THE BEHAVIOR OF BANKERS TOWARDS PROFIT AND LOSS SHARING CONTRACTS: A Modified Theory of Planned Behavior Approach

This paper aims at determining the psychological factors of bankers in implementing Profit and Loss Sharing (PLS) financing contracts in Islamic banks. The phenomenon of low occurrence of PLS financing motivated the researcher to write this paper. The theory of modified planned behavior is used to explain the research problem. The research data were obtained from 139 employees of Islamic banks in Indonesia and processed using SEM-PLS. The results show that trust has an effect on attitudes; subjective attitudes and norms affect intentions; perceived behavioral control has no effect on intention and behavior; intention influences behavior; and the perception of inefficiency has a negative effect on the relationship between intention and behavior. This paper contributes to the elaboration of the psychological factors affecting the behavior of bankers in implementing PLS. The implications of this paper for designing the strategies to improve the bankers' behavior in PLS financing are: the management must boost the bankers' confidence for the PLS financing benefits; the stakeholder’s social pressure is needed; the authority of bankers in implementing PLS financing needs to be increased; the bankers' access to the partner’s financial reports needs to be expanded; and the regulators need to give appreciation to Islamic banks that have a larger PLS portfolio.


Introduction
Islamic bank is a financial institution entity that has been operated by using the sharia principles as the basis (Anwer et al. 2020, 12;Martiana et al. 2022, 3). In contrast to conventional, profit-oriented banks, Islamic banks have social goals and concern for the environment through the practice of sharia values (Agriyanto 2015, 79;Sufyati 2021, 526). Islamic banks do not only have an intermediary function in the financial sector, but also have a social justice function. As baitul mâl, Islamic banks have economic instruments for the distribution of wealth through the collection and distribution of zakat, infâq and alms. This instrument leads Islamic banks to have a social function. (Ipandang & Djaoe 2022, 89). In addition to having a social function, Islamic banks also operate to look for profit. In gaining profit, Islamic banks certainly implement sharia rules. Sharia rules UlUl AlbAb: Jurnal Studi Islam, Vol. 23, No. 2, 2022 create justice for the contracting parties. Contracts in sharia are different from conventional contracts which tend to exploit one party through a fixed interest system. All contracting parties in Islamic banks play the same role as cooperation partners. Cooperation contracts in Islamic banks are called mud{ ârabah and mushârakah. The parties in the mud{ ârabah and mushârakah contracts are called the owners and managers of capital. Islamic banks act as the owners of capital and the partners act as the managers of capital. Islamic banks and partners have risks, related to the capital portion for the sustainability of the business run by partners. As capital managers, partners are not burdened with the obligation to pay fixed interest, but profits are divided based on the agreement. The amount of profit should not be determined in advance, but should be based on the realization of the business run by the partners. This concept makes mud{ ârabah and mushârakah contracts have fair value and differentiate them from conventional banks. (Supriatna et al. 2020, 238). In addition to the positive side, PLS also has negative sides such as concerns about moral hazard, fraud and other activities that are not in the best interests of the banking industry (Ascarya 2009, 88). Entrepreneurs can act opportunistically through hiding the profitability of the project by placing more burden on Islamic banks (Fakir et al. 2020, 285).
The Financial Services Authority of Indonesia (OJK) reports that the average proportion of PLS financing to total financing is only about 30 percent.
Year after year, there is no significant increase. The aforementioned Sharia banking data demonstrates that ideals are not always in line with reality. The PLS system is not supported by empirical evidence, according to the prevalent view in society regarding the primary distinction between Islamic and conventional banks. It is worrisome that a long-term low PLS financing will alter the community's perception of Islamic banks. In the absence of a substantial proportion of PLS financing, Islamic banks may be perceived as conventional banks. Observing low PLS financing from the bankers' behavior in making financing decisions is inextricable. Bankers appear to dislike PLS financing more than other forms of financing. Therefore, the issue in this study is the low implementation of PLS financing in Indonesian Islamic banks. The successful