Download - Impacts of CPO-Export Tax on Several Aspects of Indonesian CPO Industry

1

IMPACTS OF CPO-EXPORT TAX ON SEVERAL ASPECTS OF INDONESIAN CPO INDUSTRY

Wayan R Susila*****

* Indonesian Institute for Estate Crops,Bogor, Indonesia.

Impacts of CPO-Export Taxon Several Aspects of

Indonesian CPO Industry

INTRODUCTION

Like rice, crude palm oil (CPO) isconsidered a strategic commodityin the Indonesian economy. Firstly,as a raw material for the maincooking oil consumed inIndonesia, its price plays animportant role in determining theinflation rate of the Indonesianeconomy (Badan Pusat Statistik,2001; Arifin and Susila, 1998;Amang, 1995). Secondly, palm oilindustries provide employment formore than two million people(Direktorat Jenderal Bina Produksi

Perkebunan, 2003). Thirdly, it is asource of foreign exchange earningas its export has exceeded morethan USD 1 billion since 1997.While the area of land cultivatedwith oil palm in 1988 was 862 900ha, it sharply increased to around4 million hectares in 2002,implying an 11% per annumgrowth rate. As a result, theproduction of CPO increased byaround 10% per annum, from342 700 t in 1988 to 7.97 milliontonnes in 2002 (Direktorat JenderalBina Produksi Perkebunan, 2003).Similarly, the use of CPO as a raw

ABSTRACTABSTRACTABSTRACTABSTRACTABSTRACT

To control domestic supply and price of crude palm oil (CPO) andcooking oil, the government of Indonesia has imposed CPO-export taxsince August 1994. As the CPO industry plays an important role inIndonesian economy, the imposition of the tax is expected to havesubstantial impacts on various aspects of the industry, such as oninvestment, production, trade, farm income and welfare distribution.The main objective of this study is to assess these impacts using aneconometric model of the industry. The results of the study reveal thatthe export tax policy has inhibited the growth rate of investment,production, export and farm income. On the other hand, this policyhas been an effective instrument to control domestic CPO and cookingoil price. Moreover, this policy has caused a substantial welfare transferfrom producers to consumers and the government. To compromise theseconflicting impacts, an alternative CPO tax formula is also proposedin this paper.

2

OIL PALM INDUSTRY ECONOMIC JOURNAL (VOL. 4(2)/2004)

material for cooking oil sharplyincreased by around 10% perannum.

The CPO industry is expectedto play a greater role in theinternational market for oils andfats. Basiron (2002), Sultoni andSusila (1998) and Pasquali (1993)projected that the growth rate ofCPO production would be thefastest among edible oils. CPO hasbeen predicted to over takesoyabean oil as the biggest oiltraded in the world oil and fatmarket. The market developmentof CPO will even be faster becauseof the success of the UruguayRound (Susila et al., 2002; Barton,1993).

Considering its importance,Indonesia has launched policies tooptimize the development of theindustry, such as supplystabilization. The most importantpolicies have been on export tax,first implemented in August 1994to stabilize and secure the domesticsupply and price. The export taxrate when first introduced waslinearly related to the CPO price,ranging from 40%-60% of thedifference between the CPO priceand minimum export price to betaxed. From 4 July 1997 toFebruary 1998, the rate waschanged to 5% of the CPO price.Due to a sharp increase in the CPOprice and substantial depreciationof the rupiah, the governmentbanned export in the beginning of1998. Then, this policy wasreplaced by an export tax of 60%which has since been graduallyreduced to 30% in July 1999 andapproximately 4% in 2002.

This policy is expected to havea substantial impact on variousaspects of the industry, such asinvestment (area), production,consumption, trade, domesticprice, added value, farm incomeand welfare distribution. On theother hand, the magnitude anddistribution of the impacts willcolour the future of the industry.

Therefore, an estimation of themagnitude and distribution ofthese impacts is important in aneffort to formulate an appropriateexport tax rate relatively fair toproducers, consumers and thegovernment.

This paper has two mainobjectives, namely: (1) to assess theimpacts of CPO-export tax onIndonesian CPO industry, coveringimpact on domestic price,investment (area), production,consumption, export, employment,added value, cooking oil price,government revenue, producersurplus and consumer surplus; and(2) to propose an alternative CPO-export tax formula which isrelatively fair, either from produceror consumer point of view.

RESEARCH METHOD

Theoretical Model and ModelSpecification

Simulation approaches on theeconometric model of the industrywere used to assess the impacts ofCPO-export tax on various aspectsof the industry. The econometricmodel developed in this paper isbasically a modification of a modelpreviously developed by Susila etal. (1995). The main modelmodifications are a re-specificationof the model, level of aggregationand the use of a simultaneousequation system approach for allequations (the previous modelused single equation for eachcountry and simultaneous equationsystem for the world market).

The use of a simultaneousequation system approach isexpected to yield better estimatesbecause this approach is consideredmore appropriate in dealing with asystem of commodity market inwhich some variables aresimultaneously related or inter-dependent (Koutsoyiannis, 1977;Pindyck and Rubinfield, 1987).Due to this advantage, various

studies used this approach forcommodity market modelling andpolicy analysis. Dradjat (2003)used this approach to assess theimpacts of trade liberalization onIndonesian rubber, cocoa, coffee,and tea industry. Abidin (2000)and Ernawati (1997) used thisapproach to estimate the impact oftrade liberalization on Indonesiansugar industry. Finally, Zulkifli(2000) used this simultaneousequation system to assess theimpact of CPO trade liberalizationon Indonesian CPO industry.

