Investigating the asymmetry effects of crude oil price on renewable energy consumption in the United States

The reduction in oil prices might make crude oil a cheaper alternative to renewable energy (RE). Given this, the present paper examines the effect of fluctuation of oil prices on the use of RE in the United States (US) during the period 1970 to 2018. We constructed two nonlinear autoregressive distributed lag (NARDL) models to examine the effect of the positive and negative oil price shocks on the use of RE in the US. The RE consumption is taken as the dependent variable and the gross domestic product (GDP), Brent crude prices, population density, trade openness, and price index as independent variables. The result revealed that the rise in crude oil price, GDP, and population density will increase RE use in the short run and in the long run as well. Moreover, the study finds that any decrease in oil prices will decrease RE use in the short run and its effect will eventually diminish in the long run. On the policy front, it is suggested that US should raise its energy security by reducing its dependency on imported crude oil and increase the role of RE through the imposition of taxes on oil and increase the base of production and consumption through a series of measures.


Introduction
In the last four decades, the dependency on energy has increased much faster than ever before. The world witnessed a paradigm shift in the source of energy from wind, water, and coal to oil and natural gas during this period. Fossil fuel played a substantial role in the source of modern energy generation as most of the traditional energy generating sources was replaced. However, rise in the prices of oil and the environmental consequences of greenhouse gas emission highlighted the importance of an alternative source of energy, a replacement to the use of crude oil. More than a decade now, the importance of renewable sources of energy has increased across the globe, including in the US. The political commitment of G7 countries and the European Union to future sustainability of energy through the availability of RE shows its importance (European Commission News, May 2015). Even the RE initiatives of the G7 countries aim at making energy accessible to all in the African region by improving RE by 2030 (United Nations 2015).
Many countries started to encourage the use of alternative source of energy and supported technological innovation, mechanism and policies to enhance the production of RE (Dogan et al. 2021). The prime objective of these efforts is to increase the availability of energy for all through the global RE system at a cheaper and affordable price along with protecting the environment. As it is a cheaper alternative and has low carbon emission, the US has witnessed a rise in both production and consumption of RE in recent years (World Bank 2020). Even the expansion of renewables has often surpassed the expected target (Schmalensee and Bulovic 2015); and in particular, the growth of solar energy has surpassed the target and projection of the International Energy Agency's (IEA). While the production and supply of renewables is on a rising trend, the fluctuation in the price of crude oil, in particular, a decrease in crude oil prices has left several questions unanswered in the hands of the academia and policy makers. It is evident that between 2011 and 2020 (April), the price of the crude oil declined from a high of about $120 per barrel to as low as $21 per barrel. Though the oil price has increased thereafter, but in general, the declining trend in the oil prices in the last one decade might affect the demand for RE and derail the ambition of a carbon-free energy future.
To be precise, the falling price of the crude oil might create two possibilities. The first is based on the notion that any fall in the price of crude oil would increase its demand. This might have some serious consequences on the future demand for RE. In other words, a reduced oil price would again increase the demand for oil, substituting the alternative use of the renewable resources in the US. The second possibility is that, the cheaper the crude oil, the more funds that the country will have to promote RE projects, which in turn, increase the demand for RE sources. Consequently, based on these two possibilities, this paper is motivated to investigate whether the increase or the decrease in crude oil price would affect the demand for RE in the US.
The US is the world's largest crude oil consumer and the second-largest producer of CO 2 emission. Therefore, moving toward crude oil might have serious consequences on world environmental degradation. Moving toward RE sources is important to compact greenhouse gas emission that mostly comes from the consumption of fossil fuels. Therefore, the core objective of this research is to explore the impact of oil price fluctuations on the use of RE in the US.
RE consumption in the US is on an increasing trend. The US is among the few countries in the world having the best resources along with the knowledge of innovation and financing abilities. Since the beginning of late 18 th century, fossil fuel has played a major role as the source of energy for the US economy. However, increased concern about the greenhouse gas emission and the US becoming a major global player in the production and consumption of cleaner energy have shifted the outlook since the beginning of the current century.
Toward the late nineties, energy consumption in the US shifted to RE, mostly derived from the wind, solar and biofuels. The share of RE in the total energy consumption in the US increased from just about 3% in the late nineties to 7.5% in 2010 and to 11% in 2019. The international Renewable Energy Agency (IRENA) predicts that the share of RE in the US could increase to 27% by 2030 with more than 50% coming from the power sector. The power sector which accounts for more than 55% in the total energy consumption at present shows an increasing share being generated from renewable sources.
