“The impact of the capital structure on Iraqi banks’ performance”

The current paper aims to investigate the effect of the capital structure on the profitability of a panel of eighteen Iraqi listed banks from 2009 to 2018. Furthermore, the unbalanced panel data approach (fixed effect and random effect) is utilized to explore the influence of capital structure on banks’ profitability. This study’s findings point out that the banks’ performance in terms of return on assets has a significant positive association with equity to assets ratio, liabilities to assets ratio, and bank size. On the other hand, long-term debt to assets ratio, short-term debt to assets ratio, and total debt to assets ratio showed a significant negative effect on banks’ performance. This study highlights new facts for an enhanced understanding of the capital structure and its association with banks’ performance in developing economies like Iraq. This study is considered one of the earliest studies of its types by determining the Iraqi banks’ optimal structure and examining capital structure’s impact on their performance. Nevertheless, the study contributes significantly to theoretical literature, policymakers, and industry so that conventional Iraqi banks can boost their performance.


INTRODUCTION
The capital structure is considered a difficult issue faced by financial decision-makers because of its association with many investment decisions.The low net current value for any project planned for investment therein may result from the capital's high cost.This can happen because of the increased financial leverage, which comes from the expansion of the dependence on debt.It may lead to non-profit due to the expansion of the dependence on debt.Management needs to identify the components of capital structure positively associated with profitability (Al-Zubaidi & Salamah, 2014; Watson & Head, 2007).Nevertheless, banks differ from others like service, commercial, and industrial companies in the level of their dependence on external funds.This is due to different fund resources and the nature of their different activities.Therefore, banks resort to a comparison between finance sources regarding risks, costs, returns, time, participation in management, obligations of interest, and installments.The optimal financial structure was achieved by maximizing the business valuation, as this valuation is essential for the debt value and the equity value (Al-Zubaidi & Salamah, 2014).
In achieving optimal performance, banks can use an assortment of strategies and techniques.The main strategies are through the capital structure.The association between the capital structure and the banks' profitability is right of huge importance to all banks.The capital structure's decision is of great importance to banks in particular because the banks are sensitive to the change in leverage due to the decrease in their capital to total assets (Al-Kayed, Zain, & Duasa, 2014).Furthermore, banks' capital requirements are regulated, and these sets of banks are different from other firms.Moreover, it is noted from previous studies that many studies dealing with the impact of the capital structure on banks' profitability still are very small, so this study is distinguished from previous studies that it is the first study of this kind in Iraq, which examines a broad sample of the influence of capital structure on banks' performance.Meanwhile, previous studies did not address the capital structure of Iraqi banks and their importance.The current study tries to fill this knowledge gap by examining the impact of capital structure on Iraqi banks' performance.

