Content uploaded by Truly Santika
Author content
All content in this area was uploaded by Truly Santika on Apr 25, 2019
Content may be subject to copyright.
Accepted paper in journal World Development (17 April 2019)
Does oil palm agriculture help alleviate poverty? A
multidimensional counterfactual assessment of oil palm
development in Indonesia
Authors:
Truly Santika 1,2,3,4,*, Kerrie A. Wilson 1,5, Sugeng Budiharta 1,6, Elizabeth A. Law 1,7, Tun Min Poh 2,
Marc Ancrenaz 2,3,8, Matthew Struebig 4 & Erik Meijaard 1,2,3,4
Affiliations
1 ARC Centre of Excellence for Environmental Decisions, The University of Queensland, Brisbane, QLD,
Australia
2 Borneo Futures, Brunei Darussalam
3 IUCN Oil Palm Task Force
4 Durrell Institute of Conservation and Ecology (DICE), University of Kent, Canterbury, CT2 7NR, United
Kingdom
5 Institute for Future Environments, Queensland University of Technology, Brisbane, Australia
6 Purwodadi Botanic Garden-Indonesian Institute of Sciences, Jl. Surabaya-Malang Km. 65, Pasuruan,
Jawa Timur, Indonesia
7 Norwegian Institute for Nature Research (NINA), Trondheim, Norway
8 Kinabatangan Orang-utan Conservation Programme, Sandakan, Sabah, Malaysia
* Corresponding author:
Truly Santika
E-mail: trulysantika@gmail.com
1
Abstract
Palm oil producing countries regularly promote the positive impact of palm oil agriculture on poverty
alleviation, despite limited evidence about the contribution of this crop on village well-being. Past
evaluations that quantify the social impact of oil palm are dominated by localized studies, which
complicate the detection of generalizable findings. Moreover, only a few of these evaluations are based
on rigorous case-control studies, which limits the robustness of the conclusions. Here we examined the
association between the development of oil palm plantations and change in objective or material well-
being between 2000 and 2014 across villages in Kalimantan, Indonesian Borneo. We applied a matching
method to evaluate the impacts of oil palm plantations across different aspects of well-being, accounting
for varying time delays in the accrual and realization of benefits after plantation development. Our study
reveals that the social impacts of oil-palm plantations are not uniformly positive, nor negative, and have
varied systematically with biophysical locations and baseline socioeconomic conditions of nearby
communities prior to oil palm development. Plantations developed in villages with low to moderate forest
cover, in which the majority of communities already relied on market-oriented livelihoods, were
associated with improved socioeconomic well-being compared to villages without oil palm development.
However, we found the opposite for plantations developed in remote villages with higher forest cover, in
which the majority of communities previously relied on subsistence-based livelihoods. Overall, oil palm
growing villages were more associated with reduced rate of improvement of social and environmental
well-being compared to villages without oil palm development, regardless of location and baseline
community livelihoods. Our findings highlight an urgent need for careful evaluation and planning in the
development of oil palm agriculture in remote forested areas. For oil palm regions that have been
developed, our study shows that unsustainable livelihoods, increased socioeconomic disparity, and
environmental issues remain major challenges.
Keywords: human well-being; impact evaluation; Indonesia; landscape analysis; multidimensional poverty;
oil palm
2
1. Introduction
In recent decades there has been an enormous increase in the demand for oils and fats worldwide due to
growing populations and wealth: the global consumption rose from 80 million to 213 million tonnes
between 1990 and 2015 (FAO 2016). Palm oil production has boomed in response to this demand, with
the global production increasing from 14 million to 63 million tonnes over the same period (FAO 2016).
