The Impact of a Fiscal Policy on the Macroeconomic Performance, Agricultural Sector and Poverty

1. METHODOLOGY

The impact on fiscal policy can be evaluated through computable general equilibrium model (CGE). The basic model will be used in the research is a combination of ORANI-F (Horridge et al. (1993), INDOF (Oktaviani, 2000), WAYANG (Wittwer, 1999) and ORANIGRD (Horridge, 2002). The modification of these models will be built to more focus on the macroeconomic policy change, which is emphasised on the fiscal policy change on poverty. Because the new model is different from the basic model, this model will be called Indonesian Poverty Model.

The impact of the reduction of fuel subsidies on the household welfare may channel through two ways: expenditure side and income side. The reduced subsidies on fuel, for instance, will push the price of fuel facing either industries for intermediate inputs or households for final demand. The type of households that suffers from such a shock is urban and rural low income households. Both households rely heavily on fuel (kerosene) for their cooking. This impact is very direct one, because households demand the fuel directly from the market. On the indirect side, some industries such as transportation sector and fertiliser industries demand fuel for their input factor. An increased price of fuel will impact of the demand for fuel in both industries and it is expected that the prices of transportation services and fertiliser will be higher than that of with subsidy. Again, low income household will experience a decline in their real income. Even low income rural households may have a worse decrease in their real income. Because in the fertiliser industry, approximately 30 % of total production cost is cost of fuel, subsidy cuts on fuel will definitively push the price of fertiliser to rise. The low income rural households whose income is mainly agricultural income encounter a difficult situation. They may reduce their demand for fertiliser due to the high price of fertiliser. However, it will lead to a decline in their agricultural output. Consequently their income may decrease. Or maintaining the current level of fertiliser application requires an additional input cost. It is estimated that their welfare may decline and the incidence of poverty may also rise.

The government decided to open the fuel market to the international market. This implies that the price of fuel at the domestic market will be subject to the fuel price fluctuations at the international level. However, the government realises the difficulties faced by the consumers when the domestic prices fluctuates following the price development at the international level. Figure 4.1. shows that to stabilise the fluctuation of the fuel price, the government provides amount of subsidy to maintain the price level at Pd (PwABPd). As the price at the international level rises and the government maintains the level of the domestic price, the government is required to increase the subsidy. Following hardship times faced by the government to finance the development reflected by the limited sources of income, the government faces a growing budget deficit every year. To reduce the budget deficit, the government decided to reduce the level of fuel subsidy and other subsidies (electricity, telecommunication and fertilisers). It is acknowledged that abolishing fuel subsidies at these times is impossible. Therefore the government reduces the level of subsidies periodically by setting up a new level of fuel price. In modelling issues, two possible simulations will be considered. The level of subsidies will be reduced by a required level, leaving the price level endogenous (the government provides reports on annual subsidy on certain products) or the price levels are adjusted periodically, leaving the level of subsidy endogenous. The later might be considered as a traditional approach since the direct target is the consumers.

Figure 4.1. The level of fuel subsidy

The transmitting of the subsidy cuts on the domestic and international price is exhibited in Figure 4.2. Following the reduction of fuel, the producers, e.g. farmers have to re-organise their production technology. The reduce their demand for fuel from xo to x1. This change in the fertiliser application will affect the level of output. Without no subsidy, the output is at the yo level. Because the price of fuel is higher than that of no subsidy, the current level of the output is y1. As a response to the reduced supply in the market, the current supply function is S’ with the current domestic price, PD1. It is seen that households in general face a higher output price than they did before. Given an unchanged nominal income, there will be a declining household real income. Assuming no export taxes on the agricultural sectors, a change in domestic prices will be channelled to the change in export price proportionally. Given a constant exchange rate, the export price is now higher than before.

Figure 4.2. The Impact of Subsidy Cut on Fuel on the Domestic and Export

Prices.

As the export prices rise, the country will lose their international competitiveness. To remedy such a problem, it may depreciate its currency in order to improve their competitiveness. If there is no depreciation of its national currency, those who are dependent on the export growth such as agricultural households will be badly hurt. As the decline in the demand for export occurs, the price of agricultural sector is expected to drop. As a result, more poverty incidence in the rural area will be observed.

As discussed earlier, there will be contracting sectors following the subsidy cut of fuel. Figure 4.3 shows the impact of contracting sectors on the demand for labour and the wage level. As the sectors contract (Q1 to Q2), the demand for labour will be declined from N1 to N2. Another consequence is that the wage level will be lower. The decline in the demand for labour in the contracting sectors implies that some households will be losing their income. Their household’s welfare even becomes worse as the income from that sector is the main source of income. As we see industries largely operate in the urban area. Therefore urban households will be encountering the decline in their welfare not due solely to the decline in their income but also due to the increase in the prices of some outputs.

Figure 4.3. Output and the Demand for Labour.

