COMPLIANCE BEHAVIOR MODEL OF PAYING ZAKAT ON INCOME THROUGH ZAKAT MANAGEMENT ORGANIZATIONS

This research intended to determine the factors encouraging muzakki to pay their zakat on income through Zakat Management Organizations (ZMO). The responses of 251 respondents dominated by Semarang city residents were analyzed. The result showed that attitudes of muzakki, subjective norms, religiosity, and intention had a significant impact on zakat compliance behavior in contrast with the perceived behavioral control variable. On the other hand, those four independent variables affected significantly on intention. It can be concluded that intention can be used as intervening, especially for the perceived behavioral control which has a perfect indirect correlation to behavior. The contribution of those independent variables toward zakat compliance behavior was 64.8% (R-square was 0.648), while the rest was influenced by other variables that were not analyzed in this research. ZMO has improved its services by simplifying the method of zakat payment. Nevertheless, they still need to give the added value through the good of amil governance and provide the different value from the professional and integrated distribution of zakat funds. Still, the education and socialization towards muzakki are necessary. This research also shows there can be an intervention of the muzakkis’ institutions/workplace which can be analyzed further.


INTRODUCTION
The provisions of zakat have been set in the Qur'an among others in Surah At Taubah 103 and Al Baqarah 267 which explain the importance of zakat and zakat position as one of the pillars of Islam. When referring to the history of Islam in the days of the Prophet (PBUH) and his companions, we can see that if there are Muslims who have been obliged to pay zakat but they are not willing to pay, the Prophet will fight them (Narrated by Bukhari and Muslim in (Kurnia & Hidayat, 2008)). The prophet (PBUH) also has provided examples of special officers to regulate the problem of zakat called as amil zakat (Hafidhuddin, 2006: 125).
In Indonesia, the study of zakat continues to be carried out from the discussion of the object of property which must be paid as zakat with the good fund management method which is stipulated in Law No. 23 of 2011 on the Management of Zakat. However, the management of zakat funds in Indonesia has not performed optimally (Ibrahim, 2011;Wahid, et. al., 2014). This situation was shown by the potential of accumulating national zakat funds in 2015 amounting to Rp 286 trillion, but the national Zakat, Infaq, Sodaqoh (ZIS) accumulated by the Zakat Management Organization (ZMO) in 2015 reached Rp 3.17 Trillion or 1.3 % of potential (BAZNAS, 2017). It indicates a wide open gap between the potential of zakat accumulation in Indonesia and the realization of ZIS funds that has been collected by ZMO.
Some researchers have conducted various researches on zakat especially in Malaysia, ie the research on the compliance of paying zakat gold with respondents of staff and teachers at the Universiti Utara Malaysia (UUM) (Sukri, Wahab, & Hamed, 2016). The result of his research reveals that attitude is a factor that can affect the behavior. However, the other states otherwise by proving that attitude has no significant effect although their target respondents are the same (Azman & Bidin, 2015). On the other side, the religiosity of a person does not become a significant factor in the behavior of paying zakat. This causes confusion because zakat itself basically is the proof of religious obedience/religiosity (Azman & Bidin, 2015;Sukri et al., 2016). Their conclusion is opposed by many researchers who proves that religiosity has an effect on the behavior of paying zakat (Mastura & Zainol, 2015;Sobana, Husaeni, Jamil, & Saepudin, 2016). Those contradictory results indicate that there is still a research gap.
The results of previous studies that contradicted each other and the magnitude of the gap between the potential of zakat Indonesia and the collection of national zakat could indicate the lack of compliance of the Muslim community on the behavior of paying zakat through amil zakat. Based on this condition, the authors intend to conduct research related to the Compliance Behavior Model of Paying Zakat on Income through Zakat Management Organizations. Expectantly, the formulated model can explain the behavior of community compliance (muzakki) in paying zakat through the Zakat Management Organization (ZMO), i.e. by making intention as an intermediate attitude, subjective norms, perceived behavioral control, and religiosity on compliance behaviors pay zakat income through the Zakat Management Organization (ZMO).