With certain assumptions, in 2020, the US Energy Information Administration forecast that the RE production would increase from 18% in 2019 to 31% in 2050. The electric power sector, the largest source of energy to the US has undergone a drastic shift in its production from fossil fuel and nuclear power to wind and solar sources. Data shows that the share of renewables in electricity production registered an increase from 11% in 2010 to 14% in 2014 and 20% in 2020. With increased renewable electricity production and continuous fall in the price of natural gas, the wholesale prices of electricity has fallen in the US market. Increased production of RE and a fall in the prices of natural gas and other petroleum products could pose a threat to the consumption of renewables. However, data shows a satisfactory trend in the consumption of RE as it has increased from 6.5% in 2007 to more than 12% in 2019. Over the years, the development of technology, declining cost, and improving performance is making solar and wind more popular and competitive than other sources of energy. The production of energy (e.g. electricity) from these sources will minimize emission and reduce the cost of energy; and its supply is unlikely to decline over time.
The policies in regard to RE at the state and central levels are also encouraged and implemented so as to popularize its use. States such as California, New Jersey, and Massachusetts have enacted new laws for RE supply and consumption, for e.g., California enacted policies to supply 100% cleaner energy by 2045, New Jersey and Massachusetts have prepared a blueprint to produce 50% and 35% of the energy from renewable sources, respectively. The effort of these governments could increase RE production; however, its use will primarily depend upon the prices of the available alternative energy and the strictness of the law in future.
Given that each country has different economic characteristics, the panel result might not deliver a clear or an accurate conclusion. Moreover, majority of the studies examined the positive effect of oil price shocks (linear effect of oil prices) on RE by ignoring the negative effect of oil price shocks. Given this, the present study used the nonlinear ARDL model to find the nonlinearity effects of oil prices (increase or decrease) on RE use in the US.

Literature review
Many studies have examined the main macroeconomic determinants that impact energy use in general. The most utilized variable is the GDP growth (Omri and Kahouli 2014;Kyophilavong et al. 2015;Rafindadi and Ozturk 2016;Shahbaz et al. 2016;Dogan and Aslan 2017;Ben Jebli et al. 2019;Mujtaba et al. 2020 and so forth). The relationship between GDP growth on air pollution is well-examined by the previous studies and the effects are clearly underlined. Many previous studies concluded that the increase in economic activities increases environmental degradation more in developing and emerging economies than the developed economies. This can be linked to the environmental Kuznets curve (EKC) hypothesis. At the early stages of economic development, the country is basically a polluting industrial economy until a certain stage of economic development; then the economy will transform into a clean service-based economy. Moreover, when income is higher, citizens will have more preference for better environmental quality (Dinda 2004;Erdogan et al. 2020).
Foreign direct investment (FDI) is a widely utilized pollution indicator (Omri and Kahouli 2014;Mudakkar et al. 2013;Azam et al. 2015Azam et al. , 2019Khan and Ozturk 2020;Abdo et al. 2020;. FDI can have different effects on environmental degradation depending on the type of investment. Basically, large industrialized countries tend to send their dirty factories to less developed nations with less stringent environmental regulations, cheap labor, and resources. Therefore, FDI increases environmental degradation mostly in less developed countries than the developed ones. Similarly, the development of the financial institution is proven to be an important indicator of environmental degradation (Sadorsky 2010;Zaidi et al. 2019;Nasir et al. 2019 and so forth).
Population is an important indicator of environmental degradation as the world population is growing rapidly far more than the earth's ability to support. Therefore, the rise in population has proven to increase environmental degradation in most of the countries (Omri and Kahouli 2014;Azam et al. 2015;Liu et al. 2019, Mujtaba et al. 2020 and so forth). Despite the differences in methodologies and the countries concerned, most of the existing studies have established that variables such as GDP, FDI, financial institutions, and population have significant consequence on the use of energy. Besides these, the impact of energy prices on energy consumption is not new, as found by many scholars in their empirical analysis (Sadorsky 2010;Aguirre and Ibikunle 2014;Murshed and Tanha 2020;Apergis et al. 2021).
