Green Finance and Industrial Low-Carbon Transition: A Case Study on Green Economy Policy in Kazakhstan

Abstract

The transition to a low-carbon (LC) economy is a major challenge for governments around the world. This article aims to investigate the most effective market and governmental initiatives to facilitate the industrial sector’s shift to a less carbon-intensive economy. According to our analysis, the Green Economy Policy (GEP) has the potential to reduce industry carbon emissions (CEs) in some areas by promoting energy transition, rather than focusing on developing short-term reduction methods. We found that the GEP decreased pilot sites’ industrial carbon intensity (CI) by an average of 7.88%, and this reduction persisted after many robustness checks. The favorable impact of the GEP differs based on population size (large and small populations) and geographic location (eastern, central, western, northern, and southern regions). Also, it is critical to emphasize how crucial green financing (GF) is to ease the energy transition.

1. Introduction

Governments worldwide now have to deal with the issue of global warming since innovative and sustainable economic development depends on completing a low-carbon transition (LCT). As a result, various nations have set goals for becoming carbon neutral. In light of this, this paper aims to investigate the most effective market- and government-led approaches for achieving an LCT. By doing so, we hope to broaden the field of study on carbon reduction tactics, give developing nations a fresh theoretical framework for putting an LCT into practice, and assist in resolving conflicts between various current climate policies. The amount of greenhouse gas (GHG) emissions keeps rising. Due to the COVID-19 (coronavirus) pandemic, global energy-related carbon dioxide (CO₂) emissions decreased somewhat in 2020. However, in 2021, they increased again, reaching an all-time high of 36.3 gigatons (Gt) annually (Figure 1). According to models, current policies are insufficient and would likely result in a 2.6–2.9 °C increase in temperature. Countries will require ambitious packages of new policies that coordinate and catalyze the complete decarbonization of their economies if they are to fulfil their obligations and realize the goals of the Paris Agreement.

To meet the Paris Agreement’s lower thermodynamic threshold (LTTG), global GHG emissions must drastically peak, decline, and eventually reach zero by around 2060. Global CO₂ emissions from fossil fuels must reach zero by roughly 2050 [1,2].

Based on Kuramochi et al.’s work (2018), there is a need to invest in decarbonizing the energy supply, replacing fuel with electricity, and implementing efficiency measures in buildings, industry, and transportation. This also involves implementing low-carbon agricultural practices, as well as preserving and expanding forests and other natural carbon sinks [3].

The implementation of such policies will require changes in infrastructure, lifestyle, and behavior. These changes will involve transitioning from driving to using public and active transportation, developing more sustainable and livable cities, adopting plant-based and healthier diets, improving material use and recycling, implementing circular economy principles, and educating current and future workers about the green economy (GE). Based on the relationship between growth and finance, it is essential to establish sustainable sources of financing to ensure long-term economic development. Sustainable development is further supported by the implementation of green finance (GF), environmental conservation, the adoption of green technology, and the increased use of renewable energy. The Paris Agreement recognizes financing and supporting environmentally friendly technologies and practices as a crucial component in the global fight for sustainable development, among many other factors that determine environmental sustainability [4]. Furthermore, GF is a crucial financial tool to support environmentally friendly businesses and technology to bridge the gap between supply and demand and subsequently reduce harmful emissions. Financial products such as the Climate Credit Card, Small Business Administration Express Loans, green bonds (GBs), green mortgages, green commercial real estate loans, and green home equity programs are the result of this process, which involves financial intermediaries and markets [5].

Kazakhstan has made significant progress in its pursuit of green energy, with collaborations with international financial organizations. The nation has set bold objectives for addressing climate change, with the aim of reaching carbon neutrality by 2060. To achieve this, it aims to reduce its greenhouse gas emissions and shift to renewable energy, as outlined in the nation’s Green Energy Concept and Carbon Neutrality Strategy for 2060.

The main mechanisms of Kazakhstan’s transition to a “green economy” are as follows:

• The sustainable use of water resources;
• The sustainable development of agriculture;
• Reducing the energy intensity of gross domestic product;
• Development of the electric power sector;
• Improving environmental culture and business responsibility;
• The development of “green” financing.

Reaching this objective necessitates a significant reform of the nation’s energy framework. Currently, Kazakhstan depends on coal for more than 70% of its electricity production. However, the nation aims for renewable energy to account for 50% of its electricity by 2050 and 15% by 2030. Kazakhstan has achieved remarkable advancements, leading to a revision of its original 2030 goal of 10% in 2021. As of early 2024, the country had 146 green energy facilities, which comprised 59 wind farms, 45 solar power installations, 39 mini-hydroelectric stations, and 3 biogas plants, collectively generating a total capacity of 2880 megawatts. Over the past ten years, the growth of Kazakhstan’s renewable energy industry has received considerable backing from international financial institutions through long-term funding.

GF is undoubtedly the best type of funding for Kazakhstan’s economic initiatives. As the country’s primary financial coordinator, The Astana International Financial Centre (AIFC) is working to accelerate GF’s introduction in the Republic of Kazakhstan as a top priority [6].

MFCA will help promote the expansion of GF by giving businesses a platform for issuing green bonds, helping them enhance energy efficiency, decrease adverse environmental consequences, increase the efficient use of already-existing natural resources, and mitigate and adapt to the effects of climate change.

The MFCA regulator has already adopted the relevant legislative acts. This means that Kazakhstan now has a legal framework in place for the issuance of green bonds. Currently, a “green” taxonomy is being developed, which will classify sustainable and “green” project categories and economic activity types. To establish a viable market for green finance, including green bonds and green banking, an ecosystem of green taxonomy, information disclosure, financial products, and incentives is required. Kazakhstan has already developed a green and sustainable finance roadmap. The topic of green economy (GE) growth is extensively covered in both the “Kazakhstan-2050” strategy and the Concept of Transition to Green Economy of the Republic of Kazakhstan. These documents outline the economic sectors requiring “green” financing, target indicators, and the necessary level of investments. The Kazakhstan government is creating a database of green projects, regulating the carbon emissions trading system, and creating a regulatory framework for GE and CF. The CF concept, including green standards and GB instruments, is being developed by the country with the support of developed countries and international organizations. To implement the GF tool, banks and other financial institutions will assess the compliance of projects with sustainability principles, determine the presence and severity of environmental and climate risks, and apply various assessments, performance indicators and classifications.

Investment banks and financial institutions can consult sustainable finance experts when evaluating green and sustainable projects. A resource for assessment, analysis, and research on environmental, social, and governance (ESG) issues is Sustainalytics. To facilitate financing efforts, including the provision of sustainable credit, Sustainalytics offers bond issuers and borrowers the opportunity to obtain an ESG rating license.

In addition, Sustainalytics works with pension funds and other financial institutions to ensure that the projects they finance are aligned with sustainability goals and that the socially conscious and environmentally friendly GBs issued by its clients comply with the principles of the International Capital Market Association (ICMA) [6,7]. Since its establishment in 2020, the CF Center (GFC) has served as the lead executive organization for developing GF in Kazakhstan.

