Electricity Consumption, Renewable Energy Production, and Current Account of Organisation for Economic Co-Operation and Development Countries: Implications for Sustainability

Abstract

This paper uses the bootstrapped Granger Causality Testing approach to investigate the relationship between electricity consumption, renewable energy production, and the current account of the six OECD countries. One of the main advantages of using this approach is that it captures the cross-section dependence in our sample and applies the Seemingly Unrelated Regression (SUR) to examine the causality relationship between the variables. The empirical findings show the presence of cross-section dependence in our sample as the six Organisation for Economic Co-operation and Development (OECD) countries share resources, capabilities, and key competencies. Notably, a unidirectional causality exists running from electric power consumption to the current account of the USA. The current account balance causes electric power consumption in the case of France and Switzerland. The tri-variate causality relationship between electricity consumption, renewable energy production, and current account balance could not be established in the case of Germany, Finland, or the UK.

1. Introduction

One of the most recent developments in energy innovations is integrating renewable energy generation into electricity production [1,2,3]. Completely substituting non-renewable sources of energy generation with renewable sources has become a challenge not only for the OECD countries but for all economies around the world. Electricity consumption in Switzerland, Germany, France, the United Kingdom, the USA, and Finland is rising. It is the responsibility of governments and the private sector to invest in innovative methods of renewable energy generation so that this rise in demand for electricity can be adequately satisfied by the electricity providers [4]. Generally, renewable energy accounts for hydro, solar, wind, tide, geothermal, and wave sources for energy generation [5]. Electricity providers can use such natural sources for renewable energy generation.

Additionally, the OECD countries are also using biofuels to meet the increase in demand for electricity, and living organisms provide an essential source of biofuel. In addition to this, municipal waste also contributes to renewable sources of energy generation. Municipal authorities are responsible for collecting and disposing of waste materials at a central location for energy generation [5,6]. There are many reasons why the OECD countries are pushing for renewable energy generation for electricity production. First, natural sources of energy generation are the most cost-effective sources for electricity generation in the long term. For instance, the continuous use of hydro, solar, wind, tide, geothermal, and wave energy for electricity generation will not lead to the depletion of energy sources [7,8]. Second, renewable sources of energy generation are a gift from nature to humanity, and these sources of energy generation do not have negative repercussions for the natural environment. For instance, using oil for electricity generation will increase greenhouse gas emissions, adversely affecting the health of the environment and the welfare of people [7,8,9].

Although there are numerous advantages to using renewable sources of energy generation, countries are finding it difficult to switch from non-renewable sources of energy generation to renewable sources. One of the main reasons for this is that the initial cost associated with financing this switch is extremely high as new infrastructure has to be set up, and human resources will have to be trained and equipped with new knowledge about using renewable sources of energy generation [10,11,12,13,14,15]. According to Banos et al. [16], the rapid advancement of information computer technology will help us to design models that can help effectively mitigate the unreliability of renewable energy sources of electricity generation. Importing and exporting electricity from neighbouring countries is common in the OECD countries. Countries such as France, Switzerland, and Germany can easily import and export electricity from their neighbouring countries. This implies that electricity consumption and renewable energy production contribute to the current account balance. The OECD countries are working towards maintaining a sustainable current account balance as a healthy current account balance helps to achieve a better standard of living and maintain optimal economic growth in the short and long run [17].

Against this backdrop, this paper makes three main contributions to the existing energy literature: (i) This study investigates the relationship between electricity consumption, renewable energy production, and the current account balance of the six OECD countries. (ii) Traditional cointegration and causality tests assume that the estimation models automatically account for cross-section dependence in the sample. However, this is not necessarily true, as the cross-section dependence test has to be undertaken to determine the presence of cross-section dependence in the sample. Once cross-section dependence is confirmed, the bootstrapped Granger Causality Test can be applied to determine the causality relationship between the variables. The bootstrapped Granger Causality Test is superior to the traditional Granger Causality Test and cointegration models as it is based on cross-section SURs rather than ordinary least square (OLS) regression. A close examination of the existing literature shows that limited studies have investigated the multivariate causality between electricity consumption, renewable energy production, and current account balance. Most studies have explored the relationship between electricity consumption and economic growth or renewable energy production and economic growth. For instance, Destek and Aslan’s [18] study based on 17 emerging countries used a bootstrapped causality test to examine the relationship between renewable energy consumption and economic growth. Their empirical findings confirm that renewable energy consumption has caused economic growth in South Korea, Greece, and Peru. Yıldırım et al. [19]’s study based on energy consumption and economic growth of 11 countries used the bootstrapped autoregressive metric causality approach to examine the nexus between energy consumption and economic growth. They also confirmed unidirectional causality from energy consumption to economic growth for the Turkish economy. The findings from this study confirm that unidirectional causality exists between electric power consumption and the current account of the USA. The causality also runs from the current account balance to the electric power consumption of France and Switzerland. There was no tri-variate causality relationship in the case of Germany, Finland, and the UK.

The rest of the present study is organised as follows: Section 2 provides an overview of electricity consumption, renewable energy production, and the current account balance of the six OECD countries. Section 3 reviews the literature. Section 4 provides an overview of data collection. Section 5 outlines the econometrics model. Section 6 presents and discusses the research findings and Section 7 provides conclusion and policy implications.

2. Overview of Electricity Consumption, Renewable Energy Production, and Current Account Balance

2.1. Electricity Consumption

It is intriguing to notice the rapid evolution of electricity consumption in Finland, France, Germany, Switzerland, the United Kingdom, and the USA. The growth in the electricity consumption of these six OECD countries can be attributed to several factors. These factors are industrial evolution, growth in population size, and a rapid increase in the number of businesses operating in OECD countries [17]. An increase in the demand for electricity implies an increase in the consumption of electricity [17].

Furthermore, most OECD countries are using a combination of renewable and non-renewable energy sources to generate electricity. The electricity generation in the United Kingdom is mainly from gas, followed by coal, nuclear, and renewables. Gas generates half of the total demand for electricity in the UK, while coal fire generates around one-third of the electricity [20]. Similarly, in the USA, around 67% of electricity is generated by fossil fuels [21]. Contrary to the practice in the USA and UK, out of the total electricity demand in Finland, 80% is produced domestically while 20% is imported from Nordic countries. In 2015, 45% of the total electricity used in Finland was produced using renewable sources [22]. Like Finland, around 20% of the electricity used in France is generated by renewable sources [23]. According to the Federal Ministry of Economic Affairs and Energy [24], as of the year 2016, renewable energy accounts for 31.5% of the gross electricity generation in Germany. The case of Switzerland is quite different from the UK, the USA, Finland, France, and Germany, as around 60% of their energy is produced from renewable energy. Figure 1 shows the electricity consumption in Finland, France, Germany, Switzerland, the United Kingdom, and the USA.

