Evaluating the Effectiveness of the Bird-in-Hand-Dividends Policy in the Stability of Jordanian Listed Banks

The present paper is empirically scrutinized the long and short-run causalities, which are running from the bird-in-hand dividends policy towards investors' preferences as proxied by banks’ stability. Through analyzing a quarterly data set covering the period Q1/1996-Q4/2018; results from the ADF test proved that the series variables became stationary only after including the first difference. However, although the Johansen test showed long-run integrations among variables; findings from the single equation of the error correction model asserted that there are no long-run causalities running from dividends’ policy towards investors’ preferences as captured by the Z-Score index “ZSI”, bankometer model or market capitalization. By contrast, results from the Wald test proved that except for earnings per share and retained ratio; the solvency of banks is found to be significantly responding to the change in dividends payout ratio. However, since there are short-run correlations among dividends’ stability, investors’ preferences and banking stability, the study concluded that the ZSI is significantly related to investors’ attitudes towards banks' decisions regarding dividends’ payments.


Research Problem and the Key Question(s)
Due to the significant role of banks in economic growth and development, there are numerous studies concerned with the potential determinants of banks' stability and banks' performance. For instance, the study of Ali (2016) revealed that the liquidity of banks as captured by liquid assets to total assets and total deposits to total assets is significantly affected by the rates of interest and inflation, price to earnings ratio and the market-to-book value ratio, over the long-run. However, Audo (2014) debated that the liquidity of Kenyan banks as captured by liquid assets to liabilities is not related to inflation rates. By contrast, Abdul Karim, et al. (2016); Wahid and Dar (2016) confirmed that banks' stability is statistically impacted by inflation, GDP and interest rates. Inconsistently, Diaconu and Oanea (2015) argued that banks' stability is not influenced by the GDP growth rate. Anyway, since the current research is also focused at measuring the impact of dividends in investors' preferences, the study presents a critical appraisal for the potential determinants of investors' behaviour. In this regard, Haron and Azmi (2006) showed that investors' behaviour is positively impacted by interest rates. However, studies including Ali (2020); Mojgan and Ali (2011) confirmed that investors' behaviour is significantly relating to dividends ratios such as EPS, DYR, and DPS. Consistently, Azam and Kumar (2011) proved that the E/P and the GDP growth rate are significantly affecting the stock index of the Karachi stock market as proxied by investors' behaviour. Studies including Bressan (2017) stated that the non-deposit leverage in the large American banks is positively associated with the dividends payout ratio, while the deposit leverage is found to be negatively correlating with the decisions of banks to pay dividends. Although, the available literature concerned with identifying the determinants of banks' performance and investors' behaviour, however, there is no evidence to support the effect of dividends policies in the stability of Jordanian banking system. Thus, to update the available literature as well as assisting investors to understand the role of dividends' policies in banks' stability, the study aims at evaluating the impact of the bird-in-hand dividends policy in the stability of Jordanian listed banks as proxied by the ZSI, market capitalization, and banks' solvency throughout the period Q1/1996-Q4/2018. Moreover, the core aim of this research will be mainly accomplished through answering the following main questions: ISSN 1923-4023 E-ISSN 1923 study discussed the irrelevance and relevance dividends policies as well as the impact of dividends policies in banks' performance. However, though the former dividends policy assumes that a firm's value is not related to dividends' polices, the later argues that a firm's dividends policy is significantly impacting a firm's value and stock prices. In other words, the bird-in-hand policy believes that the change in dividends is significantly affecting a firm's value through impacting stock prices. This view was initially supported by Lintner (1962) and Gordon (1963) as they detected a significant correlation between dividends and a firm's value. Similarly, since investors of this theory prefer dividends over capital gains, which are to some extent seen to be uncertain; authors including (Lundgren, 2016) asserted that a firm's value is significantly related to dividends' policies. Nevertheless, the irrelevance dividends policy argues that investors do not rely on dividends, while making their investment decisions. In this context, authors including Miller and Modigliani (1961) debated that investors behave rationally, especially in perfect stock markets, since all information is available for them and can be accessed equally without costs. However, through evaluating the effect of dividends policy in the performance of Kenyan listed banks, Lilian (2016) showed that banks' performance as proxied by ROA is insignificantly related to dividends per share. By contrast, Agyei (2011) argued that the performance of Ghanaian banks as captured by banks' value is significantly influenced by the dividends payout ratio during the period 1999-2003. Likewise, through investigating the effect of dividends policies in the market value of Kenyan banks, (Luvembe, et al., 2014) showed that the value of banks is positively relating to both dividends payout ratio and corporate earnings, over the period 2006-2010. Consistently, Amidu (2007 revealed that banks' return on assets is positively related to dividends policy and growth in sales, while variables like dividends payout ratio and leverage ratio are found to be negatively impacting the return on assets ratio. Paradoxically, Jepkemoi, et al. (2019) argued that the dividends payout ratio is insignificantly related to the financial performance of Kenyan commercial banks, as captured by banks' capital structure and liquidity. Furthermore, in an attempt to ascertain the influential variables of dividends payout's decisions (Imran, et al., 2013), proved that the decisions of dividends payout as captured by the current year dividends are statistically impacted by earnings per share, previous dividends payouts, capital ratio and bank's size. Additionally, the study showed that dividends payouts of the listed Pakistani banks are negatively responding to banks' cash flow.