2. Literature Review

The 2022 report of the Intergovernmental Panel on Climate Change (IPCC) states that the ability of many people to adapt to the effects of climate change is undermined by the continuing effects of colonization. Some studies consider the former Soviet Union as an empire and its former member states as postcolonial states. According to Heathershaw (2010), postcolonial theory is essential to understanding the discourse of the nation-state and its consequences, and “the postcoloniality of the Central Asian states is an essential aspect of the survival of the nation-state today” [8].

In Central Asia, Bhavna Dave engages directly with Kazakhstan and refines it into a postcolonial idea. She also points out that a comparison with the African and Asian cases of “elites and commoners” is important, given the extent to which they accepted and internalized the linear logic of Soviet development categories, ethno-racial stereotypes, and the obsession with civilization, which she considers “insulting” [9]. This is the contradictory attitude that Central Asians have towards their postcolonial identity. It shows how the problems of the Central Asian states (and their derived discourses) can be illuminated through a postcolonial approach (the hybridity of institutions and the adaptation strategies of the subaltern) while at the same time taking into account the hostility of the subjects towards this very way of thinking.

According to Dubuisson (2020, p. 9) [10], discourses on the former Soviet region continue to influence environmental policy decision-making and public attitudes towards land protection, natural resource conservation, and management in Kazakhstan.

Aldashev et al. (2017) and Kudaibergenova (2016) in their research also consider Kazakhstan and the subsequent development of the country and the actions of the state under the prism of postcolonialism [11,12].

2.1. Measuring the LCT in Kazakhstan

Currently, three general energy transitions have been observed worldwide. The first transition saw coal replace wood as the primary energy source. The second transition involved oil replacing coal as the primary energy source. The third transition entails the global effort to switch from fossil fuels to renewable energy sources. As of 2018, fossil-fuel resources provided 80% of the world’s energy, with 36% coming from petroleum, 13.2% from coal, and 31% from natural gas [13].

Driving the LCT of energy systems requires the widespread adoption and utilization of LC technologies [14].

Reviewing highly cited ESI papers on LC technology diffusion in our literature sample, we found that researchers mainly focus on renewable energy technologies, pollution prevention technologies, and other technologies that facilitate the energy transition.

A few particular technologies regularly examined in our literature sample are shown in Figure 2.

Scholars have focused on generation and storage technologies related to the utilization of renewable energy. The renewable energy sources most widely studied in the literature are solar and wind energy, which are also considered the most promising renewable technologies to be utilized in large quantities in the near future [15,16].

In their 2016 study, Karatayev et al. examined the production of energy from both conventional and renewable resources. Conventional resources included coal, oil, natural gas, and uranium, while renewable resources included hydropower, wind, solar power, and bioenergy. The study identified barriers to the development of renewable energy, such as low electricity tariffs, transmission losses, inefficient technologies, and weak government regulatory and legal systems [17].

The relationship between energy use and pollution in Kazakhstan’s environment was examined by Raihan et al. According to their results, long-term CO₂ emissions appear to be positively and significantly impacted by the energy consumption of fossil fuels. Moreover, Kazakhstan’s environmental sustainability is negatively impacted by rising fossil-fuel energy consumption. Their empirical findings suggest that using renewable energy is crucial to reducing CO₂ emissions in Kazakhstan. It is predicted that raising Kazakhstan’s overall energy mix’s proportion of renewable energy sources will decrease the nation’s carbon emissions. These estimates show that renewable energy consumption negatively and significantly impacts CO₂ emissions [18].

Industrial structure and resource abundance [19]; urban agglomeration considering data on nighttime lighting [20]; technological innovation and industrial agglomeration [21]; green innovation [22]; GBs as LC and climate-resilient investments [23]; and green credit [24] are some of the factors affecting the LCT that have been studied in previous studies.

This is in contrast to single-factor efficiency metrics like CEs per capita (Zheng et al., 2019) [25], energy consumption per capita [26], and CEs per unit of GDP [27].

The index system approach is used in some other studies to quantify the LCT. An LC economic index reflecting seven dimensions of (1) economic development, (2) energy patterns, (3) society and life, (4) carbon and the environment, (5) urban transportation, (6) solid waste, and (7) water was created by Tan et al. (2017) using the entropy weight method. Deng and Yang (2019) used the entropy weight method to combine the five factors of resource conservation, pollution reduction, industrial upgrading, productivity growth, and sustainable development to create an industrial LCT index [28,29]. Sun et al. (2020) assessed South Asia’s performance in sustainable development by creating an indicator of sustainable development based on the three dimensions of (a) environment, (b) energy, and (c) economy [30].

Poberezhskaya et al. (2021) argue that authoritarian and semi-authoritarian developing nations offer compelling case studies due to their ability to control economic and environmental resources while seeking support to combat climate change. Despite Kazakhstan’s involvement in international climate programs at a local level, the government has historically prioritized exploiting fossil fuels and mining uranium over efforts to address climate change [31].

Koch et al. (2021) analyzed the government policies of Russia and Kazakhstan regarding renewable energy, the geopolitics of renewables, and the promotion of sustainability based on the two countries’ government policies and investments in renewable energy [32].

2.2. GF

The term “GF” describes the provision of funds for initiatives that positively impact the environment [33]. As an alternative, “climate finance” describes financial resources specifically set aside for programs that lessen global warming. The common meaning of all terms discussed here is financial resources allocated to sustainable growth [34].

Adopting a national policy across all sectors simultaneously will increase the effectiveness of GF development [35]. Every economic sector—agricultural, industrial, and service—should adhere to the same national strategy. GF must be included in ecologically sound projects to support rural regeneration. To accelerate green development, governments should aim for high environmental quality and support other industries by utilizing various financial instruments. Hence, it is believed that the characteristics of GF development are highly significant [36,37].

CF is a transformative force for ecological development in two ways. First, it promotes safe environmental practices and products among entrepreneurs. Second, it reduces pollution by replacing energy-intensive equipment with energy-efficient equipment. By making farmers aware of environmentally friendly products, CF development reduces poverty. Green and organic farming should be encouraged among farmers, and environmentally harmful ingredients should not be used in crop production [38].

The environment determines the standard of living. Pollution in all its forms is on the rise due to advanced industry, impacting all living things. Environmental degradation is a serious issue that is only getting worse, and it is evident that this toxic environment needs to be cleaned up while keeping CF fully in mind. All of these harmful environmental consequences adversely influence the ultimate objective of human existence, which is to maintain human health. There is no denying that the current environment mankind finds itself in is hostile to its hopes for a better life [39].

2.3. GF and Its LC Implications

Reducing CEs’ harmful effects on human health and the environment depends on funding clean and renewable energy projects. CF incorporates sustainable factors into a financial decision-making procedure. It funds energy-saving, resource-efficient, and greenhouse gas emission-reduction technologies to address ecological and sustainability concerns. This investigation has a noteworthy historical context [40].

