Financial Risk Indicators on the Performance and Stability of Banks: Evidence from Jordanian Banks (2018–2024)

Abstract

This study investigates the key determinants of bank stability and profitability in commercial and Islamic banks listed on the Amman Stock Exchange (ASE) in Jordan, with a focus on credit risk and capital adequacy during the period 2018–2024. Using panel data from 15 banks, the study applies fixed effects regression models with clustered standard errors. Liquidity is proxied by the loan-to-deposit ratio (LDR), credit risk by the loans loss provisions-to-total loans ratio, and capital strength by the equity-to-assets ratio, alongside a COVID-19 dummy and an interaction term between liquidity and credit risk. Financial performance and stability are measured using return on assets (ROA), return on equity (ROE), and the logarithmic Z-score. The findings indicate that credit risk has a significant negative effect on both bank performance and financial stability, whereas capital adequacy exerts a positive and significant effect. The COVID-19 pandemic negatively affected financial performance and stability, while liquidity (LDR) shows no significant direct effect. The interaction between liquidity and credit risk was statistically insignificant across all estimated models, suggesting that credit risk remains the dominant determinant regardless of liquidity conditions. The study highlights the importance of effective credit risk management and strong capital buffers in enhancing bank resilience. It contributes to the literature by providing recent evidence from the Jordanian banking sector and by incorporating multiple performance measures, a pandemic shock variable, and risk interaction effects to better understand bank stability within a unified empirical framework for an emerging banking market.

1. Introduction

Bank stability is a very essential requirement for the existence of a strong and healthy economy, as banking institutions are the most important economic sector in any country and financial system and it plays the role of intermediary for economic activity in the market (Altarawneh & Shafie, 2018). The lack of bank stability or the occurrence of any crises in the banking sector can adversely affect the other economic sectors in the country, especially in emerging markets such as the case of Palestine (Altarawneh & Shafie, 2018).

In the Jordanian context, the banking institution plays an essential role in financing retailers and businesses. Thus, the existence of good soundness and a high performing banking sector has a constructive and positive effect on economic prosperity, economic growth and sustainability. Recently, the banking sector in Jordan has experienced constant challenges and unexpected turmoil due to COVID-19 in 2020 that tested considerably the risk management of banking institutions and their stability. This study concentrates on the period 2018–2024. Likewise, it investigates the way that significant risk factors affected the financial performance of banking institutions and their stability throughout an unstable period. The Jordanian banking sector represents an important case for analysis because it operates within an emerging market environment characterized by regional economic uncertainty and a bank-centered financial system.

Banking institutions suffer several types of risks, especially credit risk and liquidity risk as they are the most severe risks that affect bank financial performance and bank liquidity. According to Peykani et al. (2025) credit risk is the most likely cause of losses because of debtors defaulting on their obligations, leading to NPL and LLP. Liquidity risk includes the risk that a bank cannot meet its short-term financial commitments, commonly measured using indicators such as loan-to-deposit ratios. The inefficient and poor credit risk management can severely decrease financial performance and bank stability. The experiences of several banks around the globe have shown that severe exposure to credit risk and liquidity risk can lead to permanent economic distortion. Thus, supervisory and legislative organizations have used frameworks, for instance the Basel III accords, to increase the effectiveness and efficiency of credit risk and liquidity risk management to ensure the existence of sufficient capital shields and liquidity coverage to maintain financial stability. On the other hand, the extent to which these risks affect bank financial performance and bank stability can differ by economic context and over time, affirming the necessity of further studies to explore this topic in financial markets such as Jordan.

The COVID-19 pandemic represented a major external shock to banking systems worldwide. In Jordan, the pandemic reduced economic activity, increased expected credit losses, and negatively affected bank profitability. These developments highlighted the importance of risk management practices and raised important questions regarding the resilience of banks during periods of economic uncertainty.

Thus, this study’s purpose is to analyze the effect of significant risk indicators on bank financial performance and bank stability of banking institutions in Jordan throughout 2018–2024, comprising the COVID-19 pandemic period. This study concentrates on five important factors: credit risk, liquidity risk, capital adequacy, a COVID-19 dummy variable, and the interaction between liquidity and credit risk. Examining the interaction between liquidity and credit risk is particularly important because deteriorating credit quality may intensify funding pressures during periods of economic uncertainty, thereby affecting both bank profitability and financial stability.

This study contributes to the existing literature by providing recent empirical evidence from the Jordanian banking sector. Despite the importance of the banking sector within the MENA region, limited evidence has jointly examined bank performance, financial stability, and the interaction between liquidity and credit risk in emerging banking markets. While extensive evidence exists for developed banking systems, relatively few studies have explored these relationships within the MENA banking market.

Financial stability, solvency, and liquidity represent related but distinct concepts. Liquidity refers to a bank’s ability to meet short-term obligations, whereas solvency reflects its ability to absorb losses and maintain adequate capital over the long term. Financial stability is a broader concept that captures the overall resilience of banks to financial distress. In this study, liquidity is proxied by the loan-to-deposit ratio, capital strength is measured by the equity-to-assets ratio, and financial stability is proxied by the logarithmic Z-score. Bank stability is measured using the logarithmic form of the Z-score (log Z-score) to reduce skewness and improve estimation properties.