Definition of Zakat
Zakat is part of the property with certain requirements (nishab and haul) in which Allah SWT obligates the owner to hand over to the person whose right to receive it i.e. mustahiq (Hafidhuddin, 2006). Zakat embodies the vision of life in a society where the money is not a goal; it is a means for a higher purpose such as for the more evenly of wealth distribution (Al-Faizin, Insani, & Widiastuti, 2017;Belabes, 2017), and for the welfare of the people (Parisi, 2017). Thus, zakat is not just a gift from a wealthy person to a person who cannot afford, but the essence of zakat is the right of mustahiq which must be issued from the wealth of muzakki. It refers to Quran Surah at-Taubah (9: 103) and Narrated Sunnah of Bukhari and Muslim.

Zakat on Income
Contemporary Fiqh scholars review the objects of property that must be paid because of looking at the development of the current era in which in the various works or sources of income in the past time were not as complex as nowadays. One of the figures concerns to this situation is Sayyid Qutub (d. 1965 AD) in Hafidhuddin (2006: 94-95) who interprets Surah Al Baqarah (2: 267) that the essence belonging to the object of zakat includes all the results of human effort which are good and lawful. Both are given by Allah SWT from within and on the earth, and the provisions and levels as having been explained in the Sunnah of the Prophet that can be known by explicit explanation and by qiyas results. In 2011, Firdaus and Beik (2011: 23) analyzed the potential of zakat in Indonesia which reveals that the second largest zakat in Indonesia is from zakat income.
Indonesia has arranged zakat fund management regulation from the calculation, collection, distribution, and reporting as stated in Number 23 of The 2011 Law on Zakat Management. The calculation of zakat income is managed on the Regulation of the Minister of Religious Affairs (MRA) of the Republic of Indonesia Number 52 of 2014 on Terms and Procedures for Calculating Zakat Mal and Zakat Fitrah and the Utilization of Zakat for Productive Enterprises, Article 1 verse 17, and Chapter III paragraph 8 Article 26 and Article 27 concerning Zakat on Revenues and Services. Referring to MRA's Regulation, it is known that nishab of zakat income is worth 653 kg of grain or 524 kg of rice, with zakat content of 2.5%, while in article 27 it is reaffirmed that the zakat of income and services are made when the income and services are received and paid through amil zakat.

Compliance Behavior of Paying Zakat
Behavior in Ajzen's theory of the planned behavior which is a pre-planned response or activity tends to be carried out (Ajzen, 2005). This is in accordance with the behavior of paying zakat income in which actually implements the behavior of Muslim. People need to know the right way of their calculations that have been obliged to pay the zakat. The behavior that is referred in this study is more devoted to the behavior of a person in complying with his obligations as a muzakki, i.e. people who must pay zakat income, through the Zakat Management Organization (compliance behavior of zakat payment on income), or can be taken for the simple terms to be Zakat Compliance Behavior / ZCB. The construct of behavioral compiler refers to Sukri et al. (2016) and added with Sudirman's indicator (in Ridwan, 2016) emphasizing that not only zakat fitrah but also zakat income must be fulfilled.

Attitude of Muzakki
Referring to the explanation by Ajzen (2006) in, it can be said that muzakki's attitude is the perception, belief, or judgment of muzakki to the behavior in which it can be based on strong beliefs whether they are good or bad behavior to do (Ramdhani, 2011: 63). Solomon (2017: 298-299) explains that people who have an attitude with greater conviction tend to perform certain behaviors. In this case, it is the behavior of paying zakat income through the institution of zakat (Zakat Management Organization/ZMO).
H1: Muzakki's attitude has a significant effect on the behavior of obedience to pay zakat income through the Zakat Management Organization (ZMO). Ajzen (2006) explains that subjective norms are accepted social pressures to bind or not to bind individuals in certain behaviors. The subjective norms of Fishbein and Ajzen (2010: 131)

Perceived Behavioral Control
Perceived behavioral control refers to individual perceptions of his ability to implement a behavior (Ajzen, 2006). Based on this explanation, it can be said that the perceived behavioral control in this study is how much individual's confidence which are the existing factors can affect behavior in applying the behavior of paying zakat on income through the Zakat Management Organization? If muzakki does not feel capable of performing such behavior due to inadequate capacity (unable to utilize the provided facilities), muzakki will be very difficult to perform the behavior of paying zakat through ZMO.
H3: Perceived behavioral control has a significant effect on Behavior of Paying Zakat through the Zakat Management Organization (ZMO).

