Bank capital and risk in emerging banking of Jordan: a simultaneous approach

Abstract

Financial risk has received increasing attention from policymakers and financial institutions. Therefore, the present study examines the relationship between capital and risk for Jordanian banks by using data from 2010–2019. The study employs fixed effect, random effect, GMM, and 3SLS. Our findings show that the capital requirement regulation has a positive impact on capital and risk rates. Moreover, the study also concludes that Jordanian banks hold more than the minimum regulatory capital requirements laid down by Basel II, III, and the CBJ. The banking sector increases its capital adequacy by raising its liquidity and reducing its tendency to take risks. Our results indicate a highly significant negative relationship between Jordanian commercial bank capital and risk. Liquidity risk, ROA and stock market capitalization are positively related to bank capital. The results of the study suggest that Jordanian banks should be involved in higher-risk lending actions and help increase competition in the banking sector.

Impact Statement

The banking sector plays a vital role in economic growth and bank capital serves as a buffer at the time of economic shock. Similarly, risk management is also crucial for the sustainable financial sector and economic development. Thus, this study employs a simultaneous approach in developing a relationship between bank capital and risk in the Jordanian banking sector. Our findings show that the capital requirement regulation has a positive impact on capital and risk rates. Our results also indicate a highly significant negative relationship between Jordanian commercial bank capital and risk. Studying the association between capital and risk offers insights into how banks manage and reduce different risks, helping to develop efficient risk management strategies. After studying this relationship banks can also optimize their capital allocation strategies and can shed a light on how capital can affect the lending power and credit availability to boost the finance in industries and thus contribute to economic growth.

Keywords:

• Capital requirements
• risk-taking behaviour
• capital regulations
• Jordan

REVIEWING EDITOR:

• David McMillan, University of Stirling, Stirling, UK

SUBJECTS:

• Economics
• Finance
• Business, Management and Accounting

JEL CLASSIFICATION:

• G21
• G32
• G33

1. Introduction

Financial risk has received increasing attention from policymakers and financial institutions, as large losses have resulted in many large banks failing, particularly in the last global financial crisis, which may require effective risk management in the banking sector. These trends have had adverse effects in the banking sector services, prompting bank regulatory to take various steps to control the market. The recent financial crisis highlighted the need for the financial system’s stability, particularly in relation to bank risk and capital control as an external source. In the banking sector, understanding the connection between capital and risk decisions is very relevant. Therefore, the adjustment and correcting regulations in the banking area, the investigation of the underlying mechanisms should receive a big attention. In this paper, we will explore the most important sources of risks in the Jordanian banking sector. The adjustment and correction of banking regulations should therefore focus great attention on research into the underlying mechanisms.

During the study period, the capital adequacy ratio (CAR) and share of the core capital of tier one is very closer in Jordanian banking system, indicates the highest quality capital component and the most capable of absorbing financial risk and losses. Moreover, the stock of shareholders in licensed banking companies grew 3.6% in 2015. These results show the sustainability of the banking system 's solvency. The capital share of immigrants in the total capital of licensed banks in the Jordanian banking system was approximately 49% at the end of 2015. Legal reserve requirements by CBJ were originally 12% before going down to the outstanding rate of 7%, as part of the expansionary measures adopted to encounter the global financial crisis since 2009.

In conclusion, we can say that the Jordanian banking system did not face any problems in implementing the minimum capital requirement, Jordanian Banking sector easily satisfy the Basel capital standards and the CBJ capital standards which is higher than the Basel standards. The increase in the Jordanian banking system capital over time has improved the credit facilities offered to the individual, private and public investment. On the other hand, we can conclude that how the Jordanian banks took advantages from its high CAR to withstand the unstable political environment and risk. The high level of CAR has an impact on the decrease of non-performing loans because of the leverage ratio increased to 12.7 percent, which was the minimum level set by the CBJ 6%. As a result of their appropriately high CARs, which are the highest in the region, the Jordanian banking system is not likely to suffer any problems or difficulties in implementing Basel III capital requirements.

Previous studies elaborate that regulation significantly affect the decisions regarding risk and capital. Bank failures led the Basel Committee for Banking Supervision (SCBS) in 1989 to announce in the wake of the recent subprime bubble crisis that risk-based capital criteria are implemented to authorize the banks’ asset portfolio capital requirements (minimum 8%) to meet risks. In order to enhance capital quantity and quality in the bank sector, the BCBS proposed a new regulatory for capital framework.

The authorities implement the Capital Regulation to guarantee that the chance of default is kept to a low in order to limit bank risk with respect to the stability of the financial system. Kahane (1977) claimed that by setting a minimum capital requirement for banks, new capital laws will reduce bank risk and ensure the stability of the banking system. The capital regulations enforced by the authorities could lead to distortions in banking behavior (Blum, 1999). The banks must adjust their risk and capital to the optimal amount of leverage, which means that banks will also raise the risk if it raises its liquidity to meet the new capital requirements (Koehn & Santomero, 1980). Regulators can precise this by introducing measures to restrict the risk of banks and increase their supervisory and supervised banking systems (Kahane, 1977; Kim & Santomero, 1988; Gennotte & Pyle, 1991). Changes in regulation positively affect both capital and risk decisions (Shrieves & Dahl, 1992; Jacques & Nigro, 1997); while few studies reported the negative impact of the change in the regulation on bank capital ratio (Stolz, et al., 2003; Skully et al., 2009). Whereas few reported a positive and insignificant effect of regulation on bank capital (Van Roy, 2005).

Another factor that is likely to have a negative impact on risk is bank size, as large banks have lower risk exposure due to their increased diversification (Lindquist, 2004; Van Roy, 2005). Numerous studies on the factors influencing the capital ratio have discovered a negative relationship between size and capital ratio (for example, see Jacques & Nigro, 1997; Das & Ghosh, 2004; Van Roy, 2008; Berger et al, 2008; Floquet & Biekpe 2008; Skully et al., 2009). While others found a positive impact of bank size on capital (Gropp & Heider, 2010). The capital ratio is also thought to be significantly influenced by liquidity, though some studies have found that there is no relationship between the two variables (Jokipii & Milne, 2011; Athanasoglou). There are fewer studies that support the idea that banks keep liquidity on hand to act as a shock absorber, which would eliminate the need for additional capital (Jokipii & Milne, 2011). Earnings are another element that influences a bank’s capital ratio, with high profitability causing banks to increase their capital ratio. Banks’ less efficient decisions lead to higher risk exposure since efficiency levels impact both capital and risk decisions.

It is very essential to explore the capital-risk relationship as well as how the regulation influences the risk and capital changes in the banking system. It is very essential to study the development of the banking sectors, where banks are the main source of funding. This study has a potential implication for banking development, financial stability and development of emerging economies. Studying the association between capital and risk in these situations offers insights into how banks manage and reduce different risks, helping to develop efficient risk management strategies. After studying this relationship banks can also optimize their capital allocation strategies, and can shed a light on how capital can affect the lending power and credit availability to boost the finance in industries and thus contribute in economic growth.

