Review of livestock self sufficiency in indonesia

Journal of Southeast Asian Economies Vol. 30, No, 3 (2013), pp, 294-308 ISSN 2339-5095 print / ISSN 2339-5206 electronic

DOI: 10.1355/ae30-3e

Determinants of Relative Demandfor Imported Beef and a Reviewof Livestock Self-Sufficiency in

Indonesia

Risti Permani

This paper reviews the challenges facing Indonesia's self-sufficiency programme. It analysesthe determinants of change in relative demand for imported beef by using the Vector ErrorCorrection Model (VECM) based on annual data from 1992 to 2010. It also investigates thelong-run relationships between relative domestic price and relative import quantity to predictthe impact of decreased reliance on imported beef using Impulse Response Functions (IRFs).The results suggest that increased income in Indonesia is associated with increased relativedemand for imported beef. A shock in relative import quantity, as a result of a governmentdecision to cut beef import quotas, for example, would have long-term impacts on relativedomestic price.

Keywords: Livestock, Indonesia, self-sufficiency. Vector Error Correction Model (VECM), ImpulseResponse Functions (IRFs).

I. Introduction

The beef industry is becoming more economicallyand politically important in Indonesia. Therecent past was characterized by the Indonesiangovernment's ambition to achieve self-sufficiencywith regard to this good.' Beef self-sufficiencyhas been primarily motivated by the revitalizationprogrammes in agriculture, fishery, and forestryintroduced in 2005. In the past, the term "self-sufficiency" specifically referred to self-sufficiencyin rice (Mears 1984). More recently, Indonesia's

growing income has led to a diversification of foodconsumption as the population experiences changingtastes, dietary intake, and purchasing power. Whilerice remains an important commodity, there is anincreasing demand for non-staple and high-proteinfood commodities such as beef. Another drivingfactor has been increased reliance on imported livecattle and, in contrast, increased productive femalecattle slaughter. In Indonesia, "excessive" relianceon imports is often perceived to be a result ofthe government's failure to stimulate domestic

Journal of Southeast Asian Economies 294 Vol. 30, No. 3, December 2013

© 2013 JSEAE

production rather than as a means of freeing upresources for more competitive activities.

Indonesia must increase its domestic productionfrom 67 per cent of domestic consumption in 2010to 90 per cent by 2014 (Sarwindaningrum 2009).Additionally, the cattle population including thosethat are part of the breeding cycle must increasefrom 12 million in 2009 to 14.6 million by 2014(Sarwindaningrum 2009). Although historicaltime series data has shown that such a target isan ambitious one given the high volatility in beefproduction, the Indonesian government believesthat the 2011 livestock census, reporting cattlepopulation at 14.8 million, is a positive indicationthat the target is achievable.

Yet, self-sufficiency is not simply about havingadequate cattle, especially if one takes sustainabilityinto account. Various sources suggest that a GDPgrowth of about 6 per cent per annum has led toa 3 to 5 per cent increase in beef consumption.This figure is expected to rise given increasedurbanization, modem retail penetration, and anexpanding middle class (estimated at 30 millionpeople in 2012). In the past, the deficit was metby live cattle imports, including slaughter andfeedlot cattle, and boxed beef imports. However,the current self-sufficiency programme restrictsthese imports, in particular boxed beef. The importquota for boxed beef was slashed from 100,000tonnes in 2011 to 34,000 tonnes in 2012 and wasaimed at providing greater opportunities for localbeef producers. Despite massive governmentprogrammes to assist smallholder cattle producers,only some regions such as West Nusa Tenggarahave seen successful results. In the meantime, beefprices continue to increase. While beef distributorsand the Ministry of Agriculture have differentopinions regarding the cause of this price hike,the increase contradicts the objective of becomingself-sufficient in beef: to promote food security ofwhich providing access to affordable food is a keycomponent.

Against this backdrop, this study aims atreviewing several policy options to achievingself-sufficiency in beef production. In addition, itinvestigates how the ratio of total imported beef todomestic beef production responds to changes in

the relative prices and income in the long and shortterm by applying the Vector Error Correction Model(VECM) using annual data from 1991 to 2010.Relative price variation may result from changes innational price levels, tariff reductions, and globalprice hikes. This study also adopts the ImpulseResponse Functions (IRFs) model to observecausality among the variables. The outcomes areuseful for gauging Indonesia's response to a shockin trade flows due to changes in price and quantity.This information may help predict the impact ofcurrent government approaches to livestock self-sufficiency on the domestic economy, both in theshort and long term.

Few studies comprehensively review Indonesia'slivestock sectors, particularly in relation to thegovernment's self-sufficiency programme. Twoexceptions are Hadi et al. (2002) and Vanzetti etal. (2010). The first study presents an excellentreview of government policies that have been andshould be implemented, as well as a single-countrysimulation experiment to estimate the effects ofvarious policy options such as changes in tariffand changes in technical production capacity inlivestock sectors (Hadi et al. 2002). In a similarvein, but using a multi-country computable generalequilibrium model, GTAP, Vanzetti et al. (2010)look at the feasibility of achieving livestock self-sufficiency. Both studies conclude that achievingself-sufficiency through tariffs on imported beefand imported cattle is very costiy and beef self-sufficiency can better be achieved through researchand development to increase the productivity ofnative cattle in both breeding and fattening.

