Geopolitical Risk and Energy Transition in Russia: Evidence from ARDL Bounds Testing Method

Abstract

One of the current debatable global problems is climate change or global warming as crucial geopolitical risks. The progress of energy transition by considering geopolitical risk has not been considered seriously yet. This paper contributes to the literature by modeling and analyzing energy transition patterns in Russia with emphasis on geopolitical risks factor as a giant fossil fuels producer using the ARDL bounds testing method over the period of 1993–2018. The main results proved long-run negative impact of economic growth, population growth and inflation rate on energy transition of Russia, while CO₂ emissions, geopolitical risk, exchange rate and financial openness have positive impacts on energy transition movement in the country. Furthermore, we found out that in the short-run, the relationship between energy transition improvement and economic growth, CO₂ emissions, population growth and inflation rate is negative, while geopolitical risk, exchange rate and financial openness are the only variables which accelerate energy transition in the country. As major concluding remarks, Russia’s policy makers should draw attention to the long-run energy plans in the country. Furthermore, lowering dependency of Federals’ budget to the oil and gas revenues would be a useful policy to reduce negative impact of economic growth on energy transition movement in the country. Another recommendation is to determine rapid decarbonizing policies in the country.

1. Introduction

The fact of climate change or global warming, which has increased since the industrial revolution, has significantly affected geopolitical risk potential of countries leading to more interest in using green energy sources. [1] proved that green energy sources bring many advantages over non-renewable energy for international security and peace. Particularly, the green energy sources may not make the resource curse in energy producer countries [2] and lower regional and global geopolitical competition over domination in non-renewable energy markets. Hence, according to [3,4], countries are going to move to change their energy basket to combat climate change and lower geopolitical risks.

Therefore, it can be seen that the global energy sector is going through change of paradigm. The value of energy projects has always been assessed by economic factors and its effective use. However, due to increasing awareness of danger of greenhouse emissions to the environment, the environmental factor began to be taken into account as a geopolitical risk, in the evaluation of energy projects, such as power plants, oil and gas projects, etc. It is an interesting and meaningful development in the energy industry that market participants are paying attention to the level of CO₂ emission of energy projects. Moreover, the share of renewable energy at electricity sector and energy mix in many countries has been growing globally. Development of technologies and state policies supporting the use of renewables and clear energy sources are helping renewables become more competitive. Decreasing the emission of CO₂, which accelerates climate change, is increasingly becoming an essential part of state policies. It is also called decarbonization. Decarbonization, more broadly, can be described as energy transition.

Energy transition is a structural change in the energy sector, moving away from usages of fossils fuel to energy sources which emit no or less greenhouse gases. In other way, it is also often described as sustainable development. At its heart is the need to reduce energy-related CO₂ emissions to limit climate change [5]. While sustainable solutions are envisioned for the future, many societies are still under high-carbon and high-pollution energy regime borne by fossil fuels.

Energy transitions are a multidimensional, complex, non-linear, non-deterministic, and uncertain phenomenon and, therefore, they are difficult to characterize using a small number of features. They require a transformation of actors and their conduct, of markets, and a change in the existing regulations and policies [6,7]. Historically, a new energy source or technology has displaced another because it can produce either cheaper services or services with superior attributes, such as being cleaner, easier or more flexible. Previous energy transitions have been the result of the development of a better technology or the emergence of a new source of energy with superior technological attributes, so the world shifted from biomass to coal, from coal to oil, and from oil to natural gas [8].

Energy transition is described as decarbonization and is a tool to lower geopolitical risk which can affect each national economy in a different way as every country is not in the same situation. While everyone agrees that reducing greenhouse gas emissions is important for the environment, it is inevitable that countries that heavily depend on imports of fossil fuels are more motivated in decreasing CO₂ emission levels than economies where share of natural resources is high. Countries which are not producers of hydrocarbon resources are also more inclined to take part in developing renewable resources and technologies in order to reduce dependence on energy imports, and have shares in this developing industry. Russia is one of the biggest hydrocarbon producers along with being the largest exporter of energy resources. In this paper, energy transition in Russia will be considered, as well as how various factors affect the share of renewable energies in the Russian economy.

This research contributes to the literature on geopolitical risk, energy transition and energy security. It is deviated from earlier literature by introducing a new theoretical energy transition pattern and using geopolitical risk index in energy transition modeling.

The rest of paper is organized as follows: Section 2 discusses energy transition and geopolitical risk in Russia. Next, a brief literature review is represented. Theoretical model of our research is explained in Section 4. Section 5 contains data description and empirical model specification. The next section discusses empirical findings and finally, the paper concludes in Section 7.

