Information Asymmetry, Group Cohesiveness, and Budgetary Slack: A Moderating Effect of Budgetary Participation

Objective – The purpose of this study is to examine whether information asymmetry and group cohesiveness affect budgetary slack with budgetary participation as a moderating variable. Design/methodology – This research was conducted in a public hospital in Banda Aceh, Indonesia. This study uses primary data in the form of questionnaire which were filled by 38 respondents who were involved in the hospital budgeting process. The samples are selected through simple random sampling methods. The Moderated Regression Analysis (MRA) was utilized to analyze the data. Results – The result of this study shows that partially information asymmetry and group cohesiveness influence budgetary slack at a public hospital in Banda Aceh, Indonesia. By using moderated regression analysis with interaction test, the budgetary participation is able to moderate information asymmetry and group cohesiveness towards budgetary slack at a public hospital in Banda Aceh.


Introduction
A hospital is a health service institution that conducts wide-ranging individual health services that provide inpatient, outpatient and emergency services. Public hospitals are hospitals managed by the government, regional government and non-profit legal entities based on Indonesian Government Law No. 4, (2009) concerning hospitals which explains that hospitals have several functions, namely: (1) the delivery of medical treatment and health recovery services in accordance with hospital service standards, (2) maintenance and improvement of individual health through complete second and third level health services according to medical needs, (3) organization of education and training of human resources in the context of increasing the ability to provide health services, (4) organizing research and development and screening of health technology in the framework of improving health services by taking into account the ethics of health science.
The regional general hospital in Banda Aceh city is has been transformed into a regional public service agency (BLUD). The regional public service agency is an agency within the government established to provide services to the community in the form of supplying goods and /or services without prioritizing profit in carrying out its activities based on the principles of efficiency and productivity. The regional public service agency is also part of the regional government apparatus, with the legal status not separate from the regional government. In managing its finances, the BLUD uses the financial management pattern of the public service agency or PPK-BLU which is a financial management pattern that provides flexibility in the form of flexibility to implement sound business practices in improving services to the public in order to advance the general welfare and improve the life of the nation. The privilege given to the BLUD is Jalaluddin*, Juanda, Lia Utami JAROE VOL. 3(1) due to special demands, namely to improve the quality of service from the BLUD. The business plan budget (RBA) is an embodiment of the preparation of the budget prepared by the BLUD for use by the BLUD itself.
The budget is important for an entity; the key component of planning is the budget, namely financial plans for the future that identify the actions needed to achieve them. Every for-profit and non-profit entity can benefit from the planning and control given by the budget (Hansen & Mowen, 2009). Budgeting in the public sector is a political process. In this case, the budget is an instrument of accountability in the management of public funds and the implementation of programs financed with public funds (Mardiasmo, 2004:61). In short, it can be said that the public budget describes the financial condition of a public organization which includes information on the expenditure budget, revenue, and carried out activities. The budget must be made in line with the general objectives and mutually support the objectives of each division.
The budget contains actions or steps that will be taken to realize these objectives. The budget also functions as a planning and control tool so that managers can carry out organizational activities more effectively and efficiently. The budget as a planning tool plays a role in planning, financing and income at an accountability center that will be achieved by local governments within a certain period of time by carrying out various activities that have been previously determined. As a control tool, the budget plays a role in evaluating the performance of managers by looking at the extent to which managers can achieve the targets set in the budget.
Budgeting and budget execution involve many parties, ranging from middle level management to lower level management or can be referred to as principals and agents. The problem with middle and lower level management involvement in the budget preparation process is the emergence of budgetary slack. Budgetary slack can also be interpreted as the difference between the reported budget and the budget that is in accordance with the best estimate for the company, which is a lower revenue budget and budgeting for expenditure that is higher than the actual estimate (Sugiwardani, 2012).
This can be seen from one phenomenon that occurred at Waled Hospital in Cirebon, where the phenomenon is related to the alleged corruption of the misuse of the 2012 Jamkesmas budget which is used for personal interests, which should be intended for public services. In addition, there were also irregularities in the salary distribution system for employees at Waled District Hospital that were very different from those in other agencies. This is evidenced by the fact that the salary of a hospital director is around Rp. 100 million per month, thus making the local government (Pemda) must review the payroll system at Waled District Hospital. However, after being traced back, it was reported earlier that there were allegations of corruption carried out by a number of staff of the Waled District Hospital regarding the misuse of Jamkesmas in 2012 (Jabarpublisher.com, 2017).
Budgetary slack can be avoided by the establishment of a budgeting system that is expected to improve the performance of managers and subordinates, namely participatory budgeting. Brownell (1982) \ defines budgetary participation as a process whereby the individuals in it are involved and have an influence on the preparation of budget targets whose performance will be evaluated and may be valued for the achievement of their budget targets. Budgeting participation also adds information that can reduce the ambiguity of roles that support performance improvement. When subordinates have better information than their superiors, participatory management control systems allow subordinates to disclose personal information that can be included in the budget when their performance is assessed (Baiman and Evans, 1998).
From this description, budgeting participation can play a role as a moderating variable that has a good influence to strengthen or even weaken each independent variable and the dependent variable. The strengths or weaknesses of the effect depend on how many lower-level managers are involved and will participate in the budgeting process. If there is involvement of lower-level managers in budgeting, then the budget goals will tend to be personal goals of managers that will result in greater conformity of objectives, and vice versa (Hansen & Mowen, 2009:448).
In connection with the study background, this study uses independent variables consisting of information asymmetry and group cohesiveness. In addition, to distinguish this study from previous research, the author uses one additional variable, namely budgetary participation. These variables function as moderating variables and cannot be directly observed but instead measurement proxies are required that serve as indicators in the study. Therefore, researchers have an interest in conducting research with the title "Effect of Information Asymmetry and Group Cohesiveness on Budgetary Slack with Budgetary participation as Moderating Variable".