Theoretical model of CPOmarket was specified to consist of10 blocks/subsystems, that are,four blocks of main producing/exporting countries (Malaysia,Indonesia, Nigeria, and LatinAmerica), four blocks of mainconsuming/importing countries(West Europe, China, Pakistan,and Egypt), one block of the restof the world, and one block of theworld CPO market. Theproduction share of the fourproducing countries has beenaround 86% of world production,while the consumption share of thefour consuming countries has beenaround 72%. This implies that thelevel of aggregation is consideredto be relatively justified.

A simplified theoretical modelby assuming the model consistingof one block exporting country(Indonesia), one block importingcountry (West Europe), and theworld market block, is illustratedin Figure 1. Figure 1 shows thehypothetical relationships betweenvariables in the model.

As seen in Figure 1, Indonesiablock consists of seven equationsas follows:

INPOAt = a0+a1RPORBP+a2INE45+a3INI45+a4D0+a5INPOAt-1+U1...........(1)

INPOQt = b0+b1INPOPt+b2INPOAt+b3T+U2......................................................(2)

INPOCt = c0+c1INPOPt+c2 INYt+c3D2+c4 D3+U3....................................(3)

3


INPOXt = d0+d1(1-INTAXt)WDPOPt*INEt+d2INPOQt+d3INPOXt-1+U4....(4)

INPOSt =e0+e1(WDPOPt-WDPOPt-1)+e2INPOQt+e3INPOCt+e4INPOSt-1+U5....................................................(5)

INPOMt = INPOCt+INPOXt+INPOSt-INPOQt-INPOSt-1............................(6)

INPOPt = f0+f1(1-INTAXt) WDPOPt*INEt+f2INPOQt+f3INPORPt-1+U7........(7)

INPOA = oil palm mature area ofIndonesia (1000 ha)

INPOQ = palm oil palm productionof Indonesia (1000 t)

INPOC = palm oil consumption ofIndonesia (1000 t)

INPOX = palm oil export of Indonesia(1000 t)

INPOM = palm oil import of Indonesia(1000 t)

INPOS = palm oil stock of Indonesia(1000 t)

INPOP = domestic price of palmoil (Rp/kg)

RPORBP = [(1-INTAXt-4)WDPOPt-4+(1-INTAXt-5) WDPOPt-5]/(INRBPt-4+ INRBPt-5)

= price ratio of palm oil and rubberwith time lag four and five years

WDPOP = world palm oil price(USD/t)

INTAX = CPO export tax (%)

ING = Indonesia gross domesticproduct (USD million)

INI = Indonesian interest rate (%per annum)

INN = Indonesian population(million)

INE = Indonesian exchange rate(rupiah/USD)

INY = Indonesian income percapita (USD/capita)

WDRBP = world rubber price(USD/t)

INRBP = domestic rubber price(Rp/kg)

T = vintage of crops which have

optimal production or trend ofgovernment support

D0 = dummy variableD = 0 before1979D = 1 after 1979 to 2010

D0 represents the governmentpolicy to promote oil palmdevelopment (Perusahaan IntiRakyat project and credit subsidy)

D2 = dummy variableD2= 0 before 1980D2 = 1 after 1980 to 2010

D2 represents the governmentpolicy to control domestic supplythrough production allocationpolicy (domestic allocation policy)

D3 = dummy variableD3 = 0 before 1991D3 = 1 after 1991 to 2010

D3 repersents domestic tradeliberalization policy

Note: subscript 45 means sum ofvariable at time lag 4 and 5example: INPOP45 = INPOPt-4 +INPOPt-5

Figure 1. General theoretical model of crude palm oil (CPO).

GNP Popula- tion

Consumption

Import

Stock

Export

Tax

Domestic

Price

Production

Mature Area

Government

Policy

Interest

Rate

Other Crop

Prices

Exchange

Rate

T

Export

World Import

World

Price

World

Stock

World

Export

World

Production

World

Consumption

GNPPopula-

tion

Consumption

Substitute

Cooking Oil

PricesImport

Stock

Exchange

Rate

Export

Note:

= Endogenuous

= Exogenuous

Substitute

Cooking Oil

Prices

BLOCK OF INDONESIA BLOCK OF WORLD BLOCK OF EEC

4


Equation (1) shows that maturearea of oil palm plantation isaffected by the price ratio of palmoil and rubber, exchange rate,interest rate, government policyrelated to palm oil development(D0), government policy onliberalization of edible oil domesticmarket, and previous mature area.Moreover, since time lag betweeninvestment (planting activities)and mature area is around four tofive years, then this time lag is usedin Equation (1). The role of theCPO-export tax will be captured inthe model as the producer price isdefined as the world price minustax. The relationship betweenexchange rate and mature area isexpected to be positive, while thatof mature area and interest rate tobe negative. The governmentpolicies (D0) to support oil palmdevelopment through PerusahaanInti Rakyat (PIR) project , a kindof nucleus estate smallholderproject, and credit subsidy toprivate estates since 1979, areexpected to be positively related todevelopment of the oil palm area.

Equation (2) indicates thatproduction is determined by thecurrent mature area, CPO price,government policies, and exporttax rate. The CPO price, maturearea, and government policies areexpected to be positively related toproduction. On the other hand,the tax rate is expected to benegatively related to production,since tax depresses the CPOdomestic prices.

Like most consumptionequation, consumption of CPO isexplained by domestic price, priceof its substitute products (coconutoil), gross domestic product(GDP), and population (Equation3). The higher the GDP, numberof population, and substitutingproduct prices, the higher will bethe CPO domestic consumption.On the other hand, the higher theCPO domestic prices, the lower

will be the CPO domesticconsumption. The governmentpolicies to control CPO export byforcing the producers to allocatepart of their production todomestic market since 1980 orknown as domestic allocationpolicy (D2) and domestic tradeliberalization policy launched in1991 (D3) are expected to have apositive impact on consumption.