implementation of Islamic banking operations is the realization of justice, unity, and even the distribution of human welfare. Compared to mushârakah-based products, Sharia bank products with mud{ ârabah and mushârakah agreements are highly relevant for expanding welfare and reducing poverty. However, Islamic banks in Indonesia have a small share of PLS financing implementation. The focus of the research is the subpar performance of Islamic bankers in the implementation of PLS financing. The Theory of Planned Behavior explains the behavior of bankers when implementing appropriate PLS financing (TPB) (Ajzen 1991, 181). TPB is a well-established theory for analyzing individual behavior that is used to analyze bankers' behavior when implementing PLS financing.
TPB represents the evolution of Reasoned Action/TRA. Initially, according to TRA, individual behavior can only be predicted based on intentions; attitudes and subjective norms can be used to predict intentions. However, TRA has the drawback that it cannot predict voluntary behavior (Ajzen, 1991, 181). Voluntary behavior is when an individual can choose whether or not to engage in behavior. For instance, the implementation of PLS financing in Indonesian Islamic banks is primarily determined by the free will of manager-level bankers. Ajzen (1991) modified the TRA by introducing a perceived behavioral control variable to predict behavior and intention. The variable of perceived behavioral control is added because one's conduct is influenced by the factors of difficulty or ease. By presenting TPB's perceived behavioral control variables, the TRA was created. According to the Theory of Predictive Behavior, a person's behavior can be predicted by behavioral intention and perceived behavioral control. There are three antecedents to the behavioral intention: attitude, subjective norms, and perceived behavioral control. All attitudes, subjective norms, and perceived behavioral controls have a trust-based foundation. Psychological theory has been adopted to explain economic behavior such as investment decision-making (Bakar & Yi 2016, 321). TPB has also been used to explain the behavior of sustainability financial statements presentation (Thoradeniya et al. 2015, 111), purchase intention in halal marketplace (Fuadi et al. 2022, 100), farmers that use Islamic finance for agricultural production (Abid & Jie UlUl AlbAb: Jurnal Studi Islam, Vol. 23, No. 2, 2022, behavioral intention to adopt Islamic banking services in Malaysia (Shith et al. 2021, 30), customers' behavior in using Islamic P2P lending (Rofiqo et al. 2022, 76), and dividend policy decisions (Liao et al. 2022, 2). The research shows that although the Theory of Planned Behavior is derived from psychological theory, it has been used to explain the reasons why people are engaged in certain behavior in various fields, including finance. This argument underlies the use of Theory of Planned Behavior to explain the behavior of bankers in implementing financing with PLS contracts.
PLS financing is a contract between the capital owner (s} âhib al-mâl) and the manager (mud{ ârib) or financing partner. The PLS financing system creates agency issues for Islamic banks and their financing partners. As has been revealed, the agency relationship between Islamic banks and financing partners can result in information asymmetry and conflicts of intent (Jensen & Meckling 1976, 5). Information asymmetries occur as a result of the fact that financing partners understand the business more than Islamic banks do due to the cooperation results. The problem of the emergence of asymmetry in Islamic banking has been the subject of numerous studies. Ahmed discovered that the imposition of fines is influenced by the tardiness and dishonesty of partners in mud{ ârabah contracts. By monitoring and aligning incentives, PLS contracting can reduce transaction costs and minimize asymmetric information. Islamic banks, with the authority to engage in partner management affairs, incur greater monitoring expenses than Islamic banks (Ahmed, 2002, 102). These monitoring costs render PLS financing inefficient, thereby discouraging Islamic banks from focusing on PLS financing. If the benefits (return) are less than the costs incurred to obtain these benefits, PLS financing is required.
In the theory of Planned Behavior model, this study presents the perception variable of inefficient PLS financing. Modifications are made to the Theory of Planned Behavior Model to address the research question of what factors contribute to the low implementation of PLS financing at Islamic banks. The variable perception of inefficient PLS financing is moderately high, but the intentions for profit-sharing contracts is low. The assumption that bankers will be reluctant to implement PLS financing is caused by agency costs, which increase contract costs.

