Causes for Deforestation in Indonesia: Corruption and Palm Tree Plantation

Abstract

Indonesian biodiversity is threatened by massive deforestation. In this research paper, claims that deforestation in Indonesia is caused by corruption and supported by crude palm oil production is verified using time series analysis. Using Engel Granger cointegration test, three time series of data, specifically corruption perception index, rate of deforestation and price of crude palm oil are inspected for a long-run relationship. Test statistics suggests that there is no long-run relationship among these variables. Authors provide several explanations for this result. For example, corruption in Indonesia, as measured by CPI is still very high. This may mean that forest cover loss is possible even though there is a positive change in corruption level. According to the results, crude palm oil price has also no effect upon forest cover loss. This is likely due to very low shut-down price of crude palm oil for which production is still economical.

Full text

Asian Social Science; Vol. 11, No. 27; 2015
ISSN 1911-2017 E-ISSN 1911-2025
Published by Canadian Center of Science and Education
120
Causes for Deforestation in Indonesia: Corruption and Palm Tree
Plantation
Osama Eldeeb1, Petr Prochazka1 & Mansoor Maitah1
1 Faculty of Economics and Management, Czech University of Life Sciences, Czech Republic
Correspondence: Mansoor Maitah, Faculty of Economics and Management, Czech University of Life Sciences
Prague, Kamycka 129, 165 21 Praha 6 – Suchdol, Czech Republic. Tel: 420-23-438-2139. E-mail:
maitah@pef.czu.cz
Received: July 23, 2015 Accepted: August 3, 2015 Online Published: November 20, 2015
doi:10.5539/ass.v11n27p120 URL: http://dx.doi.org/10.5539/ass.v11n27p120
Abstract
Indonesian biodiversity is threatened by massive deforestation. In this research paper, claims that deforestation in
Indonesia is caused by corruption and supported by crude palm oil production is verified using time series
analysis. Using Engel Granger cointegration test, three time series of data, specifically corruption perception
index, rate of deforestation and price of crude palm oil are inspected for a long-run relationship. Test statistics
suggests that there is no long-run relationship among these variables. Authors provide several explanations for
this result. For example, corruption in Indonesia, as measured by CPI is still very high. This may mean that
forest cover loss is possible even though there is a positive change in corruption level. According to the results,
crude palm oil price has also no effect upon forest cover loss. This is likely due to very low shut-down price of
crude palm oil for which production is still economical.
1. Introduction
It is estimated that that corruption in the forestry sector reached over Rp 273 trillion in 2011 in Indonesia.
(Jakarta Globe 2013) Corruption by government officials linked to natural resource extraction is leading to
billions of dollars in state losses annually in Indonesia (Jakarta Globe 2013). One of the most pressing
environmental issues possibly related to corruption in Indonesia is deforestation. Deforested land is transformed
usually to palm tree plantations. Therefore, it is necessary to examine a relationship between deforestation and
palm trees plantation, respectively corruption.
1.1 Deforestation in Indonesia
Deforestation in Indonesia not only possess a significant threat to local biosphere through biodiversity reduction
but also influences global climate. Tropical deforestation from developing countries, including Indonesia,
contributes to emissions of greenhouse gases. (Margono et al., 2014) This is mainly carbon dioxide, the main
cause of global warming. Primary forest clearing also leads to the loss of biodiversity due to the destruction of
unique tropical forest habitats. (Margono et al., 2014)
Present understanding of forest change within Indonesia is unclear. A report from the United Nations Food and
Agricultural Organization (UNFAO) claims that the rate of deforestation in Indonesia is approximately 0.31
million hectates per year from 2000 to 2005. Same report claims that from 2005 to 2010 roughly 0.69 millions of
hectares were lost. A report from the Indonesian Ministry of Forestry provides an estimate of approximately 0.4
million hectares lost between 2009 and 2011. These estimates vary significantly and therefore a study has been
conducted that uses spatial as well as temporal data of digital imaginary to show actual deforestation dynamic. A
study done by Reference shows that Indonesia is a country with the highest rate of increasing forest covers loss
from 2000 to 2012. (Margono et al., 2014) Deforestation is also cited by (Vidal, 2015) where two areas of the
Indonesian islands of Sumatera and Kalimantan (the Indonesian part of the island of Borneo) are used as an
example to show fastest forest cover losses in the humid tropics. This has according to (Vidal, 2015) implications
for carbon dynamics, biodiversity conservation, and local livelihoods.
1.2 Palm Trees Plantations in Indonesia and Deforestation
Agricultural commodities, particularly energy one have gained a lot of significance in international trade (Svatos
et al., 2013; Svatos & Smutka, 2012). There is an increase in aggregate global demand for palm oil. This increase