This shows that the electricity consumption of Finland increased by 714.02% from the year 1960 to 2014, followed by a 374.43% increase for France, a 343.39% increase for Germany, a 220.86% increase for the USA, a 144.51% increase for Switzerland, and a 112.65% increase for the United Kingdom [25].

2.2. Renewable Energy Production

Globally, developed and developing economies are focusing on finding renewable means of energy generation. There are different sources of renewable energy, and some of these sources include biomass, hydropower, geothermal, wind, and solar [21]. Using renewable sources to generate energy has become a global development issue rather than a national issue. Statistics show that the use of renewable sources to generate energy is sharply increasing. Three resources have become common for generating renewable energy. These sources are biofuels, solar, and wind [21]. There are several advantages of using renewable sources to generate energy, and these advantages attract many countries to switch from non-renewable to renewable sources of energy generation. One of the main advantages of renewable energy is that it significantly reduces greenhouse gas emissions. In the last decade, greenhouse gas emissions have been responsible for the effects of climate change faced by many small island developing countries [21]. The climate change effects triggered by greenhouse gas emissions have resulted in a rise in seawater that has engulfed houses and agricultural land in many small island developing countries.

Moreover, some disadvantages of renewable energy generation are the lack of availability and reliability of renewable energy sources and the high costs of generating renewable energy [21]. For instance, strong cloud cover, droughts, and weak wind velocity reduce the reliability of renewable energy. Figure 2 shows renewable energy production by the six OECD countries.

Figure 2 shows that renewable energy production in the United Kingdom increased by 6164.65% from the year 1960 to 2017, followed by a 3976.67% increase for Germany, a 580.55% increase for France, a 264.05% increase for the USA, a 154.99% increase for Switzerland, and a 112.62% increase for Finland.

2.3. Current Account Balance

The current account of the six OECD countries is a vital indicator of the macroeconomic performance of each of the countries. The role of the current account is to record the import and export of goods and services in an economy. It is called the current account because one records the trade in goods and services for the current period [26]. It is essential to have a positive current account balance. A positive current account indicates that a country lends to other countries worldwide. In contrast, a negative current account indicates that the country borrows from other countries worldwide [27]. To achieve a sustainable balance on the current account, policymakers need to strengthen trade policies and foreign exchange reserves and reduce inflation rate and exchange rate volatility [28]. The current account deficit of the USA recorded for the current period is becoming a growing concern as high levels of current account deficit negatively affect investor confidence in the economy [29]. Similarly, the UK’s current account deficit is also becoming a growing concern as its imports are increasing faster than its exports [30].

Moreover, Finland and France’s account situation is deteriorating, while Germany and Switzerland have been recording account surpluses. In 2018, Germany’s current account surplus was USD 289,186.70 (million), and Switzerland’s was USD 74,032.99 (million). The current account deficit of the US in the year 2018 was USD 490,978 (million), followed by the UK at USD 123,318 (million); France’s, USD 17,864.50 (million); and Finland’s US 3800.20 (million). Figure 3 shows the current account of the six OECD countries.

Figure 3 shows that the current account balance of the six OECD countries has worsened since the 1970s. A rapid increase in imports, global economic shocks, financial crises, exchange rate volatility, and a lack of research and development for new product development are some of the factors that have deteriorated the current accounts of these six OECD countries.

3. Literature Review

3.1. Electricity Consumption–Renewable Energy Production Nexuses

The research on electricity consumption and renewable energy production has received much attention from academics and practitioners. Interestingly, the existing studies have yet to examine the causality between electricity consumption and renewable energy production nexuses. However, many studies have qualitatively discussed the issues concerning electricity consumption and renewable energy production and have made worthy conclusions [10,12,13,14,15,31].

To begin with, Lund [32] and Lund and Kempton [33] highlighted that oil supply is subject to high volatility; therefore, renewable energy sources can be used to control the impact of volatility in oil supply on electricity generation. They further emphasised that the US authorities must generate renewable energy effectively and efficiently. Hass et al. [13] highlighted that the European Union countries are using several policy measures to increase the use of renewable energy to generate electricity. Two of these commonly implemented policies are Tradable Green Certificates and Feed-in-Tariffs. Ibitoye and Adenikinju [14] argued that Nigeria is one of the poorest countries in the African region, with the lowest per capita energy consumption. The demand for electricity consumption in Nigeria can be effectively met by using renewable sources of energy generation. Strbac [34] highlighted that, with the development of renewable sources of energy generation, a broader application of them to the demand side of electricity production can be effectively managed in the UK. Denholm et al. [35] emphasised that policies on the storage of renewable energy should be integrated with policies on electricity consumption and generation. For instance, due to inconsistencies in the availability of wind, solar, and water energy, it may be easy to develop technologies that can store renewable energy that can be used for electricity generation. Roscoe and Ault [36] emphasised that smart meters in the UK can automatically change the demand for electricity by automatically adjusting renewable energy supply. Apergis and Payne [37] examined the relationship between non-renewable and renewable electricity consumption and the economic growth rate for 16 emerging market economies within a multivariate panel framework from 1990 to 2007. The research findings confirmed that a unidirectional causality exists between economic growth and renewable power consumption in the short run, in addition to a bidirectional causality between renewable electricity consumption and economic growth over time. Non-renewable electricity usage has a bidirectional causal relationship with short-term and long-term economic growth. According to Richardson [38], using electric vehicles and renewable energy sources will decrease carbon emissions from the power generation sector. Finn and Fitzpatrick [12] found that the demand side of electricity can be managed by using real pricing techniques for renewable energy generation. Bento and Moutinho [39] found that renewable electricity production reduces carbon dioxide emissions. Bélaïd and Youssef [40] found that renewable electricity consumption has significantly improved the environmental quality of Algeria. However, the current renewable electricity generation levels are insufficient to influence the carbon dioxide emissions reduction target significantly. Shukla et al. [41] argued that renewable electricity generation can be enhanced in South Asian countries by setting up renewable electricity production systems in remote islands. Atems and Hotaling’s [10] study, based on a sample of 174 countries, found a statistically significant relationship between renewable electricity generation and the economic growth rate. Lin and Chen [15] argued that, in the short run, energy pricing does not drive renewable energy technological innovation. Sharif et al. [31] found that renewable energy production hurts environmental degradation. The literature reviewed above confirms that none of the existing studies have investigated the causality between electricity consumption and renewable energy production. Bildirici and Kayikci [42] investigated the causal and dynamic relationship between current account balance, renewable energy production, energy imports, renewable energy consumption, and economic growth in the 1976–2019 period for the G20 countries through the panel Fourier bootstrapping ARDL model. This study’s findings confirmed a unidirectional causality between energy imports to renewable energy consumption and the current account deficit to renewable energy production. If countries reduced their dependence on energy imports, it would have a multiplier effect on improving the quality of the environment through the production of more renewable energy and reducing current account imbalances.