Religiosity
In the discussion of compliance to pay zakat, the factor of religiosity cannot be ignored because people who have a high level of religiosity should be obedient to pay zakat (Kamil, Zainol, & Ram Al Jaffri, 2012). This is confirmed by Sukri et al. (2016) that Islam is not only about religion but also the way of life. Therefore Muslims are expected to follow every command of Allah SWT including terms of paying zakat.
H4: Religiosity has a significant effect on Compliance Behavior Paying Zakat Income through the Zakat Management Organization (ZMO).

Intention
Many previous studies have concluded that intention is a very influential factor in predicting a behavior (Ajzen, 2006;Fishbein & Ajzen, 2010). Thus, it can be said that the expected individuals who have a positive intention toward certain behaviors will do the behavior. Ajzen (2005) explained in the theory of planned behavior that intention is an indication of a person's tendency to perform certain behaviors intentionally with a specific purpose.
The intention in the context of this research is the intention or inclination of a person to pay zakat income through ZMO. It is known that the zakat income has a certain calculation (nishab). So, paying zakat income also needs to begin with some preparation which indicates that muzakki intends to pay zakat income through ZMO. On the Islamic view, the intention is defined as the beginning or foundation of behavior. It is crucial to determine how the implementation of behavior both explicitly and inwardly (purpose) of muzakki. Therefore, Kurnia and Hidayat (2008) also reinforce the assumption of intention that can be a tendency to encourage someone to do a certain behavior. Ajzen (2006) in his Theory of Planned Behavior also explained that attitudes, subjective norms, and perceived behavioral control will be able to affect behavior with intentions as intermediaries or intervening.
Based on the previous exposure, it can be concluded that attitude variables, subjective norms, behavioral control, and religiosity can also affect intentions which are stated on the 5 th hypothesis until the 8 th hypothesis. The ninth hypothesis refers to the theory belong to Ajzen (2006) and Muhammad and Saad (2016) studed that intention may affect the application of certain behaviors.

RESEARCH METHOD
The type of this research was explanatory research with a quantitative approach. The target of respondents was Muzakki who paid zakat income through Amil Zakat / Zakat Management Organizations (ZMO) in Semarang city. Determination of sample using margin of error calculation with maximum significance was 0.1 (10%). The sampling technique was accidental sampling. Components of the questionnaire in this study were 47 closed statements and six open questions to represent six latent variables with Likertliked scale ranging from 1 to 7.
The obtained data from the questionnaire were analysed using Partial Least Square (PLS) analysis (Ghozali & Latan, 2015). The steps of testing empirical model of research were based on the PLS, namely: 1) construct of a causal link diagram of causality between constructs for both latent variables and between latent variables with indicators and estimate of the specific value of each variable, 2) validity test, 3) the outer model reading with convergent validity, discriminant validity, and composite reliability techniques, and 4) inner model reading.

RESULT AND DISCUSSION
In this study, there were 251 respondents who paid both through amil zakat of National Board of Zakat (BAZNAS) and zakat management unit which was an extension of the zakat management agency. The variation of respondents by demography can be seen in Table 4.1.
Based on the demographic data of respondents, it can be seen that female muzakkis were easier to decide to pay zakat through ZMO especially after various efforts conducted by amil zakat to socialize the institution, such as promoting zakat by the efficient transaction, the services given by ZMO, and the programs for muzakki. By age, it is known that the majority of muzakkis were millennial Indonesians on the productive age who always wants the ease of transactions for various purposes. ZMO officers also saw their opportunity to attract mustahiq to pay zakat through them. This phenomenon was reinforced by the experience of paying zakat of the majority which was less than one year. It also shows that the young generation still needs education and socialization related to zakat and management by ZMO. In fact, most of muzakkis had high intellectual level based on their last education level. Most of them were scholars as proven by Sobana et al. (2016) on the compliance of Muslim merchants for payment of zakat maal.