2. Literature Review and Hypotheses Development

The risk management and its relationship to bank performance and bank stability is one of the highly researched topics in accounting and finance due to the importance of banking sector stability in economic growth and prosperity. The relationship between risk management, bank performance, and financial stability is commonly explained by the risk–return trade-off theory. While higher risk may generate greater returns, excessive risk-taking can increase losses and undermine profitability and stability. Therefore, effective risk management seeks to balance profitability objectives with financial resilience.

Empirical evidence generally supports this argument. Altarawneh and Shafie (2018) found that higher levels of non-performing loans were associated with lower bank profitability, Similar evidence from the Nigerian banking sector suggests that both internal and external banking factors significantly influence profitability (Sayedi, 2014). while Peykani et al. (2025) reported that greater credit risk reduced bank stability by increasing insolvency risk. When credit quality deteriorates, banks are required to recognize expected losses through higher loan loss provisions, which directly reduce profitability and may weaken capital positions (Khoffash & Awwad, 2025).

Therefore, both theoretical arguments and prior empirical evidence suggest that higher credit risk is expected to adversely affect bank performance and financial stability. Accordingly, the following hypothesis is proposed:

H1. 

Credit risk has a negative effect on bank performance and bank stability.

Liquidity is commonly measured by the loan-to-deposit ratio (LDR), which reflects the extent to which deposits are transformed into loans. While moderate lending activity may enhance profitability by increasing income-generating assets, excessively high LDR levels may reduce liquidity buffers and increase exposure to funding pressures. According to the risk–return trade-off theory, banks must balance profitability objectives with adequate liquidity management.

Empirical evidence regarding the effect of liquidity on bank performance and stability remains inconclusive. Altarawneh and Shafie (2018) found no significant relationship between liquidity and bank profitability in Jordanian banks. In contrast, Mehdi et al. (2025) reported that liquidity was a significant determinant of bank performance in MENA banking systems. Similarly, Peykani et al. (2025) found that stronger liquidity positions were associated with improved bank stability.

These mixed findings suggest that the effect of liquidity may vary across banking systems and economic environments, creating the need for further investigation in the Jordanian banking sector. Accordingly, the following hypothesis is proposed:

H2. 

The loan-to-deposit ratio significantly influences bank performance and bank stability.

Capital strength reflects a bank’s ability to absorb unexpected losses and maintain financial resilience. According to the capital buffer theory, well-capitalized banks are generally more stable and better positioned to withstand adverse economic conditions. Strong capital levels may also support profitability by reducing financial distress risk and enhancing stakeholder confidence.

Empirical evidence generally supports a positive relationship between capital strength and bank outcomes. Berger and Bouwman (2013) reported that higher capital levels enhanced bank performance, particularly during periods of financial stress. Similarly, previous studies have shown that stronger capitalization contributes to greater financial stability and lower insolvency risk (Laeven & Levine, 2009; Alkhazali et al., 2024). Evidence from Palestinian banks also indicates that capital adequacy is positively associated with banking profitability and financial performance (B. A. Awwad, 2021).

Although some studies suggest that excessively high capital ratios may reduce return on equity because equity financing is more costly than debt financing (Areghan et al., 2021), the overall empirical evidence indicates that the benefits of strong capitalization generally outweigh these potential costs.

Accordingly, the following hypothesis is proposed:

H3. 

Capital strength has a positive effect on bank performance and bank stability.

2.1. COVID-19 Pandemic and Bank Performance/Stability

The world has witnessed a severe natural experiment in 2020 with the rise in COVID-19, which severely hit the banking sector around the world. Thus, there was a generally negative effect of this pandemic on bank stability and bank financial performance as this sector witnesses a decrease in its activities. In addition, we take into consideration the effect of this pandemic on bank stability. A large shock could dilute the bank’s capital equity, decreasing its Z-score, particularly if losses were considerable. On the other hand, given policy support and short-term supervisory reductions (loan delays, use of capital conservation buffers, etc.), the effect on bank stability may be less marked or postponed in relation to the instant hit on financial performance. The banking sector in Jordan experienced a severe but short-term decline in financial performance in 2020. (European Investment Bank [EIB], 2022; B. S. Awwad & Razia, 2023).Thus, the following hypothesis was developed:

H4. 

There is a negative effect of the COVID-19 pandemic on bank financial performance and bank stability.

2.2. Interaction of Liquidity and Credit Risk

The interaction between liquidity and credit risk has received increasing attention in banking research because these risks may reinforce each other during periods of financial stress. Banks experiencing deteriorating credit quality may face higher funding pressures, while weak liquidity positions may limit their ability to absorb credit-related losses. Consequently, the combined effect of high credit risk and constrained liquidity may have a stronger adverse impact on bank performance and financial stability than either risk individually.

Empirical evidence on this interaction remains relatively limited, particularly in emerging banking markets. Previous studies suggest that liquidity shortages may amplify the negative consequences of credit deterioration and increase financial vulnerability (Tanwar, 2024). However, the magnitude and significance of this interaction remain inconclusive across different banking environments.

