"
ABSTRAKA tax reform was made by the Government of Indonesia in 1984. However, inefficiency seem to be still prevalent in the Indonesian tax system. Comparison with other countries during the period 1991 to 1994, for example, indicates that the Indonesia tax ratio is in a lower position. This thesis is intended to study the effect of the 1984 tax reform on the efficiency of the Indonesian tax system.
The level of taxation model calculates the tax ratio needed when the rate of economic growth has been determined. This model basically uses the Harrod models on economic growth as a starting point and modifies it by including tax variables. Tax capacity model correlates selected macro economic variables to the tax variables to obtain the optimum capacity of collecting tax. The tax elasticity model correlates tax elasticity with other selected economic variables. The important thing in this model is the effort to separate the growth of the tax into automatic and discretionary one. Optimization model has basically the same features with the tax capacity model, i.e., to find out the optimum tax function using certain selected variables. In the optimization model, however, objectives and constraints which are not considered in the tax capacity model are included.
The general equilibrium model includes the tax variables into the economic general equilibrium model. The econometric model developed in this thesis is basically a tax capacity type of model.
The efficiency of tax system in this study is developed using the concept of optimum "input-process-output" relationship. Output is the optimum tax collection. Input will be represented by selected economic variables. Taxes are assessed on economic activities. These activities will be reported in the macroeconomic information system where those selected economic variables are part of them. In addition to input-output relationship, the growth of taxes may also be affected by a discretionary variables (process factor). The discretionary variables, include, among others, tax policy, tax administration, tax personnel and environment. In this econometric tax modeling, the discretionary variables will be represented by a dummy variables representing tax reform.
The approach used in this study will be, first, to develop a simultaneous econometric model. The improvement on the Indonesia tax system will be tested using the model above through its dummy variable. The selected economic variables will be classified into group of activities which consist of: (1) aggregate demand; (2) balance of payment; (3) monetary; (4) government budget and; (5) aggregate supply. Variables Y (Gross domestic product), C (Consumption), I (Investment), X (Export), M (Import) and GR (Government Revenue) are selected from the aggregate demand. The balance of the payment group will be represented by X (Export) and M (Import). The monetary and government budget are represented by M2 (supply of money which indicates the economy's liquidity) and the government revenue. The aggregate supply will be represented by Y (Gross Domestic Product), number of employment (N) and Investment (I).
Description of symbols in the equations could be found in the main chapters of this thesis.
In addition to the econometric model, a non statistical analysis will also be made to support the statistical evaluation. The analysis comprises of qualitative, quantitative and correlative analysis. The qualitative analysis compares the substance of the new law against the old one. It is concluded, based on this analysis, that tax paid by the taxpayers may not decrease, although less tariff was introduced under the new law. This statement applies both for income as well as value added taxes. This conclusion has the implication that the increase in the government tax revenues will be dependent upon tax administration and law enforcement. Besides, the taxpayers' awareness and compliance will also play a role in the growth of tax revenue.
The quantitative analysis focuses on the growth and structural changes of tax revenue. There are two variables evaluated i.e., the tax revenue itself and the number of taxpayers. The average annual growth rate of tax revenue per taxpayer is 5.7% for income tax and 57.5% for value added tax. 61% of the growth rate of income tax is primarily due to the increase in the member of taxpayers while the remaining 39% is due to increase in the volume of activities. The value added tax has the reverse situation. The growth rate of value added tax is primarily due to increase in volume of activities (91 %) and the remainder is caused by the increase of taxpayers. Based on this analysis certain preliminary findings could be drawn: (1) value added tax collection is more efficient than the income tax or; (2) the effective tax rate of value added tax is higher than the income tax.
The growth analysis indicates that the value added tax grew faster than income tax. The annual growth rate of income tax were 30% and 23.5% respectively for 10 years before and after tax reform. On the other hand, the percentages for value added tax were 24% ten years before tax reform and 37.5% ten years after that. The consequences of the different growth rate above were the changes in the structure of tax revenue.
Direct taxes as a proportion to total tax revenue decreased from 41% during the period of 10 years before tax reform to 40% ten years after that. The proportion of income tax also decreased from 37.5% to 36.5% during the same period. Value added tax, on the other hand, has a different situation. The proportion of value added tax to total tax revenue has been increasing from 19% during 10 years before tax reform to 35.5% ten years after that. Meanwhile, the proportion of indirect taxes to total revenue increased from 59% to 60% during the same period.
The correlative analysis was done by relating tax revenue with selected economic variables i.e., Gross Domestic Product (Y), Export (X), Import (M), and supply of money (MD). Three types of taxes were evaluated i.e., income tax, value added tax and total tax revenue. Two method of analysis were used i.e., ratio analysis and point of elasticity. The conclusions reached based on the above analysis are:
a. The increment of value added tax revenue due to tax reform was higher than the increment of income tax.
b. Tax reform causes tax structure more regressive.
c. Tax reform seems to increase the efficiency of the Indonesia tax system.
The above conclusion is supported by data such as the fact that ratio of income tax to gross domestic product has increased from 2.54% ten years before tax reform into 3.31% ten years later. The percentages for value added tax were 1.32% before tax reform and 3.23% ten years later. The point elasticity of income tax t0 gross domestic product has increased from 1.12 to 1.42 during the same period. The related numbers for value added tax are 0.08 and 2.35, respectively.