2. Energy Transition, Geopolitical Risk and Energy Security in Russia

According to BP Statistical Review of World Energy 2019, Russia is the biggest exporter of natural gas, second biggest exporter of crude oil and third biggest exporter of coal in the global market in 2018. During the last decades, Russia has played an important role in the global non-renewable energy markets by having significant shares of global energy production.

Table 1. Russia’s energy production (% of world production), 2000–2018.

Energy Sources	2000	2005	2010	2015	2018
Oil	8.83%	11.73%	12.46%	12.02%	12.07%
Natural gas	22.35%	21.39%	18.99%	16.69%	17.30%
Coal	5.54%	4.91%	4.31%	4.68%	5.50%

Source: Authors’ compilation from [9].

represents the share of crude oil, natural gas and coal production of Russia in the world during 2000–2018. As shown in Table 1, Russia has contributed to approximately 12%, 17% and 5% of global oil, natural gas and coal productions, respectively, which show the giant place of this country in the world’s energy markets.

Globally rising renewables targets and the transition towards a decarbonization paradigm are regarded in Russia as a significant threat to hydrocarbon export revenues, and thus to Russian economic security. National economy is dependent on fossil fuel production. In 2017, hydro carbons provided 25% of GDP and 39% of the country’s federal budget revenues [10]. The share of thermal power generations in electricity generation in the world is 74% in 2018 and that of renewables including hydro is 26% (Figure 1). In Russia, the share of thermal power generation in 2018 is 83% and the rest is renewables including 16% of hydro (Figure 2). The abundance of cheaply available natural and hydro resources is well reflected in the electricity generation mix in the country.

Russian economy has expanded more than 3-fold from $516 billion in 1990 to $1.658 trillion in 2018 (Figure 3). The climate agenda, technological progress and the availability of new technology solutions are able to dramatically increase the efficiency of the energy sector and transform its traditional way of functioning, the desire among all countries to ensure the competitiveness of their national economies and boost their development of affordable energy, and last but not least, the need to increase energy security [10].

Russia’s export value in 2018 reached $449 billion. Among that, oil and oil product exports were $256 billion. Russia exported oil worth $207 billion, gas worth $49 billion and coal $17 billion in 2018. These three fossil fuel products account for over 50% of Russian total exports (Figure 4). Export revenue from oil is about four times larger than that of gas. Naturally, oil sales revenue contributes several times more for the national economy.

As a major exporter of natural resources, Russia’s energy security is closely related to the stability of demand and diversification of buyers. Due to geographical proximity from Western Siberia, the country’s main region of hydro-carbon production, Russia’s primary market always has been European consumers. The majority of gas is exported to European consumers and oil is in a similar situation but to less of an extent. After the deteriorating political relations with the Western countries, Russia accelerated diversification of its market and strengthen its ties with Asian countries. It is called “Pivot to the East”. China has emerged as an important partner for cooperation and Russia’s main target market for demand diversification. It led to the conclusion of a long-term gas supply agreement between the two countries, the first pipeline gas export project to Asia, Power of Siberia. In 2018, Russia exported 247 BCM of gas and almost 80% of gas was exported to the European market through pipeline. 12% went to CIS countries and 10% by LNG (Liquefied Natural Gas) (Figure 5). For Russia, it is important to find consumers other than European countries, and China and LNG will play an important role for this purpose.

Regarding, Russian oil exports, the situation is a little different. China is the number 1 importer of Russian oil with 27% of oil export revenue of the country. At the same time, Korea and Japan are ranked 5th and 7th with 6% and 2% share each. Even though European countries are the largest buyers of Russian oil, its share is smaller than that of gas (Figure 6). Oil is easier to diversify than gas, thanks to its convenience of transportation and storage.

3. Literature Review

The issues of global energy transition and environmental issues related to CO₂ emission has been actively discussed globally. For instance, [12,13,14,15,16] have discussed the importance of energy transition to have a better future regarding human health and environment.