SHARE | Volume 8 | Number 1 | Jan -Jun 2019
Although the respondents were still young, they had high maturity that can be seen from the composition of the majority of respondents who were married. The high spirit of muzakki to pay zakat is also indicated by the consistency of muzakki for paying zakat although nishab of the several respondents did not reach the requirement for zakat income. It also happened for charity-giving by low-income households (Arsyianti, Kassim, & Adewale, 2017).
Based on the answers of the open questions, the result was that some respondents still cut their income personally to be distributed to other Amil Zakat Institutions that they trusted even though their income were deducted by the company as zakat payment through the payroll system. It can indicate the high public awareness to clean their property from the rights of others and their high willingness to help others.
The following table shows the matrix of indicators for each variable. This table is provided in conjunction with outer model analysis for testing the validity of the proxies. The table is named outer loading factor. Outer loading factor describes the factors that dominated the variables and the factors that could not represent the variable well. As the consequence, it should be taken out from the comprehensive model. Therefore, the good factors should not be less than 0.5 (Ghozali & Latan, 2015).
The attitude of muzakki showed the dominant factor was the possibility to do because of the easiness not the urgency level. Thus, the respondents still could pay their zakat through other ways such as giving it directly to the mustahiq. The existence of ZMO was still not in an urgent level. This temporary interpretation was supported by the dominant factors of perceived behavioral control. It was the capability of mustahiq itself to utilize the convenient facilities from ZMO. In addition, the facilities were parts of the society that existed in order to be accepted by the mustahiq by using it. In another word, it was part of society's wishes.

SHARE | Volume 8 | Number 1 | Jan -Jun 2019
The most factors that dominated religiosity were from practice and knowledge dimensions. As we know, zakat payment implementation is one of the practices that must be done by moslems. It is also written on five pillars of Islam. Moreover, the implementation of zakat indeed must be accompanied with the understanding of zakat both about the calculation and the principle of zakat itself.

Outer Model Results (Measurement Model)
Data analysis was begun by creating a path diagram. Furthermore, outer model analysis was done to test the validity and reliability. The results of outer model process are presented in Figure 1.
The Outer Model results show the consistency of the construct validity of the latent variable after being tested by looking at outer loading, discriminant validity-cross loading, and t statistics on outer model results, i.e. validity value was above 0.5 (Ghozali & Latan, 2015). The reliability can be seen in the results of composite reliability and Cronbach's alpha in Table 4.3. Based on Table 4.3, it is shown that the six (6) latent variables have the value of Cronbach's Alpha and Composite Reliability of more than 0.7. Thus, it can be concluded that the data reliability was achieved. Therefore, the data processing could be continued to the next stage of the inner model analysis based on the outer model or measurement model that gave good results.

Inner Model Results (Structural Model)
Inner model analysis was used to test the existing hypothesis or to test the correlation of latent variables which could be seen on the t-statistics or the level of significance of influence (P-value) which is presented in Table 4.4. Based on these results, it can be seen that only the perceptual relationship of behavioral control to the compliance of paying zakat whose hypothesis was rejected (H3 rejected) from the nine Hypotheses in this study while other eight hypotheses were accepted. They were the correlation of hypotheses of the attitude of muzakki, subjective norms, and religiosity toward zakat compliance behavior. The correlation of hypotheses of the attitude of muzakki, subjective norms, perceived behavioral control, and religiosity toward the intention.
Because of the regression coefficient of each variable, regression equation can be arranged as follows. In the data processing, the magnitude of the influence of independent variables towards the dependent variables can also be seen which also can be presented in Table 4.5. Based on Table 4.5., It can be seen that the contribution of the four independent variables to the intention was 72.7% which means 27.3%. Others were affected by other variables which were not examined in this study. Meanwhile, the contribution of the influence of the four independent variables along with the influence of the four independent variables through intention was as an intervening variable to the behavior of obedience paying zakat which was 64,8% and 35,2% . Others were affected by other variables which were not examined in this research.

Intervening Test Results
Intervening in this study was the variable of intention which was the intermediary from independent variable of attitude of muzakki, subjective norm, perceived behavioral control, and religiosity with the dependent variable of zakat compliance behavior. Based on the significance test that was analyzed in the inner-model sub-chapter, it is known that the intention had a significant effect on the behavior of compliance to pay zakat. So, the function of intention as a mediator could be tested further. The indirect effect test or the use of intermediate intentions from the four independent variables to the dependent variable can be seen in Table 4.6. Tests of indirect influence between attitudes of muzakki and compliance behavior of paying zakat through intention can be done by comparing the regression coefficients directly to indirect regression coefficient which is already presented in the original sample (O) column.

