Sustainable Banking and Bank Stability in Nigeria: Empirical Evidence from Deposit Money Banks

Abstract

We investigated the impact of sustainable banking practices on bank stability in the Nigerian banking sector. We focused on data from 2012 to 2022, which were extracted from the balance sheets of deposit money banks in Nigeria. We employed the Dynamic Ordinary Least Squares (DOLS) estimator with E-Views to analyze the data. Our findings show that environmental emissions and waste reduction have minimal effects on bank assets, capital adequacy, and liquidity, though they do not directly cause financial instability. Investments in environmental innovation reduce asset growth and increase liquidity constraints but lower non-performing loans, emphasizing a trade-off between sustainability and stability. Environmental resource use efficiency remains neutral regarding asset stability and capital adequacy but poses liquidity challenges. Social welfare investments have little impact on asset growth and profitability, potentially reducing financial stability. Human resource development improves capital adequacy and liquidity strengthening bank stability, while community investments aid societal growth but create liquidity pressures. Macroeconomic factors like GDP growth and inflation are significant, yet economic growth does not always increase bank assets, whereas inflation increases non-performing loans. Sustainable banking in Nigeria is evolving; therefore, there is a need for robust regulation, financial incentives for compliance, a high level of awareness, and alignment between banking operations and sustainability principles.

1. Introduction

The need to conduct a comprehensive study on the impact of sustainable banking on bank stability is essential in Nigeria as the country serves as a focal point for other developing economies. Nigeria is Africa’s largest economy, with a Gross Domestic Product (GDP) of USD 477.38 billion in 2022 (World Bank, 2023; National Bureau of Statistics, 2023), a diverse financial sector (International Monetary Fund, 2023), and economic realities of many emerging markets that attract international investors, making it a primary destination for foreign investment (Michael, 2023; Mogaji, 2023). In addition, Nigeria’s banking sector mirrors other developing economies in terms of financial regulations, sustainability practices, and environmental challenges like India, Kenya, Malaysia, and South Africa. Despite weak governance and regulatory gaps, Nigeria remains a key model for emerging markets. Its financial sector attracts global investors while integrating sustainable banking reforms. The results of this study will not only inform green or socially responsible investors but also serve as a compelling market signal for other developing countries who had agreed with the Sustainable Development Goals of the United Nations.

Sustainable banking is a concept in sustainable finance that aims to foster a sustainable global economy (Busch et al., 2016; Bătae et al., 2021). Sustainable banking entails the provision of financial services that not only generate profit but also emphasize environmental, social, and governance (ESG) factors in line with their traditional financial activities as financial institutions (Dikau & Volz, 2021; Menicucci & Paolucci, 2023). Thus, stakeholders in financial institutions, such as regulatory bodies, deposit money banks, and bank customers, among others, must ensure that they integrate sustainable practices into all aspects of banking operations—from lending and investment to day-to-day operations—invest in green finance products, and incorporate climate and sustainability risk assessments (Yip & Bocken, 2018; Rume and Islam, 2020). Sustainable banking practices can enhance long-term bank stability by managing risks connected to ecological degradation, bad governance practices, and social inequality.

Several scholars, including Nizam et al. (2019), Simpson and Kohers (2002), and Cornett et al. (2014), have explored the relationship between corporate social responsibility (CSR) and banking sector performance, but findings remain mixed. For instance, Soana (2011) argued that banks engaging in social and environmental practices may experience lower financial performance compared to those that do not conform to ESG principles. Meanwhile, other studies have found no clear link between sustainability regulations and bank stability in countries like Bangladesh, China, and Nigeria (Oni, 2016). On the other hand, empirical evidence suggests that sustainable banking can enhance bank stability (Friede et al., 2015; Khan et al., 2016; Giuzio et al., 2019). Specifically, Friede et al. (2015) asserted that banks with strong environmental, social, and governance (ESG) performance tend to have lower risk exposure and a more stable financial outlook. Nevertheless, there are few studies on sustainable banking and bank stability in developing countries.

Some anecdotal evidence also suggests a positive link between sustainability and bank stability (Weber et al., 2016; Weber, 2014; Hornbeck, 2012), emphasizing banks’ ability to tap into opportunities in the green economy (Bebbington et al., 2024). This underscores how sustainable banking practices contribute to bank stability by positively influencing key financial indicators, such as asset quality, non-performing loans (NPLs), risk-adjusted return on assets (RAROA), the loan-to-asset ratio, capital adequacy ratio, the liquidity coverage ratio, and the credit risk ratio. Sustainable banks improve asset quality and reduce NPLs through stringent screening processes, lending primarily to well-managed and resilient businesses (Weber, 2016; Scholtens, 2009). This approach strengthens risk management strategies, enhances risk-adjusted returns, and ensures long-term financial stability (Cort & Esty, 2020).

Furthermore, sustainable banks maintain balanced loan-to-asset ratios and strong capital adequacy and liquidity coverage ratios, which support asset diversification and ensure financial resilience (Jeucken, 2001; Becchetti et al., 2015a; Scholtens, 2009). By thoroughly evaluating the creditworthiness of borrowers and project viability, these banks achieve lower credit risk ratios (Weber, 2016). While sustainable banking does not directly control inflation, it mitigates its impact by promoting economic stability through long-term investments and reducing speculative financial activities, ultimately fostering inclusive and environmentally sustainable GDP growth (Schoenmaker & Schramade, 2019). However, the impact of sustainable banking on bank stability remains a subject of ongoing debate, especially in developing countries where banks operate in unpredictable and maladapted financial institutions.

Nigeria is a developing economy, and many of its banks have adopted sustainable practices, showing promising results in terms of stability and growth since their introduction to the nine sustainable banking principles in 2012 (Central Bank of Nigeria (CBN), 2012, 2019). However, implementing sustainable banking practices comes with challenges, particularly in balancing short-term financial pressures with long-term sustainability objectives. Some stakeholders believed that prioritizing sustainability could undermine profitability, creating a perceived trade-off between financial performance and sustainability (Revelli & Viviani, 2015; Shi et al., 2025).

Interestingly, Nigeria adopted sustainable banking principles in 2012. It is necessary to evaluate its impact on deposit money bank stability, taking cognizance of the fragility of Nigeria’s banking system, which has experienced government intervention in the form of bailouts, short-term profit orientation, and a lack of comprehensive enforceable laws on sustainable banking practices. In addition, a limited number of empirical studies have examined the relationship between sustainable banking practices and bank stability in Nigeria. Therefore, this study takes cognizance of the principle-based approach, theoretical-based approach, and risk-based approach in measuring bank stability, making it a unique contribution to the literature. By doing so, we seek to bridge the gap in the existing literature and understand the current position of the adoption of sustainable banking principles among deposit money banks in Nigeria.

2. Literature Review

There are several theories related to stakeholder theory, legitimacy theory, financial intermediation theory, resource-based view (RBV) theory, institutional theory, and others. However, this study applied stakeholder theory as a theoretical framework in examining the impact of sustainable banking on bank stability in the Nigerian banking sector due to its suitability and robust framework in addressing environmental and social factors. More specifically, the theory emphasizes the interests of different stakeholders in the decision-making process, which aligns with the holistic and integrative framework of the ESG principles. Freeman et al. (2018) asserted that stakeholder theory supports a holistic approach to business, aligning with the objectives of the ESG principles that emphasize the consideration of different stakeholders’ interests, especially those related to environmental sustainability and social equity.