However, energy transition in Russia is not a widely discussed issue. [10] provided an overview of Russian energy policy in the context of the global energy transition and the authors predicted that Russia, ranking fourth in the world in primary energy consumption and carbon dioxide emissions, adheres to the strategy of “business as usual” and relies on fossil fuels. [17] discussed future energy trade in geopolitics of energy transition between Russia and the European Union. The major concluding remark is that focusing on less state-centric cooperation and involving more non-state actors can accelerate the progress of energy transition in energy trade between the EU and Russia. [18] argued that the main driver of Russia’s renewable energy policy is to achieve the economic benefits related to the manufacturing of green equipment, and the focus on industrial development rather than the decarbonization of the power sector clearly appears from the decision of the Russian government to tie renewable energy subsidies to stringent local content requirements, in particular, solar energy benefits from a subsidy regime that is favorable to local manufacturers. [19] described foresight for the energy priority in Russia. He argued that one of the main energy priorities of Russia is the transition to an environmentally friendly energy industry. In addition, Russian renewable energy industry is actively growing from a very low base and requires substantial R&D investments. [20] explored the relationships between carbon emissions and their main determinants such as energy consumption, real income, international trade, level of education and level of urbanization in the Russian Federation, employing data for the period 1991–2016. [21] investigated the relationship between economic development and environmental pollution among Russian regions based on the concept of Environmental Kuznets Curve. [22] attempted to explore the nexus between oil consumption, economic growth and carbon dioxide (CO₂) emissions in three East Asian oil importing countries (i.e., China, South Korea and Japan) over the period 1980–2013, by using the Granger causality, Johansen cointegration test, Generalised Impulse Response functions (GIRF) and variance decompositions. Testing the EKC hypothesis, some studies explain changes in environmental quality purely by economic growth [23,24,25,26,27]. However, recently more studies have begun to identify additional factors that may play important roles in environmental degradation processes. For example, energy consumption is considered a key determinant of environmental quality [28,29,30,31,32,33,34]. Numerous studies agree that energy consumption is mostly responsible for CO₂ emissions [35] and directly or indirectly plays the principal role in environmental problems [36].

4. Theoretical Background

A variety of commodities are produced daily in the world, and both renewables and non-renewables as energy sources are employed as a major production input in the process of commodities production, if we assume that an economy comprises only two sectors of industry and household (residential sector), who are the main commodities consumers. In other words, demand of these two sectors for different commodities shapes demand for energy resources.

We start with the industry’s energy demand. Equation (1) represents the production function of industry, assumed to be in the form of Cobb-Douglas:

$$Y_t^I=F\left(K_t,L_t,E{T}_t^I\right)=K_t^{\alpha }{L}_t^{\beta }{\left(E{T}_t^I\right)}^{\left(1-\alpha -\beta \right)}$$

(1)

Here, $Y^I$ is the total output of industry, K is the capital input, L denotes the labor input, $E{T}^I$ represents energy inputs of industry production which are considered as energy transition (share of renewables to non-renewables).

By considering constant return to scale, α is the elasticity of production of capital, β is the elasticity of production of labor, and the elasticity of production of energy resources is equal to 1 − α − β.

In the sector, firms are maximizing their profit as follows:

$$Max{\pi }_t=P_t^Y{Y}_t^I-r_t{K}_t-w_t{L}_t-e_t\left(P_t^E+T_t\right)E{T}_t^I$$

(2)

where $\pi$ is the sector’s profit, $P^Y$ is the price of the final products of industry, $r$ denotes the interest rate of capital, $w$ denotes the wage rate, $e$ denotes the exchange rate, $P^E$ denotes energy price and $T$ denotes the transportation environmental costs (we can use CO₂ emissions as a proxy for this variable). We are assuming that the country of study (Russia) is importing renewable energy technologies (e.g., solar photovoltaic module, Li-Ion cells, wind power generators, etc.) for the energy transition. If the distance between exporter of renewable energy technologies and importer (Russia) is far, this will have more carbon footprint. This is the reason that we used CO₂ emission as a proxy of the environmental transportation costs. Exchange rate is also matter as the imported goods are denominated in non-Ruble currencies (e.g., US$). In the supply chain of renewable energy technologies such as solar modules, each country is not the sole producer of all the components and needs to import components from different countries; hence, the exchange rate matter [37].

Equation (3) shows the first order condition of profit with respect to $E^I:$

$$\frac{\partial {\pi }_t}{\partial E{T}_t^I}=\left(1-\alpha -\beta \right)\frac{P_t^Y{Y}_t^I}{E{T}_t^I}-e_t\left(P_t^E+T_t\right)=0$$

(3)

The energy transition demand is represented in Equation (4):

$$E{T}_t^I=\left(1-\alpha -\beta \right)\frac{P_t^Y{Y}_t^I}{e_t\left(P_t^E+T_t\right)}$$

(4)

As shown, industry’s energy transition demand is a function of the elasticities of production of labor and capital, the real output of industry sector, the price of energy, the exchange rate, and the transportation environmental cost (we can consider CO₂ emissions as a proxy for this variable).