This is made clear by increased funding for economically sound and environmentally friendly projects. Financial institutions have been increasing their holdings of GF investments while reducing their share of fossil fuels and other assets with high climate risk to mitigate the impact of environmental pressures and the risks associated with industrial transition. An emerging alternative asset class with enormous profit potential is GF. As investors adopt sustainable and environmentally friendly ideas, the green sector is growing, and the fossil energy sector is shrinking [41].

Consequently, research exploring the connection between GF and LC indicators indicates that GF has a practical impact on reducing CE. Nonetheless, there is still significant potential for further development in the literature on this topic. On one side, many results do not rely on causal inference techniques, which means that estimates may be skewed. Conversely, these studies primarily rely on a segment of GF or a holistic GF index as a substitute variable for GF, which hinders a precise evaluation of its effects.

This establishes a foundation for additional exploration in this paper, utilizing external shock theory and the difference-in-differences (DID) approach.

H1: 

GF directly contributes to reducing carbon emissions.

In addition to focusing on environmental concerns, CF aims to achieve both “steady growth” and “reducing pollution”. Hence, the more underdeveloped industries have to spend on environmental projects and pollution control and the greater their production capacity, the more pronounced the clean industry’s competitive advantage is, and the more stringent CF rules become. Thus, companies can alter their mode of production and boost green productivity by using CF to improve capital allocation, risk dispersion, and market oversight. According to Taghizadeh-Hesary et al. (2020), CF helps people develop an idea of what constitutes sustainable consumption, raises public awareness of environmental protection, and motivates people to purchase environmentally friendly goods—all contributing to developing environmentally inclusive societies. CF, therefore, helps to achieve a reduction in CEs [42].

Restricting finances forces high-carbon industries to look for ways to change their development trajectory. The concept of endogenous growth has allowed technological advancements to drive the growth and evolution of companies. First, high-carbon businesses use technological innovation to treat pollutants released into the atmosphere via carbon decomposition and carbon capture. Then, through technical collaboration, the clean industry can grow and accomplish reductions in carbon dioxide emissions. As per the research conducted by Sun and his colleagues (Sun et al., 2020), green technological innovation is the main driver of efforts to conserve energy, reduce emissions, and promote sustainable development.

This has been determined through a systematic or formal investigation [43].

H2: 

GF is correlated with the development of an LC economy.

Green lending can impact the establishment of an LC enterprise in four primary ways. Initially, it can supply funding for the growth of such an enterprise. Due to the substantial amount of long-term capital needed for LC industry projects, financial institutions can effectively combine capital resources to ensure enough funding to drive industry growth. Additionally, by steering industry advancements towards sustainable practices, CF can help reduce overcapacity in traditional industries and accelerate modernization by improving the supply structure of manufacturing components [44].

This can optimize the effective utilization of resources; promote investment in sustainable, eco-friendly industries; incentivize the allocation of dormant capital towards innovation; and facilitate the transition to more environmentally friendly business practices. Increased transparency on environmental impacts will urge companies to operate carbon-neutrally. To some extent, this will curb high-emission business practices, raise their environmental consciousness, and control how they operate. Increasing access to green information can boost the efficiency of urban GE by improving the city’s ability to innovate technologically and reduce industrial pollution [45].

H3: 

GF plays a beneficial role in developing an LC economy by promoting innovation in environmentally friendly technologies.

According to transaction cost theory, structural change encourages financial institutions to invest low-cost capital in green and low-carbon industries to promote industrial upgrading. CF supports industries by favoring high-value-added tertiary industries over highly polluting enterprises, based on their environmental policies, to channel social funds with structural effects. This policy ensures the improvement of industrial organization and curbs the rapid growth of low- and medium-end manufacturing sectors. Yet, CF enables the key benefit of LC environmentally friendly industries to be achieved by rearranging resources for optimal use and overcoming obstacles in both industry and geography.

The mechanism of CF encourages innovation in businesses from both internal and external perspectives. Wang, X. and Wang, Y. (2021) discovered that through internal and external incentives, as well as the capacity to foster enterprise innovation, CF could realize the exchange of capital or information among green economic subjects [46]. Based on theoretical analysis, Ma et al. (2020) concluded that the primary factor limiting the innovation of green technologies is the financing problem [47]. To foster successful green technology innovation from the standpoint of financial institutions and governments, they suggested a financial services system. Huang et al. (2021) discovered that green credit guidelines could generally stimulate corporate green technology innovation. Rather than imposing financial restrictions, green credit guidelines primarily restricted the advancement of green technology by decreasing the amount of debt financing [48].

3. Policy Background

The President of Kazakhstan signed a decree on 30 May 2013 regarding the country’s transition to a green economy [49]. From that moment on, the implementation of the project “Kazakhstan: Green Financial System” began. The project aims to bridge the gap between Kazakhstan’s financial system and GE’s needs. GF is a catalyst for the transition of Kazakhstan’s economy to a “green” path of development, namely to a GE with sustainable development. Also, Nursultan Nazarbayev addressed the issue of developing a GE in the strategy “Kazakhstan-2050” [50].

Kazakhstan is an energy-exporting country that receives stable revenues from the oil and gas sector. These primary industries are recognized as a priority for export. The primary sector has significantly contributed to the development of Kazakhstan’s national economy. In the past, fossil fuels provided almost all available energy. In 2020, coal comprised half of the overall energy supply, with natural gas (31%) and oil (18%) coming next in significance. Renewable energy currently accounts for less than 2% of the total and comprises only a small fraction (Figure 3).

Residential energy usage has surged, accounting for 33% of total final energy consumption in 2020. In 2020, it became the biggest consumer sector, surpassing the industry sector with a 32% market share. Transportation accounted for 18% of the total energy consumed, while services and other sectors comprised 16% of the energy used [51].

Kazakhstan heavily relies on extractive industries for its economy, making it vulnerable to fluctuations in the global commodity market. The country’s economy consumes a large amount of energy compared to other nations worldwide [52]. When extracting oil, aggressive methods are often used that lead to the destruction of the natural system on a large scale. There is a loss and degradation of natural capital. About 75% of the territory of Kazakhstan is at high risk of environmental destabilization due to the problems of desertification, increasing emissions of toxic substances from mobile and stationary sources, waste storage, and other negative consequences [53].