Given the limited evidence and the importance of understanding how multiple risks jointly affect banking outcomes, the following hypothesis is proposed:

H5. 

The interaction between liquidity and credit risk significantly influences bank performance and bank stability.

3. Methodology

3.1. Sample and Data Collection

The final sample consists of 15 banks listed on the Amman Stock Exchange (ASE) during the period 2018–2024, including 11 commercial banks and 4 Islamic banks. The commercial banks are Arab Bank, Jordan Ahli Bank, Bank of Jordan, Cairo Amman Bank, Jordan Kuwait Bank, Invest Bank, Arab Jordan Investment Bank, Housing Bank for Trade and Finance, Jordan Commercial Bank, Capital Bank of Jordan, and Société Générale Jordan. The Islamic banks are Jordan Islamic Bank, International Arab Islamic Bank, Safwa Islamic Bank, and Al Rajhi Bank Jordan.

Banks were included based on the availability of complete financial information required for the analysis. No observations were excluded due to missing data, resulting in a balanced panel dataset. The data were collected from publicly available annual reports and official disclosures issued by the sampled banks and the Amman Stock Exchange.

The explanation of choosing this period to explore both the years before the arise of the pandemic against its period and initial recovery. Thus, giving the researcher the opportunity to analyze both normal circumstances and the effect of the crisis.

3.2. Variables and Measurements

3.2.1. Dependent Variables: The Study Utilized Three Measures as Proxies of Financial Performance and Bank Stability

• Return on Assets (ROA): It is the net income divided by total assets. It reflects the extent of the efficiency of the firm in using available assets to create profit. It is an interesting measure of overall profitability from the perspective of all capital providers.
• Return on Equity (ROE): ROE is calculated as net income divided by total shareholders’ equity. It measures the returns generated from shareholders’ investments in the bank and reflects the efficiency with which management utilizes equity capital to generate profits. ROE is sensitive to capital structure, as greater leverage may increase shareholders’ returns during favorable economic conditions. Therefore, ROE provides a shareholder-oriented measure of financial performance.
• Z-score: The Z-score is a composite index usually manipulated to evaluate bank financial stability and insolvency risk (Beck et al., 2007). The Z-score = (Equity/Assets + ROA)/σ(ROA), where Equity/Assets is the capital-to-asset ratio, ROA is as defined above, and σ(ROA) is the standard deviation of ROA for each bank. Interestingly, the Z-score evaluates the number of standard deviations that a bank’s return would have to decline below the mean to reduce its equity, so a greater Z-score shows a lesser prospect of indebtedness (more stability) (Laeven & Levine, 2009). A greater Z-score means the bank is more stable. Due to the relatively short panel used in this study, the σ(ROA) is calculated over the available years for each bank; we recognize this as an estimate because of limited data, but it has to distinguish banks by risk level. The Z-score is utilized in logarithmic form in the regression (log Z-score) to decrease skewness. Although the Equity/Assets ratio constitutes one component of the Z-score calculation, the inclusion of capital measures as explanatory variables in Z-score models is common in the banking stability literature (Atari et al., 2026; Berger & Bouwman, 2013; Laeven & Levine, 2009). The variable is retained to examine the direct association between capital strength and bank stability.

3.2.2. Independent Variables

• Loan-to-Deposit Ratio (LDR): It is the ratio of total customer loans divided by total deposits of customers. It is a very important measure of liquidity risk. A greater LDR shows that the banking institution has lent out a greater percentage of customers’ deposits, keeping a minor stake of customers’ deposits un-lent (or another liquid assets) to cover withdrawals or unexpected financing needs. An LDR above 100% indicates that the loans of the bank are greater than the customers’ deposits, indicating dependence on other financing sources. We suppose LDR to capture the way that the liquidity management impacts bank financial performance and risk. The LDR is stated as a ratio.
• Credit risk is measured by the ratio of loan loss provisions to total loans (LLP/Total Loans), which reflects the extent of expected credit losses recognized by banks relative to their lending portfolio.
• Provision-to-Loans Ratio: Provision-to-loans ratio is calculated as the loan loss provisions divided by total loans. The provision to total loans ratio is used as a measure of credit risk. A greater ratio means that a bank anticipates or has incurred greater losses on its credit portfolio relative to its size, indicating lower quality assets. Under IFRS 9, banking institutions understand that provisions are based on anticipated credit losses. Thus, the provision-to-loans ratio captures both got and expected credit issues. The researcher uses the annual provisioning expense or changes in loan loss allowance compared to loans if available; when just the end-of-year allowance is disclosed, the researcher confirms steadiness in explanation. The greater the provision-to-loans ratio, the greater income is being set aside to cover expected loan losses that promptly decrease net income and indicate credit risk problems. It is expected to be a negative relationship between provision to total assets and ROA/ROE/Z-score as this ratio reflects the weakness of the credit portfolio.
• Equity-to-Assets Ratio: This ratio measures the strength of the capital. It is calculated by dividing the total shareholders’ equity over total assets of the bank. It is also called the leverage ratio. A greater ratio means the banking institution depends less on corporate leverage and depends greatly on shareholders’ equity. The equity-to-assets ratio aligns strictly with the Common Equity Tier 1 (CET1) capital ratio in spirit (though CET1 ratio basically uses risk-weighted assets in the denominator; our measure is un-risk-weighted for easiness and comparability across banks). We anticipate greater equity-to-assets to associate with greater bank stability and possibly with superior normal performance (still it could decrease ROE automatically).
• COVID-19 Dummy: In order to control for the impact of COVID-19, COVID-19 was included as a dummy variable in this study that assigns the value (1) for the years 2020, 2021 and (0) otherwise. Whereas the effects of the pandemic remained in 2021, several economic measures and bank performance metrics recovered in 2021. The COVIID-19 dummy captures any broad-based impact on the performance of banks common to that year, after controlling for other variables. It is expected to be a negative effect of the COVID-19 pandemic on bank financial performance and bank stability.
• Interaction (credit risk × liquidity risk): The researcher measured the interaction between liquidity and credit risk by multiplying the LDR and the provision-to-loans ratio. This ratio is important to examine if the effect of the credit risk on bank financial performance and stability is influenced by the liquidity position in the bank. The existence of a negative coefficient of the interaction of credit risk and liquidity risk on bank financial performance and liquidity indicates that the combination of high LDR and high provisions is more harmful than the sum of the effects separately.