Emphasis on the key influences of RE use is relatively a newly focused area. Much less is known about the impact of energy price, particularly the oil price on RE use, though several other aspects of RE are studied and debated intensively. In context to the relation between the oil price and the RE, Omri et al. (2015), Aguirre andIbikunle (2014), Ferrer et al. (2018), Cao et al. (2019) found that oil price negatively affects RE consumption and investment. Majority of these studies investigating the relationship between oil price and use of RE considers a panel of countries rather than a single country for analysis. For example, Coban and Topcu (2013) while studying in context of EU27 countries find a negative relation of oil price with the RE demand. The study by Aguirre and Ibikunle (2014) on 38 countries of the EU OECD and BRICS finds that oil price has a marginal negative impact on consumption of RE. The study by Doytch and Narayan (2016), involving 74 countries, shows evidence of negative influence of the energy price index on RE consumption in a model based on the Blundell-Bond dynamic panel estimator. Bondia et al. (2016) in a multivariate framework find that in the long run, oil prices have no effect on RE consumption, but influences it in the short run. Similarly, Dogan et al. (2021) find that oil price has weaker relation with RE consumption than RE production. However, Murshed and Tanha (2021) in the context of South Asian economies during the period 1990to 2018 have found that crude oil prices influenced the RE transition.
Though most of these studies find a weak or negative relationship between oil price and the use of RE, its intensity varies in different studies. Despite the evidence of a negative relationship between the two variables, these studies fail to provide a clear explanation behind such a relationship. On the other hand, some of these studies suggest the reason for such negative relationship could be the small sample size. Some others suggest that the relationship is because of the nature and pattern of energy consumption of these countries, industrial or otherwise. Moreover, some other scholars suggest that the negative relationship between the two variables is because crude oil and RE are complementary goods. However, scholars such as Lin and Omoju (2017) found that the surge in oil prices resolves an increase in the demand for renewables. Such a relationship exists because RE sources are a substitute for crude oil; therefore, the rising prices of crude oil will make RE a cheaper alternative.
Particularly, in the context of the US, there are many of studies (Menyah and Wolde 2010; Dogan and Ozturk 2017; Dogan and Turkekul 2016; Ben Youssef 2020, etc.) that examine different aspects, particularly the renewable/ nonrenewable energy use and CO emission. For example, Menyah and Wolde (2010) using a modified Granger causality test, studied the relationship between carbon dioxide (CO 2 ) emission, renewable and nuclear energy consumption and real GDP for the period 1970 to 2007. Dogan and Ozturk (2017) studied the influence of GDP, RE and non-RE consumption on CO 2 emission in the EKC model for the period 1980 to 2014. In another study, Dogan and Turkekul (2016) investigated the causal relationship between CO 2 , energy consumption, GDP, trade openness, urbanization, and financial development for the period 1960 to 2010. Ben Youssef (2020) studied the impact of foreign research and development (R&D) spillovers on RE consumption and pollution using the database of the period 1980 to 2016. Despite the availability of a number of studies in the context of the US, we have not come across any specific study, which properly examined the relation between the crude oil prices and RE in the US alone. This motivates us to empirically examine this topic in the US context. However, there are many existing studies, which used the panel data to find an overall result for a group of countries (Inglesi and Dogan 2018;Erdogan et al. 2020;Mujataba and Jena 2021, etc.) and some including the US. Some other studies specific to the US found the impact of oil prices on CO 2 emission (Hammoudeh et al. 2014;Boufateh 2019;Ullah et al. 2020, etc.). For example, Hammoudeh et al. (2014) using the daily sample from July 2006 to November 2013 under a quantile regression framework find that increase in crude oil price reduces CO 2 substantially in the US. Boufateh (2019) using the NARDL bound testing approach over the period 1976 to 2013 suggests that positive (negative) fluctuations in crude oil prices reduces (increase) CO 2 emission in the US. Similarly, applying NARDL for the period 1981 to , Ullah et al. (2020 find that positive shocks in diesel prices in the US have a significant negative impact on carbon emission in the long run. Given this, empirical studies on the topic under the present investigation have not been properly covered for the US. However, a recent study by Guo et al. (2021) in the context of G7 countries has explored a short-term and long-term dynamic relationship between the oil price and consumption of RE using the NARDL model. The result reveals that for Canada, the US, and Italy, an increase in oil prices has a stronger impact on RE consumption than a negative change. Our study differs from this in terms of the data coverage, variable used, and the methodological specifications to capture RE consumption. While Guo et al. (2021) used linear and NARDL, we constructed two NARDL models to examine the effect.

