Introduction

In the modern era, energy consumption is an essential element of an economy and development like blood in the human body (Ziaei 2015). Economic and environmental issues such as financial crises energy and environment are core concerns of new millennium (Ziaei 2015). An increase in carbon dioxide (CO2) emission diminishes the industrial production (Mahmood et al. 2019). A major portion of economic growth comes from energy which creates the interdependence among energy consumption, financial sector, trade openness, and economic development (Hafeez et al. 2018; Nasreen and Anwar 2014). The world energy consumption has increased by 1.38% from 2013 to 2014 (WDI 2018). Asia is a populace and developing region along with rapid economic growth (Albiman and Sulong 2017). Asian economies are going to finance 8 trillion US $ in infrastructure until 2020 (Hafeez et al. 2018). Hence, it needs to explore the features of energy consumption for Asia due to many reasons. Firstly, the economic growth-energy consumption nexus is crucial to understand for developing countries to flourish and grow. Secondly, the additional energy demand is compulsory to generate and fulfil the desires level of goods and services of the fast-growing population. Thirdly, the Asian economies trace the increasing trend in economic growth which require the additional energy consumption (Hafeez et al. 2019a). Fourthly, the abundant amount of energy consumption is required to sustain the business operation and industrial growth (Hossain 2012; Hafeez et al. 2019a). Moreover, Asian countries have less the energy consumption disparity level, and it is a useful tool to diminish the environmental degradation in South Asia (Hafeez et al. 2019a). Lastly, the studies of Hafeez et al. (2018) and Mahalik et al. (2017) suggest that production level and financial development are directly linked with the environment.

The financial development is a vital pillar of economic growth to establish economic efficiency in the modern world (Hafeez et al. 2019b). The studies of Sadorsky (2010) and Hafeez et al. (2019b) define the financial development such as “an increase in financial activities through domestic credits by private, and financial banking sectors subsequently, foreign direct investment, and market capitalization.” Financial development indicators have direct linkages with energy demand and economic activities (Hafeez et al. 2019b; Sadorsky 2011). It also diminishes the information cost and enriches the return on investment by saving-investment channel to enhance the capital flow and production activities, hence, an increase in energy demand (Sadorsky 2010, 2011). The dominance of energy consumption shock is significant on stock market in Europe as compared to East Asia (Ziaei 2015). Therefore, a comprehensive and operational financial policy is obligatory regarding energy consumption to accomplish sustainable development (Kahouli 2017).

In recent times, financial development is becoming the curative instrument of economic growth and demand for energy consumption (Hafeez et al. 2019c). An increase in investment and renewable and nonrenewable energy consumption uplifts the economic growth in South Asia (Rahman and Velayutham 2020). In 2017, the world’s financial sector is growing at a pace of 5.61% (Hafeez et al. 2019a). It also affects the energy consumption and demand by various transmission mechanisms such as (i) direct channel, it provides the credits and loans to consumers on appropriate conditions which enriches the consumption level and buying more luxurious goods and uplifts the aggregate energy consumption (Hafeez et al. 2018; Sadorsky 2011); (ii) business effects, easy loans for building new production plants increase in capital goods, and business growth as result additional energy demand (Sadorsky 2010); and (iii) wealth effect, the expected profits create the investment opportunities in stock market that augment the economic growth and energy demand expansion (Ozturk and Acaravci 2013).

In the modern era, the nexus of energy consumption and financial development are getting attention from energy experts and researchers. Consequently, various studies have conducted the research works to explore the linkage including financial development and energy consumption across the globe. The available literature provides two schools of thoughts considering the influence of financial development on energy consumption. The first school of thought recommends that financial development enriches the energy consumption (Islam et al. 2013; Al-mulali et al. 2013; Al-Mulali and Sab 2012; Çoban and Topcu 2013; Mahalik et al. 2017; Xu 2012; Sadorsky 2010, 2011). For example, the study of Mahalik et al. (2017) concludes that financial development uplifts the energy consumption and environmental degradation for Saudi Arabia. Relatedly, Xu (2012) also suggests the positive marginal impacts of financial development on energy consumption in 29 Chinese provinces. Financial development indicators have a positive association with energy consumption (Tang and Tan 2014; Al-mulali et al. 2013; Al-Mulali and Sab 2012; Sadorsky 2010). The research work of Çoban and Topcu (2013) explores that financial development is inflating the energy consumption and heterogenous impact in old and new members of European Union respectively. Likewise, Sadorsky (2011) suggests that financial development by stock market turnover also inflates the energy consumption in 9 Central and Eastern Europe economies.