Sustainable banking practices and bank stability appear to be a double-edged sword, making it essential to understand their dynamics to achieve financial stability and promote resilience amid different phases of sustainable banking practices that will enhance bank stability in the long run. Additionally, there are few studies that have investigated this phenomenon in developing countries.

Nevertheless, a few studies, such as Samson and Tukur (2024), examined the effect of bank sustainable performance on bank stability in Nigeria using ESG principles to proxy sustainable banking, while earnings per share, leverage, and firm size were used to proxy bank stability with the aid of panel data regression. Their findings revealed that governance performance, social performance, and environmental performance have a negative and insignificant effect on the bank stability of deposit money banks in Nigeria. Salim et al. (2023) examined the impact of sustainable banking practices on bank stability using a panel dataset of 473 banks in 74 countries. They used corporate environmental performance (CEP) and corporate social performance (CSP) to proxy sustainable banking practices. Their findings revealed a negative relationship between sustainable banking practices, as measured by CEP, and bank stability, as measured by non-performing loans (NPLs). However, the effect was insignificant for small and large banks, as well as for banks in countries with low environmental scores. Additionally, there was no significant relationship between CSP and bank stability, whereas financial product safety, a component of CSP, had a positive effect on bank stability.

Karim et al. (2022) examined sustainable banking regulations before and during the coronavirus outbreak, focusing on the moderating role of financial stability in Pakistan between 2006 and 2020 using the System-Generalized Method of Moments. The dependent variables were proxied by return on assets, return on equity, and Tobin’s Q, while the capital adequacy ratio, total debt ratio, deposit ratio, loan ratio, long-term debt ratio, age, size, and growth opportunities were used as independent variables, with financial stability serving as the moderating variable. Their findings revealed that the capital adequacy ratio, deposit ratio, and loan ratio were positively significant, while leverage ratios were negatively related to profitability and market return. Furthermore, their findings indicated that in both the pre-pandemic and pandemic eras, most banks in Pakistan improved their deposit ratio, profitability, and capital adequacy, while financial stability played a significant moderating role by reducing default risk and enhancing the effectiveness of sustainable banking operations. Torre Olmo et al. (2021) used a two-step System-GMM on 1236 banks across 48 countries between 2015 and 2019 to examine the impact of sustainable banking practices on bank profitability and bankruptcy risk, focusing on the effects of market power and efficiency on bank profitability. Their findings revealed that the adoption of sustainable banking principles enhanced profitability. It was further revealed that market power influences profitability among traditional banks more than sustainable banks. The study further showed that both sustainable and conventional banks experienced higher levels of cost scale efficiency, which led to an increase in their profitability. However, there is no significant relationship between sustainable banking and insolvency risk. Their findings revealed that the conventional bases of bank profitability are inapplicable in explaining the superior profits of sustainable banks.

Amadi et al. (2021) examined banking system stability as a prerequisite for financing the Sustainable Development Goals in Nigeria using annual time series data from 1992 to 2019 with the aid of the Autoregressive Distributed Lag (ARDL) model. SDGs 8 and 9 were used to proxy the dependent variable, while variables such as GDP per capita, commercial banks’ loans to small-scale enterprises, and liquid assets to total assets of banks were used to proxy the independent variables. The study’s findings revealed a positive relationship between banking system stability and the funding of SDGs 8 and 9. However, there was a negative relationship between non-performing loans and other variables in the study. The findings suggest that the stability of Nigerian banks would enhance the funding of the Sustainable Development Goals and contribute to banking sector development. Toader et al. (2018) examined corruption and banking stability using evidence from emerging economies between 2005 and 2012. They analyzed data from 144 commercial banks across 17 countries in Central and Eastern Europe using a panel ordinary least squares estimator. Their findings revealed that banks achieve stability, increase credit growth, and reduce non-performing loans in a low-corruption environment. Interestingly, the study further revealed that banks operating in countries that do not comply with corporate governance principles experience a greater impact of corruption.

Nasreen et al. (2017) examined financial stability, energy consumption, and environmental quality using evidence from South Asian economies between 1980 and 2012 with the aid of a multivariate framework. Their findings indicate that financial stability enhances environmental quality. However, in the long run, continuous increases in energy consumption and economic growth negatively affect environmental quality. The results also align with the environmental Kuznets curve (EKC) hypothesis, which suggests that there is an inverted U-shaped relationship between income and environmental quality. Additionally, the study provides evidence of a one-way causal relationship between financial stability and CO₂ emissions in Pakistan and Sri Lanka. These insights offer policymakers valuable guidance in developing integrated financial, economic, and energy policies to mitigate the adverse effects of environmental pollution.

Dwumfour (2017) examined banking stability in sub-Saharan Africa using OLS-PCSE, FE regression, and SYS-GMM with country-specific data from 32 sub-Saharan African countries from 2000 to 2014. The study used the Z-score, the ratio of non-performing loans to gross loans (NplGross), and the ratio of bank regulatory capital to risk-weighted total assets (Regcar) to proxy the dependent variable, while foreign entry, market power (five-bank asset concentration), competition (Boone indicator), banking access (number of commercial bank branches per 100,000 adults), profitability (net interest margin), diversification (non-interest income to total income), institutional quality, banking crises, and macroeconomic variables (inflation) were used to proxy independent variables. The findings revealed that banking spread (net interest margin—NIM) is the core means of achieving stability during crisis periods, especially among foreign banks. The results further show that diversification has a positive impact on stability (Z-score) but not at a significant level. Likewise, stability could be improved if large banks are well regulated in concentrated markets.

Ziaei (2015) used a sample of 13 European and 12 East Asian and Oceanian countries to investigate the effects of financial indicator shocks on energy consumption and carbon dioxide (CO₂) emissions from 1989 to 2011. Their findings revealed that in these countries, the effects of energy consumption and CO₂ emission shocks on financial indicators such as private sector credit were not very pronounced. However, the strength of energy consumption shock on the stock return rate in European countries was greater than that in East Asian and Oceanian countries. In addition, East Asian and Oceanian countries experienced shocks to the stock return rate, which was influenced by energy consumption. Bordon and Schmitz (2015) evaluated financial stability as a precondition for the financing of sustainable development in emerging and developing countries, finding that financial stability depends on national financing conditions, financial system structure, and sector-specific risks, particularly in energy. Managing financial risks is vital, requiring emerging economies to navigate financial complexities while ensuring sustainable investments. Strengthening international financial regulation and coordination are necessary to contain systemic risks without delaying sustainable development financing. Balancing financial stability with long-term development efforts remains a key challenge for policymakers in developing economies.

Mircea (2014) used a qualitative methodology to evaluate the existing literature and previous case studies to determine whether sustainable banking is a banking solution, focusing on financial stability and environmental responsibility. His findings revealed that despite regulatory inconsistencies and a lack of adequate institutional frameworks to support sustainable banking initiatives, the integration of environmental, social, and governance (ESG) principles has enhanced sustainable banking and economic resilience and reduced credit risk, subject to comprehensive policy implementation and stakeholder commitment.