Model and data
Relying on the findings of past studies (Coban and Topcu 2013;Doytch and Narayan 2016;Mujtaba et al. 2020 and, we utilized the following two models specification (to examine the robustness of the results) to examine the variables that determine RE consumption during the period 1970 to 2018. Different from the previous studies, the nonlinear ARDL is utilized to examine the effects of the positive and the negative shocks in oil prices on RE consumption.
Here, lnRE is the natural log of RE consumption (millions of kilowatt-hours), lnOP is the natural log of Brent crude price (US$ per barrel), lnPD is the natural log of population density (per sq. km of land area), lnTRD is the natural log of trade openness measured in millions of constant US dollars and is calculated as (Export+Import)/GDP, lnPI the natural log of price index and ε t is error term with the assumption of normal distribution 1 .
The above model represents basic RE consumption. It contains GDP as an indicator of income and price index, since changes in the price level might affect the demand (consumption) of RE. Moreover, the price of crude oil is also included, which represents the price of an alternative source of energy, the increase in crude oil prices will make RE a more attractive alternative. Population density is a determining factor of RE as the increase in population will increase energy use in general. Therefore, high population density might increase the consumption of RE. Moreover, the increase in the country's price level will cause a decline in the consumer purchasing power, which will cause a reduction in the consumption of RE.
The data of RE consumption per capita was obtained from the Energy Information Administration (2016), GDP, Brent crude oil prices, population density, and price index were retrieved from the World Development Indicators (2020), and trade openness was obtained from the OECD database (2020).

Residual augmented least squares (RALS) unit root test
The first step is to check the stationarity; therefore, the RALS unit root test of Meng et al. (2014) is used. The specification of the model is as follows The null of nonstationarity is based on ϕ=0 and the ordinary least square approach is utilized to generate the tests statistic. Z t is a vector that houses the exogenous variables, which is expressed as Z t ¼ 1; t; D * 1t ; …; D * Rt ; DT * 1t ; …; DT * Rt Â Ã 0 . Whereas, the dummy D * 1t ¼ 1 for t ≥ T B + 1, i = 1, …, R, and 0, otherwise, and D * 1t ¼ t−T Bi for t ≥ T B + 1 and 0 otherwise. T Bi capture the locations of the breakpoints and δ are the coefficients that are derived from the regression of Δy t on ΔZ t . e S * t capture the transformed term of the detrended variables, . Consistent with the work of Lee et al. (2012), the transformation is needed to make the method more robust. b w t is the information on non-normal errors and its presence transform the test into a RALS framework. In the Lagrange multiplier (LM) test of Lee et al. (2012), γ = 0. The lagged terms of ΔS t− j are incorporated into the analysis in order to cater for the incidence of autocorrelation 2 . The next step is to examine the nonlinear ARDL as suggested by Shin et al. (2011) 3 . This test is used to reveal the asymmetric (nonlinear) relationship in the short-run and the long-run. Basically, the nonlinear ARDL is able to examine whether the positive shocks of the independent variables have the same effect as their negative shocks on the dependent variables.
Contrasting earlier studies, the findings of this research reveal whether the short-run and long-run positive shocks of the independent variable are similar to the short-run and long-run negative shocks. The nonlinear ARDL models are presented as follows: Where, t represent time; − and + symbolizes the positive and negative shocks for oil prices. ε t is the error term with the assumption of normal distribution.
α + and α − are the long run parameter and z 1 is the vector regressor which is explained as: z t + and z t − are the positive and negative partial sums which is expressed below: Asymmetric error correction model (AECM) is as follows: where . The nonlinear ARDL framework has the same procedures as the linear ARDL proposed by Pesaran et al. (2001). The estimation of Eq. (10) the null hypothesis ρ = ∅ + = ∅ − = 0. Furthermore, in nonlinear ARDL, the Wald test is employed to find the long run coefficient by ∅ + = ∅ − as well as the short run coefficient as μ + = μ − . Lastly, the cumulative dynamic multiplier effects of z + and z − on y t is presented as follows: where; k→∞; the m þ k →α þ ; m − k →α − where the long run asymmetric α + and α − is already calculated, and thus, can be used below:

Empirical results
Before the NARDL is utilized, it is important to confirm that all the variables are not stationary at the second difference. The unit root tests results confirmed that all the variables are stationary at the first difference; therefore, the NARDL can be performed. The next step is to generate the positive and negative shocks for oil price variables; and finally, the NARDL can be performed 4 . The empirical outcome commenced by testing the unit root structures of the variables, which is described in Table 1. We applied the Meng et al. (2014) unit root test for our assessment; we also reported the results of Lee et al. (2012) test. The procedures suggested by Dawson and Strazicich (2010) are employed to decide on the ideal lag. When the variables are subjected to RALS test, there is evidence that the series are stationary at first difference. The results from the LM test reported in Table 2 are substantially similar to the output from the RALS test. It is noted that about 30% are located in the latter part of 2000s. This is not surprising given that the period largely corresponded with the US financial crisis, which triggered the worst form of economic recession in the US ever since the great depression of the 1930s. To conduct a robustness check for the stationarity tests, we further used two additional unit root tests provided by Kapetanios et al. (2003) and Kruse (2011) 5 ' 6 . The outcomes of the two tests, which are reported in Table 2, are all stationary at first differences.
