“Banking system stability: A prerequisite for financing the Sustainable Development Goals in Nigeria”

ARTICLE INFO Agatha Amadi, Kehinde A. Adetiloye, Abiola Babajide and Idimmachi Amadi (2021). Banking system stability: A prerequisite for financing the Sustainable Development Goals in Nigeria. Banks and Bank Systems, 16(2), 103-118. doi:10.21511/bbs.16(2).2021.10 DOI http://dx.doi.org/10.21511/bbs.16(2).2021.10 RELEASED ON Wednesday, 02 June 2021 RECEIVED ON Saturday, 02 January 2021 ACCEPTED ON Wednesday, 12 May 2021


INTRODUCTION
The banking system in Nigeria continues to experience unimpressive and unstable financial conditions, mainly due to the hugely increasing portfolio of non-performing loans (NPLs), which is transferred from one year to another and written off without sufficient recovery efforts (Oleribe & Taylor-Robinson, 2016). This leads to capital and shareholders' fund (SHF) erosion, impairment of liquidity, poor asset quality and consequent reduction in profitability and solvency levels. These conditions have consequently necessitated the monetary authorities initiating several banking reforms aimed at addressing the problems of unstable and unsound banks in the country. According to Bebeji (2013), despite several banking reforms in Nigeria, the problems of banking instability caused by non-performing loans seem endless as the soundness of more banks continues to be threatened, so that near failure and collapse would have been recorded had it not been for the Central Bank's intervention. The case of Polaris (former Skye) Bank is instructive where the Central Bank of Nigeria (CBN) had to finance up to the tune of 700 billion (about USD 2 billion) to keep it from failing. Antoniades et al. (2019) confirmed that the main contributors to the financial crunch experienced by the banking system in Nigeria were non-adherence to prudential guidelines, incorrect financial policy and extreme risk-taking. Although the assumption is that the banking system flourishes and survives more on risk-taking, the risks faced by banks must be efficiently managed to prevent failure and collapse (Enders & Remig, 2014;Grober, 2007). However, stability in the banking system is attained when the banking system has the ability to withstand shocks arising from internal and external economic imbalances or any kind of volatility in their business operations.
The financial system regulators are aware of the overwhelming negative consequences of loss of confidence in the banking system (Ingves, 2016; Thomson, 1992). Hence, regulators worldwide rate banking stability as a top supervisory and regulatory policy objective. Perhaps, after South Africa and Egypt, Nigeria is next with a growing banking sector in Africa. Unlike the developed economies, significant issues in the literature that are often underreported are the factors that determine the banking system stability in emerging economies, owing to the fact that developed economies have more sophisticated financial structures than those economies yet to mature (Ozili, 2018 The Sustainable Development Goals agenda was fully incorporated into the economic and environmental sustainability agenda, alongside the aspirations for progressive societies. The SDGs are more committed and ambitious, in particular in seeking to eradicate rather than diminish hunger, poverty and as a world-wide agenda. The 17 SDGs are to be nationally owned and are already universally accepted (Jeucken, 2001). There is also a great deal of energy and momentum surrounding the achievement of these goals with national ownership and a will to bring together the public sector (government), private sector, civil society and individuals to achieve the goals with financing targets.
While the public capital is scarce in supply, private finance is constrained by uncertainty, and risk and return requirements. There is a need for serious improvement of managing financial complexities of banks (Nwachukwu, 2014;Soludo, 2004). Furthermore, the banking system through its intermediation role can mobilize funds that account for part of the investment and funding the SDGs (Ziolo et al., 2018). The questions are: Do banks see or understand their role in the financing need for the accomplishment of the SDG objectives? Will the hostile business environments hinder banks from effectively carrying out this funding responsibility? How will the regulatory authorities ensure that complying with this funding role will not lead to infringement of the regulatory requirements? Finally, are deposit money banks (DMBs) in Nigeria ready to finance SDGs?
The answers to the outlined queries are key for effective development of the strategies that will ensure a sound and sustained banking system and ensure a successful and efficient mobilization of adequate capital that will be used in financing SDGs throughout its duration without a threat to the system. Generally, only about 48% of SDGs investment requirements can actually be covered in emerging economies. For example, in 2016, Multilateral Development Banks (MDBs) reportedly mobilized a total of USD163.6 billion in private co-funding with low-income economies representing 4% (USD 5.9 billion). Middle-income economies were a significant amount of 40% (USD 65.2 billion), while high-income economies alone represent 56%, (USD 92.5 billion). The estimate revealed that the funding gap is above USD 2.5 trillion for all emerging economies and about USD 1.3 trillion specifically for African countries (Adams, 2017;Bordon & Schmitz, 2015;Akintoye & Opeyemi, 2014;Jaiyesimi, 2016;Kharas et al., 2014;Haigh, 2012;Jeucken, 2001). This study aims to contribute to the achievement of the Sustainable Development Goals, especially SDGs 8 and 9, in Nigeria by demonstrating the ability of the banking system, through their intermediation role, to finance the SDGs while maintaining the required level of stability and soundness.