However, the second school of thought concludes that financial development is deflating the energy consumption evidence from the studies of Ouyang and Li (2018) for 30 Chinese provinces, Kahouli (2017) for South Mediterranean countries, Farhani and Solarin (2017) for the USA, Jalil and Feridun (2011) for China, Tamazian and Rao (2010) for transitional economies, and Mielnik and Goldemberg (2002) for 20 developing countries, respectively. An increase in advance and green technologies due to financial development is useful to optimize energy consumption and sustain a better environment quality (Ouyang and Li 2018; Yuxiang and Chen 2011). Similarly, energy intensity also reduces energy usage (Mielnik and Goldemberg 2002). A bidirectional causality also exits including energy consumption and financial development (Akhtar et al. 2016; Chtioui 2012). Lastly, the studies of Ozturk and Acaravci (2013) for Turkey and Kakar (2016) for Pakistan and Malaysia do not find evidence of relation among energy consumption and financial development.

The aforementioned available literature spotlights an inclusive linkage between energy consumption and financial development. Thus, it provokes various research questions. Some key determinants of nexus of energy consumption-financial development are not incorporated which can influence in the direct or indirect channel. The econometric approach of existing studies did not tackle the endogeneity, heterogeneity, and cross-sectional dependence issues, respectively, and provides biased and unclear estimates. Hence, the marginal impact of financial development on energy consumption is needed to re-examine. To incorporate these gaps, this study contributes to rethink and compute the nexus of energy consumption, economy, and financial development indicators in Asian economies by utilizing the robust panel estimation approach to tackle the endogeneity, heterogeneity, and cross-sectional dependence issues.

Furthermore, the innovative aspects of present study analysis can be stated as follows: (i) this is pioneer research work to scrutinize the financial development indicators, energy consumption, and economy, for Asian economies to propose suitable policy recommendation; (ii) foreign direct investment (FDI) and urbanization are incorporated in empirical models of financial development indicator energy consumption nexus to avoid the specification biasness; and (iii) robust DSUR (dynamic seemingly unrelated regression) methodology is used to evaluation the marginal impacts. Moreover, the cross-sectional dependence (CD), CIPS, and CADF panel unit root tests to investigate the CD and integration order of panel of Asian economies. While the Wester-Lund co-integration test is used to assess the presence of long-run linkage. Lastly, the causal direction is checked through the heterogenous DH (Dumitrescu and Hurlin) causality test.

Data collection and conceptual framework

Data collation

A balanced panel of 32 Asian economies is chosen from 1990 to 2017 based on data available to measure the marginal impacts of financial development indicators (FDI) on energy consumption (EC). The dataset of all under-considered variables is extracted from the World Bank database of World Development Indicators (WDI 2017). Natural logarithm of all under-considered variables is appropriated to estimate elasticities and meaningful interpretation. Twenty-five out of 33 Asian economies are participating in One Belt and Road initiative (Hafeez et al. 2018). The study contains 32 Asian economies, namely, Armenia, Azerbaijan, Bahrain, Bangladesh, Bhutan, Brunei Darussalam, China, Georgia, Indonesia, Iran, Israel, Kazakhstan, Kuwait, Kyrgyz Republic, Lebanon, Malaysia, Maldives, Mongolia, Myanmar, Nepal, Oman, Pakistan, Philippines, Qatar, Russian Federation, Saudi Arabia, Sri Lanka, Tajikistan, Turkey, United Arab Emirates, Vietnam, and Yemen. Energy consumption (EC) is quantified by total energy use (kg of oil equivalent) per capita (Liu et al. 2019; Hafeez et al. 2019a). Three indicators, domestic credit by financial (FDCF), private (FDCP), and banking (FDCB) sector subsequently, are used to the gauge financial development for Asian economies (Hafeez et al. 2019a, b). GDP at constant 2010 US$ is used to compute the economy size (Liu et al. 2019; Hafeez et al. 2018, 2019c). Urban sprawl is quantified through a ratio of the urban population to the total population (Hafeez et al. 2019b; Islam et al. 2013). By following the study of Neequaye and Oladi (2015), the foreign direct investment is calculated by net inflows as a per cent of GDP.