Next, we can consider energy demand of household with the following utility function:

$$U_t=\left(C_t,E{T}_t^H\right)=\frac{1}{1-\gamma }{\left(C_t\right)}^{1-\gamma }+\frac{1}{1-\delta }{\left(E{T}_t^H\right)}^{1-\delta }$$

(5)

where $U$ and $C$ denote utility and consumption (non-energy commodities) of households at time t, respectively. $E{T}_t^H$ represents energy transition demand of households at time t.

Households are maximizing their utility with respect to their budget, which is the constraint, as shown in Equation (6):

$$S.t.P_t^C{C}_t+e_t\left(P_t^E+T_t\right)E{T}_t^H=Y_t^H$$

(6)

where $P^C$ denotes price of non-energy goods, $P^E$ denotes the energy prices denominated in US$, and $T$ denotes the transportation environmental costs (we considered CO₂ emissions as a proxy for this variable). $Y^H$ is total income of the households. It should be mentioned that in the above utility function, We consider population as a proxy of $C$. When the population growth rate increases the consumption of goods (energy and non-energy) increase. This is the reason that we included population in our empirical model as a control variable.

In order to maximize the utility function of households for defining the factors that determine energy demand, we develop the Lagrange function, as in Equation (7):

$$\mathsf{\Gamma }=U\left(C_t,E{T}_t^H\right)-\lambda \left\{P_t^C{C}_t+e_t\left(P_t^E+T_t\right)E{T}_t^H-Y_t^H\right\}$$

(7)

Obtaining the first-order conditions with respect to the $E{T}^H$, $C$, and $\lambda$ results in Equations (8)–(10):

$$\frac{\partial \mathsf{\Gamma }}{\partial E{T}_t^H}={\left(E{T}_t^H\right)}^{-\delta }-\lambda \left\{e_t\left(P_t^E+T_t\right)\right\}=0\to E{T}_t^H=f\left(e_t\left(P_t^E+T_t\right),Y_t^H\right)$$

(8)

$$\frac{\partial \mathsf{\Gamma }}{\partial C_t}=C_t{}^{-\gamma }-\lambda \left\{P_t^C\right\}=0$$

(9)

$$\frac{\partial \mathsf{\Gamma }}{\partial \lambda }=P_t^C{C}_t+e_t\left(P_t^E+T_t\right)-Y_t^H=0$$

(10)

As shown in Equation (8), a household’s energy transition demand is a function of exchange rate, price of energy, transportation environmental costs (CO₂ emissions), and the income level of the households. The total energy transition demand is equal to the combined energy demand of households and industry (Equation (11)).

$$E{T}_t=E{T}_t^I+E{T}_t^H$$

(11)

Therefore, the total energy demand is a function of different factors, as shown in Equation (12):

$$E{T}_t=f(P_t^E,T_t,e_t,Y_t)$$

(12)

where $P^E$ is the energy price which can be represented by inflation rate, and $T$ denotes the CO₂ emissions as a proxy for transportation environmental costs, $e$ is the exchange rate, and $Y_t$ is the total gross domestic product of the economy, which depends on the income level of households ($Y^H$) and the total output of the industry ($Y^I$).

Therefore, we will employ inflation rate, CO₂ emissions, exchange rate, economic growth from theoretical background and geopolitical risk, population growth and financial openness as three control variables. It should be mentioned that since our variables are based on our theoretical model, the quality control of them may be approved.

5. Data Description and Model Specification

Econometrically, the general function of empirical model is written as following:

$$E{T}_t=f\left(IN{F}_t,CO{2}_t,EX{C}_t,GR{O}_t,PO{P}_t,FI{N}_t,GE{O}_t\right)$$

(13)

To attain coefficients in the form of elasticities, we transform all the variables into logarithms and our empirical model changes as follows:

$$\begin{array}{cc}\hfill LE{T}_t& ={\beta }_0+{\beta }_1\ast LIN{F}_t+{\beta }_2\ast LCO{2}_t+{\beta }_3\ast LEX{C}_t+{\beta }_4\ast LGR{O}_t\hfill \\ & +{\beta }_5\ast LPO{P}_t+{\beta }_6\ast LFI{N}_t+{\beta }_7\ast LGE{O}_t+{\mu }_i\hfill \end{array}$$

(14)

Here, ET_t denotes energy transition of Russia in time t. INF_t and CO2_t show inflation rate and CO₂ emissions in Russia in time t. EXC_t represents exchange rate in Russia in time t, while GRO_t indicates economic growth of Russia in time t. Finally, POP_t, FIN_t and GEO_t are population growth, financial openness and geopolitical risk in Russia, while μ_t is the error term assumed to be normally distributed with zero mean and constant variance.

