9lg $gulvrq - lpem

Vid Adrison

ISSN 2356-4008

LPEM-FEUI Working PaperChief Editor: I Kadek Dian Sutrisna RatherEditors: Riatu M. Qibthiyyah, Maria AgrivaSetting: Rini Budiastuti, Libertina Judith C.

Copyright 2013, Institute for Economic and Social ResearchFaculty of Economics University of Indonesia (LPEM-FEUI)Salemba Raya 4, Salemba UI CampusSalemba Raya 4, Salemba UI CampusJakarta, Indonesia 10430Phone: +62-21-3143177 Fax: +62-21-31934310Email: [email protected]: www.lpem.org

LPEM-FEUI Working PaperISSN

Deforestation in Decentralized and DemocraticIndonesiaVid Adrison1*

AbstractThis study investigates the effect of forestry decentralization on deforestation in Indonesia. A theoretical modelis constructed to explain the effect of decentralization on deforestation, and tested using the same dataset asused in [1]. We find the following evidence; (1) Forestry decentralization increases deforestation rate, (2) Exceptfor production forest, there are statistical evidence linking deforestation with district head election, (3) Localgovernment spending on environmental is ineffective in reducing deforestation.

KeywordsDeforestation — Decentralization — Democratization — Asia — Indonesia

1Institute for Economic and Social Research, Faculty of Economics Universitas Indonesia (LPEM-FEUI)*Corresponding author: [email protected], [email protected]

Contents

Introduction 1

1 LITERATURE REVIEW 2

1.1 Theoretical prediction and empirical evidence ofdecentralization . . . . . . . . . . . . . . . . . . . . . . . 2

1.2 Decentralization and deforestation in internationalcontext . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.3 Decentralization and deforestation in Indonesiancontext . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2 DATA SOURCES AND EMPIRICAL SPECIFICATION3

2.1 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . 32.2 Empirical specification . . . . . . . . . . . . . . . . . . 3

3 FINDINGS 4

3.1 Finding from in-depth interviews . . . . . . . . . . . 43.2 Results from econometric estimation . . . . . . . . 5

4 SUMMARY AND POLICY RECOMMENDATION 5

4.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 54.2 Policy recommendation . . . . . . . . . . . . . . . . . 5

References 6

INTRODUCTIONIndonesia forest has decreased substantially from 1.1 millionsquare km in 1990 into 0.944 million square km [2]. GivenIndonesia has the eight largest forest area and fourth largestpopulation in the world, Indonesia has a potentially impor-tant role in determining the green-house gas emission in theworld.

Some existing studies found that decentralization and lo-cal democratization contribute to deforestation in Indonesia.

For instance, [3] concludes that regulatory flaws between ex-isting Regional Government Law No. 32/2004 and ForestryLaw No. 41/1999. [4] also found that regional autonomypushed illegal logging even further by “legalizing” them, asthe local government applied charges/fee for transportingtimber. Recent study by [1] found that increased numberof jurisdiction and local district election also play a role indeforested Indonesia.

The effect on deforestation as found by existing literatureis contradictory to what the proponents of decentralizationand local democratization believes. Theoretically, decentral-ization will bring greater benefit to local economy than cen-tralized regime as it brings the government closer to thepeople. There are several channels that make decentraliza-tion beneficial such as inter-jurisdictional competition [5],informational advantage [6] and better preference homogene-ity [7]. On the other hand, local democratization increasesthe chance of bringing greater benefit as only the politicianthat can represent the local needs the best will be electedthrough local election (theoretically).

The discrepancy between theoretical versus actual resultsof decentralization and local democratization are realized bythe policy makers, both parliament and executives. This canbe reflected by the fact that some decentralization-relatedlaws – such as law No. 32/2004 on Regional Governmentand law No. 33/2004 on Fiscal Balance – are listed in the2009-2014 National Legislation Program (Program Legis-lasi Nasional/Prolegnas) of Indonesian Parliament (DewanPerwakilan Rakyat/DPR). In addition to the above laws, lawNo. 41/1999 on Forestry is also listed in the Prolegnas. Asof November 2013, none of these laws have been amended.

This study aims at investigating the effect of forestrydecentralization on deforestation in Indonesia1. In order to

1Notice that forestry resource management in Indonesia has gone

Deforestation in Decentralized and Democratic Indonesia — 2/8

answer the research question, we review existing relevantliteratures on deforestation and decentralization, conduct in-depth interviews with relevant stakeholders at the nationaland local policy makers, conducting empirical test througheconometric modeling.

