Introduction

Due to the continuous economic progress, the world is facing problems of environmental degradation (Tariq 2017; Mehmood 2020b, a). Therefore, UNFCCC has set up frameworks for cooperation to deal with the climatic problems. In the race to gain economic growth, countries are consuming more energy (Mehmood and Tariq 2020). The world is still consuming fossil fuels to increase industrial production (Mehmood 2020a). Almost 85% of energy is generated by fossil fuels (Sadorsky 2009), which is responsible of 57% carbon emissions in the world (Sen and Ganguly 2017). To achieve sustainable development, it is necessary to reduce CO2 emissions. Therefore, there is a need of the hour to use renewable energy in order to boost economic growth without harming the environment (Brown et al. 2018). Renewable energy is considered the best indicator to enhance economic growth (Bhattacharya et al. 2017; Shahzad et al. 2017; Eren et al. 2019). Renewable energy resources include wind, solar, and biomass. These sources are cost-effective and eco-friendly. Moreover, these energy resources reduce air pollution and have the potential to reduce poverty by providing access in remote areas (Kamran 2018; Gielen et al. 2019). A large number of studies have investigated the impacts of nonrenewable and renewable energy on CO2 emissions. They found negative relationship between renewable energy and CO2 emissions (Paramati et al. 2017; Dong et al. 2018; Sarkodie et al. 2019) and positive relationship with GDP but the potential of renewable energy to reduce air pollution and enhance economic growth is still questioned (Apergis et al. 2010; Adams et al. 2018). Similarly, some research studies also found the same impacts on CO2 emissions of renewable and nonrenewable energy consumption (Sarkodie and Adams 2018). The adverse effects of renewable energy on climate are due to the poor technology and transmission system (Shah et al. 2020). Therefore, hampering the adverse impacts of renewable energy can be minimized by adopting efficient technology and proper transmission systems (Mirza et al. 2009; Shahzad et al. 2017).

Developing nations are facing multiple challenges in order to enhance their economic growth. Growing economic activities require more energy consumption and developing nations are consuming more fossil fuels. The fossil fuels are depleting rapidly and its prices are fluctuating. These fluctuating prices affect the developing nations adversely. These developing nations are also facing the problems of low institutional quality (Bellakhal et al. 2019).To improve the economic and social conditions, it is necessary that government respond to the requirements of the population (Shah et al. 2020). In countries with higher institutional quality, energy use is increasing economic growth (Adams et al. 2016). Therefore, economic growth is affecting CO2 emissions differently in developing and developed world. Resultantly, the role of institutional quality towards economic growth and environmental problems is still under consideration, which needs serious attention.

Hence, to answer the abovementioned problem, this research study aims to find out the impacts of different socioeconomic factors effecting environmental pollution in South Asian countries. The regional countries are getting economic growth with dirty industries like petrochemicals and steel. According to Guardian (2012), India ranked at the 3rd position in terms of CO2 emissions from energy use, Pakistan at 33rd, Bangladesh at 57th, Sri Lanka at 90th, and Nepal at 137th in the world. Institutional quality can improve air quality by implementing strict environmental policy (Mehmood 2020b). Developing nations are facing the drastic impacts of environmental changes. In this regard, institutional quality can be an effective tool for South Asian countries to lower environmental degradation.

This study employed Cobb-Douglas production function to probe the association between institutional quality GDP and energy institutional quality nexus. The Cobb-Douglas function is extensively used to probe the different perspectives of economic growth. This study employed economic activity as a production of different factors: nonrenewable energy, renewable energy, capital formation, and institutional stability. Additionally, to analyze the nexus of energy-institutional stability-environmental quality, Diet and Rosa’s environmental function is applied. This model computes the effects of population and environmental degradation technologies (Dietz and Rosa 1997).

Despite the available literature finding the linkages between energy and environment, this research makes significant contribution to the existing literature by incorporating South Asian countries. Experiential findings can vary for different countries due to the different economic conditions. Therefore, this work will enhance the understanding of economic and environmental conditions of the South Asian countries. Secondly, this study utilizes the Diet’s environmental function to find the environmental effects of economic growth along with political and socioeconomic variables. Firstly, it is argued that sustainable development can be achieved through strong institutional quality. Secondly, this study utilizes the traditional energy and renewable energy consumption to find its impacts on CO2 emissions. Thirdly, this research utilizes the technique of ARDL to provide important policy instruments for individual South Asian countries.

After the introduction in first section, literature review is in second section. Thirds section consists of methodology and data. Results and discussions are in 4th section. Conclusion and policy instruments are in the last section.

