The impact of public–private investment in energy on environmental degradation: evidence from major investment countries

The objective of the present study is to explore the impact of public–private investment in energy, foreign direct investment, urbanization, and renewable and non-renewable energy consumption on environmental degradation in major investment countries during the period 1998Q4–2018Q4. In doing so, the cross-sectional dependence test and CIPS panel unit test were employed to identify the cross-sectionally dependency and the integrational properties/stationarity among the variables. Furthermore, we opted for Westerlund (2007) panel cointegration test to check the long-run association among the variables. To achieve the short-run and long-run elasticities, we have recommended cross-sectional-autoregressive distributive lag (CS-ARDL). The study outcomes revealed that public–private partnership in energy is negatively and significantly impacting CO2 emissions in both the short run and the long run. Furthermore, foreign direct investment and urbanization are negatively related to CO2 emissions, while renewable energy is positively affected it. However, the coefficients are insignificant. Moreover, non-renewable energy has a positive and substantial influence on CO2 emissions. Lastly, study outcomes offer several policy insights to develop investment in public and private partnerships in the energy sector to reduce CO2 emissions in major investment countries.


Introduction
Energy is a critical input factor in the production function.In fact, energy is used in every stage of economic activity, including the production and consumption of goods and services.Energy, we could say, is the lifeblood of the economy (Zheng et al. 2022;Ummalla and Goyari 2021).
Furthermore, the global energy demand is continuously increasing due to an increase in population growth, urbanization, and industrialization (Ibrahiem 2020;Ummalla and Samal 2018).It is predicted that global energy consumption will increase by approximately 48% by 2040 (Paramati et al. 2018).However, global warming and climate change are the biggest challenges for attaining sustainable development in this twenty-first century.Much pressure on countries to fight against climate change and protect the environment.According to the UN COP26 climate conference, many nations have promised to reach the target of achieving net zero emissions.Furthermore, the conference and IPCC report also recognized that limiting the global average temperature to 1.5 °C requires rapid reductions in global greenhouse gas emissions (Murshed et al 2022a, b;UNCC 2021;Bhattacharya et al. 2020).
Access to electricity and energy services is becoming a crucial aspect for most countries in the world.One person among the five is living without modern energy services in the world.Nearly 2.7 billion people depend on coal, wood, charcoal, and animal wastes for cooking and heating.According to World Bank (2022), globally, 733 million people still lack access to electricity, and 2.4 billion people continue to cook with fuels that are hazardous to their health and the environment.Even though the 7th point of UN Agenda 2030 clearly indicated that "Affordable and Clean Energy ensure reliable, sustainable, modern and affordable access to energy for all."Even the Paris Agreement, article 9 states that developed countries should support developing countries in the financing of climate in various ways, including both private and public, and provide US 100 billion dollars by 2020 and additional funds by 2025 (Salman et al. 2022;Raiser et al. 2020).
We can observe the double-edged problem in the energy sector.First, many organizations are compelling developed and developing countries to reduce CO 2 emissions by promoting renewable energy use.Second, millions of people worldwide do not have access to electricity.To address climate and access to electricity, government alone cannot frame the rules and regulations.Therefore, private sector intervention is needed to tackle these global issues.
The government of any country does not have full pledged fiscal resources to develop and promote sustainable energy sources to meet the increasing demand for energy.Therefore, over the last two decades, public-private investment partnership has gained attention in the energy sector.This is significant because the need to increase energy generation capacity is expected to continue the countries' economic growth targets.The private sector can also bring technology and expertise, which can improve efficiency.Collaboration between private and public actors plays a key role in energy investment decisions.With understanding, parties from both sectors can compensate each other to their mutual advantage by sharing risk and return.Furthermore, it can be an attractive massive investment and a win-win situation for both government and private parties.Newcomb et al. (2013) proposed that decentralized energy generation involving private parties could be a game changer in future energy production.Furthermore, many research papers have suggested the role of public-private partnerships in energy in the case of reducing CO 2 emissions (Ummalla and Samal 2019; Ummalla and Samal 2018).
The significance of public-private partnerships in energy has attracted attention in the energy-growth literature.Few studies have empirically analyzed the impact of public-private partnerships in energy on CO 2 emissions.For example, Kirikkaleli et al. (2022) in Bangladesh; Akinsola et al. (2022) and Ahmad and Raza (2020) in Brazil; Cheng et al. (2021), Anwar et al. (2021), andShahbaz et al. (2020) in China; Chunling et al. (2021) in Pakistan; Kirikkaleli and Adebayo (2021) in India; and Udeagha and Ngepah (2023) in South Africa.At the panel study level, few studies are conducted by incorporating other macroeconomic variables.For instance, Adebayo et al. (2021) in East Asia and the Pacific region.However, Adebayo et al. (2021) neglected the role of FDI in the reduction of emissions.Tabash et al. (2023) in a panel of emerging and growth-leading economies.More specifically, no study examined the impact of public-private partnerships in energy on CO 2 emissions in top energy investment countries by incorporating FDI, renewable and non-renewable energy consumption.Therefore, the present paper can fill the research gap in the existing literature.
This paper contributes to the existing body of literature in two ways.First, this is the first study to explore the effects of PPP investment in the energy sector and FDI inflow as explanatory variables, including disaggregate energy use (renewable and non-renewable energy) on environmental degradation in the world's top PPP investment countries.This is a novel study of the energy-growth nexus.The advantage of this study is that the findings will be more valuable and helpful to other countries' governments and policymakers.Second, the previous studies conducted on time series.But we utilize panel data set to get a more accurate estimation of model parameters and more temporal dynamics of the relationship, which cannot be addressed by a single time series data.Only few studies, Adebayo et al. (2021) and Tabash et al. (2023) conducted on panel data with conventional econometric methodology.They have neglected to apply robust second-generation econometric techniques.Finally, therefore, the study employed several unique econometric techniques considering endogeneity, cross-sectional dependency, and heterogeneity across variables among the selected countries.Based on the CS-ARDL, our study confirmed that public-private partnership in energy significantly reduces CO 2 emissions.Furthermore, foreign direct investment and urbanization are negatively related to CO 2 emissions, while renewable energy has an insignificant positive impact on CO 2 emissions in selected countries.
The rest of the paper is structured as follows.The "Review of literature" section reviews the previous studies on public-private partnership in energy and CO 2 emissions nexus and FDI and CO 2 emissions nexus.The "Data measurement and econometric methodology" section describes the data and methodology.The "Econometric results and their interpretation" section provides the empirical results and analysis.Finally, the "Concluding remarks and policy suggestions."