Conceptual framework

This research work is an exertion to explore the nexus of energy consumption and financial development for Asian economies. An established financial sector of the economy helps to reduce the borrowing cost and financial risk and create investment opportunities. It also provides incentives to investors to develop advance energy-efficient outputs. Financial development indicators are major facilitators of economic growth and international trade which encourages pollution through energy consumption mechanism (Katircioglu 2012; Hafeez et al. 2019c). It also builds the economic efficiency and strengthens the stock market and foreign direct investment (FDI) flows (Katircioglu et al. 2018). FDI empowers companies to enlarge current resources and create new plants (Neequaye and Oladi (2015). The higher rate of urbanization determines to arise in energy demand for consumer goods (Hafeez et al. 2019b; Charfeddine and Ben Khediri 2015).

The aforesaid activities encourage the energy demand for growing business and development. Hence, the financial sector of the economy has a different linkage with energy consumption. For instance, easy access loans from the banking sector will provide opportunities for consumers to buy luxuries goods (Hafeez et al. 2018; Sadorsky 2010). Investment also enhances the economic growth in South Asia (Rahman and Velayutham 2020). Luxury goods boost energy consumption and encourage the aggregate demand for energy (Sadorsky 2011). For producer perspective, financial development provides easy conditions loans to expand production units by buying and installing the new pieces of machinery and advance technology, hence, energy demand surge (Hafeez et al. 2018, 2019b). As energy use enhances the industrial production in Pakistan (Mahmood et al. 2019).

Therefore, this research work encompasses the financial development indicators to evaluate the nexus of energy consumption and financial development for Asian economies. Regarding Asia, the literature on energy consumption-financial development nexus is inadequate. Hence, to do this, the research work computes the following energy demand functions as follows:

$$ E{C}_{it}={\beta}_{0 it}+{\beta}_{1i}F{D}_{it}+{\beta}_{2i} GD{P}_{it}+{\beta}_{3i} FD{I}_{it}+{\beta}_{4i}U{S}_{it}+{e}_{it} $$
(1)

where “i” = 1,2, …..33 denotes Asian economies, “t” is the time-span, and EC, FD, GDP, FDI, and US indicate the energy consumption, financial development indicators, gross domestic product, foreign direct investment, and urban sprawl, respectively. After incorporating financial development indicators, the energy demand function is stated in Eqs. 2–4.

$$ \mathrm{Model}\ 1:\kern0.5em E{C}_{it}={\beta}_{0 it}+{\beta}_{1i} FD C{F}_{it}+{\beta}_{2i} GD{P}_{it}+{\beta}_{3i} FD{I}_{it}+{\beta}_{4i}U{S}_{it}+{e}_{1 it} $$
(2)
$$ \mathrm{Model}\ 2:\kern0.5em E{C}_{it}={\beta}_{5 it}+{\beta}_{6i} FD C{P}_{it}+{\beta}_{7i} GD{P}_{it}+{\beta}_{8i} FD{I}_{it}+{\beta}_{9i}U{S}_{it}+{e}_{2 it}, $$
(3)
$$ \mathrm{Model}\ 3:\kern0.5em E{C}_{it}={\beta}_{10 it}+{\beta}_{11i} FD C{B}_{it}+{\beta}_{12i} GD{P}_{it}+{\beta}_{13i} FD{I}_{it}+{\beta}_{14i}U{S}_{it}+{e}_{3 it}, $$
(4)

In models 1–3, FDCF, FDCP, and FDCB denote the financial development through financial sector, private sector, and banking sector subsequently.

Computational approach and results

Prerequisite, unit root, and co-integration analysis

Cross-sectional dependence (CD) analysis is a groundwork before the selection of panel unit roots tests to avoid forecasting errors (Pesaran 2004; Rauf et al. 2018; Liu et al. 2018; Yasmeen et al. 2018). Table 1 demonstrates the CD analysis and concludes that Asian economies are dependent and variables of the present study have cross-sectional dependence. Therefore, the traditional unit roots tests are not able to provide robust results. Subsequently, the second-generation panel unit root tests, CADF and CIPS, robust for heterogeneity and cross-sectional dependence, are applied to check the order of integration of financial development (FD) indicators, gross domestic product (GDP), energy consumption (EC), FDI, and urban sprawl (US). The outcome of the panel unit analysis is reported in Table 2. It infers that financial development indicators (FDCF, FDCP, FDCB), energy consumption, GDP, FDI, and urban sprawl are integrated at first difference.