The statistical test performed, using time series data of 1973174 to 1993194, concluded that the model is not fit to be used for estimation. Revision to the model, using logarithmic form, come up with the new one as follows:
(1) In Typph = 5,75 - 1,25 In Y-0,11 In I.2 + 0,54 In X + 0,40 In MD
(0,00) (0,00) (0,01) (0,00) (0,00) Calculated F: 0,00 Adj. R-Squared: 0,79
(2) In TYPPN = 0,36 In 1.2 + 0,38 In X - 0,82 In M + 0,14 In MD - 0,25 TR
(0,00) (0,00) (0,00) (0,01) (0,07) Calculated F: 0,00 Adj. R-Squared: 0,87
(3) In TYOT = 4,56 - 0,80 In Y + 0,55 In X (0.00) (0,00) (0,00) Calculated F: 0,00 Adj. R-Squared: 0,76
(4) In C 0,90 In Y (0,00) Calculated F: NIA Adj. R-Squared: 0,96
(5) In I = -3,33 + 1,18 In Y (0,00) (0,00) Calculated F: 0,00 Adj. R-Squared: 0,99
(6) In X = 0,80 In MIGAS (0,00) Calculated F: N/A Adj. R-Squared: 0,87
(7) In M = 0,88 In Y (0,00) Calculated F: NIA Adj. R-Squared: 0,96
(8) In MD = -9,21 - 0,45 In r + 2,07 In Y (0,00) (0,01) (0,00) Calculated F: 0,00
Adj. R. Squared: 0,95
(9) In GR = 2,42 + 0,93 In FA (0,00) (0,00) Calculated F: 0,00 Adj. R-Squared: 0,99
(10) In Y = 0,17 In N + 0,82 In I (0,00) (0,00) Calculated F: 0,00 Adj. R-Squared: 0,99
It should be noted that with such revision, the type of the model has been changed from tax capacity to tax elasticity model. Significant results were obtained for all equation in the model during the statistical test using Seemingly Unrelated Regression (SUR). Interpretation of the coefficients of the tax equations in the revised model concludes that:
a. Income tax has a negative elasticity relationship with gross domestic product (Y) and Investment two years lag (L2). Elasticities between income tax and export (X) and supply of money (MD) are positiive.
b. Value added tax has a negative relationship with import (I) and tax reform (TR). Positive elasticity was obtained between value added tax and two years lag investment (L2), export (X) and supply of money (MD).
c. Other taxes has a negative elasticity relationship with gross domestic product (Y) and a positive relationship with export (X).
The implication of the above results can be summarized below:
a. Tax reform has an effect on the collection of income tax. The growth of this tax was basically due to automatic growth instead of discretionary one. Meanwhile, although tax reform has an effect on the collection of value added tax, the effect was negative, meaning that tax reform did not improve the efficiency of the tax system.
b. The growth of income tax did not have a relationship with the growth of gross domestic product and two years lag of investment. This is an abnormal situation which could be interpreted that the efficiency of the income tax collection can still be improved. The positive relationship between the growth of income tax and export and supply of money is deemed to be appropriate.
c. Value added tax grew negatively if it is related to the growth of import. Additionally, this model indicates that the growth of value added tax does not have any relationship with the growth of gross domestic product. These two phenomena seem to be abnormal. It could be an indication that the efficiency of the value added tax collection can still be improved. The positive relation between values added tax and export, two years lag investment and supply of money is deemed to be appropriate.
d. Other tax has a negative relation with the growth of gross domestic product. Efficiency improvement is still probable with this kind of tax. Positive relationship with export is deemed to be appropriate.
This study comes up with certain recommendations as follows:
1. The efficiency of income and value added tax collection can still be improved. Improvement should be made on tax administration, law enforcement and certainty and clarity on rules and regulation.
2. The policy on final withholding on income tax should be implemented prudently. This policy may cause the tax system more regressive. The tax object selected should be focused on those related to individual taxpayers rather than corporate taxpayers. Additionally, the final tax withholding should be assessed on the lower income group representing the mass taxpayers.
3. Tax model should be used in the projection of tax revenue. By doing this, more justification could be provided when determining. the target for tax revenue. Additionally, this model could be used as a tool for analyzing the effects of any policies issued by the Government relating to the variables (sectors) included in the model. Preferably the tax model should be combined with the general equilibrium model of the Indonesian macro economy.
4. Tax reform has been proven as being able to increase tax revenue. It is recommended that similar reforms could be made on other taxes and non tax revenue. Attention should be made on non tax revenue, because there is a great potential to develop revenue from this sector. Pricing of the Government services should be reconsidered. At present the pricing of such services does not consider the cost of providing it mainly because it is assumed that the cost would be recovered through taxes. In the context of globalization, however, reconsideration of government services pricing is a must. By doing this, the efficiency of the whole economy may be increased. It should be noted, however, that a cross subsidy concept should also be considered in the pricing process.
5. This study also indicates that the structure of tax revenue is becoming more and more regressive. Attention should be made on income tax. Tax collection efficiency should be improved. The tax payers awareness and compliance program should be focused on this tax. The extensification program should always be continued. Meanwhile, tax rules and regulation should always be kept updated. Law enforcement should be focused on middle class individual (corporate) tax payers.
6. This study also conclude that the tax reform does not have a significant impact to the efficiency of income tax collection system. The learning period needed to reach optimum condition need to be extended. It is therefore recommended that fundamental changes should be avoided. Efforts should continually be made on the improvement of the present system, both internally and externally. The internal improvement includes updating of rules and regulations, computerization of data system and procedures and staff development. External improvement includes, extensification program, law enforcement and integration with other supporting systems such as legal and accounting.
We hope that this study will benefit the readers and stimulate other more comprehensive studies to be made."
