1. Introduction
The high-speed economic expansion in China in the last few decades has been accompanied by some severe problems, including resource scarcity, high energy demand, and environmental damage that needs an immediate solution. As a direct input into the industry and through the numerous services it offers, the natural environment is crucial to every economy. Minerals and fossil fuels are examples of environmental resources that directly assist in the production of commodities and services, while, on the other hand, it causes an increase in the carbon level in the air and degrades the ecosystem. The extent of environmental stress and degradation has become the topic of debate at all economic and development platforms, and it is now obvious that the current level of the influence of human activities on the environment cannot be sustained into the future without causing unacceptable harm to life on our planet. Future generations will be swamped by environmental deterioration and social collapse if we fail to transform our self-destructing economy into one that is environmentally sustainable. However, the most effective means of achieving sustainable development is to restrict economic and population expansion while simultaneously encouraging technological advancement to lessen the environmental impact of human activities.
A report by NOAA (2021) [
1] revealed that the average rise in the ocean and land temperature since 1880 has been 0.07 degrees Celsius per decade. The pace of temperature increases since 1981 has been 0.18 degrees Celsius on average, which is more than twice as in prior decades. Extreme weather results in heat waves, floods, and droughts, as well as water system disruptions, which are all caused by global warming [
2]. According to Pörtner et al. (2022) [
3], global warming will exceed 1.5 °C in the near future, several climatic disasters will become unavoidable, and humanity and ecosystems will face numerous risks as a result. People who live in areas that are extremely affected by climate change number around 3.3 to 3.6 billion. Numerous species are threatened by climate change. Actions such as controls in socioeconomic growth, susceptibility to exposure, and adaptation are required to keep rising temperatures below 1.5 °C, which will significantly cut down on the expected losses and harm from global warming to both ecosystems and humans. The main factor which is responsible for the rise in temperature and climate change is CO
2 emission. Exposure to CO
2 emissions is also linked to several health issues, many of which are inflammatory, such as respiratory acidosis. In addition, it can modify the makeup of the bones and kidneys, cause physiological and behavioral abnormalities, and also cause oxidative alterations [
4]. Reduction in CO
2 emissions will surmount these health challenges.
Since China acceded to the WTO, its economy has rapidly grown, and significant changes have been made to its industrial structure. The major goals of economic reforms in China are to change and modernize the industrial structure as the country goes through a critical phase of economic transition. In Asia, Japan is the only developed nation as, per the report WESP (2022) [
5], China is still a developing economy even though it has the second-largest economy in the world. There are 4.71 billion people living in Asia, making up around 59.76% of the world’s population, of which only China accounts for 18.47% [
6]. It is difficult for governments of Asian countries to meet the energy needs of the continent’s enormous population virtually, but it is also a worldwide issue in terms of CO
2 emissions. China is the country with the highest CO
2 emissions, contributing 33% of the world’s CO
2 emissions in 2020 (11.9 billion tonnes) [
7]. According to Dale (2021) [
8], only China accounted for 59 percent of Asia’s CO
2 emissions, so serious action needs to be taken by considering different ways. Wang et al. (2020) [
9] highlighted that China invested heavily and used a lot of fossil fuel energy until 2014. However, by the end of the same year, in an effort to move toward a greener growth path, it stopped increasing its use of fossil fuels and increased its use of renewable energy sources. Sustainable investment in the energy sector ensures sustainable development as well as mitigating pronounced levels of air pollution in order to alleviate both air pollution and the sustainability of growth [
10].
Figure 1 shows an upward trend of per capita carbon emissions in China. The graph shows a drastic increase in carbon emissions after 2001.
Over the past twenty years, information and communication technology has grown rapidly. Previous empirical studies suggested that one way to boost efficiency while lowering energy usage is to employ information technology to support economic growth. Utilizing information and communication technology to cut back on commuting will also help reduce CO
2 emissions and ultimately pollution. The internet may increase energy efficiency, and the decline in energy consumption has a favorable impact on environmental quality [
11]. The usage of ICT may increase production effectiveness and decrease material product consumption, which in turn lowers energy demand and lessens the environmental load. The ICT industry itself consumes 4–6% of all the power produced worldwide in 2020. According to Ferreboeuf et al. (2019) [
12], the carbon footprint associated with digital activities accounts for around 4% of global emissions, or 2 gigatonnes of CO
2, out of which terminals account for 66% of digital emissions, followed by networks (19%) and data centers (15%). According to Lorincz et al. (2019) [
13], by 2030, just 1.97 percent of the world’s carbon emissions will come from the ICT sector. Instead, the broad use of ICT technology will make it possible for other industries to drastically cut their carbon emissions. ICT technologies can decouple economic development from emissions increase by 2030 by reducing global CO
2 emissions by 20%, which is ten times the carbon emissions of the ICT sector as a whole (Telecom lead, 28 February 2022 “MWC 2022: ICT is not a contributor to carbon footprint, but an enabler of carbon handprints, says Huawei”).