Past studies show the heterogeneous link between risk and bank capital. Some of them found a negative relationship and some found the positive relationship. Starting with the inverse relationship, the explanation focuses on deposit insurance as a dominant bank behaviour, resulting in lower capital and higher levels of risk (Black et al., 1978; Kareken & Wallace, 1978; Merton, 1972). While regulatory pressure can explain the positive link between bank capital and risk, where the regulation encourages banks, the capital of which has increased to maintain more risky investments and a higher level of risk increases the capital level of a bank (Buser et al, 1981). In this article, we will explore the most significant factors of risk in the banking sector of Jordan. Present study leads to a new problem focused on the relationship between bank risk and capital using mixed methods. In addition, this study shows the impact of new capital requirements regulations on the bank risk. For this purpose, simultaneous equation models are used to examine the relationship between bank capital and risk intensively and distinctively, using FE, RE, GMM, and 3SLS.

Therefore, the current study offers an empirical analysis into how risk and capital are affected by banking sector regulation. The remainder of the study is organized as follows: A brief overview of the literature is presented in part 2, a brief history of Jordan’s banking industry is given in section 3, and data, variables, and methodology are discussed in section 4. Section 5 of the report discusses the empirical findings. Section 6 provides a summary of the findings.

2. Literature review

Soundness is a key factor in the banking system’s success, especially the capital regulation that aims to control the different risks that bank faces. Capital ratio is one of the major decisions the bank is focusing on to ensure banking sector stability. Capital and risk are one of the central topics in the literature on banking and research, following the most recent financial crisis. This section concentrates on the major literature works, which investigated the role of regulation on capital and risk determinations of banks. Many empirical studies have studied the actions of the banks to prove the impact of regulation on risk and capital. The sound impacts of capital requirements in literature have been thoroughly deliberate and theories claim that unwarranted capital requirements can lead to excessively high-risk management (see Kahane, 1977; Koehn & Santomero, 1980; Kim & Santomero, 1988).

Banks seem to change their capital ratios through capital raising and lessening the risk, as regulatory pressure increases (Jacques & Nigro, 1997; Shrieves & Dahl, 1992). While, some studies contrast to this evidence (Stolz et al., 2003; Van Roy, 2005). Increased capitalization improved the average return on assets (Trujillo-Ponce, 2013), while in contrast, Goddard et al. (2004) found the negative relationship between bank capital and return on assets for the European banking system.

According to several earlier research, bank capital and risk changes frequently have a sizable impact on the banks’ ex-ante capital level (Iwatsubo, 2007; Distinguin et al., 2013). Numerous earlier research revealed that banks with low capital levels tend to increase their risk when capital ratios decline, thereby increasing the likelihood of default. Rime (2001) has identified that Swiss banks have a positive and significant impact on the risk-taking of banks in line with the minimum regulatory capital requirement and the regulatory pressure. Stolz et al. (2003) shows that banks with low capital ratio try at the same time to raise capital and reduce risk to coordinate capital and risk. In addition, Konishi and Yasuda (2004) from the Japanese banks found that adding a criterion for capital adequacy results in lowering risk-taking.

Homölle (2004) found that certain banks have increased risk for assets due to the introduction of strict capital requirements. Hussain and Hassan (2005) determined that capital regulation did not improve the share ratios but did minimize portfolio risks for banks in developed countries. Jokipii and Milne (2008) found a positive association between capital adequacy and risk, with bank capitalization affecting capital adjustment and risk. Alkadamani (2015) found that regulatory pressure positively impacts bank capital during the financial crisis. Risk-based banking investment requirements’ implications on wealth were examined by Eyssell and Arshadi (1990). According to their research, trading banks’ capital decreased as a result of the new risk-based capital’s creation. Due to low capital ratios, banks have also experienced the greatest income losses.

Shrieves and Dahl (1992) examined how capital and risk are related in a sizable sample of the U.S. banking sector spanning the years 1983 and 1987. They noticed that the disclosure of risk and the amount of capital are both inter-related. In addition, Blum (1999) explored the positive association between capital and risk within banks, suggests that risk-taking activity is often limited not solely due to regulatory control but by bank owners or managers. Ediz et al. (1998) looked at the impact of risk-based capital regulations using quarterly data from the UK from 1989 to 1995. The study’s findings suggested that maintaining capital is a useful technique for enhancing bank stability without influencing banks’ lending decisions. The findings demonstrate that the UK capital requirements appear to be a desirable regulatory instrument since they improve the stability of the banking system without influencing banks’ lending decisions.

Park (1999) examined the effects of the asymmetric information and capital adequacy on portfolio decisions by use of OLS and 2SLS in US. The empirical findings suggest that asymmetrical information influences bank decisions. Similarly, Berger (1995) explains that capital requirements based on risk had a minor effect on growth of credit. He also found that banks with low capital tend to reduce loans. Rime (2001) found a positive association between risk changes and capital ratios in 154 banks, with regulatory pressure positively affecting the RWA capital ratio but not significantly affecting bank risk. They also explored that change in capital is negatively related with capital to RWA and capital to total assets but positively related with risk.

Francis and Osborne (2010) examined banks’ risk-based capital ratio decisions and the impact of regulatory capital requirements on capital ratios. They found that regulations affect capital management practices and capital ratios fluctuate throughout the economic cycle. Empirical studies in the U.S. banking sector have found that both capital and risk-based capital ratios are significant predictors of stock return volatility and bank Z-scores. The capital ratio is found to be better than the risk-based capital ratio, as it predicts stock-return volatility and bank Z-scores, aligning with previous research (Hogan, 2015).

A study was conducted on the tobacco industry in Pakistan to find out how financing decisions affect corporate liquidity and growth in tobacco industry. It was found that tobacco industry makes use of debt financing and making use of this leverage to enjoy high profits (Salman, 2019). Salman and Munir (2012) conducted a study on the cement industry in Pakistan for investigating debt and equity financing impact on business performance. It was found that equity financing has positive impact on business performance. Another study was conducted on pharmaceutical industry of Pakistan to analyse performance by using financial analysis. It was revealed that performance was improved over a period of time as this is considered as one of the indicator for the stakeholders (Salman & Qamar, 2011). There is also literature available on green finance and green behaviour and how it impacts the performance. A study was conducted on Fashion industry of Malaysia and it was revealed that consumers’ green behaviour impacts the corporate performance. This study helps to understand the influence of consumer’s behaviour on corporate performance (Ali et al., 2020).

Abbas et al. (2019) studied the relationship between bank capital buffer on risk in US and study revealed that the capital buffer has varying and negative effects on total risk during the crisis, pre-crisis, and post-crisis. Da Fonseca Nicolay et al. (2018) found that during economic booms, banks reduce capital buffers and increase risk exposure. On the contrary, Mahdi and Abbes (2018) determine a positive bidirectional link between Islamic banks’ capital and risk. Their findings indicate the risky nature of this bank category, which is mostly due to the Sharia-compliant contracts used, such as Moudharaba and Moucharaka contracts. Furthermore, a study by Moudud-Ul-Huq (2019) found that higher capital buffers and efficiency promote each other, lowering the likelihood of bank failure (due to non-performing loans), even though increased risk halts efficiency. As a result, banks’ greater stability increases capital buffers. While fewer studies found no significant impact of bank capital on risk (e.g., Bitar et al., 2018).