Review of literature
There is a scarcity of literature on PPP in energy.Many of the studies focus solely on time series.However, based on previous research, we divided the literature review into two sections: (i) PPP in the context of energy and CO 2 emissions; and (ii) FDI and CO 2 emissions.
PPP in energy and CO 2 emissions nexus Shahbaz et al. (2020) investigated the nexus between public-private partnerships in energy and CO 2 emissions in China over the period 1984-2018.Utilizing bootstrapping autoregressive distributed lag modelling (BARDL), the study found that public-private partnerships promote CO 2 emissions in the while it reduces CO 2 emissions in the long run.However, technological innovations reduce it.Furthermore, they also revealed that FDI increases CO 2 emissions in China.Kirikkaleli and Adebayo (2021) explored whether public-private partnerships in energy reduced CO 2 emissions in India during 1990Q1 and 2015Q4.Their results showed that public-private partnerships in energy positively contributed to CO 2 emissions in the long run.Furthermore, the results documented that public-private partnership investment in energy and renewable energy substantially leads to CO 2 emissions at various levels in India.Ahmad and Raza (2020) studied the impact of public-private partnerships investment in energy and technological innovations on climate change in Brazil from 1984 to 2018.They reported that public-private partnerships investment in energy reduces environmental quality via raising CO 2 emissions.Whereas technological innovations decrease it in the long run.The results also show that FDI positively influences environmental quality by decreasing CO 2 emissions.However, public-private partnerships investment in energy mitigates CO 2 emissions, technological innovations decrease it in the short run, and FDI has no significant influence on CO 2 emissions.By employing FMOLS and DOLS tests, Khan et al. (2020) proved that public-private partnership investment in energy reduces environmental quality by raising consumption-based CO 2 emissions in China in the long run during 1990Q1-2017Q2.Whereas renewable energy consumption reduces it.Later, frequency domain causality test results revealed that public-private partnership investment in energy causes CO 2 emissions throughout the frequency.Utilizing non-parametric causality in quantiles and linear Granger causality techniques, Raza et al. (2021) examined the causal nexus between public-private partnership investment in energy and CO 2 emissions in selected developing countries over the period January 1998-December 2016.No relationship is established between public-private partnerships investment in energy and carbon emissions using the linear Granger causality test.But outcomes of the non-parametric causality in quantiles revealed that PPP investment in energy leads to CO 2 emissions for considered countries.It suggests that public-private partnerships investment in energy retards environmental quality by increasing CO 2 emissions.Anwar et al. (2021) demonstrated the asymmetric impact of PPP investment on transport CO 2 emission in China by employing the quantile ARDL approach and causality test, spanning 1990Q1-2018Q4.Their results reported that PPP investment has an insignificant impact on CO 2 emissions in lower quantiles.However, negative and significant in higher quantiles in the long run.The quantile Granger causality test results revealed a bidirectional causality between PPP in transport energy and CO 2 emissions from the transport sector.Van Song et al. (2022) analyzed the impact of PPP investment and ecological innovation on environmental abatement in the USA during 1990-2015.The quantile ARDL results showed that PPP investment in energy promotes environmental abatement via increasing CO 2 emission and haze pollution like PM2.5.However, ecological innovation and GDP square help to mitigate environmental abatement.Cheng et al. (2021) studied the role of energy productivity and public-private investment in energy in achieving carbon neutrality targets in China, spanning the period 1991Q1-2017Q4.They reported that public-private partnerships hampered carbon neutrality targets in China by realizing high carbon emissions.However, development in energy productivity, renewable energy, and technological innovation can achieve it by reducing CO 2 emissions.Furthermore, their spectral Breitung and Candelon Causality test revealed that public-private partnership leads to CO 2 emissions in China.Adebayo et al. (2021) demonstrated that renewable energy consumption and technological innovation reduce CO 2 emissions.Public-private partnership investment in energy and economic growth enhanced CO 2 emissions in East Asia and Pacific Region during 1992-2015.Later, applying a frequency domain causality test, they found that technological innovation, public-private partnership investment in energy, and renewable energy consumption cause CO 2 emissions in the long term.Using the ARDL test, Chunling et al. (2021) revealed that public-private partnership investment in energy degrades environmental sustainability in Pakistan over the period 1992-2018.Furthermore, they also showed that technological innovation, economic growth, and trade openness enhance ecological footprint both in the short and long run.Kirikkaleli et al. (2022) found that public-private partnership investment in energy has a positive impact on CO 2 emissions along with economic growth, trade openness, and FDI in Bangladesh during 1997-2019.A recent study by Tabash et al. (2023) reported that public-private partnership investment in energy along with FDI have negative impact on CO 2 emissions in a panel of emerging and growth-leading economies during 2000-2019.However, Udeagha and Ngepah (2023) found reverse results in the case of South Africa during 1960-2020.