According to the “Kazakhstan 2050” strategy, Kazakhstan will shift from basic raw material exports to collaborating on energy processing and sharing new technologies. In this regard, the country’s share of oil exports will decrease to 30% by 2050. GE is based on clean and “green” technologies. Its development will help prevent the emergence of an environmental crisis that has become widespread in post-industrial countries. GE aims to rectify the economical consumption of depletable resources and employ the rational use of inexhaustible resources, including wind, tidal, water, and solar power, biofuels, and others. In turn, the development of GF will allow Kazakhstan to make such a transition. Kazakhstan ratified the UN Framework Convention on Climate Change (1992, Rio de Janeiro, Brazil) on 4 May 1995. The Kyoto Protocol, adopted in Japan in 1997 and ratified in 2009, was followed by the Paris Agreement on climate change, signed in France in 2015 and ratified in 2016. Nursultan Nazarbayev extensively discussed creating a sustainable economy in the “Kazakhstan-2050” policy. Afterwards, the President of Kazakhstan approved the decree outlining the plan for transitioning the Republic of Kazakhstan to a “GE” on 30 May 2013. By the year 2050, as outlined in the “Kazakhstan-2050” strategy, there will be a significant increase in GDP, the creation of over 500,000 jobs, the development of new industries and services, and a guarantee of high living standards for the population of the country. The concept pinpoints which economic sectors need GF, sets goals for them, and then determines the necessary investment amounts. Kazakhstan’s government is working on establishing laws and regulations for the initiatives of green economy and finance, including setting up a system for trading CO₂ emissions and compiling a database of environmentally friendly projects. International agencies and wealthy nations are offering guidance on environmentally friendly investments and assisting in creating a framework for Green Financial Systems (including GB instruments) and environmentally conscious standards. There are several types of green financing instruments: emissions trading (achieving low CEs), the Green Climate Fund (financing green projects of countries, carried out mostly through contributions from developed countries), green lending (lending to eligible green projects with a lower interest rate of repayment), and other special instruments. Green loans are the most common and effective tool for financing projects, given the ease of processing for both the lender and the loan recipient. “Green” loans offer favorable conditions for the loan recipient.

Green loans make up about 39% of the total portfolio of the Green Climate Fund. One of the most used GF instruments in world practice is GBs. GBs are debt securities utilized to repay or support new or current environmentally friendly initiatives. GB investments are mainly focused on the long term, with a projected return on investment expected in 10 to 20 years [53].

Investments in renewable energy (34%) are the most popular financing area in the global GB market. In Kazakhstan, the total volume of investments in CF as of mid-2016 amounted to EUR 364 million. Since 2006, the European Bank for Reconstruction and Development (EBRD) has allocated about EUR 1.335 billion for implementing green projects in Kazakhstan. In 2014, the EBRD signed the first large-scale project to finance and build wind power plants with the Republic of Kazakhstan [54].

GBs are a debt instrument initially launched by the UNDP in Kazakhstan in August 2020. Thanks to the efforts of the AIFC’s GF Center, a stable debt market was established in Kazakhstan, as acknowledged by the UNDP. After that, three additional sets of ESG bonds were released: two offerings from the Asian Development Bank totaling USD 32.5 million and one from Damu totaling USD 2.4 million. The amount of leverage UNDP utilizes to support the initial issuance of GBs is impressive: every dollar of donor funds allocated to subsidize GBs and showcase their viability has resulted in USD 175 being raised through bond issuances thus far [55]. In 2015, the second largest renewable energy project was signed. Kazakhstan has about 55 renewable energy facilities, with a total capacity of 336 megawatts. Most of these facilities have attracted investments of KZT 58.5 million. According to the plans of the Ministry of Energy, by the end of 2020, approximately 52 renewable energy sources with a total capacity of 2 gigawatts will be commissioned, to which investments amounting to almost KZT 1 trillion will be directed. The Green Climate Fund has approved financing for USD 110 million to implement projects in Kazakhstan under the EBRD Renewable Energy Framework Program [56,57].

In December 2016, the Advisory Council on the development of CF in Kazakhstan was formed. A website (www.greenfinance.kz accessed on 15 May 2024) was launched to develop the GF system in Kazakhstan. The collection of materials and negotiations with all interested parties took place in 2016. The following year, in 2017, we studied international experience in the field of green financing and the interest of global investors in investing in the GE of Kazakhstan. The potential and capabilities of the financial sector of Kazakhstan to implement the Paris Agreement were investigated. In June 2017, an economic and financial model was developed to estimate the demand for investments and services following Kazakhstan’s commitments. A review of the country’s financial policies and regulations was carried out, and recommendations were made to stimulate investment and services in GE. The final report was presented in September, and a seminar was organized during the EXPO 2017 exhibition [58].

Kazakhstan is attracting international sources of green financing. The EBRD is a prominent investor in Kazakhstan and carries out various specialized programs in the country. Since 2007, Kazakhstan has been awarded funds from the Global Environment Facility annually for projects related to sustainable development, environmental management, and environmental change. The field of cystic fibrosis is currently in a phase of growth and progress. International environmental organizations are dedicated to establishing guidelines for disclosing information, devising strategies, and offering practical advice on sustainable financing.

Under the President of the Republic of Kazakhstan’s Decree and the Constitutional Law of Kazakhstan, the International Financial Center “Astana” was established in the country’s capital in 2015 to regulate the financial industry. The AIFC was officially launched in 2018 after establishing a complete financial center infrastructure, creating key institutions and organizations, and implementing the required legal and regulatory framework for the commencement of operations by the first registered participants at the AIFC premises. The AIFC represents the top standards and methods found in the most prominent financial hubs around the globe, including New York, London, Hong Kong, Singapore, and Dubai. A common law system based on principles, legislation, and legal precedents from England and Wales was introduced at the AIFC site in the post-Soviet space for the first time.

To establish a strong legal and regulatory framework, over 80 AIFC laws have been implemented, including financial regulations that align with international standards and best practices recognized by the global investment community. Today, the economic center location provides opportunities for obtaining capital, a variety of financial services, and investment options. It also offers beneficial tax, visa, and labor regulations that help attract more participants to the AIFC and support its growth.

Over 1700 individuals from 70 countries, including top global investment institutions, financial corporations, and banks, have recognized the advantages and distinctiveness of the AIFC ecosystem. The current environment for offering different services and conducting business has played a key role in making Astana a top regional financial center and investment hub, attracting investors and market players globally [59].

4. Methodology

Various approaches are employed to determine the treatment effect of the GEP. Initially, a comprehensive array of time-dependent controls are implemented, encompassing factors such as degree of urbanization, economic status, research and development expenditure, energy usage, and industrial level. These controls aim to minimize uncertain effects as much as they can. Secondly, we consider the GEP to be an external shock and employ a difference-in-differences (DID) approach to remove endogeneity, allowing us to assess the impact of the GEP accurately. Third, we perform various robustness checks to validate the reliability of our benchmark estimation, including an analysis of parallel trends through an event study. Fourth, we explore variations in treatment effects using grouped regressions and examine both long-term and short-term impacts by breaking down the CI. We assembled and created a panel sample encompassing 16 regions in Kazakhstan from 2014 to 2022, using it as a basis for our empirical study, which led to the following conclusions. Initially, the GEP lowered the industrial CI in chosen areas by 7.88%, and this finding remains consistent even after conducting various robustness checks. Additionally, the adverse impact of the GEP differs depending on geographical area and population size. We add to the literature in the following four areas. Initially, we determine the impact of the GEP using a causal inference technique that can address the endogeneity issue. Additionally, existing research often relies on specific components of GF, like GBs or a GF index, as proxy variables. This approach inhibits a precise evaluation of the overall effects of GF. We examine this effect using the DID approach, which helps mitigate issues related to inconsistencies in measurement scale, thereby providing a precise evaluation of the influence of green finance. Third, our DID approach enables us to analyze the variations in policy impacts between regions where policies are implemented and those where they are not, facilitating comparisons with other climate policies. Additionally, we examine the strategies that the GEP can utilize to facilitate a LCT for businesses, discovering that advancing a sustained energy transition is the primary approach to achieve this goal.