Information Asymmetry
In economics, there is a condition called information asymmetry or information imbalance. Anthoni & Govindarajan (2001) state that information asymmetry arises in agency theory, where the principal gives authority to subordinates (agents) to manage the company they own. Because of the delegation of authority and the separation of duties from the principal to the agent, the supervisor cannot always know the actual activities carried out by his subordinates due to limited information about the factual circumstances of the unit or center of responsibility that is managed by subordinates. This condition then causes a phenomenon called information asymmetry.
Information asymmetry according to Dunk & Nouri (1993) occurs because there are agents who have more information than other parties do (principal). From this explanation it can be concluded that information asymmetry is an imbalance of information held by subordinates and superiors about a unit of responsibility in an organization.

Group Cohesiveness
Both formal and informal groups can have closeness or similarity in attitudes, behaviour, and achievement. This closeness is referred to as "group cohesiveness" which is generally associated with the encouragement of members to stay together in the group rather than the urge to force members out of the group (Gibson & Donnely, 1993). Furthermore, Robbins (2002) defines group cohesiveness as a level that describes the members attracted to each other and motivated to remain in the group. The concept of cohesiveness is important for the understanding of organizational groups. The degree of cohesiveness can have positive or negative effects depending on how well the group's goals are consistent with formal organizational goals.
If cohesiveness is high and the group accepts and agrees with the formal goals of the organization, then group behavior will be positive in terms of the formal organization. But if the group is very cohesive but the goals are not in line with formal organization, then the group's behavior will be negative in terms of formal organization.