LITERATURE REVIEW
The stability of the banking system connotes a system that is at its upmost performance level and functions without being impaired or degraded by disruptions or volatility from its external and/ or internal environment (Adams, 2017). Besides, banking stability is mostly considered a continuum, especially in terms of being consistent with multiple combinations of the fundamental elements of finance (Novotny-Farkas, 2016). Globally, the investment required to attain the SDGs are huge, with the majority of the emerging economies, specifically because of their poor infrastructural development, thereby, making the scale of existing financial flows insufficient (Pisano et al., 2012;Sadiq & Mushtaq, 2015). The only channel for closing the resulting funding gap for the attainment of SDGs' objectives is via a sustained stability in the banking system, which will be able to provide and stand the pressure of finance requirements (Stenberg et al., 2017;Weber, 2014

The link between the banking system and the SDGs
The SGDs of interest to this study are 8 and 9. SDG 8 is particularly concerned with the funding of the real economy for at least 7% growth with full productive employment that emphasizes real growth rate for an employed person. SDG 9 focuses on infrastructure development, innovation and industrialization. Banks need to tune up to be sustainably involved in SDG 9 as they have been involved in SDG 8 as matter of practice. However, three key aspects are noticeable in the link between sustainable development and the banking system. The first involves the environmental regulations introduced that affect the banking system in several ways. The second issue concerns credit risk management over which the-banking system has continued to be reactive rather than proactive regarding challenges with respect to non-performing loans growth. The third is with respect to the stakeholders' pressures as it deals with the bank's reputational risk (Baranes, 2009; Egede & Lee, 2007; Thompson & Cowton, 2004). Figure 1 provides a framework for a stable banking system and SDG funding, showing key operational activities in which banks must be involved consistently to remain sustainable in order to successfully provide sufficient funding for the SDGs up to 2030.

Review of the empirical literature
The first discussion is on banking stability, while the second is on funding the SDGs. Dwumfour (2017) examined banking stability in sub-Saharan Africa. Using Z-score to proxy stability, the result showed that crisis and high percentage of foreign banks reduce bank stability, while diversification positively affects stability. The results support largely the competition-fragility view. Therefore, the less the competition amongst banks during crises periods, the more banks achieve stability. Similarly, Ozili (2019) used aggregate data to analyze the determinants of banking stability. The result revealed that the level of non-performing loans, bank efficiency and regulatory capital are the major determinants of banking stability in Nigeria. Toader et al. (2018) studied the influence of corruption on banking stability in European emerging markets. The result indicates that lower levels of corruption impact positively bank stability leading to reduced credit losses with more realistic credit growth. The result showed further that the stability of banks operating in the countries that have not adopted corporate governance have higher impact of corruption. In addition, Sere-Source: Author's compilation (2020).

