The Influence of Transformational Leadership, Extrinsic Work Motivation and Work Discipline on Employee Performance Agricultural Extension and Human Resource Development Agency (BPPSDMP) Ministry of Agriculture

This research is used to determine whether there is transformational leadership, work extrinsic motivation and work discipline on the performance of BPPSDMP employees of the Ministry of Agriculture. By paying attention to the level of control of human resources (HR), an agency is expected to make a major contribution in improving employee performance. This view also uses primary data. The data collection technique used was to use a questionnaire received by distributing questionnaires via google form to 104 employees at the Ministry of Agriculture's BPPSDMP. The information analysis approach used in this review is multiple linear regression evaluation, which is operated through the SPSS 23 application. Based on the results of this study, it shows that the variables of transformational leadership, work extrinsic motivation and work discipline have a significant effect on employee performance. That partially transformational leadership variables have a significant effect on employee performance. That the extrinsic work motivation variable has a dominant effect on employee performance. And the work discipline variable has a significant positive effect on the performance of BPPSDMP employees at the Ministry of Agriculture.


INTRODUCTION
An association or organization created with the full intention of obtaining as much benefit as can reasonably be expected.The owners usually require the assets of the organization owned to be supervised by management or management services.Administration will be able to the owners to detail the assets of the organization run by executives through financial statements.The results of financial recording are a means to present in a structured manner the company's financial performance, the company's financial position, and and the entity's cash scheme that is useful for most users or readers of financial statements in making economic decisions (PSAK, 2017: 1).
Companies see taxes as an additional cost burden that can reduce their profits.Therefore, the company is expected to take measures to reduce its tax burden.In this case, the taxpayer or owner of the company wants to pay as little tax as possible.(Suandy, 2008) and one way is to do tax planning to predict factors that affect yield control.Tax planning is one of the functions of tax administration to calculate the amount of tax to be paid, as well as a way to reduce taxes.(Astutik, 2016) states that companies carrying out tax planning are used to make tax savings in accordance with the provisions of tax regulations.
Deferred tax assets are a major factor in determining an organization's profit management.According to Sukirno, et al (2009: 244) deferred tax assets occur when time differences cause positive corrections, so that the tax burden according to accounting is smaller than according to tax regulations.Meanwhile, in accordance with Financial Accounting Standard (PSAK) No. 46 2017, the revision of deferred tax assets is the amount of tax recovered in the future due to temporarily deductible assets and compensation for residual losses.Companies are always trying to reduce the amount of taxable profits to lower the tax bill.
Previous research on profit management has been widely conducted, on the impact of tax planning, tax-deferred expenditures and company size on profit management.(Paramita et al., 2021) explained that tax planning has an impact on profit management, while deferred taxes and company size have no impact on profit management practices, then (Khalifah, 2019) explained Tax planning, deferred tax burden, and company size have an effect on profit management.Previous studies gave inconclusive results, so researchers wanted to reevaluate the data in different units.and also the Covid-19 pandemic which caused several sectors to experience a significant increase in the company's profit, researchers conducted this study entitled "The Effect of Tax Planning, Deferred Tax Burden and Company Size on Profit Management in Manufacturing Companies in the Goods and Consumer Sub-Sector Listed on the Indonesia Stock Exchange in 2016-2020".
This study aims to analyze several variables in manufacturing companies in the goods and consumption sub-sector listed on the Indonesia Stock Exchange during the 2016-2020 period.First, this study will analyze the effect of tax planning on profit management practices in the context of these manufacturing companies.Second, the study will also evaluate the impact of deferred tax burden on profit management practices on same-sector companies.In addition, this research will involve an analysis of the effect of company size on profit management practices, opening insights related to how company size can affect the dynamics of profit management in manufacturing companies in the goods and consumption sub-sector on the Indonesia Stock Exchange.

