The impact of the Indonesian income tax reform: A CGE analysis

Economic Modelling 31 (2013) 492–501

Contents lists available at SciVerse ScienceDirect

Economic Modelling journal homepage: www.elsevier.com/locate/ecmod

The impact of the Indonesian income tax reform: A CGE analysis Hidayat Amir a, John Asafu-Adjaye b,⁎, Tien Ducpham c a b c

Centre for Fiscal Management, Jalan Dr. Wahidin No. 1, Jakarta 10710, Indonesia School of Economics, The University of Queensland, St Lucia Q4072, Australia School of Tourism, The University of Queensland, St Lucia Q4072, Australia

a r t i c l e

i n f o

Article history: Accepted 21 December 2012 Keywords: Indonesia Tax reform Economic growth Poverty CGE analysis

a b s t r a c t This study evaluates the impacts of Indonesia's recent income tax reforms on key macroeconomic variables, as well as the impacts on poverty and income distribution. It was found that the reductions in personal income tax and corporate income tax increase economic growth under a balanced budget assumption. The policy reforms also lead to a small reduction in the incidence of poverty. However, the policies also lead to an increase in income inequality because the tax cut is more beneficial to households in the highest income categories. It is recommended that future tax cuts should target the urban and rural poor. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Governments use tax systems as policy instruments to achieve a variety of objectives, among them are income redistribution, economic stabilisation, providing public goods, and fostering economic growth. The combination and importance of these objectives vary for each country, and usually depend on their respective political and economic backgrounds. As such, the design of the tax system is different for each country. While a good tax system must be efficient and equitable, in reality, these two goals may conflict with each other. For example, a tax system may be efficient – in the sense of causing minimum distortions in the economy, thus fostering economic growth – but it also may be inequitable in its effects on the income distribution. This trade-off has long been a central debate in developing a theory of optimal taxation; for example, see studies conducted by Mirrlees (1971), Feldstein (1973), Sandmo (1976), Samuelson (1986), Martimort (2001), and Krause (2005). The issue of how fiscal policy affects dynamic efficiency (and, in turn, growth) and redistribution has attracted much attention in applied macroeconomics. Many studies have explored how tax policies affect a country's economic growth rate. For example, Gemmell (1988) examined some effects of taxation on economic growth within the Keynesian tradition of least developed countries. He concluded that the relationship between taxation, savings, and growth was complex, and that the theory that taxation significantly influences economic growth is unjustifiable. In the setting of neoclassical economics, Engen and Skinner (1996) analysed the effect of taxes on US economic growth, using a theoretical approach based on the Solow growth model, and an ⁎ Corresponding author. Tel.: +61 7 33656539; fax: +61 7 33657299. E-mail address: [email protected] (J. Asafu-Adjaye). 0264-9993/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.econmod.2012.12.018

empirical approach based on the country's historical economic record. They concluded that the design of the tax system was likely to exert a modest, but cumulatively important influence on long-term growth rates. However, it is worth noting that countries whose tax structures are more efficiently administered and legally enforced are more likely to enjoy faster economic growth rates than countries without (Engen and Skinner, 1996). Other studies that describe the relationship between taxes and economic growth include Goulder and Summers (1989), Easterly and Rebelo (1993), Mendoza et al. (1994), Stokey and Rebelo (1995), Auerbach (1996), and Lee and Gordon (2005), among others. Between them, some have suggested that taxes negatively correlate with economic growth; some suggest that correlations between the two are relatively small or insignificant, while others found the opposite to be true. In addition, many studies have analysed how economic growth provides the optimal impact for poverty reduction or income distribution. In the 1990s and 2000s, the analyses of developing countries highlighted the importance of a high level of economic growth in accelerating poverty reduction (e.g., see Bourguignon, 2003; Dollar and Kraay, 2002; Essama-Nssah, 2005; Kraay, 2006; Ravallion and Chen, 1996; Son, 2004; Son and Kakwani, 2008; Timmer, 2007). Levels of inequality are also a factor to consider. It has been found that economic growth is less efficient in reducing poverty in countries with high levels of inequality, or where growth benefits the ‘non-poor’ more (Persson and Tabellini, 1994; Ravallion, 1997, 2001). The term ‘pro-poor growth’ became popular due to the argument that poverty reduction requires both rapid economic growth and an equal distribution of income. Other studies that have analysed the issues of taxation policy, economic growth, and poverty or income distribution have revealed that the relationships are complex (Auerbach, 1996; Eicher et al., 2003).

H. Amir et al. / Economic Modelling 31 (2013) 492–501

Understanding the wider impact of taxation policy in a given economy is important, not only to strengthen basic theories but also to aid policy development on how to achieve a more efficient and equitable taxation system. Most of the studies that are based on optimal tax theory – beginning with the first study by Mirrlees (1971) – have taken a more theoretical approach that has often produced very little practical advice to inform policy-making (Sørensen, 2007). Many of the previous empirical studies have adopted partial equilibrium approaches that have failed to estimate the full impacts of taxation policy. There is therefore the need for a more comprehensive approach that takes into consideration the various interrelationships between all actors in the economy, in order to more realistically estimate the economic effects and distributional consequences of any tax policy changes. In view of the foregoing, the main objective of this study is to identify and quantify the direction and magnitude of the effects of the Indonesian government's recent tax policy reforms on Indonesia's economy. More specifically, the study looks at the macroeconomic impacts of reducing the marginal tax rate on personal income tax (PIT) and the introduction of a flat tax rate on corporate income tax (CIT). The analysis will evaluate the impacts of tax policy reform at both the macro- and micro-levels. The former will include impacts on aggregate variables such as economic growth, employment, government revenue and so on. The latter will include sectoral impacts, impacts on household welfare, as well as income distribution. This study makes the following contribution to the existing literature. It is the first to use a CGE model to specifically analyse the modern taxation system in Indonesia. This is done by combining the latest national Input–Output (IO) table with a Social Accounting Matrix (SAM) to evaluate the impact of Indonesia's recent tax reforms at both the macro- and micro-levels. This information is supplemented with data from the National Socioeconomic Survey (Susenas) to provide a rich database that facilitates the analysis of the impacts of the tax policy on poverty and income distribution in Indonesia. The paper is structured as follows. After this Introduction, Section 2 presents a brief review of the 2008 income tax reform in the context of Indonesia's economic development. Section 3 offers a general description of the main features of our model highlighting the database construction, structure of production, final demand, composition of the institutions in the economy and the model. Section 4 describes the policy simulations and the magnitudes of the policy shocks, while Section 5 analyses the simulation results. Finally, Section 6 presents the conclusions and policy recommendations. 2. Income tax reform in Indonesia Indonesia is classified as a lower middle-income country with a Gross Domestic Product (GDP) of US$540.27 billion and a nominal per capita GDP of US$2349.38 in 2009. Industry (manufacturing and nonmanufacturing) is the largest economic activity and accounted for 46.7% of GDP in 2009, followed by services (39.2%), and agriculture (14.1%). However, agriculture employs more people than the other sectors, accounting for 41.2% of the total workforce of 99.6 million people, followed by services (39.9%), and industry (18.8%) in 2008 (World Bank, 2010). For nearly three decades, Indonesia's economy grew at an average annual rate of 7.2%, until the Asian financial crisis in 1997. In 1993, in light of its economic performance, Indonesia was classified as one of Asia's newly industrialising countries by the World Bank. Indonesia's rapid economic growth was accompanied by a steady decline in poverty and a rapid increase in investment. Rapid industrialisation transformed Indonesia from an economy once highly dependent on agriculture to a newly industrialising economy. At the end of 1997, when the monetary crisis hit Southeast Asia, the Indonesian economy suffered a major development setback. Its economy contracted by 13.1% in 1998, which was about double the reduction in Malaysia and Thailand (Hill, 2000).