Banking Stability Critical Appraisal
The review of previous studies showed that there are many researchers concerned with exploring the influential variables of banks' performance and banks' stability. In this regard, researchers including Diaconu and Oanea (2015) argued that the stability of Romanian banking system as measured by the ZSI is not associated with the GDP rate, while it is conversely responding to the change in loans/ deposit ratio. However, the study delivered that the profitability of banks as captured by ROA is significantly impacted by GDP growth rate. Nevertheless, according to Wahid and Dar (2016), the stability of Islamic and commercial banks is statistically affected by bank's size, equity/total assets and income diversification. However, the study revealed that banks' stability as gauged by the ZSI is negatively linked with inflation. However, through classifying Malaysian banks into three main categories, namely, commercial banks, Islamic banks and overall banking industry, (Abdul Karim, et al., 2016), revealed that the stability of commercial and overall banking industry models as represented by the ZSI is significantly related to interest, inflation and the GDP growth rates, during the long-run. Anyway, through slighting the light for identifying the influential factors of banks' liquidity and banking performance (Audo, 2014) showed that the liquidity of Kenyan commercial banks, as captured by liquid assets to liabilities is not affected by the rate of inflation. By contrast, Ali (2016) debated that the liquidity of Jordanian commercial banks as captured by liquid assets/total assets and total deposits/total assets is significantly co-integrated with interest and inflation rates, price/earnings ratio and market/book value ratio, over the long-run. Nevertheless, the liquidity of banks as proxied by liquid assets/total deposits and total loans to total deposits is found to be insignificantly correlated with market fundamentals. By contrast, Fadare (2011) debated that banks' liquidity as gauged by total loans/total deposits is significantly impacted by the monetary policy interest rate and the lagged loans to total deposit ratio. Additionally, studies like Subedi and Neupane (2013) asserted that the liquidity of Nepalese banks as captured by liquid assets/total assets and ROA is negatively related to inflation rate, short-term interest rate, GDP growth rate and the capital adequacy ratio. Consistently, Riaz and Mehar (2013) confirmed that the profitability of Romanian banks as captured by ROA is significantly associated with interest rates, while insignificantly related to total deposits/total assets and banks' credit risk. In the same vein, Irungu (2013) affirmed that the performance of Kenyan banks as measured by ROA is positively impacted by interest rate, liquid assets/total assets, GDP growth rate and the saving deposits' ratio. Beyond that, studies including Haron and Azmi (2006) showed that the GDP growth, inflation and interest rates are significantly impacting the performance of Malaysian commercial banks as measured by banks' deposits. However, through measuring the impact of macroprudential policies, banks' characteristics and banks' ownership in banks' stability as captured by the Z-score index; (Oktaviyanti and Purnawan, ISSN 1923-4023 E-ISSN 1923-4031 2019 found that the stability of banks is significantly impacted by variables such as the requirement reserve and the GDP growth rate.