Intention towards Zakat Compliance Behavior
Based on the results of the hypothesis 9 test, it can be concluded that the intention has a significant effect on the behavior of muzakki to pay zakat through Zakat Management Organization. This finding is consistent with the theory (Ajzen, 2006). The muzakki plans to implement zakat paying behavior through ZMO and efforts to contact ZMO that are the dominant factors in reflecting on intention. Nevertheless, the average value of the two indicators is still lower than the normative reflection of intention. It shows that there is still a wide open effort to improve the intention that is by making muzakki start preparing a mature planning and communicating to muzakki.

Attitudes toward Behavior and Intention
Attitudes affect the behavior and intentions. These can be seen from hypothesis 1 and hypothesis 5 which were accepted. For the direct and indirect effects, attitudes have a partial indirect effect on behavior through intention. This finding supports the theory Ajzen (2006) and the results of previous research (Huda, Rini, Mardoni, & Putra, 2012;Sukri et al., 2016).
The positive attitude of muzakki to the intention and behavior of paying zakat through ZMO is dominated by the perception of muzakki who feels the behavior is very likely to be done because of its ease. If the ZMO wants to improve the positive perception of muzakki, it can improve its facility, synchronize it with technological progress (Sobana et al., 2016), and the provided convenience and service. As the result, muzakki feels pleasure for having transaction in ZMO. This is in line with the majority of the current population that was dominated by productive-aged society and millennial generation who requires the practicality of behaving with technological sophistication.
Nevertheless, these findings also show that the added value of paying zakat through ZMO is more emphasized on the provided facilities such as the usefulness of paying zakat through ZMO and the importance of paying zakat through ZMO that have not become a priority consideration of muzakki to decide to pay zakat through ZMO. It also shows that equitable distribution, optimization of zakat fund management, and integrity of ZMO, according to muzakki, have not become the added values as well as the different value of ZMO. Moreover, good management of zakat funds is a form of responsibility which is not only to be addressed to Muslims but also to Allah SWT, and it is the main purpose of amil zakat (Saad, Aziz, & Sawandi, 2014;Wahab, Zainol, & Bakar, 2017).

Subjective Norms toward Behavior and Intention
The subjective norm affects behavior and intention. It is known from hypothesis 2 and hypothesis 6 that were accepted. For the direct and indirect effects, subjective norms had a direct influence as good as indirect influence on behavior through intention. These findings support the theory (Ajzen, 2006), and the previous research results (Ahmad, Nor, & Daud, 2011;Azman & Bidin, 2015;Mastura & Zainol, 2015;Muhammad & Saad, 2016;Al-Mamun, A., & Haque, A. (2015).
This direct correlation between subjective norms on the behavior of obedience and paying zakat shows that the perception of the surrounding environment of muzakki has an impact on the final decision of muzakki to pay zakat through zakat management organization. The expectations of people in the muzakki environment are indicators that dominate subjective norms (Wahyuni, 2017;Zagrali, 2017).
When people have big expectations or big trends to the individuals, they can be forced by the individual. So, the individual feels the need to perform the behavior expected by the external parties in social life. The average expectation of people around that is low indicates the effect of subjective norms on zakat paying compliance behavior through ZMO that can be optimized by improving the expectations of people around muzakki. Nevertheless, it is also known that the environment around muzakki makes muzakki take decision to pay zakat to each individual in general because zakat behavior tends to be like a personal matter.
Based on this situation, ZMO can still optimize the influence of subjective norms by doing the socialization and cooperation to community, scholars, leaders or other groups to improve education and socialization about zakat. This cooperation must be done in collaborative strategy with the ZMO that becomes the agile agency to receive the zakat at the time while the scholars or leaders give their appeals to pay zakat (Ummulkhayr, Owoyemi, & Cusairi, 2017). In addition, ZMO can provide zakat ambassadors to educate the pilgrims or society at the beginning of the Islamic events.
Then, building a good opinion of ZMO in the wider community is also an important point that can be done by ZMO. So, people can have a positive thought on ZMO. The tendency of figures for showing their positive perceptions of ZMO can also improve the intention of the community to implement the behavior of paying zakat through amil zakat.
The regression coefficient of subjective norm on behavior with intention as the intermediate variable was positive but it was lower than the regression coefficient between subjective norms and behavior directly. Intentions that could not be good intermediate variables can be further analyzed based on respondents' demographics in this study.
Based on the description of some respondents, it is known that some muzakki paid zakat through salary facility that is one method to build the habit of paying zakat through amil zakat. The facility is cooperation between ZMO and good institutions where muzakki works. The facility can be utilized by muzakki upon the approval of the muzakki himself. In other words, the influence of the respondent's family, other external factors that can make muzakki perform zakat paying behavior through the Zakat Management Organization are from the agencies where they work.
The appeals or policies of the agency tend to improve the implementation of certain behaviors by individuals based on the norm factor although the individual has not had sufficient intention. It is expected that the appeal or suggestion from the institution and other external parties should not only be normative but also combined with the socialization and education of the institution to improve comprehension about zakat. So, the implementation of obligatory worship behavior which is part of the five pillars of Islam does not only become a form of routine activity of obligations but also a real worship.
On the other hand, the strong influences of the institutions shape behavior to the muzakkis for paying zakat through ZMO. One of them is caused by an agency intervention. This phenomenon happens many times in Indonesia. Several numbers of agencies that give appeals to its employees to pay zakat through the auto debit of salary facilities. The intervention of this party tends to be the influence of the amplifier and the weaker which in the theory of planned behavior can be referred to as actual behavioral control (Ajzen, 2006).