RESEARCH METHOD
This study uses multiple linear regression analysis method to evaluate the impact of independent variables (tax planning, deferred tax burden, and company size) on dependent variables (profit management) in goods and consumption subsector companies, including food and beverages, pharmaceuticals, beauty equipment, household utilities, and household appliances and others, listed on the Indonesia Stock Exchange for the 2016-2020 period.Data is obtained through questionnaires for primary data and annual financial statements of the company concerned for secondary data.The study population includes companies of these subsectors, with sampling using purposeful sampling to ensure criteria for stability and completeness of data.Data collection techniques involve documentation and literature studies, with descriptive analysis for tax planning variables, deferred tax burden, and company size.Operational definitions and calculation formulas are included to understand variables.Multiple linear regression and correlation analysis, F statistical tests and t tests are used to evaluate the relationship and significance of variables, as well as classical assumption tests such as normality, multicollinearity, heteroscedasticity, and autocorrelation to validate models.

RESULTS AND DISCUSSION
This study used a sample of 96 companies consisting of 24 companies from 2016 to 2020.The dependent variable used in this study was profit management.While the independent variables used in this study are tax planning, deferred tax burden, and company size, their effect on company performance has not been studied.Then it can be described according to the table and explained as follows.Profit management which is the dependent variable in this study shows an average or mean value of 0.0107 with a standard deviation of 0.012448.Profit management with the lowest value (minimum) of -0.030 and the highest value (maximum) of 1.15.Judging from the range, it shows that the distribution of data for profit management can be said to be good, this is shown by the value of the standard deviation which is smaller than the average value.The lower the value of the profit distribution generated, the higher the level of profit management carried out by the company.
The independent variable tax planning (X1) shows an average or mean value of 0.6911 with a standard deviation of 0.16974.Tax planning with the lowest value (minimum) is 0.04 and the highest value (maximum) is 0.91.Judging from the range, it shows that the distribution of data for tax planning can be said to be not good.This is shown from the standard deviation value that is greater than the average value.The independent variabek deferred tax expense (X2) shows an average or mean value of 0.0256 with a standard deviation of 0.24492.Deferred tax expense with the lowest value (minimum) of -0.01 and the highest value (maximum) of 2.40.Judging from this range, it shows that the data for deferred tax expenses can be said to be good.This is shown from the value of the standard deviation which is smaller than the average value.
The size of the independent variable company (X3) shows an average or mean value of 28.3528 with a standard deviation of 2.02903.Managerial ownership with the lowest value (minimum) of 21.64 and the highest value (maximum) of 32.27 seen from the range shows that the distribution of data for company size is good.This is shown that the standard deviation value is smaller than the average value.Before performing multiple linear regression testing, it is important to ensure the regression model's assumptions are met.The classical assumption test is the basic assumption that must be met in a regression model.The first classical assumptions that need to be passed to get a good regression model are the normality test, the multicollinearity test, the autocorrelation test, and the heteroscedasticity test.

Normality test
The normality test is used to test whether the recursive model, bound variables and independent variables are normally distributed or not.Test normality using Kolmogorov-Smirnov.Regression models that tend to be accurate are those that are normally or close to normal distributed.(Ghozali, 2016) the criteria for normal residual values are graph investigations and statistical tests.In this test, analysts using Kolmogrov Smirnov's statistical test with the following estimates: If the probability value of kolmogorof smirnov > a signification level of 5% (0.05) , then the distribution of data is said to be normal.If the probability value of Kolmogorov smirnov < a signification level of 5% ( 0.05) , then the distribution of data is said to be abnormal.

Table 1. Kolmogorov-Smirnov One Sample Test Results before outlier
From the results of the One Sample Kolmogorov-Smirnov Test, the level of siginficance can be seen at the value of Asymp.Sig (2-tailed) where the three independent variables do not meet the significance level of > 0.05.There are several ways to be able to omit data that has not been normally distributed, one of which is to exclude some outlier data excluded from the analysis.According to Ghozali (2006: 41) outliers are cases or data that have uni characteristics that look very different from other observations and appear in the form of extreme values for either single variables or combinations.In this test, the amount of data processed (N) is 105 then after being done after the release of the remaining outlier data into 96 data which results in the table data level of the normality test for one-sample Kolmogorv-Smirnov Test 96 samples it can be found that the Monte Carlo Sig.(2-tailed) value of 0.055 if observed has exceeded 0.05, which previously had a significance of 0.000.based on the table obtained the level of significance or Monte Carlo Sig.(2-tailed) in the third Variables greater than 0.05 can be concluded that all data has been distributed normally.The multicollinearity test is designed to find out whether the regression model finds a correlation between independent variables.The method used to detect multicollinearity is to use the values of tolerance and variance inflation factor (VIF). (Ghozali, 2016;103).The VIF limit is 10 if the VIF value is above 10 then multicollinearity occurs.