LITERATURE REVIEW
The capital structure explains how the company raises its resources to extend its corporate operations.The capital structure is a mixture of property and debt funds of all kinds, based on the company's financial decisions.Decades ago, a modern theory of the capital structure was well-known after Modigliani and Miller study (Modigliani & Miller, 1963).The Modigliani and Miller theory is the cornerstone of modern company finance.The cost of bankruptcy, the cost of management, the variability of information, and the market's efficiency, company value is not influenced by how it was funded, whether by issuing shares or by getting debt from external sources.The second proposal states that the firm's leverage does not influence debt proportion to equity.
The liabilities to assets ratio (LAR) is one of the most used indicators in measuring the degree of external financing sources in the banks' financial structure and determining the amount of debt per dinar to total assets.Additionally, this proportion indicates the size of the bank's potential risks in terms of the debt burden.If this ratio is greater than 100%, this indicates that the bank's debt is greater than its assets, and this means that it will face the risks of being unable to meet its obligations, especially long-term, which exposes the bankruptcy and liquidation problems.If this percentage is low, then this indicates that in its activities, the bank relies more on self-financing than it depends on debt, and one can say that the bank has assets that exceed the value of its debt (Alhamdan & Alqadah, 2013).However, the higher LAR ratio points out that the bank is loaned up, and its liquidity is low (Mikhan & Jain, 2007).Saeed and Badar (2013) and Umar et al. (2012) discovered a significant negative association between LAR and firm performance in Pakistan.However, Alhamdan and Alqadah (2013), Abor (2005), and Khashrama and Qaqish (2000) pointed out that the LAR has a positive association with firm performance.Thus, the LAR hypothesis can be written as follows: H1: Liabilities to assets ratio (LAR) is significantly positively related to banks' performance.
The high capital helps banks become quite safer during the liquidation and decrease their external fund resources, leading to their enhanced performance (Shamki, Alulis, & Sayari, 2016).Prabowo (2018), Alhamdan and Alqadah (2013) say that banks' financing costs will be few and depend on external financing to increase the profits achieved when banks are capitalized well.Many researchers examined the association between equity to assets proportion (ETA) and banks' performance was measured by ROA and ROE (Jadah, Alghanimi, Al-Dahaan, & Al-Husainy, 2020; Prabowo, 2018; Al-Farisi & Hendrawan, 2011).It is because the large size of equity will reduce capital risk.Nevertheless, the low capital ratio may affect public trust in the bank and its safety.Generally, there are many advantages to the large equity size by providing many services for the bank's customers, reflecting their increasing profitability (Bashir, 1999) (2007) found that the long-term debt to gross assets ratio has a negative effect on the Jordanian banks' profitability.In the same vein, Abor (2005) discovers a significant adverse association between LTDAR and the profitability of companies in Ghana.Hence, the LTDAR hypothesis is written as follows: H3: Long-term debt to gross assets ratio (LTDAR) has a significant negative association with banks' performance.
Numerous studies were carried out using shortterm debt to assets ratio (STDAR) to measure the capital structure.Moreover, the influence of shortterm debt to assets ratio (STDAR) on banks' performance appears to be positive in some previous studies.For example, Salteh, Ghanavati, Khanqah, and Khosroshahi (2012) discovered a significant positive influence of STDAR on Iranian firms' performance.Similarly, Arbabiyan and Safari (2009) pointed out that the STDAR is positively significantly related to banks' profitability using a panel of 100 companies from 2001 to 2007.Nevertheless, other studies discover a negative association between STDAR and banks' performance, for instance, Z. Ramadan and I. Ramadan (2015) explored a significant negative association between short-term debt ratio and a company's performance in Jordan.Likewise, Salim and Yadav (2012), using a panel of 237 Malaysian corporations from 1995 to 2011, pointed out a negative association between STDTA and firm profitability.Similarly, Zeitun and Tian (2007) discovered that the short-term debt to assets ratio is negatively associated with the banks' profitability in Jordan.

Thus, STDAR hypothesis is written as follows:
H4: There is a negative association between shortterm debt to assets and banks' profitability.
The results of previous research are inconsistent.Some researchers have pointed out that the total debt to assets (DTA) ratio and banks' profitability have significantly negative association.
Nevertheless, other researchers have discovered an insignificant association between DTA and banks' performance.In the panel of 100 companies from 2001 to 2007, Arbabiyan and Safari (2009) found that the total debt to assets ratio (DTA) is related positively to banks' performance.
Likewise, Abu-Rub (2012) found a positive association between total debt to gross assets ratio (DTA) and the company's profitability.Similarly, Salteh et al. ( 2012) stressed that 28 Iranian companies' performance is positively linked to the DTA.Nonetheless, Taani (2013) discovered an insignificant association between the capital structure and Jordanian companies' performance.Meanwhile, a major inverse association between the proportions of debt and corporate profitability was explored in Pakistan by Memon, Bhutto, and Abbas (2012).Similarly, Muritala (2012) found that the company's debt level and income are substantially adverse in Nigeria.The same negative impacts on the banks' profitability have been documented by Zeitun and Tian (2007).Additionally, Salim and Yadav (2012) discovered a significant negative relationship between DTA and firms' profitability using 237 Malaysian firms from 1995 to 2011.Thus, the fifth hypothesis may be written as follows: H5: There is a significant negative association between the proportion of total debt to gross assets and the Iraqi banks' performance.

Data sources and sample
The current study emphasizes the capital structure effects on the commercial listed banks' prof-

Model specification
The current study utilizes the panel data approach analysis because the panel data regression was the most appropriate.After all, it considered changes over time to performance indicators, banks' samples' different characteristics, and temporary changes in the banking environment (Bortolotti et al., 2002).The multiple regression approach is utilized to analyze the association between capital where ROA -return on assets, ROE -return on equity, β 1 -β 8 -coefficients of explanatory variables LAR -liabilities to total assets ratio, EAR -equity to assets ratio, LTDAR -long-term debt to assets ratio, STDAR -short-term debt to assets ratio, DTA -total debt to assets, BSZ -bank size (natural log of the total assets), ASG -asset growth (The current year's assetslast year's assets / current year's assets), INF -inflation, ε it -error term.

