy, and quality in the sample.

Table 4. Pairwise correlations.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
REC (1)	1
Paris (2)	0.12 *	1
CO2 (3)	−0.48 *	−0.10 *	1
Forest (4)	0.29 *	0.01	−0.18 *	1
Open (5)	−0.16 *	0.02	−0.03	−0.01	1
Credit (6)	−0.32 *	0.13 *	0.60 *	−0.15 *	0.04	1
FDI (7)	−0.18	0.07	0.38	−0.15 *	−0.23 *	0.39 *	1
Natural (8)	0.03	−0.12 *	−0.17 *	0.05	0.12 *	−0.17 *	−0.127 *	1
Corr (9)	0.07 *	−0.06	0.56 *	−0.06	0.12 *	0.48 *	0.08 *	−0.30	1
GE (10)	0.25 *	−0.06	0.72 *	−0.07	0.23 *	0.56 *	0.15	−0.03 *	0.94 *	1
PS (11)	0.15 *	−0.05	0.58 *	0.07	0.18 *	0.39	0.01	0.26 *	0.82 *	0.81 *	1
VIF		1.77	1.05	1.10	1.10	1.77	1.33	1.08

This table presents the pairwise correlations. The sample comprises 552 observations from 2010 to 2022. * Indicates statistical significance at the 5% levels, respectively.

reports the Pairwise correlation matrix of the variables used in our empirical regression. As expected, a positive relationship exists between the dependent and independent variables, supporting (H1). In addition, there is a negative and significant correlation between CO₂ emission and REC. It is also shown that Forest area is positively correlated with REC. Only two control variables are not correlated with REC, which are FDI and Natural. Gujarati [57] asserts that a coefficient of correlation that is less than 0.8 may not be subject to a serious multi-collinearity problem. The high correlation exists between GE and Cor with a coefficient of 0.94. Given the high degree of collinearity between the governance indicators, we estimated each variable in a separate empirical model.

To further our analysis of multicollinearity, the variation inflation factor (VIF) values are below 10 [58], and the mean of VIF is 1.31, which is below 2 [59]. In summary, the VIF test and correlation analysis confirmed the absence of multicollinearity problems in our sample.

4. Econometric Specification and Empirical Analysis

4.1. Econometric Specifications

To address endogeneity and omitted variable concerns, we use the GMM estimator proposed by Arellano and Bover [60] and developed by Blundell and Bond [61]. The regression takes the following form:

$$\begin{array}{c}{REC}_{it}={\alpha }_1{REC}_{it-1}+{\beta }_0+{\beta }_1{Paris}_{i,t}+{\beta }_2{CO2}_{i,t}+{\beta }_3{forest}_{i,t}+{\beta }_4{Open}_{i,t}+{\beta }_5{Credit}_{i,t}+{\beta }_6{FDI}_{i,t}\hfill \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}+{\beta }_7{Natural}_{i,t}+{\mu }_t+{\delta }_i+{\epsilon }_{i,t}\hfill \end{array}$$

(1)

where ${REC}_{it}$ is the dependent variable representing the consumption of renewable energy (% of total energy); the subscript i represents each country, t represents the year, ${REC}_{it-1}$ is the lagged dependent variable, and ${Paris}_{i,t}$ is dummy variable takes 1 after 2016, 0 otherwise. ${CO2}_{i, t}, {forest}_{i,t}, {Open}_{i,t}$, ${Credit}_{i,t}{, FDI}_{i,t}$, and ${Natural}_{i,t}$ are described in Table 1. ${\beta }_0$ is the intercept term; ${\beta }_1$, ${\beta }_2$, ${\beta }_3{, \beta }_4{, \beta }_{5 }{{, \beta }_6, and \beta }_7$ are the coefficient vectors of the estimated parameters; ${\alpha }_1$ is the first-order autoregressive coefficient of the dependent variable; ${\mu }_t$ is an unobserved year specific effect; ${\delta }_i \mathrm{i}\mathrm{s} \mathrm{a}\mathrm{n} \mathrm{u}\mathrm{n}\mathrm{o}\mathrm{b}\mathrm{s}\mathrm{e}\mathrm{r}\mathrm{v}\mathrm{e}\mathrm{d} \mathrm{c}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{y} \mathrm{s}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c} \mathrm{e}\mathrm{f}\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{t}; \mathrm{a}\mathrm{n}\mathrm{d} {\epsilon }_{i,t}$ is the error term.