Research Method
The population of this study was all bankers in Islamic banks that could implement PLS financing. The sampling technique used was nonprobability sampling. This technique was best suited to the objectives of the study. The size of the study sample was decided based on the Cohen approach (Hair et al. 2012, 325). Minimal research samples decided with the Cohen approach have considered the statistical power and effect size. The minimum sample is determined by looking at the most significant number of arrows regarding one construct or variable in the research model (Cohen 1992, 156). Based on the empirical model of this study, the most significant number of arrows regarding the construct of intention implies PLS financing, which are up to 3 pieces. The Cohen table provides a clue that the three arrows concerning the construct have an expected significance level of 5% and a minimum R2 of 0.10 with 124 respondents.
The data analysis used quantitative methods with the SEM-PLS approach. The data management used WarpPLS statistical software. The arguments that suggest the use of WarpPLS to process data were: first, the research objective is to explore or expand the theory; secondly, WarpPLS can process complex models and can directly display the results of the coefficients and p-values for moderating variables. The resampling setting of this study is "stable" because the stable method generates the p-value that approximates the value generated by other sampling (bootstrapping, jackknifing, and blindfolding). Indeed, the disclosure of the choice of settings is essential because it will cause different results (dramatic effects) on the PLS output (Kock 2011, 3).
The estimation of a measurable model precedes hypothesis testing. The measurement model or external model seeks to examine the relationship between unobserved, indirectly measurable latent variables (Hair et al. 2012, 333). The reliability and validity of measurable models are assessed. Composite reliability and Cronbach's alpha are used to measure reliability. Two components of construct validity were examined: (1) convergent validity (loading factor and AVE) and (2) discriminant validity (AVE root and correlation between latent variables).
Structural models are also referred to as models that concentrate on the hypothetical or path relationships between latent variables (Hair et al. 2012, 332). The structural model is viewed from three different perspectives. The first one is the measure of the Determination Coefficient (R-Square) for each endogenous latent variable in the structural model with critical criteria (0.75), moderate criteria (0.50), and weak criteria (0.25). The second one is a measure of predictive relevance or Q (Squared). The last one is the effect size (F-Squared effect size), the absolute value of the individual contribution of each latent predictor variable to the value of the R-Squared criterion variable. There are three distinct effect sizes: small (0.02), moderate (0.15), and large (0.35).
In this study, six hypotheses are proposed. Hypotheses testing is based on the significance level and coefficient value. Therefore, the hypothesis is accepted if the significance value is below 5% and the coefficient value corresponds to the direction of the hypothesis. In contrast, the hypothesis is not accepted if the significance value is greater than 5% but the coefficient value is not.
Questionnaires were distributed through direct delivery to Islamic bankings in Indonesia using shipping services and through online using Google forms. The data collection for Islamic banking in Central Java is carried out online, using mail, and direct visits.  Table 2 shows that out of 201 questionnaires distributed, 154 or 76.6% were returned. Of the returned questionnaires that could be processed, there were 139 questionnaires. The research data of 139 respondents were sufficient to be processed in the research model since the calculation of the minimum data requirements for the research model sticks to 124 respondents (Cohen 1992, 99).
The purpose of the confirmatory factor analysis is to evaluate the validity and reliability of each construct or latent variable. A construct is considered reliable if it has a composite reliability value>0.70. Validity is divided into two, namely convergent validity and discriminant validity. Convergent validity is met if it has a factor loading value>0.70, significant at the level<0.05 and has an average variance extracted (AVE) value of 0.50. Discriminant validity is met if it has a root value of AVE>0.50 and the cross-loading value is lower than the construct loading value (Hulland 1999, 105).