Hypotheses' Development
This paper is empirically assessed the effectiveness of bird-in-hand dividends policy in banks' stability as captured by the ZSI, MC and the bankometer model. Thus, in an attempt to accomplish the core aim of this research, the study focuses on testing the following main hypotheses:


The review of previous related studies revealed that the way of how companies decide to distribute dividends exerts a vital role in affecting a firm's value and investors' behaviour. Thus, due to the importance of dividends in the process of decision making, there are many researches concerned with testing the impact of dividends in banks' performance and investors' behaviour. For instance, in an attempt to examine the correlation between dividends payout ratio and banks' performance Agyei (2011) confirmed that banks' performance is significantly related to the payout ratio. Additionally, Lintner (1956); Gordon and Shapiro (1956) revealed that the value of firm is significantly affected by dividends per share ratio. Accordingly, the researchers argued that investors prefer to acquire a company's share through focusing on dividends instead of looking for capital gains. Further dividends' theories debated that shareholders' wealth or a firm's value is insignificantly related to dividends policies (Miller and Modigliani, 1961). Thus, through relying on the above discussion, the present research formulated the set of the following null hypotheses: H 01 : The stability of Jordanian listed banks is insignificantly related to the bird-in-hand dividends policy.
This main hypothesis is divided into the following sub-hypotheses: H 01.1 : The stability of Jordanian listed banks is insignificantly related to the bird-in-hand dividends policy during the long-run.
H 01.2 : The stability of Jordanian listed banks is insignificantly related to the bird-in-hand dividends policy during the short-run.


The critical appraisal of previous researches revealed that there are several studies concerned with exploring the impact of dividends' policies on investors' attitudes. For example, a study by Ali (2020) hypothesized that the behaviour of Amman stock exchange's investors is significantly related to the volatility in the EPS and dividends payout ratio during the long and short-terms. As a result, the study showed significant correlations of investors' perceptions along with dividends' policies during the long-run. Previously, the researcher delivered that banks' liquidity as a proxy for investors' behaviour is significantly related to the change in the price to earnings ratio (Ali, 2016). Additionally, studies including Abdul Karim, et al. (2016) revealed that banks' stability as captured by the ZSI is significantly related to the interest, inflation and GDP growth rates, during the long-run. However, Mojgan and Ali (2011) confirmed that investors' behaviour is significantly related to EPS and DPS. Consistently, Azam and Kumar (2011) proved that the E/P and the GDP growth rate are significantly affecting the stock index of Karachi stock market as proxied by investors' behaviour. Although, the available literature slighted the light for identifying the determinants of banks' performance and investors' behaviour, however, there is no evidence to support the effect of dividends policies in the stability of Jordanian banking system, or the validity of investors' perceptions to mediate the relationship of dividends policies along with banks' stability. Therefore, the following hypothesis is also going to be tested: H 02 : Investors' preferences are not significant to mediate the long or short-run causalities, which are running from dividends' stability towards the stability of banks as captured by the ZSI.

Research Methodology
This research evaluated the effectiveness of the bird-in-hand dividends policy in the stability of Jordanian listed banks as captured by the ZSI, BM and MC. For this purpose the study intended to analyze a time series data relates to all Jordanian listed banks, which are 15 in total, during the period Q1/1995-Q4/2018. However, due to data inconsistency the study restricted to deal with a quarterly data set covered the period Q1/1996-Q4/2018, as well as the Safwa Islamic bank was excluded from the sample of this research. However, since the present research is mainly focused on testing hypotheses, those are primarily built through relying on the available literature, the positivism paradigm was adopted to achieve the core aim and objectives of this research. Therefore, the study used a set of empirical measurements like the ADF, Johansen Co-integration, ECM and the Granger Causality Wald tests in order to analyze a quarterly data set covering the period Q1/1996-Q4/2018. The used data are majorly drawn from the financial statements as well as financial reports, which are published by the statistical bulletins of ASE, the official sites of Jordanian central bank and the site of global economy (ASE, 2020; CBJ, 2020; the Global Economy, 2020). Moreover, since the study relies on time series variables, after running the descriptive test; the study applied the Augmented Dickey Fuller test to check for data stationarity. Thus, since the variables are found to be integrated of the same order I(I), the study restricted to run the Johansen co-integration test. Thereafter, the ECM and the Wald tests were applied to find-out whether banks' stability is being granger caused by the bird-in-hand dividends policy during both the long and short-terms.