This paper is organized as follows; in the next section,we discuss the existing literature of decentralization from the-oretical and empirical aspects, and the effect decentralizationon forestry resource management in international context aswell as Indonesian context. Section 3 discusses data sourceand empirical strategies. The findings from in-depth inter-views and empirical results are discussed in section 4, andfollowed by conclusion and policy recommendation in sec-tion 5.

1. LITERATURE REVIEW1.1 Theoretical prediction and empirical evidence

of decentralizationDecentralization is defined as the transfer of authority and re-sponsibility for public function from the central governmentto subordinate or quasi-independent government organiza-tion of the private sectors [8]. There are several types ofdecentralization, namely; political, administrative, fiscal andmarket decentralization.

From theoretical standpoint, the effect of decentralizationon public service delivery is ambiguous. In the proponentsides, there are several channels that make decentralizationleads to improved public service deliveries, such as increasedinter-jurisdictional competition [5], informational advantage[6], better preference homogeneity [7] increased direct ac-countability as local government is only responsible for acertain task that is specific to their jurisdiction [9].

On the other hand, decentralization may results in de-creased public service delivery. The channels may comefrom coordination breakdown among bureaucrats which leadto excess rent extraction which leads to a decrease in thequality of public goods and service [10] and low capacity oflocal bureaucrat [11].

The empirical evidence on the effect of decentralizationis also mixed. The impact of decentralization on growth arefound negative in some studies, for instance, [12] on China,[13] for United States, and [14] on developed and developingcountries. In contrast some other studies found positive effectof decentralization on economic growth, for instance [15]on United States, [16], cross country analysis by [17], [18]and [19]. On public goods provision, [20] find that localcompetition and public participation has increased efficiencyof public goods provision at the village level in Indonesia.On corruption, negative correlation between decentralizationwith corruption is found by [21] and [22], while [23] foundthat federalist countries tend to have higher corruption.

through centralized-decentralized-and recentralized regime.

1.2 Decentralization and deforestation in interna-tional context

Numerous studies have been conducted to investigate theeffect of decentralization on natural resources managementand the results vary from one country to another. Decentral-ization is found to have positive impact on forest investmentin Guatemala [24], Bolivia ([24]; [25]), Kenya and Mexico[25], while for Uganda it is found negatively correlated [25].For the case of Indian Himalayas, [26] found that forestsmanaged by local communities are 20–30% less lopped.

[24] conducted a comparative analysis of forestry de-centralization in Bolivia and Guatemala and they find thatdecentralized political power and fiscal power in Guatemalais more likely to encourage the mayors to invest more ontheir staffs and capital in the forestry sector compare to rela-tively stringent Bolivia. A mixed result is also found by [25]on the case of Bolivia, Kenya, Mexico and Uganda. Theirempirical finding indicates that decentralization in followedby increase forest investment in Uganda, and Mexico, a slightincrease in Bolivia, while for Kenya forest investment fallssignificantly.

Study by [27] and [28] indicate the importance of insti-tutional factor when designing decentralization in naturalresource management. [27] conducted three case studies onNueva Vizcaya province in Philippines and found that con-flicting authorities in forestry sector with almost identicalresponsibilities lead to ineffective policy implementation.[28] use a panel of 124 countries from 1980 to 2004 andconclude that countries with abundant natural resources withlow democratic tend to have higher tendency to corrupt astheir governments take on rent-seeking activities.

1.3 Decentralization and deforestation in Indone-sian context

The Regional Government Law No. 22/1999, which is consid-ered the milestone for new era of decentralization in Indone-sia, does not list forestry sector as the affair to be carried outby the central government. And neither does Regional Gov-ernment Law No. 32/2004 which replaced law No. 22/1999.The existing Forestry Law No. 41/1999, which was passedfour months after law No. 22/1999, assigns the central gov-ernment to carry out forestry resource management function.However, ministerial decree allows governors and districtheads to issue small scale forestry extraction licenses2.

These regulations were later revoked by the issuanceof Government Regulation No. 34/2002. Such regulatoryproblems – i.e., flawed division of authorities between cen-tral and regional government and inconsistencies betweenlaws – have been considered as the contributing factor todeforestation in Indonesia by [3].