4.1. Research Area

The world’s largest landlocked nation is Kazakhstan. With 18.7 million people, it is an upper-middle-income nation. Almaty and Nursultan (the capital) are the two largest cities in the nation. Moreover, 90% of the country’s land area is flat, except for a few tall mountains in the east and southeast. Accounting for 60% of the region’s GDP, the country is the largest economy in Central Asia and a major wheat producer and exporter (USAID, 2017).

Kazakhstan has abundant natural resources, including coal, oil, and gas. The country’s GDP growth rate is closely correlated with greenhouse gas emissions due to its high reliance on coal and oil. Most greenhouse gas emissions associated with producing mineral raw materials come from three categories of sources: cement, lime, and glass (Figure 4). Ammonia and calcium carbide are produced in the chemical industry. Cast iron, steel, blast-furnace coke, and ferroalloys (ferrochrome, ferrosilicon, ferrosilicon chrome, and ferro silicomanganese) are produced via ferrous metallurgy; aluminum, lead, and zinc are produced via non-ferrous metallurgy. The manufacturing and use of asphalt, solvents in paint products, and lubricants are examples of non-energy fuel products and solvents.

Industrial output increased prices by 4.4% between 2014 and 2023. The construction industry experienced a 9.8% growth, and the processing industry saw a 14.5% growth. At the same time, mining and quarrying saw a significant decline of 11.4%, while supplies of electricity, gas, steam, hot water, and conditioned air fell by 6.25% (Figure 5). Short-term economic indicators for 2020 suggest that the national economy has improved slightly, especially in the industrial and service sectors.

4.2. Econometric Model

The difference-in-differences (DID) method is an effective econometric technique for estimating the true impact of an intervention. When it is uncertain whether an intervention will lead to a specific outcome, the DID method involves implementing the intervention in the treatment group while excluding the control group. After the intervention, the differences in outcomes between both groups are examined. The intervention is considered effective if there is a statistically significant difference in outcomes between the treatment and control groups. DID is a quasi-experimental technique used to measure the causal effects of non-randomly selected interventions. It is widely employed in various branches of economics to assess the impact of policy changes. To use the DID method, we must ensure that we can run a version of the natural experiment. We need a clear picture of what interventions need to be carried out. We need to identify treatment and control groups. We also need information about the schedule that marks the start and end of the procedure. A timeline helps identify the characteristics of the groups before and after the intervention. Thus, at the end of the intervention, four pieces of information are available:

• Characteristics of the control group before the intervention.
• Characteristics of the treatment group before the treatment.
• Characteristics of the control group after the treatment.
• Post-treatment characteristics of the treatment group.

These four data points make it easier to see how each group changed before and after the treatment. Significant differences in the outcomes of the treatment group and the control group before and after treatment can be identified using DID techniques [60].

The following specific settings are used in the DID method to estimate the influence of GF on industrial CI accurately:

$${\mathrm{C}\mathrm{I}}_{\mathrm{p}\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{G}\mathrm{F}}_{\mathrm{p}\mathrm{t}}+{{\mathrm{X}}^{\prime }}_{\mathrm{p}\mathrm{t}}\mathsf{\eta }+{\mathsf{\lambda }}_{\mathrm{p}}+{\mathsf{\mu }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{p}\mathrm{t}}$$

(1)

where the result ${\mathrm{C}\mathrm{I}}_{\mathrm{p}\mathrm{t}}$ indicates the decoupling of industrial CI and CEs with industrial economic development for region/state p in year t.

The treatment variable ${\mathrm{G}\mathrm{F}}_{\mathrm{p}\mathrm{t}}$ is the result of the interaction between the dummy variables ${\mathrm{T}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}}_{\mathrm{p}}$ and ${\mathrm{P}\mathrm{o}\mathrm{s}\mathrm{t}}_{\mathrm{t}}$.

The value of ${\mathrm{T}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}}_{\mathrm{p}}$ is 1 if state p is part of the GEP treatment group; otherwise, it is 0. If GF is introduced after year t, the value of ${\mathrm{P}\mathrm{o}\mathrm{s}\mathrm{t}}_{\mathrm{t}}$ is 1; if not, it is 0. ${\mathrm{X}}_{\mathrm{p}\mathrm{t}}$ represents a vector of changing state-level controls that may impact industrial CI, as explained below.

The annual fixed-effects term ${\mathsf{\mu }}_{\mathrm{t}}$ captures macro shocks and environmental regulations. ${\mathrm{C}\mathrm{O}}_2$ emission limits in the green concept ${\mathsf{\epsilon }}_{\mathrm{p}\mathrm{t}}$ is the stochastic error term.

${\mathsf{\lambda }}_{\mathrm{p}}$ represents a fixed-effects variable specific to the local area, which accounts for factors that do not change over time, such as the location and size of the area. What we want to focus on is the ${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$ coefficient in Equation (1). The unbiased estimate of ${\mathsf{\alpha }}_1$ using the DID framework relies on the assumption that trends are parallel. This means that before introducing the GEP, there should be similar results between the treatment and control groups, with their developmental patterns running in parallel.

To show the parallel pre-trend, we analyze the relationship between the GEP and the DID method through an event study. This study takes place in the following setting.

$${\mathrm{C}\mathrm{I}}_{\mathrm{p}\mathrm{t}}={\mathsf{\alpha }}_0+{\sum }_{\mathrm{m}=-6}^3{\mathsf{\alpha }}_{\mathrm{m}}{\mathrm{T}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}}_{\mathrm{p}}\times {\mathrm{P}\mathrm{o}\mathrm{s}\mathrm{t}}_{{\mathrm{t}}_0}+{{\mathrm{X}}^{\prime }}_{\mathrm{p}\mathrm{t}}\mathsf{\eta }+{\mathsf{\lambda }}_{\mathrm{p}}+{\mathsf{\mu }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{p}\mathrm{t}}$$

(2)

Here, ${\mathrm{P}\mathrm{o}\mathrm{s}\mathrm{t}}_{{\mathrm{t}}_{0+\mathrm{m}}}$ is a collection of placeholders that represent the period spanning nine years before and after the year GEP implementation occurred, encompassing the three years following the implementation year. Equation (1) also applies to other configurations. To avoid complete multicollinearity, we set the base year of the regression to m = −1.