China is undertaking serious investments in ICT, which results in boosting the country’s economy by generating income, opening up employment opportunities, and exporting goods. China’s spending on ICT infrastructure increased by 21.8 percent annually from 1978 to 2018, which is near twice the country’s spending on non-ICT infrastructure. Comparatively, China’s service sector uses ICT more than the industrial sector [
14]. Transition in ICT will not only impact the economic factors but also help the ecological aspect through resource efficiency and renewable energy. Webb (2008) [
15] asserts that the environmental benefits of ICT services may outweigh their negative effects on greenhouse gas emissions. Internet penetration in China reached 74.4 percent by June 2022, with 1.05 billion people using the internet. The Chinese government’s interest in the potential of ICT to accelerate economic growth and alter industrial methods took place as the country entered a period of an economic “New Normal.” In order to integrate ICT with the energy system, the national energy administration sponsored 55 “Internet plus” smart energy demonstration projects in 2017 [
16].
Figure 2 shows the number of internet users in millions over time as an indicator of ICT development in China. China had its first web server in 1994, and by 1997 there were 620,000 internet users. The figure shows that the number of internet users increased drastically after the year 2005. China’s household broadband penetration was just 39% in 2013, and in 2021, there were more than 1.02 billion internet users in China, which shows the fast expansion of ICT in the Chinese economy.
Understanding ICT’s environmental effects is essential for addressing climate change concerns. Global economies, communities, and the environment have all seen a significant transformation as a result of ICT. According to the study of Park et al. (2018) [
17], the use of the internet primarily has a huge impact on increasing energy efficiency and thus reducing the environmental pollution. Ulucak and Khan’s (2020) [
18] investigation explored that ICT has three different environmental effects: use effect, substitution effect, and cost effect. During the usage effect, production, distribution, and installation of ICT equipment occurs and energy consumption increases, while, on the other hand, e-waste creation and ICT equipment emit carbon and affect the environment adversely. The substitution effect prevails when ICT improves the effectiveness of existing mechanisms as well as energy production and usage patterns and results in decarbonization. The cost effect occurs when demand for products and services rises as a result of price declines due to ICT-induced technological progress. The cost effect speeds up carbon emissions and thus affects the environment negatively.
Regarding how ICT affects energy use and the environment, there is a disagreement in the literature. The impact of ICT on energy use in developing nations has only been briefly examined by researchers. To the best of our knowledge, the literature does not provide empirical evidence on the asymmetric environmental effect of ICT on sector-wise energy consumption in China. This study attempts to investigate the environmental effect of ICT on energy use in the industrial, residential, and industrial sectors of China and assesses the threshold effect of ICT on environmental quality. The empirical evidence on the association between ICT, energy consumption, and the environment in China will help policymakers to establish environmentally sustainable strategies regarding the penetration of ICT in different sectors of China.
The remaining research is as follows: The section “Literature Review” examines previous literature. The third segment of this study explains the model and methodology, while the fourth part explores the results and analysis of the empirical study. The “Conclusion and policy suggestion” section summarizes the research’s conclusions and makes policy suggestions in light of the findings.
4. Results and Discussion
The statistical overview of the variables is shown in
Table 2, which contains the average value, minimum and maximum values, and standard deviation as a measure of dispersion. Skewness and kurtosis are used to assess if the variables are normally distributed or not.
According to the statistical summary, the average carbon emission in China during the period of 1990–2021 was 6.15e+06 kt with a standard deviation of 3.17e+06. The value of skewness was 0.194, which is closer to zero, showing symmetry in the distribution, while the value of kurtosis was lower than 3. The average value of ICT was recorded as 3.54e+08 with a standard deviation of 3.58e+08. The value of skewness was 0.56, whereas the value of kurtosis was 1.874, which shows a deviation from the standard normal distribution. The average values of the regime-dependent variables RE, INE, and TE were 1.25e+07, 2.62e+07, and 6.54e+06, respectively. The values of skewness and kurtosis show that these variables are not normally distributed. The average population in China during the period under study was 1.3e+09 with a standard deviation of 8.3e+07, while the values of skewness and kurtosis show that distribution is slightly normal, and the average GDP was 4.89e+12 with a standard deviation of 4.87e+12.
