The Design of Income Tax System Responding To The Middle Class Growth, and Its Effects on Income Distribution

Anda Nugroho

Researcher, Fiscal Policy Office of Ministry of Finance, Indonesia

Email:

Rita Helbra Tenrini

Researcher, Fiscal Policy Office of Ministry of Finance, Indonesia

Email:

ABSTRACT

A large portion of the Indonesian population is entering the middle-class as its economy is growing rapidly. Currently, middle class in Indonesia is the third-largest in the world. There are about 74 million middle classes in Indonesia, and this number will double by 2020. In the other hand, economic growth also creates a problem of rising inequality. Inequality in Indonesia is worsened as the Gini index increasing from 0.308 in 1999 to 0,41 in 2011. Both conditions, rising middle class and increasing inequalities create a challenge for policy maker to design optimal personalIncome Tax (PIT) system that can capture the tax potential from middle class growth and at the meantime improving the inequality. In developing countries like Indonesia the tax system has been aimed at increasing government revenues as for the past 10 years, the personal income tax revenue has increased from 19,5 trillion rupiah in 2002 to 83,3 trillion rupiah in 2012. More than that, the income tax is also supposed to be used as a public policy instrument to alter after-tax income distribution.

The purpose of this paper is to design a personal income tax system that can capture the increase in tax potential as the middle class growth and also promote better income distribution in Indonesia. Using microsimulation and Computable General Equilibrium (CGE) approach, we quantitatively analyze the way proposed PIT system affect government revenue and alter the inequality of the income distribution. First, we propose some PIT systems and quantify the way they affect after tax income by using the micro data. Next, we employ CGE model and execute the simulation to calculate the effect of proposed PIT systems on the Indonesian economy. We use Indofiscal (Amir, 2011), a CGE model of the Indonesian economy with a focus on fiscal policy. The model has capability of evaluating a range of fiscal policy, including personal income taxes. The result will help the policy makers to design a better income tax system in responding to the current situation of middle income growth and rising inequality.

Keywords: Middle Class, Personal Income Tax, Inequality

  1. INTRODUCTION

Tax revenue has an important role for the Indonesian government, since 70% of government total revenue came from tax. Income tax revenue has the bigest portionof total tax revenue, therefore the volatility of income tax revenue will also impact the government budget. Government have to maintain the stability of income tax revenue as it is the main source of government revenue while using it as a tool for fiscal policy. Income tax consists of three type of tax: corporate income tax, personal income tax, and other income tax. This research focussing on the personal income tax policy since it has the largest impact to the income distribution.

Many research assuming that Indonesia still has a lot of personal income tax potential, considering more than 250 million populatation all over Indonesia. Furthermore a large portion of the Indonesian population now is entering the middle-class as its economy is growing rapidly. Nielsen, a leading media research said that currently, middle class in Indonesia is the third-largest in the world. There are about 74 million middle classes in Indonesia, and this number will double by 2020.

The main objective of this study is to analyze to design a personal income tax system that can capture the increase in tax potential as the middle class growth and also promote better income distribution in Indonesia. This research would give an input for the indonesian government in designing proper personal income tax systemfor the country.The rest of the paper is organized as follows. Section-2 describes the middle class in Indonesia and it’s tax potential. The methodology is discussed in section-3. Section-4 describes the result of the policy simulation.

  1. MIDDLE INCOME CLASS IN INDONESIA AND IT’S TAX POTENTIAL

According to Indonesia Economic Quarterlyreports(WorldBank, 2011), theconcept ofthe middle classcan bedefined in variousaspects such as socio-cultural aspect and socialaspect. In thesocio-cultural aspect, the middle classgenerallyhave a wealthy and prosperouslifein the community. The middle classalsohave agreater probability thanlow-incomehouseholdstobuyhome appliancessuch asTV, refrigerators, motorcycles orcarorown a propertylikehouse.Those who in themiddle classwill bemore ability toseekandconsumehighereducation, an advanced health careservices andrecreational.Whilein the socialterm, most economistsuseincomeorexpenditurepatterntodefinethe middle class.