Literature review

Energy holds a fundamental role in achieving economic growth. Renewable energy resources are best alternatives to obtain sustainable development. From the last few decades, the research community has investigated the energy environment nexus and stressed upon the usage of renewable energy. Among the earlier studies (Sadorsky 2014) investigated are the linkages between renewable energy and CO2 emissions. The author conducted a panel study for 18 developing countries over the data of 1994–2003. Their findings revealed that GDP is increasing renewable energy use in the long run. In a study by Apergis and Payne (2014), the nexus between renewable energy and CO2 emissions was investigated for OECD countries. The study confirmed the long-run association between renewable energy and CO2 emissions. Similarly, the linkages between renewable energy, economic growth, and CO2 emissions were probed by Sadorsky and Perry, (2009) for MENA countries for the period of 1975 to 2008. This panel study showed that renewable energy is important to gain economic growth without hurting the environment. Sebri and Ben-Salha (2014) validated the long-run association between trade, GDP, renewable energy and CO2 emissions for BRICS over the data of 1971 to 2010. The ARDL approach confirmed that renewable and nonrenewable energy consumption are increasing GDP in India and South Africa. Moreover, feedback causality was found between renewable, nonrenewable energy, and CO2 emissions. Renewable energy had a negative effect on CO2 emissions. Zeb et al. (2014) probed the association between poverty, natural resources, GDP, and renewable energy in SAARC countries. The co-integration tests confirmed the long-run linkages between estimated variables.

Institutional quality can play a significant role in controlling environmental pollution (Sarkodie and Adams 2018). They employed ARDL approach for data of South Africa and validated the environmental Kuznets curve. The use of renewable energy was suggested by their study. Dong et al. (2017) advocated that natural gas and renewable energy are reducing CO2 emissions and validated environmental Kuznets curve in BRICS. For a panel of 128 countries, Dong et al. (2019) depicted that renewable energy is lowering CO2 emissions in the long run. Akadiri et al. (2019) analyzed the impacts of GDP and energy consumption on CO2 emissions in South Africa. The results showed that environmental pollution was not a result of energy use and GDP. Similarly, Dong et al. (2018) analyzed the impacts of population, energy use, and economic growth on CO2 emissions in a panel of 128 economies. They found that economic growth and population are increasing CO2 emissions significantly.

On the other side, some studies found that there might be the same effects of nonrenewable and renewable energy consumption on CO2 emissions. In this regard, Farhani and Shahbaz (2014) argued that nonrenewable and renewable energy resources increase CO2 emissions in MENA countries. Same results were found by Bilgili et al. (2016) in 17 OECD countries over the period of 1997–2010. In another study, Mert and Bölük (2016) found that both kinds of energy are contributing towards CO2 emissions in 16 European countries. Moreover, the effects of renewable energy on CO2 emissions were less than the effects of nonrenewable energy. The author stated that after a threshold level, renewable energy use starts to increase CO2 emissions. They argued that the positive association between renewable energy and CO2 emissions is due to the lack of efficient technology to store renewable energy. Therefore, the positive effects of renewable energy on CO2 emissions can be controlled by efficient storing capacity and transmission technologies (Shah et al. 2020).

In the literature of resources management, the role of institutions is getting stance. In this line, Al-Mulali and Ozturk (2015) probed the effect of political stability on ecological footprints along with other variables of trade, energy use, and urbanization in 14 MENA countries. The econometric results confirmed that political stability reduces environmental pollution significantly. In another study, Bhattacharya et al. (2017) found that economic freedom and renewable energy reduce environmental pollution. Lastly, Adams and Klobodu (2017) found that bureaucratic quality and democracy hold an important role in controlling air pollution. Sharif et al. (2019) investigated the nexus of renewable and nonrenewable energy–CO2 emissions in 74 economies. They found that nonrenewable energy is degrading the environment but renewable energy use is improving the environmental quality. Sharif et al. (2017) probed the electricity generation-GDP nexus in Singapore. They found that electricity generation is increasing GDP. Sharif et al. (2020a) analyzed the nonrenewable-renewable energy use and ecological footprints in Turkey. The scholars found that renewable energy is reducing ecological footprints. Sharif et al. (2020b) investigated the impacts of renewable energy use on environmental degradation in top polluted economies. They found the positive role of renewable energy in increasing air quality. Aziz et al. (2020) investigated the role of forest area and renewable energy towards ecological footprints in Pakistan. Forest area and renewable energy are found to lower ecological footprints.