FDI and CO 2 emissions nexus
Numerous research has been conducted around the worldwide to find the connection between FDI and CO 2 emissions.Some have argued that there exists a positive and negative relationship among these variables.For instance, Waqih et al. (2019) examined FDI and carbon emissions in SAARC Region countries from 1986 to 2014.Using panel ARDL and FMOLS tests, they found that FDI has a negative impact on CO 2 emissions while energy consumption significantly positive impact.Ahmad and Raza (2020) examined the relationship between foreign direct investment (FDI) on CO 2 emissions in Brazil during 1984-2018.According to the findings, FDI impedes environmental quality by increasing CO 2 emissions in the long run.However, in the short run, the relationship is negative and insignificant.Shahbaz et al. (2020) also found that foreign direct investment increased CO 2 emissions in China over time from 1984 to 2018.However, it had a negligible short-term impact on CO 2 emissions.Demenaa and Afesorgbor (2020) found that foreign direct investment has a negative impact on CO 2 emissions using a meta-analysis of 65 primary studies.Bakhsh et al. (2021) investigated the relationship between FDI inflows and four CO 2 emission indicators in 40 Asian economies from 1996 to 2016.The empirical results revealed that FDI inflows have a positive impact on CO 2 emissions.Overall, FDI inflows significantly improve environmental quality by lowering CO 2 emissions.Balli et al. (2021) reported that increased FDI inflows had a negative impact on air quality in Asia-Pacific Economic Cooperation (APEC) countries from 1981Q1 to 2021Q1.The panel causality test results suggested that these two variables had bidirectional causality.Kirikkaleli et al. (2022) discovered that FDI, along with public-private partnership investment in energy, economic growth, and trade openness, has a positive impact on CO 2 emissions in Bangladesh during 1997-2019.
To summarize, numerous research has been undergone around the world to explore the connection between the FDI inflows, renewable and non-renewable energy, and CO 2 emissions However, very few studies have investigated the impact of public-private partnership investment in energy on CO 2 emissions in a time series framework, along with other controlled variables.Furthermore, no studies on PPP investment in the energy sector and CO 2 emissions have been conducted using the top investment countries in a panel setup.Finally, the analysis was deliberated by taking the CS-ARDL examination during 1998Q4-2018Q4.