Variables Measurements
For the purpose of accomplishing the core aim of this research, the study employed a set of quantitative-continuous explanatory and dependent variables. These variables and their measurements are described as shown in Table 1.

Earnings per Share=
Where, the BNP relates to the average of banks' net profit, divided by the average of banks' outstanding shares, during the time "t".

Retained Ratio=
It relates to the average of banks' net profit minus the average of distributed profits, related to the net profit's average.

Dividends Stability_Dummy
It relates to banks' decisions regarding dividends, and it took 1 when the relationship of EPS along with DPS is found to be positive, while -1 when it is found to be negative. DS_DUM =+1 or -1 → ↔

BS=
Where, BS represents banks' solvency, CA is the summation of banks' capital ratio, TETA stands for total equity/ assets ratio, CAR is the capital adequacy ratio, NPL refers to non-performing loans, CI is the cost of income ratio, and the TLTA represents the total loans/total assets ratio and TNLB relates to the total number of listed banks during the time t.

Z-Score Index=
Where, the ∑ROA represents banks' return on assets ratio, the ∑Equity relates to banks' equity, the ∑Assets refers to banks' total assets, and the σ.ROA stands for the standard deviation of return on assets ratio.
Where, ∆ is the difference operator of Jordanian listed banks' stability as captured by the ZSI, BM and MC and it is used as a main proxy for investors' preferences. : i= 1... 4 relate to the short-run dynamics, and they are the parameters to be estimated. While the , parameters relate to the long-run multipliers. i is the first lagged period of the error correction term, and refers to the white noise error term with usual properties. However, to check the likelihood of utilizing investors' preferences as a mediator that to mediate the relationship between dividends' stability and the ZSI; the following models are going to be tested: Where, the IPs refers to investors' preferences and it is employed as a mediator. Thus, after confirming that there is a significant correlation between dividends' stability and ZSI, as well as dividends' stability along with IPs, model number (4) is going to be considered to check whether IPs is valid to adjust the relationship of dividends' stability along with banks' stability as captured by the ZSI.

The Study Findings and Discussion
In an attempt to assess the effectiveness of the bird-in-hand in the stability of Jordanian listed banks, the current research employed a set of advanced empirical techniques like the descriptive test, ADF, Johansen, ECM and the granger causality Wald tests, in order to analyze a quarterly time series data covering the period Q1/1996-Q4/2018. Consequently, results from these tests are demonstrated as shown in tables 2-12: This table presents results from descriptive test, which is conducted to explain the variables' data in terms of means, standard deviations, variances, skewness and kurtosis. The above Table 2 reports that throughout the study's period the used variables are found to exhibit significant variations in terms of magnitude. However, except for EPS, DPR and RR the standard deviations of other variables are found to be relatively high. In addition, variables including the growth in stock prices, EPS, DPR, RR and banks' solvency are found to be highly skewed to the right. On the other hand, since kurtosis's values of the ZSI, MC and dividends' stability are found to be less than 3 the variables' data seemed to have flatness, and they are lighter-than-normal tails. By contrast, variables including stock's growth, EPS, DPR, RR and banks' solvency are found to be sharply peaked and having heavier-than-normal tails.
The following table presents results from ADF, which is used to check for data stationarity. Note: *** indicates that the time series variables are significant at 5% level of significance. Table 3 confirmed that when the used series variables tested at levels, they are found to be non-stationary. However, in order to avoid the problem of getting spurious results the variables are converted into the first difference. Thus, after including the first difference the variables became stationary, as well as integrated of the same order I(I).