Results and Discussions
The composite reliability, loading factor, and average variance extracted (AVE) values are displayed in Table 3. Each variable in the research demonstrates a composite reliability value that is greater than 0.70. These results suggest that the instrument used to measure the variable has a high degree of reliability. In addition, reliability indicates that the instruments used to measure the variables in this study produced identical results each time measurements were taken.  Table 3 demonstrates that the AVE value for all variables is greater than 0.50, indicating that the research instrument has a high degree of convergent validity. An instrument that satisfies the convergent validity criterion can collect data with the same pattern in order to measure the same construct. The cross-loading value is lower than the construction loading value, as shown in Table 4. The cross-loading value indicates that the criteria for discriminant validity have been satisfied. The indicators of discriminant validity can also be gleaned from the results of the root value of AVE>from correlations with other constructs. The results of the AVE roots in the diagonal column indicate that the AVE roots for all variables are greater than the correlations of the other constructs. Thus, the cross-loading and AVE root values indicate that the discriminant validity of the instruments in this study is met. The structural model analysis is utilized to test hypotheses. The outcomes of data processing with the WarpPLS software are depicted in fig. 2 Figure 2 displays the relationship between the paths as follows: Trust influences the attitude of implementing PLS financing with a coefficient of 0.31, significant at the 0.01 level; Attitudes influence the intention in implementing PLS financing with a coefficient value of 0.18, significant at the 0.05 level; Subjective norms influence the intention in implementing PLS financing with a coefficient of 0.55, significant at the 0.01 level; Perceived behavioral control has no effect on the intention in implementing PLS financing with a coefficient value of 0.09, significant at the level of 0.14; Perceived behavioral control has no effect on the behavior of implementing PLS financing with a coefficient of 0.10, significant at the 0.16 level; Intention influences the behavior of implementing PLS financing with a coefficient value of 0.74, significant at the 0.01 level; Efficiency moderates the negative intention relationship to the behavior of implementing PLS financing with a coefficient of -0.15, significant at the level of 0.10. The conclusions of each hypothesis test are presented in table 5.  (Sholihin & Ratmono 2013, 25). According to table 6, the R-squared attitude of implementing PLS financing is 0.097, which means that the belief in implementing PLS financing can explain 9.7% of the variance in the attitude of implementing PLS financing. R-squared intention in implementing PLS financing of 0.587 indicates that the predictor variables of attitude, subjective norms, and perceived behavioral control can explain 58.7% of the variance within the intention of implementing PLS financing. R-squared behavior implements PLS financing of 0.678, which indicates that variations in behavior can be explained by intention, perceived behavioral control, and moderating the variable efficiency by 67.8%. Table 6 also includes a measurement of the Q-indicator of predictive validity. Q-Squared can evaluate the predictive validity or significance of a set of latent predictor variables with respect to a criterion variable. Good predictive validity requires a Q-Squared value to be greater than zero. Because Q-Squared SIKAPBH is 0.102, MINATBH is 0.583, and PRIBH is 0.682 or greater than zero, the estimation results for this research model indicate that its predictive validity has been good.

The Influence of Trust on the Attitude towards Implementing PLS Financing
Empirical testing of the research accepts the first hypothesis, which states that trust influences the attitude towards implementing PLS financing. The argument of accepting the first hypothesis states that trust influences the attitude of financing because Islamic banking bankers already UlUl AlbAb: Jurnal Studi Islam, Vol. 23, No. 2, 2022 have high confidence in the PLS financing as the results. The empirical evidence of the theory of Planned Behavior research explains that belief in objects or behaviors will affect attitudes. Therefore, the theory of Planned Behavior can explain the effect of trust on the attitude of PLS financing. The results of this study support the results of previous studies, which found out that belief in behavior or objects influences attitudes (Davis 1989, 321;Ives et al. 1983, 788). These results are also in line with the research conducted in the field of Islamic banking which pointed out that the factor of religiosity or trust had an effect on attitude (Abid & Jie 2022, 3). Religion has an effect on behavioral intention to adopt Islamic banking services in Malaysia (Shith et al. 2021, 43). This fact explains that a banker's positive or negative attitude is determined by the belief in the benefits of financing with a PLS contract. These results indicate that bankers need to continue to improve their understanding and benefits of financing with PLS contracts. Islamic Banking bankers also need to understand that the purpose of Islamic banking is not just for profit. Islamic Banking bankers need to understand that Islamic banks also have a goal to promote social justice and promote Islamic financial instruments such as zakat, infâq, s{ adaqah.