Before conducting estimation of Equation (14), we have to test the stationarity characters of the variables. Due to the probability of presence of structural breaks, we conducted [38] test which proposed three following models to check the hypothesis of one-time structural break in the series:

$$\Delta x_t=a+a{x}_{t-1}+bt+cD{U}_t+{\displaystyle {\displaystyle \sum }_{j=1}^k}{d}_j\Delta x_{t-j}+{\mu }_t$$

(15)

$$\Delta x_t=b+b{x}_{t-1}+ct+bD{T}_t+{\displaystyle {\displaystyle \sum }_{j=1}^k}{d}_j\Delta x_{t-j}+{\mu }_t$$

(16)

$$\Delta x_t=c+c{x}_{t-1}+ct+dD{U}_t+bD{T}_t+{\displaystyle {\displaystyle \sum }_{j=1}^k}{d}_j\Delta x_{t-j}+{\mu }_t$$

(17)

Here, dummy variable is represented by DUt indicating mean shift occurred at each point with time break while trend shift variables are shown by DTt.

Next step after checking the stationarity properties is to performing an estimation which is done by the ARDL bounds testing approach introduced by [39] to find out long-run relationship between energy transition and independent variables. In addition, the reliability of estimations will be tested by ARCH LM test for higher order autocorrelation, White test for homoscedasticity and Ramsay RESET test for misspecification of model.

Most of the data were extracted from the BP statistical review of 2019 and World Bank database. Additionally, the data for Geopolitical Risk Index (GPR) are gathered from Matteo Iacoviello website. [40] introduced the Geopolitical Risk Index (GPR) for countries based on automated text-research results of the digital archives.

The primary descriptive statistics of data for Russia are represented in

Table 2. Descriptive statistics of data, 1993–2018.

Variables	Unit	Mean	Maximum	Minimum	Std. Dev.
Economic Growth	%	1.73	10	−12.56	5.71
Exchange rate	LCU per US$	30.90	67.05	0.99	16.84
Energy transition	%	3.91	4.37	3.55	0.21
CO2 emissions	Metric tons per capita	11.40	13.10	10.10	0.78
Population growth	%	−0.11	0.21	−0.46	0.22
Inflation rate	%	67.65	874.24	2.87	174.55
Financial openness	LCU per US$	0.52	0.69	0.41	2.14
Geopolitical Risk Index (GPR)	-	105.92	220.07	47.67	27.23

Source: Authors’ compilation from [11].

.

As shown in Table 2, economic growth of Russia has a mean of 1.73% over 1993–2018. Moreover, it takes a maximum of 10% (in 2000) and a minimum of −12.56% (in 1994). Moreover, as its national currency value against US dollars, it has a mean of 30.90 LCU per US$, whereas its maximum and minimum values happened in 2016 (67.05 LCU per US$) and 1993 (0.99 LCU per US$), respectively. In regards to energy transition, it has an average of 3.91% and its maximum and minimum happened in 1993 (4.37%) and 2011 (3.55%), respectively. In addition, CO₂ emissions in Russia have an average of 11.40 metric tons per capita (According to World Bank databases, in this period, the global average of CO₂ emissions is nearly 4.41%). In regards to population growth, Russia has a negative mean of −0.11% over 1993–2018, while the average of financial openness in this country is only 0.52. The Geopolitical Risk Index (GPI) of Russia has an average of 105.92 between 1993 and 2018. Comparing with other countries like India, Brazil, Indonesia with GPI < 100, Russia has a quite high GPI which highlights the importance of applying different strategies like energy transition to lower this index.

To check out the existence of primary relationship between dependent (energy transition) and explanatory series (economic growth, exchange rate, CO₂ emissions, population growth, inflation rate, geopolitical risk and financial openness) in Russia, the Pearson correlation test between variables is conducted. The results of this test are represented in

Table 3. Pearson correlation results.