This paper extends the GTAP analysis (Vanzettiet al. 2010) by adopting the Impulse ResponseFunctions (IRFs) model. The use of the IRFsallows us to simulate the impacts of a shock ona selected variable. In particular, the model isable to trace out the response of the relative beefimport quantity in the VEC system to shocks inthe relative price (and vice versa) and projectthe effect of this one-time shock for future timeperiods. In addition, it takes into account tariffand income per capita in estimating demand forimported beef, which were not considered inVanzetti et al. (2010). Also important is a brief


discussion on several policy options to achieveself-sufficiency. While their studies provide valu-able insights to the issues, Hadi et al. (2002)'s andVanzetti et al. (2010)'s focus on international tradepolicy seems to slightiy overlook an analysis ofdomestic problems.

This study finds that as Indonesia's income percapita increases, domestic demand for importedbeef has increased. The IRFs suggest that a shockin relative import quantity, for example as a resultof the current government's decision to cut beefimport quota, would have long-term impacts onrelative domestic price.

The remainder of the paper is organized asfollows. Section II reviews challenges and policyoptions to achieve the livestock self-sufficiencytarget. Section III investigates the determinants ofimport substitution and IRFs. Finally, section IVconcludes with a brief discussion on the limitationsof the study presented in this paper and offerssuggestions for future research.

IL Challenges for Achieving LivestockSelf-sufficiency

The concept of self-sufficiency is closely related tofood security but the two terms differ. Accordingto the Government of Republic of IndonesiaRegulation {Peraturan Pemerintah) No. 68 Year2002, food security is a food-sufficient conditionfor households determined by the availability ofadequate food in terms of: quantity; quality; safety;equality; and affordability. Food supply can comefrom either domestic production or other sources.Self-sufficiency, on the other hand, is defined as acondition in which at least 90 per cent of domesticdemand for food is met by domestic production(Ilham 2006). Hence, while food securityemphasizes the importance of access to food,self-sufficiency requires sufficient domestic foodproduction capacity to meet domestic demands.In other words, self-sufficiency is an importantbut not a necessary condition to achieving foodsecurity. However, with increasing internationaldependence, the sufficiency of local supply is notonly determined by local production but is also

affected by the world market and governmentpolicies on trade.

Achieving self-sufficiency through improvedproductivity is ideal but challenging. A previousstudy identifies some "technical" constraintswithin the livestock sectors across Asian countries,including: limited land; a shortage of skilledlabour and resources to control animal disease;the lack of investments; the general lack of relatedand supporting industries; and the unavailabilityof conducive conditions governing the productionand marketing of ruminant meat (Rutherford1999). Indonesia deals with similar issues.

First, the supply of land in Indonesia isproblematic. Indeed, this was the main reason citedby the Minister of Agriculture to explain why self-sufficiency cannot be achieved in 2010 (Wahyuni2009). Hutabarat and Kustiari (2009) identify thatone major problem faced by Indonesian agriculturalsectors is increased demand for agricultural landin addition to increased demand for water andirrigation. A 2005 survey on livestock in NusaTenggara Barat (NTB) province showed thatthe limited land supply has resulted in farmerskeeping their cattle confined (Yusdja et al. 2005).This method may impact beef quality.

Second, labour supply and welfare conditionsare problematic in Indonesia. There has beenrising concern about the low proportion of youngworkers in the livestock sector. Of the 4.3 millionworkers in the livestock sectors, 25 per cent arebetween the ages of 10 and 25, 35 per cent arebetween 25 and 44, and the rest (39 per cent)are over 44 (General Directorate of LivestockServices — Ministry of Agriculture Republic ofIndonesia 2008). Cattle farmers are also facingfinancial constraints. Indeed, a majority of cattlefarmers in Indonesia are "keepers" or those whokeep cattle as assets rather than "producers". Thismay contribute to high productive female slaughterrates. The Indonesian government has introducedvarious credit schemes to control the rates, yet theresults of these schemes are not well-documented.

Another major problem with achieving self-sufficiency is strong pressure from the internationalcommunity to keep domestic markets open.


Indonesia's trading partners, particularly ASEAN,may object to the goal as there is an agreement onfood security through the ASEAN Integrated FoodSecurity (AIFS) framework. The AIFS framework,though mainly focused on grain staples, alsoincludes some relevant livestock considerationsaround the stated cooperation objectives ofincreasing sustainable food production; promotingconducive markets and trade for agriculturalcommodities and inputs (such as animal breeds);ensuring food stability; and promoting availabilityand accessibility to agricultural inputs (Vanzettiet al. 2010).

To become self-sufficient in beef, the Indonesiangovernment has initiated a number of livestockdevelopment programmes. However, the mainobjective of these various programmes is notclearly stated. Hadi et al. (2002) conclude that anylivestock development strategy undertaken by thegovernment should: (i) improve the incomes ofsmallholder producers; (ii) encourage a sustainableand efficient domestic production capacity; and(iii) satisfy the growing demands of Indonesia'sconsumers for beef in ways that improve theoverall performance of the economy.

Preliminary observations and consultations withstakeholders suggest a range of policy options thatthe Indonesian government can choose from. Theoptions can be generally divided into two broadcategories: trade and productivity improvement.This study focuses on trade but a brief overviewof productivity improvement policy is presentedbelow.