4. 1. Sector Aggregation

The model will not use all sectors, which are available in the current Indonesian Input-Output Table. Sectors specified in the model can be seen in Table 2.3. The aggregation of sectors as described above refers to the objectives of this research. One of the research objectives is to examine the impact of the subsidy reduction of fuel on the macroeconomic parameter. Therefore, refinery oil, fertilizer and transportation sectors are explicitly disaggregated. Explicitly disaggregating the transportation sector is due to the fact that this sector demands fuel. Any increase in the price of transportation cost will have a domino impact of the price of other commodities. According to the Central Bureau of Statistics transport cost or the price transportation means contributes to the inflation rate. Therefore, examining the impact of the subsidy reduction of fuel requires such a sectoral disaggregation. In addition the number of sectors is considered to be manageable and relevant to the research objectives.

4.2. Model Structure

The notation system in the model parallels the ORANI-F (Horridge, 1997), INDOF (Oktaviani, 2000), WAYANG (Wittwer, 2002) system. It was established by the ORANI modelers to be both efficient in use and easy to interpret. The system of notation uses lower case letters for names of variables in the linear equations and upper case for the coefficients of these equations. Upper case is also used to refer to other coefficients and parameters and for the names of sets of items. Lower case is used for the indices of items, which form a set. Finally, in presenting the model, it is often convenient to present the underlying non-linear equations themselves, and thus in terms of the levels of variables rather than percentage changes. Variables in this form are written in upper case but in bold. Names may be shown in italics or non-italics, with no change in meaning, as convenient for presentation.

Sets and indices

Variables and coefficients often apply to sets of items. There are nine major sets, all with upper case names. The sets and the indices, which are generally used to refer to an item from the sets, are as follows:

industries (IND) i

source (namely domestic or foreign) (SRC) s

occupation (OCC) o

commodities (COM) c

household (HH)h

There are four sets defining subsets of commodities:

margin type commodities (MAR)m

(ie., commodities facilitating flows of other commodities)

non-margin type commodities (NONMAR)n

traditional export commodities (TRADEXP)c

non-traditional export commodities (NTRADEXP)c

Finally, there is a sequence set:

a sequential list of integers (YEARS)t

Variables

Each variable name is alphanumeric. The generic form of a name is "basicnamesubscripts_sets". Here "basicname" is the primary indicator of the meaning of the variable. It follows a general structure as described below. Often a variable is applicable to a set of items. To refer to the individual item from a set, one or more subscripts specifying the element are added to the "basicname". For example, "basicnamecs" refers to a variable applying to a commodity c from a particular source s. Rather than identifying a particular individual item by its index, an individual items may itself be named. For example, a commodity can be sourced from domestic or foreign sources. For referring to a variable pertaining to, say, the domestically sourced commodity, "basicnamec"dom"" is sometimes used, where dom is the name of the source. The item name, enclosed in "", is used as the subscript. Finally, some variables are aggregates or averages over the various items in a set. This type of variable is indicated by appending an underscore character (as a subscript) with one or more following indices identifying the set or sets of items over which the aggregate or average is formed. For example "basicnameci_s" refers to a variable applying to a commodity c in a particular industry i and it has been formed by aggregating or averaging over all sources

The basicname component of the names of most variables consists of two or three parts, as follows:

1) The type of variable is indicated in a letter or letters. The main ones are:

atechnical change

delordinary or absolute (rather than percentage) change

fshift variable

pprice (000 Rp)

pfprice, foreign currency

ttax

wmonetary value( 000 Rp)

x quantity

Additionally, a number of minor variable types have relatively limited application. These are

employemployment

levactual level rather than change of variable value

phiexchange rate

qnumber of households

utilityhousehold utility level

2) The digits 0 to 6 indicate the activity or purpose of use.

1current production

2investment

3consumption

4export

5other (government)

6inventories

0either all users, or user distinction irrelevant

The inclusion of the digit, even when the particular use is irrelevant (the 0 digit) to the variable serves to separate the alphabetic characters from components (1) above and (3) below.

3) Three or more letters forming a descriptor for the variable. The descriptor adds clarity of meaning to a name where necessary.

accumaccumulation

basbasic, not including margins or taxes

capcapital

cifimport at border price

impimports (duty paid)

lab labor

lndland

luxlinear expenditure system (supernumerary part)

marmargins

octother cost tickets (e.g., other miscellaneous production

cost)

primprimary factors as a group (land, labor or capital)

purat purchasers’ prices

sublinear expenditure system (subsistence part)

tartariffs

taxindirect taxes

tottotal or average over all inputs for some use or activity

Variables used in the study are listed in Table 4.1.

Table 4.1. List of Variables Used in the Study

COM x SRC

Shifter for longrun investment rule

Source: Horridge et al. (1993), Wittwer (2002) and Oktaviani (2000), Horridge (2002) adapted and appended by the writer.