3.2. Electricity Consumption–Current Account and Renewable Energy Production–Current Account Nexusses

The rising current account deficits of many developed countries have become a hotly debated topic in the academic literature [43,44,45]. There are numerous studies published on the current account and its relationship with trade [46,47], exchange rates [15,48], inflation [49,50], and the economic growth rate [51,52,53]. Interestingly, no existing studies have explored the electricity consumption–current account nexus and the renewable energy production–current account nexus. The current account deficit imposes a threat on the macroeconomic stability of the nation. A high level of imports compared to exports is one factor worsening the current account balance of many countries. Countries facing severe issues related to their current account deficit are not focusing on export-led growth; instead, they focus on importing goods from households and business warehouses to meet domestic consumption.

Some of the key authors who have examined the relationship between electricity consumption, imports, and exports are Adams and Shachmurove [54], Jebli et al. [55], Rafindadi and Ozturk [56], and Al-Bajjali and Shamayleh [57]. For example, Glomsrød et al. [58] found that the government of Norway implements policies which bring changes to the overall electricity consumption of Norway to achieve a sustainable balance in the current account. Adams and Shachmurove [54] argued that China’s rapid industrialisation would increase oil imports, thus reducing China’s current account balance. Narayan and Smyth [59] investigated the relationship between electricity consumption, exports, and GDP by using data from the Middle Eastern countries. They could not find any relationship between electricity consumption and exports. Jebli et al. [55] noted the presence of unidirectional causality running from renewable energy consumption to imports of the 25 OECD countries. Rafindadi and Ozturk [56] found that a 1% rise in Japanese exports and imports increases Japanese electricity consumption by 0.921% and 0.2193%, respectively. Al-Bajjali and Shamayleh [57] argued that Jordon’s policymakers must invest in green energy projects and impose stringent bans on importing less energy-efficient electrical appliances to address the electricity demand crisis. Shahbaz et al. [60] found that human capital and export diversification are the two new key determinants of energy demand in the USA.

Furthermore, a few authors examined the relationship between renewable energy production and current account. For example, Gnansounou [61] highlighted that the high volatility of energy generation imports and export prices adversely affects the transition of energy technology. Weber et al. found that Chinese carbon dioxide has doubled in the last five years [62]. The empirical results indicated that one-third of Chinese carbon dioxide emissions were due to meeting the export demand. Consequently, renewable energy production is needed to meet the demand for energy by the private tradable sector. Based on this argument, we can postulate that renewable energy positively impacts exports. According to Jiang et al. [63], China’s quest towards achieving a low-carbon economy will have implications for importing and exporting goods. The demand for imports and exports will increase as new products are developed to increase public awareness of energy savings and greenhouse gas emissions. Dai et al. highlighted that renewable energy production may not incur substantial macroeconomic costs [64]. The economic cost that may be incurred in the short run is offset by the benefits of using renewable energy in the long run. Andini et al. found that import substitution is a key determinant of the positive impact of renewable electricity power generation projects [65]. Zeren and Akkuş found that non-renewable energy consumption causes trade openness [66].

The above discussion indicates that only a handful of studies have discussed issues concerning electricity consumption, renewable energy production, and current accounts. However, these studies lack empirical rigour to confirm the causality between electricity consumption, renewable energy production, and current account balance. This study uses the bootstrap approach to Granger Causality to examine the causality between these three variables. Unlike traditional causality and cointegration approach models, the bootstrap Granger Causality Test accounts for cross-section dependence as the estimation model is based on cross-section SURs rather than OLS. This ensures the causality test produces reliable and efficient statistics for valid inferences. Bildirici and Kayikci [42] investigated the causal and dynamic relationship between current account balance, renewable energy production, energy imports, renewable energy consumption, and economic growth in the 1976–2019 period for G20 countries through the panel Fourier bootstrapping ARDL model. This study’s findings confirmed a unidirectional causality between energy imports and renewable energy consumption and the current account deficit and renewable energy production. If countries reduced their dependence on energy imports, it would have a multiplier effect on improving the quality of the environment through the production of more renewable energy and reducing current account imbalances.

4. Data Collection

The data used in this study were collected from the OECD and World Bank Databases [5,17,25,67,68]. Six OECD countries have been considered: Finland, Germany, Switzerland, France, the United Kingdom, and the USA. There are three reasons for selecting these six countries for empirical analysis. First, these six countries have demonstrated high economic resource dependence on electricity production and renewable energy generation. There are several ways in which resources and key competencies are shared among these six countries. One of the main ways includes sharing innovative ideas, skilled labour, financial resources, and the goods and services used in producing electricity and generating renewable energy. Second, the selection of these six OECD countries is based on the availability of data for current account balance (CACC), electric power consumption (EPC), and renewable energy production (RENEN).

Table 1. Variables with Definitions.

No.	Variables	Definition of Variables	Database
1	Current Account Balance (CACC)	The current account variable is usually expressed in millions of US dollars and as a percentage of GDP. This indicator records a country’s international transactions with the rest of the world. This indicator is measured in millions of US dollars.	OECD
2	Electric Power Consumption (EPC)	This measure captures the difference between electricity production from power plants, combined heatless transmission, and own use of electric power by the power plants. Electric power consumption is measured in Kilowatt Hours (KwH) per capita.	World Bank
3	Renewable Energy Production (RENEN)	This measure captures the overall total contribution of renewables to the total primary energy supply. Renewable energy production is measured in thousand Tons of Oil Equivalent (TTOE).	OECD

Source: Created by the Authors (2019).

shows the variables with the definitions used for empirical analysis.

The data for the variables were extracted from the World Bank and OECD databases. These two databases provide the most reliable and accurate sources of data. One of the main advantages of using the OECD and World Bank databases is that such databases provide in-depth data definitions and are continuously updated using reliable sources. The data for the three variables used in this study were collected from 1960 to 2018.