Shahbaz (2013) investigated whether financial instability increases environmental degradation in Pakistan. The author revealed that a positive relationship exists between financial stability, economic growth, energy consumption, and environmental degradation. The results further show that one of the dominant factors that harms environmental quality is energy consumption. Richard (2010) used a dataset involving 16 developed and 20 developing countries to evaluate the relationship between financial instability and CO₂ emissions with the aid of static and dynamic models. The findings revealed that there is a positive relationship between financial instability and environmental degradation. The author further established that economic growth and population density are the main factors that increase environmental pollution in the selected countries. Their findings also affirmed the environmental Kuznets curve (EKC) hypothesis, which states that there is a relationship between environmental quality and economic development.

An empirical review showed that most existing studies from developed countries focused on the relationship between financial stability, environmental sustainability, sustainable development, and bank performance, with mixed findings. However, there is an empirical gap in the literature examining the direct impact of sustainable banking practices on bank stability in Nigeria, a developing nation that has implemented sustainable banking principles since 2012. Interestingly, most of the existing studies were from developed economies where sustainable banking has been integrated into financial stability frameworks. Since Nigeria has unique economic, regulatory, and institutional challenges, like those of other developing countries, it is therefore expedient to examine whether the adoption of sustainable banking principles has had any tangible impact on bank stability considering the volatility of Nigeria’s financial sector that is subjected to external market shocks and regular government intervention. The remainder of this study is structured as follows: Section 3 presents a description of the data and methodological approach used, while Section 4 entails the results and findings. Finally, Section 5 presents a discussion of the findings, and Section 6 presents the conclusions and recommendations.

3. Materials and Methods

3.1. Data Description

Methodological Framework: Estimation Technique, Model, and Variable Specification

We extracted the data from the annual reports of the selected twelve deposit money banks in Nigeria, while the GDP growth rate and inflation rate were collected from the Central Bank of Nigeria Statistical between 2012 and 2022. This period captures key regulatory reforms like the 2014 CBN Code, economic fluctuations and the adoption of Basel III guidelines, as well as oil price crash and naira depreciation between 2015 and 2017. In addition, it reflects governance stability, post-2008 financial crisis recovery, the pandemic and digital era (2020–2022), and the commencement of the implementation of sustainable banking principles by the Central Bank of Nigeria in 2012.

3.2. Model Specification

This study examines the long-term relationship between sustainable banking practices and bank stability based on the previous work of Salim et al. (2023). It is interesting to note that Salim et al. (2023) utilized the Generalized Method of Moments (GMM) estimator. However, our study adopts the Dynamic Ordinary Least Squares (DOLS) estimator. Both GMM and DOLS are designed to address econometric concerns such as endogeneity, heteroscedasticity, and autocorrelation. Nevertheless, DOLS is preferred in this study based on the data pertaining to a single country, making the cross-sections relatively homogeneous and eliminating concerns about cross-sectional dependence (Ghaderi et al., 2023).

Furthermore, the DOLS estimator, introduced by Stock and Watson (1993), offers significant advantages over the traditional Ordinary Least Squares (OLS) method. By incorporating both leads and lags of the first differences in the explanatory variables, DOLS effectively addresses issues of endogeneity, simultaneity bias, and serial correlation, thereby producing more reliable long-run estimates (Muduli et al., 2022; Insah & Ofori-Boateng, 2012). In this study, panel co-integration time series properties tests were conducted to verify the appropriateness of the use of panel DOLS to estimate the long-term relationship between sustainable banking practices and bank stability among banks in Nigeria (Alam et al., 2021). The panel DOLS equation according to Stock–Watson is shown as Equation (1) as follows:

$$Y_{it}={\beta }_0+{\beta }_1{X}_{it}+\sum _{j=-m}^n{\phi }_j∆X_{i,t-j}+{\partial }_{it}$$

(1)

where Y_it, X_it, and ∆X are the dependent variable, matrix of explanatory variables, and matrix of the first difference in explanatory variables, respectively. ${\beta }_0$, ${\beta }_1$, and ${\phi }_j$ are the estimated parameters, and m and –m are the orders of lead and lag, respectively.

In line with the objective of this study, the DOLS equation for the long-term relationship between sustainable banking practices and bank stability among banks in Nigeria is expressed as Equation (2):

BS_it = α + β_KX_it + e_it

(2)

where BS_it represents the bank stability for a number of banks, i, over a specific annual period, t. α is the constant term, while β_K and X_it are the coefficients of sustainable bank practices, denoted as the error term in the model.

In this study, sustainable banking practices are categorized into two groups: key variables and control variables. Environmental sustainability in banking is denoted as ENV, while social sustainability is represented as SOC. Meanwhile, the control variables include macroeconomic factors such as the GDP growth rate and inflation rate. This information is expressed in Equation (3):

$${BS}_{it}=\alpha +{\beta }_1{ENV}_{it}+{\beta }_2{SOC}_{it}+{\beta }_3{GDPGrowth}_{it}+{\beta }_4{Inflationrate}_{it}+{\in }_{it}$$

(3)

Unlike the study of Salim et al. (2023), this research provides a more detailed breakdown of the environmental factor into three key dimensions: environmental emissions and waste reduction (env_erm), environmental innovation (env_inu), and environmental resource use efficiency (env_ru). Similarly, the social factor is further divided into three components: social workforce (soc_wf), social human rights (soc_Hr), and social community (soc_Com). Equation (4) incorporates the environmental and social dimensions and is expressed as follows:

$$\begin{array}{c}{BS}_{it}=\alpha +{\beta }_1{Env\text{\_}em}_{it}+{\beta }_2{Env\text{\_}Ru}_{it}+{\beta }_3{Env\text{\_}In}_{it}+{\beta }_4{Soc\text{\_}wf}_{it}+{\beta }_5{Soc\text{\_}Hr}_{it}+{\beta }_6{Soc\text{\_}Hr}_{it}\hfill \\ +{\beta }_7{GDP\text{\_}Growth}_{it}+{\beta }_8{Inf}_{it}+e_{it}\hfill \end{array}$$

(4)

In addition to expanding the explanatory variables, it is interesting to note that this study evaluated bank stability with three different approaches, such as principle-based, theoretical-based, and risk-based measures (Abdul Karim et al., 2019; Salim et al., 2023). The principle-based approach is guided by Basel I, Basel II, and Basel III which, in this study, are represented by assets, the capital adequacy ratio, risk-adjusted net interest margin, and liquidity coverage rate. Meanwhile, the theoretical-based and risk-based approaches measure bank stability using the Z-score and non-performing loans, respectively. Therefore, we present six DOLS regression models as follows:

$$\begin{array}{c}{Assets}_{it}=\alpha +{\beta }_1{Env\text{\_}eml}_{it}+{\beta }_2{Env\text{\_}Ru}_{it}+{\beta }_3{Env\text{\_}In}_{it}+{\beta }_4{Soc\text{\_}wf}_{ic}+{\beta }_5{Soc\text{\_}hr}_{it}\\ +{\beta }_6{Soc\text{\_}Com}_{it}+{\beta }_7{GDpGrowth}_{it}+{\beta }_8{Inf}_{it}+e_{it}\end{array}$$