Export is expected to bepositively related to the CPO worldprice, production, pervious export,exchange rate, and negativelyrelated to export tax rate (Equation4). Assuming that stockmanagement is part of speculativeactivity, the volume of stock isexpected to relate positively toprice difference between currentprice and previous period price andproduction. Moreover, stock istheoretically related toconsumption and previous periodstock (Equation 5).

In the Indonesian block, importis assumed to be a residual variableas the function of Indonesianimport is just to fulfil the deficit indomestic market. To accommodatethis phenomenon, the stockequation is represented by residual-identity equation (Equation 6).

Domestic price is basically amarket-integrated approachimplying that the domestic price isstrongly influenced by the worldprice (Equation 7). As seen in theequation, the domestic price isnegatively related to the tax rateand production and positivelyrelated to exchange rate andprevious period price.

With various adjustments toaccommodate specific characteristicsof each country/block, all sevenequations can be applied to eachcountry. For example, area andproduction equation will beeliminated in West European blocksince it is not a producing country.Moreover, West Europe has noexport tax, but persistently

imposes import tax to protect itssunflower oil industry. Thus, allseven equations could be used asprototype equation in each block,except the world market block.Model specifications and theresults of estimation for each blockare reported in Susila et al. (2002).

Except for price equation, allequations in the world marketblock are summation of its relevantvariables (Equation 8-Equation 13)below. For example, Equation (8)shows that world or total maturearea is summation of mature areaof all producing countries.

WDPOAt = INPOAt+MLPOAt+ NIPOAt+LAPOAt+CHPOAt+RWPOAt.........(8)

WDPOQt = INPOQt+MLPOQt+NIPOQt+LAPOQt+CHPOQt+RWPOQt.........(9)

WDPOCt = INPOCt+MLPOCt+NIPOCt+ L A P O C t+ E C P O C t+ C H P O C t+PKPOCt+EGPOCt+RWPOCt.........(10)

WDPOXt = INPOXt+MLPOXt+LAPOXt+ECPOXt+RWPOXt......................(11)

WDPOMt = INPOMt+MLPOMt+NIPOMt+LAPOMt+ECPOMt+CHPOMt+PKPOMt+EGPOMt+RWPOMt.....(12)

WDPOSt = INPOSt + MLPOSt + NIPOSt+LAPOSt+ECPOSt + CHPOSt + PKPOSt+EGPOSt+RWPOSt.......................(13)

WDPOPt = ae0+ae1WDPOSt-1+ae2WDPOMt+ae3WDPOXt+ae4WDPOPt-1 ................................(14)

where:

1. The first two letters of the labelon the dependent variablerepresent country or world(IN: Indonesia, ML: Malaysia,NI: Nigeria, LA: LatinAmerica, CH: China, EC:European EconomicCommunity, PK: Pakistan, EG:Egypt, RW = rest of the worldcountry, WD: world);

2. The next two letters of thelabel represent the commodity(PO = palm oil); and

5


3. The last letter of the labelsrepresents variable (A: matureArea, Q: production, C:consumption, X: export,M: import, S: stock, P: Price)example: WDPOX: world CPOexport, MLPOQ: palm oilproduction of Malaysia.

The CPO world price equationis specified as a behaviouralequation, not as a market clearingequation. Following this, theworld price is expected to bepositively influenced by worldimport, price of other oils (soybeanoil and sunflower oil), movingaverage effects (WDPOP15).Moreover, world export andprevious stock are expected tonegatively influence world price(Equation 14).

Model Identification, Estimation,and Simulation

Model identification used inthis study was order condition.With 52 endogenous variables(equations), 73 pre-determinedvariables, and around 3-10explanatory variables in eachequation and using ordercondition, we know that the modelis definitely over-identified.Insights learned from previousstudies suggest that the use of rankcondition will end up with thesame conclusion as that of ordercondition. Therefore, rankcondition was not applied in thisstudy.

Given that the model was over-identified, 2SLS method ofestimation was applied.Koutsoyiannis (1977) stated thatunder the circumstance of theexistence of model misspecification,missing of relevant variables,multicolinearity and autocorrelationerror, 2SLS tends to yield morerobust estimates. Moreover, 2SLSmethod is arguably the simplestmethod among methods suited toover-identified model.

Based on previous export taxrates, four scenarios associatedwith the tax rates were analysed inthis study, namely:

Scenario I is the basicscenario, that is a scenarioin which the governmentwould not have imposedthe CPO export;

Scenario II is conducted toevaluate the impacts ofexport tax on variousaspects of Indonesian CPOindustry since itsimplementation (1994-1999). This scenario,therefore, is based on theimposition of actual taxrate as the government haddecided in the 1994-1999periods. The differencesbetween the results ofScenarios I and II are theimpacts of the export taximposition on the industryin the 1994-1999 periods;and

Scenarios III, IV, and V isused to predict the impactsof export tax on variousaspects of Indonesian CPOindustry using timehorizon of the year 1999-2010. Scenario III is basedon an assumption that theexport tax rate in that timehorizon would be 40%.Using similar approach,Scenarios III and IVrepresent 40% and 60% taxrate, respectively. Thedifferences between theresults of these scenariosand Scenario I are theimpacts of the export taximposition on the industryin the year 1999-2010periods, respectively.

On the basis of the magnitudeand distribution of the impacts, theeffectiveness of the export taxpolicy was evaluated. Based onthis evaluation and various factors

related to the consumer andproducer, such as the number offarmers/consumers consumption/income share, and theory ofsecondary right, an alternativeexport tax rate was formulated.