Control Variables: In our fixed-effects model design, bank-specific fixed effects act to control for time-invariant differences across banks; for instance, variances in management quality, business models, or market niche that is several banks might constantly have greater margins because of emphasis on definite lending sectors, or greater cost structures. Bank fixed effects control for unobserved time-invariant bank-specific characteristics such as business models, ownership structure, management practice and any other stable bank characteristics such as ownership structure, etc. The study does not comprise explicit time fixed effects for each year since the time dimension is short and the study includes the COVID dummy to capture the one major systemic shift; including year dummies for every year could soak up much of the variation, and the pre-2020 years did not have clear distinct shudders that would require year dummies. The study does not include year fixed effects because the panel covers a relatively short period (2018–2024), and the COVID-19 dummy captures the major common shock affecting the banking sector during the study period. In addition, including a full set of year dummies may substantially reduce the available degrees of freedom given the limited sample size. Nevertheless, the exclusion of year fixed effects and additional macroeconomic controls should be considered a limitation of the study.

No further macroeconomic controls (such as GDP change or interest rates) are examined, somewhat because of the limited number of time periods (which makes it hard to estimate macro effects) and partly due to the COVID dummy indirectly captures the macro recession in 2020.

Overall, the regression specification is mainly motivated by the crucial bank-level risk variables and the pandemic shock dummy, with bank fixed effects to deal with unobserved heterogeneity.

3.3. Model Specification and Estimation

The study estimates panel regression models for each dependent variable (ROA, ROE, and Z-score) using the fixed-effects estimator. The general model can be specified as follows:

$$\begin{array}{cc}\hfill Y_{it}=\alpha +{\beta }_1\left({\mathrm{LDR}}_{it}\right)& +{\beta }_2\left({\mathrm{Provision}}_{it}\right)+{\beta }_3\left({\mathrm{Equity}/\mathrm{Assets}}_{it}\right)+{\beta }_4\left({\mathrm{COVID}}_t\right)\hfill \\ & +{\beta }_5\left(\mathrm{LDR}\times {\mathrm{Provision}}_{it}\right)+{\mu }_i+{\epsilon }_{it},\hfill \end{array}$$

where $Y_{it}$ is the dependent variable for bank $i$ in year $t$ (alternatively ROA, ROE, or log Z-score in separate regressions). $\alpha$ is a constant term. ${\mu }_i$ represents the bank-specific fixed effect (a dummy for each bank $i$ that captures all time-invariant characteristics of that bank). ${\epsilon }_{it}$ is the idiosyncratic error term.

The composite error term is specified as:

εit = μi + νit

where μi represents unobserved bank-specific fixed effects and νit is the idiosyncratic disturbance term. In the dynamic framework, fixed effects are removed through transformation.

All continuous independent variables are either in ratio form or percentage form and are used as such in the regressions. Although the Equity/Assets ratio is one component of the Z-score calculation, the Z-score is a broader composite measure that also incorporates profitability and earnings volatility. Therefore, it is not a linear function of the Equity/Assets ratio alone. Including Equity/Assets as an explanatory variable allows the analysis to examine the direct association between capital strength and bank stability beyond its mechanical role within the Z-score measure.

Estimation technique: The study used the within-bank fixed effects estimator. This statistical analysis approach is explained by the expected correlation between the fixed characteristics of the bank and the regressors. For example, the business model of the bank might affect both its distinctive LDR and its performance; thus, a random-effects model (that postulates no such correlation) would be unfitting. A Hausman test was used to compare fixed vs. random effects, and it favored the fixed-effects model (p < 0.05, assuming that the fixed effects estimates are considerably different—and so preferable—given expected endogeneity of regressors with bank effects). Also, given the study’s interest in controlling for unobserved heterogeneity, fixed effects is a natural choice. Given the relatively small sample size and short time dimension of the panel (15 banks over seven years), advanced dynamic estimators such as GMM were not employed because they may generate instrument proliferation and unstable estimates in small-sample banking panels (Roodman, 2009).