(5)

$$\begin{array}{c}{CAR}_{it}=\alpha +{\beta }_1{Env\text{\_}eml}_{it}+{\beta }_2{Env\text{\_}Ru}_{it}+{\beta }_3{Env\text{\_}In}_{it}+{\beta }_4{Soc\text{\_}wf}_{ic}\hfill \\ +{\beta }_5{Soc\text{\_}hr}_{it}+{\beta }_6{Soc\text{\_}Com}_{it}+{\beta }_7{GDpGrowth}_{it}\hfill \\ +{\beta }_8{Inf}_{it}+e_{it}\hfill \end{array}$$

(6)

$$\begin{array}{c}{RANIM}_{it}=\alpha +{\beta }_1{Env\text{\_}eml}_{it}+{\beta }_2{Env\text{\_}Ru}_{it}+{\beta }_3{Env\text{\_}In}_{it}+{\beta }_4{Soc\text{\_}wf}_{ic}\hfill \\ +{\beta }_5{Soc\text{\_}hr}_{it}+{\beta }_6{Soc\text{\_}Com}_{it}+{\beta }_7{GDpGrowth}_{it}\hfill \\ +{\beta }_8{Inf}_{it}+e_{it}\hfill \end{array}$$

(7)

$$\begin{array}{c}{LCR}_{it}=\alpha +{\beta }_1{Env\text{\_}eml}_{it}+{\beta }_2{Env\text{\_}Ru}_{it}+{\beta }_3{Env\text{\_}In}_{it}+{\beta }_4{Soc\text{\_}wf}_{ic}\hfill \\ +{\beta }_5{Soc\text{\_}hr}_{it}+{\beta }_6{Soc\text{\_}Com}_{it}+{\beta }_7{GDpGrowth}_{it}\hfill \\ +{\beta }_8{Inf}_{it}+e_{it}\hfill \end{array}$$

(8)

$$\begin{array}{c}{Z-score}_{it}\hfill \\ =\alpha +{\beta }_1{Env\text{\_}eml}_{it}+{\beta }_2{Env\text{\_}Ru}_{it}+{\beta }_3{Env\text{\_}In}_{it}+{\beta }_4{Soc\text{\_}wf}_{ic}\hfill \\ +{\beta }_5{Soc\text{\_}hr}_{it}+{\beta }_6{Soc\text{\_}Com}_{it}+{\beta }_7{GDpGrowth}_{it}\hfill \\ +{\beta }_8{Inf}_{it}+e_{it}\hfill \end{array}$$

(9)

$$\begin{array}{c}{NPL}_{it}\hfill \\ =\alpha +{\beta }_1{Env\text{\_}eml}_{it}+{\beta }_2{Env\text{\_}Ru}_{it}+{\beta }_3{Env\text{\_}In}_{it}+{\beta }_4{Soc\text{\_}wf}_{ic}\hfill \\ +{\beta }_5{Soc\text{\_}hr}_{it}+{\beta }_6{Soc\text{\_}Com}_{it}+{\beta }_7{GDpGrowth}_{it}\hfill \\ +{\beta }_8{Inf}_{it}+e_{it}\hfill \end{array}$$

(10)

3.3. Methodological Framework

3.3.1. Panel Unit Root Test

To determine the stationarity and order of integration of all variables, this study utilizes three different unit root tests. The first is the LLC test, introduced by Levin et al. (2002), which examines the presence of a unit root across cross-sections. Additionally, this study applies the IPS and ADF tests, developed by Im et al. (2003) and Maddala and Wu (1999), respectively, to assess unit roots at the individual level. The results of these panel unit root tests for the variables included in this study are presented in

Table 1. Variables description and measurement.

	Description	Acronyms	Measurement/Justification
Dependent Variables	Asset	Asset	The natural logarithm of the volume of total assets.
	Non-Performing Loans	NPL	Non-performing loans are computed as impaired loans divided by gross loans.
	Risk-Adjusted Net Interest Margin	RANIM	The risk-adjusted net interest margin (RANIM) is a fundamental financial metric utilized within the banking industry to comprehensively assess the profitability and risk exposure associated with a bank’s core lending and investment activities (Chen & Peng, 2019).
	Capital Adequacy Ratio	CAR	The bank’s available capital is expressed as a percentage of its risk-weighted credit exposures. ${\displaystyle \frac{Tier1Capital+Tier2Capital}{Risk\text{\_}WeighedAssets}}\times 100$
	Liquidity Coverage Ratio	LCR	This is a metric used to ensure that a bank has enough high-quality liquid assets to cover its total net cash outflows over a 30-day stress period. ${\displaystyle \frac{High-QualityLiquidAssets\left(HQLA\right)}{TotalNetCashOutflowsover30days}}\times 100$
	Credit Risk	Z-Score	The Z-score is a metric used to assess the risk of insolvency for deposit money banks (DMBs). It is calculated by taking the sum of the return on assets and the capital adequacy ratio and then dividing it by the standard deviation of the return on assets. This formula is commonly referenced in the literature (Ha & Quyen, 2017; Majumder & Li, 2018).
Sustainable Banking Practices	The weight of each component is divided by the items found in the financial statement.
Independent Variables	Environmental	ENV	The relative sum of category weights for the environmental categories. It is based on three dimensions: ENV_Ru (resource use efficiency), ENV_Em (emission and waste reduction), and ENV_In (environmental innovation). ENV_Ru = a bank’s efficiency in reducing the use of materials, energy, or water and its capacity to find more eco-efficient solutions for business processes. ENV_Em = a bank’s commitment and effectiveness in reducing environmental emissions and waste in operational activities. ENV_In = a bank’s capacity to reduce environmental burdens and costs for its clients and to create new opportunities for eco-designed products and services.
	Social	SOC	The relative sum of category weights for the social responsibility categories. It is based on four dimensions: SOC_Wf (workforce), SOC_Hr (human rights), and SOC_Com (community). SOC_Wf = a bank’s effectiveness in achieving job satisfaction and a safe and healthy workplace while developing both equal and diverse opportunity. SOC_Hr = a bank’s effectiveness in respecting fundamental human rights conventions. SOC_Com = a bank’s commitment to being a good citizen, respecting business ethics, and protecting public health.
Control Variables	Inflation Rate	IR	It measures the rate of rise in the prices of goods and services.
	GDP Growth Rate	GDPgr	It is the change in the GDP from one period to another. $\begin{array}{l}\left(\right(GDP\text{\_}currentperiod-GDP\text{\_}previousperiod)/\\ GDP\text{\_}previousperiod)\times 100\end{array}$

Source: Researchers’ compilation, 2024.

.

The results in

Table 2. Results of panel unit root tests.