Perceived Behavioral Control towards Behavior and Intention
Perceived behavioral control has no effect on behavior but it affects intention. it is known from Hypothesis 3 that was rejected while Hypothesis 7 was accepted. For the direct-indirect effect, perceived behavioral control has a perfectly indirect effect on behavior through intention. These findings support the theory (Ajzen, 2006), and the previous research results (Bidin, Othman, & Azman, 2013;Huda et al., 2012).
The indirect influence between the perceived behavioral control and the behavior of obedience paying zakat through intentions as intermediate variables can be concluded that intention can be a good intermediate variable between the perceived behavioral controls with the behavior of obedience for paying zakat. The indirect influence of the perceived behavioral control on the behavior of obedience for paying zakat through intentions is a perfect indirect influence because the regression coefficient of perceived behavioral control toward behavior through intention is more than the absolute value of regression coefficient of perceived behavioral control directly. Direct influences between perceived behavioral control and behavior are insignificant. As if viewed from the direction of influence, the perceived behavioral controls that previously had a negative direction toward behavior become positive toward the behavior through intention.
The theory fully explains that the perceived behavioral control has an effect on behavior with intention as an intermediary (Ajzen, 2006). In the context of zakat paying behavior, it can be interpreted that muzakki will not implement the behavior without intention even though the average value of the ability of muzakki to implement high paying behavior of zakat is low. This is in accordance with the results of this study that the perceived behavioral control affected behavior through intention although the belief of muzakki states that paying zakat is a personal decision of each. Then, if it is not followed by the intention, the ability of muzakki will not be utilized by muzakki to implement the behavior of paying zakat through amil zakat.
High independent muzakki needs to be balanced with the ability to access various information and ease of zakat through ZMO to improve muzakki's intention in performing zakat paying behavior through ZMO. It is important to improve muzakki's intentions through perceived behavioral control. It is considered that the perceived behavioral control is the factor that has the greatest correlation to intention compared to the other three independent variables. In an effort to improve muzakki's intention to perform the behavior of paying zakat through amil zakat, the ZMO can socialize various facilities of ease of paying zakat through amil zakat and provide socialization to access various information expected by muzakki such as information of zakat collection, distribution of zakat fund, and audit result management of zakat fund management.
Based on the summary of open-ended questions related to the perceived behavioral control, it is known that out of 118 respondents who answered, 46 stated they have no obstacles in performing the behavior of paying zakat through amil zakat while the remaining 72 argued otherwise.
A small number of respondents stated that they have obstacles coming from each individual, such as intention, awareness, and time of muzakki to pay zakat through amil zakat. However, 10 out of 12 people who thought they enjoy the benefits of ease of paying zakat through amil zakat such as; digital facilities, especially zakat pickup facilities (to anticipate their dereliction and improve bonding and closeness between amil and muzakki). Other two (2) along with other 60 respondents still felt that they have obstacles derived from the amil zakat such as the lack of socialization of the zakat management organization for the introduction of the functions and benefits of their existence, the lack of educational information about the facilities, the ease with which they can use, the lack of quality of services to enhance the closeness or friendship with muzakki, and the lack of information related to the transparency of reports on the fundraising and distribution of funds and other information that can demonstrate the quality of amil zakat. That is also has been proven by the external factors of muzakki such as the transparency of the ZMO itself that affects the tendency to compliance the zakat payment (Sobana et al., 2016).
Obstacles that tend to be external factors of muzakki that are not immediately addressed will make muzakki cancel to pay zakat of their income through amil zakat. Meanwhile, if amil zakat is more intense in conducting socialization, education, and optimization of good management as described in the previous paragraph, it can improve the trust of muzakki with Zakat Management Organization. Consequently, the tendency of muzakki to pay zakat income through Zakat Management Organization is greater (Mustafa, Mohamad, & Adnan, 2013).
On the other hand, there were other variables that were not included in this study, but they had an influence on the perceived behavioral control and could be a moderator of behavior, i.e. the actual condition variable that can control the behavior or the real state of existence if the perceived behavioral control is a measure of capacity within the individual, the actual conditions that exist by looking at the real conditions in the field, including the existence of good interventions that can strengthen or weaken the implementation of zakat paying behavior through ZMO.
Other external factor that can make muzakki cancel to pay through ZMO is the other distribution option that has become a habit in the general public, i.e. paying zakat directly to the recipient (mustahiq). External factor can refer to actual behavioral control or existing factual conditions. It can be a factor in the imposition of zakat paying behavior through amil zakat due to the flexibility of muzakki to pay his zakat whether directly or through ZMO. On this matter, ZMO needs to increase the value of difference, especially related to the management and distribution of zakat funds.