493

By 1999, the Indonesian economy had recovered and it grew at a very modest rate of 0.8%, increasing steadily up to 2008, but still below the average level of the three decades preceding the crisis. According to the previous method of estimation used by national statistics agency (BPS—Statistics Indonesia), the high economic growth during those three decades was responsible for reducing poverty from 40.1% (equivalent to 54.2 million people) of the population in 1974 to 11.3% (22.5 million people) in 1996. On the other hand, the new methodology puts this figure at 17.5% (34 million). 1 The new methodology was introduced in 1996 due to the increasing nature of interregional commodities. The 1997 Asian financial crisis caused the number of poor people to surge to 49.5 million (24.2%) in 1998. Various policy measures undertaken successfully reduced this figure to 38.7 million (23.4%) in 2000. Since then, the poverty rate has been relatively stable at about 15% (see Fig. 1). Another problem facing the Indonesian economy is unemployment. According to BPS—Statistics Indonesia, the unemployment rate from 1980 to 1996 was only about 3%, but after the 1997 financial crisis this figure jumped to 6.3% in 1999, reaching a peak of 10.5% in 2006. By 2008, it had declined to 8.5% (BPS, 2009b). These statistics suggest that although the previous high level of sustained economic growth reduced the number of poor people significantly, it failed to improve the equality of income distribution or significantly reduce unemployment. This indicates that poverty, income distribution, and unemployment are still major threats to Indonesia's future economic development. Within the last decade, the Indonesian government has aimed to continually reform its tax system in order to adapt it to achieve its fiscal policy objectives. After the 1997 financial crisis, Indonesia focused on its economic recovery. For five years, Indonesia's economy was faced with low growth, high unemployment and inflation, and financial distress. The government has faced several challenges in its efforts to achieve fiscal sustainability, particularly given the continual decline of revenue from oil and natural gas, and its commitment to gradually reduce foreign debt (Ikhsan et al., 2005; Nasution, 2002). On the revenue side, the Indonesian government has no other choice but to effectively mobilise revenue from taxes. Taxes have a great potential to be the main source of government funding. Tax revenue increases can be achieved by improving tax administration, expanding the tax base, or by increasing tax rates. Ikhsan et al. (2005) concluded that there is still an opportunity to increase Indonesia's national tax revenue without increasing the tax rate. Not only is the tax ratio to GDP still relatively low compared with other developing countries, but also the number of registered tax payers who actually pay taxes is still very low in relation to the population of the country. Typically, tax systems in developing countries are not efficient due to a lack of modern tax administration and a limited number of tax payers. In Indonesia, this situation is made worse by the high level of tax avoidance. These factors have been discussed by Gillis (1985), Marks (2003a, 2003b), and Ikhsan et al. (2005). In their study, Ikhsan et al. (2005) concluded that Indonesia tax reforms undertaken prior to the twenty-first century were successful in increasing government revenue, but were not at the optimal level required. Table 1 shows trends in the composition of government revenue from 1980 to 2008. After radically reforming its postcolonial tax system in 1984, Indonesia's tax revenue increased significantly from 5% of GDP in 1980–81 to 9.9% of GDP in 1995–96 when the second reform was carried out. From that time, the ratio of tax revenue to GDP remained relatively stable, staying below 13% until 2007 and reached 13.3% in 2008.

1 BPS—Statistics Indonesia measures poverty using a basic needs approach. Poverty is defined as the inability to meet basic needs i.e. food and non-food. According to this approach, the poor are the people who have an average expenditure per capita per month below the poverty line of Rp151,997 (about US$1.55 PPP a day) for the year 2006 (BPS, 2009a; World Bank, 2006).

494

H. Amir et al. / Economic Modelling 31 (2013) 492–501

60.00 50.00

million (old method)

million (new method)

% (old method)

% (new method)

40.00 30.00 20.00

2009

2007

2008

2005

2006

2004

2003

2002

2001

2000

1999

1998

1996

1993

1990

1987

1981

1984

1980

1978

0.00

1976

10.00

Source: BPS-Statistic Indonesia Fig. 1. National poverty statistics.

However, the tax revenue share in total government revenue increased sharply from only 27.1% to 67.9% over the period from 1980–81 to 1995–96, and has remained relatively stable since then. On the other hand, the proportion of nontax revenue (including oil and natural gas revenue) declined from 72.9% to 32.1% during the same period. Total income tax revenue from personal and corporate income taxes increased significantly from only Rp0.9 trillion in 1980–81 to Rp327.5 trillion in 2008, equivalent to a 23.3% increase in total revenue and grants. Total income tax revenue in 2008 contributed about 50% of total tax revenue. Value added tax and luxury sales tax (VAT) receipts also increased significantly from Rp0.4 trillion in 1980–81 to Rp209.6 trillion in 2008.

VAT's contribution in 2008 was about 32% of the total tax revenue. In total, income tax and VAT contributed more than 81% of total tax revenue or about 55% of total government revenue in 2008. As a consequence of this significant contribution, the government focused on these two taxes (i.e. income taxes and VAT) when reforming the taxation system in the 2000s. In 2008, the Indonesian parliament approved the government's proposal to amend the income tax laws. One of the essential amendments was adjusting marginal tax rates in PIT and applying a single (flat) tax rate to the CIT. Table 2 shows the adjustment of the marginal income tax rate in the new 2008 income tax legislation. This amendment is part of the tax reform and fiscal adjustment programme that started in

Table 1 Composition of Indonesian government revenues 1980/1981–2008. 80/81 (% of GDP) Total revenue and grants Non tax revenue Tax revenue Domestic taxes Income tax VAT and luxury sales tax Land and building tax Excise Other taxes International trade taxes Surplus/deficit (% of total revenue and grants) Total revenue and grants Non tax revenue Tax revenue Domestic taxes Income tax VAT and luxury sales tax Land and building tax Excise Other taxes International trade taxes Surplus/deficit Memorandum items: Total revenue and grants GDP nominal (Rp trillions) Annual GDP growth (%) External debt, total (% of GDP)