5. Econometrics Methodology

5.1. Cross-Section Dependence Test

Globalisation has drastically increased the level and degree of financial integration experienced by economies. Through globalisation, economies can share financial resources, and intergovernmental economic organisations act as intermediaries in bringing the economies together to share innovative ideas, capabilities, and know-how. The mutual agreements amongst international trading blocs have increased the movement of the skilled workforce across national and international boundaries. Before the advent of globalisation, it was less likely for economic shocks to be easily transmitted from one country to another. However, globalisation has significantly increased economic integration by making it easier for economic shocks to be transmitted from one country to another. This idea has been well polarised by Robertson and Symons [69], Pesaran [3], Anselin [70], and Baltagi [71]. This requires testing and accounting for cross-section dependence in residuals before undertaking any test. The test for cross-section dependence can be easily captured by Equation (1):

$${CACC}_{it}={\alpha }_i+{\beta }^{\prime }{EPC}_{it}+{\beta }^{″}{RENEN}_{it}+{\epsilon }_{it} \mathrm{f}\mathrm{o}\mathrm{r} i=1,2\dots i_n;\hspace{1em}t=1,2\dots \mathrm{T},$$

(1)

$${EPC}_{it}={\alpha }_i+{\beta }^{\prime }{CACC}_{it}+{\beta }^{″}{RENEN}_{it}+{\epsilon }_{it} \mathrm{f}\mathrm{o}\mathrm{r} i=1,2\dots i_n;\hspace{1em}t=1,2\dots \mathrm{T},$$

(2)

$${RENEN}_{it}={\alpha }_i+{\beta }^{\prime }{EPC}_{it}+{\beta }^{″}{CACC}_{it}+{\epsilon }_{it} \mathrm{f}\mathrm{o}\mathrm{r} i=1,2\dots i_n;\hspace{1em}t=1,2\dots \mathrm{T},$$

(3)

In Equations (1)–(3), electric power consumption (EPC_it), current account balance (CACC_it), and renewable energy production (RENEN_it) are k × 1 vector of all regressors in our empirical model. The null hypothesis, in this case, is as follows:

$$H_o:p_{ij}=p_{ji}=cov\left({\epsilon }_{it},{\epsilon }_{jt}\right)=0 for i\ne j$$

versus

$$H_o:p_{ij}=p_{ji}\ne 0 for some i\ne j$$

This study uses a Cross-Section Dependence test for panel data models with varied sample sizes. The null hypothesis assumes that $\left({\epsilon }_{it},{\epsilon }_{jt}\right)$ are independent and identically distributed over time across cross-sectional units. Alternatively, $\left({\epsilon }_{it},{\epsilon }_{jt}\right)$ may be correlated across cross-sections. However, the assumption of no serial correlation still holds. Traditionally, it was assumed that panel data models had interdependent disturbances, but recent studies have proved that this is false [3]. Cross-section dependence tests are crucial for determining the degree of cross-section dependence in our sample size [3]. There are several serious implications of ignoring or assuming that cross-section dependence naturally exists in the sample without testing for cross-section dependence [72]. Two of these implications, well cited and discussed in the existing literature, are incorrect test statistics and estimator efficiency loss. The five cross-section dependence tests are considered for empirical purposes are as follows [72]:

• Breusch-Pagan Chi-square;
• Pearson LM Normal;
• Pearson CD Normal;
• Friedman Chi-square;
• Frees Normal.

The cross section dependence test procedure in Eviews 13 runs all five tests. The cross-section dependence test determines whether observations in a panel dataset are independent across distinct cross-sectional units (such as persons, corporations, and countries). There are seven distinct steps involved in running a Cross Section Dependence Test. These are (1) data preparation, (2) panel regression estimations, (3) residual calculation, (4) the cross section dependence test, (5) interpretation, (6) a robustness check, and (7) reporting.

5.2. Bootstrapped Granger Causality Test

Before undertaking the causality test, it is essential to undertake the cross-section dependence test to determine the degree of cross-section dependence in the sample. Once the cross-section dependence test rejects the null hypothesis of cross-section independence, we can apply the bootstrap approach to Granger Causality. There are several advantages of using the bootstrap approach to Granger Causality. For instance, the bootstrap approach to Granger Causality is a more efficient estimator based on SUR rather than OLS. In the presence of cross-section dependence, SURs are more efficient than OLS. We have a panel of six (less than ten) OECD countries for empirical purposes. Most of the series in our sample dates from 1960, which implies that the time dimension is significant; cross-correlations of the errors that exist in our sample can be modelled by using the seemingly unrelated regression equation framework that was developed by Pesaran [3] and Zellner [73]. The bootstrap approach to Granger Causality does not require unit root tests or cointegration tests. However, we will still conduct unit root tests to determine the variables’ stationarity. Equation (4) captures the equations for the bootstrap approach to Granger Causality:

$$\begin{gathered} {CACC}_{1,t}={\partial }_{\mathrm{1,1}}+\sum _{l=1}^{ml{CACC}_1}{\beta }_{\mathrm{1,1},l}{y}_{1,t-l}+\sum _{l=1}^{ml{EPC}_1}{\gamma }_{\mathrm{1,1},l}{EPC}_{1,t-l}+\sum _{l=1}^{ml{RENEN}_1}{\gamma }_{\mathrm{1,1},l}{RENEN}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{\mathrm{1,1},l}{z}_{1,t-l}+{\epsilon }_{\mathrm{1,1},t} \\ {CACC}_{2,t}={\partial }_{\mathrm{1,2}}+\sum _{l=1}^{ml{CACC}_1}{\beta }_{\mathrm{1,2},l}{y}_{2,t-l}+\sum _{l=1}^{ml{EPC}_1}{\gamma }_{\mathrm{1,2},l}{EPC}_{2,t-l}+\sum _{l=1}^{ml{RENEN}_1}{\gamma }_{\mathrm{1,1},l}{RENEN}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{\mathrm{1,2},l}{z}_{1,t-l}+{\epsilon }_{\mathrm{1,2},t} \\ {CACC}_{N,t}={\partial }_{1,N}+\sum _{l=1}^{ml{CACC}_1}{\beta }_{1,N,l}{y}_{N,t-l}+\sum _{l=1}^{ml{EPC}_1}{\gamma }_{1,N,l}{EPC}_{N,t-l}+\sum _{l=1}^{ml{RENEN}_1}{\gamma }_{\mathrm{1,1},l}{RENEN}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{1,N,l}{z}_{N,t-l}+{\epsilon }_{1,N,t} \end{gathered}$$