Variable	Panel Specification	Unit Root Test	Levels	First Difference	Order of Integration
Assets	Common (pooled) effects Individual effects	LLC IPS Fisher-ADF	−5.029 *** −0.623 25.33	−2.966 *** 3.656 *** 61.292 **	I(1)
CAR	Common (pooled) effects Individual effects	LLC IPS ADF	−22.091 *** −6.296 *** 54.281 ***	11.563 *** 4.441 *** 64.326 ***	I(0)
RANIM	Common Individual	LLC IPS ADF	−2.916 *** 0.548 17.332	−9.920 *** −4.705 *** 69.938 ***	I(1)
LCR	Common Individual	LLC IPS ADF	−5.886 *** −1.504 * 36.560 **	69.938 *** −1.735 ** 48.858 ***	I(1)
Z-Score	Common Individual	LLC IPS ADF	0.638 −0.651 27.37	−2.409 *** −3.309 *** 55.47 ***	I(1)
Env_em	Common Individual	LLC IPS ADF	0.069 0.413 18.026	−3.69 *** −2.50 *** 32.07 ***	I(1)
Env_ru	Common Individual	LLC IPS ADF	−3.371 *** 1.284 * 32.95	4.952 *** −3.538 *** 54.82 ***	I(1)
Env_In	Common Individual	LLC IPS ADF	−8.185 *** −5.07 *** 75.972 ***	−22.28 *** 13.34 *** 139.56 ***	I(0)
Soc_wf	Common Individual	LLC IPS ADF	−6.025 *** −2.638 *** 45.42 ***	12.64 *** 7.07 *** 92.95 ***	I(0)
Soc_hr	Common Individual	LLC IPS ADF	−1.51 1.66 ** 41.304 ***	8.81 *** −5.191 *** 72.45 ***	I(1)
Soc_com	Common Individual	LLC IPS ADF	−2.496 *** −1.096 29.97	−5.673 *** −3.94 *** 59.48 ***	I(1)
Ir	Common Individual	LLC IPS ADF	−8.63 *** −3.06 *** 49.32 ***	−8.63 *** −3.31 *** 55.34 ***	I(0)
Gdpgr	Common Individual	LLC IPS ADF	−6.709 *** −1.839 * 35.63 *	−7.70 *** −3.221 *** 54.264 ***	I(0)

Note: Unit root test abbreviations: LLC, IPS, and Fisher-ADF represent Levin, Lin, and Chu; IM, Pesaram, and Shin; and Fisher Augmented Dickey–Fuller (ADF). *, **, and *** denote the rejection of the null hypothesis that the series has a unit root at 10%, 5%, and 1% statistical significance levels, respectively. I(0) and I(1) indicate an integrated order of zero and an integrated order of one, respectively. Source: Authors’ computation from EViews 10 Extract, 2024.

indicate that none of the variables are stationary at their level form, but they achieve stationarity at first-order differencing, I(1). The panel unit root tests reveal a mixed order of integration, with some variables at level I(0) and others at first difference I(1). Notably, all dependent variables become stationary after first differencing, making the panel DOLS regression method the most suitable estimation technique for this study.

3.3.2. Panel Co-Integration Test

We conducted a panel co-integration test to verify the long-term relationship between sustainable banking practices and bank stability. Importantly, the DOLS panel co-integration regression method was carried out using the Kao test to ascertain the panel co-integration existence among the variables in this study.

Table 3. Panel Kao co-integration test results.

Variables	Test Statistics	T-Statistics	Prob.
ALL	ADF	−1.994 **	0.027

Notes: ** denote the rejection of the null hypothesis of no co-integration at 5% statistical significance levels, respectively. ALL represents assets, including non-performing loans (NPLs), CAR, RANIM, LCR, Z-score, env_em, env_ru, env_in, soc_wf, soc_hr, soc_com, ir, and gdpgr. Source: Authors’ computation from EViews 10 Extract, 2024.

presents the results of the Kao co-integration test, which confirm the presence of a long-term relationship among all the variables in this study. The ADF t-statistic value of −1.994 is statistically significant at the 5% level, leading to the rejection of the null hypothesis of no co-integration. Consequently, the alternative hypothesis, which asserts the existence of co-integration, is accepted.

3.3.3. Cross-Sectional Dependence Tests

A cross-sectional dependence test was conducted to determine whether the cross-sections are interdependent or function independently.

Table 4. Results of cross-sectional dependence tests.

Cross-Sectional Dependence (CSD) Tests	T-Statistic	p-Value
Breusch–Pagan LM	105.541	0.001
Pesaran Scaled LM	3.4416	0.001
Bias-Corrected Scaled LM	2.8416	0.005
Pesaran CD	2.4856	0.013

Note: Extracts from EViews 10. BPLM: Breusch–Pagan LM; PSLM: Pesaran Scaled LM; PCD: Pesaran CD.

presents the results of the cross-sectional dependence tests for the 12 selected deposit money banks in Nigeria over the study period of 2012 to 2022. The findings in Table 3 indicate that all four cross-sectional dependence tests—Breusch–Pagan LM, Pesaran Scaled LM, Bias-Corrected Scaled LM, and Pesaran CD—yielded values that are not statistically different from zero at the 1% significance level. Therefore, the null hypothesis of no cross-sectional dependence is rejected, and the alternative hypothesis confirming the presence of cross-sectional dependence is accepted.

4. Results and Findings

4.1. Results of Descriptive Statistics

Table 5. Panel descriptive statistics.

Variable	Mean	Std. Dev.	Minimum	Maximum	Skewness	JB	Observations
ASSET	12.113	0.381	11.39	12.80	−0.28	5.01 *	120
CAR	11.064	33.31	−201.60	69.80	−3.73	1560.69 ***	120
RANIM	0.043	0.02	0.004	0.13	0.72	36.28 ***	120
LCR	45.371	16.99	14.40	132.30	1.83	225.98 ***	120
Z-SCORE	−0.324	6.054	−35.04	1.65	−5.20	3711.41 ***	120
NPL	4.982	3.69	0.17	22.50	2.94	653.78 ***	120
Env_em	0.404	0.25	0.00	1.00	0.84	14.32 ***	120
Env_ru	0.509	0.24	0.00	1.00	−0.02	2.22	120
Env_In	0.506	0.259	0.00	1.00	−0.12	5.26 *	120
Soc_wf	0.603	0.269	0.00	1.00	−0.11	3.65	120
Soc_hr	0.574	0.308	0.20	1.00	0.36	11.65 ***	120
Soc_com	0.568	0.299	0.30	0.16	0.21	9.17 ***	120
Gdpgr	2.417	3.484	2.48	−1.79	−0.29	4.56 *	120
Ir	11.857	3.031	3.03	8.00	0.72	10.45 ***	120

Source: Authors’ computation from EViews 10 Extract, 2024. * and *** denote rejection of the null hypothesis of normality (based on the Jarque-Bera test).

presents the descriptive statistics for the variables used in this study. First, the variables with the lowest and highest mean values are credit risk (Z-score) at −0.324 and the liquidity coverage ratio at 45.371, respectively. Second, the minimum and maximum values of all variables differ, indicating variability across the study period. Third, the skewness values show both positive and negative skews, with none of the variables exhibiting a perfectly symmetrical (zero skew) distribution, confirming that all variables are asymmetrically distributed. Fourth, the Jarque–Bera test results are statistically significant at the 1% level, suggesting that the null hypothesis of normal distribution is rejected for most variables, except for environmental innovation (Env_inu) and social responsibility to the community (Soc_com), which follow a normal distribution. Lastly, this study comprises a balanced panel dataset with 120 observations for all variables from 2012 to 2022. However, the sample size remains relatively small since both the number of years and cross-sections are less than 30.

4.2. Estimation Results

Table 6. Estimated long-term relationship between sustainable banking practices and bank stability in Nigeria.