Hence, the role of renewable energy is not clear in reducing CO2 emissions. Moreover, there is very little literature available, which investigated the potential effects of institutional quality on environmental pollution in South Asia. Therefore, this study fills the gap in the available literature by investigating the energy-institutional quality-environment nexus for individual South Asian countries.

Methodology

The prime objective of this study is to investigate the nexus of renewable energy, institutional quality, and CO2 emissions for South Asian countries. For this purpose, we convert the annual data into quarterly data of 1996Q1–2019Q4. Quarterly data provides robust results (Shahbaz et al. 2013). Following previous literature, we include nonrenewable energy as a comparison of renewable energy. To find the environmental effects of different socioeconomic factors, we adopted Diet’s and Rosa’s climatic function. This function was developed by Dietz and Rosa (1997), according to which different socioeconomic factors affect environmental quality. We included institutional factor to the equation by following the research of Paramati et al. (2017):

$$ CO{2}_t={\beta}_0+{\beta}_1{REN}_t+{\beta}_2{EN}_t+{\beta}_3{INS}_t+{\beta}_4{POP}_t+{\beta}_5{G}_t+{\beta}_6{CF}_t+{\epsilon}_t $$
(1)

All variables are converted to their natural logarithmic form to avoid any problem of distributional properties. This study utilizes the latest econometric technique of ARDL. This technique provides robust results for small samples considering different lag orders. The following are the ARDL equations for long-run and short-run estimations:

$$ \mathrm{lnCO}{2}_t={\upbeta}_0+\sum \limits_{n=1}^p{\partial}_n lnCO{2}_{t-n}+\sum \limits_{o=1}^{q1}{\delta}_o{lnREN}_{t-o}+\sum \limits_{p=1}^{q2}{\varphi}_p{lnEN}_{t-p}+\sum \limits_{r=1}^{q3}{\mu}_r{lnINS}_{t-r}+\sum \limits_{s=1}^{q4}{\varnothing}_s{lnPOP}_{t-s}+\sum \limits_{u=1}^{q5}{\hbox{\pounds}}_u{lnG}_{t-u}+\sum \limits_{v=1}^{q6}{\mu}_v{lnCF}_{t-v}+{\varepsilon}_t $$
(2)
$$ \Delta {(lnCO)}_t\kern0.5em ={\beta}_0+\sum \limits_{n=1}^p{\partial}_n\Delta \left({lnCO}_{t-n}\right)+\kern0.5em \sum \limits_{o=1}^{q1}{\delta}_o\Delta \left({lnREN}_{t-o}\right)+\sum \limits_{p=1}^{q2}{\varphi}_p\Delta \left({lnEN}_{t-p}\right)+\sum \limits_{r=1}^{q3}{\mu}_r\Delta \left({lnINS}_{t-r}\right)+\sum \limits_{s=1}^{q4}{\varnothing}_s\varDelta \left({lnPOP}_{t-s}\right)+\sum \limits_{u=1}^{q5}{\hbox{\pounds}}_u\varDelta \left({lnG}_{t-s}\right)+\sum \limits_{v=1}^{q6}{\varphi}_v\Delta \Big({lnCF}_{t-v}+{\vartheta z}_{t-1}+{\varepsilon}_t $$
(3)

where lnCO2, lnREN, lnEN, lnINS, lnPOP, lnG, and lnCF represent carbon dioxide emissions, nonrenewable energy, institutional quality, population growth, GDP, and capital formation, respectively.

Results and discussion

For any econometric analysis, a suitable model selection requires to determine the order of integration for all time series data. For this purpose, this study employed unit root test with structural breaks. Table 1 is showing the results of unit root test; it can be observed that the data of Pakistan, Bangladesh, and Nepal have mixed order of integration but data for India and Sri Lanka is stationary at first difference I(1). It is also important to note that none of the variable is stationary at second difference I(2), which leads to the econometric analysis of ARDL approach.

Table 1 Unit root test
Full size table

The autoregressive distributed lag (ARDL) approach is widely used (Mehmood 2020b) and provides some advantages over other econometric methods of Johansen and Juselius (2009). Moreover, other econometric techniques utilize more than 1 equations for long-run estimation but ARDL uses only one equation. This technique is also applicable to the data of mixed order of integration (Balsalobre-Lorente et al. 2018). Moreover, this method also avoids the problems of serial correlation and endogeneity (Shah et al. 2020).