To examine the moderating effect of governance indicators, we introduce an interaction term for the association between governance indicators and the Paris Agreement. The second regression takes the following form:

$$\begin{array}{c}{REC}_{it}={\alpha }_1{REC}_{it-1}+{\beta }_0+{\beta }_1{Paris}_{i,t}+{\beta }_2{CO2}_{i,t}+{\beta }_3{forest}_{i,t} +{\beta }_4{Governance indicator}_{i,t}\hfill \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}+{\beta }_5{Governance indicator}_{i,t}*{Paris}_{i,t}+{\beta }_6{Open}_{i,t}+{\beta }_7{Credit}_{i,t}+{\beta }_{8}FDI+{\beta }_9{Natural}_{i,t}+{\mu }_t+{\delta }_i \hfill \\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}+{\epsilon }_{i,t}\hfill \end{array}$$

(2)

where governance indicator (The WGI framework includes six broad dimensions of governance indicators known as voice and accountability, regulatory quality, rule of law, control of corruption, political stability, and governance effectiveness. In this paper, we used only three dimensions) presents political stability or control of corruption or governance effectiveness, i represents each country, and t is the time period defined by [62]. These indicators estimate governance indicators (ranging from −2.5 (weak) to 2.5 (strong) governance performance) collected from ICRG. ${Governance indicator}_{i,t}\times {Paris}_{i,t}$ is the interaction between the Paris Agreement and governance for country i at year t.

Using the principal component analysis, we create an index composite based on the three indicators that represent institution quality (IQ). The application of the exploratory factor analysis on the IQ variable shows a value of KMO equal to 0.724. This value is greater than 0.5, which is acceptable and a significant Bartlett test of sphericity (p = 0.000 ≤ 0.05). Pallant [63] reports that a Cronbach’s alpha (Cα) value above 0.6 is considered a highly reliable and acceptable index; in our case, this index is equal to 0.879. Therefore, all these tests confirm the factorization principle.

To estimate the effect of the Paris Agreement on renewable energy, ${\beta }_1$ and ${\beta }_5$ cannot be interpreted independently, as ${\beta }_1$ captures the effect of the Paris Agreement on renewable energy when the quality of the institution is zero. To estimate the total effect of the Paris Agreement, a partial derivative of Equation (2) is calculated as ${\displaystyle \frac{\partial REC}{\partial Paris}}={\beta }_1+{\beta }_{5}Governance indicator$.

As the marginal effect depends on the level of institutional quality, the marginal effects at the average, minimum, and maximum levels of institutional quality are estimated. The results are reported at the bottom of

Table 5. Dynamic panel data regression results.

	Model 1 (1)	Model 2 (2)	Model 3 (3)	Model 4 (4)	Model 5 (5)
Paris	0.38 *** (0.002)	0.38 *** (0.002)	1.09 *** (0.000)	0.90 ** (0.03)	0.41 *** (0.000)
PS		1.13 *** (0.000)
Paris × PS		−0.48 *** (0.003)
GE			1.66 *** (0.000)
Paris × GE			−0.91 *** (0.005)
Crr				1.84 *** (0.000)
Paris × Corruption				−0.59 *** (0.000)
IQ					1.08 *** (0.000)
IQ × Paris					−0.32 *** (0.000)
CO2	−0.23 *** (0.001)	−0.44 *** (0.000)	−0.47 *** (0.000)	−0.55 *** (0.000)	−0.60 *** (0.000)
Forest	0.05 *** (0.000)	0.05 *** (0.000)	0.06 *** (0.000)	0.05 *** (0.000)	0.06 * (0.100)
Open	0.05 *** (0.000)	0.03 *** (0.001)	0.02 * (0.08)	0.02 ** (0.041)	0.01 *** (0.000)
Credit	0.007 ** (0.018)	0.009 *** (0.003)	0.002 (0.466)	0.000 (0.768)	0.003 (0.245)
FDI	1.53 × ${10}^{13}$ (0.925)	5.00 × ${10}^{13}$ (0.755)	8.29 × ${10}^{13}$ (0.606)	1.10 × ${10}^{12}$ (0.486)	1.04 × ${10}^{12}$ (0.510)
Natural	0.02 (0.192)	0.01 (0.526)	0.02 (0.265)	0.03 (0.107)	0.04 ** (0.042)
Constant	2.332 *** (0.002)	0.061 *** (0.000)	0.747 *** (0.000)	1.572 * (0.082)	2.471 *** (0.008)
Marginal effect
Min	0.38	1.517	2.027	1.501	1.374
Mean	0.38	0.308	0.362	0.504	0.41
Max	0.38	−0.46	−0.39	−0.545	−0.594
Year effect	Yes	Yes	Yes	Yes	Yes
Country effect	Yes	Yes	Yes	Yes	Yes
Hansen test p-value	1.00	1.00	1.02	1.03	1.00
AR (2) test p-value	0.52	0.42	0.48	0.57	0.51