85/86

90/91

95/96

2000

2005

2006

2007

2008

18.5 13.5 5.0 3.6 1.9 0.8 0.2 0.7 0.0 1.4 −2.4

19.3 12.3 7.0 6.2 3.2 1.7 0.2 1.0 0.1 0.8 −2.7

15.0 6.3 8.7 7.7 3.1 3.2 0.3 0.9 0.1 1.0 −2.2

14.6 4.7 9.9 9.1 4.2 3.7 0.4 0.7 0.1 0.8 −0.2

14.7 4.9 9.7 9.2 5.2 2.6 0.2 1.0 0.1 0.6 −4.2

17.9 5.3 12.5 12.0 6.3 3.7 0.7 1.2 0.1 0.5 −0.5

19.1 6.9 12.3 11.9 6.3 3.7 0.7 1.1 0.1 0.4 −0.9

17.9 5.5 12.4 11.9 6.0 3.9 0.8 1.1 0.1 0.5 −1.3

19.8 6.5 13.3 12.6 6.6 4.2 0.6 1.0 0.1 0.7 −0.1

100.0 72.9 27.1 19.5 10.1 4.4 0.9 3.9 0.3 7.5 −12.8

100.0 63.7 36.3 31.9 16.5 8.9 1.0 5.2 0.4 4.4 −14.2

100.0 42.2 57.8 51.2 20.6 21.6 2.0 6.1 0.9 6.6 −14.4

100.0 32.1 67.9 62.5 29.0 25.1 2.9 5.0 0.5 5.4 −1.6

100.0 33.7 66.3 62.5 35.5 17.7 1.6 6.7 0.7 3.9 −28.9

100.0 29.9 70.1 67.0 35.4 20.5 4.0 6.7 0.4 3.1 −2.9

100.0 35.9 64.1 62.1 32.7 19.3 3.8 5.9 0.4 2.1 −4.6

100.0 30.6 69.4 66.4 33.7 21.8 4.2 6.3 0.4 3.0 −7.0

100.0 32.9 67.1 63.4 33.4 21.4 3.2 5.2 0.3 3.7 −0.4

9.1 48.9 8.7 26.8

18.7 97.0 3.5 42.0

31.6 210.9 9.0 61.1

66.3 454.5 8.4 61.5

152.9 1042.3 4.9 87.4

495.2 2774.3 5.7 46.5

638.0 3339.2 5.5 35.9

707.8 3949.3 6.3 32.2

981.6 4954.0 6.1 30.1

Sources: Ministry of Finance (various years). Note: In 2000, the period of state budget changed. From 1980 to 1999, the period of state budget was 1 April to 31 March, from 2000 onwards; it has been 1 January–31 December. For 2000, it was only nine months (1 April 2000–31 December 2000).

H. Amir et al. / Economic Modelling 31 (2013) 492–501 Table 2 Marginal income tax rate adjustmenta. Income range (Rp million)

Personal income tax

0–25 25–50 50–100 100–200 200–250 250–500 >500

5 10 15 25 35

Old marginal tax rate (%)

Corporate income tax Old marginal tax rate (%)

New (flat) tax rate (%)

5

10

28% (from 1 Jan 2009) 25% (from 1 Jan 2010)

15

15 30

New marginal tax rate (%)

25 30

Source: Directorate General of Tax, Ministry of Finance. a The reform was effective from 1 January 2009.

495

purchase their materials from domestic or imported sources, whichever gives the cheaper price. If the price of material from the domestic market increases and becomes more expensive relative to the imported material, producers will substitute imported goods for domestically produced goods. The degree of substitution is governed by Constant Elasticity of Substitution (CES) (or Armington) parameters. In the second nest, the cost of the demand for primary factors is minimised using a CES function. Similar to the procedure in the intermediate demands, producers substitute more expensive inputs (capital or labour composite) with cheaper ones. In the lowest level, the cost of the labour composite demand is minimised using a similar CES function to combine the 16 labour types of inputs. The lowest cost labour types will substitute the more expensive of labour types in order to minimise the total cost of labour usage. 3.2. Investment demand

the early 2000s (see Brondolo et al., 2008), and is still being implemented with a VAT and luxury sales tax reform on the agenda. 3. Description of the model Some aspects of our model were based on ORANI-G (Horridge, 2003) and the Applied General Equilibrium Model for Fiscal Policy Analysis developed by Yusuf et al. (2008). We also incorporated useful information from the 2005 Indonesian SAM, particularly the part regarding transactions between agents in the economy. The theoretical structure of our model is based on the Johansen approach, in which the equations are linearised using percentage changes instead of the levels of variables. This is also the approach used by most Australian CGE models such as ORANI (Dixon et al., 1982) and MONASH (Dixon and Rimmer, 2002). In terms of extending the household categories to have adequate features on poverty and income distribution analysis, we adopted the approach used by Yusuf (2007) with several modifications. The database for our model consists of 24 commodities and industries, four types of margins (trade, road transportation, air and water transportation, and transportation support), two sources (domestic and imports), 16 types of occupations and 200 household classifications based on the percentiles of income distribution: 100 for rural and 100 for urban areas. The database was consolidated from three key data sources: (a) the 2005 Indonesian IO Table; (b) the 2005 Indonesian SAM; and (c) the 2005 National Socioeconomic Survey (Susenas). All the data were published by BPS—Statistics Indonesia. We used two main steps to consolidate the three data sources into the final model database. Firstly, we expanded the household categories in the 2005 SAM and the 2005 IO table using information from the Susenas. Secondly, we combined the extended 2005 IO Table with the extended 2005 SAM to produce a model database with detailed household information.

The structure of final demand for investment by industries is very similar to the structure of production. At the top level of the nest, investment by industry is assumed to have a linear relationship among inputs. To some extent this reflects the prevailing technology of the industry, as when the industry expands it would require certain types, and proportions, of inputs to for the production process. However, investors are assumed to minimise their capital formation costs by choosing the cheaper source between domestic and imported sources. 3.3. Household demands There are 200 different household categories in model, and it is assumed that each household maximises its utility by choosing the commodities to be consumed subject to their budget constraints. The nesting structure for household demand is nearly identical to that for investment demand. The only difference is that composite commodities are aggregated by a Klein–Rubin utility function (Horridge, 2003), resulting in a linear expenditure system (LES). In a similar fashion, households are assumed to select the cheapest cost of goods and services for their final consumption from the domestic and imported sources using the CES functional form. The household utility function only determines the composition of commodities demanded by the households to maximise their utility. The total of household consumption in an economy is determined by (a) the number of households in each category, which is assumed to be constant in this study, and (b) total household disposable income or household income minus the level of the PIT rate. More detail on the household income equations will be discussed in the section on institutions in the economy below. 3.4. Export demands

3.1. Structure of production The nested structure of production in the model is illustrated in Fig. 2. The industries in the model are single output industries, using as inputs domestic and imported commodities, primary factors and other costs. The primary factors of production include capital and 16 labour types as mentioned earlier. Output is produced through a three-level nesting process. In the top level, the production of output in each industry requires intermediate inputs, primary factors and ‘other costs’. ‘Other costs’ refers to all production taxes/subsidies and payroll taxes. All of these inputs are combined via a fixed-proportion relationship in a Leontief function to produce commodities. At the lower level of the production structure, there are two nests: import/domestic composition of intermediate inputs and primary factor proportions. In the first nest, the intermediate input demands for producers follow a cost minimisation function through an imperfect substitution of domestic and imported goods using the Armington assumption (Armington, 1969). To minimise costs, producers choose to

There are two types of export demands: traditional and nontraditional exports. For the traditional exports, each commodity is modelled to have a separate downward sloping demand curve. For non-traditional exports, each commodity is assumed to move in line with aggregate non-traditional exports, which is modelled as a downward sloping demand curve. 3.5. Economic agents This section of the model is a significant departure from the standard ORANI-G type model. Here, we incorporate information taken from the 2005 Indonesian SAM to capture the flow of funds and transactions between agents in the economy. There are four types of economic agents in the model: households, firms, government, and the rest of the world (ROW). Households are a source of factors of production and receive income from their ownership of these factors. Household income can also be derived from transfers received from governments, firms, ROW and