In terms of productivity improvementprogrammes, the biggest challenge for the Indonesiaself-sufficiency target is breeding. The number ofbreeding units has not shown a consistent increaseover the years. In 2007, there were only ten cattlebreeding units across Indonesia. Given the aim toboost cattle population by over 20 per cent, sucha number is highly inadequate. Foreign investmentin breeding establishments is still low. Accordingto the National Statistics Agency (BPS), onlyone unit out of 317 small and large livestockestablishments with cattle had investments from aforeign company.

Breeding activities in smallholder systems arealso problematic. A consultation with West NusaTenggara-based scientists suggests that there arefive key components of a successful smallholderbreeding program: effective insemination;seasonal mating; early weaning; effective cattleconfinement management; and good animalhusbandry practices. Substantial challenges remainin the last component. For example, the animals'food supply is subject to limited land. In mostcases, farmers use their land to support their ownfood consumption rather than that of their cattle(Winarso 2009).

Switching the focus to trade. Table 1 presentsseveral policy options. In recent years Indonesiahas been experiencing a significant decline inlive cattle and beef imports. The share of beefconsumption coming from imported live cattlehad decreased from about a quarter of totalconsumption in 2010 to 11 per cent in 2012 andboxed beef decreased from one-third in 2010 to6 per cent in 2012. The figure is likely to falleven further as the government continues topromote its beef self-sufficiency programme.Importing breeder cattle has therefore been seenas "the second best option". However, recentdevelopments in the breeding cattle trade have notbeen encouraging. In September 2012, a seriousdispute emerged between the Indonesian andAustralian governments with Indonesian officialsrejecting more than 11,000 breeding cattle fromAustralia (Alford 2012). The dispute followedanother incident where Indonesian officialsprevented 2,000 Australian cattle from enteringthe Jakarta port in August 2012.

One key issue in the dispute is the confusionover distinguishing breeder cattle form therest. According to the Ministry of AgricultureRegulation Number 54, the three categories ofcattle (breeder, feeder, and slaughter cattle) willbe subject to different import duties. Feeder andslaughter cattle will be charged a 5 per cent importduty, while there is no import duty for breeders.The August 2012 incident was triggered by theIndonesian officials' reluctance to recognize thatthe cattle being shipped were breeders, despite


TABLE 1Trade Policy Options to Achieve Beef Self-Sufficiency

Policy options

Impose tradebarriers (i.e., importquota, total importban, tariffs) for allimported cattleincluding breeding.feedlot, andslaughter cattle.

Increase beefsupply by importingfeedlot cattle.

Import breedingcattle.

Pro.s

• To immediately achieve self-sufficiency in livestock given asignificant drop in imports andconsumption.

• To protect local beef producers.• Imports are easily controlled i.e..

only coming into the countrywhen needed.

• To reduce the risks of animaldiseases spread from othercountries (in the case of totalimport ban).

• To provide employmentopportunities in feedlot sectors.

• Local consumers including beefprocessing small industries (e.g..meatball sellers, etc) have accessto an affordable and consistentsupply of beef.

• Imported breeding cattle can beallocated to smallholder systemsusing government support schemes.thus increasing employmentopportunities and livelihoods ofsmallholder beef producers.

• To achieve more sustainable self-sufficiency targets by increasingthe cattle population.

• This option may improve thegenetics of the Indonesian cattlepopulation.

Cons

• Increased beef prices in the short-run due to supply shortage; in thelong-run, the ability to meetdemand depends on productivitygrowth rates i.e., whether they canexceed consumption growth rates.

• Can damage trade relationshipswith partner countries.

• Allocation of import permits maycreate rent-seeking opportunities.

• Discourage private sector(local and foreign) to invest infeedlot sectors.

• Feedlot sectors may have problemsin finding breeding cattle.

• This option provides less incentivefor improvement in breedingactivity.

• Smallholder feedlot systems mustcompete with large-scale feedlotcompanies (although this mayprovide incentives for them toimprove productivity).

• This option provides less incentivefor improvement in breedingacfivity.

• There is little room or incentivefor Indonesia to improve genetics.although preliminary observationssuggest that this is not a majorissue.

• Recent import protocols restrictthis option in relation to pedigreecertificate requirements.

• The government's requirement thatcommercial feedlots keep breedercattle in order to obtain an importpermit is economically inefficient.

• Farmers may not have sufficientknowledge and capability toproduce the planned outcomes.

SOURCE: Author's compilation based on discussions with various stakeholders.


claims made by the importers that the cattle were"breeding slaughter cattle". The certification fromthe Australian Brahman Breeders' Associationdeclaring that the cattle were breeders (not feedersas claimed by the Indonesian government) was notaccepted. Indonesian Customs refer to RegulationNumber 19 which requires individual pedigreecertification for each animal (The Republicof Indonesia Ministry of Agriculture 2012). Abreeding heifer brought into Indonesia typicallycosts US$1,000 while a pedigreed animal may beup to 200 per cent more expensive due to the lackof supply (Alford 2012).