Before ending the estimation, the researcher examined for multi-collinearity among independent variables. The VIFs were all less than 5, signifying that multi-collinearity is not a serious issue.

4. Results

4.1. Descriptive Statistics

Table 1. Descriptive statistics of key variables (2018–2024).

Variable	Mean	Std. Dev.	Min	Max
ROA (Return on Assets)	0.0105 (1.05%)	0.0052	−0.002 (−0.2%)	0.0214 (2.14%)
ROE (Return on Equity)	0.095 (9.5%)	0.0445	0.000	0.185 (18.5%)
Z-score	4.73	1.20	2.3	7.1
LDR (Loans/Deposits)	0.75	0.10	0.55	0. 94
Provision Ratio (LLP/Loans)	0.0054	0.0066	0.000	0.03
Capital Ratio (Equity/Assets)	0.105	0.020	0.065	0.144

presents the descriptive statistics of the examined variables across all bank-year observations (105 observations: 15 banks × 7 years):

Commercial and Islamic banks listed on the Amman Stock Exchange (ASE) reported an average ROA of approximately 1.05% and an average ROE of 9.5% during the period 2018–2024. The average equity-to-assets ratio was approximately 10.5%, indicating a relatively well-capitalized banking sector. The average loan-to-deposit ratio (LDR) was 75%, ranging from 55% to 94%, reflecting variations in liquidity positions across banks. Loan loss provisions averaged 0.54% of total loans, although some banks reported provisioning levels of up to 3% during the peak of the COVID-19 pandemic. The mean Z-score was approximately 4.7, suggesting a moderate level of financial stability across the sampled banks. Notably, the COVID-19 period (2020–2021) was associated with an industry-wide decline in both ROA and ROE, accompanied by higher provisioning levels. This pattern is reflected in the minimum ROA value of −0.2% recorded in 2020. The negative minimum ROA reflects temporary losses experienced by some banks as a result of increased loan loss provisions and weaker economic activity during the pandemic.

The Jordanian banks have a mean of ROA approximately (1%) with a significant decline in 2020 as several banks had ROA about 0% or a little positive in that year) and recapture by 2021–2022. ROE had an average 8–10% over 2018–2024, again decreasing in 2020. The average equity-to-assets ratio was nearly 11–12%, showing a high-capitalized banking sector. Loan-to-deposit ratios differ significantly through banks, amounting from nearly 60% for the most liquid banks to 90% for greater loaned-up banking institutions. The provision-to-loans ratio had an average of approximately 2–3% (as a yearly expense to loans), with greater points in 2020 when several banking institutions took large provisions. The cross-sectional standard deviation of Z-scores assumed that several banking institutions were considerably more stable (Z-scores above 50 in some cases) whereas a small number of them had less bank stability (Z-scores in the 20s), subject basically to their capital ratios and earnings instability.

4.2. Correlation Matrix

Table 2. Correlation matrix of key variables.

	ROA	ROE	Z-Score	LDR	Provision	Cap. Ratio
ROA	1.000
ROE	0.872	1.000
Z-score	0.621	0.412	1.000
LDR	–0.105	0.021	–0.235	1.000
Provision (LLP/Loans)	–0.485	–0.536	–0.560	0.069	1.000
Capital Ratio (Equity/Assets)	0.542	–0.312	0.633	–0.198	–0.203	1.000

presents the correlations among the variables. The table shows that ROA and ROE are positively correlated with each other and with the capital ratio, whereas they are adversely correlated with the provision ratio (credit risk). Interestingly, the provision ratio is negatively correlated with Z-score and performance measures, confirming that greater credit risk tends to correspond with little financial performance and bank stability. LDR has an insignificant adverse correlation with Z-score and a very small negative correlation with ROA, but its correlation with ROE is less obvious. The capital ratio is positively correlated with Z-score (as anticipated, as both ROA and capital enter the Z-score numerator) and reasonably with ROA but demonstrates a negative correlation with ROE (as high equity decreases leverage, other things remain unchanged, decreasing ROE). These pairwise correlations already hint that greater credit risk decreases bank financial performance, and strong capital buffers correspond with higher stability.

4.3. Regression Analysis

None of the examined independent variables is so greatly correlated as to demonstrate multi-collinearity; that is, there is a low correlation between credit risk and liquidity risk, 0.07.

Table 3. Panel regression results (fixed effects).

Independent Variable	Dependent: ROA	Dependent: ROE	Dependent: Z-Score
Loan-to-Deposit (LDR)	−0.0012	+0.0154	−0.48 *
Provision/Loans (Credit Risk)	−0.0825 **	−1.013 **	−13.5 **
Capital/Assets (Cap. Ratio)	+0.0457 *	−0.392 *	+28.7 **
COVID-19 Dummy (2020–2021)	−0.0038 ***	–0.047 ***	−1.05 **
LDR × Provision (Interaction)	−0.0081	−0.115	−2.20
Bank Fixed Effects	Yes	Yes	Yes
Observations (N)	105	105	105
R-squared (within)	0.48	0.52	0.45

Coefficients represent impacts on ROA, ROE (in fractional terms), and Z-score. Significance: *** p < 0.01, ** p < 0.05, * p < 0.1. Bank dummies included but not reported. The fixed-effects estimator was selected based on the Hausman test results, which indicated that the fixed-effects specification was more appropriate than the random-effects model.

shows the panel regression estimates for the effect of liquidity risk, credit risk, and capital strength on bank performance and stability. Fixed effects for banks are included (absorbed) in these models. We report coefficient estimates with significance levels (*** p < 0.01, ** p < 0.05, * p < 0.1. respectively).