496

H. Amir et al. / Economic Modelling 31 (2013) 492–501

OUTPUT

Leontief

Good 1

-- up to --

Good G

Primary factors

CES

CES

CES

Domestic good 1

Imported good 1

Domestic good G

Imported good G

Capital

Other Costs

Labour Composite

KEY CES

Functional form

Inputs or outputs

Labour Type 1

Labour Type 2

-up to-

Labour Type O

Source: Adopted from Horridge (2003) Fig. 2. Structure of production.

from other households. Household after tax income is equal to disposable income, and taxes are a percentage of household income based on the marginal income tax rate structure. Part of disposable income is spent and the rest will be saved. Corporate income consists of the revenue from ownership of capital less corporate income tax, and transfers from other institutions. Corporate expenditure includes payments or transfers to other institutions and the balance is defined as corporate savings. Total government revenue is defined as the sum of receipts from the following sources: (i) indirect taxes; (ii) revenue from import tariffs; (iii) household income tax (PIT) revenue; (iv) corporate income tax (CIT) revenue; (v) transfers from foreign parties; and (vi) revenue from government-owned production factors. Government expenditure consists of expenditure on goods and services, as well as interest payments to domestic and foreign parties. Other expenditures made by the government are in the form of subsidies to industries. Finally, government budget balance (surplus) is defined as total government revenue less total government expenditure. For ROW, foreign income is defined to include revenue of the rest of the world from ownership of production factors, payments received from imported commodities and transfers from other institutions. Foreign expenditure consists of spending for exported commodities, payments to production factors and transfers to other institutions. The balance is defined as foreign savings.

On the other hand, in the long run closure, the labour wage rate is fully flexible to keep the economy in full-employment. Labour can move across sectors and different types of occupations. In addition, the capital stock is allowed to change and move across sectors. Variables that are assigned to be exogenous in the short- and long-run simulations are tax rates, transfers between institutions and technological change. In the policy experiments, we run the simulations under two different conditions: with and without balanced budget conditions for the shortand long-run scenarios. The balanced budget (or budget neutrality) condition is defined as the situation where government spending and revenue move in such a way as to maintain a constant budget balance. The non-balanced budget condition is when revenue and spending do not necessarily move in line. Without the balanced budget condition, the tax rate reduction policies will increase the government's budget deficit for the current year. To close the increment in the budget deficit, the government must seek additional budget financing, which is usually from issuing government bonds or from incurring domestic or foreign debt. As a result, the government will bear the burden of repayment in the subsequent years. However, under the balanced budget condition, the potential additional burden on the government's budget can be avoided by reducing the level of spending.

3.6. Closure

Three key policy scenarios were examined. These were: (i) impacts of the adjustment of the marginal personal income tax rate structure; (ii) impacts of the application of a flat corporate income tax rate of 28%; and (iii) the effects of the simultaneous implementation of these two polices. We considered each of these scenarios under both the balanced budget and non-balanced budget conditions. An important issue to address prior to undertaking the policy experiments was to determine the magnitude of the shock to apply in each scenario. In the Indonesian tax system, income taxes consist of taxes on different kinds of income stipulated in the different articles of the Income Tax Law. Personal income tax is governed by Income Tax Article 21 (salary and wages tax) and Income Tax Article 25/29. Corporate

In a comparative static CGE model, such as the one used in this study, time is represented in terms of the short run and the long run. In the short-run closure, it is assumed that there is not enough time for the capital stock to adjust following a policy shock and thus there is no new investment. Capital and investment are therefore assumed to be fixed. The rate of return adjusts to reflect the changes in the demand for capital. The short-run closure also assumes that this time frame is not long enough for contractual labour to adjust. Hence, the real wage rate is fixed. Changes in aggregate employment therefore reflect the demand for labour.

4. Policy scenarios and magnitudes of the shocks

H. Amir et al. / Economic Modelling 31 (2013) 492–501

income tax is governed by Income Tax Article 25/29, while some taxes are paid by both individuals and corporations under Income Tax Article 22 (import and non-import), Article 23, Article 26, and other types of income taxes, such as final tax, departure tax, and oil and gas taxes. Income Tax Article 21 and Article 25/29 (Individual and corporation) are subject to the new changes in the income tax rate structure, while others are not affected. The magnitudes of the shocks to be applied were determined in a number of steps. Firstly, based on tax returns data obtained from the Directorate General of Tax, we estimated the proportions of the amount of taxes that are affected by the changes in the marginal PIT and CIT rates. These were found to be 92.9% and 35.6%, respectively.2 Secondly, we estimated the impact of the marginal PIT rate changes on household tax payments. Examples of the magnitudes of the shocks for some of the household categories are shown in Appendix Table 1. Thirdly, we estimated the impact of applying the single CIT rate on corporate income tax payments. From our calculations (see Appendix Table 2), we determined that the reduction in CIT payment is 5.4%. Finally, we estimated the magnitude of the shock for the CIT rate change as a product of the change in the CIT payments (−5.4%), the proportion of taxes affected by the tax rate changes (35.6%), and the effective CIT rate in the 2005 Indonesian SAM (30%). This results in a calculated shock of −0.57% for the CIT policy change. 5. Simulation results This section presents a summary of the long-run simulation results on the effects of the income tax policy under the balanced budget and non-balanced budget conditions. 5.1. Macroeconomic impacts Table 3 reports results for the macroeconomic effects of the tax reform under the two scenarios. We first consider the first half of the table which reports results for the balanced budget condition, starting with the reduction in the PIT rate (SIM1). Under the balanced budget assumption, government consumption expenditure declines by 0.409% in order to maintain the budget neutrality position. The reduction in government demand lowers the output of the public (government) sector, thus releasing labour to other sectors of the economy. The lower demand for labour is reflected by a decline of 0.046% in the average real wage rate, which lowers the cost of production. This, in turn, enhances the external competitiveness of firms, resulting in an increase in exports of 0.15%. At the same time the lower income taxes generate higher demand from the household sector due to the higher disposable income, resulting in a 0.418% increase in real private consumption. The effect of the higher demand from the household sector offsets the decline in public sector output and puts upward pressure on the real wage rate. The strong income effect also results in a 0.071% increase in the demand for imports. Real investment increases by 0.096% in response to the higher demand for capital required for domestic production. The increase in real private consumptions, real investment and net exports results in real GDP increasing by 0.185%. The second simulation (SIM2) involves a reduction in the CIT rate. Again, the immediate effect of the policy is a reduction in corporate tax revenue. The government responds by decreasing its expenditure by 0.135% in order to keep the budget balanced. The reduction in the output of the government sector reduces the demand for labour. The lower demand for labour from the government sector leads to a reduction in the real wage rate under the full employment assumption. In the same mechanism discussed previously, the lower corporate tax rate generates more disposable income in the economy thus inducing higher

2

Details of these calculations are available from the authors upon request.