Despite a number of major challenges, theIndonesian government has been focussing itsresources to achieve livestock self-sufficiency by2014 through various programmes. Yet, the successof these programmes should be assessed solelybased on the self-sufficiency ratio. A commonmeasure of self-sufficiency in livestock can bemisleading because it ignores the types of cattlethat the domestic industry produces. The self-sufficiency ratio is normally defined as the ratioof production to production plus imports minusexports. One main problem with this approach isthat imports may come in various forms. In thecase of beef, Indonesia not only imports beef inboxes but also live cattle for various purposes,including slaughter, feeders, and breeding animals.Not all live cattle imports, for example feeders andbreeding animals, can be added to "imported beefcalculations of the self-sufficiency ratio. Thiswould overestimate total imports for consumptionpurposes. Further, cattle brought to Indonesia asfeeders and breeding animals will reduce the self-sufficiency ratio. In contrast, failure to includeslaughter animals would lead to an underestimationof beef imports, and thus an over-estimated self-sufficiency ratio will be obtained.

Figure 1 shows different assumptions on theimport component. SSRl is where domestic meatproduction is divided by total meat productionplus meat imports minus meat exports. Weexclude live cattle imports for slaughter inthe calculation. SSR2 includes all live cattieimports in the calculation by assuming one livecattle weighs 224 kilograms of meat once it is

slaughtered.^ SSRl suggests a 100 per cent ratioin 2010, while SSR2 suggests a 75 per cent ratio.The corresponding import dependency ratio toSSR2 i.e., IDR2 suggests an average increase inimport dependence in the last decade, from 10per cent in 1998 to 30 per cent in 2009 beforedecreasing to 25 per cent in 2010.

The sustainability of Indonesia's livestock self-sufficiency programme should be of concern. Figure2 shows that over the last few decades, increasesin stocks of live cattle have been significantlyassociated with live cattle imports.' From the late1960s to the 1970s, Indonesia was exporting livecattle and buffaloes to several countries includingSingapore and Hong Kong, which may explain adrop in the stocks in 1969 (Figure 2). As domesticdemand increased, the Indonesian governmentstopped live cattie exports in 1979, which mayexplain a significant increase in stocks in the early1980s.

In the days following Australia's live cattieexport ban to Indonesia in mid-2011, the Indonesiangovernment expressed its confidence that domesticbeef price would not be affected by changes inimports. The same confidence seemed to informits decision to cut import quotas in 2011. Thenext section investigates the causal relationshipsbetween domestic prices and imported beef. Theoutcome is useful to test whether the argumentfor insignificant effects of live cattle import ondomestic price holds.

III. Empirical Analysis

This section uses the VECM to further exploredeterminants of relative demand for imported cattie.The analysis is important to assess how responsiveIndonesia's terms of trade are to fluctuation inprices. In addition, it takes into account changesin tariff rates and income per capita. Data aretaken from various sources (FAO 2012; WorldBank 2012a, 2012e)."' Table 2 presents descriptivestatistics.

The paper employs three alternative methods:(i) Ordinary Least Squares (OLS); (ii) the PartialAdjustment Model (PAM); and (iii) the VectorError Correction Models (VECM) (Kapuscinski


FIGURE 1Self-Sufficiency and Import Dependency Ratios (1961-2010)

1960 1970 1980 1990 2000 2010Year

1960 1970 1980 1990 2000 2010Year

SSR1 SSR2 IDR1 IDR2

and Warr 1999). The specifications are as followsfor country j at time i:

where, log refers to the (logarithm)

relative demand for imported beef including

P (') Ibeef from live cattle import; log -^^— refers

to the (logarithm) relative domestic price; andindicates the difference operator. For simplicity,the logarithm of relative demand for imported beefis denoted as IogQI_QD, while the logarithm ofrelative domestic price is denoted as logPD_PLThe difference operator and the lagged operatorwill be denoted as "D" and "L", respectively.

OLS estimates may be able to produce unbiasedand consistent estimates but they cannot captureany dynamic relationship between imports,domestic production, and prices. Given the time-series data we use in the analysis, it is most

Journai of Southeast Asian Economies 300 Voi. 30, No. 3, December 2013

nGURE2Live Cattle Stocks and Imports (1961-2010)

A

il

' 1960 1965 1970 1975 1980 1985Year

1990 1995 2000 2005 2010

Stocks(heads) Import quantity (heads)

likely that the estimates are inefficient due toautocorrelation. The model cannot control for anycommodity-specific effects either. For example,the Indonesian government might protect domesticbeef production more than the government protectslive cattle. The inclusion of the level of relativedemand for imports in the previous period mightbe able to capture such time-variant commodity-specific effects, and is therefore the reason behindthe use of the PAM model.

The problem with the PAM method, however,is the autocorrelation of the error terms due to thepresence of a lagged dependent variable. It couldyield biased estimates of elasticity of substitution

2 i

(ßj ) . More specifically, if the coefficient for ßj is

larger than 1, the autoregressive estimates arenon-stationary. If this is the case, then stationarity

(i.e, when the series has constant mean andstandard deviation over time) can be achieved bysimple differencing or some other transformation.However, relative demands for imports in thelong-run are often "drifting together" (i.e.,cointegrated) with the relative price index atroughly the same rate.