Independent Variable	Dependent Variable: Bank Stability Estimation Method: Panel DOLS
	Assets	CAR	RANIM	LCR	Z-Score	NPL
Env_em	−0.064 (−0.54)	−2.897 (−0.465)	−0.002 (−0.184)	−1.615 (−0.251)	2.326 (0.707)	2.397 (1.437)
Env_In	−0.230 * (−1.78)	−0.472 (−0.069)	−0.005 (−0.486)	−11.049 (−1.560)	0.135 (0.037)	−1.405 * (−0.466)
Env_Ru	0.076 (0.70)	3.181 (0.557)	0.008 (1.08)	−5.458 (−0.919)	2.441 (0.810)	1.219 (0.798)
Soc_wf	0.012 (0.12)	−2.113 (0.400)	0.007 (0.960)	−0.871 (−0.163)	−1.942 * (−0.696)	−1.315 (−0.930)
Soc_hr	−0.029 (−0.32)	6.216 (1.291)	−0.004 (−0.559)	7.994 * (1.611)	3.312 (1.303)	1.894 (1.470)
Soc_com	−0.065 (−0.34)	3.546 (0.659)	0.003 (0.401)	−6.600 * (−1.182)	−0.959 (−0.338)	−0.413 (0.287)
GDP Growth	−0.046 *** (−2.69)	1.135 (1.251)	−0.002 * (−1.74)	−0.154 (−0.162)	0.396 (0.827)	0.289 (1.191)
Inf	−0.026 * (−1.75)	0.809 (1.044)	+0.00001 (0.011)	−0.423 (0.525)	0.277 (0.679)	0.445 ** (2.148)
No. of observations	121	121	121	121	121	121
No. of cross-sections	5	5	5	5	5	5
R-squared	0.779	0.868	0.676	0.523	0.487	0.332
Long-run variance	0.06	169.06	0.0003	179.29	47.09	12.762
F-statistics	3.42 ***	5.83 ***	1.93 ***	1.05	0.911	0.475

*, **, and *** denote 10%, 5%, and 1% levels of significance respectively.

presents the estimated results of the Dynamic Ordinary Least Squares (DOLS) regression, analyzing the impact of sustainable banking practices on bank stability in Nigeria. The findings in Table 6 reveal that environmental emissions (Env_em) have a negative but statistically insignificant impact on assets (−0.064), as well as on the CAR (−2.897), RANIM (−0.002), and LCR (−1.615). This shows that banks with higher environmental emissions may experience slight reductions in their asset base, capital adequacy, and liquidity. Nevertheless, the insignificant positive effects on the Z-score (2.326) and NPL (2.397) indicate that banks with high emissions do not necessarily face greater financial instability or higher non-performing loans.

In addition, environmental innovation (Env_In) revealed a statistically significant negative impact on assets (−0.230) and NPL (−1.405). This suggests that investments in environmental innovation may lead to a decrease in asset growth and a reduction in non-performing loans. The large negative effect on the LCR (−11.049) suggests that banks engaging in environmental innovation may face liquidity constraints due to the initial high costs associated with sustainable investments. However, its effects on the CAR (−0.472), RANIM (−0.005), and Z-score (0.135) are insignificant, implying a limited direct impact on capital adequacy, risk-adjusted margins, and overall bank stability.

Interestingly, resource use efficiency (Env_Ru) has a positive but statistically insignificant effect on assets (0.076), the CAR (3.181), and RANIM (0.008), suggesting that banks exposed to environmental risks may maintain their asset base and capital adequacy while slightly improving their risk-adjusted margins. However, its negative impact on the LCR (−5.458) implies that banks facing higher environmental risks might experience liquidity pressures. The insignificant positive effect on the Z-score (2.441) and NPL (1.219) suggests that resource use efficiency does not significantly impact overall financial stability or loan performance.

Furthermore, social welfare (Soc_wf) has a positive but insignificant effect on assets (0.012) and the RANIM (0.007), suggesting that investments in social welfare may not significantly enhance asset growth or risk-adjusted net interest margins. Its negative effect on the CAR (−2.113) and LCR (−0.871) suggests that banks investing more in social welfare might experience slight reductions in capital adequacy and liquidity, though the effects are not statistically significant. However, there is a negative significance on the Z-score (−1.942, p < 0.10). This implies that higher social welfare investments could be linked to a decrease in overall bank stability. In addition, its negative but insignificant impact on NPLs (−1.315) suggests that it causes a possible reduction in non-performing loans, though the effect is weak.

In addition, social human resource (Soc_hr) has a negative but insignificant effect on assets (−0.029) and the RANIM (−0.004), suggesting that investments in human resource development may not significantly affect asset growth or risk-adjusted net interest margins. Nevertheless, its positive effect on the CAR (6.216) and LCR (7.994) indicates that banks that prioritize human resource development may experience improved capital adequacy and liquidity. Additionally, its positive effect on the Z-score (3.312) and NPL (1.894) shows that human resource investments may enhance financial stability and loan performance, though the effects are not statistically significant at this early stage of sustainable banking principles in Nigeria.

Social community investment (Soc_com) shows a negative but insignificant effect on assets (−0.065) and the Z-score (−0.959), suggesting that banks investing in community development may not experience significant asset growth or improved financial stability. However, its positive but insignificant effect on the CAR (3.546) and RANIM (0.003) implies that community investments may not substantially enhance capital adequacy or risk-adjusted margins. However, its negative and statistically significant impact on the LCR (−6.600, p < 0.10) indicates that high community investments might contribute to liquidity challenges. Finally, its insignificant negative effect on NPLs (−0.413) suggests a weak association with loan performance.

GDP growth has a statistically significant negative effect on assets (−0.046), suggesting that higher economic growth may not necessarily lead to an increase in bank assets due to alternative investment opportunities outside the Nigerian banking sector. Similarly, its negative and weakly significant effect on the RANIM (−0.002) shows that economic growth slightly reduces risk-adjusted net interest margins as a result of increased competition or lower lending rates in a growing economy. However, GDP growth has an insignificant positive effect on the CAR (1.135), Z-score (0.396), and NPL (0.289), suggesting no strong link between economic expansion and these indicators of bank stability.

Inflation (Inf) has a weakly significant negative effect on assets (−0.026), implying that rising inflation could slightly reduce the asset base of banks, possibly due to declining real asset values or reduced lending. However, it has a statistically significant positive effect on NPLs (0.445), indicating that higher inflation is associated with an increase in non-performing loans, likely due to borrowers struggling with rising costs and repayment burdens. Inflation’s effects on other indicators, such as the CAR (0.809), Z-score (0.277), and LCR (−0.423), are statistically insignificant, suggesting that there is no strong impact on capital adequacy, overall financial stability, and liquidity.

4.3. Post–Estimation Tests

Table 7. Post-estimation test.

Post–Estimation Tests	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Correlogram of Residual Squared	6.12	5.90	97.12 ***	34.92 ***	216.23 ***	8.222
Variance-Inflated Factor (VIF)	4.45	7.58	3.09	2.10	1.95	1.50
MAPE	8.75%	23.81%	24.13%	18.02%	110.50%	40.2%

*** denote 10%, 5%, and 1% levels of significance, respectively.

presents the results of post-estimation tests for all the models in this study. Firstly, the correlation coefficients of residual squared are statistically different from zero for all models except Models 1, 2, and 5. This indicates that these three models do not exhibit serial autocorrelation, confirming that assets, the capital adequacy ratio (CAR), and non-performing loans (NPLs), which serve as proxies for bank stability, are reliable for inference. Secondly, the variance inflation factor (VIF) test results for all models are below the threshold of 10, signifying the absence of multi-collinearity among the explanatory variables.