Dependent Variable		Independent Variables
	Economic Growth	Exchange Rate	CO2 Emissions	Population Growth	Inflation Rate	Financial Openness	Geopolitical Risk
Energy transition	−0.02	0.19	0.03	−0.15	−0.29	0.18	0.23

Source: Authors’ compilation from SPSS 19.

as follows:

According to Table 2, there is a negative relationship between energy transition and economic growth, population growth and inflation rate in Russia, while CO₂ emissions, geopolitical risk, exchange rate and financial openness are variables indicating positive correlation with energy transition in this country over the period 1993–2018. Due to the nature of the Russian economy as an oil-based economy [41], its GDP and economic growth significantly depend on fossil fuel production and exports. Hence, it is expected that this increase in economic growth decelerates the energy transition movement in Russia. In addition, we expect that the exchange rate has a positive coefficient, meaning that by any depreciation of Russian Ruble against US dollars, cost of improving environmentally friendly energy projects in this country raises leading to a lowering of renewable energy consumption and development in Russia. Moreover, the positive impact of CO₂ emissions on the energy transition movement in Russia is expected. This positive impact has been confirmed by numerous studies such as [42] in cases of 17 OECD countries and [43] in cases of 85 developed and developing economies. In regards to population growth, we may expect the adverse impact of it on energy transition in Russia. Any growth in population may lead to a higher aggregate demand (e.g., in energy sector, transportation, and commodities market) which needs a higher level of fossil fuel consumption in Russia. In addition, it is expected that the inflation rate has a negative impact on energy transition progress in Russia. Any increase in general level of prices may lead to cost increase of renewable energy projects in Russia which can be considered as a major obstacle for energy transition movement in the country. Finally, positive impact of financial openness on energy transition is expected. A higher financial openness means easier ways of financing renewable energy projects in Russia by international institutions or companies. In regards to geopolitical risk, by increasing the speed of energy transition, Russia can lower the threat of natural curse [2] and lower the speed of climate change (according to [9] website, Arctic sea ice is now decreasing at a rate of 12.85% per decade which is an important geopolitical risk for Russia).

Table 4. Expected signs of variables’ coefficients.

Variables	Expected Sign
Economic growth	Negative (−)
CO2 emissions	Positive (+)
Inflation rate	Negative (−)
Population growth	Negative (−)
Financial openness	Positive (+)
Exchange rate	Positive (+)
Geopolitical risk	Positive (+)

Source: Authors’ compilation.

represents the expected signs of coefficients of our independent variables:

6. Results

To find out the relationship between independent variables and energy transition in Russia, the ARDL bounds testing method was conducted. One of the main pre-requirements of this method is that all the series should be integrated at I(0) or I(1) or I(0)/I(1). To check this pre-requirement, the Zivot-Andrews structural break test was applied and

Table 5. Zivot-Andrews structural break test results.

Variable	At Level		At 1st Difference
	T-Statistic	Time Break	T-Statistic	Time Break
Economic Growth	−3.583	1993	−10.443 *	2014
Exchange rate	−4.281	1997	−8.288 *	1997
Energy transition	−3.791	2009	−9.829 *	2009
CO2 emissions	−3.476	1997	−10.101 *	1993
Population growth	−5.010	2014	−9.029 *	2009
Inflation rate	−3.582	2009	−11.593 *	1993
Financial openness	−4.118	2009	−8.391 *	1997
Geopolitical risk	−3.665	2009	−10.292 *	2009

Note: * shows significance of variable at 5% level. Source: Authors’ compilation.

reported its results as follows:

The results, shown in Table 5, revealed that all the variables are integrated at 1st level or I(0). Therefore, we conducted the ARDL bounds testing method to investigate the relationship between energy transition, economic growth, exchange rate, CO₂ emissions, population growth, inflation rate and financial openness in Russia in the presence of structural break in the series over the period of 1993–2018. To run the ARDL bounds testing estimation, we needed to select an appropriate lag length (to compute F-statistics for comparing with critical bounds) which was done based on the minimum value of AIC (Akaike Information Criterion).

Table 6. ARDL bounds cointegration testing results.

Estimated Models	Optimal Lag Length	Structural Break	F-Stats.
FGro (Gro|ET, EX, CO, POP, Inf, FOP, GEO)	5,5,5,5,5,6,6	1993	3.593 **
FET (ET|Gro, EX, CO, POP, Inf, FOP, GEO)	5,5,5,5,5,5,5	2009	3.616 **
FEX (EX|Gro, ET, CO, POP, Inf, FOP, GEO)	6,5,5,5,6,6,5	1997	4.728 *
FCO (CO|Gro, EX, ET, POP, Inf, FOP, GEO)	5,5,5,6,6,6,5	1997	4.639 *
FPOP (POP|Gro, EX, CO, ET, Inf, FOP, GEO)	5,6,6,6,5,5,5	2014	2.322
FInf (Inf|Gro, EX, CO, POP, ET, FOP, GEO)	5,5,5,5,5,5,6	2009	1.932
FFOP(FOP|Gro, EX, CO, POP, Inf, ET, GEO)	5,5,5,5,5,5,5	2009	1.550
FGEO(GEO|Gro, EX, CO, POP, Inf, ET, FOP)
Significant Level	Lower Bounds I(0)	Upper Bounds I(1)
1% level	2.841.83	3.972.87
5% level	2.32	3.38
10% level	1.83	2.87