The negative effect of decentralization on deforestationis also found by [4] and [1]. [4] conducted case studies in

2Governor can issue forestry extraction licenses up to 10,000 hectares(Kepmenhutbun No. 312/Kpts-II/1999), while district head can issue up to5,000 hectares (Kepmenhutbun No. 051/Kpts-II/2000).


Berau district of East Kalimantan and Kotawaringing Timurof Central Kalimantan. They found that regional autonomyhas pushed illegal logging activities even further by “legal-izing” them through applying local fees/charges. A recentstudy by [1] covers four major islands in Indonesia (Sumat-era, Kalimantan, Sulawesi and Papua) and concludes thatincreased political jurisdiction has positive correlation withdeforestation rate. They also find that deforestation occur athigher rate prior in the year prior to local election.

The effect on community payoffs logging contract in adecentralized regime is investigated by [29]. Using a surveyof 60 communities in East Kalimantan, they find that whilesome communities have successfully claimed their share, oth-ers continue to lose their forests to industrial interest for verylittle financial compensation. They conclude that the commu-nities payoff depend on the communities’ valuation over theforest. Higher valuation and greater ability to self-enforce arepositively correlated with payoffs. Communities for whomself-enforcement is costly – either they have high opportu-nity of time or low collective actions – have difficulties inclaiming an effective share in logging benefit.

2. DATA SOURCES AND EMPIRICALSPECIFICATION

2.1 Data sourcesIn order to empirically test the effect of forestry decentraliza-tion and local democratization on deforestation, we employthe same dataset as [1]. They use satellite image from Moder-ate Resolution Imaging Spectroradiometer (MODIS) datasetto calculate the changes of forest area by type from 2000 to2008 for all districts in four major islands, namely Sumat-era, Kalimantan, Sulawesi and Papua. They use the smallestscale of satellite imaging, which one pixel equals to 250mx 250m in the actual size. This means that one pixel equalsto 6.25 hectares. Multiplying the number of changes in thetotal forest area by year with a factor of 6.25 would provideus the annual change in forest area in hectares. This wouldserves as dependent variable for deforestation equation.

For the independent variables, we use data from Ministryof Finance Ministry of Home Affairs and BPS. The BPS datacovers socio economic variables which includes the shareof agriculture, mining and manufacturing sector in non-oiland gas Gross Regional Domestic Product (GRDP) whichreflect the local economic structure as the proxy for non-forestry sector resources, mean years of schooling (as theproxy for citizen control for quality of public goods). Thedata from Ministry of Finance includes the share of localown source revenue in total district revenue (as the proxy fordistrict fiscal capacity), the share of environmental spendingin total district spending (as the proxy for local preferencetoward environment) and the share of infrastructure spendingin total district spending. As [1] find statistical evidence thatdeforestation occurred at higher rate prior to local electionyear, dummy variable for lagged one and two years prior to

direct local election are included. The descriptive statisticsfor all variables are depicted in Table 1.

2.2 Empirical specificationThe empirical specification to investigate the effect of forestrydecentralization and local democratization on deforestationis as the following

Extractionit = Zβ +α1Forestit−1 +α2Decentralizationt

+ α3LocalElectionit+1

+ α4LocalElectionit+2 + ci +uitWhere (1)

Extractionit : Level of forestry extraction of district i in timet;Z : Set covariates for non-forestry endowment, which in-cludes:

• Per capita non-oil and gas GRDP to reflect the level ofeconomic development. As forestry sector is one com-ponent in GRDP, using the contemporaneous valueof per capita GRDP introduces endogeneity problem.Thus, as the solution we use lagged one year of percapita non-oil and gas GRDP.

• Contribution of agriculture, manufacturing and miningsectors in non-oil and gas GRDP.

• Mean years of schooling (as the proxy for local hu-man capital quality that enable them to for control thequality of public goods).

• Share of local own source revenue in total districtrevenue (as the proxy for district fiscal capacity). Weuse lagged value to indicate whether or not districtwith low fiscal capacity in will extract more forestryresource in the next period in order to receive higherrevenue sharing from forestry.