Lastly, the Mean Absolute Percentage Error (MAPE) values for Models 1, 2, and 5 are 8.75%, 23.18%, and 40.2%, respectively. These percentages indicate strong, accurate, and reasonable forecasting capabilities for the models. Notably, Model 1 (assets) demonstrates the most robust and reliable forecasting ability, making it the most dependable variable for assessing bank stability in this study.

5. Discussion of Findings

The findings on environmental emissions (Env_em) suggest that while higher environmental emissions are associated with slight reductions in bank assets, capital adequacy, and liquidity, these effects are statistically insignificant, aligning with the studies of Friede et al. (2015) and Nizam et al. (2019). This implies that Nigerian banks engaging in environmentally unfriendly practices may not experience immediate financial instability given that sustainable banking is still in its early stages in Nigeria (Samson & Tukur, 2024). However, given the global shift towards sustainability and increasing regulatory pressure, banks with high emissions may face long-term financial risks due to potential penalties, reputational damage, and reduced investor confidence. The insignificant positive effect of environmental emissions on the Z-score and NPL suggests that, for now, banks with higher emissions do not necessarily face higher default risks, aligning with the study of Capelle-Blancard and Petit (2019). Nevertheless, the practical economic implication of these findings shows that there is a need for Nigerian banks to proactively transition toward greener investments to attract global investors.

The results on environmental innovation (Env_In) indicate that banks investing in environmental innovation experience a decline in asset growth and liquidity, suggesting that the high initial costs of sustainable banking investments may put financial pressure on banks in the short term, aligning with previous studies (Weber, 2017; Samson & Tukur, 2024). Nevertheless, the significant negative impact of environmental innovation on NPLs implies that such investments contribute to improved loan performance, possibly by attracting more creditworthy and sustainability-conscious borrowers. Interestingly, this finding aligns with global evidence that sustainable banking enhances long-term stability by reducing exposure to high-risk sectors. Conversely, the large negative impact on liquidity (LCR) suggests that banks implementing green initiatives may face short-term liquidity challenges, requiring supportive policies such as green financing incentives, tax reliefs, and holidays, aligning with the study of Gangi et al. (2019).

The findings on environmental resource use efficiency (Env_Ru) reveal that improved resource efficiency has a positive but statistically insignificant impact on bank assets, capital adequacy, and risk-adjusted margins, aligning with the studies of Nizam et al. (2019) and Scholtens (2009). This suggests that banks adopting better resource use practices—such as energy efficiency and reduced resource waste—can maintain stable asset growth and capital adequacy. However, the negative impact on liquidity (LCR) reveals that banks investing in resource efficiency may experience short-term liquidity pressures, possibly due to the costs of adopting new technologies or restructuring operations (Samson & Tukur, 2024). These findings align with the study of Clark et al. (2015). In addition, the insignificant impact on overall stability (Z-score) and non-performing loans (NPLs) suggests that banks have yet to fully capitalize on resource efficiency to enhance long-term financial stability. The practical economic implication is that it is expedient for Nigerian banks to expand their investments in resource-efficient technologies while ensuring adequate liquidity buffers to manage short-term financial pressures. This shows that financial services and products that are environmentally conscious—such as green financing and lending, sustainable infrastructure financing, the promotion of renewable energy projects, energy efficiency projects, green bonds, and the financing of waste management and recycling—are still in their early stages.

The findings on social welfare (Soc_wf) suggest that investments in social welfare programs have a positive but insignificant impact on asset growth and risk-adjusted net interest margins, aligning with the previous study of Clark et al. (2015). This shows that while social welfare initiatives—such as financial inclusion programs, poverty alleviation, and corporate social responsibility (CSR) efforts—may contribute to a bank’s public image and customer engagement, they do not immediately translate into significant financial benefits. However, the negative effect on the capital adequacy ratio (CAR) and liquidity coverage ratio (LCR) reveals that banks that allocate substantial funds to social welfare may experience a slight decline in their capital reserves and liquidity levels, aligning with the studies of Becchetti et al. (2015b) and Cornett et al. (2014).

Moreover, the statistically significant negative impact on the Z-score shows that higher investments in social welfare could be linked to reduced overall bank stability, possibly due to increased operational costs without immediate financial returns on investment. Despite its negative but insignificant impact on non-performing loans (NPLs), social welfare initiatives might still contribute to reducing loan defaults in the long run by improving customers’ financial well-being. This finding contradicts the study of Bouslah et al. (2018), who argued that financial constraints imposed by social investments could limit a bank’s ability to assess and mitigate credit risk effectively, potentially leading to higher default rates that can later translate into non-performing loans.

Furthermore, the findings on social human resource investment (Soc_hr) indicate that investing in human capital has a negative but insignificant effect on asset growth and risk-adjusted net interest margins, suggesting that expenditures on employee training, health benefits, and capacity building may not immediately boost a bank’s financial performance, aligning with the study of Buallay et al. (2017). However, the positive and significant effect on capital adequacy (CAR) and liquidity (LCR) suggests that banks that prioritize human resource development are likely to maintain better capital reserves and liquidity positions, possibly due to enhanced productivity, efficient risk management, and operational efficiency, aligning with the study of Rahman and Akhter (2021) and the human capital and resource-based view (RBV) theory.

In addition, the positive effect on financial stability (Z-score) and non-performing loans (NPLs) implies that well-trained employees contribute to better risk assessment and adequate credit management, leading to improved overall bank stability and loan performance. This aligns with risk management theory, which posits that the ability of financial institutions to identify and mitigate risks can have a direct effect on bank stability and resilience, as well as the findings of Githaiga (2021). However, these effects are not statistically significant.

The results on social community investment (Soc_com) reveal that community development spending has a negative but insignificant effect on asset growth and financial stability (Z-score). This shows that while banks invest in social infrastructure, educational programs, and small-business development, these investments do not immediately enhance the financial performance of deposit money banks in Nigeria. This finding aligns with the studies of Buallay (2019) and Cornett et al. (2014).

However, the positive but insignificant effect on capital adequacy (CAR) and risk-adjusted net interest margins (RANIM) shows that community development initiatives might contribute to an enhanced bank brand value and customer loyalty but do not directly translate into improved financial ratios in terms of performance (Wu & Shen, 2013; Tong et al., 2024). However, the significant negative impact on liquidity (LCR) indicates that extensive community investments might strain a bank’s short-term liquidity, likely due to the costs associated with funding these projects without immediate financial returns. The insignificant negative effect on NPLs suggests that while community investment might improve borrowers’ economic conditions, the impact on loan performance remains weak in most cases, aligning with the study of Shen et al. (2022). This finding contradicts Buallay (2022), who argued that banks with higher CSR expenditures, such as community investment initiatives, usually have a lower level of non-performing loans.

