Since the industrial era, ecological inequality has caused an increase in carbon dioxide (CO2) and other Green House Gas (GHG) effects which ultimately gave rise to global environmental issues. The impact of disasters experienced by various countries, such as floods, droughts, storms, excessive global warming, deforestation and other environmental damage, motivates us to think seriously about various ways to overcome these problems. The government, business and industry actors and society in general, all have a role to play in combating environmental destruction and building a sustainable environment ((Rashid & Uddin, 2019); (Hanif et al., 2020); (Mengyao, 2020); (Akhter et al., 2021)) . Investors are also starting to consider their share ownership in companies that have a bad image due to the environmental problems they create. Likewise, banking can also indirectly influence the consequences caused by environmental polluting companies.

Forests & Finance released data in April 2022 showing investment funds and credit distributed to 23 mining companies operating in the three largest tropical forest areas in the world. These findings show that banks and investors disbursed credit amounting to 37.7 billion USD during 2016-2021, 43% of which (16 billion USD) to companies in Southeast Asia. Then 61 billion USD in the form of shares and bonds in mining companies, 55% of which (39 billion USD) is earmarked for companies in Latin America. Interestingly, these credits were received by several mining companies located in Indonesia, and even became the largest credit recipients from 2016 to 2021, namely PT. Inalum is also owned by BUMN (Forest & Finance, 2022-04-12). Like companies in general, the banking industry is also not free from its connection with environmental problems ((Trehan, 2015); (Kapoor et al., 2016)). Banks can choose to lend money to clean or dirty industries. Due to their key role in providing capital to all economic sectors, banks and other financial institutions have great leverage in the transition to a greener economy (Cui et al., 2018).

Green credit in Indonesia can be said to be a new issue in the banking industry. This is because not all banks in Indonesia have implemented and reported the development of their green credit. Several banks have reported the amount of green credit they have through Sustainable Business Credit (KUBL) which is reported in each bank's annual sustainability report. The amount of green credit for each bank tends to increase and decrease differently. Researchers assume there are several factors that can influence the amount of green credit, such as bank size, bank capital adequacy, bank operational efficiency, bank liquidity, and the number of bad loans owned by the bank.

Implementing a green credit business in a bank will require high capital. According to Zhang (2020), this capital is spent to cover costs that must be prepared by banks to implement green credit. Banks with large capital are also assumed to have a lot of human resources and adequate technological infrastructure, so they are able to run a green credit business well. The green credit business is currently a positive trend that has received support from many parties, including the government. This will cause green credit to have quite high demand. The government can also support green credit businesses through various incentives such as tax incentives for green businesses. However, in some developing countries, the government cannot always provide support through these incentives, so banks will prepare these incentives through large bank capital (He et al., 2019). This capital will also be prepared by the bank to maintain the CAR ratio while the bank provides high green credit loans. Bank capital also usually shows the size of the bank's coverage. Large banks usually have large capital.

Green credit will require quite a lot of costs. These costs include the bank's expenditure on human resource training, from the managerial level to the bank's frontline employees, related to green credit (Wanting, 2020). This is because the green credit business is a new initiative, so the mechanism for implementing green credit has not been widely implemented by developing countries, including Indonesia. Apart from that, these costs are also needed to prepare adequate technological infrastructure (Yin & Matthews, 2018). Another cost required is the cost of promoting a green credit business. Many people still don't want to switch from traditional credit and other non-green businesses, so promoting green credit will help improve the bank's reputation in the community. (Ho et al., 2019) explain that high green credit can cause low financial efficiency of a bank. This is due to the high costs that a bank needs to incur for the green credit business. However, these large costs incurred did not affect the decline in bank operating profits, thereby reducing the bank's operating efficiency ratio.