Data Sources and Descriptions

In general, there are two groupsof data used in this study, namely,palm oil and macro-economicrelated data. The first group of dataconsists of oil palm mature area,production, consumption, export,import, stock, and edible oil prices.Data sources for these are Oil World(1987-2002), Direktorat JenderalBina Produksi Perkebunan, StatistikKelapa Sawit (1991-2003) andBadan Pusat Statistik (2001).Macro-economic related dataconsist of various data such aspopulation, income per capita,gross domestic product, exchangerate, and interest rate. The mainsource of these data is InternationalMonetary Fund (1986-2002).

For future values of explanatoryvariables used in the scenario, twoapproaches were applied. If thefuture values (predicted values) areavailable, such as for populationand gross domestic product thatare provided by FAO or IMF, thenthese data were used in thescenario. Otherwise, some timeseries approaches, such as movingaverage or autoregressive modelswere applied to estimate thepredicted value of the explanatoryvariable.

RESULTS AND DISCUSSIONS

Results of Model Estimates

The results of estimates aregenerally in accordance withtheories/hypotheses. This can beseen from the sign and magnitudeof estimates, which are generally asexpected. Moreover, the modelestimates are also fairly robust,

6


indicated by 25 equations out of 39estimated equations (13 identityequations do not need to beestimated) have coefficient ofdetermination larger than 90%.Only nine equations havecoefficient of determination lowerthan 75%. Moreover, with acriterion of root mean squarepercentage error to validate themodel, it can be concluded that themodel is adequate for simulationmodelling. A more comprehensivediscussion on model validation canbe found in Susila et al. (2002).

Equation (15) shows that thereare four explanatory variables thatsignificantly influenced investment(mature area). They are price ratioof CPO and rubber, D2, D3, and T.These four variables can explainaround 92.65% of mature areavariation. The domestic allocationpolicy (D2) had a negative impacton area development whileliberalization of CPO trade inIndonesia (D3) had supported thedevelopment of oil palmplantation. Moreover, variable Tthat also represents the trend ofgovernment support policy has asignificant and positive impact onpalm oil plantation in Indonesia.

INPOAt = -761.05 + 345.33 RPORBP(0.044) (0.286)

226.77 D2 + 321.73 D3(0.070) (0.068)

+ 62.85 T .......................(15)(0.001)

R2 = 92.65%

The price ratio of CPO to rubber,which is significant in Equation(15) indicates that there has been acompetition in resource (land) usebetween oil palm and rubber. Thecompetition has occurred mainly inSumatera and Kalimantan, whichare traditionally known as rubberand oil palm plantation centres. Thepositive sign of the price ratiocoefficient indicates that theincrease of CPO price or a decreasein rubber price will increase oil

palm plantation with four to fiveyear time lag.

Production behaviour can simplybe explained by mature area and timetrend (T). These two variables canexplain around 98% of productionvariation. CPO price has aninsignificant effect on production, acommon characteristic of productionresponse of estate crops. In otherwords, price changes had notresponded to production changes, butby investment decision (areaexpansion). T in Equation (16)mostly represents an increase in yielddue to plantation composition basedon plantation age (vintage). Theexpansion of oil palm since 1970 hascaused a continuous increase ofvintages which are in maximum yield(Susila, 1997).

INPOQt = -521.49 + 1.96 INPOA (0.003) (0.000)

77.77 T .........................(16)(0.000)

R2 = 98.41%

As expected, consumption of CPOhas been positively related to incomeper capita as most domestic use ofpalm oil is for cooking oil (Equation17). Moreover, government domesticallocation policies (D2) and tradeliberalization policy (D3) also hada significant impact on the increaseof CPO domestic consumption. Byusing domestic allocation policy,the CPO producers were forced toallocate part of their production todomestic market. Therefore,government has a control todomestic supply and price that hada positive impact on domesticconsumption. Trade liberalizationpolicy was aimed at improvingdomestic market efficiency byreducing the role of the statetrading company (BULOG) and topromote participation of privatecompanies. This policy has apositive impact on domesticconsumption. Moreover, theconsumption of CPO has not beensignificantly affected by its price as

cooking oil is considered as a basicneed.

INPOCt = 16.47 0.000024 INXP(0.828) (0.892)

+ 0.66 INY + 359.15 D2 +(0.000) (0.001)

253.25 D3 ....................(17)(0.105)

R2 = 96.62%

Equation (18) shows thebehaviour of Indonesian CPOexport that is mainly influenced byCPO export price and production.A 1% increase of export pricecauses a 0.253% increase in export,implying that export is inelastic toprice. Since export tax isnegatively related to export price,increases in export tax decreasesvolume of export. The role ofproduction in export equationindicates that international markethas been the main destination ofIndonesian CPO production. Onthe other hand, Indonesian stockhas been relatively small, due tohigh risk and cost of holding stock,compared to its potential profitgain.

INPOXt = 24.78 + 0.00044 INPOPXT +(0.233) (0.033)

0.42 INPOQ 0.23 INPOXt-1(0.011) (0.414)

+ 191.53 D3 .................(18)(0.437)

R2 = 96.62%

The estimates of stock andimport are not as robust as theprevious equations. As an example,signs of the estimates of stockequation are consistent withhypotheses, but are notsignificantly explained through thevariation of stock. Yet, importequation can only explain by itsmoving average term implying thatthere is no variable that canexplain the behaviour of import,except for some previous values ofimports. This can be justifiedbecause the import volume hasbeen relatively small (Indonesia is

7


one of the main CPO exporters) sothat import is just a residualvariable.

INPOMt = 35.52 + 0.63 INPOME ..(19)(0.235) (0.030)

R2 = 96.62%

INPOSt = 24.78 + 0.00044 INPOPXT(0.235) (0.033)

+ 0.42 INPOQ 0.23 INPOXt-1(0.011) (0.414)

+ 191.53 D3 ..................(20)(0.437)

R2 = 96.62%

The world CPO price can wellbe explained by world current andprevious stock, import, export, andmoving average effect (Equation21). All these four variablesexplain around 90% of world pricefluctuation. As expected, worldstock and export are negativelyrelated to price while import ispositively related to price.