Results from
This table describes results from Johansen test, which is conducted to check the long-run integration among dividends' polices and banks' stability. Since results from the ADF test showed that the time series data became stationary as well as integrated of the same order I(I), the Johansen cointegration test was performed, and subsequently as it is shown in table 4 the variables of the bird-in-hand dividends policy are found to be significantly integrated with investors' preferences, over the long-run. Thus, since results confirmed that banking stability as captured by the ZSI, BM and MC is significantly co-integrated with the dividends' measures; the study moved on to run the ECM to check whether there are long-run causalities running from dividends policies towards banks' stability (see table 5-7).
This table describes the long-run causalities, which are running from dividends' policy towards banks' stability as captured by the Z-score index. As shown in Table 5, since results from the ECM showed that the error term is negative, while the P-value is not significant; the study confirmed that the stability of banks as proxied by the ZSI is not being granger caused by the quarterly change in EPS, DPR, or the RR during the long-run. Which in other words mean that investors of Amman stock exchange do not rely on the change of dividends' payments when they think to acquire or liquidate their shares at the Jordanian listed banks.
This table describes the long-run causalities, which are running from dividends policies towards banks' stability as captured by Bankometer model. The above Table 6 confirmed that there are no long-run causalities running from the quarterly change in EPS, DPR, or the RR towards banks' stability as gauged by the bankometer model. Which in other words mean that banks' decisions regarding paying dividends do not affect investors' preferences as captured by the stability of Jordanian listed banks, over the long-run.
This table describes the long-run causalities, which are running from dividends policies towards banks' stability as captured by market capitalization.   The present study is also aimed at checking whether investors' preferences act as a mediator to adjust the relationship of dividends' stability along with the ZSI as a proxy for banks' stability. Thus, since results from table 9 showed a long-run integration of the ZSI along with dividends' stability, and similarly a long-run integration was also detected between investors' preferences and dividends stability; the study shifted to regress the ZSI on dividends' stability, as well as investors' preferences on dividends' stability separately (see results in tables 10 "Panel A & B").
This table shows results from the ECM, which is employed to regress banks' stability on dividends stability and investors preferences on dividends' stability separately. As shown in Table 10, after regressing the stability of banks on dividends stability, as well as investors' preferences on dividends stability; results from the ECM showed that the ZSI as well as investors' preferences as captured by the growth of banks' stock prices are separately, and significantly related to dividends' stability during the long-run. Which in other words mean that a 1% increase in dividends' stability decreases the stability of banks by 5.608. Therefore, the results indicated that the enhancement of dividends stability at the Jordanian listed banks is negatively impacting banks' stability as captured by the ZSI. By contrast, the results showed that a 1% increase in dividends stability motivates investors of Jordan to invest into the banks. which in other words, the findings showed that the 1% enhancement in dividends stability leads to a 15.7% increase in the stock prices of Jordanian listed banks. Moreover, since the above results confirmed that dividends stability is significantly impacting the stability of banks, as well as investors' preferences; the study can now proceed to test whether investors' preferences are valid to mediate the long-run causalities, which are running from dividends stability towards the stability of banks as proxied by the ZSI. Since the above Table 10 proved that investors' preferences is valid to be utilized as a mediator, the Johansen test was applied to find out whether investors' preferences are able to adjust the long-run integration of dividends' stability along with banking stability. However, although results from Table 11-panel A, proved that there are long-run integration among dividends' stability, investors' preferences and the ZSI; results from the ECM as demonstrated by panel B confirmed that the stability of banks is not being granger caused neither by dividends stability nor by investors' preferences. Which in other words mean that when investors' preferences were included in the model as a mediator, no relationship between dividends' stability and the ZSI was detected. Based on these results the study indicated the stability of Jordanian banks is not impacted by the nature of relationship between dividends' stability and investors' preferences. However, to check whether investors' preferences are valid to adjust the short-run correlation of dividends' stability along with the ZSI; the granger causality Wald was finally applied. Although results from the ECM as shown in Table 11 confirmed that the stability of banks is not being granger caused by dividends stability or investors preferences, however, the above Table 12 proved that the stability of banks as proxied by the ZSI is significantly related to dividends' stability and investors' preferences in the short-run. Therefore, the study found that since dividends' stability has a power to impact the growth of stock prices it can be concluded that the change in stock prices play a vital role to affect investors' decisions, and subsequently that will affect the stability of Jordanian listed banks as captured by the ZSI. Accordingly, investors' preferences are found to be valid to mediate the relationship of dividends' stability along with the ZSI, during the short-run.