The correlation analysis is given in
Table 3. Correlation measures the degree of linear association between variables. The correlation coefficient values show that CO
2 emissions are positively related to other variables of the model, and the correlation of carbon emissions with industrial energy consumption, transport energy, GDP, and population are relatively strong. ICT is positively related to all variables except population, while the value of the coefficient shows a weak relationship between them. ICT is strongly correlated with GDP, while its relationship with energy consumption is weak, which indicates the hat relationship between ICT and energy consumption is nonlinear. Residential energy consumption is uncorrelated with industrial energy consumption, GDP, and transport energy consumption, while it is highly correlated with population. Industrial energy consumption is found to be positively related to transport energy consumption as the industrial activities require mobility of labor and raw material and thus affect transportation. Industrial energy consumption is also positively related to GDP and population. Transportation energy consumption is positively related to GDP and population. GDP and population are also correlated because an increase in the population affects aggregate demand, which stimulates economic activities.
Table 4 reports the VIF values for the regressors. The lower value of VIF is 1, and there is no cap on its upper value. The rule of thumb for the low-to-moderate level of correlation is the value of VIF between 1 and 5. Since the VIF value for none of the variables exceeds 5, we conclude that a severe problem of multicollinearity does not exist.
Before moving to the estimation of the parameters of the model, the stationarity of the series under examination is first tested because we are dealing with time series data. We applied KPSS and ADF-GLS tests on the logged series to examine stationarity.
Table 5 lists the results of the unit root tests. The KPSS test’s results indicate that the null hypothesis of no unit root was accepted, however, the ADF-GLS test indicated that the null hypothesis of non-stationarity is rejected. Thus, based on the unit test results, we conclude that the variables are stationary at the level.
The result of threshold regression is reported in
Table 6. Results indicate that an increase in economic growth affects carbon emissions negatively and the value is significant at a 1 percent level of significance, which implies that environmental quality improves with the increase in economic expansion initially. This means that, over time, the benefits of economic growth for mitigating CO
2 emissions will be achieved. The coefficient of the quadratic term is positive and significant, implying degradation in the environmental quality, whereas the cubic term is carrying a negative sign, thus we have found an inverted N-shaped EKC for the Chinese economy. Our finding is in line with the result of Li et al. (2019) [
80], who explained that the N-shape of EKC is congruent with the stages of China’s economic development. Before modernization and industrialization, carbon emissions declined with economic growth; after that, with the increase in industrialization and urbanization, the Chinese economy moved towards a low to moderate economic expansion, which has resulted in an increase in carbon emissions and created other environmental issues in the country, and kept rising until the economy reached a turning point; finally, as the growth of GDP rises, carbon emissions decline and lead the economy towards sustainability. Since it seems improbable that the turning point will be achieved shortly, environmental regulations are necessary to counteract unexpected environmental impacts and hasten the advent of the turning point for the economy.
The coefficient of the population is positive and significant at a 5 percent level of significance, suggesting that with every 1 percent increase in the population, carbon emissions will increase by 2.805 percent. China’s fertility rate has been steadily declining since 1995 as a result of the family planning efforts that have been made and the policy that has been continuously implemented since then. However, despite the population structure taking a smaller and smaller share of the global population each year, China’s population was still growing. There is now less room for people to live because of the deterioration of the ecological environment and changes in how land is used. The amount of carbon dioxide produced by energy use was steadily rising along with the rising demand for energy from a huge number of people [
81]. As a result, China’s population continues to be the primary driver of its carbon emissions. It is likely that China’s population, natural resources, and environment will continue to face significant challenges in the future. Therefore, the Chinese economy needs to manage the population by putting the national family planning policy into practice and by paying special attention to the effect of population size on the environment.
The coefficient of residential energy consumption given the ICT level below the threshold value is 0.673, which is positive and significant at 10 percent, that is, a one percent increase in energy use in the residential sector causes carbon emissions to rise by 0.673 percent. On the other hand, the environmental effect of residential energy consumption moves towards deterioration given the level of ICT above the threshold level. With the increase in the use of ICT devices among households, the demand for electricity increases, which puts upward pressure on total energy demand. The invention of smartphones and portable devices including laptops and digital cameras with WiFi-enabled internet connectivity has provided accessibility to a larger range of activities [
82,
83]. Moreover, ICT uses energy, particularly when operating or installing equipment that requires electricity, thus it causes an increase in demand. ICT has facilitated many people to use social media applications such as YouTube and other websites as a source of earnings. TV and radio have been displaced by laptops and smartphones to watch movies and listen to music, which has led to an increase in ICT-related energy in the residential sector. Through the widespread adoption of E-life due to ICT advancements, activities previously carried out outside the home, for example, teleworking, online shopping, dining out, learning, and doctor consultations, are now conducted inside. This result is in line with the findings of Sadorsky (2012) [
84] and Yu et al. (2020) [
85], who observed that, in China, even after accounting for the energy savings from avoided travel, the rising popularity of online activities and services would result in a rise in household energy consumption of almost 18.1%.