The relationship between risk and capital ratio is also widely discussed in us. Previous findings show that capital ratio is positively related to risk while liquidity is negatively related to banks risk in US commercial banks (Abbas et al., 2021; Abbas & Ali, 2022). In another study on US commercial banks Abbas and Ali (2020) found that risk-based capital is negatively related to banks risk while non-risk capital is positively related to bank risk.

Previous studies have either examined the relationship of bank capital on risk and performance, capital requirements on bank risk, or impact of risk on bank capital and performance related measures. But this study has contributed in the previous knowledge by adding a simultaneous relationship between bank capital and risk, especially in the Jordan. Based on the literature mentioned above. The following alternate hypotheses have been suggested for Jordanian banks.

H1: There is a positive and significant impact of capital on bank risk taking.

H2: There is a negative and significant impact of risk taking on bank capital.

3. Capital and risk in the Jordanian banking sector

Despite of increasing risk in global financial sector and issues in the Middle East region, such as the political and economic issues that impacted the Jordanian economy, Jordan has continued to perform satisfactorily. Jordan’s real GDP achieved moderate growth of 2.4% in 2015 (WDI). During the global financial crisis, certain banks around the world lacked the capital and liquidity they needed to face risks. CAR proved to be a valuable instrument to increase the capacity of banks to withstand financial shocks and high risks. The Basel Committee for Banking Supervision made a number of changes concerning Basel II at the end of 2010 with the purpose of strengthening the capital base of banks in qualitative and quantitative terms, raising liquidity rates, and increasing their risk-taking capacity (Basel III).

As far as Jordanian banking is concerned, the sector is generally stable. In addition to the convenient levels of liquidity, the Jordanian banking system has the highest and most satisfactory capital levels in the Middle East and North Africa (MENA) region, which allows it to cope with shocks and great risks. At the end of 2015, Jordan’s financial sector comprised licensed banks with total assets equal to JD48.0 billion (USD 67.6 billion). The assets of licensed banks made up 94.1% of the total assets. The assets of licensed Jordanian banks reached JD 45.2 billion (USD 63.6 billion)¹, representing 169.7% of GDP in the last of 2015.

Jordan’s banking system starts implementing Basel II capital adequacy standards in 2008. The Jordanian banking system meets CBJ and Basel 2 requirements, with a draft instruction for implementing Basle III capital requirements. The capital adequacy ratio and Tier 1 core capital share are similar, with the Tier 1 capital ratio accounting for 97.17% of the CAR. This high-quality capital component is capable of handling financial risk and shocks, ensuring banks’ safety and durability. The capital adequacy ratio for Jordanian banks was above the Basel equity ratio of 8%. That shows that banks have a strong capability, firstly, to absorb any loans or shocks that could arise by internal or external factors that cannot be controlled.

The banking system in Jordan, however, had sufficient capital to counter any financial crisis consequences. Year after year, there has been a substantial increase in the risk of lending to this sector in the ratio between individuals’ debt to their income and net wealth, requiring banks to consider the risk of lending and expand their study to address this risk evolution. More than 35,6% of the total facilities provided by banks accounted for real estate facilities. The real state guarantees cover 151% of the volume of property, which improves banks’ capacity to deal with risks.

The banking and financial sectors is generally stable in Jordan. In the Middle East and North Africa, Jordan boasts a strong banking system that can withstand high shocks and risks, as well as convenient liquidity and profitability levels. The following signs indicate that the Jordanian banking system is resilient and secure and can withstand shocks and high risks:

• The high capital adequacy ratio in the MENA region (after the UAE). CAR in the Jordanian banking industry ranged from 15.9% to 19.1% in years 2003–2015 (CBJ Reports).

• As far as the liquidity ratio is concerned, Jordan’s banking system enjoys a stable liquidity role as highly liquid reserves accounted for around 151.3% at the end of 2015 compared to 179.6% in 2003, which is more than the minimum ratio set by the CBJ of 100% and representing relative stability in the amount of bank liquidity.

• In 2015, the ROA for the banking sector in Jordan was 1.1% compared to 0.7% in 2003. The return on equity (ROE) amounted to approximately 10.3% by the end of 2015, compared to 9.9% in 2003. The financial sector has achieved a high degree of profitability. Jordan’s banking system’s rate of return on assets was 1.3 percent in 2015, comparatively low from 1.4% in 2014 and the return on equity was 10.3% in 2015 as compared to 11% in 2014. The low ratio is partly attributed to the rise in bank income tax rates, in addition to the risk-averse approach followed by Jordan’s banking sector (CBJ Reports).

• Finally, the Jordanian banking system has shown a strong solvency ratio, which increased from 7.5% in 2003 to 12.70% in 2015 (the Basel minimum leverage ratio was 3%).

4. Data

The article aims to analyse the two models below, showing the association between capital and risk in Jordan’s commercial banks between 2010 and 2019 and shed light on how the Jordanian commercial banks react to the regulation on their capital. The data were collected from the Amman Stock Exchange, Association of Banks in Jordan, the Central Bank of Jordan, and the commercial Jordanian bank’s annual reports.

4.1. Variable description

In this study, profitability measures and bank-specific variables are considered as key variables. This article examines the impact of regulatory pressure on Jordan’s banks’ capital and risk. To test this relationship, the data from 11 commercial banks of Jordan were collected over the period of 2010–2019. Eleven banks (all local commercial banks in Jordan) are already listed in the Amman Stock Exchange (ASE). Hence, the performance of these banks is highly important for shareholders. Due to the availability of all information for all banks, balanced panel data is used for analysis. In total, the sample contains 110 observations. Bank-specific information (micro-information), industry-specific variables are mainly gathered and collected from banks’ annual reports published in Bloomberg, ASE, CBJ, and Jordan Bank Association (ABJ). Unstructured interviews with CBJ risk managers and various commercial banks were also conducted. The purpose of these interviews was to provide the Jordanian banking system with a broad understanding of the risk-based capital requirements. The budget sheets, income statement, and notes were examined to obtain data on the variables mentioned in the model. For selected variables, all values taken are in JD. Eleven banks representing the majority of the assets in the Jordanian banking system were chosen based on the data that were available. The list of the variables used are presented in (Table 1). The study uses STATA software for analysing the data and producing the regression results.