Note 1: Gro, ET, EX, CO, POP, Inf, FOP and GEO represent economic growth, energy transition, exchange rate, CO₂ emissions, population growth, inflation rate, financial openness and geopolitical risk, respectively. Note 2: * and ** indicate significant at 1% and 5% at levels, respectively. Source: Authors’ compilation.

represented the findings of the ARDL bounds testing method for our model with consideration of structural break in the series. The findings express that the computed F-statistics are greater than upper critical bounds at 5% and 1% levels, so we used energy transition, economic growth, exchange rate, CO₂ emissions, population growth, inflation rate and financial openness as predicted variables. Furthermore, the structural breaks stem in the variables of energy transition, economic growth, exchange rate, CO₂ emissions, population growth, inflation rate and financial openness in 2009, 1993, 1997, 1997, 2014, 2009, and 2009 respectively. The results proved that the series are cointegrated for a long-run linkage between energy transition, economic growth, exchange rate, CO₂ emissions, population growth, inflation rate and financial openness in the case of Russia.

Considering the presence of long-run relationship between variables, we investigated marginal effects of economic growth, exchange rate, CO₂ emissions, population growth, inflation rate, geopolitical risk and financial openness on energy transition. The estimated results are reported in

Table 7. Long-run and short-run relationship analysis.

Dependent Variable	Analysis	Independent Variables	Coefficient	p-Value
Energy transition	Long-run	Economic growth	−0.289	0.00
		Exchange rate	0.192	0.04
		CO2 emissions	0.332	0.00
		Population growth	−0.140	0.00
		Inflation rate	−0.319	0.01
		Financial openness	0.451	0.00
		Geopolitical risk	0.278	0.02
Energy transition	Short-run	Economic growth	−0.023	0.00
		Exchange rate	0.008	0.00
		CO2 emissions	−0.119	0.05
		Population growth	−0.053	0.00
		Inflation rate	−0.152	0.02
		Financial openness	0.255	0.01
		Geopolitical risk	0.313	0.00
Short-run Diagnostic Tests
Test	F-stats		p-value
Chi-2 Arch Arch LM test for higher order autocorrelation	2.381		0.13
Chi-2 White For homoskedasticity	1.492		0.13
Chi-2 Ramsay For misspecification of model	1.792		0.12

Source: Authors’ compilation.

.

According to the estimated coefficients, represented in Table 7, we can highlight the following findings:

(i)

Short-run analysis

• Russian economic growth negatively influences on energy transition of the country in the short-run. A 1% increase in economic growth of the country leads to energy transition reduction by nearly 0.02%, meaning that improvement of national production in Russia cannot play a useful role to replace fossil fuel energy sources with green energy ones.
• The short-run relationship between exchange rate and energy transition is found to be positive and statistically significant indicating that a 1% appreciation in Russian Ruble is linked with a 0.008% increase in energy transition. Appreciation in Ruble slightly hurts the export flows of this country, particularly exports of oil and gas. Therefore, the local SMEs of Russia may find suitable opportunity to develop their projects in related to expansion of green energy productions.
• The results confirm that CO₂ emissions have negative short-run contribution to energy transition in Russia. A 1% increase in this variable leads to decrease of energy transition in Russia by nearly 0.11%.
• The short-run impacts of population growth and inflation rate on energy transition movement in Russia are negative and a 1 percent increase in them is linked to a 0.05% and 0.15% reduction in energy transition, respectively.
• Coefficient of financial openness is positive and statistically significant denoting that in the short-run, a 1% increase in financial openness degree in Russia may lead to an increase of energy transition of the country by approximately 0.25%.
• The short-run impact of geopolitical risk on energy transition movement in Russia is positive and statistically significant. A 1% increase in geopolitical risk may lead to nearly 0.31% increase in energy transition of Russia in the short-run.