497

Table 3 Macroeconomic effects (% changes). Variable

Real GDP Real private consumption Real investment Real government consumption Real exports Real imports Consumer price index GDP price index Average real wage

Balanced budget

Non-balanced budget

PIT

CIT

PIT + CIT

SIM1

SIM2

SIM3

PIT

CIT

PIT + CIT

SIM4

SIM5

SIM6

0.185 0.418

0.084 0.018

0.270 0.436

−0.113 0.404

−0.005 0.014

−0.118 0.418

0.096 −0.409

0.035 −0.135

0.131 −0.544

−0.025 0.051

−0.001 0.002

−0.026 0.053

0.151 0.071 −0.072

0.162 −0.004 −0.077

0.314 0.067 −0.149

−0.448 0.095 0.216

−0.016 0.003 0.008

−0.464 0.098 0.224

−0.087 −0.046

−0.094 −0.050

−0.181 −0.096

0.263 0.138

0.009 0.005

0.272 0.144

Source: Model simulation results.

household consumption. However, because not all of the CIT reduction is translated into increase in household disposable income that is as high as the PIT case, the increase in household consumption is not as high compared with SIM1. In this case, household consumption increases by only 0.018%. The lowering of the CIT rate reduces the average real wage which is not fully offset by the upward inflationary pressure induced by higher household demand. Therefore, there is a decline in production cost that makes domestically produced goods relatively cheaper than the imported ones. Consequently, there is a decrease in import demand (−0.004%) compared with the increase in import demand in SIM1, and an increase in exports. Overall, real GDP increases at lesser rate (0.084%) compared with SIM1. In comparison, both the PIT and CIT reductions in the first instance will reduce domestic production costs and increase exports. However, the difference between the two policies is that the income effect in SIM1 is much stronger than in SIM2 as the income generated from the tax rate reduction in SIM1 (Rp13.6 trillion) is much higher than in SIM2 (Rp4.3 trillion) as shown in Table 4. Furthermore, the PIT reduction is more directly related to final consumption than the CIT reduction; therefore the former provides a greater stimulus to household demand, resulting in an increase in economic growth which is more than twice as much as in the latter. The second half of Table 3 shows results for the unbalanced budget condition in which government spending does not have to be reduced to match the reduction in revenue. The implicit assumption here is that the government would run a budget deficit by borrowing and the payment is deferred into the future period. The results for simulations 4, 5 and 6 show a similar pattern. The initial reductions in the marginal tax rates effectively inject more income into the economy. Household consumption increases in all cases, but is much stronger in the PIT case than in the CIT tax as explained previously. Under the non-balanced budget condition, both real government consumption and real household consumption increase, putting upward pressure on the domestic price level and the average real wage. This raises domestic production costs, causing domestic industries to lose their competitiveness in the international market. Consequently, exports decline in all the three scenarios. As a result of that, demand for capital, and subsequently investment, also falls. Under the non-balanced budget condition, the price effect is much stronger than the income effect compared with the balanced budget condition due to the additional demand from the government sector; there is no reduction in the real wage rate in the government sector to offset the upward inflationary pressure. This is reflected by the substitution away from domestically produced goods towards imported goods. As a result, imports increase in all three scenarios (see results of SIM4, SIM5 and SIM6 in Table 3). The important implication of these results for the nonbalanced budget condition is that on top of the initial loss of government revenue due to the tax cut, total government revenue will decline further

498

H. Amir et al. / Economic Modelling 31 (2013) 492–501

Table 4 Fiscal impacts (Rp billion). Budget condition

Revenue

Balanced budget Indirect tax Import tariffs Personal income tax Corporate income tax Government transfer Foreign Factor ownership Government consumption Subsidies (industries) Subsidies (imported commodities) Transfers to households Transfers to corporations Government saving (deficit) Total

SIM1 (PIT) 497.5 50.0 −13,575.7 989.0 −71.4 −0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 −12,111.6

Non-balanced budget

SIM4 (PIT)

Indirect tax Import tariffs Personal income tax Corporate income tax Government transfer Foreign Factor ownership Government consumption Subsidies (industries) Subsidies (imported commodities) Transfers to households Transfers to corporations Government saving (deficit) Total

253.0 58.4 −13,441.6 −300.6 214.6 2.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 −13,213.4

Expenditure

Revenue

0.0 0.0 0.0 0.0 −71.4 −10.2 0.0 −12,194.4 194.4 102.3 −101.0 −31.3 0.0 −12,111.6

SIM2 (CIT) 82.6 −0.5 −29.2 −4299.7 −76.8 −1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 −4324.7

Expenditure

Revenue

Expenditure

0.0 0.0 0.0 0.0 −76.8 −10.9 0.0 −4177.4 83.3 −0.6 −108.6 −33.6 0.0 −4324.7

SIM3 (PIT + CIT) 580.6 49.4 −13,599.1 −3317.9 −148.0 −1.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 −16,436.8

0.0 0.0 0.0 0.0 −148.0 −21.1 0.0 −16,373.2 278.0 101.8 −209.3 −64.8 0.0 −16,436.8

SIM5 (CIT) 0.0 0.0 0.0 0.0 214.6 30.6 0.0 2100.0 −99.5 116.4 303.4 94.0 −15,972.8 −13,213.4

8.5 1.8 19.7 −4682.9 7.7 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 −4645.0

SIM6 (PIT + CIT) 0.0 0.0 0.0 0.0 7.7 1.1 0.0 73.2 −4.3 3.2 10.9 3.4 −4740.2 −4645.0

261.8 60.3 −13,425.6 −4980.9 222.4 2.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 −17,859.1

0.0 0.0 0.0 0.0 222.4 31.7 0.0 2174.2 −103.8 119.6 314.6 97.4 −20,715.4 −17,859.1

Source: Model simulations.

due to loss of output, which inevitably leads to an even worse government fiscal deficit position. 5.2. Fiscal impacts As discussed earlier, the immediate impact of the PIT policy change (SIM1) is a decline in the government's revenue from personal income taxes. As shown in Table 4, under the balanced budget condition, PIT revenue declines by Rp13,576 billion. This is partially offset by the increase in the other tax revenues. For example, corporate income tax, indirect tax and import tariffs increase by Rp989 billion, Rp498 billion and Rp50 billion, respectively. The positive impacts on CIT, indirect tax, and import tariffs also indicate that there is an increase in production/ corporate profits, household consumption, and the level of imports. The immediate impact of the CIT policy (SIM2) is a decline of Rp4299.7 billion in corporate income tax revenues. The policy also has impacts on other government budget components. For example, PIT revenue decreases by Rp29.2 billion as a result of the decline in wages, which is required to keep the economy in full-employment. Indirect tax increases by Rp82.6 billion, reflecting the increase in household consumption. There is a slight decline in import tariffs of Rp0.5 billion, which indicates that the level of imports declines slightly. Under the combined policy (SIM3), the magnitudes of the impacts reflect the cumulative impacts of the two policies. The combined policies result in a Rp16,437 billion decline in government expenditure. Under the non-balanced budget condition, the level of deficit increases by Rp15,972.8 billion from the PIT policy and by Rp47 40.2 billion from the CIT policy. The combined policy (SIM6) results in a Rp20,715.4 billion increase in the government deficit. 5.3. Sectoral impacts Table 5 shows results for sectoral output and price effects resulting from the policy changes. Under the balanced budget policy condition,

outputs of all commodities increase, with the exception of government services which decline. The larger increases include wood (wood and wood products), mining (coal, ore and oil mining) and textiles (textile and wearing apparels), which increase by 0.680%, 0.598% and 0.564%, respectively. Under the CIT policy (SIM2), these commodities still remain in the top three in terms of increase in output but with a slightly different rank. Mining registers the largest growth with an increase of 0.554%, followed by wood and wood products with 0.377% and textiles with 0.339%. The changes in the rankings are due the fact that SIM2 is mainly driven by an export stimulus due to cheaper labour costs while SIM1 has the additional household demand on top of the export stimulus. The Restaurant sector is a typical example of this difference (0.47% in SIM1 and 0.025% in SIM2) as it is not traditionally an exporting sector. In terms of the price effects, the balanced budget policy condition effectively reduces all commodity prices. The decline in prices from the tax reforms is mainly due to the lower cost of factors of production due to lower demand for labour from the government sector in order to keep the budget balanced.