RESULTS
Before the model test, the descriptive statistics, Pearson correlation, Breusch-Pagan LM test, and Hausman specification test were used to fulfill regression requirements.

Descriptive statistics
Descriptive statistics are illustrated in Table 2 for the variables of the analysis.Table 2

Correlation
The matrix for the correlation between dependent and independent variables is used.Moreover, the association between independent variables shows there is no strong correlation between them.To study the correlation strength, Pearson correlation matrix for the sample banks' variables is discussed in Table 3. Table 3 demonstrates a correlation matrix for all study variables.The highest correlation was shortterm debt to assets ratio at 0.638.For all study variables, Table 3 illustrates a correlation matrix.
The short-term debt to assets ratio had the highest correlation of 0.638.It is found that the ROA is positively associated with liabilities to gross assets ratio, equity to assets ratio, bank size, asset growth, inflation, etc.However, ROA is negatively associated with the long-run debt to assets ratio and the short-run debt to gross assets ratio.Table 3 also illustrates that the ROE has been negatively correlated with long-term debt to gross assets ratio, short-term debt to gross assets ratio, total debt to gross assets ratio.Nevertheless, ROE was positively correlated with liabilities to gross assets proportion, equity to assets ratio, bank size, asset growth, inflation.4.

Regression analysis with ROE
In this subsection, panel data multiple regressions are utilized to test the association between capital structure and ROE: normality test, serial correlation test, multicollinearity test, and heteroscedasticity test for ROE model.
The output of the diagnostic tests indicates that there was no issue with the data.Moreover, the Breusch-Pagan LM test and Hausman specification test were used to fulfill the requirements of regression.The findings of diagnostic tests refer to this model does not suffer from any problems.Furthermore, the outputs of the Breusch-Pagan LM test show that the p-value is lower than 0.05; this indicates that the fixed effect approach or random effect approach is suitable.Furthermore, the Hausman test shows that the p-value is higher than 0.05.This indicates that the random effect approach is a suitable methodology to run the ROE model.The panel random effect for ROE results is illustrated in Table 5. Note: **, * show significance at 1% and 5% level, respectively.

Robustness check
To improve the robustness of the research findings, the association between the structure of capital and bank performance was analyzed further.It needs to be mentioned that the tables are not viewed because they would occupy so much space.Firstly, the authors also check if the relationship between capital structure and banks' performance is non-linear.Here, all the variables are quadratically described in Eq. ( 1) and Eq. ( 2).In the not tabulated findings, the fixed effect and random effects guesses of amended Eq. ( 1) and modified Eq. ( 2), respectively, with the quadratic terms, find insignificant coefficients on any of the quadratic capital structure variables.This result suggests then that the association between the capital structure and banks' profitability is linear.Secondly, we used alternative measures for bank performance.Thus, the authors use alternative measures for ROA and ROE, measured as return (before tax) on total assets and return (before tax) on equity, respectively.Finally, the authors re-test the primary model's regressions by using alternative the proportion of total debt to gross assets, with debt to capital proportion.The key results are closed to those reported in these cases found in Tables 4 and 5.