4.2. Methodology

We utilized the simultaneous equation model established by Jacques and Nigro (1997) and Rime (2001) to evaluate the link between capital and risk. By adding the process of adjustment to bank capital, the model is updated in line with capital regulation, and risk levels are split into two variables (endogenous and exogenous). The model is written as below: (1) $$\Delta {\text{CAR}}_{\text{it}}=\Delta {\text{dCAR}}_{\text{it}}+{\text{X}}_{\mathrm{i},\mathrm{t}}$$(1) (2) $$\Delta {\text{RISR}}_{\text{it}}=\Delta {\text{dRISK}}_{\text{it}}+{\text{Z}}_{\mathrm{i},\mathrm{t}}$$(2) where ΔCAR_it and ΔRISK_it are the observed variations in capital and risk levels for bank ‘i’ in period ‘t’, respectively, while X_it and Z_it are the exogenously defined variables. The ΔdCAR_it and ΔdRISK_it are endogenously determined variations in capital and risk that are developed using the framework for partial adjustment. The partial adjustment framework uses discretionary changes in capital and risk (ΔdCAP_i,t and ΔdRISK_i,t) to help banks aim for optimal capital and risk levels. Banks adjust their capital and risk to their targets when exogenous shocks remove real target levels, observing only partial changes to their target level. The discretionary capital-risk changes are proportional to the difference between target and ‘t-1’ level. (3) $$\Delta {\text{dCAR}}_{\text{it}}={\alpha }_0\left(\text{CAR}{*}_{\text{it}}–{\text{CAR}}_{\text{it}-1}\right)$$(3) (4) $$\Delta {\text{dRISK}}_{\text{it}}={\beta }_0\left(\text{RISK}{*}_{\text{it}}–{\text{RISK}}_{\text{it}-1}\right)$$(4) where, CAR*_it and RISK*_it denote the banks’ target level of capital and risk. By substituting equations ‘3’ and ‘4’ into equations ‘1’ and ‘2’, the observed capital and risk changes will be written as follows: (5) $$\Delta {\text{CAR}}_{\text{it}}={\alpha }_0\left(\text{CAR}{*}_{\text{it}}–{\text{CAR}}_{\text{it}-1}\right)+\text{Xi},\mathrm{t}$$(5) (6) $$\Delta {\text{RISK}}_{\text{it}}={\beta }_0(\text{RISK}{*}_{\text{it}}–{\text{RISK}}_{\text{it}-1)}+\text{Zi},\mathrm{t}$$(6)

Eqs. (5)(5) $$\Delta {\text{CAR}}_{\text{it}}={\alpha }_0\left(\text{CAR}{*}_{\text{it}}–{\text{CAR}}_{\text{it}-1}\right)+\text{Xi},\mathrm{t}$$(5) and (6)(6) $$\Delta {\text{RISK}}_{\text{it}}={\beta }_0(\text{RISK}{*}_{\text{it}}–{\text{RISK}}_{\text{it}-1)}+\text{Zi},\mathrm{t}$$(6) show that observed changes in capital and risk depend on the target level of capital and risks along with the lagged value of capital and risk.

Additionally, Eqs. (7)(7) $$\Delta {\text{CAP}}_{\text{it}}={\alpha }_0+{\alpha }_1\text{REGi}{,}_{\mathrm{t}-1}+{\alpha }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\alpha }_3{\text{ROA}}_{\mathrm{i},\mathrm{t}}+{\alpha }_4\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}-{\alpha }_6\text{CAPi}{,}_{\mathrm{t}-1}+{\epsilon }_{\mathrm{t}}$$(7) and (8)(8) $$\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1\text{REGi}{,}_{\mathrm{t}-1}+{\beta }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\text{LLP}}_{\mathrm{i},\mathrm{t}}+{\beta }_4\Delta {\text{CAP}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}–{\beta }_6{\text{RISK}}_{\mathrm{i},\mathrm{t}-1}+{\text{v}}_{\mathrm{t}}$$(8) , which depend on desired capital and risk levels, estimate changes in capital and risk during the time period ‘t’. Bank capital and risk are influenced by exogenous and endogenous variables, which makes it challenging to predict them and are the vectors of coefficients that need to be estimated, while ԑ_t and v_t, respectively, are error terms of the two equations. (7) $$\Delta {\text{CAP}}_{\text{it}}={\alpha }_0+{\alpha }_1\text{REGi}{,}_{\mathrm{t}-1}+{\alpha }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\alpha }_3{\text{ROA}}_{\mathrm{i},\mathrm{t}}+{\alpha }_4\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}-{\alpha }_6\text{CAPi}{,}_{\mathrm{t}-1}+{\epsilon }_{\mathrm{t}}$$(7) (8) $$\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1\text{REGi}{,}_{\mathrm{t}-1}+{\beta }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\text{LLP}}_{\mathrm{i},\mathrm{t}}+{\beta }_4\Delta {\text{CAP}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}–{\beta }_6{\text{RISK}}_{\mathrm{i},\mathrm{t}-1}+{\text{v}}_{\mathrm{t}}$$(8)

The equations indicate that changes in risk and capital levels in period ‘t’ are influenced by differences between the target level and actual capital and risk in the previous period.

The study used panel GMM, and 3SLS. The two Eqs. (7)(7) $$\Delta {\text{CAP}}_{\text{it}}={\alpha }_0+{\alpha }_1\text{REGi}{,}_{\mathrm{t}-1}+{\alpha }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\alpha }_3{\text{ROA}}_{\mathrm{i},\mathrm{t}}+{\alpha }_4\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}-{\alpha }_6\text{CAPi}{,}_{\mathrm{t}-1}+{\epsilon }_{\mathrm{t}}$$(7) and (8)(8) $$\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1\text{REGi}{,}_{\mathrm{t}-1}+{\beta }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\text{LLP}}_{\mathrm{i},\mathrm{t}}+{\beta }_4\Delta {\text{CAP}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}–{\beta }_6{\text{RISK}}_{\mathrm{i},\mathrm{t}-1}+{\text{v}}_{\mathrm{t}}$$(8) are estimated using the GMM because of the endogeneity problem, which is used by Kleff and Weber (2008), Iwatsubo (2007), and (Hussain & Hassan, 2005). In theory, portfolio risk and capital are interrelated, i.e. in the capital equation, portfolio risk is endogenous. The study explores the endogeneity of the lagged dependent variable and other right-hand variables in the bank capital ratio. It uses 3SLS equations to estimate a dynamic model within GMM. The study also addresses the potential endogeneity problems of bank capital, highlighting the existence of serial correlation of risk. To address these issues, a GMM estimator, established by Arellano and Bond (1991), is used to estimate the dynamic panel model. This estimator considers both exogenous and endogenous changes in capital and risks, eliminating serial correlation, heteroscedasticity, and endogeneity. GMM is suitable for time series, panel, or cross-sectional data.