Notes: This table presents the regression estimated by GMM regression for the sample of 24 countries over 2000–2022. Column 1 presents the linear dynamic panel data. Columns 2, 3, and 4 present results regarding the moderating role of political stability (column 2), governance indicator (column 3), corruption (column 4), and on the relationship between the Paris Agreement and renewable energy deployment. Column 5 presents results regarding institutional quality measured by principal component analysis. The p-values appear in parentheses below the estimated coefficients. ***, **, * refer to the 1%, 5% and 10% significance levels, respectively.

. In this sense, the ratification of the Paris Agreement could improve the consumption of renewable energy below an institutional quality threshold equal to −${\beta }_1$/${\beta }_5$ if ${\beta }_5$ < 0. Since ${\displaystyle \frac{\partial REC}{\partial Paris}}$> 0 ⇒ ${\beta }_1+{\beta }_{5}Governance indicator>0$ ⇒ $Governance indicator$ < −${\beta }_1$/${\beta }_5$.

To examine the validity of the orthogonality assumption system GMM, the Hansen test of overidentification and the Arllano and Bond tests for second-order and higher-order several correlation AR (2) tests are employed, given that the system GMM technique relies on an internal instrument.

4.2. Main Results

The results of GMM (Table 5, column 1) show a positive and significant effect at the 1% threshold between the Paris Agreement and renewable energy deployment, supporting hypothesis H1. Based on this result, we suggest that by signing the Paris Agreement, countries feel more responsible to achieve the international agreement’s goals. Consistent with institutional theory, they are more likely to promote renewable energy to meet the citizens’ expectations and maintain a good reputation. These results are in line with the findings of Liu et al. [11], Przychodzen and Przychodzen [28], and Dogan et al. [21]. These authors show that the Kyoto Protocol positively affects renewable energy development.

Regarding institutional quality, governance indicators positively affect renewable energy deployment, consistent with H2. Columns 2–4 of Table 5 show that a 1% increase in PS, GE, and CRR leads to an increase in REC by 1.13%, 1.66%, and 1.84%, respectively. We suggest that political stability and the control of corruption stimulate renewable energy deployment. This supports the findings of Saba and Biyas [21], who suggest that political stability is a determinant of renewable energy deployment. These findings also corroborate those of Alsaleh and Abdul-Rahim [64] for 28 European Union (EU) countries.

Furthermore, using principal component analysis reveals that the institutional quality index (Table 5, column 5), which is constructed by the three governance indicators, statistically and positively affects renewable energy deployment. This can be explained by the fact that institutional activities function sufficiently toward renewable energy use. Table 5, columns 2, 3, 4, and 5 show that PS, GE, and CRR negatively moderate the relationship between the Paris Agreement and renewable energy deployment, which leads to the rejection of hypothesis H3. In other words, the effect of the Paris Agreement on renewable energy deployment is less pronounced in the presence of good governance indicators. There could be two possible explanations. The first could be related to a shift to another venue of climate management. We suggest that these countries are likely to shift their focus to other climate policy strategies, such as carbon pricing systems and nuclear power investments. For example, The European Union Emissions Trading System (EU ETS) is the world’s largest carbon market, covering multiple sectors across EU member states [65]. Digitemie and Ekemezie [65] suggest that the revenue generated from carbon taxes can be used to fund climate mitigation and adaptation efforts or to provide rebates to low-income households. Empirically, Ortas and Álvarez [66] test the effectiveness of the EU’s environmental management policies. The authors find that its carbon emissions trading scheme reduced carbon emissions and ensured that the EU complied with the targets of the Kyoto Protocol. Other scholars have also confirmed the effectiveness of carbon trading in developed countries [67]. Another climate policy option is to invest in nuclear power, as Guglyuvatyy [68] suggested for Australia to meet the carbon emissions-reduction targets outlined in the Paris Agreement. According to Venizelou and Poillikkas [69], the role of nuclear energy in the hydrogen economy has become increasingly interesting, especially as innovations in the field have enabled continued cost control. Consequently, countries with established nuclear infrastructures have favored pink hydrogen over green hydrogen, potentially slowing down investments in renewable energy. This necessitates substantial investment and a greater reliance on nuclear power. This pursuit of diversification can be explained by the high level of renewable energy deployment in developed countries, making marginal investment in this sector less profitable.