(4)

and

$$\begin{gathered} {EPC}_{1,t}={\partial }_{\mathrm{1,1}}+\sum _{l=1}^{ml{EPC}_1}{\beta }_{\mathrm{1,1},l}{y}_{1,t-l}+\sum _{l=1}^{ml{CACC}_1}{\gamma }_{\mathrm{1,1},l}{CACC}_{1,t-l}+\sum _{l=1}^{ml{RENEN}_1}{\gamma }_{\mathrm{1,1},l}{RENEN}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{\mathrm{1,1},l}{z}_{1,t-l}+{\epsilon }_{\mathrm{1,1},t} \\ {EPC}_{2,t}={\partial }_{\mathrm{1,2}}+\sum _{l=1}^{ml{EPC}_1}{\beta }_{\mathrm{1,2},l}{y}_{2,t-l}+\sum _{l=1}^{ml{CACC}_1}{\gamma }_{\mathrm{1,2},l}{CACC}_{2,t-l}+\sum _{l=1}^{ml{RENEN}_1}{\gamma }_{\mathrm{1,1},l}{RENEN}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{\mathrm{1,2},l}{z}_{1,t-l}+{\epsilon }_{\mathrm{1,2},t} \\ {EPC}_{N,t}={\partial }_{1,N}+\sum _{l=1}^{ml{EPC}_1}{\beta }_{1,N,l}{y}_{N,t-l}+\sum _{l=1}^{ml{CACC}_1}{\gamma }_{1,N,l}{EPC}_{N,t-l}+\sum _{l=1}^{ml{RENEN}_1}{\gamma }_{\mathrm{1,1},l}{RENEN}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{1,N,l}{z}_{N,t-l}+{\epsilon }_{1,N,t} \end{gathered}$$

(5)

and

$$\begin{gathered} {RENEN}_{1,t}={\partial }_{\mathrm{1,1}}+\sum _{l=1}^{ml{RENEN}_1}{\beta }_{\mathrm{1,1},l}{y}_{1,t-l}+\sum _{l=1}^{ml{EPC}_1}{\gamma }_{\mathrm{1,1},l}{EPC}_{1,t-l}+\sum _{l=1}^{ml{CACC}_1}{\gamma }_{\mathrm{1,1},l}{CACC}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{\mathrm{1,1},l}{z}_{1,t-l}+{\epsilon }_{\mathrm{1,1},t} \\ {RENEN}_{2,t}={\partial }_{\mathrm{1,2}}+\sum _{l=1}^{ml{RENEN}_1}{\beta }_{\mathrm{1,2},l}{y}_{2,t-l}+\sum _{l=1}^{ml{EPC}_1}{\gamma }_{\mathrm{1,2},l}{EPC}_{2,t-l}+\sum _{l=1}^{ml{CACC}_1}{\gamma }_{\mathrm{1,1},l}{CACC}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{\mathrm{1,2},l}{z}_{1,t-l}+{\epsilon }_{\mathrm{1,2},t} \\ {RENEN}_{N,t}={\partial }_{1,N}+\sum _{l=1}^{ml{RENEN}_1}{\beta }_{1,N,l}{y}_{N,t-l}+\sum _{l=1}^{ml{EPC}_1}{\gamma }_{1,N,l}{EPC}_{N,t-l}+\sum _{l=1}^{ml{CACC}_1}{\gamma }_{\mathrm{1,1},l}{CACC}_{1,t-l}+\sum _{l=1}^{ml{z}_1}{\vartheta }_{1,N,l}{z}_{N,t-l}+{\epsilon }_{1,N,t} \end{gathered}$$

(6)

In Equations (4)–(6), the electric power consumption (EPC_it), current account balance (CACC_it) and renewable energy production (RENEN_it) are presented for each country i with the period represented by t; l is the number of lags and ${\epsilon }_{1,N,t}$ refers to the disturbances that are contemporaneously correlated across equations. A two-step procedure is used to apply the bootstrap approach to Granger Causality. These two steps are explained below.

• Step One: A regression model with cross-section SURs is estimated to generate a bootstrapped sample of residuals for EPC, CACC, and RENEN.
• Step Two: Use a bootstrapped sample of EPC_i,t*, CACC_i,t*, RENEN_i,t* for the Granger Causality Test, Wald test, and bootstrap confidence intervals.

It is expected that a two-way direct relationship will be estimated between electric power consumption (EPCit), current account balance (CACCit), and renewable energy production (RENENit). All variables were converted into log form before the analysis was conducted to determine the relationship between electric power consumption, current account balance, and renewable energy production.

6. Empirical Findings and Their Discussions

The descriptive statistics presented in

Table 2. Descriptive Statistics.

Country	CACC		EPC		RENEN
	Mean	St. Dev.	Mean	St. Dev.	Mean	St. Dev.
Switzerland	53,191.34	20,386.56	6310.608	1711.158	3712.778	973.8371
Germany	116,675.9	131,240.9	5506.047	1813.188	10,621.14	11,283.07
Finland	3842.653	5903.715	10,325.26	5047.088	6412.898	1995.258
France	−3218.202	20,250.78	5122.382	2123.503	14,581.92	5580.355
United Kingdom (UK)	−30,313.03	43,809.32	4908.430	1019.614	2752.680	4112.099
United States of America (USA)	−194,224.4	234,365.0	10,318.05	3027.765	92,337.46	30,781.27

show that Switzerland (USD 53,191.34 million), Germany (USD 116,675.9 million), and Finland (USD 3842.653 million) have a positive and healthy mean current account balance as compared to France (USD −3218.202 million), the United Kingdom (USD −30,313.03 million), and the United States of America (USD −194,224.4 million). The unsustainable current account balances for France, the United Kingdom, and the United States of America can be addressed in several ways. One of the common ways of addressing this is investment in the production of renewable energy. Initially, investment in renewable energy generation may increase or decrease the current account balance. This is determined by the worth of the goods and services that are imported and exported for renewable energy production. If exports exceed imports, the current account balance will decrease and vice versa. Out of the six OECD countries, the highest renewable energy production is for the USA (92,337.46 TTOE), followed by France (14,581.92 TTOE), Germany (10,621.14 TTOE), Switzerland (3712.778 TTOE), and the UK (2752.680 TTOE). Electric power consumption is highest for Finland (10,325.26 KwH), followed by the USA (10,318.05 KwH), Switzerland (6310.608 KwH), Germany (5506.047 KwH), and France (5122.382 KwH).

The results of ADF statistics test presented in

Table 3. ADF unit root analysis.