Globally, oil palm plantations have expanded from 6 to 18 million hectares over the last two decades, and
now account for 0.4% of the world’s permanent cropland (FAO 2016). Indonesia and Malaysia is currently
the world’s leading palm oil producer, supplying approximately more than 80% of the commodity globally
(FAO 2016). In Indonesia, plantations have predominantly been developed in Sumatra and Kalimantan
(Indonesian Borneo), with rapid expansion in recent decades (Fig. A1 and A2 in Appendix A; Directorate
General of Estate Crops Indonesia 2015). Prior to 2000, oil palm plantations were developed mainly on
barren land (e.g. burnt-over areas) or on established farmland, but conversion of forest to plantations has
become increasingly common in recent years due to increasingly scarce non-forest land for agriculture
(Fig. A3a; Santika et al. 2015; Gaveau et al. 2016b). Agricultural expansion into forested areas could have
profound impacts on local and traditional communities who depend highly on forest produce and income
(Sheil et al. 2009).
In 2017, the Indonesian oil palm industry was estimated to employ 3.8 million people (Directorate
General of Estate Crops 2017), ca. 2.4% of the total Indonesian work force (Allen 2016), and the
Indonesian government increasingly promotes oil palm cultivation as a way to alleviate poverty and
advance development in remote forested landscapes (Cooke 2012; Potter 2012; Li 2016). Other
developing and emerging countries, such as Brazil, Peru, Colombia, Nigeria, Gabon, Ghana, and Rwanda,
are following similar steps (Villela et al. 2014; Byerlee et al. 2017; Meijaard et al. 2018). However, despite
driving macro-level economic growth in many tropical countries, there are mixed reports about the social
impact of oil palm agriculture on the life of communities in production areas. While positive impacts on
household income and consumption, as well as local development and employment have been observed,
numerous social and environmental costs of oil palm have also been reported (reviewed in Table B1 in
Appendix B). For example, in some areas the promised benefits and compensation of oil palm cultivation
has not reached all smallholders (Sheil et al. 2009; Rist et al. 2010), and consultation with indigenous
peoples has not always occurred (Colchester et al. 2006), leading to escalating conflicts with local people
(Abram et al. 2017; Wakker et al. 2004). Expansion of the oil palm industry in some areas has increased
income benefits mainly among wealthy farmers and skilled migrants while marginalising others, leading to
social disparities (Obidzinski et al. 2012, 2014; Zen et al. 2016). In addition, depletion of land and water
resources, and deterioration of air and water quality due to unsustainable oil palm practices have further
burdened communities in some locations (Phalan 2009; Sheil et al. 2009; Wells et al. 2016).
Past evaluations of the social impact of industrial-scale oil palm plantations in developing
countries have typically been appraised in localized studies, either within district, province or state (Table
B1). A few exceptions are Obidzinski and colleagues (2012) who studied the perceptions of the social
impact of oil palm plantations among three indigenous communities in the province of West Kalimantan,
Papua, and West Papua, and Castiblanco, Etter and Ramirez (2015) who studied the social impact of oil
palm across municipalities in Colombia. Furthermore, only a few of these past evaluations were based on
rigorous case-control studies (Table B1). In complex social and environmental settings, there are many
factors other than the intervention that may cause or moderate change, so quantifying the impact of an
intervention is impossible without identifying an appropriate comparator or counterfactual of what would
have happened in the absence of that intervention (Ferraro 2009). Additionally, studies that apply a
counterfactual analysis approach for evaluating the social impact of oil palm are mostly based on data
from Sumatra, and geographically biased around transmigration areas in the province of Riau and Jambi
(e.g. Alwarritzi et al. 2015; Euler et al. 2017; Gatto et al. 2017; Krishna et al. 2017) (Table B1). These
provinces are not only recognized as the hotspots of Indonesia's recent oil palm boom, but also the
hotspots of transmigration programs during the New Order regime in 1965-1998 (the period from Pra-
Pelita to Pelita IV), when at least 680,000 transmigrants from Java relocated in both provinces combined
3
(Junaidi 2012; Palupi et al. 2017). Thus, the baseline socioeconomic context and socio-political history of
the oil palm growing areas associated with these studies likely have limited transferability of the resulting
conclusions to other oil palm areas where recent migration was considered to be less prevalent prior to oil
palm developments, such as in many parts of Kalimantan and Papua.