Budget
Budget is a statement of the estimated performance to be achieved over a certain period of time expressed in financial measures (Mardiasmo, 2005:61). According to Falikhatun (2007), budget is a management tool in allocating the limited natural resources and sources of funds owned by the organization to achieve goals. Kenis (1979) states that budget is not only a financial plan regarding costs and revenues in a center of responsibility, but can also function as a means of control, coordination, communication, performance evaluation, and motivation.
Based on the above explanation, it can be concluded that budget is a planning and control within the company in detail and systematically can be used by line management to manage the organization in the future, especially one year.
According to (Bastian, 2001:80) budget functions include: JAROE VOL. 3(1) 1) Budget is the final result of the work plan preparation process.
2) Budget is a blueprint for activities to be carried out in the future.
3) Budget as an internal communication tool that connects various work units and work mechanisms between superiors and subordinates. 4) Budget as a tool to control work units. 5) Budget as a tool for motivation and persuasion of effective and efficient actions in achieving the organization's vision. 6) Budget is a political instrument. 7) Budget is an instrument of fiscal policy. Milani (1975) explains that budgetary participation in a company is the level of involvement and influence of individuals in the preparation of the budget. For more detail on involvement and influence in the preparation of the budget argued by Milani (1975), namely:

Budgetary Participation
1) How far the budget is affected by the involvement of managers.
2)Reasons for superiors at the time of the budget revision process.
3) Frequency of expressing initiatives, giving proposals and or opinions about the budget to superiors without being asked. 4)How far managers feel they have influence in the final budget. 5) The interests of managers in their contribution to the budget. 6)Budget frequency is discussed by superiors when the budget is prepared. Based on the above explanation, it can be concluded that budgetary participation is the participation of managers and subordinates communicatively in the budget preparation process, where the information needed by managers can be given by subordinates in actual, so that managers can make good decisions in a budget without prioritizing the interests of managers only but also the interests of subordinates and the company as a whole.

Budgetary Slack
One problem in budgetary participation is budgetary slack (Hansen & Mowen, 2004:377), which is closing the budget when a manager deliberately estimates low income or raises costs. Managers do this so that budget targets can be achieved, so that the manager's performance looks good. Budgetary slack is the difference between the stated budget and the best budget estimate that can be honestly predicted (Veronika & Krisnadewi, 2008). In a state of budgetary slack, subordinates tend to advance the budget by lowering revenues and increasing costs compared to the best estimates proposed, so the target will be easy to achieve. Budgetary slack can occur in the condition of information asymmetry due to the existence of biased information between superiors and subordinates that can make budget implementers to carry out budgetary slack.

Effect of Information Asymmetry on Budgetary Slack at General Hospitals in Banda Aceh
Dunk & Nouri (1993) defines information asymmetry as a condition where the information possessed by subordinates exceeds the information possessed by their superiors, including both local and personal information. Dunk & Nouri (1993) examined the effect of information asymmetry on the relationship between budgetary participation and budgetary slack. Anthoni & Govindarajan (2001) states that the condition of information arises in agency theory, where the principal (owner/supervisor) authorizes the agent (manager/subordinate) to regulate the activities of the company they own. Information asymmetry is a condition where the owner/supervisor does not have enough information about the performance of the agent/ subordinate so that the supervisor cannot determine the subordinate's contribution to the company's actual results (Falikhatun, 43   2007). Information asymmetry arises because managers get more information from within the company.

The Effect of Group Cohesiveness on Budgetary Slack at General Hospitals in Banda Aceh
Group cohesiveness is a formal and informal group that has closeness or similarity in attitude, behaviour, and achievement. Groups with a strong unity attitude are generally more effective in decision-making situations compared to groups where there is a lot of internal conflict and a lack of a spirit of cooperation among their members. Cohesiveness is very important for the understanding of organizational groups, where the level of cohesiveness can have both positive and negative effects depending on how well the group's goals are in line with formal organizational goals. If cohesiveness is high and the group accepts and agrees with the formal goals of an organization, then the group's behaviour will be positive in terms of the formal organization, but if the group is very cohesive and the goal is not in line with the formal organization, then the group's behaviour will also be negative in terms of the formal organization (Falikhatun, 2007).

Effect of Budgetary Participation on Information Asymmetry with Budgetary Slack in General Hospitals in Banda Aceh
Participation means taking part. Participation is also interpreted as a symptom of democracy, where people are involved in planning and implementation and also share responsibility according to their level of maturity and level of obligation. Participation is a process that involves several parties in setting common goals which will have an influence on the parties involved in the future.
The definition of participation in budgeting is explained in more details by (Ompusunggu & Bawono, 2006) as a collaborative budgeting process in decision making, involving two or more groups that have an influence on decision making in the future.