The VECM model aims to distinguish thelong-run relationship between the two variables(potentially drifting together) and the short-run dynamics, i.e., deviations of relativedemand for imports from its long-run trend anddeviations of relative price-index from its long-run trend (Engle and Granger 1987). The term

\{t-\)- log

Grangerf;pf(i)1l^— refers to the error

\p)(t))]correction term. The elasticity of substitutionis estimated based on the coefficient y\ which


TABLE 2Descriptive Statistics, 1991-2010

Definitions VariablesPeriod

1991 to 1996

316,851.3033,395.8131,477.6535,877.653,344.50290.53

2,367.18925.15

4.822.040.860.03

1 0.160.01

1997 to 2002

336,728.6016,709.6560,963.3127,527.541,923.20630.31

1,439.23193.10

1.060.160.870.030.150.01

2002 to 2010

392,934.9052,161.35100,794.6052,601.534,589.221,150.171,802.40255.73

1.250.370.840.060.170.02

Domestic beef production (tonnes) QD

Total beef import (including live QIcattle)(tonnes)''"

Domestic beef producer price PD(US$/tonne)

Imported beef price(unit value) PI(US$/tonne)

Live cattle tariff (simple average) TARIFF

Live cattle stock (% world's total STOCK_live cattle stock) PERCENT

Worid's GDP per capita GDPC_RATIO(Constant 2000 US$)

NOTE: a. Total livestock import is the sum of meat and live cattle imports in tonnes, where one head of imported livecattle is presumed to weigh 224 kilograms (see endnote 3 for details).GDPC_RATIO is the ratio of Indonesia's GDP per capita to the world's average of GDP per capita.SOURCE: Author's calculation based on data from various sources (FAO 2012; The World Bank 2012a, 2012b) . Thesecond row in each cell (in bold) represent standard deviation.

captures the short-run relationship between relativedomestic price and relative demand for imports.Coefficient y^ tells us the proportion of thedisequilibrium which is corrected with eachpassing period. This coefficient should be negativeand less than the absolute value of 1, indicating

its re-equilibrating properties. If y j = 0 , then the

process never re-equilibrates and if y j = - 1 , re-

equilibration happens in period I. Table 3 presents

the stationarity property of the data and the

cointegration test suggesting the use of VECM.As shown in Table 4, estimates are obtained

from three different specifications: OLS, PAM andVECM. For VECM, two different approaches areused. Column (3) of Table 4 presents results froma two-step VECM. The lagged error-correction(EC) term is derived from residuals based on theOLS regression in column (1) of Table 4. Notethat the term indicates the speed at which it returns

to its equilibrium level. It is negative as expected,suggesting the process re-equilibrates. Column (4)of Table 4 presents a one-step VECM in which theerror correction term consists of lagged log QI_QDand lagged logPD_PI. Based on columns (3) and(4), the 95 per cent confidence interval of thecoefficient for relative domestic price is between0.35 and 1.17.

Next, the impulse response function (IRF) isderived to illustrate relative beef imports' responseto a shock in relative price. Figure 3(i) suggeststhat one unit shock in relative domestic price leadsto a 0.3 decrease in relative import demand inPeriod 1, then gradually increases in imported beefdemand in Period 2 before subsiding in Period 7onwards. Due to increased domestic beef prices,consumers would lower beef consumption, leadingto an initial decline in demand for imported beef.If the market is unrestricted, in Period 2, thehigher competitiveness of foreign beef producers


TABLE 3Stationarity and Cointegration Test

Statistics (p-value in parentheses)

Decision

Augmented Dicky-Fuller testfor stationarity

Ho = variable has a unit root(non-stationer)

Price variable Quantityvariable

-2.477 -2.914(0.339) (0.157)

Accept Ho Accept HO

Johansen test forcointegration

Ho = max number of

cointegration relations = 10.296

Accept Ho

NoTE.s: 5 per cent critical value of Johansen statistics (N = 18, lags = 2) is 3.76

TABLE 4Determinants of Relative Live Cattle Import, 1991-2010

Method

Dependent variable

logPD_Pl

L.logQLQD

D.logPD_PI

L.logPD_PI

L.EC

Constant

Implied coefficient of elasticity

Adjusted R̂Durbin-Watson statisticsNumber of parametersDurbin-Watson {HQ = no serial

correlation^Number of observations

OLSbit

logQLQD

1.275(1.737)

-2.830***(^.924)

1.275

0.2340.3192

Reject «0

20

PAMbit

logQLQD

0.979**(2.963)0.659***

(4.546)

-1.165***(^.496)

0.979

0.7581.7273

Accept Ho

20

VECM_two_stepbit

D.logQLQD

0.931***(6.064)

-0.400***(-5.855)

0.195*(2.269)

0.931

0.7931.9483

Accept Ho

19

VECM_one_stepbit

D.logQLQD

_0.409***(^.575)

0.762**(3.959)0.199

(0.728)

-0.768**(-2.970)

0.762

0.8121.9364

Accept Ho

19

NOTES: Durbin-Watson critical valuesDW(2,20) = 1.100-1.537; DW(3,20)

(lower and upper bounds) at 5 per cent level of significance:= 0.998-1.676; DW(3,19) = 0.967-1.685; DW(4,19) = 0.859-1.848.