The bank’s financial performance and bank stability were examined during 2018–2024 using a fixed effects panel model. This study emphasized three essential risk factors: credit risk, liquidity risk, and capital strength measured by the equity-to-assets ratio, plus a pandemic dummy for the COVID-19 shock. Here is a thorough discussion of the way that each factor affects a bank’s financial performance and bank stability.

• Liquidity risk (loan-to-deposit ratio): There is an insignificant effect of the loan-to-deposit ratio on ROA and ROE. Coefficients on LDR were insignificant and not distinct from zero, indicating that moderate differences in liquidity status amongst Jordanian banks (most LDRs clustered around 70 –90%) did not translate into significant variances in bank financial performance. However, in the Z-score model, LDR exhibits a negative coefficient that is statistically significant at the 10% level, suggesting that higher LDR is associated with a modest decline in bank stability. Overall this assumes that banks commonly worked within contented liquidity ranges, conceivably because of regulatory liquidity requirements or judicious asset liability management. Therefore, we cannot support Hypothesis 2 in its strong form. On the other hand, the result does not inevitably suggest that liquidity is insignificant; it may be that the outcome is non linear, or that liquidity issues merely coincidence with other shocks that have extreme funding stress. Previous empirical studies on Jordan for 2010–2014 found an insignificant effect of LDR on ROA, which strengthens the idea that liquidity risk has been well managed in this market.
• Credit risk (provision/loans)—The provision-to-loans ratio is a good predictor of decreased financial performance and bank stability. Under the first model, that is, the ROA regressions, the provision/loans has a negative coefficient; this indicates that an increase in provisions by a small fraction promptly decreases ROA. Also, results show anticipated credit losses; thus, greater provisions captivate operating income and diminish profitability. Also, under the second model, which is ROE regression, the provision/loans has a negative coefficient (significance at the 5% level), confirming that credit issues decrease returns to shareholders as well. Interestingly, the Z-score regressions revealed a significant negative coefficient on the provision ratio, indicating that as credit risk upsurges, the solvency buffer declines and liquidity risk increases. Cautiously, this is spontaneous: declining asset quality enforces banking institutions to reserve income to shield non-performing loans, dipping retained earnings and shareholders’ equity. The adverse effect on Z-score demonstrates that credit risk is not merely some earnings issue but a stability issue; if provisions unexpectedly confound, liquidity can be endangered. Generally, these findings strongly support Hypothesis 1 and are in line with previous empirical studies that correlate greater non-performing loans or loan loss provisions with weaker bank performance.
• Capital strength (equity-to-assets ratio): The analysis shows a positive effect of capital strength on the ROA as well as a strong positive effect of equity-to-asset ratio on Z-score. Under the first model, the ROA model, a greater equity ratio considerably upsurges ROA, which supports the notion that good capitalized banks have less financing costs and more ability to create better financial performance. The result indicates that a 1% upsurge in the equity ratio increases ROA by many basis points, signifying that an increase in equity-to-assets ratio increases financial performance by providing a buffer against losses and by indicating financial trustworthiness to depositors and investors. Under the second model, which is the ROE model, there is a negative effect of the equity-to-assets ratio and ROE. This indicates the leverage trade off: more capital can decrease return on shareholders’ equity despite it increasing net income; the gains in ROA are offset by a greater equity base in the denominator. However, the Z-score model indicates that a greatly significant positive coefficient on equity as the Z-score formula integrates ROA with the equity ratio, a greater equity buffer promptly increases the liquidity metric and upsurges the number of standard deviations earnings would have to decrease to eliminate capital. This approves that capital adequacy is a very important factor influencing bank stability and emphasizes the significance of Basel III style buffers in developing markets.
• COVID-19 dummy: There is a negative effect of the COVID-19 pandemic on ROA and ROE, indicating that financial performance severely decreased throughout the crisis period. Calculations assume ROA of banking institutions decreased by approximately half a percentage point on average and the ROE decreased by many percentage points in 2020 in comparison with other years. These decreases indicate that the economic closure, fragile loan progress and a flow in provisions as banking institutions expected credit defaults. On the other hand, in the Z-score model the COVID dummy is statistically significant. This indicates that even though earnings decreased, the liquidity of banks did not decrease decidedly. Strong capital mitigates and regulatory restraint (for instance postponed loan reimbursements and capital buffer release) is expected to segregate banks from a liquidity crisis. So, Hypothesis 4 is supported: COVID-19 had a strong negative effect on financial performance as well as a negative effect on bank stability once other variables were controlled.
• Interaction between liquidity and credit risk (LDR × Provision):

There is an insignificant and negative effect of the interaction between liquidity risk and credit risk on financial performance in models (ROE, ROE and bank stability). All coefficients on the interaction term are negative but statistically insignificant, indicating that liquidity situations did not magnify or diminish the effect of credit risk on bank financial performance and bank stability. There are several explanations for this result. Firstly, banking institutions with greater credit risk are inclined to uphold conservative liquidity situations. Thus, severe integrations of great LDR and great provisions did not take place. Secondly, banking institutions had access to various other financing sources such as central bank facilities, interbank markets that prohibited liquidity deficiencies from impairing credit losses, and regulators may have inhibited risk taking such that high credit risk and liquidity stress could not correspond. Consequently, Hypothesis 5 is not supported, although the negative sign is directionally consistent with expectations.