RESEARCH METHODS
The stability of the banking system is measured as a composite index, derived from a standardized process and weighted mean of three indices, which includes sub-index of economic climate, banking soundness and bank vulnerability to externalities (Cheang & Choy, 2010 where ω p -weight attached to a single sub-index indicating its comparative importance.
Statistical normalization transforms the indicators to a mutual scale with a zero mean, and its standard deviation equals one. The standard deviation is the scaling factor, and the model is as shown below. Using the statistical method, the BSSI is obtained by computing the weighted averages of the three sub-indices that emerged from the normalization process, namely: where X t depicts the value of the indicators (X) during the period t, µ represents the mean, and σ indicates the standard deviation.
The BSSI is developed through weighting: where r = {s, v, c}; µ i represents responses that reverted high in during the consolidated responses from U and FGD, which depicts the correct number of indicators in each sub-index.
The adapted model is stated as follows: where GDPPC and CBLSE (the indicators of SDGs) represent the dependent variable vector, Y denotes the regressors (the banking system stability indicators), p implies the lagged structure, ∆ represents the variation in terms, β 1 and β 2 indicate the long-run coefficients, α 1 and α 2 are the short-run coefficients, and ε t represents the stochastic error term.
Explicitly, the models are estimated as follows: where ∆ indicates the first-difference operator and p is the maximum lag order. The existence of co-integration amongst variables is tested using the F-statistics. The coefficients: ϕ, σ, α, λ, β, ∂ and α that match the dynamic forces of the short-run model, however δ 1 , δ 2 , δ 3 , δ 4 , and δ 5 represent the long-run association.
Therefore, as co-integration is present, long-run measurement will be determined, which is stated as: In an effort to identify the correct ARDL model, Liew (2004) used a selection lag criterion, the Akaike Information Criterion. According to Liew (2004), this type of information criterion could be used if the sample size is below 60. This is a maximum lag of 2 as established by Pesaran et al. (1999). After this, the study models the short-run equations derived from the ECM:  The short-run model coefficients are usually constants that explain the underlying forces of the model and highlight meeting point of the model at ∂. This indicates that the re-parameterization of errors generated at a specific period is corrected in a subsequent period, while the long-run association tests will be inconclusive where F statistic lies between the upper and lower bounds.

Data description and estimation methods
The study used annual time series data capturing the domestic and global financial crisis, including several periods of structural reforms in Nigeria.     Table 3 shows the correlation analysis of the variables used to check the likelihood of the existence of multicollinearity in the ARDL result. The correlation coefficients result of the indicators of SDGs are in agreement with the a priori expectation. Table 4 depicts the unit root test. The result shows that commercial banks' loans to small enterprises, gross domestic product per capita, liquid assets to total assets, total non-performing loans to total gross loans and external assets to total assets seem not to reject the null hypothesis "no stationary" at levels, whereas return on assets and total loans to total deposits were stationary at levels I(0). However, after numerous re-statements on differencing and the length of lag, the series were seen to have finally rejected the null hypothesis at first difference I(1). This implies that the first difference of the series is mean reverting and stationary. Hence, commercial banks' loans to small scale enterprises, gross domestic product per capita, liquid assets to total assets, financial deepening, total non-performing loans to total gross Loans, Credit to Private Sector and external assets to total assets are integrated of order I(1). Thus, this argument stimulates the co-integration test to evaluate whether or not the linear grouping of the considered stability and SDGs indicators yields any stationary residual. Table 5 shows the order of the ARDL models I & II selected by AIC for the co-integration test. The results indicate that computed F-statistics for both models are higher than the upper bound critical values. Thus, the null hypotheses of no co-integration at the 5% significance level is rejected. This supports the idea of a stable and unique long-run Table 3. Correlation matrix Source: Authors' compilation (2020). LCBLSE  GDPPC  EATA  LATA  ROA  FND  LCPS  TLTD  TNTL   LCBLSE  1.0000  ------