Journal of Southeast Asian Economies 303 Voi. 30, No. 3, December 2013

FIGURE 3Impulse Response Functions

-.5-

IRF. logPD_PI, logQLQD

o 5 10 15 20 25 30 35 40 45 50

PeriodGraphs by irfname, impuîse variable, and response variable

IRF, logQLQD, logPD_PI

Ô-

0 5 10 15 20 25 30 35 40 45 50

PeriodGraphs by irfname, impulse variable, and response variable

compared to local producers would lead to anincrease in demand for imported beef.

Looking at the causal relationship, Figure 3(ii)shows that the effect of one unit shock in relativeimport quantity on relative domestic price ispositive. A shock in relative import quantity canbe due to more relaxed trade policy, decreaseddomestic production capacity, etc. A 1 unit increasein relative import quantity leads to a 0.285 increasein relative domestic prices in Period 1 but in Period2, the effect thins out and stabilises at Period 7.

Taken together, there is a strong indication ofa two-way causal relationship between relativeimport quantity and relative domestic price.Furthermore, a shock in each variable seems tohave significant (i.e., non-zero) effects in thelong run.

Relative demand for imported beef may beaffected by more variables than relative domesticprice. However, the short time-series data usedin this study restricts the number of right-handside (independent) variables that can be included.Preliminary regressions and a literature reviewhighlight the importance of income measures inpredicting changes in demand for imports. Thisstudy therefore includes relative income per capitai.e., the ratio of Indonesia's GDP per capita tothe world's average of GDP per capita (GDPC_RATIO). Unobserved determinants of relativedemand for imported beef may be correlated torelative domestic price. Therefore, the study usesthe Instrumental Variable (IV) method. To proxya change in relative domestic price (D.logPD_PI),it takes the available stock of cattle in Indonesia


as a percentage of the world's total cattle stock(STOCK_PERCENT). Another variable is tariff.This study uses data from the World IntegratedTrade Solution (WITS) database, which suggestthat tariff rates (both simple-averaged and trade-weighted tariff rates) for livestock up to 2010 hadbeen relatively low.

The first-stage regression is presented inTable 5. Both TARIFF and STOCK_PERCENTare significantly and negatively associated withchange in relative domestic price. This indicatesthat any intervention affecting both indicatorscould be effective in controlling the change inrelative domestic prices. A negative link betweenSTOCK_PERCENT and D.logPD_PI indicatesthat increased supply of livestock might lead tohigher competitiveness of domestic beef.

One may question whether relative domesticprice is more responsive to a change in increasingrelative herd size or imposing tariff. Accordingto the FAO statistics, between 2007 and 2008,STOCK_PERCENT increased by 0.05. Given itscoefficient, this would lead to a 0.19 decrease inD.logPD_PI. For TARIFF, it increased by 0.56between 2006 and 2007, implying a 0.10 decreasein D.logPD_PI. In other words, to provide thesame effect on relative domestic price as a 0.05increase in STOCK_PERCENT, the governmentmust increase tariffs by 100 per cent. Further work

might be useful to analyse whether increasingrelative domestic productivity by 0.05 or increasingtariffs by 100 per cent is more feasible. However,taking into account the inefficiency created bytrade barriers and sustainability of agriculturalinnovation, it seems that increased productivityis more preferable. The results also show thatincreased relative income per capita is associatedwith increased relative domestic price.

Table 6 presents the second-stage regression.The benchmark model in column (a) is taken fromthe last column of Table 4. Compared to column(a), the IV model in column (b) produces a slightlysmaller coefficient i.e., 0.630. GDP per capitaratio is positively associated with relative demandfor imported cattle. This supports argument thatmany middle-upper Indonesians households preferimported cattle to domestic products.'̂

rv. Concluding Remarks

This paper has investigated the determinants ofchange in demand for beef and cattle importsrelative to domestic cattle breeding in Indonesia.The results suggest that a relative increase in GDPper capita is associated with increased demand forimported beef. This study also finds that imposingtariffs could affect the relative domestic pricebut not to the extent that increased production

TABLE 5First-stage Regression

Dependent variable;D.logPD_PI Coef Std. Err. p-value

L.logQI_QDL.logPD_PI

L.TARIFF

GDPC_ratioL.STOCK_percentConstant

-0.394-1.097-0.189

36.466-3.806

-2.370

0.0370.135

0.0363.769

1.236

0.998

-10.43-8.13-5.19

9.67

-3.08

-2.38

0.0000.0000.000

0.0000.009

0.034

NOTES: Number of observations is 19; F-statistics = 82.32 (p-value = 0.0000);Adj R-squared = 0.8849; Root MSE = 0.1922


TABLE 6Second-stage Regression

Dependent variable: D.logQIjQD

D.logPD_PI

L.logQLQD

L.logPD_PI

GDPC_ratio

Constant

Adjusted R̂Number of observations

(Column (b))

Benchmark(a)

0.762**(3.959)-0.409***

(-6.575)0.199

(0.728)

-0.768**(-2.970)

0.81219

IV-model(b)

0.613**(3.135)-0.483***

(-7.681)-0.051

(-0.168)14.974*(2.028)-3.188**

(-2.614)0.849

19

NOTES: Instrumented variable is D.logPD_PI.

capacity does (relative to the world's productioncapacity).