		ADF Test
Country	Variables	Levels	t-Statistics
Switzerland	CACC	I (0)	−2.558
	EPC	I (0)	−3.069 **
	RENEN	I (0)	−0.694
	CACC	I (1)	−3.496 **
	EPC	I (1)	−2.219
	RENEN	I (1)	−11.229 ***
	Decision	CACC: I (1) EPC: I (0) RENEN: I (1)
Germany	CACC	I (0)	−0.231
	EPC	I (0)	−3.231 **
	RENEN	I (0)	2.339
	CACC	I (1)	−4.971 ***
	EPC	I (1)	−5.331
	RENEN	I (1)	−5.951 ***
	Decision	CACC: I (1) EPC: I (0) RENEN: I (1)
Finland	CACC	I (0)	−3.221 **
	EPC	I (0)	−2.638 *
	RENEN	I (0)	−1.422
	CACC	I (1)	−5.406 ***
	EPC	I (1)	−0.054
	RENEN	I (1)	−8.913 ***
	Decision	CACC: I (0) EPC: I (0) RENEN: I (1)
France	CACC	I (0)	−3.221 **
	EPC	I (0)	−2.638 *
	RENEN	I (0)	−1.423
	CACC	I (1)	−5.406 ***
	EPC	I (1)	−0.054
	RENEN	I (1)	−8.913 ***
	Decision	CACC: I (0) EPC: I (0) RENEN: I (1)
United Kingdom (UK)	CACC	I (0)	−0.577058
	EPC	I (0)	−3.914622 ***
	RENEN	I (0)	6.484
	CACC	I (1)	−7.331 ***
	EPC	I (1)	−4.695 ***
	RENEN	I (1)	2.276
	RENEN	I(2)	−7.285 ***
	Decision	CACC: I (1) EPC: I (0) RENEN: I (2)
United States of America (USA)	CACC	I (0)	−0.718148
	EPC	I (0)	−3.357462 **
	RENEN	I (0)	1.091886
	CACC	I (1)	−6.120418 ***
	EPC	I (1)	−6.890528 ***
	RENEN	I (1)	−6.796180 ***
	Decision	CACC: I (1) EPC: I (0) RENEN: I (1)

Note: ***, **, and * represent significance at 1%, 5%, and 10%, respectively. All analyses were undertaken based on automatic lag selection.

show that the current account balance is level and stationary for Finland and France, at least at a 5% level of significance. Similarly, electric power consumption is level and stationary for Switzerland, Germany, Finland, France, the UK, and the USA, at least at a 5% level of significance. Renewable energy production is stationary for Switzerland, at least at a 5% significance level. The rest of the variables are stationary at the first difference, except renewable energy production for the UK which is stationary at the second difference.

The results of the cross-section dependence test shown in

Table 4. Cross-section dependence analysis.

Equation:
Null hypothesis: Cross-sectional independence
	CACC			EPC			RENEN
Test	Statistic	d.f.	Prob.	Statistic	d.f.	Prob.	Statistic	d.f.	Prob.
Breusch-Pagan Chi-square	102.4401	15	0.0000	96.85849	15	0.0000	78.71164	15	0.0000
Pearson LM Normal	14.86887		0.0000	13.84980		0.0000	10.53666		0.0000
Pearson CD Normal	−2.357699		0.0184	5.712588		0.0000	5.004555		0.0000
Friedman Chi-square	12.86686	61	1.0000	92.34041	61	0.0059	148.5631	61	0.0000
Frees Normal	1.623827		0.0000	1.678362		0.0000	1.613497		0.0000
All analysis was undertaken based on automatic lag selection.

Source: Eviews 8.0 (2019).

reveal that the null hypothesis of cross-sectional independence is rejected for current account balance, electric power consumption, and renewable energy production for the six OECD countries. This indicates that the six OECD countries in our sample depend on each other in several ways. First, these six OECD countries are aggressively involved in trade-related activities, sharing research-related capabilities and key competencies [55,74]. Second, these countries have always encouraged cross-border investments, thereby encouraging liberalisation and a free flow of capital movements from one country to another. The level of financial integration in the last decade has increased significantly as there are numerous multinational companies headquartered in the USA, and their subsidiaries are based in Switzerland, Germany, Finland, France, the United Kingdom (UK), and the United States of America (USA) [75].

Nonetheless, the opposite case is also true as numerous companies headquartered in Switzerland, Germany, Finland, France, and the United Kingdom have subsidiaries in the USA. Third, the OECD countries in our sample have been working towards achieving international milestones related to meeting the increasing demand for electricity production by using renewable sources of energy generation. These OECD countries have been exchanging knowledge, technological know-how, and key competencies in producing renewable energy as efficiently as possible [17]. The common renewable sources of energy generation these countries are tapping into are waves, tide, wind, geothermal, solar, and hydro [17]. This shows that the six OECD countries in our sample depend on each other in several ways, and the cross-section dependence test also captures this effect.

The bootstrapped Granger causality test results for the six OECD countries are presented in

Table 5. Bootstrapped Granger Causality Analysis (EPC vs. RENEN).

	Countries	H0: EPC Does Not Cause RENEN	H0: RENEN Does Not Cause EPC
		Bootstrap F-Statistics	Bootstrap F-Statistics
1	Switzerland	0.07997	2.90581 *
2	Germany	1.42841	0.39343
3	Finland	0.23104	0.98235
4	France	0.64613	1.97479
5	United Kingdom (UK)	0.52344	1.19431
6	United States of America (USA)	2.35128	1.27885

Note: * represent significance at 10% level. All analyses were undertaken based on automatic lag selection.

. The empirical results show that the null hypothesis that electricity power consumption does not cause renewable energy production is rejected for Switzerland, Germany, Finland, France, the United Kingdom (UK), and the United States of America (USA). The null hypothesis that renewable energy production does not cause electric power consumption is rejected only in the case of Switzerland, at p < 10%.

Table 6. Bootstrapped confidence interval critical values (EPC vs. RENEN).

	Countries	Bootstrap Wald Test T-Statistics	Bootstrap β Coefficients	H0: EPC Does Not Have an Impact on RENEN
				Bootstrap Confidence Interval Critical Values
				90%		95%		99%
				Low	High	Low	High	Low	High
1	Switzerland	14.52163 ***	0.487285 ***	0.431109	0.543462	0.419981	0.554590	0.397630	0.576941
2	Germany	5.505355 ***	2.471924 ***	1.720240	3.223608	1.571337	3.372511	1.272266	3.671582
3	Finland	6.536499 ***	0.219662 ***	0.163403	0.275922	0.152258	0.287066	0.129874	0.309450
4	France	11.17913 ***	2.063843 ***	1.754775	2.372911	1.693551	2.434135	1.570582	2.557103
5	United Kingdom (UK)	3.776071 ***	1.457062 ***	0.811075	2.103049	0.683110	2.231015	0.426093	2.488031
6	United States of America (USA)	15.73048 ***	8.701787 ***	7.775699	9.627875	7.592248	9.811326	7.223788	10.17979
	Countries	Bootstrap Wald Test T-Statistics	Bootstrap β Coefficients	H0: RENEN Does Not Have an Impact on EPC
				Bootstrap Confidence Interval Critical Values
				90%		95%		99%
				Low	High	Low	High	Low	High
1	Switzerland	12.54597 ***	1.469294 ***	1.273233	1.665354	1.234395	1.704192	1.156389	1.782198
2	Germany	6.184959 ***	0.157270 ***	0.114701	0.199839	0.106268	0.208272	0.089331	0.225209
3	Finland	6.306914 ***	1.974258 ***	1.450208	2.498308	1.646398	2.602118	1.137895	2.810621
4	France	11.34860 ***	0.353262 ***	0.301149	0.405374	0.290826	0.415697	0.270093	0.436431
5	United Kingdom (UK)	3.412491 ***	0.151879 ***	0.077370	0.226389	0.062610	0.241149	0.032965	0.270794
6	United States of America (USA)	16.30379 ***	0.102582 ***	0.092048	0.113115	0.089962	0.115202	0.085771	0.119393