DISCUSSION
Tables 4 and 5 indicate that the total ratio (LAR) liabilities have a significant positive association with both banks' performance measures (ROA and ROE).This outcome does not support H1.Moreover, these results consistent with the findings of Alhamdan and Alqadah (2013), Abor (2005), and Khashrama and Qaqish (2000).
Nevertheless, this result contrasts with other studies (Saeed & Badar, 2013;Umar et al., 2012) who found negative influences.The plausible reason for this result is that increased use of this debt to finance investments in banks will increase the level of profit.Furthermore, the positive influence of a higher liability to gross assets ratio on profitability operates solely through the influence of leverage (because the low proportions of lending are artificial).Profitability eventually is an accounting concept that reflects efficiency levels as much as redistributive or economic policies.
The outcomes of Tables 4 and 5 point out a significant positive influence of the proportion of equity on Iraqi banks' performance (ROE, ROA).The benefits of equity funding are that the corporation does not need collateral, and the risk of financial loss is not raised by additional share funding.These findings indicate that the second hypothesis of this study is supported.Moreover, this finding corresponds to previous research (Jadah et al., 2020;Prabowo, 2018), which showed that highly profitable banks are more likely to rely on internally generated funds and share capital as their financing source.In practice, well-capitalized banks face lower bankruptcy costs, which reduces the cost of financing or has fewer external financing needs resulting in higher performance.Furthermore, this finding's plausible reason is the high proportion of capital to assets; the return on assets will also increase.This indicates that the capital used is required for the internal bank for operating activities in generating operating income and other investment activities.This result is supported by Sufyan's argument (2011) that banks face lower costs when well-capitalized then financing costs and dependence on external financing to generate a higher decline in profitability.According to Shamki et al.
(2016), high capital makes banks relatively safer in liquidation, reducing dependence on external financing and hence increasing profits.To sum up, the main sign of the positive coefficient estimate for equity-to-assets indicates the efficient management of banks' capital structure.
Based on the findings illustrated in Tables 4 and 5 2012) who identified a positive influence.This indicates that the rise in the long-term debt ratio is linked to reduced profitability.This shows that long-term debt is relatively expensive, which leads to reduced banks' profitability using high rates of long-term debt.Furthermore, the LTDAR ratio is compatible with other leverage metrics and suggests a loss of earnings by rising long-term debt.Besides, banking should be funded by internal funds that affirm the principle of pecking order (Myers & Majluf, 1984).Long-term debt is not likely to be important to such banks to produce income, which could be consistent with the Modigliani and Miller (1958) theorem of irrelevance in the capital structure.On the other hand, it may be important to the banks' decision-making but is considered a source of insufficient funding to increase profits.
From the estimation results of Tables 4 and 5, one can see that both ROA and ROE are adversely affected by short-term debt to gross assets ratio.These findings are in line with the study's predic-tion of a negative association between short-term debt to gross asset ratio and the bank's profitability.Moreover, this finding agreed with prior works of Z. Ramadan and I. Ramadan (2015), Salim and Yadav (2012), Zeitun and Tian (2007) who found a negative influence.This indicates that there is lower profitability in connection with the rise to the short-term debt ratio.This shows that short-term debt is more costly than equity, and therefore the use of a large share of short-term debt decreases the banks' profitability.The significant negative association discovered in the current research in line with the pecking-order theory (Myers & Majluf, 1984).
The findings of Tables 4 and 5 show that the association between total debt to gross assets and ROA and ROE calculate bank profitability are statistically negative.Therefore, the fifth hypothesis of this study will be confirmed.In Iraq, therefore, banks typically benefit more from lower leverage.This means that commercial banks can reduce overall debt and use domestic funds in particular to finance higher profitability.The results of this analysis thus correspond to Myers and Majluf's (1984) pecking-order theory.Moreover, this result is in harmony with Muritala (2012), Salim and Yadav (2012).This shows these banks with a high debt ratio collect borrowing costs.This decreases net profit, which results in a reduction in return on assets and return on equity.
As for control variables, the regression results in Tables 4 and 5 of this study illustrate that bank size positively significantly influenced both banks' profitability measures (ROE and ROA).This finding is in the same line with the work of Jadah et al. (2016).Bank size is generally used as a measure of the economy of scale of the banking industry.The plausible reason behind the positive influence of bank size on banks' performance is that the banks benefit from the economy of scale.Nevertheless, asset growth and inflation do not have any influence on banks' performance.

CONCLUSION
In financial management, capital structure is a highly delicate issue, which affects profitability.The key objective of the current paper is to analyze the debt calculated the capital to assets ratio, equity to gross assets ratio, overall debt to gross assets ratio, and the proportion of short-term debt to overall assets on banks' performance measured by return on assets and return on equity.A panel of 18 banks listed is ISE from 2009 to 2018 is used, following bank size controls, asset growth, and inflation.The results point out that the banks' performance measures are substantially responsive to measures of capital structure.The capital structure plays a significant role in the understanding of Iraqi banks' performance.

Table 1 .
The study variables summary Note: ROA: return on assets; ROE: return on equity; LAR: liability to total assets ratio; EAR: Equity to gross assets ratio; LTDAR: long-term debt to gross assets ratio; STDAR: short-term debt to gross assets ratio; DTA: total debt to gross assets; BSZ: bank size; ASG: asset growth; INF: inflation.

Table 3 .
Pearson correlation matrix

Table 4 .
The influence of the capital structure on banks' performance measured by ROA Note: **, * show significance at 1% and 5% level, respectively.

Table 5 .
The impact of the capital structure on banks' performance measured by ROE