• Share of environmental spending in total district spend-ing (as the proxy for local preference toward environ-ment). We also use interaction term to see the impactof changes of local preference toward forestry due toforestry decentralization;

Forestit−1 : Total forest stock in previous year, which reflectsthe maximum amount that can be extracted this year;Decentralizationt : Dummy for forestry decentralization,which takes value of one for 1999–2002 period, and zeroafter 2002;LocalElectionit+1 : Dummy for one year before election,which takes value of one if it is one year before election, zerootherwise;LocalElectionit+2 : dummy for two years before election,which takes value of one if it is two years before election,zero otherwise;ci : Unobserved heterogeneity of district i;uit : Idiosyncratic error of district i in time t.

Our parameter of interest is α2, where positive and statis-tically significant result indicates that the implementation ofdecentralization - i.e., local government are given the author-ity to issue licenses for forestry extraction – leads to higherextraction of forestry resources.


There are five categories of forest which we estimate,namely convertible forest, conversion forest, production for-est, protection forest and other forest. We also estimate thetotal rate of deforestation for all categories. The level offorestry extraction is measured in natural logarithm for easeof analysis. Since there are district without particular typeforest, thus the deforestation rate is zero, the dependent vari-able is measured using log(deforestation+1) to avoid manymissing observations. For robustness check, we run usingrandom effect and fixed effect for each category of dependentvariable.

3. FINDINGS3.1 Finding from in-depth interviewsIn addition to discussion with national parliament and centralgovernment agencies, we have also conducted in-depth inter-views with relevant stakeholders in four districts receiving

large amount of forestry revenue sharing, namely Seruyanand Katingan in Central Kalimantan, and Kutai Barat andBerau in East Kalimantan. The respondents include forestryoffice, environmental office, revenue office, planning agencyand parliament at both province and district level.

Some important points from in-depth interviews withlocal stakeholders are as follow. First, the utilization of refor-estation fund (Dana Bagi Hasil Dana Reboisasi/DBH DR)– which is a part of forestry revenue sharing – is very low.Government Regulation No. 35/2002 is considered very re-strictive by local forestry offices, such that they decide notto use reforestation fund to avoid being held against the law.Second, local forestry offices claim they do not have suffi-cient fund to carry out their assigned function. On average,the annual budget for local forestry office is between IDR 4–6billion (USD 400–600 million). Third, inconsistencies be-tween forest map and ground fact is a common phenomenon.For instance, a village that has been established for years is


categorized as forest zone by the central government. Fourth,deforestation usually takes place within a particular distancefrom the forest boundaries. Thus, if a forest is seen intact inthe outside boundary, it does not mean that whole forest isstill intact, as. Fifth, agriculture and mining sector are consid-ered as significant contributor to deforestation as there havebeen massive conversions of forest zone to non-forest andsome mining activities take place even in conservation area.

3.2 Results from econometric estimationAs we can see in Table 2, we find that forestry decentral-ization before 2002 leads to higher deforestation rate. Thisresult is consistent using all type of forest. We also find thattwo years before local district head election, deforestationoccurred at a higher rate in convertible and protection forest,while for conservation forest deforestation increased one yearbefore local district head election. This finding is consistentwith [1], where they find that the deforestation rate occurs athigher rate before election.

The share of mining sector is also positive and statisti-cally significant for conservation forest and other forest. Thisfinding is also consistent with our interview results wheremining activities often take place in conservation forests. Wedo not find the evidence supporting the changes in local pref-erence toward environment can reduce the deforestation rate.This is shown by none of the interaction variable betweenenvironmental spending and forestry decentralization is neg-ative and statistically significant. This supports our in-depthinterview results that the budget allocation for local forestryoffice is very small such that they cannot effectively reducethe deforestation rate.

The regression result using fixed effect specification isavailable in Table 3. Unlike random effect specification, de-centralization variable only significant in total forest, pro-tection forest and other forest while for convertible and con-servation forest are insignificant. We also do not find thesupport for the effectiveness of local government spendingtoward environment in reducing deforestation rate. Similarly,we find evidence of the positive relationship between districthead election with deforestation for convertible, conservationand other forest.

4. SUMMARY AND POLICYRECOMMENDATION

4.1 SummaryThe empirical results provide several interesting findings.First, there are statistical evidences supporting that the defor-estation rate is higher due to the implementation of forestrydecentralization in 1999–2002. Thus, from environmentalperspective, the decision of central government to withdrawthe license for forestry extraction is justified to avoid evenhigher deforestation rate.