Table 1 CD analysis
Full size table
Table 2 Unit root analysis
Full size table

Previously, the existing studies were employing the traditional panel co-integration tests such as Johansen, Kao, and Pedroni. But these tests do not incorporate the cross-sectional dependence and provide biased co-integration outcomes. Hence, the present research relies on Westerlund co-integration test to identify the existence of co-integration between financial development indicators (FDCF, FDCP, FDCB), energy consumption, GDP, FDI, and urban sprawl (Westerlund 2007). The Westerlund co-integration test can tackle the cross-sectional dependence which has concluded by CD analysis in Table 1. Both G and P statistics are statistically significant and infer the presence of co-integration in all models in Table 3.

Table 3 Westerlund co-integration analysis
Full size table

Marginal impacts in Asia

The prime purpose of this research is to evaluate the long-run marginal impacts of financial development indicators (FDCF, FDCP, FDCB) on Asian energy consumption by using GDP, FDI, and urban sprawl as control variables. The estimation methods of earlier studies such as panel ARDL, fixed and random effect modelling, and generalized method of moments (GMM) do not consider the cross-sectional dependence and provide biased and inconsistent estimates. Therefore, the second-generation estimator such as DSUR (dynamic seemingly unrelated co-integration regressions) is applied to compute the marginal impacts by handling the heterogeneity, biasness, and cross-sectional dependence issues (Dogan and Seker 2016). The long-run marginal impacts are reported in Table 4. The natural logarithm is applied to all under-considered variables. So the estimated magnitude of FDCF, FDCP, FDCB, GDP, FDI, and US statistically indicate the elasticities of energy consumption related to financial development (by financial, private, and banking sector), FDI, GDP, and urban sprawl, respectively. The marginal impact infers that GDP has statistically insignificant on energy consumption in Asian economies.

Table 4 DSUR results
Full size table

Columns 2, 4, and 6 designate that financial development indicators are deteriorating energy consumption. The estimates infer that a 1% increase in financial development to financial, private, and banking sectors will decline the energy consumption by 0.06, 0.06, and 0.07% subsequently. It signifies that indicators of financial development are prevailing components to control the energy demand. For instance, negative externality of energy use infers energy inefficacy in Pakistani economy (Mahmood et al. 2019). On the other hand, Ouyang and Li (2018) argue that financial development comprehensively reduces the economic growth across the 30 Chinese provinces. One possible reason for this is that Asian economies are utilizing energy-efficient production approaches and green financing. For Malaysian economy, the energy efficiency attainment is recommended to reduce energy usage through financial development (Islam et al. 2013), while energy disparity decreases environmental degradation in South Asia (Hafeez et al. 2019a). Asian economies have green technology installation and primary industries which needs less energy consumption and hence decline in energy demand (Nasreen et al. 2017). Another reason is that the introduction and implementation of environment acts and protection laws are also another reason to control energy demand for a better environment.

Regarding the nexus of energy consumption and FDI, the marginal impact of FDI on energy consumption is significantly positive and stimulates energy consumption by 0.4% in each case. It indicates that FDI is enhancing energy consumption. One potential understanding of this result is that the FDI stimulates the production level through new investment which upsurges the energy demand (Rauf et al. 2018; Hafeez et al. 2018). In Asia, most economies are emerging and tracing the development path and do not care about the energy-efficient technology usages. Concerning the link of energy consumption and urban sprawl, the estimates infer a positive and strong impact of urban sprawl on energy consumption. It signifies that a 1% increase in urban sprawl will enhance the energy consumption by 6.9, 7.1, and 7.3% subsequently in models 1–3. This outcome reflects that it is a natural phenomenon to migrate from rural to urban areas for the sake of better education, daily usage facilities, and job and life standard. The aforementioned activities need the energy to fulfil their task, hence a rise in energy demand. This result is similar to the study of Liu et al. (2017) and Mahalik and Mallick (2014).

The GDP has an insignificant impact on energy consumption in Asia. Moreover, three financial development proxies are modelled to assess estimates validity. Models 1 to 3 infer that all financial indicators have a coherent negative impact on energy consumption in Asia. It concludes that alternative measures of financial development cannot vary the impact of nature on energy consumption.