www.ccsenet.org/ass Asian Social Science Vol. 11, No. 27; 2015
121
is obvious especially in the last several years. (World Watch, 2015) Palm oil is the most common, globally
produced palm. Its yield belongs to one of the highest when compare to other crops. Regarding the costs of
production and refining, palm oil belongs to the least expensive oils. (World Watch, 2015). Production areas
devoted to palm oil belong currently to the most significant contributors of rain forest destruction in Indonesia.
According to UNEP (2015), there is a possibility that no later than 2032, most rainforest in Indonesia would be
irreversibly degraded. If we examine the speed and volume of deforestation in the last six years, this prediction
may seem to be overly optimistic, as oil palm plantations and biodiesel refineries are getting more and more
popular. Newest estimates even conclude that only 2 percent of rain forest can be preserved by 2022 with even
less forest being saved in the lowlands of Indonesia. (UNEP, 2015)
1.3 Corruption in Indonesia
Corruption linked to the forestry, plantation industries and mining leads potentially to billions of dollars in state
budget losses each year in Indonesia. Natural resources are in such a way never-ending income for corrupt
officials (Jakarta Globe, 2013). In this case, we are talking about systematic corruption which is created
regardless who is the corrupted person. This is the number one cause of corruption in the environmental sector
(Winbourne, 2002). Another cause is relatively a lot of rights issued to government representatives, lack of
accountability for these representatives and also little possibility to adhere to transparency in their decisions.
Private interest of such public officials in revenues that can be obtained from environmental resources is much
stronger than public interests in environment. (Winbourne, 2002) Government controlled sectors in Indonesia are
especially energy, road, railway and port projects. (Jakarta Globe, 2013) Total volume of corruption can be
manifested by an example of Indonesian reforestation fund that lacks more than $5.2 billion that was supposedly
lost in 1993 to 1997 due to bad management and fraudulent behavior, according to an audit by Ernst and Young.
(E360.yale.edu 2015)
According to E360.yale.edu (2015), deforestation binge has been abetted by rampant and systemic corruption in
the Indonesian government, military, and forestry sector (E360.yale.edu 2015)
2. Analysis of Deforestation Due to Palm Trees and Corruption
According to (Koyuncu & Yilmaz, 2009) corruption contributes to deforestation. Koyuncu and Yilmaz (2009)
found a statistically significant strong positive relation between corruption and deforestation for different periods
across different countries (Koyuncu & Yilmaz, 2009). Authors in this paper re-examine this relationship in
Indonesia by utilizing time series approach of cointegration. In order to quantify corruption, a corruption
perception index is used. This is based upon an assumption that a ranking of countries according to the extent
corruption exist can be made. The corruption perception index was created in 1995 by Transparency
International. It ranks almost 200 countries on a scale of zero to 10, with zero indicating high levels of corruption
and 10 indicating low levels (Koyuncu & Yilmaz, 2009)
Source: own compilation
Figure 1. Deforestation, corruption perception index and crude palm oil price
0
0.5
1
1.5
2
2.5
3
3.5
2002 2004 2006 2008 2010 2012
200
300
400
500
600
700
800
900
1000
1100
deforest (left)
cpi (left)
palmoil (right)