Furthermore, GDP growth has a statistically significant negative impact on bank assets in Nigeria. This suggests that as the economy expands, banks do not necessarily experience a corresponding increase in their assets. A possible explanation is that higher GDP growth creates alternative investment opportunities outside the banking sector, such as direct investments in real estate, capital markets, and technology-driven financial services. This finding aligns with studies in developing economies where economic expansion often leads to financial diversification beyond traditional banking (Mesagan & Vo, 2024; Ozili, 2022). The weakly significant negative effect on risk-adjusted net interest margins (RANIM) further supports the notion that economic growth might lead to increased competition among banks or a reduction in lending rates, thereby compressing profit margins from what is expected as returns. The statistically insignificant effects of GDP growth on the capital adequacy ratio (CAR), financial stability (Z-score), and non-performing loans (NPLs) reveal that economic expansion alone does not directly enhance these indicators. This aligns with prior research showing that in emerging markets, structural challenges such as regulatory inefficiencies, exchange rate volatility, and credit market imperfections can weaken the expected positive link between GDP growth and financial stability (Scholtens, 2009; Tan & Floros, 2018; Ozili, 2022). In the case of Nigerian banks, this shows that GDP growth alone is not a sufficient factor for achieving banking stability, reinforcing the need for targeted financial policies that align with sustainable banking principles.

In addition, the findings also reveal that inflation has a weakly significant negative effect on bank assets. This suggests that rising inflation may erode asset values or reduce the capacity of banks to expand their loan portfolios. This is in line with the study of Wu and Shen (2013), who affirmed that inflationary pressures reduce the real value of financial assets and discourage long-term investments in the banking sector in developing countries. A key observation is the statistically significant positive effect of inflation on non-performing loans (NPLs). This suggests that higher inflation increases credit risk as borrowers struggle with rising costs and find it difficult to meet repayment obligations. This is consistent with the findings of Agénor and Da Silva (2013) who found that inflation often leads to deteriorating loan quality, particularly for consumer and SME loans in developing economies. Interestingly, this finding emphasizes that Nigerian deposit money banks should integrate inflationary risk into credit decision-making and risk assessment frameworks and promote conservative lending strategies in order to achieve responsible lending and financial resilience. However, inflation’s effects on other indicators, such as the CAR, Z-score, and the liquidity coverage ratio (LCR), remain statistically insignificant. This suggests that while inflation affects loan performance, it does not significantly impact overall bank stability or capital adequacy, aligning with the study of Abiodun et al. (2020) and Egbunike and Okerekeoti (2018).

Interestingly, these study findings can be attributed to several underlying factors aligned with empirical findings in the literature on sustainable banking. Most banks in developing countries have not fully integrated environmental and social considerations into their financial decision-making, as sustainable banking practices are still at an early stage and the financial benefits of sustainable banking practices often take time to materialize (Eccles et al., 2014; Weber, 2017; Ratnasari et al., 2021). In addition, Nigeria’s regulatory framework for sustainable banking has only recently gained momentum compared to developed countries, where sustainability regulations have been in place for decades (Friede et al., 2015). Other factors, such as weak regulatory enforcement, the traditional short-term profit orientation of most banks in developing countries, limited green financing options, and economic volatility, can further delay the realization of sustainable banking benefits.

6. Conclusions and Recommendation

This study investigated the influence of sustainable banking practices on bank stability by analyzing data from the balance sheets of selected Nigerian deposit money banks between 2012 and 2022. We concluded that environmental emissions currently have an insignificant impact on bank stability variables such as assets, capital adequacy, and liquidity. This suggests that Nigerian banks have not been fully complying with environmental emission regulations under sustainable banking principles. Nevertheless, banks should adopt environmentally responsible strategies as investors are increasingly conscious of sustainability due to global regulatory frameworks. The long-term financial risks associated with unsustainable banking practices cannot be overemphasized. While environmental innovation and resource use efficiency are essential for sustainable banking, their financial impact on Nigerian banks remains uncertain. However, policy interventions are needed to support the short-term financial pressures related to environmental innovation and resource use efficiency due to temporary liquidity constraints.

In addition, social sustainability investments, such as human resource development, social welfare programs, and community investments, enhance stakeholder engagement and improve a bank’s reputation but offer limited direct financial benefits. Investments in social welfare initiatives do not significantly impact asset growth or profitability but place pressure on liquidity and capital reserves. Similarly, community investment programs mainly contribute to societal impact without directly influencing financial stability. This highlights the need for Nigerian banks to adopt a balanced approach to corporate social responsibility (CSR). However, investments in human resource development demonstrate that prioritizing employee training and capacity building strengthens risk management and operational efficiency. Therefore, banks should focus more on internal human capital development to enhance financial resilience and stability.

Finally, macroeconomic variables such as GDP growth and inflation significantly influence the financial performance and stability of Nigerian banks, though with mixed effects. While GDP growth negatively affects bank assets and profit margins due to increased financial diversification and competition, it does not directly improve capital adequacy, financial stability, or loan performance. Similarly, inflation weakens asset expansion and increases credit risk by raising non-performing loans, highlighting the vulnerability of banks to macroeconomic volatility. These findings underscore the need for Nigerian banks to adopt proactive financial policies, integrate macroeconomic risks into strategic planning, and implement sustainable banking practices to enhance long-term stability and resilience.

Based on the findings, the following recommendations are made: Since sustainable banking practices are still in their early stages in Nigeria, conducting extensive research over a longer period would provide deeper insights into their true impact on financial stability. This would enable better-informed policy decisions and strategic planning. Regulatory authorities should strengthen oversight and ensure that banks fully integrate sustainability into their risk management frameworks. Just as penalties are imposed for non-compliance with banking stabilization measures, similar consequences should apply for failing to adhere to sustainable banking principles. Furthermore, many bank executives and stakeholders still perceive sustainability as secondary to financial performance. Therefore, extensive training programs, seminars, and awareness campaigns should be organized to highlight the long-term benefits of sustainable banking. These initiatives should also demonstrate how sustainability can mitigate long-term risks such as liquidity and credit risks.

To prioritize sustainable banking, financial institutions should introduce market-driven products and incentives such as green loans, sustainability-linked loans, green bonds, sustainable investment funds, green mortgages, eco-friendly car loans, sustainable supply chain financing, green savings accounts, energy efficiency financing, and carbon credit trading platforms. Additionally, tax incentives should be provided to support sustainable investments and projects. At the national level, the government must implement macroeconomic policies to control inflation as it remains a major challenge affecting both the micro and macroeconomic environment. Policies that foster economic growth and create a favorable business climate should be introduced to help banks lower operational costs and enhance shareholder value.

To achieve the nine principles of sustainable banking set by the Central Bank of Nigeria in 2012, Nigerian deposit money banks should develop financial products and services such as green bonds, ESG-linked loans, sustainable supply chain financing, green loans, sustainable investment funds, social impact bonds, and green mortgages. These initiatives would not only support sustainability efforts but also improve capital reserves, ensure stable revenue streams, attract a diverse customer base, minimize long-term risks, diversify funding sources, mitigate climate-related financial risks, and contribute to positive environmental and social outcomes. In the long run, adopting these measures would enhance the financial resilience and overall stability of banks, particularly in developing economies like Nigeria, where economic structures are still evolving.

7. Limitation

A limitation of this study is that the findings of this study focus only on deposit money banks, excluding other financial institutions such as microfinance banks and fintech firms.