Our literature review (Table B1) highlights substantial variability of the social impact of oil palm
among locations with different biophysical characteristics and baseline socioeconomic conditions of
communities. There is therefore a clear need for greater synthesis to detect generalizable findings, which
can come from landscape-based assessment over broad areas. Understanding variation in how social
benefits accrue from oil palm developments is critical to help identify opportunities to improve policy
implementation and ultimately improve outcomes for village communities (Hodbod & Tomei 2013;
Blaber-Wegg et al. 2015). Therefore, a robust evaluation of the impact of the oil palm industry requires
analyses of (1) changes in indicators of well-being due to the oil palm development compared to the
counterfactual condition without oil palm, (2) trends over time to account for potential delays in the
accrual and realization of benefits, and (3) multiple sites over broad areas to detect variations in the
impacts of oil palm across different baseline biophysical and community socioeconomic conditions.
Here we assessed the impact of oil palm plantation developments on changes in objective or
material aspects of well-being between 2000 and 2014 across villages in Kalimantan, Indonesian Borneo.
Oil palm plantations mainly include industrial scale plantations developed within concession boundaries
(including smallholder plantings operated as part of the nucleus estate system, i.e. cooperation between
company plantations and smallholders in terms of capital and labour supply) and to a smaller extent
independent smallholders (see Fig. A1a). We note that industrial scale plantations often spatially coincide
with independent smallholder plantations (Euler et al. 2016; Jelsma et al. 2017), and that our results
should be interpreted as an overall effect of oil palm development, across different production scales, on
village well-being. As plantations require some time to mature, be harvested, and provide economic
benefits (Rist et al. 2010), we investigated the change of well-being following 2-3, 6-8, and 11-14 years
after initial plantation development within the village administrative unit. We defined well-being across
five dimensions: (1) basic (i.e. living conditions), (2) physical (i.e. infrastructure), (3) financial (i.e. income
support), (4) social (i.e. security and social equity), and (5) environmental (i.e. prevention of natural
hazards). We used 17 indicators from the village level dataset Potensi Desa (PODES) as proxies for these
aspects of well-being (Table 1). PODES data were collected by the Bureau of Statistics Indonesia (BPS)
roughly every three years between 2000 and 2014 (BPS 2015), providing the best data option for this
broad-scale analysis. We recognise that more subjective, non-material, indicators exist to measure human
well-being (Costanza et al. 2007; Daniel et al. 2012). However, these are difficult to aggregate at the
village-level and are not available within the PODES dataset.
The effect of oil palm plantations on village well-being was assessed based on a counterfactual
analysis: comparing what actually happened and what would have happened in the absence of the oil
palm plantations. For this counterfactual, we employed a matching method (Dehejia & Wahba 2002) to
select a set of control villages outside oil palm growing areas that exhibited the same baseline
characteristics as villages within the oil palm areas. By doing so, we controlled for variables that could
confound the analysis, such as socio-political factors, accessibility, agricultural productivity, and recent
climate, as well as baseline village well-being and primary livelihoods (Table 2). We explored how the
effect of oil palm plantations varied across different village primary livelihoods prior to plantation
development, which includes (1) subsistence-based livelihoods, i.e. swidden farming of dryland rice (inter-
cropped with banana, maize, and cassava, and typically supplemented by market exchange for forest
and/or agroforest products) and inland fishing (Budiharta et al. 2016), and (2) market-oriented livelihoods,
including polyculture plantations (e.g. coffee, rubber, oil palm, coconut), horticulture, aquaculture,
agricultural services, and non-agricultural sectors.
4
Table 1. PODES indicators used as proxies for five dimensions of well-being: basic, physical, financial, social, and
environmental. Variable wk denotes the directional effect of the change in indicator k that defines
improvement in well-being. If wk=1, then positive change (i.e. an increase) in indicator k represents
improvement in well-being. If wk=-1, then negative change (i.e. a reduction) in indicator k represents
improvement in well-being.