4.3. Data

Our study utilized a dataset sourced from the Office for National Statistics. We compared these sources and compiled a panel sample that included 16 regions in addition to Astana, Almaty, and Shymkent in Kazakhstan from 2014 to 2022.

4.4. Dependent Variable

We determine the Competitive Index by comparing the industrial CEs to the total industrial production value. Various techniques are available in the current literature for determining ${\mathrm{C}\mathrm{O}}_2$ emissions, resulting in a range of potential values. Hence, by employing the established IPCC methodology, we determine that

$${\mathrm{C}\mathrm{O}}_2={\sum }_{\mathrm{n}=1}^{\mathrm{N}}{\mathrm{E}}_{\mathrm{n}}\times {\mathrm{N}\mathrm{C}\mathrm{V}}_{\mathrm{n}}\times {\mathrm{C}\mathrm{E}\mathrm{F}}_{\mathrm{n}}\times \mathrm{C}\mathrm{O}\mathrm{F}\times 44/12$$

(3)

where ${\mathrm{C}\mathrm{O}}_2$ represents CE; En represents the energy consumption of a specific type of energy; ${\mathrm{N}\mathrm{C}\mathrm{V}}_{\mathrm{n}}$ is the average net calorific value of that type of energy provided by the Bureau of National Statistics; ${\mathrm{C}\mathrm{E}\mathrm{F}}_{\mathrm{n}}$ is the CE factor of that type of energy provided by the IPCC; COF is the carbon oxidation factor, which is equal to 1 according to the IPCC; and 44/12 is the molecular weight ratio of carbon dioxide to carbon. Equation (3) determines the combined carbon emissions of four energy sources: coal and its derivatives, crude oil, petroleum products, and natural gas. We can obtain the carbon intensity data by dividing the total CEs by the industrial output value. In the estimation process, the confidence interval and energy consumption were considered average values for each region.

4.5. The Treatment Variable

${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$t is the interaction of the dummy variables, ${\mathrm{T}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}}_{\mathrm{p}}$ and ${\mathrm{P}\mathrm{o}\mathrm{s}\mathrm{t}}_{\mathrm{t}}$, defined above. The treatment group includes five oblasts: Karaganda Region, Pavlodar Region, Atyrau Region, Aktobe Region and East Kazakhstan Region. The year of treatment is 2017.

4.6. Controls

The STIRPAT model has been widely used in various studies and has shown numerous applications and advancements in the energy and environment sectors [61,62]. As a result, Ehrlich and Holdren initially put forward the IPAT model, which breaks down environmental impacts into factors such as population size, affluence, and technology [63]. Afterwards, Dietz and Rosa explored the fundamentals of the IPAT model and introduced the STIRPAT model, an enhanced iteration of the IPAT model [64]. This enhanced version considers the environmental consequences of variations in factors like population, wealth, and technology. It also allows for the inclusion, adjustment, or breakdown of other influential factors needed for a particular study. Furthermore, we consult the relevant literature and incorporate two extra variables into the model. Energy consumption is one factor determined by industrial energy consumption and is evaluated using the same formula as Equation (3). The second indicator is at the industry level and is determined by the proportion of industrial production compared to GDP.

4.7. Summary Statistics

Table 1. Summary statistics.

Variable	Definition	N	Mean	Std. Dev	Min.	Max.
Urbanization level	Proportion of urban population in the total population (%)	171	2.92	2.11	0	8.78
Economic level	Per capita GDP (10,000 KZT/person)	171	3,254,528	3,195,608	0	1.92
R&D investment	Internal expenditure of research and development funds (hundred million KZT)	171	1.04	1.21	0	5.65
Industry level	Proportion of industrial output value in GDP (%)	171	0.004	0.007	0	0.045
CI	Ratio of industrial CEs to industrial output value (kg/KZT)	171	0.25	0.079	0.13	0.37
Energy consumption	Industrial energy consumption (ten thousand tons of standard coal)	171	4676.52	8843.60	1479.166	29,616.46

displays the main variables’ summary statistics. The lowest value for CI observed is 0.13. This suggests that there is a separation between industrial CEs and economic growth in certain states. Nevertheless, the typical CI value is 0.25, and the highest value observed is merely 0.37, indicating that most states have not transitioned to an LC economy yet. Additionally, the analysis of regulatory data shows that energy consumption in the industry sector is notably high, particularly given its substantial impact on GDP. To achieve industrial growth, ${\mathrm{C}\mathrm{O}}_2$ reduction is needed in Kazakhstan.

5. Empirical Results and Robustness Check

5.1. Empirical Results

To better understand the impact of the GEP on CI and explain the coefficients more meaningfully, we utilize the natural logarithm of CI and non-percentage controls such as economic level, R&D investment, and energy consumption in a regression analysis. This approach helps minimize the influence of outliers in the data on regression outcomes and prevents collinearity. The estimation results of Equation (1) can be found in

Table 2. Treatment effect of the GEP on CI.

	Log CI Coefficient/Robust Standard	Log CI Coefficient/Robust Standard	Log CI Coefficient/Robust Standard
GEP	−0.0966 (0.1088)	−0.0582 (0.1097)	0.0788 (0.1326)
Urbanization level	−0.1261 (0.0145) ***	−0.1189 (0.0149) ***	−0.1191 (0.0147) ***
Log economic level	0.4320 (0.0416) ***	0.4188 (0.0418) ***	0.4181 (0.0414) ***
Log R&D investment	0.0425 (0.0200) **	0.0847 (0.0298) **	0.0829 (0.0296) ***
Log energy consumption	0.0134 (0. 0206)	0.0132 (0.0204)	0.0096 (0.0204)
Industry level	−27.5393 (3.0307) ***	−27.2271 (3.0078) ***	−27.1164 (2.9841) ***
Oblast FE	NO	YES	YES
Year FE	NO	NO	YES
N	147	147	147
Adjusted R-squared	0.4999	0.5089	0.5168

Notes: T values calculated by robust standard errors are in parentheses. Significance: *** 1%, ** 5%.

. Columns (1) through (3) each show confidence intervals and are gradually enhanced with state-fixed effects and year-fixed effects.

Our main emphasis is on the calculated coefficient of the treatment variable ${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$. The coefficient for the ${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$ variable in column 1 is −0.0966, and it is statistically significant at the 1% confidence level. This indicates that the GEP decreased the industrial CI of the treatment area by 9.66%. This result may be skewed as we did not include fixed effects.

Next, we find that the coefficient of ${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$ in column (2) is negative while continuously adding fixed effects, and the coefficient of ${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$ in column (3) is positive. The ${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$ coefficient in column (3) is 0.0788, indicating that GEP reduced the industrial CI of the treatment zone by 7.88%. This coefficient’s numerical value is quite acceptable compared to similar studies mentioned in the literature.