Banks that implement green credit are assumed to also have to maintain their liquidity ratios to remain stable. Credit provided by banks is funds deposited by other customers (DPK) which are short term and can be withdrawn at any time by the owner. Therefore, banks need to provide sufficient funds that can be released at any time. According to (Yasmin & Akhter, 2021), high credit but not balanced with good liquidity will allow payment defaults to occur, thereby reducing bank profits. If this happens, it will reduce a bank's reputation which will have an impact on reducing bank returns and profitability. In contrast, (Zhou et al., 2021) explains that if banks do not issue credit, banks can also lose the opportunity to gain profits from providing credit to bank customers.

However, like credit in general, green credit also has the risk of bad credit. when a loan is considered non-performing, the possibility of being repaid will be very low (Cui et al., 2018). Credit risk in green credit businesses can occur because green businesses require a longer process than businesses in general. Banks that implement green credit are required to reduce their bad credit ratio, so that they are not burdened with the credit risk that will result from green loans. However, Al-Qudah et al. (2021) explain that green credit will reduce a bank's credit risk. (Zhou et al., 2020) added that the credit risk of a bank's green credit will depend on the size and ownership of a bank. On the other hand, banks are also required to establish a good risk management system. Banks need to carry out stricter screening regarding green loans. Good risk management will lead to effective policy implementation (Islam et al., 2014). (Ho et al., 2019) also explains that banks with good risk control have a high amount of green credit.

Population and sample

The population in this research is 46 banks registered on the Indonesian Stock Exchange (IDX). Sampling was carried out using a purposive sampling technique which aims to obtain a representative sample using data in the form of Annual Financial Reports, KUBL Reports and Commercial Bank Annual Sustainability Reports that have been audited and published throughout 2019-2022.

Research variable

In this study there are 3 types of variables, namely one dependent variable return on assets (ROA), one intervening variable namely green credit (GC), four independent variables consisting of capital adequacy ratio (CAR), operating cost to operating income ratio (EIR), loan to deposit ratio (DR), and non-performing loans (NPL), as well as one control variable, namely bank size (BSZ). Following are the measurements of each variable:

Data analysis

This research will use data analysis techniques with the Panel Data Regression Model. This is because this research has a type of panel data, namely a combination of time series data and cross-section data. There are three panel data regression equation models, namely the Common Effect Model (CEM), Fixed Effect Model (FEM), and Random Effect Model (REM). To choose the best model, a chow-test will be carried out to choose between CEM and FEM, a Hausman-test to choose the best model between FEM and REM, and a Lagrang Muliplier-test. The regression equation is as follows:

Model 1:

Model 2:

Descriptive Statistics

Based on a sample of 35 green banks with a period of 4 years, a description of the data for each variable is obtained as follows:

Profitability as measured by ROA has a maximum value of 4.75% and a minimum of -14.75% with a mean of 1.09%, meaning that in general the sample banks produce sufficient profitability, although there are banks that experience quite large losses. Green credit provided by banks averages 20.14%, with a maximum of 65.72% and a minimum of 0.29%. Capital (CAR) with an average of 27.87%, bank liquidity (LDR) with an average of 84.40%, credit risk (NPL) with an average of 3.45% and operating efficiency (EIR) with an average of 89.35%.

Matrix correlation

The correlation matrix shows the correlation between variables, where the closer it is to 1, the closer it is to the value of 1, indicating there is a correlation between the variables, and if it is close to 0, it indicates there is no correlation between the research variables. Table 3 shows that the Pearson correlation is less than 0.70, meaning there is no correlation between the research variables.

Test the model

There are three models used to perform panel data regression. The three models are the Common Effect Model (CEM), Fixed Effect Model (FEM), and Random Effect Model (REM). To select the best model between CEM and FEM using the Chow-test, to select the best model between FEM and REM using the Hausman-test, and to select the best model between CEM and REM using the Lagrange Multiplier (LM)-test. The model test results are as follows:

Hypothesis test result

The results of hypothesis testing using the fixed effect model (FEM) with the dependent variables profitability and green credit are as follows:

Green Credit and bank profitability.