Dimension of
well-being
PODES
indicator
(k)
Description wk Response
POOR Proportions of households with poor
housing conditions *
-1 Continuous
ELCT Proportions of households with
electricity *
1 Continuous
COOK Cooking fuel for majority of households
-1 Categorical (1=electricity or
liquefied petroleum gas (LPG),
2=kerosene, 3=wood/others)
TOLT Toilet facilities for majority of
households
-1 Categorical (1=own toilet,
2=joint toilet, 3=public toilet,
4=non-toilet)
Basic
(Living
conditions)
MLNT Child malnutrition incidence in the last
year †
-1 Continuous
HEAL Distance to nearest healthcare facility -1 Continuous
PSCH Distance to nearest primary school -1 Continuous
Physical
(Infrastructure)
SSCH Distance to nearest secondary school -1 Continuous
COOP Number of active village cooperative
schemes or other schemes ‡
1 Continuous
CRDT Number of credit facilities for farmers
or communies ‡
1 Continuous
Financial
(Income support)
SIND Number of small industries (<20
employees) ‡
1 Continuous
CNFL Frequency of conflicts among
communities in the last year
-1 Continuous
AGLB Proportion of families with agricultural
wage labourers **
-1 Continuous
Social
(Security and
social equity)
SUIC Suicidal rates in the last year † 1 Continuous
WPOL Water pollution over the last 3 years -1 Categorical (1=none, 2=mild,
3=severe)
APOL Air pollution over the last 3 years -1 Categorical (1=none, 2=mild,
3=severe)
Environmental
(Natural hazard
prevention)
FLOD Frequency of floods and landslides over
the last 3 years
-1 Continuous
† per 1000 people, ‡ per 100 households, * of total households, ** of total agricultural families
5
Table 2. Variables used to assess oil-palm efficacy in improving village well-being.
Variable
Description (Static/Dynamic §) Type (Scale) Data source
SOCIO-POLITICAL
KABU Regency (kabupaten) boundaries
(Dynamic)
Categorical Indonesia Population Census
2010 (BPS 2010)
LZON Majority of legalized land use zone
(Static)
Categorical
(HP=Production Forest;
APL=Non Forest Estate)
Forest Zone Map (MEF 2010)
ACCESSIBILITY
ELEV Mean elevation (Static) Continuous (m a.s.l) SRTM 90m Digital Elevation
Database v4.1 (Jarvis et al.
2008)
SLOP Mean slope (Static) Continuous (degree) SRTM 90m Digital Elevation
Database v4.1 (Jarvis et al.
2008)
CITY Mean distance to large cities or
arterial roads (Static)
Continuous (log(km)) Provincial map from Geospatial
Information Agency Indonesia
(BIG 2016)
POPB Mean human population density per
km2 prior to of oil palm development
(Dynamic)
Continuous (log(people)) Potensi Desa (PODES) (BPS
2015)
FORB % natural forest cover a year prior to
oil-palm development (Dynamic)
Categorical
(1=0-25%; 2=25-50%;
3=50-75%; 4=75-100%)
Global Forest Change dataset
(Hansen et al. 2013) and
Indonesia’s primary and
secondary forest map (Margono
et al. 2014)
AGRICULTURAL PRODUCTIVITY
SDRY Mean long-term monthly rainfall
during dry season (Static)
Continuous (mm) WorldClim (Fick & Hijmans
2017)
SWET Mean long-term monthly rainfall
during wet season (Static)
Continuous (mm) WorldClim (Fick & Hijmans
2017)
SOIL Presence of peat soil (Static) Categorical
(1=Peat; 0=Mineral)
Peat Hydrological Area Map
(MEF 2017)
TRNS Mean distance to transmigration areas
prior to oil palm development
(Dynamic)
Continuous (log(km)) Land Cover Map (MEF 2016)
BASELINE VILLAGE SOCIOECONOMIC CONDITION
LVHD Village primary livelihood at initial
stage of oil palm development
(Dynamic)
Categorical
(SL=Subsistence livelihoods;
ML=Market-oriented
livelihoods)
Potensi Desa (PODES) (BPS
2015)
WLBNk Baseline well-being indicator k (Table
1) at initial stage of oil palm
development (Dynamic)
Either continuous or
categorical
Potensi Desa (PODES) (BPS
2015)
VILA Extent of village (Static) Continuous (km2) Potensi Desa (PODES) (BPS
2015)
RECENT CLIMATE
TDRY Mean monthly rainfall during the dry
season over the time frame assessed
(Dynamic)
Continuous (mm) TRMM Multi-Satellite
Precipitation Analysis (TMPA) v.