The findings of the study revealed that there is a strong and statistically significant relationship between financial development, income level, and ${\mathrm{C}\mathrm{O}}_2$ emissions in Kazakhstan. A 1% rise in financial development numerically corresponds to a 0.17% increase in ${\mathrm{C}\mathrm{O}}_2$ emissions. Improved financial systems in developing countries can benefit manufacturing industries and infrastructure projects that require significant energy use by increasing access to funding. This increased access to finance could potentially boost production levels and contribute to achieving economic growth similar to that accomplished with higher levels of emissions [65].

Certain challenges in developing Kazakhstan’s environmentally friendly financial system support the significance of our calculated coefficient. However, there are no standard criteria for determining green companies, leading to differing qualifications for potential investment targets. State-owned banks predominantly invest their green loans in central and state-owned enterprises, neglecting to invest significantly in private companies. However, it is challenging to distinguish eco-friendly products due to their segregation in the design, distribution, and consumption stages. Despite a high level of awareness, there is minimal application of sustainable practices, and a comprehensive set of green standards for the entire product life cycle has yet to be established.

Figure 6 displays the study event findings of Equation (2). In this graph, the dotted line indicates a confidence interval of 95%, while the connecting line for each year shows the estimated coefficient. If the confidence interval excludes zero, it indicates a statistically significant treatment effect for the GEP in that particular year. We anticipate that, prior to implementing the GEP, the projected coefficients will be near zero, showing similar developmental patterns in both the treatment and control groups. Following the implementation of the GEP, we expect the coefficients to deviate significantly from zero, indicating the effectiveness of the GEP.

The findings shown in Figure 6 support the conclusion that the null hypothesis cannot be disregarded, as the damaging impact of the GEP on CI becomes more apparent and statistically significant following the implementation of the GEP. The current costs for issuing GBs in Kazakhstan are high, and they do not offer any specific benefits for green projects to switch to projects with lower carbon emissions. Even though small- and medium-sized businesses are crucial for driving the shift to green practices, they do not have specific incentives to encourage this transformation. A comprehensive rating system for banking institutions evaluates the entire scope of their carbon financing activities, taking into account factors such as market share, annual growth rate, financial risks, and evaluation indicators such as ${\mathrm{C}\mathrm{O}}_2$ assets, carbon green projects, and products. Footprint and project income are not yet considered. Evaluation objectives remain superficial in terms of weight but do not consider quality, making them susceptible to incentive distortions, resource mismatches, and other problems.

5.2. Robustness Check

To ensure the accuracy of our benchmark results, we carried out multiple tests to confirm their reliability. Initially, we aim to pair each treatment group sample with a corresponding control group sample using propensity score matching (PSM) in conjunction with the DID estimator. On one side, the central government chooses GEPs to be implemented in certain areas, and this selection process is biased. Alternatively, the reasons for choosing may also involve considerations such as each area’s economic and financial status. The greater the quality of the location, the higher the chances of being chosen as a reward. In this hypothetical situation, the treated area could have executed the GEP with unrealistic effectiveness, potentially leading to overestimated estimates of the policy’s impact. As a result, we randomly assigned participants to the GEP treatment group to mitigate any bias that may have been caused by self-selection. Hence, the method of adjusting the radius was employed. The information is displayed in

Table 3. PSM-DID test.

	Log CI Coefficient/Robust Standard	Log CI Coefficient/Robust Standard	Log CI Coefficient/Robust Standard
GEP	−0.1543 (0.0283) ***	−0.1460 (0.0280) ***	−0.1489 (0.0277) ***
Urbanization level	0.3974 (0.3874)	0.4065 (0.3892)	0.0465 (0.5260)
Log economic level	−1.3959 (1.4438)	−1.4112 (1.4502)	0.3906 (2.2468)
Log R&D investment	2.1636 (1.2660)	2.1457 (1.2756)	−1.2917 (3.6484)
Log energy consumption	−1.8266 (6.4155)	−1.8136 (6.2931)	−1.2917 (3.6484)
Industry level	27.9645 (54.9266)	26.1644 (55.9510)	−52.3019 (103.0025)
Oblast FE	NO	YES	YES
Year FE	NO	NO	YES
N	147	147	147
Pseudo R-Squared	0.2268	0.2445	0.2659

Notes: T values calculated by robust standard errors are in parentheses. Significance: *** 1%.

and Figure 7.

The findings in Table 3 indicate that the treatment variable still has a negative estimated coefficient, even after controlling for differences between the treatment and control groups. Additionally, Figure 7 illustrates that, before being matched, there were significant discrepancies at the industry level, urbanization level, and log-economic level, with variations ranging from 0% to 50%. Following the necessary changes, the control’s total deviation was under 10%. These findings indicate that the PSM-DID estimates are valid and not influenced by any potential self-intercept bias. Next, we conduct a kernel density estimation (KDE) to consider any chance occurrences or hidden factors that could impact the confidence interval. The main objective of this estimation is to determine the specific treatment group or to approximate the political timeframe. The findings demonstrate that the calculated coefficients of the GEP in the simulation are primarily centered around zero, marking a noteworthy departure from the benchmark estimation outcomes. This outcome demonstrates that random factors do not affect the CI benchmark estimates of the GEP. KDE is a method that enables the generation of smooth curves using data points. It can also be utilized to create data points that seem to be sourced from a particular dataset. This behavior enables the creation of basic simulations that utilize actual data to represent simulated objects. The blue line represents the estimate of the baseline distribution. The KDE algorithm utilizes a variable known as bandwidth and utilizes the controls provided to adjust the range and observe how the estimation varies. In this context, the bandwidth is a parameter that helps balance the trade-off between bias and variance in the findings by smoothing out the data. The wide bandwidth range results in a smoothly skewed density distribution with inconsistent outcomes. Our KDE has a narrow bandwidth, specifically a bandwidth of 0.1003 (Figure 8).

6. Further Discussion

Heterogeneity in Treatment Effect

In this research, the implementation zones of the GEP were chosen based on emission pollution levels across the oblasts of Kazakhstan. So, five oblasts located in central, east and west Kazakhstan comprised the treatment group. These three areas possess distinct characteristics in terms of level of development, level of resource concentration, industrial structure, and level of financial development. These distinct characteristics highlight the variety within the treatment area and effectively illustrate the variations in geographical location and population size among the zones. Therefore, we divide the sample into eastern, central, western, southern, and northern regions to cover all areas of the country (

Table 4. Treatment effect of different geographical positions.