Data and its measurement
In this study, we have collected data on per capita CO 2 emissions, PPPE, FDI inflows, and urbanization from World Development Indicators, as well as renewable energy consumption and non-renewable energy consumption from the International Energy Agency (IEA).We have converted annual data into quarterly based on the linear interpolation method, which is highly utilized in the data frequency conversion method.This study has been conducted on top public-private investment in energy countries, namely, Argentina, Bangladesh, Brazil, China, India, Mexico, Malaysia, Peru, Philippines, Thailand, Turkey, and Vietnam during the period 1998Q4-2018Q4.The selection of data and countries is purely based on data availability.A detailed description of the variables, data measurement and sources are displayed in Table 1.

Cross-sectional dependency test
Before commencing any panel data analysis, one should check for a cross-sectional dependency test to whether there is evidence of cross-sectional dependency in the panel data model.Almost all countries are interrelated/connected and interdependent with each other due to the development of international trade policies, globalization, and international energy policies that may have spillover effects.As a result, there may be crosssectional dependence among the countries.Generally, we presume that error terms of panel data analysis are independent of each other.However, they are cross-sectionally dependent on the panel data model.Estimating the panel data analysis without considering cross-sectional dependency may generate inconsistent errors in results.Hence, we have applied the crosssectional dependency test propounded by Pesaran (2004).
The CIPS test statistic equation is represented as follows: where, CDF is the cross-sectional augmented Dickey-Fuller.

CIPS unit root test
After that, we endorsed the stationarity assessment of the variables.To check the integration properties of the variables in the panel data model, we have elected second-generation unit root test, that is CIPS unit root test, as it considers (1)  Kao (1999) and Pedroni (1999) test.

Cross-sectional ARDL test (CS-ARDL)
Once the cointegration relationship among the variables is confirmed, we move our analysis to investigate the longrun and short-run impact of independent variables on the dependent variable (per capita CO 2 emissions).Furthermore, the present study has applied a cross-sectionally augmented ARDL test in order to estimate the long-run nexus among the per capita income, public-private partnership, foreign direct investment, per capita renewable and non-renewable energy, and urbanization.There may be the possibility of cross-sectional dependency among the variables due to high dependency on international trade and globalization among the economies.Estimating the model by ignoring the interlinkages/dependency among them may generate estimation bias.Application of the traditional ARDL model in the presence of CD does not settle the issues created by the possible cross-sectional dependence error and generates inconsistent estimates (Banerjee at al. 2004;Paramati et al. 2017a, b).This technique generates consistent results because it accounts for endogeneity and non-stationarity issues and solves the cross-sectional dependence and heterogeneity problems associated with it (Zeqiraj et al. 2020).Furthermore, the CS-ARDL approach accounts for the inclusion of additional lagged cross-sectional averages of both dependent and independent variables in the model.It also addresses the problem of cross-sectional correlation in the error term.To fulfill this, we have utilized the CS-ARDL test established by Chudik and Pesaran (2015).The econometric model specification for cross-sectional augmented autoregressive distributive lag (CS-ARDL) is tracked as: where, Z denotes the averages for panel cross-sections, i.e., Z = (ΔZ t , AEV t � )' where AEV is the all-independent variables.