WDPOPt = -422.82 - 0.11SWDPOS(0.000) (0.002)

+ 0.37WDPOM(0.001)

0.30WDPOX(0.003)

+ 0.76WDPOP15(0.000)

+ 0.45WDSBSFP .......(21)(0.000)

R2 = 90.52%

The magnitudes of priceflexibility which are relatively highare as expected. Price flexibility ofimport and export are 18.08 and18.37, respectively. This impliesthat a 1% import/export increasewill cause more than 18% increase/decrease in price. This is consistentwith the fact that the CPO price hasbeen highly volatile withcoefficient of variation of around34%. As an example, while theaverage CPO price in 1998 wasUSD 550/t, it sharply decreased toUSD 220/t in 1999. This sharpprice decrease was triggered by anincrease in soyabean oil productionin main CPO importing country

(the USA), causing a decrease inCPO import

Evaluation and Projection of theImpacts of CPO-Export Tax

Since the policy implementationin August 1994, this export taxpolicy has had significant impacton the industry. Within the timehorizon 1994-1999 when theeffective tax rate was around13.33%, the mature area of oil palmplantation had been reduced by2.56% per annum or around 37 000ha per annum (Table 1). Thisindicates that this policy had a sub-stantial negative effect oninvestment in the industry. As aresult of this negative investmenteffect, CPO production had alsobeen depressed by the policy. It isestimated that the policy hadcaused a loss of around 0.81% ofthe total production or around36 000 t CPO per annum.

The most devastating impact ofthe policy had been on the exportand farm income. During that timehorizon, this policy had causedexport to be 6.02% lower comparedto the no export tax scenario. Thisimplies that Indonesia hadsacrificed her export of about

147 000 t annually. Similarly, thepolicy had caused the farm incometo be lower by around 11.35% oraround Rp 400 000/ha/yr; asubstantial loss for farmers.

On the other hand, this policyhad been proven to be effective incontrolling domestic cooking oilprice. With this policy, thegovernment had been successful tokeep the cooking oil price downwhen the world CPO priceincreased or when rupiah wassubstantially depreciated. Usingthis policy, the government hadkept the cooking oil price to be7.77% or Rp 184/kg lower than itshould be. Moreover, from thegovernment point of view, asignificant tax revenue, estimatedaround Rp 5241 billion, was alsoconsidered to be a positive resultof the policy.

For the time horizon of 2000-2010, impacts of three export taxrates, namely 20% (4.78%effective), 40% (9.55% effective),and 60% (14.33% effective) wereforecasted. The impacts of thesetax rates are summarized in Table2. In term of mature area, anincrease of 1% effective tax ratedecreases mature area by around0.15%, which is equivalent to a

TABLE 1. EVALUATION OF THE IMPACTS OF CPO-EXPORT TAX(1994-1999)

Impacts of Impacts ofExplanations Unit Mean export tax 1% increasing

implementation of EET(%) (%)

Mature area 000 ha 1 444.86 -2.56 - 0.19Production 000 t 4 483.95 -0.81 - 0.06Export 000 t 2 371.03 -6.20 - 0.47CPO price Rp/kg 1 524.54 -8.58 - 0.64Cooking Rp/kg 2 366.85 -7.77 - 0.58oil priceFFB price Rp/kg 342 -8.58 - 0.64Gross Rp/ha/yr 3 512 116 -11.35 - 0.85margin

Note:EET: effective export tax.

8


decrease of 3145 ha. For example,if the tax rate imposed in year1999-2010 is 4.75% effective taxrate, then the mature area will belower by around 0.76% or around16 000 ha per annum. This impliesthat the imposition of the exporttax has significantly depressed thedevelopment of oil palm plantationin Indonesia.

In addition to its negativeimpact on mature area, the exporttax has also depressed production,although at a relatively lower rate.In general, a 1% increase ofeffective tax rate is projected todecrease production by 0.09% perannum or around 5700 t perannum. As seen in Table 2, if thegovernment imposed a 14.33%effective rate, production decreasesby around 84 000 t per annum.

The negative impact of thispolicy is more substantial in termof export volume. A 1% increaseof the tax rate decreases Indonesianexport by around 0.41% or around14 000 t per annum. If forexample, the government imposesa 9.55% effective tax rate, then theexport loss will be around 133 000t per annum. This indicates thatthis policy could be an effectivepolicy instrument to controlsupply in domestic market.

On the other hand, this policyhas provided substantial benefit to

consumers. Table 3 shows that theimplementation of this policy hascaused domestic CPO and cookingoil price to be lower than theyshould be. For example, anincrease of a 1% effective tax ratedecreases CPO price by 1.13% orRp 26.89/kg. If the governmentimposes an effective export tax of14.33%, then the domestic pricewill be about Rp 360/kg lowercompared to that without exporttax.

As a consequence of CPOdomestic price being depressed bythe tax, the price of cooking oil hasalso decreased. An increase of a 1%effective tax rate will cause adecrease of 1.03% or Rp 37.35/kgprice of cooking oil. This showsthat this policy has been effectiveto control the price of cooking oil.Thus, if the government intends toprotect cooking oil consumersfrom price increase and possiblyfluctuation in the world market,then this policy could be aneffective alternative.

Producers, mainly smallholders,have suffered a great deal due to thepolicy. As the domestic price ofCPO is depressed by this policy, thefarm gate price of the farmersproduct (fresh fruit bunch or FFB),declines substantially. If thegovernment imposes a 4.78%effective tax rate, then FFB price

will be lower by around 6.61% orRp 35.23/kg (Table 3). In general,a 1% increase in tax will reduceFFB price by 1.13% (Rp 6/kg).