Once perceiving the integration properties of the variable, we proceeded with the nonlinear ARDL test to inspect possible long-run link in the series. Therefore, the bound test for cointegration is implemented; the outcomes are revealed in Table 3. The outcomes show the F-statistics is 9.287871 for model 4 and 4.539995 for model 5, which is greater than the critical values of I0 Bound and I1 Bound at the entire levels of significance, rejecting the null hypothesis of no cointegration. Therefore, it is clear that a long-run link exists among the estimated variables in both models.
As cointegration is present, the short-run and long-run estimations of the nonlinear ARDL can be performed. The short-run and long-run estimation is obtained in Table 4 and 5; the short-run results reveal that the rise in oil prices will increase RE use, while the fall in oil prices will reduce the use of RE. Moreover, the GDP, population density and trade openness will increase RE consumption significantly. However, price index will decrease the RE consumption significantly.
The following is done to perform the long-run estimation. The results in Table 4 for model 4 were similar to the short-run outcomes as the increase in oil prices, GDP, and population increase RE use. However, the rise in price level decreases the use of RE. Moreover, the results revealed that the effect of the decrease in oil prices on RE consumption will diminish in the long run.
The long run results presented in Table 5 for model 5 shows the different outcomes as increase/decrease in oil price and GDP losses their significance in the long run. Conversely, trade openness and price level have significant positive and negative effect on RE consumption in the long run.
The results (based on model 4) revealed that the increase in oil prices (similar outcomes to Lin and Omoju 2017), GDP (similar results were found by Doytch and Narayan 2016;Rafindadi and Ozturk 2016;and Rafindadi and Ozturk 2016) and population density (similar outcome to Omri and Kahouli 2014;Azam et al. 2015) will have a permanent effect on increasing RE consumption in the US. On the other hand, the rise in the price level (both model 4 and 5) will reduce consumption of RE in the US (similar to Doytch and Narayan 2016). Moreover, the result for model 5 revealed that trade openness positively contributes to the increase in RE consumption in the short and long run (similar results were found by Omri et al. 2015). Lastly, the results show that the decrease in oil prices will affect RE consumption in the short run but its effect will diminish in the long run. Therefore, the decrease in oil price will have a permanent effect on RE consumption in the US.
Moreover, to test the reliability of model 4 and 5, we have conducted the LM test for serial correlation, heteroskedasticity test as well as normality of the model. The results for the LM test revealed that the equations accept the null hypothesis of no serial correlation and the heteroskedasticity test shows that we can accept the null hypothesis of homoscedasticity. Lastly, there is evidence for normality. Therefore, the outcomes of this study are reliable.
The stability of the model as well as the short-run and long-run multipliers of oil price increases and decreases have been evaluated. The results revealed that the models of this study are stable and it takes several years for the impact of both increases and decreases in the oil prices to be fully felt (Appendix 1 and 2).

Conclusion
This research investigated the effect of the rise and fall in oil prices on RE in the US from 1970 through 2018. To study it systematically, a nonlinear ARDL model for RE use was 5 These two tests provide for nonlinearity. The detailed information on these tests have been documented in Kapetanios et al. (2003) and Kruse (2011). 6 Consistent with the advice of Kapetanios et al. (2003), we have used the demeaned version of the test. 舃***and ** and * indicate 1% and 5% and 10% significance level. The maximum lag length is set at 2 since we are using an annual dataset. () contains the optimal lag length. The optimal lag length based on the Aikaike Information Criteria.
established utilizing GDP, the price of crude oil, population density, and price index as independent variables. The results in general revealed that the rise in crude oil price, GDP, and population density will increase RE use in the short and long run. However, the increase in price level will reduce the RE use. Moreover, the decrease in oil prices will decrease RE use in the short run and its significant effect will eventually diminish in the long run. GDP is a clear indicator of income; and thus, the increase in GDP will eventually increase the demand for RE. Therefore, with the increase in GDP, the US will have enough resources to finance, invest, and produce more RE.