Cross-sectional dependency test
It is very essential to conduct a cross-sectional dependency test before commencing any of the panel data analyses.The main motive for undertaking the CD test is to identify whether there is a cross-sectional dependency among the variables.
Because in recent days, all the developing and developed countries are interrelated and interdependent with each other with the eve of economic globalization, regional connectivity and implication of several international energy policies, and so on.Therefore, there may be a high chance of crosssectional dependency among the variables.We have checked the cross-sectional dependency among the selected panel countries by using the CD test developed by Pesaran (2004).
Table 2 represents the cross-sectional dependency test results.
The null hypothesis of cross-sectionally independent is rejected in comparison to the alternative hypothesis of crosssectionally dependent.It is shown that panel countries are cross-sectionally dependent on each other.Any shocks that led to changes in selected variables of these countries may transmit to other countries.Hence, this presence of transmission effect among the countries leads to the existence of cross-sectional dependency among the variables.
Other reasons for conducting the CD test on panel data analysis include variables that are cross-sectionally dependent.In the presence of cross-sectional dependence, the implication of the first-generation panel unit root test is found to be inappropriate.As a result, we empirically verified whether variables in the selected countries are crosssectionally dependent or not. (2)

CIPS unit root test results
After confirming cross-sectional dependence among the panel countries, we applied the second-generation unit root, which accounts for the cross-sectional dependence.For this purpose, we have utilized the CIPS panel unit root test propounded by Pesaran (2007).The outcomes are reported in Table 3.The result reflects that the null hypothesis of no panel unit root test is rejected over the alternative hypothesis of the presence of the panel unit root test.We found that there is a presence of a panel unit root test.It suggests that all the variables are integrated of order I (1).It indicates that variables are non-stationary at the level.However, variables are found to be stationary after the first difference.

Panel cointegration test
Our subsequent step is to inspect the cointegration relationship among the variables after establishing stationarity among the variables.Hence, we have utilized Westerlund (2007) panel cointegration test to identify the long-run association among the per capita CO 2 emissions, public-private partnership investment in the energy sector, foreign direct investment, per capita renewable and non-renewable energy, and urbanization.The outcomes of the cointegration test are reflected in Table 4.The null hypothesis of no cointegration is rejected and accepted the alternative hypothesis of the presence of cointegration among the variables at the 1% level of significance, suggesting there is a presence of cointegration connection among the considered variables.The results from one-panel G t and one group mean statistics display the cointegration association while G a exhibits no cointegration among the underlying variables.