There are threeapproachestodefining themiddle classbased onexpenditureorincomehouseholds: (i) the absoluteapproach, (ii) the relative approach, and(iii) a combinedapproach. The absoluteapproachdefinesthe middle classata certainlevel ofexpenditure; thosewhohaveincome(or spending) on thespecific upper and lower limitsconsideredasmiddle class. The relative approachemphasizes thelevel ofhouseholdincome orexpenditurerelative tothe others. Numbers ofpercentileexpenditurepercapitais oftenusedforthe relative approach. The combinedapproachis a mixofabsolute and relativeapproaches.

Choicebetweenthe absolute approaches, relativeandcombineddepends on the purposeof the analysisthat will be done. For example, in KharazandGertz(2010) saysthat theabsolutapproachis right to beusedasthe singlemeasure to compare the size ofthe middle classbetween countries. Cannon(1980) concludedthat therelative approachwould bemore appropriatetoassess the"state" or"welfare" of the population,due tothe relativehierarchyis veryimportant for thewelfareorstatusofthe middle class.In other words, people consider themselvesasa member of middle classby comparingtheirselfwithothers. This studyusesthe the absoluteapproachin order todo an analysisofchangesin thenumber of peopleindifferent incomeclasses.This study using the absolute approach whichdefine middle classes based on certain spending levels according to the World Bank definition. The middle class are those with per capita spending between $ 2 to$ 20 per day. This approach is also used by Asian Development Bank (2010) .

FIGURE-1: Proportion of income class 2007-2010

Source : Tenrini and Anda,2013

Tenrini and Anda (2013) have calculated tax potential from the rising of middle income class in Indonesia.They found that more than 50% from all population in Indonesia come from middle income class. Graph 1 shows the result from that study in.The portion of middle income class increasing from 51,4 percent in year 2007 to 54,9 percent in 2010, while the portion of low income class decreasing from 48,5 percent in 2007 to 44,8 percent in 2010. This condition indicates that there is an increasing tax potential for personal income tax because the rising of middle income class.

FIGURE-2: Proportion of consumption of each income class 2007-2010

Source : Tenrini and Anda,2013

Figure 2 shows the consumption proportion for each income class for 2007-2010.It is found that the middle inclome class has is the biggest consumption among others. It is contribute for more than 70% of total consumption. The portion of the middle class income consumption increasing from 73,2 percent in 2007 to 74,6 percent in 2010. This condition also indicates that there is an increasing tax potential for personal income tax because the rising of middle income class.

Currently, middle class in Indonesia is the third-largest in the world. There are about 74 million middle classes in Indonesia, and this number will double by 2020. In the other hand, economic growth also creates a problem of rising inequality. Inequality in Indonesia is worsened as the Gini index increasing from 0.308 in 1999 to 0,41 in 2011. Both conditions, rising middle class and increasing inequalities create a challenge for policy maker in doing tax collection. The government should be able to design optimal income tax system that can capture the tax potential from middle class growth and at the meantime improving the inequality.

According to Musgrave and Musgrave (1989),when the goverment doing tax collection, their also doing their three function, such as; (i) Adressing inefficiencies inamarketsystem, (ii) Carring outthe socialequity throughincomeand wealthdistribution, and (iii) Creating an economic stabilityand overcoming thefluctuationsin the economywhile ensuringemploymentandprice stability. Indonesian government using progressive rateIndonesia to carry out the function of income distributionwhich is tax rate that increasing if the tax base getting bigger. (Soemitro. 1990).

Harvey Rosen (2002)said that progressive rate correspond with vertical equity “it is widely agreed that tax system should have vertical equity : it should distribute burdens fairly across people with different abilities to pay”. Furthermore another justification about progressive tax and income distribution that is income tax should be progressive in that the higher income groups pay tax proportionately more than the lowest income group as a percentage of income (Nugraha And Lewis, 2013).