The Influence of Attitude on the Intention of Implementing PLS Financing
Empirical evidence of this research has accepted the second hypothesis, which states that attitudes affect the intention to implement PLS financing. The cause for accepting the second hypothesis is the fact that Islamic banking bankers already have a high positive attitude toward PLS financing. This empirical evidence is consistent with the Theory of Planned Behavior, which affirms that a person's attitude will affect intention. Thus, it can be concluded that the Theory of Planned Behavior can explain behavior in the field of PLS financing, especially about the influence of attitudes on the intention in implementing PLS financing. Furthermore, the results of this study are following several behavioral studies of Hays et al. (2013) who found that attitudes influenced the intentions of management accountants reporting fraud, Thoradeniya et al. (2015, 111) who discovered that attitudes influenced the manager's intention in conducting sustainability report, and Warsame & Ireri (2016) who found that attitudes affected the intention in sukuk investment. Attitude affects the behavior of farmers in using Islamic Banking (Abid & UlUl AlbAb: Jurnal Studi Islam, Vol. 23, No. 2, 2022Jie 2022. Attitude has an effect on behavioral intention to adopt Islamic banking services in Malaysia (Shith et al. 2021, 43). Attitude had a significant positive effect on the intention to contribute to Islamic P2P lending (Rofiqo et al. 2022, 76).

The Influence of Subjective Norms on the Intention in Implementing PLS Financing
Empirical evidence of this study accepts the third hypothesis, which states that subjective norms influence the intention in implementing PLS financing. This research's empirical evidence follows the Theory of Planned Behavior, which confirms that subjective norms will affect intention. Thus, the Theory of Planned Behavior can explain the behavior in implementing PLS financing. This study is in line with several previous studies which found that subjective norms had an effect on the intention in using Islamic banking services (Abid & Jie 2022, 3). Subjective norm affects intention in using Islamic P2P lending (Rofiqo et al. 2022, 76). Subjective norm also affects the interest in using the halal marketplace (Fuadi et al. 2022, 100). However, this study is not in line with the research which found out that subjective norms had no effect on behavioral intention to adopt Islamic banking services in Malaysia (Shith et al. 2021, 43).

The Influence of Perceived Behavioral Control on the Intention in Implementing PLS Financing
Empirical evidence of this research does not accept hypothesis 4 which states that perceived behavioral control influences intention in implementing PLS financing. Islamic banks cannot execute their PLS financing. Disbursement with the PLS system is usually not as easy as giving a loan with a sale and purchase agreement. The provision of financing using the PLS system must pass the financing committee screening regardless of the nominal financing. This is understandable because financing using PLS systems requires the need to focus on the level of the risk of business partners. After all, Islamic banks can bear losses by the portion of capital. Minimum PLS targets obtained by partners are also a concern. Empirical evidence of this study does not support the theory of planned behavior which affirms that perceived behavioral control influences intention. This is caused by the fact that the implementation of financing using PLS systems cannot be executed directly by sharia bankers (account officers). Thus, implementing PLS financing is not a voluntary behavior. It must pass through the financing committee's stumbling block. This study is not in line with the research which discovered that perceived behavioral control had a significant positive effect on the intention to contribute to Islamic P2P lending (Rofiqo et al. 2022, 76).