The second explanation could be related to the pressure confronted by governments during times of crisis, leading them to prioritize short-term objectives instead of long-term investments such as in the renewable energy sector. For example, the COVID-19 crisis exacerbated the opportunistic tendencies of governments [70], creating circumstances favorable to strategic opportunism, in which states pursue objectives that might be unpalatable to domestic actors or the international community [15]. To cope with social and economic pressures, authorities can shift their focus to immediate social support, rather than investing in renewable energy.

The agency theory of Meckling and Jensen [71] suggests that agents (governments) may adopt opportunistic behaviors that are not aligned with the principals (citizens). We suggest that countries with good governance exhibit opportunistic behavior. By taking advantage of their reputation for good governance, they can focus on other projects judged more important to them rather than projects on energy transition. This strategy allows them to change their attention from renewable energies to other sectors while maintaining a positive reputation, aligning with Goffman’s impression management theory [72]. Thus, despite the commitments made, these countries can maintain their good reputations by signing the Paris Agreement but exploit their robust governance systems to prioritize projects that are not necessarily aligned with their objectives. The combination of two theories, agency theory and impression management theory, leads us to explain how countries can use their reputation to appear committed to international agreements while focusing on different priorities.

These results suggest that either these countries have turned to other ways of climate management beyond the deployment of renewable energy, or that their climate-management efforts have stalled due to the COVID-19 crisis, and that they adopt opportunistic behavior to cover their urgent needs. We repeat the regressions, restricting the analysis to the pre-COVID-19 period, but the results remain largely unchanged. This finding refutes the second explanation.

To further deepen the analysis, we calculate the marginal effect of the Paris Agreement on renewable energy consumption, which depends on the level of institutional quality. For example, Model 2 in Table 5 shows that the marginal effect for political stability is equal to 0.79 (−${\beta }_1$/${\beta }_5)$. Thus, beyond this threshold, the positive effect of institutional quality ceases to exist and becomes negligible, affirming that the deployment of renewable energy increases as the level of institutional quality rises to the indicated threshold. At this threshold, the Paris Agreement will have a negative or neutral impact. However, this calculation does not allow us to determine the significance; for this reason, we proceed with the estimation using the threshold model.

Considering greenhouse gas emissions, we find a negative relationship between CO₂ emissions and renewable energy deployment. This finding is consistent with those reported by da Silva et al. [73] but not with those obtained by Uzar [20] or Bamati and Raofi [10], who reveal a positive relationship between these two indicators. Increasing CO₂ emissions leads to diminishing renewable energy deployment by 0.23%. This can be explained by the dependency on fossil fuels and the power of lobbies. This suggests that the financial and political interests related to the fossil fuel industry may harm the development of renewable energies. Unlike CO₂ emissions, forest areas have a positive impact on renewable energy production. In numerical terms, a 1% rise in forest areas increases renewable energy by 17.8%. This is confirmed by Laureti et al. [54], who found a positive relationship between these two indicators.

Focusing on control variables, Table 5 shows a significant relationship between trade openness, domestic credit, and renewable energy deployment. This is consistent with the findings of Alam and Murad [37] for OCDE countries and Akintande et al. [7] for the African context but contradicts the findings of Belaid et al. [18] for G7 countries. By contrast, FDI and natural resource rents are not significant with renewable energy consumption in all specifications.