www.ccsenet.org/ass Asian Social Science Vol. 11, No. 27; 2015
122
At the same time, a second variable is added that shows influence of palm tree plantations upon deforestation.
Assumption is made that as the price of crude palm oil goes up, and there is more incentive to deforest more land
in Indonesia. All three variables are depicted in Figure 1. The figure shows deforestation, corruption perception
index (CPI) and crude palm oil price in Indonesia between 2001 and 2012.
Data in Figure 1 show that all three time series trend upwards. This is positive in terms of CPI but quite negative
in terms of forest cover loss. Price of crude oil also slopes upward which may mean that producing palm trees
may be more prosperous.
2.1 Testing for Unit Root
As previously mentioned, it is obvious from the Figure 1 that all three time series are trending. This is a useful
piece of information for further analysis. The entire process starts with testing individual series for existence of
unit root. A null hypothesis Ho: Level series contains a unit root is tested against an alternative hypothesis Ha:
Level series does not contain a unit root. This is done using Augmented Dickey-Fuller test with the contant and
trend due to trending data. All time series are tested and results are presented in the following Tables 1,2 and 3.
Due to relatively short time series, a lag was selected to 2.
Table 1. Augmented Dickey Fuller Test for Deforestation in Indonesia with constant and trend
Model definition: (1-L)y = b0 + b1*t + (a-1)*y(-1) + ... + e
1st-order autocorrelation coeff. for e: -0.231
lagged differences: F(2, 4) = 0.498 [0.6410]
estimated value of (a - 1): -1.50846
test statistic: tau_ct(1) = -1.95282
asymptotic p-value 0.6264
The test results presented in Table 1 show that the time series needs to be differenced one to make it stationary.
The same procedure is done for the other two variables of corruption perception index and crude palm oil.
Results are presented in Table 2 and Table 3 respectively.
Table 2. Augmented Dickey Fuller Test for Corruption Perception Index in Indonesia with constant and trend
Model definition: (1-L)y = b0 + b1*t + (a-1)*y(-1) + ... + e
1st-order autocorrelation coefficient for e: -0.372
lagged differences: F(2, 4) = 3.396 [0.1374]
estimated value of (a - 1): -2.14412
test statistic: tau_ct(1) = -3.40771
asymptotic p-value 0.05028
Table 3. Augmented Dickey Fuller Test for Crude Palm Oil Price with constant and trend
Model definition: (1-L)y = b0 + b1*t + (a-1)*y(-1) + ... + e
1st-order autocorrelation coefficient for e: -0.136
lagged differences: F(2, 4) = 2.052 [0.2436]
estimated value of (a - 1): -1.44904
test statistic: tau_ct(1) = -1.59503
asymptotic p-value 0.7954
2.2 Cointegration Test
Tests presented above show that all time series have a unit root of order I(1). Now, a test for cointegration can be