	West	Center	East	South	North
	All Log CI (Coefficient/Robust Standard)
GEP	0.0222 (0.1319)	0.0223 (0.1315)	0.0472 (0.1307)	0.0405 (0.1212)	0.0308 (0.1228)
Urbanization level	−0.1297 (0.0152) ***	−0.1311 (0.0152) ***	−0.1244 (0.0143) ***	−0.1493 (0.0140) ***	−0.1497 (0.1441) ***
Log economic level	0.4356 (0.0421) ***	0.4382 (0.0418) ***	0.4203 (0.0414) ***	0.4523 (0.0383) ***	0.4578 (0.0390) ***
Log R&D investment	0.0405 (0.0208)	0.0309 (0.0215)	0.0600 (0.0230) **	0.0896 (0.0211) ***	0.0334 (0.01882)
Log energy consumption	0.0102 (0.0207)	0.0104 (0.0206)	0.0095 (0.0204)	0.0048 (0.0190)	0.0059 (0.0193)
Industry level	−26.4841 (3.5746) ***	−27.2709 (3.0226) ***	−29.1888 (3.1187) ***	−29.2185 (2.7997) ***	−33.0837 (3.0706) ***
Region FE	YES	YES	YES	YES	YES
Year FE	YES	YES	YES	YES	YES
N	147	147	147	147	147
Adjusted R-Squared	0.5030	0.5051	0.5150	0.5799	0.5682

Notes: T values calculated by robust standard errors are in parentheses. Significance: *** 1%, ** 5%.

). Land use and land transformation are significant in causing climate change as they stem from intricate interactions across various levels. Key factors include international agreements, global and local economic factors, government policies and politics, connections between urban and rural areas, actions within households, and local infrastructure. Researchers of social science investigate the intricacies of this complexity and provide different strategies for adapting and mitigating issues. These strategies consider the historical interactions between ecology and society and the relationships between the natural world and the social world over time [66].

Alternatively, we categorize the sample based on the average population median. The greater the population of an area, the greater the availability of resources it will have. One study utilized data from a panel of 10 African countries from 1980 to 2019. This research provides evidence that the continual rise in CEs has had a detrimental impact on the populations living in these countries, endangering the overall African population due to the lack of air pollution regulations across borders [67].

In areas with higher population density, there are fewer detached buildings with smaller living spaces per person, increasing the likelihood of using natural gas for heating compared to buildings in less-populated areas. These impacts, as well as others, cooperate to decrease CEs from urban residential areas. However, the differences in CEs between individuals living in urban and non-urban areas are more significant than the distinctions between the groups themselves. The variations in the levels of residential building CEs in the United States are not solely caused by urban location [68].

According to the information in

Table 5. Treatment effect of population size.

	High Population	Low Population
	Log CI (Coefficient/Robust Standard)
GEP	0.0194 (0.1312)	0.0194 (0.1312)
Urbanization level	−0.1345 (0.0156) ***	−0.1345 (0.0156) ***
Log economic level	0.4312 (0.4212) ***	0.4312 (0.4212) ***
Log R&D investment	0.0466 (0.0211) **	0.0466 (0.0211) **
Log energy consumption	0.0096 (0.0206)	0.0096 (0.0206)
Industry level	−27.3618 (3.0074) ***	−27.3618 (3.0074) ***
Oblast FE	YES	YES
Year FE	YES	YES
N	147	147
Adjusted R-squared	0.5080	0.5080

Notes: T values calculated by robust standard errors are in parentheses. Significance: *** 1%, ** 5%.

, the coefficients of ${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$ in columns (2) and (3) can be ascertained.

The results indicate that while the GEP significantly lowers CI in the east and south, it has no discernible impact on CI in the north, center, or west regions. Additionally, the ${\mathrm{G}\mathrm{E}\mathrm{P}}_{\mathrm{p}\mathrm{t}}$ coefficient is not found to be statistically significant in terms of population effect. This suggests that the GEP helps reduce CI in areas in both regions but does not consider population size.

7. Conclusions

Governments worldwide now recognize the importance of addressing climate change, and it has also become a popular research topic in academia. In the past, Kazakhstan’s energy-consuming development approach has resulted in the industrial sector making up a significant portion of GDP. This has resulted in problems like excess capacity and domination by a few big companies in the industry. Analysis of the efficacy with which climate policy has been implemented has been lacking. Based on this, the DID technique is used to investigate the influence of the GEP on the industrial sector’s acceptance of LC practices.

It is discovered that the GEP reduces the industrial CI of treatment zones by 7.88%, a percentage considered reasonable compared to similar studies in the existing literature. This conclusion still holds after undergoing a series of thorough tests, such as an event study analysis, a PSM-DID test, an evaluation to eliminate the influence of the carbon-trading policy, and a KED. Furthermore, the GEP has varying effects on different regions (east, central, west, south, and north) and populations of varying size (large and small). Moreover, the GEP mainly aims to promote the transition of industrial businesses in the treatment areas to an LC economy.

Instead of focusing on short-term solutions like end-of-pipe treatment for reducing emissions, it is better to prioritize the promotion of energy transition. The implications of our findings are significant for policy creation. Initially, the government should actively encourage the adoption of the GEP, incrementally raise the share of GF in overall social financing, and broaden the range of eligible enterprises and treatment regions for financing. Based on our analysis, the GEP has the potential to greatly decrease the CEs of industrial businesses and offer them financial assistance as they transition to an LC model. Hence, the government must enhance its dedication to the GEP and oversee the execution process.

Encouraging the promotion of suitable industries in the treatment areas will have a positive impact. By relieving financial limitations, the policy aims to motivate businesses to transition to LC production methods by investing in innovative technology or improving energy efficiency. Simultaneously, the regulator should implement appropriate laws and policies to address the drawbacks of financial policy, enhance the exchange of financial information between the government and businesses, and prevent capital from being diverted to non-LC activities.

Projects that result in businesses losing their value and becoming unproductive should be avoided. Additionally, it is important to highlight the significance of CF in facilitating the transition of energy structures. This will aid the industrial sector in undergoing a profound transformation in its development trajectory, ultimately leading to long-term sustainable LC growth. Our findings indicate that, rather than emphasizing short-term emission reduction technology, the GEP largely encourages energy transition, encouraging industrial enterprises in designated regions to shift towards LC practices.

For this reason, local governments need to prioritize this strategy, target businesses that use too much energy, and increase both the long-term funding options and size of these enterprises. Simultaneously, they ought to leverage elite enterprises’ solid standing and accomplishments to inspire additional businesses to adopt an LC approach. This will help increase focus on environmentally friendly financial policies and boost involvement in sustainable financial practices.

The collaboration between the government and businesses can offer improved green financing options.

Enterprises support the LCT and help create a cycle of positive impact. Also, the government should consider the diversity among regions and cities and create tailored GEPs for each specific group. Our analysis findings indicate that the impact of GEPs on industrial CI varies. Hence, the government must take into account regional characteristics and tailor CF policies accordingly when making scientific decisions by implementing different policies. For instance, various environmentally friendly investment funds should be created according to the ownership structures and characteristics of businesses in terms of size and scope. Alternatively, the government should offer environmentally friendly financial resources to encourage small businesses to adopt LC practices. This would involve implementing strict regulations to prevent these businesses from making hasty short-term decisions.