Effect of Budgetary Participation on Group Cohesiveness with Budgetary Slack in General Hospitals in Banda Aceh
One other behaviour related to organisation that can influence participation in budgeting is group cohesiveness (Utami, 2012). Group cohesiveness is a situation where one employee have ties with other employees who are members of a group and motivate each other to remain members of the group and jointly achieve the goals they want (Widiya & Luh, 2006). In budgeting participation, if the goals of the group which has high cohesiveness are not in accordance with the goals of the organization's management, then it can cause budgetary slack.
Based on the explanation above, the framework of this study can be seen in the figure below.

Research design
The type of research investigation in this study is causality study. This study uses primary data in the form of a questionnaire. The time horizon used in this study is cross sectional.
The population in this study is general hospitals in Banda Aceh. The sample was selected using the simple random sampling method. The number of samples in the study was 38 employees of 72 employees who participated in the preparation of the budget at the general hospitals in the city of Banda Aceh, to determine the minimum sample size based on Slovin (Umar, 2005:78).

Operational Variables Budgetary slack (Y)
Dependent variable consists of budgetary slack (Y), which is the difference in the amount of budget prepared by the responsibility center manager with the company's best estimate. Slack was measured using an instrument developed by Dunk & Nouri (1993) consisting of 4 questions.

Information Asymmetry (X1)
Information asymmetry shows the difference in information possessed by superiors and subordinates in an organization. Information asymmetry is measured using an instrument developed by Dunk & Nouri (1993), consisting of 5 questions. Low scale shows low information asymmetry, and high scale shows high information asymmetry.

Group Cohesiveness (X2)
Group Cohesiveness shows the strength of members in the group and their commitment in the group. Group cohesiveness is measured using an instrument developed from Gibson & Donnely (1993) theory consisting of four questions. A high scale indicates a high level of cohesiveness and a low scale shows a low cohesiveness.

Budgetary Participation (X3)
Budgetary participation is the involvement of managers in the budget preparation process. Participation in this study was measured using an instrument developed by Milani (1975), consisting of six questions. The use of each independent and dependent variable uses instruments developed by Dunk & Nouri (1993), Milani (1975), and Gibson & Donnely (1993), using a 5-point Likert scale, with the criteria: "Strongly Agree", "Agree", " Neutral "," Disagree "," Strongly Disagree ". Sub-variables that become measurement items will be measured by questions for each independent and dependent variable, with scores as follows:

Data Collection Technique
Primary data were collected through distributing questionnaires. Each questionnaire consisted of 1 (one) dependent variable, 2 (two) independent variables and 1 (one) moderation variable.

Research Model
The analytical tool used in this study is a moderated regression analysis to test hypotheses. The reason for using the analysis is that there is a moderating variable so that an interaction test between the independent and moderation variables is needed. So the regression model is as follows: Data analysis was performed using the SPSS program version 21.0, with a significance level of 5%.

Data Description and Analysis
According to Ghozali (2009), descriptive statistics provide a description of the data seen from the mean (standard), standard deviation, variance, maximum and minimum data that describe the profile of research data. Description of the data is done to see how the data in the study are described in descriptive statistics. Descriptive analysis was conducted in this study to find a picture of the characteristics of the variables studied.
Based on the number of questionnaires distributed to respondents (38 questionnaires), there were 37 or as many as 97.36% returned questionnaires and as many as 1 or 2.64% questionnaires that did not return. From the results of returned questionnaire that was filled out by the respondent, the final sample size for this study was 37 questionnaires (n = 37).