5.4. Impacts on poverty and income distribution Table 6 presents the effects of the tax reforms on poverty and income distribution. The table shows the levels of these indicators in two stages — before the policy (pre-simulation) and after the policy (post-simulation), as well as the change between the two stages. Poverty incidence is measured using the national statistical standard and income distribution is calculated using the Gini Index at the rural, urban and national levels. Under the PIT policy for the balanced budget condition (SIM1), the poverty incidence in the rural areas decreases by 0.06%, while it increases slightly by 0.01% in the urban areas. The effects in the two areas offset each other, resulting in an insignificant effect on poverty at the national level. On the other hand, income inequality as indicated by the Gini

H. Amir et al. / Economic Modelling 31 (2013) 492–501

499

Table 5 Sectoral impacts (% changes). Commodities/industries

Output

Balanced budget Food crops Other crops Livestock Forestry Fishery Coal, ore and oil mining Other mining Food, beverages and tobacco Textiles and wearing apparels Wood and wood products Papers, printing, transport equipment and metal Chemical, fertiliser and cement Electricity, gas and water supply Constructions Trade Restaurant Hotel Road transportation Air, water transportation and communication Transportation support services and warehouse Banking and finance Real estate and business Govt admin, defence, educ., health and social service Personal, household and other services

SIM1 (PIT) 0.202 0.378 0.496 0.483 0.331 0.598 0.279 0.354 0.564 0.680 0.477 0.374 0.373 0.337 0.397 0.476 0.252 0.410 0.447 0.417 0.456 0.517 −3.011 0.392

Non-balanced budget

SIM4 (PIT)

Food crops Other crops Livestock Forestry Fishery Coal, ore and oil mining Other mining Food, beverages and tobacco Textiles and wearing apparels Wood and wood products Papers, printing, transport equipment and metal Chemical, fertiliser and cement Electricity, gas and water supply Constructions Trade Restaurant Hotel Road transportation Air, water transportation and communication Transportation support services and warehouse Banking and finance Real estate and business Govt admin, defence, educ., health and social service Personal, household and other services

0.119 −0.282 0.401 −0.396 0.240 −1.437 −0.149 0.125 −0.662 −0.686 −0.132 −0.464 0.180 −0.106 0.022 0.453 −0.363 0.116 0.125 −0.140 0.239 0.199 0.556 0.387

Price

Output

−0.109 −0.098 −0.096 −0.083 −0.084 −0.068 −0.099 −0.086 −0.067 −0.077 −0.051 −0.060 −0.064 −0.065 −0.093 −0.094 −0.082 −0.078 −0.064 −0.080 −0.075 −0.071 −0.090 −0.063

SIM2 (CIT) 0.033 0.189 0.044 0.247 0.040 0.554 0.122 0.077 0.339 0.377 0.173 0.231 0.063 0.128 0.112 0.025 0.168 0.093 0.099 0.160 0.070 0.097 −1.043 0.013

Price

Output

Price

−0.117 −0.105 −0.103 −0.089 −0.091 −0.073 −0.107 −0.092 −0.071 −0.082 −0.055 −0.065 −0.069 −0.070 −0.100 −0.101 −0.088 −0.084 −0.069 −0.086 −0.080 −0.076 −0.097 −0.068

SIM3 (PIT + CIT) 0.236 0.567 0.541 0.730 0.372 1.156 0.401 0.432 0.904 1.059 0.650 0.606 0.436 0.465 0.509 0.501 0.420 0.504 0.547 0.578 0.526 0.614 −4.059 0.406

−0.226 −0.203 −0.199 −0.172 −0.175 −0.141 −0.206 −0.178 −0.138 −0.159 −0.106 −0.125 −0.134 −0.136 −0.193 −0.194 −0.170 −0.162 −0.133 −0.166 −0.154 −0.146 −0.187 −0.131

SIM5 (CIT) 0.327 0.294 0.288 0.250 0.253 0.204 0.299 0.258 0.200 0.230 0.154 0.181 0.194 0.197 0.280 0.281 0.247 0.235 0.192 0.241 0.224 0.212 0.272 0.190

0.009 −0.008 0.015 −0.015 0.013 −0.053 −0.006 0.008 −0.026 −0.030 −0.008 −0.018 0.005 −0.004 0.001 0.018 −0.014 0.005 0.003 −0.005 0.006 0.002 0.018 0.012

SIM6 (PIT + CIT) 0.012 0.011 0.010 0.009 0.009 0.007 0.011 0.009 0.007 0.008 0.006 0.007 0.007 0.007 0.010 0.010 0.009 0.008 0.007 0.009 0.008 0.008 0.010 0.007

0.128 −0.290 0.416 −0.412 0.254 −1.489 −0.155 0.133 −0.688 −0.716 −0.141 −0.482 0.186 −0.110 0.022 0.471 −0.377 0.121 0.129 −0.146 0.246 0.202 0.574 0.399

0.339 0.305 0.299 0.259 0.263 0.212 0.310 0.267 0.207 0.239 0.159 0.187 0.201 0.204 0.290 0.292 0.256 0.243 0.199 0.249 0.232 0.220 0.282 0.197

Source: Model simulations.

Index increases slightly in both rural and urban areas, resulting in a national level increase of 0.003%. In the CIT policy (SIM2), the poverty incidence in the rural areas decreases by 0.03%, while in the urban areas, there is virtually no change. In sum, at the national level, the poverty incidence reduces by 0.01%. However the Gini Index for this policy does not significantly change in both areas. In the combined policy (SIM3), the magnitude of the change for both indicators is the sum of the changes for both policy scenarios. Overall, at the national level, the poverty incidence decreases slightly to 14.14%, although income inequality increases by 0.003%. Under the non-balanced budget condition, in the PIT policy (SIM4), the poverty incidence decreases by 0.02% for both rural and urban areas, resulting in a 0.03% decrease at the national level. In the CIT policy (SIM5), the poverty incidence decreases slightly by 0.01% for all areas. In the combined policy (SIM6), the poverty incidence experiences a 0.03% decline for both areas and a 0.04% decline at the national level. In terms of income inequality, the implementation of the policies under the non-

balanced budget condition creates similar changes to those observed in the balanced budget condition for all the policy scenarios. Overall, the results show that under the balanced budget scenario, the policy reform reduces poverty more in the rural areas, while there is a slight increase in the urban areas. However, in the non-balanced budget there is no differential impact. Similarly, income inequality increases more in the urban areas then in the rural areas in the two cases. The relatively small impact on the poverty incidence and the slight increase in income inequality can be attributed to the fact that the tax policy is mostly beneficial to the households in the highest income groups. Any benefits to the lower income groups, including those below the poverty line, are indirect. 6. Conclusions and policy implications This paper has evaluated the effects of the Indonesian government's income tax reforms on the economy and the possible distributional effects on households. Three policy scenarios were simulated. The first