Multicollinearity Test Results
The calculation results using the SPSS program show that for the independent variable the VIF value for the control planning variable is 1.025;The VIF value for deferred tax burden is 1,134; The VIF value for the company size variable is 1.161.It is realized that the VIF value of all independent factors is below 10.It is good to assume that all independent factors are not exposed to the problem of multicollinearity.
The autocorrelation test aims to test whether in the linear gergertion model there is a correlation between the confounding error in period t with the confounding error in period t-1 (previous).If correlation occurs, then there is an autocorrelation problem.Autocorrelation arises because sequential observations over time are related to each other (Ghozali, 2016)).For autocorrelation tests, the Durbin-Watson test is used only at the sati autocorrelation level (first-order autocorrelation) and requires independent intercepts (constants).Obtained from the above, it is known that the Durbin-Watson value is 1.989 with n = 96 , k = 3, to test whether there is an autocorrelation in this test can be obtained, namely the value of dU = 1.7326 then 4-dU = 2.2674.It can be concluded that the value of d meets the requirements dU< d < 4-dU = 1.7326 < 1.989 < 2.2674 so that there is no positive or negative autocorrelation with the accepted H0 decision.
Ghozali (2016) heteroscedasticity is a condition where there is an inequality of variance of one observational residual in another observation of a linear regression model.The heteroscedasticity test aims to test whether in the regression model there is an inequality of variance from the residual of one observation to another.If the variance from the residual of one observation to another observation is fixed, then at that point it is called homoscedasticity and with different assumptions it is called heteroscedasticity.A viable regression model is one with homoscedasticity or without heteroscedasticity.The heteroscedasticity test can be performed using the scatterplot strategy.The reasons for making a choice are as follows: If there is a regular pattern of dots on a page, heteroscedasticity (change in data values) occurs.Assuming there is no clear pattern, and the points are evenly spread on the Y-axis, heteroscedasticity does not occur at that point.Multiple linear regression results are used to partially test the effect of the independent variable (X) on the dependent variable (Y).(Sugiyono, 2014;277) The results of this direct examination multiple linear regression calculation should be seen in the following table : Table 3

. Multiple Regression Analysis Results
Based on the results of regression analysis it is determined that the value of the constant is -0.126; the value of the tax planning coefficient is 0.034; deferred tax expense coefficient of 0.489; and the coefficient for the size of the enterprise is 0.004.In the opinion of Ghozali (2016: 97) ujji statistics t basically shows how much influence one independent variable individually has in explaining the dependent variasu.In this study using significant levels of 0.05 (α = 5%).ttest results can be seen in the table below: Table 4. Statistical Test Results t From the estimated statistical test t tax planning anticipates profit management using SPSS, a value of Sig = 0.183 > 0.005 is obtained so that it can be concluded that there is no influence between tax planning on profit management in manufacturing organizations of the goods and consumption sub-sector listed on the Indonesia Stock Exchange 2016-2020.From the calculation of the statistical test t deferred tax expense using SPSS, the value of sig = 0.000 < 0.05 can be concluded so that there is a significant effect of partial deferred tax burden on profit management in companies in the goods and consumption sub-sector listed on the Indonesia Stock Exchange in 2016-2020.From the statistical test of t company size using SPSS, the value of sig = 0.113 > 0.05 can be concluded so that it can be concluded that there is no influence between company size on profit management in manufacturing companies in the goods and consumption subsector listed on the Indonesia Stock Exchange in 2016-2020.
The F statistical test basically shows whether all the independent factors inputted to the model together affect the dependent variable.The level of significance or probability is 5% or 0.05.From the SPSS factual test result data, it was obtained that the significant value was stated at 0.000.When compared and the importance level is 0.05 (5%) then at that point, the significance value in the measured F test is 0.000 < 0.05.This clarifies that there is an influence between tax planning, deferred tax burden and shared company size on profit management in manufacturing companies in the goods and consumption sub-sector listed on the Indonesia Stock Exchange in 2016-2020.
The coefficient of determination essentially measures how much the ability of the model equation to explain the variations contained in the dependent variable.The R2 value lies between 0 and 1.If R2 is close to 1 (one), the stronger the model is in explaining the variation of the independent variable to the dependent variable.Conversely, if R2 is 0 (zero), then the weaker the independent variable describes the bound variable.Based on the results of the coefficient of determination, it shows an R Square value of 0.893 or 89.3%.The value of the coefficient of determination of 89.3% indicates the magnitude of the commitment to the impact of tax planning factors, deferred tax burden and company size on profit management.While the excess of 10.7% is the impact of various factors that are not explained.