In developing countries like Indonesia the tax system has been aimed an increasing in government revenues, as for the past 10 years, the personal income tax revenue has increased from 19,5 trillion rupiah in 2002 to 83,3 trillion rupiah in 2012. More than that, the income tax is also supposed to be used as a public policy instrument to alter after-tax income distribution.

  1. METHODOLOGY

This study using microsimulation and Computable General Equilibrium (CGE) approach. We quantitatively analyze the way proposed PIT system affect government revenue and alter the inequality of the income distribution. First, we propose some PIT systems and quantify the way they affect after tax income by using the micro data. Next, we employ CGE model and execute the simulation to calculate the effect of proposed PIT systems on the Indonesian economy. We use Indofiscal (Amir, 2011), a CGE model of the Indonesian economy with a focus on fiscal policy. The model has capability of evaluating a range of fiscal policy, including personal income taxes.

A CGE model is an “economywide” model because it describes the motivations and behavior of all producers and consumers in an economy and the linkages among them (Burfisher, 2011). The use of CGE models in the economy has expanded over the last 25 years. It helps us in getting a better understanding of the interactions occurs in the economy. Next subsection will describe the structure of the model and the model database mode clearly.

3.1.Model Description

This study uses Indofiscal (Amir, 2011; Amir et.al, 2013), a CGE model of Indonesia that equipped with fiscal features. The model is based on ORANI-G (Horridge, 2003) and AGEFIS (Yusuf et.al, 2008). Using Johansen approach, equations are linearized in percentage changes instead of levels of variables. This approach characterizes most of Australian CGE models such as ORANI (Dixon et.al, 1982) and MONASH (Dixon and Rimmer, 2002).

There are four institutions in the model: households, corporations, government, and rest of the world (ROW). As a source of factors of production, households receive income from the production activity. It also receives other income from other institution such as governments, corporations, ROW, and from other households. Taxes are a percentage of household income based on the marginal income tax rate structure. For the corporations, the revenue earned by this institution is caming from its ownership of production factors minus corporate income tax, and transfer from other institutions. While corporate spending goes to payment or transfer to other institutions.

For the government, revenue can be described as the sum of receipts from various sources of taxes and other revenue. Government expenditure consists of expenditure on goods and services for each commodity, and expenditure for the transfer to domestic and foreign parties. Other expenditures made by the government are in the form of subsidies on commodity goods and for industries. Last, for the ROW, foreign income is defined as revenue of the it’s ownership of production factors, payment received from imported commodities and transfer from other institutions. Foreign expenditure consists of spending for exported commodities, payment to production factors and transfer to other institutions.

Figure-3 illustrates the nested production structure in the model. Output are produced through a three-level production structure. At the top nest, the model utilized Leontief function to combine intermediate inputs, primary factors, and other costs in order to produce commodities. Using this approach, commodities are produced by fixed proportion of intermediate demands, primary factors, and other costs.

FIGURE-3: Structure of production

Source: Adopted from Horridge (2003)

At the lower level of production structure, there are nesting for intermediate input demand and primary factor demand. Demand for intermediate input are composed of domestic and imported goods using Constant Elasticity of Substitution (CES) function (Armington, 1969). By utilizing CES function, the producers choose the composition of their intermediate input from domestic and imported goods, in which both of them are imperfectly subtitutable. The demand for primary factors also utilizing CES function to choose the composition of capital and labor composite. Using the function, the composition of capital and labor are determined to minimize the production cost. In addition, each production sector is assumed to produce only a single commodity. It is also assuming constant return to scale and perfect competition in the market.

Similar to the producers that choose input and output levels to maximize their efficiency, consumers also choose their consumption to maximize their utility; by purchasing the most satisfying bundle of products given their budgets and the prices of the commodities. Household demand are for commodity composites are composed using Klein-Rubin utility function (Horridge, 2003) that differentiate the household demand into two categories: (i) subsistence demand and (ii) luxury demand.