500

H. Amir et al. / Economic Modelling 31 (2013) 492–501

Table 6 Effects of income tax policy change on poverty and income distribution. Balanced budget

Variable

Non-balanced budget

PIT

CIT

PIT + CIT PIT

CIT

PIT + CIT

SIM1

SIM2

SIM3

SIM5

SIM6

SIM4

Pre-simulation: Poverty Rural (%) Urban National Income Rural inequality Urban National

17.35 10.72 14.15 0.308 0.345 0.349

17.35 10.72 14.15 0.308 0.345 0.349

17.35 10.72 14.15 0.308 0.345 0.349

17.35 10.72 14.15 0.308 0.345 0.349

17.35 10.72 14.15 0.308 0.345 0.349

17.35 10.72 14.15 0.308 0.345 0.349

Post-simulation: Poverty (%) Rural Urban National Income Rural inequality Urban National

17.29 10.73 14.15 0.310 0.349 0.353

17.32 10.72 14.15 0.308 0.345 0.349

17.26 10.73 14.14 0.310 0.349 0.353

17.33 10.70 14.12 0.310 0.349 0.353

17.34 10.71 14.14 0.308 0.345 0.349

17.32 10.69 14.11 0.310 0.349 0.353

Change: Poverty (%)

Rural −0.06 −0.03 −0.09 Urban 0.01 0.00 0.01 National 0.00 −0.01 −0.01 Income Rural 0.002 0.000 0.002 inequality Urban 0.004 0.000 0.004 National 0.003 0.000 0.003

−0.02 −0.01 −0.03 −0.02 −0.01 −0.03 −0.03 −0.01 −0.04 0.002 0.000 0.002 0.004 0.000 0.004 0.003 0.000 0.003

Source: Model simulations.

was an adjustment in the personal income tax rate structure, the second was an application of a single rate to the corporate income tax rate, and the third was a combination of the two policies. The long-run simulations were conducted under two different assumptions, with and without a balanced budget condition. It was shown that although both policy changes (i.e. reduction in PIT and CIT) result in a reduction in tax revenue, they both have a positive impact on economic growth under the balanced budget assumption. The drivers of growth come mainly from an increase in real consumption, an overall increase in sectoral output and increase in exports due to lower production costs. Both policy changes reallocate employment from the excess labour industry (government services) to other industries. The contraction of the government sector occurs as a consequence of maintaining the balanced-budget condition. The excess labour in the government sector (which is highly labour-intensive) is absorbed by other industries at a lower wage. Because in the long run the economy is at full-employment, this results in an increase in efficiency in production, which in turn allows industries to produce more at a lower cost. The outputs of export-oriented commodities increase due to an increase in domestic and international demand. The increase in industrial output also leads to increased investment. The tax reform was shown to have positive impacts on households with a slight decrease in poverty incidence particularly under the balanced budget condition. On the other hand, there is a small increase in income inequality as shown in the higher post-simulation Gini Index. The relatively small impact on the poverty incidence and the small increase in income inequality are due to the fact that the tax policy is mostly beneficial to the households in the highest income groups. The study results pose a number of implications for reducing poverty and improving income inequality in Indonesia. Firstly, there is a need for future tax cuts to be targeted towards the urban and rural poor. This can be done through an exemption of commodity tax for the commodities that are heavily consumed by the poor or by increasing the income tax threshold from the current Rp1.5 million to Rp2 million. Secondly, there is a need for such efforts to be accompanied by complementary policies to reduce poverty in the short run. Such policies could include adjusting the level of the income threshold to protect poor households

from paying more tax. In the short run, the government could consider direct cash transfers to poor households. However, in the longer term, a cash transfer policy is not appropriate as it could be treated merely as a government handout. A more sustainable approach to reducing poverty is for the government to accelerate its rural development policy. This should include providing better infrastructure in the agricultural sectors. Such an approach would be more effective in addressing poverty because the majority of the poor households live in the rural areas and work in the agricultural sectors. In conclusion, it is useful to consider the limitations of the study and ways in which it could be extended in future work. The 2008 Indonesian tax reform did not only involve adjusting the tax rates but also complemented with other policies such as modernization of the tax administration, simplification of the tax payment and reporting systems, and improving the capacity of tax officers. All of these measures were intended to increase the tax base and improve overall tax compliance. Our model did not attempt to incorporate the increased tax base and the improvement in tax compliance due to significant modelling challenges. However, this is something future research could address. Another limitation is that we developed a comparative static model in this study. Although useful for addressing “what if” policy questions, it would be desirable to develop a dynamic CGE model in the future to undertake more complex analysis. A dynamic model is able to give a better picture of how the impacts of policy changes evolve over time. For example, using such a model, it would be possible to evaluate more complex government policies such as a decision to invest and purchase capital that would contribute to capital accumulation and growth in the future. Appendix A

Appendix Table 1 Magnitude of PIT shocks for each household category (Rp million)—Sample. Household Effective Total Tax-free Taxable Tax categories rate (%) income threshold income paid — old MTR

HHR001 HHR014 HHR024 HHR081 HHR091 HHR092 HHR093 HHR094 HHR095 HHR096 HHR097 HHR098 HHR099 HHR100 HHU001 HHU005 HHU010 HHU057 HHU091 HHU092 HHU093 HHU094 HHU095 HHU096 HHU097 HHU098 HHU099 HHU100

Tax paid — new MTR

Magnitude of shock (%)

A

B

C

D

E

F

G = (F − E) / E ∗ A∗ 92.9%

0.00 0.04 0.46 1.80 2.71 2.83 2.95 3.08 3.22 3.60 4.01 4.44 5.10 8.14 0.00 0.05 0.53 1.86 4.62 4.80 4.98 5.18 5.41 5.66 6.52 7.40 8.62 12.42

7.12 15.82 18.65 41.42 54.67 56.93 59.58 62.71 66.41 71.24 77.78 86.47 103.84 169.35 10.29 15.85 19.10 41.36 86.61 90.70 95.14 100.55 107.60 116.59 128.39 144.26 174.51 288.94

15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60 15.60

0.00 0.22 3.05 25.82 39.07 41.33 43.98 47.11 50.81 55.64 62.18 70.87 88.24 153.75 0.00 0.25 3.50 25.76 71.01 75.10 79.54 84.95 92.00 100.99 112.79 128.66 158.91 273.34

0.00 0.01 0.15 1.33 2.66 2.88 3.15 3.46 3.83 4.60 5.58 6.88 9.49 24.69 0.00 0.01 0.17 1.33 6.90 7.52 8.18 8.99 10.05 11.40 14.45 18.41 25.98 61.92

0.00 0.01 0.15 1.29 1.95 2.07 2.20 2.36 2.54 3.35 4.33 5.63 8.24 18.06 0.00 0.01 0.17 1.29 5.65 6.27 6.93 7.74 8.80 10.15 11.92 14.30 18.84 38.34

0.00 0.00 0.00 −0.05 −0.67 −0.74 −0.83 −0.91 −1.01 −0.91 −0.83 −0.75 −0.62 −2.03 0.00 0.00 0.00 −0.05 −0.78 −0.74 −0.71 −0.67 −0.63 −0.58 −1.06 −1.54 −2.20 −4.39

Source: Authors' calculations.