Outcomes of short-run estimates
Once a long-term relationship between the variables has been established, we will study the short-and long-term effects of independent variables on the dependent variables.In order to do this, we have employed the CS-ARDL test developed by Chudik and Pesaran (2015) to examine the effects of per capita income, public-private partnership in the energy sector, foreign direct investment, per capita renewable and nonrenewable energy, and urbanization on CO 2 emissions in the short and long term.The results of the CS-ARDL test are represented in Table 5.The short-run estimates revealed that public-private partnership in the energy sector has a significant negative influence on CO 2 emissions at the 10% level of significance.It indicates that a 1% increase in public-private partnership in the energy sector reduces CO 2 emissions by − 0.0022% in the short run.The increase in public-private partnership in the energy sector reduces environmental degradation by emitting CO 2 emissions.Recent rising CO 2 emissions and greenhouse gas (GHG) emissions have become serious issues for both developed and developing nations, as they worsen the environmental quality and cause numerous health issues.In order to limit CO 2 emissions, international energy agencies have imposed numerous restrictions on developed and developing countries.As a result, the governments of these countries began shifting their consumption of conventional energy sources to renewable energy sources, resulting in low carbon emissions.To achieve this goal, the government has established several policies to reduce high CO 2 emissions by shifting energy consumption from non-renewable energy sources to renewable sources, such as hydro, solar, and wind energy, and by investing funds in renewable energy sources.The funding may be impossible for the government to do on its own.Banks and private sector organizations should also step forward to take responsibility for reducing high emissions.For instance, the BRICS Energy Association and the BRICS Development Bank were established through the six BRICS Summit in 2014 in order to create the fund for infrastructure and sustainable energy development as well as the creation of an "energy policy institute" and "fuel reserve bank" for the member nations.Therefore, in order to lower CO 2 emissions and preserve sustainable energy development on a global scale, both public and private entities should step forward to engage in the energy industry through public-private collaboration.
Increased public-private partnership investment in the energy sector has the potential to reduce CO 2 emissions.Furthermore, by making adequate resources available for investment in energy sources, which leads to an increase in installed capacity and energy generation, energy efficiency and security can be achieved.Basically, the consumption of renewable energy sources reduces CO 2 emissions.Therefore, colossal investment requires in these energy sources, but a limited supply of funds may not be able to achieve the aim of low carbon emissions.Once financial security in energy sources is achieved through public and private partnership investment in the energy sector, CO 2 emissions can be reduced easily.Our empirical findings revealed that public and private partnership investment in the energy sector could lower carbon emissions in the short run.The sign and magnitude are low and significant at the 10% level for major investment countries.Because investment in public and private partnerships in energy is still less recent days and needs to rise to an adequate level to reduce CO 2 emissions, therefore, both the government and private agencies should encourage to participate each other in the investment in the energy sector to achieve the low carbon emission and sustainable economic development objectives.The outcome of empirical analysis is consistent with Shahbaz et al. (2020) for China and Ahmad and Raza (2020) for Brazil in the short run, who found that public-private partnership in the energy sector increases environmental quality by reducing CO 2 emissions, and inconsistent with Raza et al. (2021) for selected developed countries, Van Song et al. (2022) in the USA, and Chunling et al. (2021) for Pakistan who reported that no association is confirmed between the public-private investment in energy and carbon emissions, and public-private partnership in the energy sector increases CO 2 emissions, respectively.Furthermore, foreign direct investment is negatively related to CO 2 emissions.However, it is insignificant.It indicates that a 1% increase in foreign direct investment reduces CO 2 emissions by − 0.005% in the short run.This suggests that private sector investment alone is not sufficient enough to meet the resource mobilization of the energy sector through investment in energy generation and increase installation capacity.Therefore, the government should encourage to take suitable initiatives by establishing a public-private partnership in the energy sector that would attract both the public and private investors to invest funds in renewable energy sectors.Investment in these sectors by them creates low cost and a feasible atmosphere, resulting in environmentally sustainable CO 2 emissions and stable economic growth.The outcomes of this study are similar to Ahmad and Raza (2020) for Brazil and Shahbaz et al. (2020) for China in the short run.
Renewable energy is positively affected CO 2 emissions.Whereas urbanization has a negative impact on it.However, the coefficients are insignificant.Moreover, non-renewable energy has a positive influence on CO 2 emissions.It suggests that a 1% increase in non-renewable energy promotes CO 2 emissions by 0.1921% at the 1% significance level.It indicates that consumption of non-renewable energy degrades the environmental quality by emitting CO 2 emissions using coal and fossil fuel.The selection of sample countries for the analysis is basically developing countries, and most of them are highly dependent on the consumption of conventional energy sources.The consumption of these energy sources emits high CO 2 emissions.Therefore, the government should divert the consumption from conventional to non-conventional energy.The coefficient is negative and significant for ECM (−1), which indicates that the long-run equilibrium position will be restored if there is any disequilibrium takes place in the short run.It also represents a stable and long-run cointegration relationship among the variables.

Outcomes of long-run estimates
In terms of long-run outcomes, we found that public-private partnerships in the energy sector had a negative and significant impact on CO 2 emissions at the 10% level.It suggests that a 1% increase in public-private partnerships in the energy sector is responsible for a 0.0023% decrease in CO 2 emissions.Therefore, a massive investment in the energy sector through public-private partnerships helps to reduce CO 2 emissions in the long run by making adequate resources available for investment in energy sources.In the long run, the study's findings contradict those of Shahbaz et al. (2020) for China, Chunling et al. (2021) for Pakistan, and Ahmad and Raza (2020) for Brazil.However, the findings are similar to those of Anwar et al. (2021) in China.
Furthermore, non-renewable energy has a positive and significant long-term impact on CO 2 emissions.It refers to the fact that a 1% increase in non-renewable energy degrades environmental quality by increasing CO 2 emissions by 0.1347%.Renewable energy, on the other hand, benefits.Foreign direct investment and urbanization, on the other hand, are negatively related to CO 2 emissions, but the coefficients are insignificant.