As the FFB price declines dueto the tax, farm income, measuredin terms of gross margin, alsosignificantly decreases (Table 3). A1% increase in the tax rate willreduce farm income by around1.53% or around Rp 90 000/ha/annum. When export tax is14.33% effective rate, such as thatimposed in 1999, and assumingeach farmer has 2 ha of oil palmplantation, then this policy willcause each farmer to suffer a lossof Rp 2.4 million per annum, asubstantial loss for a smallholder.

The export tax policy has alsoan impact on social welfaredistribution (Table 3). This policyhas caused welfare distributionfrom producer to consumer andgovernment. If the governmentincreases the export tax rate by 1%,the producer surplus will decreaseby 1.26%. On the other hand, thistax rate increase will increaseconsumer surplus by 0.86%.Moreover, this policy will alsoincrease the government revenue.For example, the average revenuegained from this policy if theeffective tax rate is 9.55%, isaround USD 174.5 million orRp 1 221. 544 billion per annum.

TABLE 2. PROJECTION OF THE IMPACTS OF CPO-EXPORT TAX (2000-2010)

Impacts of tax policy (%) Impacts of

Explanations Unit Means Tax 20% Tax 40% Tax 60%

Ef. 4.78% Ef. 9.55% Ef. 14.33%

Mature area 000 ha 2 096.75 -0.76 -1.39 -2.05 -0.15Production 000 t 6 345.89 -0.49 -0.90 -1.32 -0.09Export 000 t 3 457.51 -2.28 -3.86 -5.51 -0.41CPO price Rp/kg 2 379.24 -6.61 -10.69 -15.12 -1.13Cooking oil price Rp/kg 3 626.61 -6.05 -9.78 -13.80 -1.03Consumption 000 t 2 728.54 0.01 -0.03 -0.06 0.00Import 000 t 94.19 0.00 0.00 0.00 0.00Stock 000 t 708.08 0.10 0.05 -0.01 0.00

Notes:Ef. = equivalent with effective export tax.EET = effective export tax.

1%increasingof EET (%)

9


In addition, total governmentrevenue from this policy in 1994-1999 was estimated to be aroundRp 5241 billion.

The two other negative impactsof this policy are a loss in terms ofadded value of the industry andemployment. If the governmentimposes a 9.55% effective tax rate,the loss in terms of added value waspredicted to be around 11.56% ofadded value of the industry.Moreover, this export tax rate isalso expected to reduce the jobopportunity in the industry foraround 1730 workers.

AN ALTERNATIVEFORMULATION OF EXPORT

TAX RATE

The results of analysis indicate thatthe implementation of the CPO-export tax has advantages anddisadvantages to the industry.Moreover, this policy also has aredistribution impact to the agentsinvolved in the industries andgovernment revenue. This policyhas caused consumers and thegovernment to be better off. Onthe contrary, producers havebecome worse off, indicated by thedecline in area, production, export,farm income and employment.

Considering the benefits andcosts of the policy, the governmentis likely to maintain this policy inthe future. As this policy hassubstantial impacts on the industry,this policy needs to bereformulated in such a way thatconsumers are fairly protectedfrom a sharp fluctuation of theinternational market, whileproducers still gain a normal profitor incentive to develop theirplantations. Following this, themagnitude/rate of CPO tax shouldconsider the following facts:

investment in palm oilplantation is a long-terminvestment and therefore,price fluctuation cannot beavoided by the investors/producers. Within a certainperiod of time, CPO pricemay well be aboveproduction cost and viceversa. Production cost inthis case includes variablecost and capitalaccumulation for re-investment or rehabilitation(sustainable developmentapproach);

using this approach,production cost (assumingthat the exchange rate is

Rp 7000/USD) is aroundUSD 362/t or Rp 2555/kg.The government alsoindicates that theminimum CPO price to betaxed since July 1999 hasbeen USD 365/t (SKMenkeu No. 181/KMK.017/1999). Theassumption aboutexchange rate is on thebasis of the real exchangerate since currentexchange that is around Rp8400/USD, is consideredunder-valued;

profits/losses stronglydepend on the world price(HE) and exchange rate(ER). Therefore, these twofactors should be explicitlyconsidered to determinethe rate of the tax. In theother words, the export taxrate has to be calculated onthe bases of the fluctuationof these variables. Thus,profit (P) = (HE*ER-2555);

when the price of CPO isbelow the production cost,the producers/smallholderssuffer from a loss. Usingworld CPO price in the lasttwo decades, it was found

1%increasingof EET (%)

TABLE 3. PROJECTION OF THE IMPACTS OF CPO-EXPORT TAX ON WELFARE (2000-2010)

Impacts of tax policy (%) Impacts of

Explanation Unit Means Tax 20% Tax 40% Tax 60%

Ef. 4.78% Ef. 9.55% Ef. 14.33%

FFB price RP/Kg 533 -6.61 -10.69 -15.12 -1.13Gross margin Rp/ha/t 5 888 299 -8.79 -14.38 -20.56 -1.53Producer surplus Billion Rp 4 348.26 -7.14 -11.76 -17.30 -1.26Consumer surplus Billion Rp 25 140.12 5.11 8.14 11.27 0.86Total surplus Billion Rp 29 488.38 3.39 5.45 7.53 0.57Value-added Billion Rp 13 150.26 -7.03 -11.56 -16.44 -1.221Labour Million 1.90 -0.49 -0.90 -1.32 -0.09CPO tax Thousand USD - 85 583 174 506 266 587 175 559CPO tax Billion Rp - 599 084 1 221 544 1 866 108 1 228 912

Notes:Ef. = equivalent with effective export tax.EET = effective export tax.