Then two Eqs. (7)(7) $$\Delta {\text{CAP}}_{\text{it}}={\alpha }_0+{\alpha }_1\text{REGi}{,}_{\mathrm{t}-1}+{\alpha }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\alpha }_3{\text{ROA}}_{\mathrm{i},\mathrm{t}}+{\alpha }_4\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}-{\alpha }_6\text{CAPi}{,}_{\mathrm{t}-1}+{\epsilon }_{\mathrm{t}}$$(7) and (8)(8) $$\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1\text{REGi}{,}_{\mathrm{t}-1}+{\beta }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\text{LLP}}_{\mathrm{i},\mathrm{t}}+{\beta }_4\Delta {\text{CAP}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}–{\beta }_6{\text{RISK}}_{\mathrm{i},\mathrm{t}-1}+{\text{v}}_{\mathrm{t}}$$(8) are estimated using 3SLS followed by Jacques and Nigro (1997) and Rime (2001). This allows for considering the simultaneity of the capital and risk regulations made by banks and for obtaining asymptotically more efficient estimates. 3SLS also provide consistent estimation of the parameters. 3SLS incorporate the cross-equation correlations and produce estimates of parameters that are asymptotically more consistent and efficient than 2SLS. Finally, the article makes a significant contribution by providing further empirical information on bank capital behaviour in Jordan.

4.3. Descriptive statistics

Descriptive statistics of all variables are presented in Table 2, which indicates that the average CAP is −0.001 with the standard deviation (SD) of 0.028 for all banks. The maximum and minimum values of CAP are 0.15 and −0.076, respectively. The average risk is 0.005 with a SD of 0.063. The minimum and the maximum values of risk are -0.190 and 0.17, respectively. The average REG is 0.036 with a SD of 0.18. The maximum value of REG is 1 and minimum is 0. Furthermore, the average size is 21.42 for all banks with a SD of 1.004. The maximum value of size is 23.97 and minimum is 20.05. We have noticed average liquidity is 63.30 with a SD of 14.04. The maximum and minimum liquidity are 101.94 and 2.512, respectively. Average ROA is 1.39 with a SD of 0.538. The maximum and minimum ROA are 2.51 and −0.176, respectively. On the other hand, the average loss provision is 15.67 with a standard deviation of 1.53. The maximum and minimum losses are 19.54 and 12.59 respectively. The average CAP _t−1 is 0.20 with a SD of 0.04. The maximum is 0.32 and the minimum is 0.135. The average RISK _t−1 is 0.68 with a SD of 0.10. The maximum is 0.96 and minimum is 0.44. The average REG_t−1 is 0.036 with a SD of 0.18. The maximum is 1 and the minimum is 0 when all banks are considered. The mean of LNSMC is 19.64 with a SD of 1.20. The maximum value of LNSMC is 23.07 and minimum is 18.16.

Table 2. Descriptive statistics of the variables.

Variable	Mean	Std. Dev	Min	Max
CAP	−.0010583	0.028816	−0.07645	0.154744
RISK	0.0052161	0.063199	−0.19084	0.1781023
REG	0.0363636	0.18805	0	1
SIZE	21.424	1.004749	20.05615	23.97595
LIQUIDITY	63.30602	14.04095	2.512486	101.9485
ROA	1.396273	.5384293	−0.17	2.51
LOSS	15.67186	1.538872	12.59134	19.54682
LAGCAP	0.2051486	0.045078	0.135579	0.3296973
LAGRISK	0.6879911	0.102412	0.44052	0.9626323
LAGREG	0.0366972	0.188886	0	1
LNSMC	19.64018	1.197167	18.16323	23.06939

Note. Max = Maximum, Min = Minimum, S.D. = Standard Deviation.

4.4. Correlation analysis

The coefficients of correlations among variables are shown in Table 3. The key interest of the analysis is to look at the association between bank risk and bank capital. Table 3 shows that capital is linked negatively to risk, size, loss, and CAP_t−1. The capital is positively related to the bank's REG_t−1, liquidity, ROA, and risk_t−1. This means that better-performing banks hold more capital. A negative relation between cap, size, and risk_t−1 with risk while positive relation with liquidity, ROA, loss, REG_t−1 and CAP_t−1. Table 3 also states that REG_t−1 has a negative relation to size, ROA and CAP_t−1, while it has a positive relation with cap, liquidity, loss, risk, and risk_t−1. Size has a negative relation with cap, REG_t−1, liquidity, risk, and risk_t−1 and a positive relation with ROA, loss, and CAP_t−1. Considering liquidity has positively relation to cap, REG_t−1, ROA, CAP_t−1, risk, and risk_t–1 and negatively related to loss. Loss has a positive relationship with REG_t−1, CAP_t−1, and risk, and negative relationship with cap, liquidity, ROA, and risk_t−1. And finally, CAP_t−1 has a negative relationship with risk_t−1. Stock market capitalization is negatively correlated with capital and positively correlated with risk of Jordan banking sector.

Table 3. Correlation among the variables.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
(1) CAP	1.000
(2) RISK	−0.591	1.000
(3) SIZE	−0.101	−0.014	1.000
(4) LIQUIDITY	0.013	0.234	−0.231	1.000
(5) ROA	0.147	0.137	−0.026	0.182	1.000
(6) LOSS	−0.129	0.057	0.699	−0.125	−0.346	1.000
(7) LAGCAP	−0.364	0.267	0.302	0.060	0.151	0.199	1.000
(8) LAGRISK	0.129	−0.312	−0.321	0.319	−0.256	−0.104	−0.453	1.000
(9) LAGREG	0.056	0.077	−0.005	0.026	−0.189	0.092	−0.255	0.018	1.000
(10) LNSMC	−0.062	0.023	0.937	−0.192	0.142	0.591	0.420	−0.354	−0.116	1.000

5. Empirical results

As mentioned earlier, the relationship between capital adequacy and risk is determined by using simultaneous equation models. This section presents the results of both models, the results obtained using the GMM estimation technique (Table 4) while the results of 3SLS are presented in Table 5 and for robustness checking by FE and RE, results are presented in Table 6.

Table 4. Empirical results for bank capital and risk for the period 2010–2019.

	(CAP)	(RISK)
Variables	GMM	GMM
CAPt−1	−0.0669
	(0.0583)
Risk	−0.2557***
	(0.0227)
Size	−0.0150***	−0.0488***
	(0.0046)	(0.0115)
Liquidity	0.0003***	0.0015***
	(0.0001)	(0.0003)
ROA	0.0132***
	(0.0026)
LAGCAP	−0.2091***
	(0.0342)
LAGREG	0.0176***	0.0545***
	(0.0063)	(0.0168)
LNSMC	0.0133***	0.0341***
	(0.0040)	(0.0091)
Riskt−1		−0.1777*
		(0.0949)
CAP		−1.6028***
		(0.1388)
Loss		0.0018
		(0.0035)
LAGRISK		−0.2063***
		(0.0446)
Constant	0.0661*	0.3986***
	(0.0352)	(0.1129)
Observations	88	88
Number of id	11	11
Wald chi2	321.38***	331.78***
Sargan Test (Prob)	20.95 (0.584)	17.21 (0.19)
AR(2)	−1.85 (0.065)	−0.82 (0.41)

*** p < .01, **p < .05, *p < .1.

Table 5. The empirical results of banks capital and risk (2010–2019).