7 (Huffman et al. 2007)
TWET Mean monthly rainfall during the wet
season over the time frame assessed
(Dynamic)
Continuous (mm) TRMM Multi-Satellite
Precipitation Analysis (TMPA) v.
7 (Huffman et al. 2007)
§ Static: vary spatially but are fixed through time, and Dynamic: vary both spatially and temporally.
6
2. Materials and Methods
2.1. Study area
Our study covered Kalimantan (540,000 km2), the Indonesian portion of the island of Borneo, comprising
five provinces and 6,600 villages in 2010 (BPS 2010). Kalimantan’s interior is largely hilly and mountainous,
but extensive lowlands and swamps occur along the coasts. A large part of the region is drained by
navigable rivers. In 2015, the population of Kalimantan was estimated as 15 million (BPS 2015), with many
mainly residing along the coastline (Fig. 1A). Communities are comprised of three broad ethnic groups: the
indigenous Dayak predominantly residing in the interior regions, the indigenous Melayu (of which the
majority are Muslim) occupying mainly the coastal regions, and recent migrants from the neighbouring
islands of Java, Madura and Bali (arriving mainly as part of the government-supported transmigration
programs since the 1960s) (Fig. 1B). Village economies in Kalimantan have traditionally been based on
subsistence agriculture (i.e. swidden farming and seasonal inland fishing) and forestry, but plantation
agriculture (both polyculture and monoculture) has become an increasingly vital sector in recent decades
(Fig. 1C).
Fig. 1. (A) Human population density (per km2), and (B) village dominant ethnic group, in Kalimantan based on
PODES 2011 (data closely reflect the results of 2010 national census). (C) Change in village primary livelihood
sectors between 2000 and 2014, which includes subsistence farming, seasonal inland fishing, forestry,
plantation (polyculture or monoculture), and other sectors (including horticulture, aquaculture, agricultural
services, and non-agricultural sectors). Black lines in the maps indicate provincial boundaries.
7
2.2. Oil palm plantations
We used spatial data on the boundaries of planted oil palm plantations every five years between 1995 and
2015 provided by Gaveau et al. (2016a). These data comprise estates located within the boundaries of oil
palm concessions, including large-scale industrial plantations, smallholders under the Perkebunan Inti
Rakyat (PIR) or Nucleus Estate Smallholder schemes in transmigration areas (McCarthy et al. 2012), and a
smaller extent of independent smallholders (outside oil palm concessions) (Fig. A1a). It is worth noting
that the proportion of industrial plantation estates in Kalimantan is much higher than in neighbouring
Sumatra (64% of all plantations in Kalimantan, 33% in Sumatra, in 2015, Fig. A1). Oil palm developments at
different spatial scales (smallholder to large industrial scale) usually coincide spatially due to their reliance
on access to palm oil mills and related services (Klasen at al. 2016), hence our analysis reflects more
broadly the effects of all oil palm developments. In addition, the proportion of oil palm plantations that
have been certified through internationally recognized certification bodies, such as the Roundtable on
Sustainable Palm Oil (RSPO), between 2000 and 2015 is relatively small (around 12% by 2015) (Fig. A1).
Kalimantan’s oil palm plantations have expanded rapidly over the last decades, covering a total area of
13,000 km2 (in 1073 villages) in 2000, tripling to 40,000 km2 (in 1980 villages) in 2015 (Fig. A2).
2.3. Indicators of well-being
Village-level data, Potensi Desa (PODES), collected by the Indonesian Bureau of Statistics (BPS) roughly
every three years between 2000 and 2014 (BPS 2015), were used as proxy indicators of five dimensions of
village well-being (Table 1, with associated questionnaires provided in Table B2). The data are the most
comprehensive information on land use, population demographics, and village infrastructure available in
Indonesia, and have been used extensively to inform government policy, as well as socioeconomic studies
(Table B3). The choice of indicators and directionality of the effects on well-being listed in Table 1
correspond to existing methodologies used to assess poverty and livelihoods, such as the
Multidimensional Poverty Index (MPI, Alkire et al. 2014), the Sustainable Livelihood Approach (SLA,
Scoones 1998), and the Nested Spheres of Poverty (NESP, Gönner et al. 2007) (Fig. A4).