Religiosity towards Behavior and Intention
Religiosity has an effect on behavior and intention. It is known from hypothesis 4 and hypothesis 8 which were accepted. For the direct and indirect effects, attitudes have a direct effect as well as indirect effect on behavior through intention. These findings support the results of research from (Mastura & Zainol, 2015;Mukhlis & Beik, 2013).
Paying zakat is the third pillar of the five pillars of Islam which makes zakat as a reflection of one's faith. Thus, the more urgent (obligatory) zakat status when compared to other sunnah worship, the more obedient performance or the level of religiosity of a Muslim. It can affect the behavior of zakat compliance.
Basically, the behavior of paying zakat is mandatory for Muslims. It is one of the Pillars of Islam. Though muzakki does not have a high enough intention to pay zakat, this condition does not reduce the obligation of muzakki to pay his zakat when it has reached zakat terms. Muslims who have a high religiosity will perform the obligation to pay zakat because of Allah SWT. With good religiosity (especially good religious scholarship), muzakki will naturally intend to go to the Zakat Management Organization and fulfill his obligations. Moreover, the common Muslims in Indonesia, especially in the city of Semarang have a tendency to comply with the provisions of Allah SWT and zakat is an absolute obligation of Muslims.
Constituent constructs of religiosity show the dimension of knowledge is the most dominating level of religiosity of a person. Understanding of fasting is a dominant reflection of religiosity of the respondent. In other words, the level of one's faith will be at a high level when it comes to fasting or approaching Ramadhan because when religious activities and religious atmosphere improve they make them easier and conscious to pay zakat (Arsyianti et al., 2017). With that momentum, socialization and education about fiqih of zakat can be done more intensively during Ramadhan month which can be followed by the emphasis of fiqih about zakat fitrah, zakat maal, along with zakat profession which refers to contemporary zakat fiqh.
Then, the knowledge that dominates is the understanding of muzakki related to the nature and benefits of paying zakat for himself, mustahiq, and Islam along with the understanding related to sharia law of zakat. Thus, it becomes very important for amil zakat to educate the Muslim community related to fiqh of zakat and fiqh of contemporary zakat. Characteristics of the majority of respondents who have experience of paying zakat less than a year, and still young and productive, strengthen the need for comprehensive education related to the benefits of zakat for Muslims, along with the development of zakat that is up to date.