Second, except for production forest, there are statisticalevidences linking that local election cycle has positive cor-relation with deforestation, even in the conservation forest

(where forestry utilization is strictly prohibited) and protec-tion forest (where timber harvesting is prohibited). However,we have to be careful in interpreting this result as the forestryis a common resource. Moreover, unlike visible public goodssuch as road infrastructure where decreasing quality is easilyobserved, deforestation is less observed by the public as thesize of forest is huge. Some of local stakeholders whom wehave interviewed state that deforestation took place in theinner side of the forest, which make it difficult for the generalpublic to observe. This means that we cannot conclude whoare the actors responsible for deforestation.

Third, increased spending on environmental is ineffectivein reducing the deforestation rate. From theoretical point ofview, this indicates that the (perceived) marginal benefit fromforestry extraction is higher than its marginal cost. Formpractical point of view, small budget for local forestry officesprevent them to effectively carry out their assigned function,among them is to perform forest monitoring.

4.2 Policy recommendationAlthough Regional Government Law No. 32/2004 was passedfive years after Forestry Law No. 41/1999 and article 237 inRegional Government Law clearly requires the forestry lawto adjust and conform its regulation with regional govern-ment law, Forestry Law No. 41/1999 has not been amendeduntil 2013. As [3] pointed out that regulatory inconsistencyplay a role in deforestation in Indonesia and this study findthat allowing regional government to issue forestry licensesleads to higher deforestation, it is important for the upcom-ing regional government law to specifically include wherethe forestry resource management should be assigned. Ifthe national government wants to reduce deforestation rate,forestry resource management should be put as a function tobe carried out by the central government.

The design of forestry revenue sharing in current FiscalBalance Law No. 34/2004 should also be reconsidered. Theforestry revenue sharing formula is intended to compensatethe reduction of forestry resources with higher revenue inthe form of license fee (Dana Bagi Hasil Iuran Izin UsahaPemanfaatan Hasil Hutan/DBH IIUPHH), volume-basedroyalty (Dana Bagi Hasil Provisi Sumber Daya Hutan/DBHPSDH) and reforestation fund (DBH DR). However, underthe current system, revenue sharing from DBH IIUPHH andDBH PSDH are not earmarked. This means the spendingallocation originated from these revenues are under localgovernment discretion. For DBH DR, the revenue sharingis earmarked. Nonetheless, we find that many district gov-ernments do not use DBH DR fund to avoid being heldresponsible as the use of DBH DR is very restrictive. Thecombination of these factors leads to insufficient fundingfor the local government to perform its function in forestmonitoring. Political democratization (direct election of dis-trict head) contributes to higher deforestation rate. Given thatresource to monitor forest is limited and the fact that forestryis a common resource, this condition provides additional


incentive to illegal forestry extraction. Thus, we suggest thatthe portion of earmarked fund in the forestry revenue sharingshould be added not only in the form of DBH DR, such thatthe local authorities have sufficient revenues to perform theirfunction and prevent illegal forestry extraction.

References[1] Robin Burgess, Matthew Hansen, Benjamin A. Olken,

Peter Potapov, and Stefanie Sieber. The Political Econ-omy of Deforestation in the Tropics. NBER WorkingPapers, 17417, National Bureau of Economic Research,Inc., September 2011. [http://www.nber.org/papers/w17417. doi = 10.3386/w17417].

[2] World Bank. World Development Indicators 2013. Num-ber 13191 in World Bank Publications. The World Bank,August 2013.

[3] Luke Lazarus Arnold. Deforestation in DecentralisedIndonesia: What’s Law Got to Do With It. Law Environ-ment and Development Journal, 4:75, 2008.

[4] Anne Casson and Krystof Obidzinski. From New Orderto Regional Autonomy: Shifting Dynamics of “Illegal”Logging in Kalimantan, Indonesia. World Development,30(12):2133–2151, 2002.

[5] Charles M Tiebout. A Pure Theory of Local Expendi-tures. The Journal of Political Economy, 64(5):416–424,October 1956.

[6] Friedrich August Hayek. Individualism and EconomicOrder. University of Chicago Press, 1948.

[7] Wallace E. Oates. Fiscal Federalism. 1972.[8] Jennie Litvack, Jessica Seddon, and Junaid (Eds.) Ah-

mad. Decentralization briefing notes. World Bank Insti-tute. Washington, D.C., 1999.

http://www.nber.org/papers/w17417



[9] Torsten Persson and Guido Tabellini. Political Eco-nomics and Public Finance. NBER Working Papers,7097, National Bureau of Economic Research, April1999. [http://www.nber.org/papers/w7097.doi = 10.3386/w7097].