Given various challenges discussed in theearlier sections and the outcome of the time seriesanalysis presented by this study, there is concernabout the long-term effects of current governmentapproaches to reaching its self-sufficiency target.As suggested by the IRFs, a 1 standard deviationnegative shock in relative import quantity as a resultof the government's import quotas, for example,will lead to a 0.3 standard deviation increase inrelative domestic price in the next period. Thelong-run effect of such a shock is also statisticallysignificant. The effect will be even greater if onetakes into account the political context of sucha protectionist trade policy. Furthermore, if theIndonesian government decides to restrict or evenban imports with subsequent significant increasesin prices, the country would face a greaterchallenge to control female cattle slaughters, whichhas already been an issue. This can happen whencattle farmers respond to decreased supply due toimport reduction by selling their cattle, includingproductive females, to benefit from the price hike.In such conditions, instead of achieving the self-

sufficiency target, Indonesia will be dealing with adrop in cattle population.

Small sample properties in this study suggestthat the results should be considered as no morethan indicative. The relatively short time seriesfor analysis constrains what could be incorporatedinto the VECM. It is, however, unlikely that thegeneral pattern of the results would change. TheVECM and IRFs present consistent results withwhat is known about the demand for importedbeef in Indonesia; that income per capita mattersand a shock in beef imports would have long-termeffects on domestic beef prices. Taking lessonsfrom studies on aid in Africa, studies using a smallnumber of samples for a time-series analysis pointto quite consistent results i.e., the significance ofaid in public finance (Fagemäs and Schurich 2004;Osei, Morrissey and Lloyd 2005). Future studiesshould review restüts in this paper using a longertime-series.

Furthermore, due to limited available data, thisstudy: (i) limits the welfare measure to income percapita without taking into account the health effectsof beef consumption; (ii) does not take into accountthe environmental impact of beef production; and


(iii) does not take into account the impacts ofchange in beef prices on its substitutes (e.g., lamband poultry) and complements (i.e., chilli to makespicy beef or flour to make meatballs). Previousstudies have found a positive link between beefconsumption and the risks of heart diseases (Eraser1994; Hu et al. 1999). Regarding environmentalimpact, a study in Europe finds that beef hasfour to eight times larger environmental impactsthan poultry per kilogram slaughtered weightas shown by various indicators including: ozonelayer depletion; aquatic ecotoxicity; use of non-renewable energy; contribution to global warming;

etc. (Weiderma et al. 2008). Internalizing theabove externalities, for example through taxingbeef cattle producers and retailers, would increasethe local price and, therefore, make local beefbecome less competitive compared to importedbeef. It would also increase the "already high"cost of the government's livestock self-sufficiencyprogrammes. In short, this study highlights theimportance of a careful cost-benefit analysis ofthe government's current approach to its self-sufficiency programmes. Considering theirimportance, these abovementioned issues shouldguide future studies.

NOTES

The author is grateful to Dr Ray Trewin, Dr David Vanzetti and Nur Rakhman Setyoko for providing valuable dataand inputs to the earlier version of this paper, and Professor Christopher Findlay and Associate Professor WendyUmberger for their continuing support. The author gratefully acknowledges Australian Centre for InternationalAgricultural Research (ACIAR) postdoctoral fellowship funding for this research through Project ADP/2005/068.The usual disclaimer applies.1. The target was initially announced during the early years of Susilo Bambang Yudhoyono's presidency. It was

subsequently postponed to 2010. In early 2009, Minister of Agriculture, Anton Apriantono announced that thedeadline must, once again, be pushed to 2014 due to a limited supply of land.

2. UN Comtrade data reveal that Indonesian imports of live cattle from Australia in 2008 stood at 494,254 headswith a net weight of 205,548 tonnes, implying an average of 416 kg per head. According to Vanzetti et al. (2010),the average slaughter weight can be derived by assuming a 210/390 ratio of slaughter to live weight. See theirstudy for details on the complexities that might arise due to differences in live weight and dressed weight.

3. The volatility in beef production is relatively low (coefficient variation or CV=0.37) as it is "smoothed" byhighly volatile imported live cattle, which then enter domestic production chains (CV=1.61).

4. Ptice indicators are expressed in unit values i.e., (import or export) values divided by quantity. See note 2 fordetails on converting live cattle weight into beef weight.

5. One may argue that as the economy grows, the Indonesians are becoming more aware of the negative healtheffects of beef consumption and, therefore, will reduce their beef consumption. According to Schroeder et al.(1996), although consumers in high income countries do not on average increase per capita beef consumptionwhen income grows, there is not enough evidence to concluding that there has been any decrease in beefconsumption either. To investigate this issue, this study adds GDP per capita and its square variables to theestimation. Unfortunately, this leads to severe multicoUineadty and the coefficients for both variables becomeinsignificant.

REFERENCES

Alford, Peter. "Jakarta fires new shots in cattle war". The Australian, 6 October 2012. Available from <http://www.theaustralian.com.au/national-affairs/jakarta-fires-new-shots-in-cattle-war/story-fn59niix-I226477655995>.

Engle, Robert F. and C.W.J. Granger "Co-Integration and Error Correction: Representation, Estimation, and Testing".Econometrica 55, no. 2 (1987): 251-76.

Fagemäs, Sonja, and Cedrik Schurich."The Fiscal Effects of Aid in Malawi". Economics and Statistics Analysis Unit(ESAU), Overseas Development Institute, 2004. Available from <http://ideas.repec.Org/p/odi/wpaper/7.html>.