PODES data represent the overall socioeconomic conditions in a village, and thus do not capture
the variation and disparity in socioeconomic indicators among different sub-villages (dusun) nor
households. Rather, the data provide a useful way to compare village administrative units over large
spatial extents. PODES data are collected from the village head offices by BPS representatives, thus the
reliability of data may vary across different villages and there is potential for bias, e.g. under-or over-
reporting. However, should this random bias propagate sufficiently, we expect that the error would be
inflated and override the true oil palm effect, and thus the estimated effect of oil palm would appear
negligible, as opposed to changing the directionality of the effect.
2.3. Analysis methodology
2.3.1. Unit of analysis
We used the village administrative unit as the spatial unit of analysis, which was defined according to the
Bureau of Statistics Indonesia in 2010 census (BPS 2010). We assessed the impact of oil palm on the
change in village well-being following 2-3, 6-8, and 11-14 years of plantation age, which allows
understanding of the complex temporal dynamics of oil palm impacts across different indicators of well-
being in different contexts. To do so, we compared the change in well-being indicators between paired
PODES censuses (Table B4).
8
Oil palm usually takes 3-4 years to mature to produce commercially viable yields (Basiron 2007;
Lee et al. 2014), hence profits from plantations are usually realized after at least 3 years. Smallholders
under the Nucleus Estate Smallholder schemes (who are normally tied to transmigration areas) typically
use these initial profits to gradually pay back investment from their nucleus company, and are only able to
pay off their debt around 6 years after planting (Feintrenie et al. 2010). For independent smallholders the
realization of socioeconomic benefits would depend on land ownership and starting capital (Lee et al. 204;
Euler et al. 2016). Regardless of the types of oil palm plantation holders, we expect the economic benefits
to be realized after the first harvesting cycle between 3 and 8 years.
Changes in indicators of social well-being, such as the frequency of conflicts, may occur either
during the oil palm development stage or several years after planting. Land grabbing disputes are
expected to manifest early in the oil palm development stage, whereas conflicts due to unfulfilled
promised benefits and compensation are expected to occur several years later (Wakker et al. 2004;
Colchester et al. 2006; Rist et al. 2010). The prevalence of agricultural wage labourers is expected to
increase dramatically during the oil palm development stage, especially when a plantation reaches its full
production level (Potter 2012). Changes in indicators of environmental well-being, such as water pollution,
are likely to occur after the oil palm infrastructure, including mills, are established and palm oil production
begins in earnest, although these impacts could take some years before they are noted (Comte et al.
2012; Obidzinski et al. 2012).
As the unit receiving treatment, we used villages where oil palm plantations were detected as the
primary land use over the full analysis periods, but not within the prior five years. As the unit for control,
we used villages where oil palm plantations were not detected as the primary land use over the range of
analysis period, nor in the prior five years. For example, to assess the effect of oil palm on the change in
village well-being between the 2003 and 2011 PODES censuses (eight years of plantation age), the unit
receiving treatment were villages where oil palm plantations were detected within the village boundaries
in 2005 and 2010, but not in 2000, and the control unit were villages where oil palm plantations were not
detected within the village boundaries in 2000, 2005 and 2010. Therefore, the number of villages in the
treated and control unit vary depending on the time frame of analysis, as shown in Table B4.
2.3.2. Confounding variables
We controlled for potentially confounding variables in the assessment of oil palm impact in terms of both
selection of villages for treatment and the outcome being measured (Table 2). To achieve this we included
variables representing: (a) socio-political factors, (b) accessibility, (c) agricultural productivity, (d) baseline
village socioeconomic conditions, and (e) recent climate.