10


that around 27% of thetimes CPO price werebelow the production costor probability (P) to getprofit is around 0.7 (P =0.7). This means that if theproducers/smallholdershave to transfer part oftheir profit to consumersand government, it is onlyaround 0.7 of the time canthat be transferred. Thiscoefficient acts as the firstweight in distributingwelfare (Equation 22);

on the basis of thesecondary right theorywhich states that profit gainof an industry is not merelyenjoyed by the peopleinvolved in the industry,but also by people, whobecause of some obstaclescannot participate in theindustry. This theory isrelevant because profit gainby producers due to eitherworld price increase orcurrency depreciation willcause a decrease inconsumer welfare. In linewith this argument, it isassumed that 75% of theprofit belongs to producersas primary right, while therest (25%) will belong toconsumers as secondaryright (SR). The mainrationale behind thisassumption is that socialcost of the estate cropindustry in so-calledreformation era is around20%-30%. For example,the government askedprivate estates to releasearound 20% of their landfor local people. For NESSchemes, the governmentstipulated that total localpeople involved in theproject is at least 30% oftotal participants;

the magnitude of the tax

should also consider thenumber of producers andconsumers, as a proxy ofpolitical power/pressuregroup. In this study, thenumber of consumer (NC)and producers togetherwith family members (NP)are 210 million and 10million, respectively; and

the magnitude of the taxshould also consider theincome share of oil palmplantation to total farmersincome (IS), and share ofcooking oil expenditure tototal household expenditure(ES). Within this study, theformer is estimated to bearound 80% and the latter tobe 4% (Badan Pusat Statistik,2001).

Following all these argumentsthen, the formulation of analternative export tax is as follows:

PE = * P * SR * (NC/NP) *(CS/IS)

= (HE*ER-2 555)*0.7*0.25*(210/10)*(4/80)......(22)

= (HE*ER-2 555)*0.1838= (HE*ER-2 555)*

18.38%....................(23)where:PE = Export tax rate (Rp/kg)HE = Export price (USD/t)ER = Exchange rate (Rp/USD)

This export tax formulation hassome fairer justifications. Firstly,the tax will be effective if theproducers gain profit, at leastenough to rehabilitate theirplantation. This representssustainable developmentargument. Secondly, the benefitsgained due to price increase orcurrency depreciation aredistributed among producers,consumers and the governmentafter considering secondary right(equity argument), the number ofproducers and consumer thatcould be a proxy of political poweror pressure group (political

argument) and the importance ofCPO in producer and consumerperspective (economic or welfareargument).

In socializing the export taxrate, the government should notapply this formula because thistends to be misunderstood. Forexample, when the governmentannounces that the export tax ratewas 30%, then many producersperceive that the magnitude of thetax was 30% of the price, not 30%of the difference between worldprice and the minimum price taxed.To avoid this misunderstanding,then the rate to be utilized shouldbe the effective tax rate. Followingthis, the export tax rate on the basisof various CPO prices andexchange rates are presented inTable 4.

As seen in Table 4, if CPO priceis USD 480/t and the exchange rateis Rp 7200/USD, then the effectiveexport tax rate will be 4.74% ofprice or around USD 22.7/t. If theprice goes up to USD 620/t and theexchange rate is Rp 9000/USD,then the effective tax rate will be9.79% of export price or aroundUSD 60.14/t.

Since the export tax rate has aneffect on the cooking oilcontribution to inflation rate, it isconsidered important to estimatethe relationship between thecontribution of cooking oil to theinflation rate with world CPO priceand exchange rate under theassumption that the effective taxrates are as presented in Table 4(Equation 24).

KONIFL = 0.00349DWDPOP +0.000224DINER ................(24)

where:KONIFL: contribution of cooking oilto Indonesian inflation rate (percent);DWDPOP: change in the world CPOprice (USD/t)DINER: change in exchange rate (Rp/USD)

11


Since 2002, the contribution ofcooking oil to the inflation rate hasbeen around 0.2266%. Forexample, if the world CPO priceincreased by USD 100/t, and theeffective tax rate is as presented inTable 4, then contribution ofcooking oil to the inflation increasewill be 0.349%. Similarly, a 1%increase in the exchange rate willcause a 0.356 increase in thecontribution.

CONCLUSION AND POLICYIMPLICATIONS

The implementation of the CPO-export tax has had varioussubstantial effects on bothefficiency/growth and welfaredistribution. This study finds that

an increase of a 1% export tax ratecauses a 0.19% decrease in maturearea/investment and 0.81% drop inproduction. More over, this policyalso leads to a 0.41% cent drop inexport. At the farm level, anincrease of 1% export tax rate leadsto a decrease of 1.53% of farmincome. In addition, this alsocauses a 1.22% decrease in addedvalue in industry and 0.09%decrease in employment. However,this policy has been proven as aneffective means to control domesticCPO price and cooking oil price.A 1% increase in the export tax ratereduces the CPO domestic priceand cooking oil price by 1.13% and1.03%, respectively. Followingthese, producer surplus declines by1.26%, while consumer surplus

increases by 0.86% for 1% increaseof the tax rate. Total export revenuegained by the government in 1994-1999 was around Rp 5 241 billion.

Considering several factorsrelated to the industries, the exporttax rate prior to 2001 which is 30%,is too high. The results of thisstudy show that the export tax rateshould be around 18% of thedifference between the world CPOprice and the minimum price taxed(USD 362/t). Under this tax rateformula, the effective tax rates willvary from 0.19%-10.98% of theworld price, depending on the CPOworld price and exchange rate.Moreover, the socialized tax rateshould be the effective tax rates,not the nominal tax rates.