FAO. EAOSTAT. Food Agricultural Statistics (FAO), 2012. Available from <http://faostat3.fao.org/home/index.html>.


Fraser, G.E. "Diet and coronary heart disease: Beyond dietary fats and low-density-lipoprotein cholesterol". AmericanJournal of Clinical Nutrition 59, no. 5 (1994): 1117S-1123S.

General Directorate of Livestock Services, Ministry of Agriculture Republic of Indonesia. "Number of AgricultureEmployee on Livestock Sub-sector based on Age Range Year 2008 by Province". 2008.

Hadi, RU., N. Ilham, A. Thahar, B. Winarso, D. Vincent, and D. Quirke. Improving Indonesia's Beef Industry. ACIARMonograph Series No. 95. ACIAR, 2002. Available from <http://www.aciar.gov.au/files/node/452/mn95.pdf>.

Hu, Frank B. Meir J. Stampfer, JoAnn E. Manson, Alberto Ascherio, Graham A. Colditz, Frank E. Speizer, CharlesH. Hennekens, and Walter C. Wlllett. "Dietary saturated fats and their food sources in relation to the risk ofcoronary heart disease in women". American Journal of Clinical Nutrition 70, no. 6 (1999): 1001-08.

Hutabarat, B. and Kustiari, R. "Data and Information on Policies Affecting Indonesian Agricultural Productivityin the Early Millennium". Indonesian Center for Agricultural Socio-Economic Research and Policy Studies(ICASEPS), Bogor, 2009.

Ilham, Nyak. "Analisis Sosial Ekonomi dan Strategi Pencapaian Swasembada Daging 2010". 2006. Available from<http://pse.litbang.deptan.go.id/ind/pdffiles/ART4-2b.pdf>. Accessed 21 April 2010.

Kapuscinski, Cezary A. and Peter G. Warr. "Estimation of Armington elasticities: An application to the Philippines".Economic Modelling 16, no. 2 (1999): 257-78.

Mears, L.A. "Rice and food self-sufficiency in Indonesia". Bulletin of Indonesian Economic Studies 20, no. 2 (1984):122-38.

Osei, Robert, Oliver Morrissey and Tim Lloyd. "The fiscal effects of aid in Ghana". Journal of InternationalDevelopment 17, no. 8 (2005): 1037-53.

Republic of Indonesia Ministry of Agriculture. "Minister of Agriculture Regulation No. 19 on Requirements of QualitySeed, Livestock Breeders and Animal Genetic Resources". The Republic of Indonesia Ministry of Agriculture,6 October 2012. Available from <http://perundangan.deptan.go.id/admin/p_mentan/Permentan%2019-2012.pdf>.

Rutherford, A.S. "Meat and milk self-sufficiency in Asia: Forecast trends and implications". Agricultural Economics21, no. 1 (1999): 21-39.

Sarwindaningrum, Irene. "Swasembada Daging Sapi 2014. Kompas, 6 October 2012. Available from <http://bisniskeuangan.kompas.com/read/20G9/11/09/05381056/Swasembada.Daging.Sapi.2014>.

World Bank. World Development Indicator Online. World Bank, lOYla. Available from <http://data.worldbank.org/>.

. World Integrated Trade Solution. World Bank, 2012è. Available from <http://wits.worldbank.org/wits/>.Vanzetti, David, Nur Rakhman Setyoko, Ray Trewin, and Risti Permani. "Home grown: cattle and beef self-sufficiency

in Indonesia". International and Development Economics. 2010. Available from <http://ideas.repec.Org/p/idc/wpaper/idecl0-04.html>.

Wahyuni, Nurseffi Dwi. "Rl Baru Swasembada Daging 2014". Detik Finance, 11 March 2009.Weiderma, B.P., M. Wesness, J. Hermansen, T. Kristensen and N. Helberg. "Environmental Improvement Potentials

of Meat and Dairy Products". JRC European Commission, 2008. Available from <http://ftp.jrc.es/EURdoc/JRC46650.pdf>.

Winarso, Bambang. "Pengembangan Ternak Sapi Potong dalam Mendukung Program Pengembangan SwasembadaDaging di Nusa Tenggara Barat". PSEKP Working Paper No. 98, 2009. Available from <http://pse.litbang.deptan.go.id/ind/pdffiles/WP_98_2009.pdf>. Accessed 20 April 2010.

Yusdja, Yusmichad, R. Sayuti, S. Hastuti, 1. Sadikin, Bambang Winarso, and Muslim. "Pemantapan Program danStrategi Kebijakan Peningkatan Produksi Daging Sapi". PSEKP Working Paper, 2005.

Risti Permani is a Research Fellow in the Global Food Studies, Faculty of Professions, University of Adelaide,Australia.

Journai of Southeast Asian Economies 308 Voi. 30, No. 3, December 2013

Copyright of Journal of Southeast Asian Economies is the property of Institute of SoutheastAsian Studies and its content may not be copied or emailed to multiple sites or posted to alistserv without the copyright holder's express written permission. However, users may print,download, or email articles for individual use.

Review of livestock self sufficiency in indonesia

Food

beef selfsufficiency

imported beef

beef production

beef industry

increased relative demand

relative domestic price

beef import quotas

relative import quantity