Note: *** represent significance at 1% level. All analyses were undertaken based on automatic lag selection.

shows that the null hypothesis that electric power consumption does not have an impact on renewable energy production is rejected for all the six OECD countries in our sample, with a significance level of less than 1%. The bootstrap β coefficients show that a 1 unit increase in renewable energy production will increase electric power consumption by 8.702 units for the USA, followed by 2.47 units for Germany, 2.06 units for France, 1.46 units for the UK, 0.49 units for Switzerland, and 0.22 units for Finland. This finding is intuitively appealing as the electricity demand is rapidly increasing, and countries must find renewable ways to generate electricity to meet their growing demand for electricity [14,35]. Similarly, the null hypothesis that renewable energy production does not impact electric power consumption is rejected for all OECD countries with a significance level of less than 1%. The bootstrap β coefficients show that a 1 unit increase in electric power consumption will increase renewable energy production by 1.97 units for Finland, followed by 1.47 units for Switzerland, 0.35 units for France, 0.16 units for Germany, 0.15 units for the United Kingdom, and 0.10 units for the USA.

Table 7. Bootstrapped Granger Causality Analysis (EPC vs. CACC).

No.	Countries	H0: EPC Does Not Cause CACC	H0: CACC Does Not Cause EPC
		Bootstrap F-Statistics	Bootstrap F-Statistics
1	Switzerland	0.89874	4.37267 *
2	Germany	0.68023	0.51526
3	Finland	1.46654	0.64658
4	France	1.53091	10.7089 ***
5	United Kingdom (UK)	0.98849	1.04918
6	United States of America (USA)	4.33835 **	0.08774

Note: ***, **, and * represent significance at 1%, 5%, and 10%, respectively. All analyses were undertaken based on automatic lag selection.

shows the six OECD countries’ bootstrapped Granger Causality Test results. The null hypothesis that electric power consumption does not cause the current account balance was rejected only in the case of the USA. On the other hand, the null hypothesis that the current account balance does not cause electric power consumption was rejected for Switzerland (p < 0.1) and France (p < 0.01).

Table 8. Bootstrapped confidence interval critical values (EPC vs. CACC).

No.	Countries	Bootstrap Wald Test T-Statistics	Bootstrap β Coefficients	H0: EPC Does Not Have an Impact on CACC
				Confidence Interval Critical Values
				90%		95%		99%
				Low	High	Low	High	Low	High
1	Switzerland	0.249044	7.611215	−46.51167	61.73410	−58.41350	73.63593	−84.44931	99.6717
2	Germany	10.25986 ***	289.8447 ***	241.3348	338.3547	231.2571	348.4324	210.2139	369.4756
3	Finland	1.452269	1.396737	−0.271020	3.064494	−0.623850	3.417325	−1.371638	4.165113
4	France	0.586445	12.77958	−25.60223	51.16138	−33.95880	59.51795	−52.09065	77.6498
5	United Kingdom (UK)	−4.512872 ***	−23.39872 ***	−32.07882	−14.71862	−33.79829	−12.99916	−37.25182	−9.54563
6	United States of America (USA)	−8.346923 ***	−56.19020 ***	−67.46010	−44.92031	−69.69258	−42.68783	−74.17650	−38.20390
No.	Countries	Bootstrap Wald Test T-Statistics	Bootstrap β Coefficients	H0: CACC Does Not Have an Impact on EPC
				Confidence Interval Critical Values
				90%		95%		99%
				Low	High	Low	High	Low	High
1	Switzerland	−0.367759	−0.001042	−0.006060	0.003976	−0.007163	0.005079	−0.009577	0.007493
2	Germany	7.517867 ***	0.002525 ***	0.001948	0.003101	0.001828	0.003221	0.001578	0.003471
3	Finland	−0.232544	−0.009669	−0.081773	0.062434	−0.09027	0.077688	−0.129356	0.110018
4	France	−0.371569	−0.000856	−0.004914	0.003202	−0.005798	0.004085	−0.007715	0.006003
5	United Kingdom (UK)	−6.125400 ***	−0.020140 ***	−0.025644	−0.014636	−0.026735	−0.013545	−0.028925	−0.011355
6	United States of America (USA)	−7.658754 ***	−0.008843 ***	−0.010776	−0.006910	−0.011159	−0.006527	−0.011928	−0.005758

Note: *** represent significance at 1% level. All analyses were undertaken based on automatic lag selection.

shows that the null hypothesis that electric power consumption does not impact the current account balance can be rejected in the case of Germany, the United Kingdom, and the United States. This indicates that a 1 unit increase in electric power consumption will increase Germany’s current account balance by 289.84 units and erode the current account balance of the UK and USA by 23.40 units and 56.19 units, respectively. Similarly, a 1 unit increase in current account balance will increase Germany’s electric power consumption by 0.003 units and decrease the electric power consumption of the UK and USA by 0.02 units and 0.008 units, respectively.

Table 9. Bootstrapped Granger Causality Analysis (RENEN vs. CACC).

	Countries	H0: RENEN Does Not Cause CACC	H0: CACC Does Not Cause RENEN
		Bootstrap F-Statistics	Bootstrap F-Statistics
1	Switzerland	0.04949	0.97759
2	Germany	1.98625	0.13346
3	Finland	0.46535	2.37416
4	France	0.00382	2.19741
5	United Kingdom (UK)	0.16612	1.08341
6	United States of America (USA)	0.60158	0.29509

Source: Eviews 8.0 (2019).

shows that the null hypothesis that renewable energy production does not cause current account balance, and current account balance does not cause renewable energy production can be rejected for all six of the OECD countries in our sample. This finding is intuitively appealing as renewable energy production is growing in the six OECD countries. However, the growth has not yet reached a stage whereby renewable energy production can become one of the major causes of current account balance in the six OECD countries and vice versa. The causality drivers of the CACC are imports, exports, net income, and net current transfers [76,77].

Table 10. Bootstrapped confidence interval critical values (RENEN vs. CACC).