Data Quality Test Validity Test Results
Validity test is done by calculating the correlation between the values obtained from questions with the overall total value of these questions, this method is often called the Pearson Correlation (Ghozali, 2016:52). Significance test is done by comparing r-count with r-table for degree of freedom (df) = n-2 with alpha (α) 0.05. In this case, n is the number of samples (Ghozali, 2016:53). The following are the results of the validity test analysis for each variable as follows: Based on table 1, it can be seen that the r-count value of all question item is greater than the r-table value at (df) = n-2 or (df) = 37-2 and the alpha value (α) 5% or 0, 05 which is 0.3202. Thus, all questions in the information asymmetry variable can be declared valid.   Table  5 explains that the value of Cronbach's alpha, variables X1, X2, moderator and Y are greater than the minimum value of Cronbach's alpha 0.6. Therefore it can be concluded that the research instrument used to measure the variables X1, X2, moderator and Y to be reliable.  Table 7 shows that the tolerance value of variables X1, X2, and moderator is greater than 0.10 while the VIF value of variables X1, X2 and moderator is smaller than 10.00 so it can be concluded that multicollinearity did not occur between independent and moderator variables in the regression model used in this study. Heteroscedasticity testing is performed using the Glejser Test, the results can be shown in table 8 where the significance value of X1, X2 and moderator is known to be greater than 0.05, so Ho cannot be rejected and it can be concluded that heteroscedasticity did not occur.

Results of Moderation Regression Analysis
In this study, MRA analysis was performed using the multiplication interaction test of two or more independent variables. The analysis is used to obtain the regression coefficient which will determine whether the hypotheses will be accepted or rejected with a significance level of 5% (0.05). The following is a moderation regression test: Model 1 Y = 16,539 -0,115X_1 + 0,315X_2 + e (1) Model 2 Y = 18,934-0,497X_1+0,111X_2+0,018X_1.Moderator+e Moderator 0.809 Y 0.737 Table 5. Reliability Test Results Table 6. Normality Test Results Table 7. Multicollinearity Test Results

Simultaneous Test Results
Model P-value 1 0.159 * α = 0.05 Table 9 shows that the P-Value > α = 0.159 > 0.05 so that H0 cannot be rejected and it can be concluded that simultaneously the information asymmetry and group cohesiveness do not affect the budgetary slack at public hospitals in Banda Aceh.

Partial Test Results
Partial test results indicate that the information asymmetry variable (X1) is smaller than α (0.05) so that H0 is rejected and it can be concluded that the information asymmetry variable (X1) affects the interest of budgetary slack (Y). The group cohesiveness (X2) variable is greater than α (0.05) so H0 is accepted and it can be concluded that the group cohesiveness (X2) variable does not affect the budgetary slack while the information asymmetry variable (X1) with budgetary participation variable (moderator) is smaller than α (0.05) so H0 is rejected. It can be concluded that the information asymmetry variable (X1) with budgetary participation variable (moderator) influences the interest of budgetary slack (Y). It can be seen in table 10 that the R square value in the first regression equation is 0.356 so it can be said that the budgetary participation variable moderates the information asymmetry and group cohesiveness effect on the budgetary slack by 35.6%.

Effect of Information Asymmetry on Budgetary Slack in General Hospitals in Banda Aceh
The results showed that the value of P-Value> α = 0.159> 0.05 so that H0 cannot be rejected and it can be concluded that the information asymmetry variable simultaneously did not affect the budgetary slack at general hospitals in Banda Aceh. This means that in preparing a budget, it is necessary to have responsible communication within it, namely internal information. The information possessed by subordinates must be conveyed to superiors, in this case superiors still have difficulty understanding all incoming information, especially on matters relating to the technical field that are better understood by subordinates. Subordinate's assistance is needed in processing the incoming information in accordance with the field of these subordinates. However, subordinates hide the personal information they have and only give distortion to superiors, resulting in subordinates can easily make a gap to the budget made with the information they have only for personal gain. Thus, the information asymmetry owned by someone can influence their desire to create a budgetary slack.
The results of this study are not in line with research conducted by (Meirina & Afdalludin, 2018). Meirina & Afdalludin (2018) found that information asymmetry has a significant positive effect on budgetary slack. And it means that if the information asymmetry increases, the budgetary slack will also increase, and vice versa. This indicates that the budgetary slack will be greater in the condition of information asymmetry, because it can encourage subordinates/executors of the budget to make budget slack.