TABLE 4. EXPORT TAX EFFECTIVE VALUE BASED ON EXPORT TAX FORMULATION 18.38%

Price Exchange rate Effective tax Value Exchange rate Effective rate(USD/t) (Rp/USD) (%) (USD/t) (Rp/USD) (%)

366-400 6 000-6 500 0.00 550-600 6 000-6 500 5.31366-400 6 500-7 000 0.19 550-600 6 500-7 000 6.28366-400 7 000-7 500 1.45 550-600 7 000-7 500 7.12366-400 7 500-8 000 2.54 550-600 7 500-8 000 7.84366-400 8 000-8 500 3.50 550-600 8 000-8 500 8.48366-400 8 500-9 000 4.35 550-600 8 500-9 000 9.05400-450 6 000-6 500 0.70 600-650 6 000-6 500 6.36400-450 6 500-7 000 2.01 600-650 6 500-7 000 7.25400-450 7 000-7 500 3.14 600-650 7 000-7 500 8.02400-450 7 500-8 000 4.12 600-650 7 500-8 000 8.68400-450 8 000-8 500 4.99 600-650 8 000-8 500 9.27400-450 8 500-9 000 5.75 600-650 8 500-9 000 9.79450-500 6 000-6 500 2.56 650-700 6 000-6 500 7.25450-500 6 500-7 000 3.73 650-700 6 500-7 000 8.07450-500 7 000-7 500 4.74 650-700 7 000-7 500 8.78450-500 7 500-8 000 5.62 650-700 7 500-8 000 9.40450-500 8 000-8 500 6.40 650-700 8 000-8 500 9.95450-500 8 500-9 000 7.08 650-700 8 500-9 000 10.43500-550 6 000-6 500 4.07 700-750 6 000-6 500 8.02500-550 6 500-7 000 5.13 700-750 6 500-7 000 8.78500-550 7 000-7 500 6.04 700-750 7 000-7 500 9.45500-550 7 500-8 000 6.84 700-750 7 500-8 000 10.02500-550 8 000-8 500 7.54 700-750 8 000-8 500 10.53500-550 8 500-9 000 8.16 700-750 8 500-9 000 10.98

12


REFERENCES

ABIDIN, Z (2000). Dampak Liberalisasi Perdagangan terhadap KeragaanIndustri Gula Indonesia: suatu analisis kebijakan. Ringkasan Desertasi.Program Pascasarjana, Ekonomi Pertanian, Institut Pertanian Bogor.

AMANG, B (1995). Peranan minyak sawit dalam memenuhi sembilankebutuhan pokok. Paper presented at Seminar Peluang dan TantanganIndustri Kelapa Sawit Menyongsong Abad XXI, Medan. 1-2 August 1995.

ARIFIN, S dan SUSILA, W R (1998). Development and prospect of palmoil industry in Indonesia. Indonesian Agricultural Research andDevelopment Journal, 20 (2): 25-32.

BADAN PUSAT STATISTIK (2001). Statistik Indonesia. Badan PusatStatistik, Jakarta.

BARTON, J H (1993). Implication of GATT of world trade in vegetableoils. Paper presented at the 1993 PORIM International Palm Oil Congress.Kuala Lumpur. 20-25 September 1993.

BASIRON, Y (2002). Palm oil and its global supply and demand prospects.Oil Palm Industry Economic Journal Vol. 2 No.1: 1-10.

DIREKTORAT JENDERAL BINA PRODUKSI PERKEBUNAN (1991-2003). Statistik Perkebunan Indonesia, Kelapa Sawit. Direktorat JenderalPerkebunan, Jakarta.

DRAJAT, B (2003). Evaluasi dan Prospek Kinerja Subsktor Perkebunanpada Era Perdagangan Bebas Dunia. Disertasi Doktor. ProgramPascasarjana, Institut Pertanian Bogor.

ERNAWATI (1997). Kajian Keragaan Gula Indonesia dan Simulasi DampakKebijakan Liberalisasi Perdagangan Gula Dunia. Thesis S2, ProgramPascasarjana, IPB, Bogor.

INTERNATIONAL MONETARY FUND (1986-2002). InternationalFinancial Statistics. International Monetary Fund. Washington D.C.

KOUTSOYIANNIS, A (1977). Theory of Econometrics. The MacmillanPress Ltd, UK.

OIL WORLD (1987-2002). Oil World Annual. ISTA Mielke GmbH.Hamburg. Germany.

PASQUALI, M (1993). Prospects to the year 2000 in the world oilseeds,oils and oilmeals economy: policy issues and challenges. Paper presentedat the 1993 PORIM International Palm Oil Congress. Kuala Lumpur. 20-25 September 1993.

PINDYCK, R S dan RUBINFIELD, D L (1987). Econometric Modelsand Ecomomic Forecasts. Mc Graw-Hill Book Co, Singapore.

13


SUSILA, W R; ABBAS, B S; HADI, P U; PRIYAMBODO, A and LUBIS,S O (1995). Model ekonomi minyak sawit mentah dunia. Jurnal AgroEkonomi, 14(2): 21-43.

SUSILA, W R (1997). Proyeksi produksi MSM Indonesia: suatupendekatan vintage. Warta Pusat Penelitian Kelapa Sawit, 5(2): 87-99.

SUSILA, W R; SUPRIONO, A and HARIANTO (2002). Dampakkeberhasilan Putaran Uruguay terhadap perdagangan dan industri kopidan kelapa sawit dunia dan domestik: dampak liberalisasi perdaganganCPO dan pajak ekspor CPO Indonesia terhadap perdagangan dan industriCPO dunia dan domestik. Kantor Menteri Negara Riset dan Teknologi,Dewan Riset Nasional.

ZULKIFLI (2000). Dampak Perdagangan Bebas terhadap KeragaanIndustri Kelapa Sawit Indonesia dan Perdagangan Minyak Sawit Dunia.Disertasi Doktor. Program Pascasarjana, Institut Pertanian Bogor.

copyright: MPOB 2004