www.ccsenet.org/ass Asian Social Science Vol. 11, No. 27; 2015
123
conducted using Engel-Granger methodology. This is used according to original Engle and Granger (1987)
article where a cointegration test is proposed. This test is made up of several sections. First, estimation of a
cointegration regression is done using ordinary least squares. This involves capturing the residual error terms.
Consequently a unit root test for these error terms is conducted. In order to proceed with the test, authors in their
paper suggest testing the null hypothesis that the error term has a unit root against the alternative that it has a root
less than unity. Since error terms are themselves estimates, new critical values have to be used. In this case, the
corrected MacKinnon critical values are used for further estimation. (Empiwifo.uni-freiburg.de, 2015)
For the analysis of cointegration a regression is set so that
yt= β1+ β2x1t+β3x2t+et
where yt is deforestation, x1t is corruption and x2t is price of crude palm oil. The test results are presented in the
Table 4 of this paper.
Table 4. Cointegrating regression- OLS, using observations 2001-2012 (T = 12)
Dependent variable: deforest Coefficient Standard error t-ratio probability value
constant -0.404088 0.207970 -1.943 0.0839 *
cpi 0.417674 0.129552 3.224 0.0104 **
palmoil -0.000207925 0.000215823 -0.9634 0.3605
Mean dependent variable 0.478333 S.D. dependent var 0.166287
Sum squared residual 0.077178 S.E. of regression 0.092603
R-squared 0.746265 Adjusted R-squared 0.689879
Log-likelihood 13.25204 Akaike criterion -20.50407
Schwarz criterion -19.04935 Hannan-Quinn -21.04266
rho -0.099565 Durbin-Watson 1.981004
Step 2: testing for a unit root in uhat
Augmented Dickey-Fuller test for uhat
including one lag of (1-L)uhat
(max was 2, criterion modified AIC)
sample size 10
unit-root null hypothesis: a = 1
model definition: (1-L)y = (a-1)*y(-1) + ... + e
1st-order autocorrelation coeff. for e: -0.056
estimated value of (a - 1): -1.10109
test statistic: tau_c(3) = -2.20657
The test shows that the p-value is large so we cannot reject the null hypothesis that the series have a unit root, i.e.
they are not cointegrated. This means that corruption as represented by the corruption perception index and price
of crude palm oil does not have a long-run relationship with deforestation in Indonesia. There may be several
reasons for this result. Authors assume that one of the reasons may be that CPI does not capture well the type of
corruption behavior that enables deforestation. Other reason may be that corruption in Indonesia is still very high
and even though some improvements are evident from the data, it still enables uncontrolled deforestation. The
reason why crude palm oil price has no impact upon deforestation may be explained by economic feasibility of
palm oil production that is large enough even with a relatively lower price of crude palm oil.
3. Conclusions
Deforestation in Indonesia is a large problem that needs to be solved. Authors in this paper try to justify claims
that deforestation in Indonesia is caused by corruption and that it is supported by crude palm oil production that

www.ccsenet.org/ass Asian Social Science Vol. 11, No. 27; 2015
124
has spread over the deforested area. Using cointegration approach three time series of data, specifically
corruption perception index, rate of deforestation and price of crude palm oil are used to find a long-run
relationship. Test results show that there is no long-run relationship among these variables. Authors acknowledge
that . there may be several reasons for this result. Firstly, corruption in Indonesia, as measured by CPI is still
very high. This may mean that uncontrolled deforestation is still enabled. Similarly, analysis shows that crude
palm oil price has no impact upon deforestation. This is probably to very low shut-down price of crude palm oil
which production is economical even at lower price levels.
References
E360.yale.edu. (2015, July 23). Indonesia’S Corruption Legacy Clouds A Forest Protection Plan By Rhett Butler:
Yale Environment 360.
Empiwifo.uni-freiburg.de. (2015). Empirische Wirtschaftsforschung Und Ökonometrie.
Koyuncu, C., & Yilmaz, R. (2011). Impact of Private Forest Ownership on Deforestation and Poverty. Quality &
Quantity, 47(3), 1657-1664.
Margono, B. A. et al. (2014). Primary Forest Cover Loss in Indonesia Over 2000–2012. Nature Climate Change,
4(8), 730-735.
Svatoš, M., & Smutka, L. (2012). Development of agricultural trade and competitiveness of the commodity
structures of individual countries of the Visegrad Group. Agricultural Economics (Czech Republic), 58(5),
222-238.
Svatoš, M., Smutka, L., Elshibani, B. A. M., & Mousbah, S. A. A. (2013). Development of visegrad countries'
agricultural trade in relation to agricultural production development. Agris On-line Papers in Economics
and Informatics, 5(1), 61-71.
The Jakarta Globe. (2013). With Decentralized Power Comes Graft in Indonesia.
UNEP-WCMC's official website. The Last Stand of the Orangutan. UNEP-WCMC.
Vidal, J. (2013). The Sumatran Rainforest Will Mostly Disappear Within 20 Years. The Guardian.
Winbourne, S. (2002, November). Management Systems International 600 Water Street SW, Washington, DC
20024 USA. Authored by: Svetlana Winbourne.
Worldwatch.org. (2015, July 23). Global Palm Oil Demand Fueling Deforestation Worldwatch Institute.
Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution
license (http://creativecommons.org/licenses/by/3.0/).