No.	Countries	Bootstrap Wald Test T-Statistics	Bootstrap β Coefficients	H0: RENEN Does Not Have an Impact on CACC
				Confidence Interval Critical Values
				90%		95%		99%
				Low	High	Low	High	Low	High
1	Switzerland	0.596420	6.381433	−12.56678	25.32965	−16.73357	29.49644	−25.84861	38.61157
2	Germany	19.40682 ***	11.51826 ***	10.49910	12.53751	10.28738	12.74913	9.845280	13.19123
3	Finland	−3.056442 ***	−2.742061 ***	−4.297761	−1.186361	−4.626885	−0.857237	−5.324429	−0.159693
4	France	−3.881570 ***	−5.769546 ***	−8.387549	−3.151544	−8.957546	−2.581546	−10.19431	−1.344782
5	United Kingdom (UK)	−15.63027 ***	−10.39552 ***	−11.50896	−9.282084	−11.72952	−9.061521	−12.17252	−8.628521
6	United States of America (USA)	−5.466601 ***	−4.883083	−6.378500	−3.387666	−6.674730	−3.091435	−7.269708	−2.496457
No.	Countries	Bootstrap Wald Test T-Statistics	Bootstrap β Coefficients	H0: CACC Does Not Have an Impact on RENEN
				Confidence Interval Critical Values
				90%		95%		99%
				Low	High	Low	High	Low	High
1	Switzerland	1.684194	0.005727	−0.000295	0.011749	−0.001619	0.013073	−0.004516	0.015970
2	Germany	15.99593 ***	0.083298 ***	0.074356	0.092240	0.072498	0.094097	0.068619	0.097976
3	Finland	−4.689594 ***	−0.151901 ***	−0.208070	−0.095733	−0.219953	−0.083850	−0.245137	−0.058665
4	France	−3.462684 ***	−0.081916 ***	−0.123583	−0.040249	−0.132655	−0.031177	−0.152339	−0.011494
5	United Kingdom (UK)	−17.21812 ***	−0.053069 ***	−0.058229	−0.047909	−0.059251	−0.046887	−0.061304	−0.044834
6	United States of America (USA)	−7.471812 ***	−0.085804 ***	−0.105029	−0.066579	−0.108838	−0.062771	−0.116487	−0.055122

Note: *** represent significance at 1% level. All analyses were undertaken based on automatic lag selection. Source: Eviews 8.0 (2019).

shows that the null hypothesis that renewable energy production does not impact the current account balance can be rejected for Germany, Finland, France, and the UK. This indicates that a 1 unit increase in renewable energy production will increase Germany’s current account balance by 11.52 units and decrease Finland’s current account balance by −2.74 units, by −5.77 units for France, and by −10.40 units for the UK. Similarly, a 1 unit increase in the current account balance for Germany will increase renewable energy production by 0.08 units but decrease renewable energy production of Finland, France, the UK, and the USA by −0.15 units, −0.08 units, −0.05 units, and −0.09 units, respectively.

One notable research finding reveals that electric power consumption directly affects the current account balance in the case of the USA. This indicates that energy used to produce electric power is one of the drivers of the current account balance in the USA. The demand for electricity in the USA has significantly increased, which must be offset by increasing electricity production. According to the US Energy Information Administration (2019a), the three major sources of electricity generation in the US are fossil fuels, nuclear energy, and renewable energy sources. To generate electricity, non-renewable sources of energy, such as fossil fuels, must be imported to offset any deficit in the supply of fossil fuels produced locally to meet the demand for energy by the providers of electricity [78]. The imports of fossil fuels for electricity generation continue to increase in the case of the USA, and it is one of the contributors to the negative current account balance of the USA. Glomsrød et al. [58] argue that the government of Norway must implement policies to make changes to electricity consumption and production to maintain a sustainable current account balance. Norway is involved in importing electricity, and this has an impact on the current account balance. As compared to the poorer countries, wealthy countries show a higher link between electricity use and wealth growth. Considering the case of the world economy as a whole, there is a larger relationship between electricity use and wealth generation than there is between overall energy use and wealth.

On the other hand, the USA is involved in importing fossil fuels to generate electricity; hence, it should reduce its import of oil and petroleum products used for electricity generation. Compared to fossil fuels, the consumption of renewable electricity is less expensive and volatile. Power markets are inherently local rather than global. In the previous decade, the average distance between residences and the nearest big electricity-producing facility was only 5 km. In the case of France and Switzerland, the current account balance causes electric power consumption, which undoubtedly confirms that import- and export-related decisions are made in light of the current account balance’s impact on electric power consumption. Rafindadi and Ozturk [56] also confirmed that Japanese imports impact electricity consumption as non-renewable energy sources have to be imported to meet the increasing demand for electricity. Another critical research finding from this study is that causality does not exist between renewable energy production and the current account balance of the six OECD countries. One of the main reasons attributed to this is that the trade of goods and services related to renewable energy production is not large enough relative to the overall trade of goods and services in order to drive the causality relationship between renewable energy production and current account balance.

7. Conclusion and Policy Implications

This study examined the causality between electricity consumption, renewable energy production, and the current account of the six OECD countries. The empirical findings confirmed a unidirectional causality between electric power consumption and the current account balance of the USA. Alternatively, the current account balance causes electric power consumption in France and Switzerland. There is no causality relationship between current account balance, electric power consumption, and renewable energy production in the case of Germany, Finland, and the UK. The bootstrap β coefficients show that a 1 unit increase in renewable energy production will increase electric power consumption by 8.702 units for the USA, followed by 2.47 units for Germany, 2.06 units for France, 1.46 units for the UK, 0.49 units for Switzerland, and 0.22 units for Finland. A 1 unit increase in electric power consumption will increase Germany’s current account balance by 289.84 units and erode the current account balance of the UK and USA by 23.40 units and 56.19 units, respectively. A 1 unit increase in renewable energy production will increase Germany’s current account balance by 11.52 units and decrease Finland’s current account balance by −2.74 units, −5.77 units for France, and −10.40 units for the UK.

The findings from this study have important policy implications. First, stringent policies should be implemented on importing high energy-consuming electrical appliances into the USA. These policies are necessary to ensure that the high demand for electricity can be sustainably met without increasing the current account deficit of the USA. Access to affordable and sustainable energy improves efficiency and productivity by empowering individuals to contribute to their communities. Apparently, it generates jobs, cash, and helps people overcome poverty [79]. It also changes the lives of families and communities by improving health, education, and communication services. Women are also empowered by lowering the time and effort required to collect and use traditional energy sources. It also promotes a cleaner environment and reduces the impact on climate change [80,81,82,83,84].

Although no causality was established in the case of Germany, Finland, and the UK, policymakers also need to learn from the experiences of the USA and promote the use of energy-efficient electrical items and technologies by households and the tradeable sector. The policymakers should use import substitution policies to increase the production of domestically produced electricity and restrict the import of electricity from the neighbouring countries, in the case of France.

One of the limitations of this study is that it is based on a small sample of six OECD countries. Future studies can conduct similar studies on a wider sample size.