The Effect of Tax Planning on Profit Management
As a result of calculations in the investigation of individual parameter significant tests statistical tests t) tax planning Anticipating profit management by using SPSS obtained sig value = 0.183 > 0.05 so it can be concluded that tax planning does not have a major impact on profit management, so H1 is not accepted.Based on normally distributed data, the results of descriptive analysis show that manufacturing companies in the goods and consumption sub-sector that are up to this study carry out profit management by avoiding a decline in profits.This is in accordance with the theory proposed by (Philips, John, Morton Pincus, 2003), where the results of descriptive statistics in this study show profit management efforts to avoid a decrease in profits indicated by the large value of the mean variable of profit management which shows a positive number (0.0107).The positive mean variable profit management figure shows the tendency of profit management efforts to avoid a decline in profits.Philips (2003) in his research shows that increasing the deferred tax burden that is an element in tax planning increases the chances of profit management to avoid a decrease in profits.

Effects of Deferred Tax Burden on Profit Management
The calculation results in this study have formulated a deferred tax burden affecting profit management, and in the results of the individual parameter significant test (statistical test t) of 0.000 < 0.005 and the value of the t test of 26,984.then it can be concluded that the deferred tax burden has a significant effect on profit management, then H2 is accepted.The results of this study are supported by the results of research (Suputra, 2017) which states that deferred tax burden affects profit management.The results of research on this variable are influential because the deferred tax burden in companies in the goods and consumer industry sector studied has a larger average so that practices in profit management are getting bigger.In this case, it means that management is successful in managing the tax burden incurred to maximize the profit to be obtained.In research (Mazini mawardi &;Haryanto, 2015) said, the higher the deferred tax burden, the greater the company does profit management to avoid losses.

The Effect of Company Size on Profit Management
The results of the review revealed that there was no large impact of company size on profit management in manufacturing companies listed on the Indonesia Stock Exchange in 2015-2020.This can be proven by the importance value obtained at the significance value of the statistical test t, which is 0.113 > 0.05.So it was concluded that the size of the company has no impact on profit management, so H3 is not accepted.In previous investigations that stated that the size of the company did not have a major impact on profit management.Based on research information, the size of the company as of the total value of assets owned encourages companies to improve profit management.As is the case with size theory, the larger the size of the company, the more likely it is to carry out profit management, since large organizations are politically prominent enough to be noticed from government agencies rather than small ones.

CONCLUSION
This study utilizes purposive sampling techniques with a total sample of 24 companies from the goods and consumption subsector on the Indonesia Stock Exchange.Data processing is carried out using SPSS software version 25, and the data used is qualified and eligible for testing.The results of research and discussion concluded that tax planning does not have a significant effect on profit management, while deferred tax burden has a significant effect on profit management.However, the size of the company has not been shown to have a significant effect on profit management.