Equation	Obs	Parms	RMSE	R^2	chi2	P
cap	99	7	0. 0209	0.4681	165.71	.000
risk	99	7	0.0439	0.5108	184.29	.000
			(CAP)		(RISK)
VARIABLES			3SLS		3SLS
RISK			−0.3759***
			(0.0319)
SIZE			−0.0189***		−.0405***
			(0.0067)		(.0145)
LIQUIDITY			0.0004**		.0014***
			(0.0002)		(.0004)
ROA			0.0072*
			(0.0039)
LAGCAP			−0.0953*
			(0.0496)
LAGREG			0.0252**		.0543**
			(0.0109)		(.0226)
LNSMC			0.0163***		.0292***
			(0.0059)		(.0112)
CAP					−1.7044***
					(.1399)
LOSS					.0020
					(.0036)
LAGRISK					−.1372***
					(.0427)
Constant			0.0722		.2715**
			(0.0562)		(.1297)
Observations			99		99
R-squared			0.4681		.5108
Chi2 (Prob > chi2)			42.11 (0.0000)

*** p < .01.

** p < .05.

* p < .1.

Table 6. Empirical results of banks capital and risk.

	(CAP)	(RISK)	(RISK)	(RISK)
VARIABLES	FE	RE	FE	RE
RISK	−0.2461***	−0.2725***
	(0.0352)	(0.0358)
SIZE	−0.0393***	−0.0186***	−0.0299	−0.0398**
	(0.0108)	(0.0071)	(0.0228)	(0.0158)
LIQUIDITY	0.0001	0.0003	0.0014***	0.0015***
	(0.0002)	(0.0002)	(0.0004)	(0.0004)
LAGREG	0.0235**	0.0187	0.0468*	0.0451*
	(0.0115)	(0.0115)	(0.0236)	(0.0234)
ROA	0.0087	0.0096**
	(0.0056)	(0.0044)
LAGCAP	−0.3891***	−0.1964***
	(0.0764)	(0.0571)
LNSMC	0.0301***	0.0169***	0.0249	0.0226*
	(0.0099)	(0.0063)	(0.0197)	(0.0122)
CAP			−1.1945***	−1.2490***
			(0.1498)	(0.1530)
LOSS			0.0070*	0.0064
			(0.0042)	(0.0042)
LAGRISK			−0.3916***	−0.2555***
			(0.0624)	(0.0509)
Constant	0.3157	0.0748	0.2241	0.3889***
	(0.2963)	(0.0588)	(0.5846)	(0.1475)
R-squared	0.6273		0.6747
Wald-Chi2/F-Stat	19.48***	99.21***	24.00***	128.96***
Hausman value	26.12***		18.95***

*** p < .01.

** p < .05.

* p < .1.

5.1. Results using the GMM Estimation method

Table 4 shows that size, risk, and lagged value of capital have significant and negative association with Jordanian commercial bank capital, while ROA has a positive association with Jordanian commercial bank. RISK model results show that size, CAP, Risk_t−1 affecting Jordanian commercial banks negatively while liquidity, stock market capitalization and regulation positively affect the Jordanian commercial banks risk and capital. The main findings show that the association between commercial bank risk and capital in Jordan is highly negative, which reduces the risk and uncertainty in Jordan’s commercial banks by implementing the new capital requirements regulations. Jordanian commercial banks consider as a well-capitalized bank stand against any future shocks which explain the negative relationship. Value of Wald-chi2 is significant for both models which reveal that both models are good fit and parameters are significantly differ from their values. Finally, chi-square value in Sargan test is not significant (p > 0.1) indicates that overidentifying restrictions are valid. Finally, auto-regressive (AR) is also insignificant for both models which indicates the absence of autocorrelation.

5.2. Results using 3SLS

The 3SLS findings for capital and risk in Jordanian commercial banks are presented in this section. Bank capital and risk-weighted assets are treated as endogenous variables in the simultaneous model. Probability value of Durbin-Wu-Hausman test on the basis of residual are significant at 1% level (chi2 = 37.35, p < .001), which reject the null-hypothesis. From these values, it is concluded that endogeneity is present among the variables. Moreover, RMSE value is also very low and near to zero which indicates that model is perfect to apply 3SLS (Table 5).

The results of 3SLS presented in Table 5. The findings of the system based on the capital-to-risk-weighted-assets (RCWA) ratio are presented in this section using the 3SLS. The first model reveals the risk impact on banking capital. The empirical results show that in the equation of capital, the bank size has a negative and significant effect on capital (β = −0.089; p < .1). These results are in line with Shrieves and Dahl (1992), Godlewski and Skully et al. (2009). This demonstrates that Jordanian major banks have greater access to financial markets, allowing them to operate with less capital. Furthermore, we can explain that the negative coefficient of smaller banks has a lower diversity compared to their larger banks.

Liquidity has a significant and positive impact on banks’ capital (β = 0.0004; p < .05). The positive relation between bank capital and liquidity was also found by Distinguin et al. (2013). The results show that Jordan’s business banks hold a high liquidity level, as established by CBJ, 100%, which positively affects Jordan 's bank capital. The ROA is also significantly and positively related to capital (β = 0.0072; p < .1). These results are compatible with Rime (2001) and Altunbas et al. (2007), which indicate that profitable commercial banks of Jordan can improve capitalisation more easily by retaining profits instead of issuing new assets in the form of security. Jordanian banks high profit margins could be attributed to the country’s underdeveloped financial sector or the banks ownership character.

Similarly, impact of the regulatory pressure has a positive impact on capital, but this relationship is less robust (β = 0.025; p < .05). This means that Jordan’s commercial banks are close to meeting minimal legal standards and are working to strengthen their capital adequacy in order to avoid regulatory penalties and provide a shield against equity shocks. The capital adequacy ratio of the Jordanian commercial banks exceeds CBJ’s minimum (12%). These results are aligned with previous studies of Aggarwal and Jacques’ (1998), Ediz et al. (1998), Zhang et al. (2008) and Akinsoyinu (2015).

The risk is negatively associated with the capital of Jordanian commercial banks at 5% level of significance (β = 0.3759; p < .01). This indicates that, in response to CBJ capital requirements, Jordanian trade banks increased their share capital by decreasing their risk levels. The negative relationship between RCWA and risk is also found by Shrieves and Dahl (1992) and Aggarwal and Jacques (1998). The lagged value of capital (CAP_t−1) is negatively related to capital indicating that increasing capital within the last year would decrease the change in capital in the current year (β = −0.095; p < .1). Results are consistent with the study of with Alkadamani (2015). Finally, stock market capitalization is positively related to bank capital (β = 0.016; p < .01). Raise in stock market capitalization increases the return on investment (Qurashi & Zahoor, 2016), which attract the investor to invest in banking sector. Another reason of this positive relationship is that stock market leads to higher stock market prices and market capitalization which results in improving financial growth and leads to the higher capital adequacy ratio.