The rise in population density will increase the use of RE for the US. It is evident that the use of energy for developed countries like the US is high as compared to the developing and emerging economies. As more energy consumption is required to generate economic growth, energy is anticipated to positively influence the consumption of different types of primary energy including the consumption of RE.
On crude oil prices, the outcome clearly shows that the rise in oil prices will increase RE use while the fall in crude oil   ***, **, * denote significance at 1%, 5%, and 10%. The critical values are based on the works of Lee et al. (2012) and Meng et al. (2014). The maximum lag length is set at 2 since we are using an annual dataset. () contains the optimal lag length.
prices will reduce RE. These results are on expected lines as the increase in crude oil prices will encourage the country to find cheaper energy alternatives such as RE. However, the decline in oil prices will reduce the use of RE as oil becomes a cheaper alternative to RE. It is normally recognized that RE is an alternative to crude oil. More use of RE will reduce environmental pollution. This would also help in moving closer to the objective of accessing affordable, sustainable, and modern energy as envisaged in the UN's sustainable development goals. With adequate availability of better technologies, it is usually more convenient for developed countries to switch from fossil fuels to RE. The basic economics theory of cross-price elasticities believes that there is positive relationship between the price of a substitute good and the demand for the other substitute good. Therefore, a positive influence of oil prices on RE consumption is anticipated as increasing oil prices will encourage households and business to decrease their oil consumption, patronize energy-efficient gadgets, and shift to RE sources. Lastly, the rise in price level will discourage the US population consumption demand for RE as higher price level lowers the value of the US dollar. Therefore, the increase in price level will decrease the country's purchasing power parity by reducing the consumption of goods and services including RE.

Policy Suggestions
From the outcome of this research, the present study makes the following recommendations. First, as the US thrives on higher GDP growth, the use of RE will play an important role in the long run, irrespective of the changes in the oil prices. At the same time, it should not be forgotten that the fossil fuel may exhaust in the long run, but GDP growth cannot be brought to a standstill. In this backdrop, additional policy options may be considered by the government to encourage the production and use of RE to make the GDP growth momentum even stronger. Second, the government may consider further reduction of taxes and provide more incentives to the units/sectors producing and using more RE. Third, the government many consider provision of more subsidy to the installation cost and expedite the licensing process associated with the RE production. Fourth, a mandatory RE purchase policy for the companies and a penalty to those who do not adhere to it will encourage more use, production and supply of RE at a competitive price.
Fifth, as the oil market is glutted and the crude oil prices are going down, it is the best time for the US government to impose tax on crude oil to raise the amount of funds available for RE research and production. This suggestion could enable the country to raise its energy security by reducing its dependency on imported crude oil and increase the role of RE as a prime energy source for the US in the future. Sixth, the US government needs to allocate more resources toward enhancing the development of RE, increasing its share in the US energy mix. Moreover, increasing the use of less water-intensive sources of RE (such as solar energy) in urban areas is essential to decrease water footprint in major cities of the US. Seventh, it is important to create a more efficient infrastructure that depends on more RE sources such as solar, wind energy, and a better rain water harvest to optimize energy production. Eighth, it is important to harmonize trade policies with other major RE producing countries such as China, India, Brazil, South Africa, and Japan to reduce trade barriers on goods that are related to RE. Finally, it is important for the government to charge upon its citizen lower taxes, low per unit electricity prices and other final products at a lower cost. To do so, affordable RE will play an important role in increasing the economies of scale and reducing the cost per unit of output for the industries in the long run.
In a broader context, it is important for every country in the long run to provide the final products and inputs to its consumer and producer at an affordable price; and hence, the use of RE will play important role in the long run for many other countries too. Further, similar country-specific studies should be carried out for other countries with the use of variables specified in our study and by including more variables to empirically establish their effect on the use of RE. Since, many countries are in the process of increasing the share of RE in the total energy mix, further research will be useful to them. Finally, country-specific future studies will draw country-specific suggestions, which would be useful on their policy fronts.