Concluding remarks and policy suggestions
In the twenty-first century, one of the primary goals of most countries around the world is to achieve high economic growth rate while also ensuring environmental sustainability.Because of rising high CO 2 emissions and emissions of greenhouse gas (GHG) are the major problems of all the developing and developed countries as it leads to degrades economic prosperity and environmental quality.Therefore, international energy agencies and developed countries are imposing many restrictions on developing countries to reduce CO 2 emissions.Several studies have undergone to find the relationship between renewable and non-renewable energy on CO 2 emissions for numerous countries.However, very few studies have tried to attempt the effects of public-private investment in energy on environmental degradation.Most of the studies have been explored on the case of a single time series basis.However, none of the studies has made an effort to start the study in a panel set-up.
Therefore, the present study examined the impact of public-private investment in energy on environmental degradation for major investment countries over the period 1998Q4-2018Q4.A cross-sectional dependence test was employed to identify whether variables are cross-sectionally dependent or not, and we found that there is evidence of cross-sectional dependency among the panel countries.Furthermore, since we found cross-sectional dependency, applying the first-generation unit root test is invalid.Therefore, a second-generation panel unit root test was applied to check the integrational properties/stationarity of the variables, and the study shows that all the variables are nonstationary at the level and stationary after the first difference.Furthermore, we opted for a panel cointegration test in order to check the long-run association among the per capita CO 2 emissions, public-private partnership investment in the energy sector, foreign direct investment, per capita renewable and non-renewable energy, and urbanization.To do so, Westerlund (2007) panel cointegration test was applied and confirmed that there is evidence of a long-run association among the variables.
To achieve the short-run and long-run effects of independent variables on a dependent variable (CO 2 emissions), we have recommended cross-sectional-autoregressive distributive lag (CS-ARDL).The study outcomes revealed that public-private partnership in the energy sector is negatively and significantly impacting CO 2 emissions in the short run and long run at the 10% level of significance.Furthermore, foreign direct investment and urbanization are negatively related to CO 2 emissions, while renewable energy is positively affected.However, the coefficients are insignificant.Again, non-renewable energy has a positive influence on CO 2 emissions.
We have drawn succeeding specif ic policy implications based on our empirical results: (1) our study found that public-private partnership in the energy sector is negatively and significantly impacting CO 2 emissions.Therefore, the government should encourage public-private partnerships in clean and renewable energy sources that increase environmental quality by reducing CO 2 emissions.(2) Government should also allow massive FDI inflows to high energy-consuming and polluting industries to use renewable energy sources from.The entry of these industries can be restricted by imposing limits on the threshold level of pollution and encouraging them to adopt low-cost, advanced, and environmentally friendly technology to use in the energy sectors.(3) The consumption of non-renewable energy impedes environmental quality by increasing CO 2 emissions.Therefore, governments of these countries should divert consumption from non-renewable energy sources, such as fossil fuel and coal, towards renewable sources, like hydro, solar, and wind energy.
The present analysis is based on the effects of PPP investment in energy sectors on CO 2 emissions at the aggregate level.However, the future study can be examined by taking PPP investment in energy sectors on CO 2 emissions and economic growth at a sectoral level.Further analysis can be done by taking developing and developed countries.

Table 1
Westerlund (2007)les and data sources -sectional dependence and heterogeneity issues in the model.In contrast, the application of first-generation or conventional panel unit root tests is ineffective and does not consider CSD in the model.Therefore, we have referred CIPS unit root test byPesaran (2007).Furthermore, we have used the panel cointegration test to estimate the long-run association among the variables.For this purpose, we have opted forWesterlund (2007)panel cointegration test.It suggests the cointegration relationship among the variables if the null hypothesis of no cointegration is rejected over the alternative hypothesis of cointegration and vice versa.The application of the Westerland cointegration test provides unbiased estimates as it considers CSD and heterogeneity issues more than the other conventional panel cointegration test, namely cross

Table 3
CIPS unit root test results *** indicates the rejection of the null hypothesis at a 1% significance level