(ii)

Long-run analysis

• Russian economic growth has negative and statistically significant long-run impacts on the energy transition process of the country. The estimation inferred that a 1% increase in economic growth is linked with a 0.28 decrease in energy transition of Russia. The main reason may be the oil-based economic structure of this country which links economic growth and non-renewable energy resources. This result is in line with [44] who proved a negative long-run relationship between clean energy and economic growth for Nigeria, while it is in contrast with [45] who revealed the positive linkage between economic growth and renewable energy consumption in Pakistan.
• The long-run relationship between exchange rate and energy transition in Russia is found to be positive, meaning that 1% appreciation of Russian Ruble against U.S. Dollars leads to increase of the energy transition process in the country by nearly 0.19%. This finding is in line with [46] who studied future scenarios of renewable transitions.
• Our empirical estimation showed that any increase in CO₂ emissions has negative and statistically significant impact on energy transition in Russia. A 1% increase in CO₂ emissions leads to increase of energy transition by approximately 0.3%. In other words, Russian policy makers consider substitution of fossil fuel consumption with renewable ones (green energy resources) as a solution for air pollution. This finding is similar to [42] who found out a negative relationship between CO₂ emissions and green energy consumption.
• Both population growth and inflation rate have a long-run negative and statistically significant impact on the energy transition process in the Russia. A 1% rise in population and price level of commodities and services in Russia is linked with a 0.14% and 0.31% decrease in energy transition. The result of positive linkage between population growth and non-renewable energy consumption is in line with [47] who proved this relationship in OECD countries.
• The coefficient of financial openness is positive indicating that an increase in level of financial openness of Russia contributes to rise of energy transition. The estimation confirmed that financial openness is a main long-run contributor to improvement of energy transition in the case of Russia.
• Finally, the estimation proves that the long-run effect of geopolitical risk on energy transition is positive and statistically significant. A 1% rise in geopolitical risk may lead to approximately 0.27% increase in energy transition of Russia in the long-run period. This finding is in line with [1] who found out that green energies may cause more small-scale conflicts but decrease the risk of large political conflicts.

7. Conclusions

This paper attempted to investigate the effects of various variables, i.e., economic growth, exchange rate, CO₂ emissions, population growth, inflation rate, financial openness and particularly geopolitical risk on energy transition (consumption of renewable energy resources to consumption of non-renewable ones) in the case of the Russian economy over the period of 1993–2018. To this end, we conducted the ARDL bounds testing method to cointegration to check cointegration among the series in the presence of structural breaks for long-run linkage.

Our findings of ARDL bounds testing approach proved that our variables are cointegrated for long-run linkage. The empirical estimations revealed long-run negative impact of economic growth, population growth and inflation rate on energy transition of Russia, while CO₂ emissions, geopolitical risk, exchange rate and financial openness have positive impacts on energy transition movement in the country. Furthermore, we found out that in the short-run, the relationship between energy transition improvement and economic growth, CO₂ emissions, population growth and inflation rate is negative, while exchange rate, geopolitical risk and financial openness are the only variables which accelerate energy transition in the country.

Moreover, the findings revealed that financial openness is the major contributor to energy transition process in the long-run, while geopolitical risk is the most important player to short-run energy transition process in the case of the Russian economy. Meaning that in the long-run, the most important issue in Russia to imply energy transition is financial flow running from banks and financial institutions (either local or foreign ones) to firms to conduct projects in relation to develop environmentally friendly energy sources, whereas different geopolitical risks such as climate change, resource curse and seeking of the EU to lowering energy dependency on Russia are the golden key of Russian intentions to increase the speed of the energy transition process.

As the major concluding remarks, Russia’s policy makers should draw attention to the long-run energy plans in the country (because the empirical results showed the more magnitudes of effects of variables on energy transition of Russia in the long-run rather than in short-run). Furthermore, lowering dependency of Federals’ budget to the oil and gas revenues would be a useful policy to reduce negative impact of economic growth on energy transition movement in the country. Another recommendation is to determine rapid decarbonizing policies in the country. Since we have found short-run negative and long-run positive effects of CO₂ emissions on energy transition, it is suggested to make a short-run climate change policy besides the long-run one. In addition, based on positive linkage between geopolitical risk and energy transition progress in Russia, it may be concluded that the current Russia-West tension growing since the 2014 Ukraine tension has been creating opportunities for this country to concentrate more on renewable energy sources than non-renewable energy ones that have strong ties with this political tension.

Finally, we can draw Figure 7 in related to the Russian energy transition model:

Overall, it is recommended for future studies to compare the energy transition model of Russia with other nations. Furthermore, using new control variables such as interest rate would make a new insight for scholars. In addition, this study suffered from some limitations such as missing data, only two sectors in the theoretical model (a multi-sector theoretical model is preferable) and a missing qualitative economic breakpoint which may be solved in future studies. Moreover, using a novel ARDL technique such as Dynamic ARDL is suggested to carry out future studies.