Based on the results of statistical tests, the significance value of Prob. The t test for the GCR variable is 0.0350 > 0.05. This shows that the GCR variable has a positive and significant effect on ROA, so that the final hypothesis in this study can be accepted. These results support several previous research results such as (Cui et al., 2018), (Zhang, 2018), (Song et al., 2019), (Yin et al., 2020), and (Yasmin & Akhter, 2021) who explain that bank green credit has a positive effect on bank profitability. The green credit business currently has a positive trend in the eyes of the Indonesian people. This is because the movement of greening in the financial sector is still not yet massive. Green credit is available and easily accessible to the public to support greening in the financial sector. It is assumed that the positive trend in recent years will influence the volume of green credit by banks, so that green credit will also affect bank profitability.

This research presents several findings, firstly, the amount of green credit distribution in Indonesia is still relatively small when compared to the total portfolio of bank credit distribution, where the average bank green credit distribution in Indonesia is only 20.14%. Second, the Indonesian government has classified MSME credit in the category of sustainable business credit or green credit in Indonesia. This classification of MSME credit becomes biased when the amount of green credit in banks in Indonesia is compared. The green credit ratio of banks that focus on providing MSME credit will appear very high when compared with green credit provided by other commercial banks. Third, of the 35 banks in Indonesia that issue green credit, there is no green credit originating from sharia commercial banks. This is a note for stakeholders, considering that several green credit categories have the same concept as maqashid sharia which is run by sharia commercial banks.

Capital adequacy, profitability and green credit.

The results of the capital adequacy hypothesis test (CAR) on green credit show a significance value of 0.0072 < 0.05. Likewise, the influence of CAR on profitability shows a p-p value of 0.0250 < 0.05. This shows that the CAR variable has a positive and significant effect on both GCR and ROA, so the hypothesis in this study is accepted. These results are in accordance with several studies from (Cui et al., 2018), (Mengyao, 2020), (Wanting, 2020), and (Huy & Loan, 2022) who explain that capital adequacy has a positive effect on green credit in banking in Vietnam. The results of this research show that to increase green credit, banks use their capital to support it. Even though green credit has greater risks, to support government programs banks still provide green credit, moreover green credit also has a positive effect on profitability.

Another result is that CAR has an effect on profitability, meaning that the amount of CAR is used by banks to support the credit provided so that it can increase bank profitability. This supports the research results of (Zhang, 2018), and (Hananto & Amijaya, 2021), who found that bank capital adequacy has a positive effect on bank profitability. CAR shows the amount of funds owned by shareholders which are managed by management and the bank. In this case, the bank is responsible for increasing shareholder value by providing good financial performance. The greater the managed CAR, the higher the bank's ROA.

Bank liquidity, profitability and green credit.

The results of the hypothesis test on the effect of bank liquidity (LDR) on green credit show a p-value of 0.4272 > 0.05. This shows that the LDR variable has no effect on GCR, so the hypothesis in this study is rejected. These results indicate that the large LDR is not able to increase GCR. This is possible because GCR is a relatively new product to support the government's green program, so the portion for GCR is still small. The results of this research support the research results of (Yasmin & Akhter, 2021) which found that LDR had no effect on green credit in banking. High bank liquidity does not guarantee a high amount of bank green credit. Even though large liquidity will result in banks having large leeway funds, and reducing the bank's opportunity to gain profits from disbursing third party funds, this has no effect on the amount of green credit issued by the bank.

Meanwhile, the effect of LDR on profitability produces a p-value of 0.0000 <0.05, which means that LDR has a positive and significant effect on profitability. The size of the LDR shows the amount of credit provided. The bank's main income is from credit, so the greater the LDR, the greater the profitability. These results support the research results of (Mengyao, 2020), which explains that bank liquidity has a positive effect on bank profitability.