Bank risk model

The results of the risk Eq. (8)(8) $$\Delta {\text{RISK}}_{\mathrm{i},\mathrm{t}}={\beta }_0+{\beta }_1\text{REGi}{,}_{\mathrm{t}-1}+{\beta }_2{\text{SIZE}}_{\mathrm{i},\mathrm{t}}+{\beta }_3{\text{LLP}}_{\mathrm{i},\mathrm{t}}+{\beta }_4\Delta {\text{CAP}}_{\mathrm{i},\mathrm{t}}+{\alpha }_5{\text{LTD}}_{\mathrm{i},\mathrm{t}}–{\beta }_6{\text{RISK}}_{\mathrm{i},\mathrm{t}-1}+{\text{v}}_{\mathrm{t}}$$(8) are presented in this section. The size of the bank has a negative and significant effect (β = −0.04; p < .01) on the level of bank risk, suggesting that larger banks are secure than smaller ones (Aggarwal & Jacques, 1998; Alkadamani, 2015). The high level of capital held by Jordanian banks shows that the exposure to credit risk can be reduced in greater diversification.

In contrast to Altunbas et al. (2007) and Repullo, who notices a significantly negative relationship of liquidity with risk levels, our study shows a highly positive impact of liquidity on Jordan’s commercial Banks risk (β = 0.0014; p < .01). Results show that holding high liquidity means investing in a highly risky asset investment to achieve high profit, so that the capital ratio remains at least as low as CBJ (12%). The loan loss provision (LLP) is positively but not significantly associated with the level of risk (β = 0.0020; p > .1) in line with Ugwuanyi (2015), who found a positive impact of LLP on risk. Mahdi and Abbes (2018) found that the provision for a loan loss does not matter to conventional loans. The results show that increased investment in risky assets increases the risk level.

The regulatory pressure (REG) has a positive and substantial impact on risk level (β = 0.054; p < .05), i.e., higher capital requirements cause banks to boost their capital ratios by taking on more risk when investing in their assets, which in turn raises risk for banks in Jordan. The behavior of Jordanian banks, which will provide greater projected returns, will raise their revenues, which will grow their capital. This might be the cause of the increase in risk. These results are inconsistent with Rime (2001), who depicted that the regulation on capital requirements has no effect on risk rates. In Jordan’s banks, the lag in risk level has a negative and significant impact on risk change. (β = −0.137; p < .01), as opposed to Alkadamani (2015), who estimated that last year’s increased risk would increase the risk shift in the current year.

The fact that capital had a very significant negative influence on risk in Jordanian commercial banks’ (β = 1.70; p < .01). It shows that these institutions increased their capital by lowering their risk exposure in response to CBJ capital rules. Shrieves and Dahl (1992) and Aggarwal and Jacques (1998) also stated the inverse relationship between risk-weighted assets (RCWA) and capital ratio. This negative relationship indicates that banks are rising their capital rates by raising their risk in response to capital requirements. Finally, stock market capitalization of positively related to risk of Jordanian banks (β = 0.029; p < .01). These results are consistent with the study of Vithessonthi (2014) who analysed the relationship between stock market and financial development on bank risk in South-East Asian countries.

5.3. Robustness checking by Pooled OLS, FE, RE Estimation method

Robustness analysis by using FE and RE are presented in Table 6. While robustness checking with addition control variables of economy (Inflation and RGDP) are also added in the model of risk and capital following the previous literature (e.g., Abbas & Ali, 2020; Isnurhadi et al., 2021, etc.). Inflation and economic growth (RGDP) are not significantly related to capital and risk in our model of analysis (Table 7). Moreover, to check the robustness of results, bank risk is also measured by Zscore [(ROA + ROE)/σROA] and results are presented in Table 7 by using GMM and in Table 8 by using 3SLS. Capital and risk (z-score) are negatively related to each other and significant only through 3SLS estimation technique.

Table 7. Empirical results of banks capital and risk (ZSCORE) with control variables estimation technique: GMM.

	(1)	(2)	(3)	(4)
VARIABLES	CAP	Z-SCORE	CAP	Z-SCORE
L.CAP	−0.2456*		−0.2483*
	(0.1293)		(0.1271)
ZSCORE	−0.0003		−0.0003
	(0.0002)		(0.0002)
SIZE	−0.0160**	−5.0431	−0.0154**	1.2164
	(0.0074)	(5.8777)	(0.0076)	(3.8627)
LIQUIDITY	0.0001	−0.0248	0.0001	−0.0753
	(0.0002)	(0.1159)	(0.0002)	(0.0931)
ROA	0.0099**		0.0093**
	(0.0042)		(0.0042)
LAGCAP	−0.3638***		−0.3618***
	(0.0673)		(0.0666)
LAGREG	0.0067	13.7747**	0.0075	11.3326**
	(0.0101)	(6.5505)	(0.0101)	(5.0715)
LNSMC	0.0150**	7.2787*	0.0145**	2.4567
	(0.0063)	(4.3716)	(0.0063)	(2.9121)
L.ZSCORE		0.0718		−0.3773
		(0.3632)		(0.2523)
CAP		3.5280		7.4193
		(49.6877)		(37.6140)
LOSS		−3.5295**		−3.5535***
		(1.4413)		(1.1322)
LAGZSCORE		−0.3185***		−0.1880***
		(0.1052)		(0.0714)
INFLATION			0.0008	0.3392
			(0.0008)	(0.4080)
RGDP			−0.0009	0.5187
			(0.0042)	(2.3918)
Constant	0.1043*	34.9273	0.1000	−8.6779
	(0.0560)	(44.0275)	(0.0612)	(30.1067)
Observations	88	88	88	88
Number of id	11	11	11	11
Wald chi2	111.17	36.98	113.02	62.61
Sargan Test (Prob)	8.33 (0.82)	6.60 (0.92)	8.02 (0.84)	21.07 (0.07)
AR(2)	−0.74	0.54	−0.69	−0.54
Prob > z (AR)	0.43	0.59	0.49	0.59

Table 8. Empirical results of banks capital and risk (ZSCORE) estimation method: Three stage least square (3SLS).

Equation	Obs	Parms	RMSE	R^2	chi2	P
CAP	99	7	0.025	0.240	33.890	.000
ZSCORE	99	7	10.830	0.201	27.430	.000
			(1)		(2)
VARIABLES			CAP		ZSCORE
ZSCORE			−0.0006**
			(0.0002)
SIZE			−0.0167*		5.0717
			(0.0087)		(3.8553)
LIQUIDITY			−0.0000		−.0645
			(0.0002)		(.0932)
ROA			0.0086
			(0.0055)
LAGCAP			−0.2968***
			(0.0662)
LAGREG			0.0100		7.4521
			(0.0138)		(5.8962)
LNSMC			0.0159**		−1.4037
			(0.0076)		(2.8924)
CAP					−72.0277*
					(39.1292)
LOSS					−3.6010***
					(1.0878)
LAGZSCORE					−.1747***
					(0.0480)
Constant			0.0945		−14.5611
			(0.0708)		(32.1700)
Observations			99		99
R-squared			0.2399		.2006
Chi2 (Prob > chi2)			46.85 (0.0000)

*** p < .01.

** p < .05.

* p < .1.

Table 1. Variables and measurement.