Additional interesting observations: In general, the fixed effects models were statistically significant, with F tests showing p values less than 0.01 for the financial performance models and approximately 0.05 for the Z-score model. Adjusted R squared values were modest (about 0.48 for ROA, 0.52 for ROE, and 0.45 for log Z-score), representing that credit risk, capital and the COVID-19 shock clarify a considerable percentage of within-bank differences in financial performance and a considerable percentage of differences in bank stability. The less R squared for Z-score may reveal the short time series and the reality that bank stability relies on longer-term dynamics not captured here. Bank fixed effects were significant, indicating that unnoticed features (business models, risk desires, management quality) affect banks’ financial performance. On the other hand, the within-bank changes in provisions, capital and COVID-19 conditions drive most of the results discussed above.

In general, the results show that credit risk management and capital adequacy are the essential determinants of bank financial performance and bank solvency among banks listed in Jordan. Liquidity risk does not demonstrate a direct influence in normal times. The COVID-19 shock had a stark but short-lived effect on returns, whereas banks’ capital buffers prohibited a liquidity crisis.

5. Discussion

The results of the study provide many insights that are in line with the previous empirical studies and theoretical implications. Furthermore, these results reflect the particular context of the banking sector in Jordan during 2018–2024. In this section of the study, the researcher explains these results considering the previous empirical studies and the stated hypotheses in this study to show the effect of credit risk, liquidity risk, and the interaction between them on financial performance and bank stability.

Firstly, there is a negative effect of credit risk on bank financial performance and bank stability. This result is consistent with empirical results that found there is a significant effect of high-quality assets on financial performance and bank stability (Altarawneh & Shafie, 2018; Drehmann et al., 2010).

The results of this study confirmed that greater loan loss provisions significantly decrease ROA, and ROE indicates that when banking institutions experience a credit fall, their financial performance is hurt considerably. This is instinctive as provisions promptly decrease income, but beyond the arithmetic impact. Furthermore, it signifies essential matters in loan portfolios that can distract management’s consideration and coerce novel lending. In addition, the reality that credit risk decreased the Z-score confirms that it is not just an earnings matter; it indicates lower financial stability and greater vulnerability to financial distress, as losses decrease capital and leave banks more susceptible. Thus, credit risk management is very important for the good soundness of banking institutions. It is in line with indications from several markets; for instance, research in developing markets such as Kenya (Aduda & Gitonga, 2011) and Nigeria (Abiola & Olausi, 2014) has correspondingly revealed that weak credit risk (high NPLs) is linked with poor bank performance (Altarawneh & Shafie, 2018; Al-Sharkas & Al-Sharkas, 2022). This study contributes to the existing literature review as it confirms that there is a negative relationship between credit risk and bank financial performance and bank stability in commercial and Islamic banks listed on the Amman Stock Exchange (ASE) in Jordan during 2018–2024, which included the COVID-19 shock, indicating that, despite being among unexpected conditions, the basic rule remained consistent: banking institutions with higher credit losses managed poorer.

The results also confirmed that there is a positive correlation between capital adequacy and bank financial performance measured by ROA and bank stability measured by Z-score. This result is very important practically and it is theoretically satisfying. The positive coefficient of the capital ratio in the Z-score model suggests that stronger capitalization is associated with greater financial stability, consistent with the role of capital as a buffer against unexpected losses. This result indicates that banking institutions in Jordan that maintain stronger capital buffers were not merely safer, that is, less expected to reach the insolvency extent, but rather they also intend to score better ROA. The positive association between capital strength and both profitability and stability is consistent with the capital buffer theory and prior empirical evidence (Atari et al., 2026; Berger & Bouwman, 2013). Stronger capitalization may enhance banks’ resilience and reduce their exposure to financial distress by providing an additional buffer against adverse shocks.

According to the loan-to-deposit ratio, the results confirmed there is an insignificant effect of the loan-to-deposit ratio on financial performance, but it has an effect on bank stability. The results indicate that the loan-to-deposit ratio did not have a statistically significant effect on ROA or ROE. This suggests that variations in liquidity conditions, as measured by LDR, were not a primary determinant of profitability differences among banks during the study period. However, the negative coefficient observed in the stability model indicates that higher LDR levels may be associated with lower financial stability, although this effect was relatively modest.

This result is in line with Altarawneh and Shafie (2018), who revealed there is an insignificant effect of liquidity risk on ROA in the banking sector in Jordan.