The Influence of Perceived Behavioral Control on the Behavior of Implementing PLS Financing
The empirical evidence of this study does not support hypothesis 4.b, which states that the perceived behavioral control influences behavior using accrual basis accounting information. The argument of not accepting hypothesis 4.b says that perceived behavioral control influences the intention in implementing PLS financing because, in reality, Islamic bankers cannot execute their PLS financing. Disbursement using the PLS system is usually not as easy as giving a loan with a sale and purchase agreement. The provision of financing using the PLS system must pass the financing committee's stumbling block regardless of the financing nominal. This is understandable because financing with PLS systems requires the need to focus on the level of risk of business partners. After all, Islamic banks can bear losses by the portion of capital. Minimum PLS targets obtained by partners are also a concern. Empirical evidence of this research does not support the Theory of Planned Behavior which states that perceived behavioral control influences intention. This is caused by the fact that the implementation of financing using a PLS system cannot be executed directly by Islamic bankers (account officers). Implementing PLS financing is not a voluntary behavior. It must pass through a stumbling block, namely the financing committee. These results are in line with research which found out that perceived behavioral control was negatively affected by the behavior of farmers to use Islamic banking (Abid & Jie 2022, 3). Perceived behavioral control has no effect on behavioral intention to adopt Islamic banking services in Malaysia (Shith et al. 2021, 43).

The Influence of the intention on the behavior of implementing PLS financing
The empirical evidence of this research supports the fifth hypothesis, which confirms that intention influences the implementation of PLS financing. The positive coefficient can be interpreted that the higher intention in implementing PLS financing will improve the behavior of implementing PLS financing. Therefore, a banker's intention in implementing financing with a PLS system may not all be realized in the form of behavior. The results of this study are in line with research stating that intention has an effect on customers' behavior to use Islamic P2P lending (Rofiqo et al. 2022, 77).

The role of moderating inefficiency in the influence of intention in the behavior of implementing PLS financing
Empirical evidence of the research accepts the sixth hypothesis. It states that there is a moderating role for inefficiency regarding the influence of intention in the behavior of implementing PLS financing. The test results show a role for negative moderation of inefficiency regarding the influence of intention in implementing PLS financing. The coefficient of interaction with a negative sign means that the higher the value of inefficiency, the weaker the influence of intention of the behavior of implementing PLS financing will be. The results of the study provide a clear picture that trust in the virtue of financing using PLS contracts does not lead to an actual behavior in implementing profit-sharing financing. Factors that reduce the behavior of implementing profit-sharing financing still exist. This is in line with several studies which have shown that moral hazard, fraud and other activities are still looming in PLS financing in Indonesia (Ascarya 2009, 88). There is a big concern that partners play the profitability of the project (Fakir et al. 2020, 285).
This research contributes to the development of the Theory of Planned Behavior in PLS financing behavior. The influence of intention regarding the behavior of implementing PLS financing depends on the nature of the fund. These results prove that agency theory assumes that humans behave opportunistically in a contract between the principal and agent. For example, the production sharing agreement is a contract between s{ âhib al-mâl and mud{ ârib. Mud{ ârib is assumed to be always opportunistic so that Islamic banking bankers consider the PLS agreement to be inefficient with indicators: (1) Financing with a PLS system contains uncertainty of income compared to other types of contracts (such as mud{ arabah and ijârah), (2) Financing with a PLS system is very risky because the business losses the partners that can be borne by the Islamic banking, (3) Financing using the PLS system is very complicated and time-consuming because it requires partners' individual supervision, (4) Islamic banks do not have the technology or mechanism that can find out the number of partners' business results quickly and efficiently for the basis of PLS, and (5) Islamic banks do not have the technology or mechanism that can find out quickly whether partners commit fraud in the calculation of PLS or not.

Conclusions
The results showed that modifications to the theory of planned behavior can explain the behavior of Islamic bankers in implementing PLS financing contracts. The trust variable has an effect on the attitude of carrying out financing with profit sharing contracts. Next, attitudes and subjective norms affect the intention to carry out financing with profit sharing contracts. The surprising finding is that behavioral control has no effect on intentions and behavior in implementing PLS financing contracts. Moreover, intentions affect the behavior of using profit sharing contracts. Meanwhile, perception of inefficiency has a negative effect on the relationship between intention and behavior using profit sharing contracts.