Variables	Notation	Measurement	Type	Source
Dependent Variables
Chang in Capital	Δ CAP	Ratio of capital to total assets (RCTA) and ratio of capital to risk weighted assets (RCWA) observed changes for bank j throughout time period t. The same ratios were employed in earlier research (see, for instance, Aggarwal and Jacques, 2001; Ediz et al., 1998; and Rime, 2001).	Bank-specific	ASE
Change in Risk	Δ RISK	Risk levels were tracked for bank j throughout the course of time, and the risk-weighted total assets to total assets ratio was calculated. The proxy has been used in prior investigations, such as those by Jacques and Nigro (1997), Aggarwal and Jacques (1998), and Rime (2001).	Bank-specific	ASE
Independent variables
Bank size	SIZE	Size (SIZE) is measured by natural log of the total assets i.e. Ln of Total Assets. In the literature the Size variable examine by many studies such as Jacques and Nigro (1997); Aggarwal and Jacques (1998); and Rime (2001); Gropp and Heider (2010); Das and Ghosh (2004); Lindquist (2004); Van Roy (2005); and Floquet and Biekpe (2008).	Bank-specific	ASE
Return on Assets	ROA	Return on Assets (ROA) is measure through Net income/Total assets. From the literature proxy has been used such as (Aggarwal and Jacques 1998; and Rime, 2001; Tanda, 2015; and Mahdi and Abbes, 2018)	Bank-specific	ASE
Liquidity	LTD	Liquidity is measure by Total Loans/ Total Deposits. To estimate bank liquidity, we use the ratio of total loans to total deposits. Both capital and risk equations contain the ratio of total loans to total deposits (LTD).	Bank-specific	ASE
Loan loss provision	LLP	It is introduced as an explanatory variable in the risk equation by Rime (2001); Gropp and Heider (2010, pp. 587-622); Cannata and Quagliariello (2006)	Bank-specific	ASE
Regulatory Pressure	REG	If a bank’s capital adequacy ratio is below the regulatory capital minimum of 12%, then REG = 1. If a bank’s capital adequacy ratio is equal to or greater than the legal minimum of 12%, then REG = 0. We can take the examples from literature such as (Awdeh, El-Moussawi, and Machrouh, 2011; and Zhang, Jun, and Liu, 2008, pp. 183-189)	Bank-specific	ASE
Stock market capitalization (LNSMC)		It is measured by multiplying stock market price and outstanding shares of the bank. Natural log (ln) of SMC is used in this study.	Bank-specific	ASE

Source: ASE, CBJ, Data-Stream and Bloomberg.

6. Conclusions

The present research is conducted to address the association between capital adequacy and risk for Jordan’s commercial banks. Intensively and uniquely, we investigate this relationship with FE, RE, GMM and 3SLS by using new techniques through simultaneous equation models. We contribute to empirical literature through the use of Jordanian banking data over the period 2010–2019, during which a more competitive environment is developed, and bank efficiency is increased.

Empirical evidence suggests that capital regulation requirement has a positive and significant impact on the capital and level of risk. Our study also concludes that Jordanian banks have more than the minimum regulatory capital requirements set by Basel II, III and the CBJ and boost their capital adequacy by growing their resources and reducing their risk-taking. The findings also show a highly significant and negative relationship between capital and risk for Jordanian commercial banks. Moreover, the study also finds the positive and significant relationship of liquidity, ROA and stock market capitalization with banks capital in Jordan and negative influence of bank size on capital and risk.

The results demonstrate that, given the significant contribution of the banking system to economic growth and development, bank managers must select an appropriate regulatory environment in order to increase the competitiveness of the banking sector and give greater consideration to important factors that will enhance capital requirements and lower risks in the Jordanian banking sector. In conclusion, research on bank capital and risk has broad implications for financial stability, economic growth, regulatory policy, and overall financial system stability. Promoting bank stability and preserving the flow of credit and economic activity must coexist in harmony, according to researchers and policymakers.

In addition to the gains from extending previous work on the risk-capital relationship, we firmly believe that our empirical results may be useful for making relevant policies for the Jordanian government and the banking regulatory authority. Jordanian banks in particular should be allowed to conduct more risk-based lending and help to increase competition in the banking sector (decrease concentration). Lastly, Jordan’s banking sector should be further established, and better control of the stock market. We are looking to adopt another research using different measures of risk.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data used in this research will be available on request. Readers can request for data on following email after publication of paper.

Additional information

Funding

The authors would like to thank Prince Sultan University, Riyadh, Saudi Arabia for their support.

Notes on contributors

Nusiebeh Nahar Falah Alrwashdeh

Dr. Nusiebeh Nahar Falah Alrwashdeh is currently working as a senior Lecturer at Cardiff Metropolitan University. She has done PhD in Economics and Finance from University of Portsmouth. Her interests are. Financial Sector, Banking Industry performance, Insurance Companies performance. Green Financing. She is also holding professional qualification in finance from Chartered Institute of Securities and Investment (CISI-UK) and currently Associate Fellow Member of CISI (UK).

Umara Noreen

Dr. Umara Noreen is working as an Assistant Professor at Finance Department, Prince Sultan University Riyadh, Saudi Arabia. She completed her PhD from Foundation University Islamabad, Pakistan in 2010. Her area of specialization is finance. She has an extensive teaching experience of 24 years at undergraduate, graduate and post graduate levels. She has 26 refereed and indexed Journal publications, 15 conference proceedings and a published book to her credit. She is a professional ISO certified trainer for the USAID program, with Asia Foundation, Islamabad Pakistan. She has also developed training manual for non-profit organizations under The Asia Foundation USAID project at CIIT. She has delivered five rounds of trainings to executives and senior level management of under this program. She has delivered a training program to the executives of Riyadh Bank, Saudi Arabia and training under Elite Monsha’t PSU program to entrepreneurs held at Prince Sultan University, Saudi Arabia.

Muhammad Hassan Danish

Dr. Muhammad Hassan Danish is an Assistant Professor of Economics at University of Management and Technology, Lahore-Pakistan. His research expertise’s comprises in the diversified field of social sciences with the mainly focus on primary data research. His work includes economics of happiness, public policy analysis and from the field of development on households and industrial economics. His expertise is micro and survey data analysis, public and social policy, econometric modelling, and microeconomic theory. Moreover, Dr Hassan streamlined the area of research with sustainable development goals of UNDP. He explores the current and more relevant issues in the research area to provide a way forward in achieving SDGs. He has excellent command of econometric software’s including Stata, SPSS, Eviews and Smart PLS for data analysis especially the primary or survey data.

Rizwan Ahmed

Dr Rizwan Ahmed is a Lecturer in Finance at University of Birmingham. Before joining University of Birmingham, he gained wide-ranging teaching and research experience as a lecturer and seminar leader in Cardiff Business School, Birkbeck, University of London, University of Huddersfield, and the University of Hull. His research interests are financial economics, Banking Industry performance, Insurance Companies performance. His expertise includes panel data analysis, and survey data analysis of corporate and banking sector.

Notes

1 https://www.cbj.gov.jo/EchoBusV3.0/SystemAssets/1a927c6c-b2a8-42db-b6a3-b6640d4d150c.pdf.