According to the effect of COVID-19 pandemic on bank financial performance and bank stability, the results confirmed that there is a negative effect of this crisis on financial performance in 2020. This result is in line with the expectations and arguments of previous studies. This study’s results confirmed that despite controlling for risk metrics, the year 2020 had a unique negative shock effect on financial performance. This can be explained due to system-wide matters such as lockdowns, decreased lending actions, margin density because of interest rate decreases, and cautionary provisioning among all banking institutions that are not fully captured by bank-specific variables.

The COVID-19 dummy was negatively and significantly associated with profitability measures, indicating that the pandemic adversely affected bank performance during the study period. The coefficient was also negative and statistically significant in the stability model, suggesting a decline in bank stability during the pandemic period.

In conclusion, concerning the nonexistence of interaction between liquidity and credit risk, this consequence is revealing about in what way these risks are evident in Jordanian banks. It proposes that credit losses are not subject to how aggressive a bank’s lending (LDR) had been, at least in terms of impairing results.

Previous studies have suggested that liquidity and credit risk may interact under certain conditions (Imbierowicz & Rauch, 2014). However, the interaction term was statistically insignificant in the present study, indicating that the effect of credit risk on bank performance and stability did not significantly depend on liquidity conditions during the study period.

The findings should be interpreted considering the study’s relatively small sample size (15 banks) and limited time period (2018–2024). Although the results are robust within the study context, caution should be exercised when generalizing the findings to other banking systems or economic environments.

Overall, the findings indicate that credit risk and capital strength were the most important determinants of bank performance and stability during the study period. By contrast, liquidity exhibited a more limited role, while the interaction between liquidity and credit risk was not statistically significant. The COVID-19 shock negatively affected profitability and was also associated with lower bank stability, highlighting the importance of effective risk management during periods of economic uncertainty.

6. Practical Implications

The findings of this study provide several practical implications for bank managers, boards of directors, and regulators in the Jordanian banking sector.

6.1. Implications for Bank Management and Boards

The results indicate that credit risk is one of the most important determinants of both bank performance and financial stability. The significant negative relationship between loan loss provisions and ROA, ROE, and bank stability suggests that effective credit risk assessment and monitoring remain essential for maintaining sustainable banking performance. Accordingly, bank managers and boards should continue to strengthen credit evaluation procedures and portfolio monitoring practices.

In addition, the positive relationship between the Equity/Assets ratio and both profitability and stability highlights the importance of maintaining adequate capital levels. The findings suggest that stronger capitalization is associated with greater resilience and improved financial outcomes.

6.2. Liquidity Considerations

The loan-to-deposit ratio did not exhibit a significant effect on profitability measures, although its relationship with stability was more limited. Therefore, liquidity management remains important, but the findings suggest that credit risk and capital strength played a more prominent role in explaining differences in bank performance and stability during the study period.

6.3. Implications for Regulators

The results underline the importance of continued monitoring of credit quality and capital adequacy within the banking sector. Given the significant effects of these factors on both profitability and stability, regulatory attention to credit-risk management and capitalization remains relevant for supporting a resilient banking system.

Finally, the negative impact of the COVID-19 shock on bank outcomes highlights the importance of maintaining prudent risk-management practices during periods of economic uncertainty.

7. Limitations and Directions for Future Research

There are a number of limitations of this study. The sample consists of (15) banks listed in Jordan, which, even though comprehensive within Jordan, is still a small sample size. Thus, the results of the study ought to be explained carefully when comparing these results with other contexts particularly large-scale banks or more volatile markets around the world that may experience different bank risk dynamics.

Another limitation relates to the exclusion of additional macroeconomic variables such as GDP growth, inflation, and interest rates. This decision was primarily motivated by the relatively short time dimension of the panel and the inclusion of the COVID-19 dummy variable, which captures the major systemic shock affecting the banking sector during the study period. Nevertheless, future studies using longer time horizons may incorporate additional macroeconomic indicators to provide a more comprehensive assessment of the broader economic conditions influencing bank performance and stability.

Future research may extend the analysis by examining larger banking samples across multiple MENA countries, which would allow investigation of cross-country institutional and regulatory differences. In addition, studies using longer panel structures may be better positioned to employ dynamic panel estimators to address potential persistence and endogeneity concerns in bank performance and stability. Such extensions could provide additional insights into how risk factors influence banking outcomes under different economic and regulatory environments.

8. Conclusions

This study examined the effects of credit risk, liquidity, capital strength, and the COVID-19 shock on bank performance and financial stability in Jordanian banks during the period 2018–2024 using a fixed-effects panel-data framework.

The findings indicate that credit risk negatively affects bank performance and stability, while stronger capitalization is associated with improved profitability and stability. The COVID-19 pandemic was negatively associated with banking outcomes during the study period. In contrast, liquidity exhibited a more limited role, and the interaction between liquidity and credit risk was statistically insignificant.

Overall, the results suggest that credit risk management and capital strength are the most important factors associated with bank performance and stability in the Jordanian banking sector. The study contributes recent evidence from an emerging banking market and highlights the importance of examining both profitability and stability when evaluating banking sector resilience.

Future research may extend this analysis by incorporating longer time horizons, additional macroeconomic variables, or cross-country banking samples to further explore the determinants of bank performance and financial stability.