H. Amir et al. / Economic Modelling 31 (2013) 492–501

501

Appendix Table 2 Estimate of the reduction in CIT payments due to the new single tax rate. Income range (Rp million)

Number tax payer

Old marginal tax rate Total tax payment (Rp million)

Avg. tax payment (Rp thousand)

Marginal tax rate (%)

Avg. taxable income (Rp thousand)

Marginal tax rate (%)

New marginal tax rate Avg. tax payment (Rp thousand)

Total tax payment (Rp million)

0–50 50–100 >100

83,959 19,053 39,235

117,456 154,936 83,016,151 83,288,543

1399 8132 2,115,870

10 15 30

13,990 70,879 7,111,233

28 28 28

3917 19,846 1,991,145

328,878 378,128 78,122,579 78,829,585

Source: Authors' calculation using 2005 Tax Return data from the Directorate General of Tax (DGT). Note: The change in CIT payment = (78,829,585–83,288,543) / 83,288,543 = −5.4%.

References Armington, P.S., 1969. Theory of demand for products distinguished by place of production. IMF Staff Papers 16 (1), 159–178. Auerbach, A.J., 1996. Measuring the impact of tax reform. National Tax Journal 49 (4), 665–673. Bourguignon, F., 2003. The growth elasticity of poverty reduction: explaining heterogeneity across countries and time periods. In: Eicher, T.S., Turnovsky, S.J. (Eds.), Inequality and Growth: Theory and Policy Implications. MIT Press, Cambridge, MA. BPS, 2009a. Data Strategis BPS. Badan Pusat Statistik, Jakarta. BPS, 2009b. Statistik Tenaga Kerja. Retrieved 8 February 2009, from http://www.bps. go.id/tab_sub/view.php?tabel=1&daftar=1&id_subyek=06¬ab=4. Brondolo, J., et al., 2008. Tax administration reform and fiscal adjustment: the case of Indonesia (2001–07). IMF Working Paper WP/08/129. Dixon, P.B., Rimmer, M.T., 2002. Dynamic General Equilibrium Modelling for Forecasting and Policy: a Practical Guide and Documentation of MONASH. North Holland, Amsterdam. Dixon, P.B., et al., 1982. ORANI: a Multisectoral Model of Australian Economy. NorthHolland, Amsterdam. Dollar, D., Kraay, A., 2002. Growth is good for the poor. Journal of Economic Growth 7 (3), 195–225. Easterly, W., Rebelo, S., 1993. Marginal income tax rates and economic growth in developing countries. European Economic Review 37 (2–3), 409–417. Eicher, T.S., et al., 2003. The impact of tax policy on inequality and growth: an empirical and theoretical investigation. In: Eicher, T.S., Turnovsky, S.J. (Eds.), Inequality and Growth: Theory and Policy Implications. MIT Press, Cambridge, MA. Engen, E., Skinner, J., 1996. Taxation and economic growth. National Tax Journal 49 (4), 617–642. Essama-Nssah, B., 2005. A unified framework for pro-poor growth analysis. Economics Letters 89, 216–221. Feldstein, M.S., 1973. On the optimal progressivity of the income tax. Journal of Public Economics 2, 357–376. Gemmell, N., 1988. Tax systems, tax revenue and growth in LDCs: a review of empirical evidence. Intereconomics: Review of European Economic Policy 23 (2), 84–90. Gillis, M., 1985. Macro and microeconomics tax reform Indonesia. Journal of Development Economics 19, 221–254. Goulder, L.H., Summers, L.H., 1989. Tax policy, asset prices, and growth: a general equilibrium analysis. Journal of Public Economics 38 (3), 265–296. Hill, H., 2000. The Indonesia Economy, Second ed. Cambridge University Press, Cambridge. Horridge, J.M., 2003. ORANI-G: A Generic Single-Country Computable General Equilibrium Model. Retrieved 22 April 2009, from http://www.monash.edu.au/policy/oranig.htm. Ikhsan, M., et al., 2005. Indonesia's new tax reform: potential and direction. Journal of Asian Economics 16, 1029–1046. Kraay, A., 2006. When is growth pro-poor evidence from a panel of countries. Journal of Development Economics 80, 198–227. Krause, A. D. (2005). Essays on the Normative and Positive Aspects of the Theory of Taxation: Tax Reform, Optimal Tax, and Tax Incidence. Unpublished Doctoral Dissertation, University of California.

Lee, Y., Gordon, R.H., 2005. Tax structure and economic growth. Journal of Public Economics 89, 1027–1043. Marks, S.V., 2003a. Personal Income Taxation in Indonesia: Revenue Potential and Distribution of the Burden. Bappenas, Jakarta. Marks, S.V., 2003b. The Value Added Tax in Indonesia: The Impact of Sectoral Exemptions on Revenue Potential and Effective Tax Rates. Bappenas, Jakarta. Martimort, D., 2001. Optimal taxation and strategic budget deficit under political regime switching. The Review of Economic Studies 68, 573–592. Mendoza, E.G., et al., 1994. Effective tax rates in macroeconomics: cross-country estimates of tax rates on factor incomes and consumption. Journal of Monetary Economics 34 (3), 297–323. Mirrlees, J.A., 1971. An Exploration in the Theory of Optimum Income Taxation. Review of Economic Studies. Nasution, A., 2002. The Indonesian economic recovery from the crisis in 1997–1998. Journal of Asian Economics 13, 157–180. Persson, T., Tabellini, G., 1994. Is inequality harmful for growth? The American Economic Review 84 (3), 600–621. Ravallion, M., 1997. Can high-inequality developing countries escape absolute poverty? Economics Letters 56 (1), 51–57. Ravallion, M., 2001. Growth, inequality and poverty: looking beyond averages. World Development 29 (11), 1803–1815. Ravallion, M., Chen, S., 1996. What Can New Survey Data Tell Us About Recent Changes in Distribution and Poverty? The World Bank, Washington, D.C. Samuelson, P.A., 1986. Theory of optimal taxation. Journal of Public Economics 30, 137–143. Sandmo, A., 1976. Optimal taxation: an introduction to the literature. Journal of Public Economics 6, 37–54. Son, H.H., 2004. A note on pro-poor growth. Economics Letters 82, 307–314. Son, H.H., Kakwani, N., 2008. Global estimates of pro-poor growth. World Development 36 (6), 1048–1066. Sørensen, P.B., 2007. The theory of optimal taxation: what is the policy relevance? International Tax and Public Finance 14 (4), 383–406. Stokey, N.L., Rebelo, S., 1995. Growth effects of flat-rate taxes. Journal of Political Economy 103 (3), 519–550. Timmer, C.P., 2007. How Indonesia connected the poor to rapid economic growth. In: Besley, T., Cord, Louise J. (Eds.), Delivering on the Promise of Pro-poor Growth. The World Bank & Palgrave Macmillan, Washington, D.C. World Bank, 2006. Making the New Indonesia Work for the Poor. The World Bank, Jakarta. World Bank, 2010. World Development Indicators. Retrieved 8 April 2010, from http://data. worldbank.org/data-catalog/world-development-indicators. Yusuf, A. A. (2007). Equity and Environmental Policy in Indonesia. Unpublished Doctoral Dissertation, The Australian National University. Yusuf, A.A., et al., 2008. AGEFIS: Applied General Equilibrium for FIScal Policy Analysis. Center for Economics and Development Studies, Department of Economics, Padjadjaran University, Bandung.