Conclusion and Recommendations
This paper concerns of assessing the effectiveness of the bird-in-hand dividends policy as represented by the EPS, DPR and RR in the stability of Jordanian listed banks as a main proxy for investors' preferences. According to this research, banks' stability was mainly gauged by utilizing variables such as the Z-score index, market capitalization and the bankometer model as a measure for banks' solvency. To achieve the core aim of this research the study employed a set of advanced empirical techniques to analyze a time series data covering the period Q1/1996-Q4/2018. Thus, since time series data are vulnerable to increase or decrease over time, the study applied the augmented dickey fuller test to check for data's stationarity as well as avoiding the problem of getting spurious regression findings. Consequently, the results confirmed that the used series variables became stationary only after including the first difference. However, since results from Johansen test revealed that the trace values are higher than their critical values; the study ran the ECM to investigate the long-run causalities, which are running from dividends policy towards banks' stability. As a result, the study failed to accept the first sub-hypothesis, which is in other words mean that the stability of banks is not related to the quarterly change in EPS, DPR, or RR over the long-run. This result disagrees with Luvembe, et al. (2014); Agyei (2011) as they confirmed that the value of banks is positively relating to dividends payout ratio. Consistently, Amidu (2007) revealed that banks' return on assets is positively related to dividends policy. Alternatively, studies including Lilian (2016) showed that banks' performance as proxied by ROA is insignificantly related to dividends per share. Based on the achieved results, the study concluded that investors of ASE don't rely on the assumptions of the relevance dividends theory, when they think to invest into the banks. In other words, investors of ASE are not following the bird-in-hand dividends policy, which states that investors prefer dividends over capital gains, which are to some extent seen to be uncertain (Lintner, 1962;Gordon, 1963).
However, though results from the ECM showed that dividends' stability is significantly impacting the stability of banks during the long-run, however, after including investors' preferences in the model this effect was totally eliminated. Thereby, the study concluded that during the long-run the correlation of dividends along with banks' stability is mostly relied on variables other than investors' preferences. Additionally, the study showed that a 1% increase in dividends' stability increases the stock prices of banks by 15.7%. Authors including Azam and Kumar (2011) proved that E/P is significantly affecting the stock index of Karachi stock market as proxied by investors' behaviour. However, Ali (2020) showed that the behaviour of Amman stock exchange's investors as captured by the stock index is not related to the change in P/E, EPS and DYR in the long-run. However, since the current research revealed that investors' preferences is significantly related to dividends' stability; the study concluded that investors of Jordan prefer to invest in the stocks, which are paying stable dividends, instead of investing in stocks which are assumed to decrease or retain dividends.
By contrast, results from the Wald test confirmed that banks' stability as measured by the BM is significantly related to the change in DPR during the short-run. In this regard, the study induced that banking stability is significantly affected by the change in banks' decisions regarding dividends' distributions. Furthermore, since results from Wald test revealed that the stability of banks relies on investors' preferences during the short-run, the study derived that when banks relies on EPS to adjust their DPR that is expected to be significantly linked with banking stability, due to the impact of dividends' stability in the way of how investors thinking towards investing in banks' stocks. Studies including Ali (2020) revealed that the perceptions of ASE's investors are significantly impacted by dividends policies during the short-run. On the contrary, the study asserted that the stability of banks as captured by MC is not being granger caused by dividends policy during the short-run. However, based on the aforementioned findings the study concluded that during the short-term investors of ASE are found to be proponents for the bird-in-hand dividends policy, while during the long-term they are found not to follow the assumptions of the relevance dividends theory. Therefore, the study recommends that it is necessary for Jordanian listed banks to rely on dividends decisions in order to anticipate investors' preferences during the short-run. Additionally, it is important for banks to be cautious of cutting dividends, because that will affect their performance, since investors of Jordan prefer to invest into the stocks, which are paying stable dividends. Furthermore, there is a necessity for further researchers to search for the other influential factors of banking stability, as well as measuring the relationship of the irrelevance dividends policy along with banks' stability and investors' reactions.