5. Additional Analysis

Dynamic Panel Threshold Regression Model

To examine the potential nonlinear impact of the set of explanatory variables on renewable energy consumption, we can extend Equation (1) to a sample split form, where the group is determined by the value of the threshold variable. The dynamic panel threshold regression model (DPTRM) can be expressed as follows:

$$\begin{array}{cc}\hfill {REC}_{it}={\alpha }_1{REC}_{it-1}& {+\beta }_0+{\beta }_{1b}{Paris}_{i,t}(q_{i,t} < {\phi }_0)+{\beta }_{1a}{Paris}_{i,t} (q_{i,t} \ge {\phi }_0)+{\beta }_2{CO2}_{i,t}+{\beta }_3{forest}_{i,t}+\hfill \\ & {\beta }_4{Open}_{i,t}+{\beta }_5{Credit}_{i,t}+{\beta }_{6}FDI+{\beta }_7{Natural}_{i,t}+{\mu }_t+{\delta }_i+{\epsilon }_{i,t}\hfill \end{array}$$

(3)

where $q_{i,t}$ is the threshold variable, which is the quality of governance indicator in this study, and ${\phi }_0$ is the threshold level.

The threshold panel regression results (

Table 6. Dynamic panel threshold regression results with governance indicators as a threshold variable.

	Model 1 (1)	Model 2 (2)	Model 3 (3)	Model 4 (4)
	Threshold PS	Threshold GE	Threshold Corr	Threshold IQ
Below threshold	0.59 *** (0.000)	2.00 *** (0.000)	0.689 *** (0.000)	0.921 *** (0.000)
Above threshold	−0.315 (0.211)	−0.02 (0.863)	−1.66 (0.189)	−0.44 ** (0.021)
CO2	−0.64 *** (0.000)	−0.64 *** (0.000)	−0.66 *** (0.000)	−0.752 *** (0.000)
Forest	0.191 ** (0.029)	0.255 *** (0.000)	0.186 ** (0.025)	0.259 ** (0.030)
Open	0.05 ** (0.02)	0.04 *** (0.010)	0.04 *** (0.000)	0.04 ** (0.040)
Credit	0.006 (0.106)	0.002 (0.474)	0.002 (0.466)	0.002 (0.489)
FDI	−3.19 × ${10}^{-13}$ (0.878)	4.91 × ${10}^{13}$ (0.813)	−3.19 × ${10}^{-13}$ (0.892)	−1.74 × ${10}^{13}$ (0.933)
Natural	−0.015 (0.566)	0.04 (0.870)	0.006 (0.823)	0.00 (0.769)
Constant	0.957 * (0.100)	2.31 ** (0.050)	0.172 (0.231)	2.00 * (0.09)
Country effect	Yes	Yes	Yes	Yes
Years effect	Yes	Yes	Yes	Yes

Note: This table provides estimations of the dynamic panel threshold regression model (DPTRM) ***, **, and * denote significance at the 10%, 5%, and 1% levels, respectively. The lag length is one, and p-values are provided in brackets.

, columns 1, 2, and 3) show that below-threshold governance indicators positively impact the relationship between the Paris Agreement and renewable energy production. When governance indicators are below a certain threshold, the Paris Agreement positively influences renewable energy deployment. This suggests that in countries with weaker governance, the Paris Agreement plays a crucial role in driving renewable energy initiatives, confirming the main findings.

However, when governance indicators are above the threshold, the Paris Agreement does not significantly impact renewable energy consumption, suggesting that these countries have shifted their focus toward other avenues of climate management beyond the deployment of renewable energy. These results highlight the importance of the quality of governance in the effectiveness of international agreements, such as the Paris Agreement. By grouping countries according to governance indicators, we find that the impact of the Paris Agreement on renewable energy projects varies according to the quality of governance. In countries with weaker governance, international policies can play a crucial role in supporting and promoting renewable energy projects. By contrast, in countries with stronger governance, the Paris Agreement has no significant effect on renewable energy. The control variables kept the same sign in all specifications.

6. Robustness Check

To check the robustness of the main findings, we assess the impact of economic wealth on renewable energy deployment as well as the moderating role of GDP per capita on the relationship between the Paris Agreement and REC. This effort is motivated by the fact that economic development is generally associated with the presence of high-quality institutions, which is confirmed by the strong correlation between the two variables in the sample. We confirm the findings using the dynamic panel threshold regression model.

As shown in column 1 of

Table 7. The impact of economic wealth on renewable energy deployment: GMM and Threshold GMM regression.