DISCUSSION
Tables 6, 7 and 8 present the long-run appraisal, the short-run analysis, and the diagnostic and stability tests. Table 6 reveals in model I that external assets to total assets, financial deepening, credit to private sector and total non-performing loans to total gross loans have a positive and significant impact on the SDG 8 funding in Nigeria at the 5% critical level. Whereas liquid assets to total assets (proxy for SDG 9), return on assets and liquid assets to total assets bore a negative but significant influence at a 5% critical level. Similarly, model II shows that external assets to total assets, liquid assets to total assets and credit to private sector have a positive and significant impact on SDG funding at 0.1 and 0.05. Credit to private sector, ROA, liquid assets to total assets and total non-performing loans to total gross loans have a negative and significant influence on SDG funding at 0.1 and 0.05 confidence levels. Largely, this shows that 10% change in external assets to total assets, financial deepening, credit to private sector and total non-performing loans to total gross loans can positively increase SDG funding by 162.13%, 0.28%, 0.86% and 0.20%, respectively. For model II, a 10% change in external assets to total assets, liquid assets to total assets and credit to private sector will positively improve SDGs funding by 24.12%, 9.63% and 0.31%, respectively. This result is in line with Bordon and Schmitz (2015), Sadiq and Mushtaq (2015) and Pisano et al. (2012) that banking stability is a key factor that can drive sustainable development funding in an economy.
The direct relationship of liquid assets to total assets, return on assets and liquid assets to total assets is an indicator of banking stability, which turned out to be negative and significant for model I. In model II, return on assets, credit to private sector, liquid assets to total assets and total non-performing loans to total gross loans have a negative significant relationship with SDGs funding. The effect is that Nigeria's banking system requires a huge variety of sustainable products and a market for diversification and expansion of Note: ***, ** and * depict rejection of null hypothesis at 1%, 5% and 10% significance levels, respectively. businesses in order to remain stable to fund the SDGs. In addition, the positive influence of FND and LCPS shows that there is a need for enabling business and operating environments that propels business development and growth to aid enhancement of Nigeria's banking stability. Thus, there is a need to drive business strategies that foster an efficient business environment that will boost the SDG agenda in Nigeria.
The short-run dynamics estimations in Table 7 shows both negative and positive associations between the lags of the regressors. The coefficient values for the ECM(-1) for both models infer that the result is in conformity with expectation, since it is negative and significant.   Figure 3 present the results of the stability and diagnostic tests. The estimated models passed all the diagnostic tests, which indicate that  the error terms have similar variance and that they are uncorrelated and normally distributed. Cumulative sum of squares and cumulative re-sults fall within the critical bounds at the 5% significance level, which means that all parameters are stable over study periods.

CONCLUSION AND RECOMMENDATIONS
This paper aims to examine the ability of banks to finance key SDG areas, especially SDG 8 and 9, while maintaining the required stability. Proxies for the SDGs were credit to the private sector, loans to small businesses (for SDG 8), while total loans to total deposits were proxied for SDG9. An analysis was conducted using panel data for 1992 to 2019. The study employed a composite index derived from a standardized process and a weighted mean of three indices, including a sub-index of economic climate, banking soundness and bank vulnerability to externalities for banking system stability indicators. In addition, the panel data with two different models were analyzed using the Autoregressive Distributed Lag (ARDL). Several diagnostic and stability tests were used to confirm the results. The test results show that the banking system can contribute to the SDG funding with banking system stability achieved and sustained. Although, the effects of return on assets on its own seems not to have enhanced sustainable development funding, the extent of responsiveness of sustainable development to the variants of ROA for both models seems to be inelastic. This implies that a proportionate variation or change in Nigeria's banking system stability would lead to a more or additional proportionate change in financing SDGs. The tests revealed that banking system stability has a significant and positive link with financing SDGs. This suggests that as banking system stability improves in Nigeria, it will enhance the capacity of the banking system generally to successfully finance SDGs throughout the estimated period. Thus, the degree of bank soundness and the consistent long-term stability are important contributing factors to successful funding of SDGs in Nigeria. Based on the results of the estimation, this study concludes that banking system stability in Nigeria can enhance the funding of the SDGs and still be stable for the performance of its total intermediation functions in the economy. A novelty of these results is that despite its seeming fragility, banks can sustainably fund SDGs 8 and 9 if it can take care of non-performing loans.
This study recommends the following: First, the banking system has to actively pursue opportunities for more innovative products allowing the bank to actually create a blended or mixed value return, which Source: Authors' computation (2020).