Operational efficiency, profitability and green credit

The results of the hypothesis test on the effect of operational efficiency on green credit have a p-value of 0.7537 > 0.05, which shows that EIR has no effect on GCR. This result is not in accordance with the hypothesis and contradicts the results of previous research such as (Zhang, 2018) and (Song et al., 2019), which found that bank operational efficiency had a positive effect on bank green credit. However, the results of this research are supported by research results from (Zumarnis & Irsad, 2023) which explains that EIR does not have a significant effect on credit distribution in banks in Indonesia. This indicates that bank operational efficiency does not have a significant effect on the amount of green credit disbursed by banks. Meanwhile, the effect of efficiency on profitability has a significant negative effect as indicated by a p-value of 0.000 < 0.05. These results support the research results of (Song et al., 2019), which explains that bank operational efficiency has a positive effect on bank profitability. Efficiency can be seen from the cost side and income side of a bank. Costs that are greater than income indicate that the bank is less efficient in its operations, and it is assumed that the bank has not achieved maximum profits. An appropriate increase in the management expense ratio will help improve operational efficiency and increase the bank's ROA.

Credit risk, profitability and green credit

The influence of credit risk on green credit produces a p-value of 0.4272 > 0.05, meaning that credit risk has no effect on green credit. These results are not in accordance with the hypothesis and are not in line with the research results of (Cui et al., 2018), and (Zhang, 2018), who found that credit risk (NPL) had a negative effect on bank green credit. However, the results of this study are supported by the results of research from (Yin et al., 2020) who found that NPLs had no effect on green credit in banking. This means that the bank's credit risk does not become an obstacle in distributing green credit. Bad credit can arise when creditors are unable to complete their obligations to the bank. Green credit services usually face higher risks, low rates of return, long-term operations and other negative factors. However, from the results of this research, bad credit in banks in Indonesia does not affect the amount of green credit issued by banks.

On the other hand, the effect of credit risk on profitability has a p-value of 0.9884 > 0.05. This shows that the NPL variable has no effect on ROA, so the eighth hypothesis in this research can be accepted. These results do not support several previous research results such as (Cui et al., 2018), (Zhang, 2018), and (Song et al., 2019) which explains that bank bad loans have a negative effect on bank profitability. However, these results support the research results of Anggraeni et al. (2019) which explains that the NPL variable does not have a negative effect on bank ROA in Indonesia. Banks that have a high volume of green credit, but have problems, can cause the bank to lose profits and profitability. Problematic credit can be in the form of third party financing that is substandard, questionable or even experiencing problems. Problem loans can arise when banks are less than perfect in analyzing credit, there is stakeholder intervention, or other macro factors that cannot be controlled by the bank. However, from the results of this research, bank bad loans have no influence on banks in obtaining bank profitability.

Green credit in Indonesia has received support from the government through the preparation of the Indonesian Sustainable Finance Roadmap as well as several regulations related to green finance and sustainable finance. It can be seen that the green credit that has been implemented by banks in Indonesia can affect bank profitability. Although this research has not shown what variables influence green credit issued by banks in Indonesia. This is because none of the variables bank capital adequacy, bank operational efficiency, bank liquidity, bank bad credit, and bank size have a significant influence on green credit in Indonesia. In contrast to the variables that influence bank profitability in Indonesia, the variables bank capital adequacy, bank operational efficiency, bank liquidity, bank size, and bank green credit have a significant influence on bank profitability. However, the bank bad credit variable does not have a significant effect on bank profitability in Indonesia.

It is hoped that the research can become a reference to support research related to sustainable finance in Indonesia. Research related to sustainable finance in Indonesia is still not widely carried out by researchers. It is hoped that this research will also support positive government policies in supporting the sustainable financial ecosystem in Indonesia and its development. Bank stakeholders can synergize with various parties such as the government & practitioners in the context of sustainable financial management strategies. Banks can also conduct various training for green credit management, expand green credit promotion, and strengthen risk management of green credit.