The bounds test results are in Table 2, which indicates that all variables are co-integrating at 1% level in all countries. Table 3 is reporting the results of long-run estimation for 5 South Asian countries. It is evident that economic growth is contributing towards more CO2 emissions in Bangladesh and Sri Lanka but economic growth in Nepal is reducing CO2 emissions significantly. A 1% increase in GDP is reducing 3.4055% CO2 emissions in Nepal. Institutional quality is reducing CO2 emissions in India, Sri Lanka, and Nepal. A 1% increase in institutional quality is reducing 0.1536%, 0.2039%, and 0.1805% in India Sri Lanka and Nepal, respectively. Interestingly, institutional quality is increasing CO2 emissions in Pakistan at 0.0232%. In Bangladesh, institutional quality has no effect on environmental pollution. Population growth also has mixed results for South Asian countries. For example, population growth is negatively associated with CO2 emissions in Pakistan and Bangladesh, but population growth is increasing CO2 emissions in Sri Lanka and Nepal, which means that the governments of Sri Lanka and Nepal have to initiate environmental awareness programs for clean environment. No significant relation was found for India. Similarly, nonrenewable energy consumption has also heterogeneous impacts upon CO2 emissions. Nonrenewable energy use is reducing CO2 emissions in Pakistan and Bangladesh. A 1% increase in nonrenewable energy use is reducing CO2 emissions at 0.8547% and 0.4363% in Pakistan and Bangladesh, respectively. Nonrenewable energy use is contributing towards CO2 emissions in India, Sri Lanka, and Nepal at the rate of 0.3315%, 1.4392%, and 14.4654%, respectively. These countries need to invest in renewable energy sources to improve their air quality. The impacts of renewable energy are negative for India, Bangladesh, Sri Lanka, and Nepal. No significant association was found between renewable energy use and CO2 emissions for Pakistan. Capital formation is reducing CO2 emissions is Pakistan and Sri Lanka but capital formation is increasing CO2 emissions in Bangladesh and Nepal. In India, this relationship was not significant. According to Table 4, GDP, institutional quality, and population growth are reducing CO2 emissions in Bangladesh and Sri Lanka in the short run. Renewable energy is also reducing CO2 emissions in Pakistan and Sri Lanka in the short run. Capital formation is reducing air pollution in Sri Lanka, Nepal, and Pakistan in the short run.

Table 2 Bounds test
Full size table
Table 3 Long-run estimates
Full size table
Table 4 Short-run estimates
Full size table

Conclusion and policy implications

Today energy has become a fundamental need; without energy consumption, no country can survive economically. Along with developed economies, developing countries are struggling more to achieve sustainable economic growth. Developing nations are facing the shortage of energy along with its environmental problems. Therefore, different strategies are being made to develop renewable energy for economic gains. In this line, this study attempts to investigate the impacts of nonrenewable and renewable energy and other socioeconomic factors on CO2 emissions in South Asian countries. South Asian countries have to rely mostly on traditional sources of energy consumption and they are facing the problems of institutional stability.

The annual data was converted into quarterly data of 1996Q1–2019Q4. The unit root test with structural breaks confirmed mix order of integration and further ARDL approach was applied to know the long- and short-run associations. The long-run association shows that economic growth in Bangladesh, Pakistan, and Sri Lanka is not sustainable. Nepal is achieving economic growth with reducing CO2 emissions. In India, Sri Lanka, and Nepal, institutional quality can play an important role in achieving cleaner production. In Pakistan and Bangladesh, population growth is also lowering CO2 emissions. Sri Lanka and Nepal need to introduce environmental awareness programs to achieve cleaner environment. In India, Sri Lanka, and Nepal, nonrenewable energy usage is contributing to environmental pollution and renewable energy is lowering it. These countries need to enhance the ratio of renewable energy to their industrial production units. Capital formation is also lowering CO2 emissions in Sri Lanka and Pakistan but capital formation is increasing CO2 emissions in Bangladesh and Nepal.

We suggest that these countries have opportunities to learn from one another to achieve sustainable development. These countries should increase their mutual relationship to formulate efficient environmental regulations to bring clean environment in the region. Moreover, they need to explore more renewable energy sources in order to reduce environmental pollution. At the same time, strict environmental policy implementation is compulsory for cleaner environment. Governments of these nations should also start some public awareness programs to create environmental awareness.

Despite the contribution to the current literature, this study has some limitations, which can be addressed by future studies. For example, this study incorporated different socioeconomic variables to know their impacts on environmental pollution. Future studies can incorporate other environmental polluting factors like ecological footprints, water pollution, etc. Moreover, future studies can employ other co-integration techniques to present novel policy tools.