Effect of Group Cohesiveness on Budgetary Slack in the General Hospital in Banda Aceh
Based on the results of statistical analysis in this study, it was found that the null hypothesis (H0) cannot be rejected and it can be concluded that group cohesiveness does not affect budgetary slack. The results of this study are not consistent with the research of Falikhatun (2007) which shows that there is a significant influence on the relationship between budgetary participation and budgetary slack. This can occur because individuals in an organization does not need to form a group called group cohesiveness to be able to create budgetary slack.

Effect of Budgetary participation on Information Asymmetry with Budgetary Slack in General Hospitals in Banda Aceh
The results of the study showed that information asymmetry with budgetary slack can be moderated by budgetary participation; therefore the third hypothesis (H3) in this study is rejected. This can be interpreted that the budgetary slack at general hospitals is strongly influenced by information asymmetry. Nevertheless, the existence of budgetary participation can strengthen and emphasize better understanding of budget objectives on all those involved and participate in the budget preparation process that reflect the objectives of the organization in order to minimize budgetary slack. So the harmony between personal goals with organizational goals can be achieved.

Effect of Budgetary participation on Group Cohesiveness with Budgetary Slack in General Hospitals in Banda Aceh
According to the results of the testing of the fourth hypothesis (H4) regarding the effect of budgetary participation on group cohesiveness with budgetary slack, X2 and Y variables can be moderated by budgetary participation, so the fourth hypothesis (H4) is accepted. Although budgetary participation can strengthen the relationship between group cohesiveness and budgetary slack, budgetary participation can also be one of the many elements that can create budgetary slack namely by the participation of agents in a budget preparation and creating a budget with a low difficulty level with purposed to easily achieving the budget targets. This can be done because this agent maintains a good image in the eyes of the principals, so that by preparing a budget with a low level of difficulty the agent can easily realize the budget and can reflect good performance in the eyes of the principals. JAROE VOL. 3(1)

Conclusions, Limitations, and Suggestions
Based on the results of tests that have been carried out on the problems formulated in the research hypothesis using multiple linear regression tests, the conclusions can be drawn as follows. First, information asymmetry affects budgetary slack in general hospitals in Banda Aceh. Secondly, group cohesiveness does not affect budgetary slack in general hospitals in Banda Aceh. Third is budgetary participation influences (can moderate) information asymmetry towards budgetary slack in general hospitals in Banda Aceh. Last but not least budgetary participation influences (can moderate) group cohesiveness towards budgetary slack in general hospitals in Banda Aceh.
This research has several weaknesses. Therefore, these limitations need to be considered in further studies in order to obtain better results in the future. The number of samples in this study is still limited to certain respondents, namely the person in charge of the budget in general hospitals in Banda Aceh and does not involve all other hospitals both inside and outside Banda Aceh. In addition to information asymmetry, group cohesiveness, and budgetary participation, there are still other variables that are considered to influence budgetary slack such as environmental uncertainty, job involvement, budget adequacy, clarity of objectives, and individual capacity and so on. However, despite the limited time and cost, comprehensive research on all these variables is not possible. The data analyzed in this study used questionnaires and the conclusions drawn were only based on data collected through the questionnaires. Hence, it will cause problems if the respondent's answer is different from the real situation. This situation is uncontrollable because it is beyond the ability of the researcher.
Based on the analysis results of the discussion, conclusions, and limitations of this study which have been presented previously, as for the suggestions that can be given among others are in order to make the research to be more representative, it is expected that the capacity of respondents in further studies can be further expanded or enlarged to include all hospitals both inside and outside Banda Aceh. Furthermore the findings of this study can be considered by practitioners and academics as important, because the high budgetary slack will create distorted behavior in the organization concerned. Furthermore, budgetary slack must be controlled or predicted early in order to increase the effectiveness of the company's budget, especially in planning and controlling activities.