Regional marginal impacts of Asia

The long-run marginal impact of Asian region is estimated and presented in Table 4. While the regional estimates are needed to sightsee the linkage of energy demand and financial development indicators within the Asian region. Therefore, this research work also computes the economy-wise long-run impact of financial development indicators by using time series dataset of Asian economies. By following the study of Hafeez et al. (2018), the FMOLS is applied to calculate the economy-wise long-run magnitudes and reported in Table 5. It uncovers that financial development indicators have a significant adverse impact on energy consumption in 7 Asian economies, namely, Brunei Darussalam, China, Georgia, Lebanon, Mongolia, Nepal, and Russia. It indicates the green financing, renewable energy usage, and rigorous policies related to energy conservations. Financial development comprehensively reduces the economic growth across the 30 Chinese provinces (Ouyang and Li 2018). On contrary, energy consumption is enhancing due to financial development in 13 Asian economies, namely, Azerbaijan, Bhutan, Iran, Israel, Kazakhstan, Kyrgyz Republic, Oman, Pakistan, Philippines, Sri Lanka, United Arab Emirates, Vietnam, and Yemen. In these economies, it signifies that green and energy efficient are needed to control energy demand in the financial sector. Rahman and Velayutham (2020) suggest that both renewable and nonrenewable energy consumption uplift the South Asian economic growth.

Table 5 Regional marginal impacts
Full size table

The urban sprawl is increasing energy consumption in 23 Asian economies except for the Maldives. It shows that a higher degree of urban sprawl requires more energy demand to fulfil the need of growing electronic appliances, health, transportation, consumer goods, housing, and education. It also found that urban sprawl is insignificant in 7 Asian economies, namely, Brunei Darussalam, Kuwait, Myanmar, Nepal, Oman, Qatar, and United Arab Emirates.

FDI is stimulating energy consumption in China, Oman, Pakistan, Tajikistan, and Yemen. Therefore, these economies need to reconsider their investment policies based on energy efficiency. Whereas, FDI has statistically an adverse impact on energy consumption in Armenia, Bhutan, Iran, Kazakhstan, Philippines, Russia, and the United Arab Emirates. It spotlights the efficient energy policies in these economies. Lastly, GDP is significantly decreasing the energy consumption in 24 Asian economies. In advanced phases, the economic growth moderates the energy demand Zaman and Abd-el Moemen (2017). It signifies that energy consumption is decreasing due to implantation of energy-efficient technology. It also indicates that Asian economies are focusing on green and renewable energy sources by maintaining the pace of production. GDP is increasing energy consumption in 2 Asian economies such as Bangladesh and Maldives.

DH panel causality

The heterogeneous panel causality test of Dumitrescu and Hurlin (2012) is applied to examine the causal linkages among energy consumption and financial development indicators. It treats each cross section individually to compute the causality by tackling the heterogeneity issue. The DH causality test outcome is reported in Table 6. It infers a bidirectional causality among energy consumption and financially developed by financial, private, and banking sector respectively. Likewise, a bidirectional causal link is found among energy consumption and GDP, while a unidirectional causality from FDI to energy consumption is also found. Lastly, urban sprawl has a bidirectional causality with energy consumption for Asia.

Table 6 DH causality results
Full size table

Conclusion

This research work modelled the marginal impact of financial development indicator on energy consumption of Asian economies including the FDI, economic growth, and urban sprawl. The latest available dataset of 32 Asian economies is used by covering the period from 1990 to 2017. A set of second-generation econometric models are implemented to compute the empirical outcomes by solving the cross-sectional dependence issues. The long-run marginal impacts and causal linkage are estimated by DSUR and the heterogenous DH causality respectively. Serval proxies of financial developed are used to examine the reliability and validity of the estimates.

The significant outcomes of present research work can be stated as follows: (1) financial development indicators are deteriorating the energy consumption in the Asian region. Conversely, economy-wise analysis provides a heterogeneous impact of financial development indicators on energy consumption. (2) Both FDI and urban sprawl are projecting an increasing trend in Asian energy consumption. (3) A heterogenous panel bidirectional causality is found between financial development indicators and energy consumption. Relatedly, GDP and urban sprawl also have a bidirectional causality with energy consumption. Lastly, a unidirectional is observed from FDI to energy consumption.

This research work also proposes some possible policies suggestions in the direction of energy consumption based on empirical findings. The administration and policy-makers may promote the energy efficiency projects through financial development indicator. Hence, the energy demand can be controlled and sustained through green financing. Secondly, the financial development by banking, financial, and private sector, respectively, hurts energy consumption. Therefore, the financial sector of Asian economies needs to be ensured by authorities to launch the energy-efficient projects through low-interest rate debts. Progressive tax policy is a suitable option for high-level energy consumption projects. Thirdly, the FDI stimulates the production level through new investment which upsurges the energy demand. Therefore, energy consumption can be reduced through investment in green energy. Fourthly, the continuous urban sprawl accrues energy demand. Thus, the government may develop the rural area and creates more opportunities to discourage urban sprawl.