The model has several closures, which mainly associated with the simulation timescale. In the short run, we assume that there is not enough time for the capital stock to adjust so that there is no new investment. Capital is sector-specific, that is fixed for each industry and cannot move between sectors. The capital rate of return may adjust to reflect the changes in the demand of capital. Hence the real wage rate is fixed, while aggregate employment can change to respond the changes in the labor market. In the long run, capital stock may adjust because there is quite enough time for the capital stock to adjust, while the capital rate of return is fixed. Aggregate employment is fixed to reflect full employment assumption in the long run, while real wage rate may adjust. In addition, there are some variables that are assigned as exogenous such as technological changes, tax rates, and transfers between institutions.

3.2.Model Database

A CGE model's database describes the circular flow of income and spending in a national economy within a specific time period. It reports the values of all goods and services that are produced and also the income generated from production activities. The database are built in which they are relevant to the research question. Researchers must decide the level of aggregation or disaggregation of the industries and commodities classification in their database. For this purpose, we modified the Social Accounting Matrix (SAM) Table used in the model to analyze the impact of each income tax design.

To dissagregate the SAM table, weuse the 2012 Household Survey (SUSENAS) as the statistical data sources. The 2012 Household Survey, give us valuable information about household characteristics in term of their income and consumtion pattern. We use the data to dissaggregate the household data in the model. It s important to capture the impact of each personal income tax design to each household income group.The SAM table dissaggregation is a crucial step, since reliable data provide a strong base for reliable policy analysis.

Table 1. Personal Income Tax Formulation

Year 2009-2012 / Year 2013-now
Regulation / Act No.36 Tahun 2008 / Act No.36 Tahun 2008
Office cost / 5% from gross income, max IDR 6.000.000 (PMK No.250/PMK.03/2008) / 5% from gross income, max IDR 6.000.000 (PMK No.250/PMK.03/2008)
Non taxable income / Rp 15.840.000
(Act No.36 Tahun 2008) / Rp 24.300.000
(PMK No.162/PMK.011/2012)
Bracket and rate /
  • untilIDR 50 million: 5%
  • AboveIDR 50 millionuntil IDR 250 million: 15%
  • AboveIDR 250 millionuntil IDR 500 million: 25%
  • AboveIDR 500 million: 30%
/
  • untilIDR 50 million: 5%
  • AboveIDR 50 millionuntil IDR 250 million: 15%
  • AboveIDR 250 millionuntil IDR 500 million: 25%
  • AboveIDR 500 million: 30%

Source : Ministry of Finance

To calculate the effective tax rate on each househod, we apply the actual tax rate on the Income Tax Act to the household survey data. There has been several changes in Income Tax ActIn Indonesia. Income tax Act first set in the Act No.7 Year 1983, then revised in the Act No.7 Year 1991, Act No. 10 Year 1994, Act No.17 Year 2000, and Act No. 36 Year 2008. There has been several adjusted in tax bracket, tax rate and non taxable income. The comparison between Act no. 17 Year 2000 and Act no.36 Year 2008 in tax formulation can be seen in the table 1.

  1. SIMULATION AND RESULTS

4.1.Policy Simulation

Ten differentscenarioshave been preparedto analyze each personal income tax system design. Each scenario is designed with specific assumptions so that we can compare all of the results of the personal income tax system designmore comprehensively

Table 2. Simulation Design

Simulation / brackets / bracket start from / rate
more / less / lower / higher / lower / higher
SIM1 / x
SIM2 / x
SIM3 / x / x
SIM4 / x / x
SIM5 / x / x
SIM6 / x / x
SIM7 / x / x
SIM8 / x / x
SIM9 / x / x
SIM10 / x / x

In the SIM1 and SIM2, we re-design the income tax bracket, so we create more brackets SIM1 and less bracket in SIM2. From the simulation result in the SIM1 and SIM2 we can analyze the impact of bracket design to tax revenue and the income distribution of the midle class. Next in the SIM3 to SIM6, we compliment the personal income tax designin the SIM1 and SIM2 with the lowest bracket shift. In SIM3 and SIM4 we shift down the lowest bracket, while in SIM 5 and SIM6 we shift up the lowest bracket. In SIM7 to SIM10 we compliment the personal income tax design in the SIM1 and SIM2 with the change in the tax rate.