	Model 1 (1)	Model 2 (2)	Model 3 (3)
	GMM	GMM	Threshold GMM
Paris	0.45 *** (0.000)	1.23 *** (0.000)	Below threshold 0.91 *** (0.000) Above Threshold −0.424 (0.234)
GDP per capita	1.53 × ${10}^{10}$ *** (0.000)	1.4 × ${10}^{12}$ *** (0.000)
GDP × Paris		−1.13 × ${10}^{-10}$ *** (0.000)
CO2	−0.28 *** (0.000)	−0.32 *** (0.000)	−0.69 *** (0.000)
Forest	0.04 *** (0.01)	0.04 *** (0.00)	0.268 ** (0.02)
Open	0.015 ** (0.04)	0.02 ** (0.04)	0.05 *** (0.00)
Credit	0.006 (0.112)	0.002 (0.265)	0.00 (0.207)
FDI	−5.18 × ${10}^{-13}$ (0.747)	−2.60 × ${10}^{-13}$ (0.873)	−9.46 × ${10}^{-14}$ (0.964)
Natural	0.003 (0.888)	0.01 (0.599)	0.006 (0.727)
Constant	0.528 (0.522)	0.02 (0.972)	3.37 (0.368)
Country effect	Yes	Yes	Yes
Years effect	Yes	Yes	Yes
Hansen test p-value	1.01	1.00
AR (2) p-value	0.74	0.69

Notes: This table reports the results of GMM regression (Column 1), the moderating role of GDP per capita on the relationship between the Paris Agreement and REC using GMM (Column 2), and dynamic panel threshold regression model (DPTRM) when GDP per capita is used as a threshold. The p-values appear in parentheses below the estimated coefficients. ***, **, refer to the 1%,and 5%, significance levels, respectively.

, the coefficient of GDP per capita is positively associated with renewable energy deployment. This suggests that economic wealth can provide funds for new projects related to renewable energy [4]. A higher GDP facilitates the allocation of resources toward regulatory costs aimed at promoting renewable energy sources [10]. These results are in line with the findings of Alam and Murad [37] and Gozger et al. [6] for the OCDE context. Furthermore, the results (Table 7, column 2) show that GDP per capita negatively moderates the relationship between the Paris Agreement and REC. In other words, the effect of the Paris Agreement is less pronounced for countries with a higher GDP per capita. Lastly, in terms of empirical results (Table 7, column 3) above the threshold of GDP, the Paris Agreement does not significantly impact renewable energy consumption. In all specifications, we confirm the main findings. We conclude that developed countries are generally well-advanced in the deployment of renewable energy prior to the signing of the Paris Agreement. Therefore, the agreement does not significantly affect existing renewable energy policies, as any additional investment in the sector would have limited marginal gain.

7. Conclusions

Despite a large stream of literature that focuses on the determinants of renewable energy deployment, little evidence has been provided regarding the impact of international agreement and institutional quality. This paper aims to extend the literature of the determinants of renewable energy deployment by examining the impact of the Paris Agreement and governance indicators on renewable energy deployment in selected developed and developing countries using annual data over the period 2000–2022. In addition, the study aims to investigate the influence of governance quality on the relationship between the Paris Agreement on renewable energy deployment. To do this, we use GMM and dynamic panel threshold regression models. We demonstrate that the Paris Agreement positively affects renewable energy consumption. In fact, the renewable energy deployment production has increased by 38% after the Paris Agreement.

Based on institutional theory [15,16], we suggest that to align with institutional pressure and social norms, as well as to preserve their positive standing in the global community, these countries prioritize promoting renewable energy sources.

Regarding institutional quality, we find that political stability, governance effectiveness, and control of corruption improve renewable energy deployment. This leads to an increase in REC by 1.13%, 1.66%, and 1.84%, respectively. However, focusing on the moderating role of governance indicators on the relationship between the Paris Agreement and REC, we reveal unexpected results showing that the impact of the Paris Agreement on renewable energy consumption is less pronounced for countries with high governance quality. This can be explained by the fact that countries are likely to shift their focus to other climate policy options, such as carbon pricing systems, nuclear power investment, and energetic efficiency. Furthermore, we find strong evidence that forest area, trade openness, and domestic credit positively affect REC, while CO₂ emissions hurt REC.

This paper presents managerial and environmental implications. Firstly, in developing countries, the quality of institutions remains important for improving renewable energy. Therefore, efforts should be made to enhance governance quality. Second, future agreements should set thresholds by action category: renewable energy, energy efficiency, etc., with periodic monitoring.

A further research avenue will be to investigate how geopolitical stability moderates the relationship between the Paris Agreement and renewable energy deployment for developed and developing countries. Another avenue of research is to examine the portfolio diversification of renewable energy for developed and emerging countries before and after the Paris Agreement using difference-in-difference estimation.