Results show that, given the level of ICT below the threshold level, a 1 percent increase in the energy consumption in the transportation sector causes carbon emissions to increase by 1.273 percent; however, given the level of ICT above the threshold value, the effect of transportation energy consumption reduces to 0.785 percent. This is due to the fact that ICTs reduce the need for in-person interactions, which relieves pressure on energy usage and transportation-related activities. Our finding is in accordance with the study of Kouton (2019) [
86] and Chatti and Majeed (2022) [
87], who are of the view that the use of ICTs makes it easier to obtain travel information, use planning tools, share transportation, work from home, assess the expenses of transportation, and make payments online. In addition, since consumers may now shop online due to e-business, automobiles are used less frequently [
88]. The use of ICT in the banking, education, health, and business sectors has made it easy for individuals to obtain information without commutation, thus reducing the use of energy in the transportation sector.
The findings of this study show that a 1 percent increase in energy consumption in the industrial sector, conditional on the low level of ICT, increases carbon emissions by 1.439 percent. The large value of the coefficient compared to the coefficient value of energy consumption in other sectors reveals that energy use by the industrial sector has a significant harmful effect on the environment. However, the value of the coefficient increases to 2.269 percent given the value of ICT above the threshold level. The results are in accordance with the findings of Nižetić et al. (2020) [
89] and Taneja et al. (2021) [
90]. With the increase in digitalization, industries are also going through digital transformation. The digital revolution in the industrial sector brings with it low-cost and energy-efficient production techniques. On the other hand, increased use of ICTs themselves may cancel out the impact of ICTs improving energy efficiency because a digitalized sector sees a rise in the use of digital components. Due to rebound effects, efficiency gains from technical advancement have failed to reduce the overall environmental burden of industrial production. Additionally, a growing amount of electricity is needed for the supporting infrastructure, such as data centers, networking hardware, cooling equipment, etc. Our result is in contrast with the findings of Liu et al. (2018) [
91], who examined that, due to China’s significant investment in improving carbon performance while altering the production technique of carbon-emitting industries, China has made an achievement in reducing carbon emissions.
5. Conclusions and Policy Implications
This study examines the effect of ICT on the environmental impact of energy consumption in different sectors of the Chinese economy. Based on the findings of this study, we infer that, in the case of the residential sector, an increase in the use of ICT above the threshold level results in environmental degradation due to energy intensity. The use of smartphones and laptops is increasing among households due to social media apps and entertainment websites. The use of ICT in the industrial sector also induces an energy rebound effect, due to which efficiency gains from technical advancement have failed to reduce the overall environmental burden of industrial production. In contrast, in the transportation sector, the use of ICT reduces energy intensity because the use of ICT by the banking, education, health, and business sectors has made it easy for individuals to obtain information without commutation, thus causing a decline in carbon emissions.
Due to the fast expansion of China’s online economy, special attention must be paid to the potential impacts of ICT and the internet era on future energy requirements. Since a high level of ICT results in a rebound effect in residential and industrial energy consumption, based on the findings of the study, we recommend that the possibility of rebound effects should be given more attention in the development of policies regarding the digitalization of the residential and industrial sectors. The negligence of the ICT-induced rebound effect while making digital transformations increases the danger of endorsing ICTs as the solution to all environmental problems. Additionally, internet policy must be developed so that its effects on adoption, access, interactions, and reach are amplified in the transportation sector. The management of green and sustainable transportation can be aided by the integration of ICTs into the transportation sector in the form of transport-sharing applications, intelligent traffic controls, and eco-driving, thus lessening environmental impacts. ICT can be utilized to help the transition to a less transport-intensive lifestyle if it is accommodated by other policies aimed at reducing demand for transportation, including high taxes.
This study analyzes the environmental effects of ICT on energy consumption in different sectors of the economy using country-level data; however, future studies may conduct the same analysis using provincial data to observe the difference in the adoption and usage of ICT across the provinces.