Conclusion
The problem in this study started with the existence of a gap phenomenon of the potential for zakat collection in Indonesia. Then, an analysis of the compliance behavior of the community to pay zakat on income as the second largest potential of zakat in Indonesia was done. The results showed that the zakat compliance behavior was mostly affected by the attitude of muzakki on zakat payments on income through ZMO. Then, by the high intention to make these payments, the level of faith of muzakki (religiosity), and the subjective norms were implied the impulse from the surrounding environment. The perceived behavioral control in the form of the ability to take advantage of the ease of transactions could only improve the conduct of behavior only if it had the intention to make the zakat payment. Thus, increasing the real funds of zakat was be done by taking into account the attitudes of the muzakki (the usefulness and practicality of zakat payment through ZMO), perceptions of the environment (supporting environment), and the level of religiosity of the muzakki (understanding of zakat).
There was a significant effect on intention towards zakat compliance behavior indicating that the Planned Behavior Theory by Ajzen can also be applied in compliance behavior of muzakki to pay zakat through Zakat Management Organization. Because of applying the behavior of paying zakat, muzakki should calculate what amount of zakat they must pay (behavior that must be planned in advance).
The religiosity factor had a direct effect on behavior and did not require intention as a mediator toward compliance behavior. This showed the firmness of zakat as part of the reflection of the Muslim's religiosity. Therefore, the muzakki does not have to wait for an intention to implement this behavior because zakat is mandatory for muzakki that must be fulfilled.
However, the attitude of muzakki towards the behavior of paying zakat through ZMO was caused by the ease of transaction as well as only motivated by environmental pressure or their superior (subjective norms). Those can be interpreted that the compliance behavior is done only for a formality while the awareness of the impact on the management of integrated and professional zakat funds by ZMO has not been noticed by them.
Thus, besides the ZMO still have to maintain their professionalism because of their responsibility towards Allah SWT, muzakki also needs to be given comprehensive education (Islamic financial literacy); about the zakat fiqh, the management of zakat during the Prophet's and companions' era, the impact of zakat and the impact of zakat management professionally, as well as contemporary zakat.
These efforts can be done by having collaboration along with; 1) social environments such as institutions, companies, and mosques as providers of place for study events/assemblies; 2) figures such as scholars, academics/science experts, and leaders, as parties who can encourage the society to pay zakat, and educate the about zakat neutrally; 3) as well as with ZMO itself who should become the agile agency to receive the zakat at the time (on the assemblies). The strategy also needs to be intensified more during the month of Ramadan where it is the month of blessing for Muslims and generally they will increasingly compete to improve their worship.
Through the comprehensive zakat education, it is expected to instill the understanding that zakat is the obligation of Muslims and the concept of zakat which can have a tremendous impact on the benefit of the people at large. So, the strong intention to pay zakat wisely will improve. By building good opinions about ZMO (good amil governance), the public can have a positive image of ZMO. A good environment for other Muslims can be achieved (other muzakki) (word of mouth strategies). Therefore, the amil zakat will be easier to socialize the digitalization of services of ZMO such as the digitalization to pay zakat, consultation services, and the publication of the report on collection and distribution of zakat funds.
At last, in-depth literature studies show a large gap between national asset and the existing potential that is not only due to the behavior of muzakki to pay zakat through amil zakat, but also in terms of reporting zakat. There are still many amils of zakat which have not yet been parts of the national reports, especially amil zakat who has not been official yet. In the data, there is a possibility of the gap that there have already been many people who become aware and pay zakat through amil zakat in fact (including amil zakat that has not been official). Moreover, zakat fund disbursement that is carried out directly by muzakki has not been taken into account in the preparation of BAZNAS outlook. This is emphasized in the 2012 Indonesia zakat outlook which emphasizes that the collection report in the outlook is based on official BAZ and LAZ data only.

Limitations of Research and Future Research Plan
The phenomenon in the field indicates the existence of intervention from the agency or office by some muzakki in performing the behavior of paying zakat through amil zakat by a way of the paycheck. Thus, the next study hopefully will cover; 1) to add these factors as moderating factors, 2) to make comparative analysis of muzakki whose payment of zakat is through salary and muzakki whose zakat payment is done independently, 3) Or to do research by focusing on one type of muzakki, especially muzakki who pay zakat through amil zakat on his own will.