[10] Andrei Shleifer and Robert W. Vishny. Corruption. TheQuarterly Journal of Economics, 108(3):599–617, Au-gust 1993.

[11] Remy Prud’Homme. The Dangers of Decentralization.The World Bank Research Observer, 10(2):201–220,1995.

[12] Tao Zhang and Heng-fu Zou. Fiscal Decentralization,Public Spending, and Economic Growth in China. Jour-nal of Public Economics, 67(2):221–240, 1998.

[13] Danyang Xie, Heng-fu Zou, and Hamid Davoodi. FiscalDecentralization and Economic Growth in the United

Sstates. Journal of Urban Economics, 45(2):228–239,1999.

[14] Hamid Davoodi and Heng-fu Zou. Fiscal Decentral-ization and Economic Growth: A Cross-Country Study.Journal of Urban Economics, 43(2):244–257, 1998.

[15] Nobuo Akai and Masayo Sakata. Fiscal DecentralizationContributes to Economic Growth: Evidence from State-Level Cross-Section Data for the United States. Journalof Urban Economics, 52(1):93–108, 2002.

[16] Justin Yifu Lin and Zhiqiang Liu. Fiscal Decentraliza-tion and Economic Growth in China. Economic Devel-opment and Cultural Change, 49(1):1–21, 2000.

[17] Ulrich Thiessen. Fiscal Federalism in Western Euro-pean and Selected Other Countries: Centralization orDecentralization? What Is Better for Economic Growth?



Discussion Papers of DIW Berlin 224, DIW Berlin, Ger-man Institute for Economic Research, August 2000.

[18] Gary M. Woller and Kerk Phillips. Fiscal Decentralisa-tion and IDC Economic Growth: An Empirical Investi-gation. The Journal of Development Studies, 34(4):139–148, 1998.

[19] Jorge Martinez-Vazquez and Robert M. McNab. FiscalDecentralization and Economic Growth. World Devel-opment, 31(9):1597–1616, 2003.

[20] Larry Chavis. Decentralizing Development: AllocatingPublic Goods Via Competition. Journal of DevelopmentEconomics, 93(2):264–274, 2010.

[21] Jeff Huther and Anwar Shah. A Simple Measure ofGood Governance and Its Application to the Debate onthe Appropriate Level of Fiscal Decentralization. WorldBank Policy Research Working Paper, 1894, 1998.

[22] Raymond Fisman and Roberta Gatti. Decentralizationand Corruption: Evidence Across Countries. Journal ofPublic Economics, 83(3):325–345, 2002.

[23] Daniel Treisman. The Causes of Corruption: A Cross-National Study. Journal of Public Economics, 76(3):399–457, 2000.

[24] Krister P. Andersson, Clark C. Gibson, and FabriceLehoucq. Municipal Politics and Forest Governance:Comparative Analysis of Decentralization in Boliviaand Guatemala. World Development, 34(3):576–595,2006.

[25] Eric A. Coleman and Forrest D. Fleischman. ComparingForest Decentralization and Local Institutional Changein Bolivia, Kenya, Mexico, and Uganda. World Devel-opment, 40(4):836–849, 2012.

[26] Jean-Marie Baland, Pranab Bardhan, Sanghamitra Das,and Dilip Mookherjee. Forests to the People: Decentral-ization and Forest Degradation in the Indian Himalayas.World Development, 38(11):1642–1656, 2010.

[27] Kulbhushan Balooni, Juan M. Pulhin, and Makoto Inoue.The Effectiveness of Decentralisation Reforms in thePhilippines’s Forestry Sector. Geoforum, 39(6):2122–2131, 2008.

[28] Sambit Bhattacharyya and Roland Hodler. Natural Re-sources, Democracy and Corruption. European Eco-nomic Review, 54(4):608–621, 2010.

[29] Stefanie Engel and Charles Palmer. Who Owns theRight? The Determinants of Community Benefits fromLogging in Indonesia. Forest Policy and Economics,8(4):434–446, 2006.

Gedung LPEM FEUIJl. Salemba Raya No. 4, Jakarta 10430Telepon: (021) 3143177 ext. 621/623;Fax: (021) 3907235/31934310

9lg $gulvrq - lpem

Documents