We used the regency administration boundaries (kabupaten) (variable KABU in Table 2) and land
use regulation (LZON) as proxies for socio-political factors. Decentralization of government functions to
regency levels has been identified as a key driver of deforestation, land degradation and conversion of
forest to agriculture in Indonesia (Resosudarmo 2004; Moeliono & Limberg 2012). Oil palm plantations in
Kalimantan are typically developed either outside the Forest Estate Zone, Area Penggunaan Lain (APL), or
inside the Forest Estate Zone in Production Forest, Hutan Produksi (HP).
Average elevation (ELEV) and slope (SLOP), proximity to large cities or arterial roads (CITY), human
population density (POPB), and forest cover (FORB) were used as proxies for accessibility. Oil palm is
typically developed in villages with particular topographical characteristics, e.g. flat or lightly undulating
land with good access to water. Villages located within proximity to major roads and towns tend to be
converted first to agriculture because the land is more accessible (Kinnaird et al. 2003; Linkie et al. 2004).
Villages within proximity to major roads and towns, higher human population density, and relatively low
forest cover are also recognized to be associated with communities with higher baseline cash income and
economic welfare (Belcher et al. 2015).
9
We used long-term seasonal rainfall patterns (SDRY and SWET), presence of peat soil (PEAT), and
distance to transmigration areas (TRNS) as proxies for agricultural productivity. The amount of rainfall
during the dry and wet seasons is the most important factor affecting agricultural productivity in
Indonesia (Oldeman & Frere 1982; Santika et al. 2017a). Soil type, i.e. mineral or peat soil, is also an
important factor driving forest conversion to agriculture (Carlson et al. 2013). The increase in agriculture
area in Indonesia had been partly attributed to an increase in transmigration sites and expansion of oil
palm plantations (Dennis & Colfer 2006).
Village primary livelihoods (LVHD) and well-being conditions (WLBN) at the initial stage of oil palm
developments, and the extent of village (VILA) were used as proxies for baseline village socioeconomic
conditions. These variables provided a baseline to control for conditions that may have biased impact
estimates. We used the monthly average of rainfall during the dry and wet seasons over a given time
frame of analysis (TDRY and TWET) as proxies for recent climate conditions (Santika et al. 2017b). These
variables are important for assessing the association between oil palm and natural hazards, such as water
and air pollution, and floods and landslides (Tosca et al. 2011; Merz et al. 2014).
2.3.3. Matching method
We employed a matching method (Dehejia & Wahba 2002) to select a set of control villages in which oil
palm plantations had not been developed and that exhibited the same baseline characteristics as villages
where plantations had been established. The matching method was performed based on nearest-
neighbour matching of propensity scores based on all variables described in Table 2 and exact matching of
the categorical baseline variables (i.e. KABU, LZON, SOIL, FORB, LVHD, and WLBN). We used matching
algorithms implemented in the R-package Matching (Sekhon 2015). We used a non-parametric
generalized boosted regression model (Friedman 2001) for binary outcomes implemented in the R-
package gbm (Ridgeway et al. 2015) to generate the propensity scores. The model allows flexibility in
fitting non-linear surfaces for predicting treatment assignment and can incorporate a large number of
covariates without negatively affecting model prediction. In various applications, this modelling approach
has been shown to outperform other methods that require model selection, specification of model
functional form, and are lacking flexibility, such as the commonly used generalized linear model (Lee et al.
2010). We used a 0.25 calliper width of the propensity scores' standard deviations in the nearest
neighbour approach, as this width was previously shown to be optimal (Austin 2011). The matching
method was applied separately for each of the 17 indicators of well-being and for each time period
analysis. We observed substantial improvement in the extent of overlapping areas of all continuous
variables between villages with and without plantation development in the matched dataset compared to
the original (unmatched) dataset (Fig. A5).
2.3.4. Assessing impact of oil palm plantations
The impact of oil palm on human-well-being was estimated by comparing the change in well-being
indicator in villages with oil palm plantations with the change in control villages without plantations. A
village i within an oil palm plantation is considered to be effective at alleviating a single indicator of well-
being k over time period t if the difference between the change in the value of the indicator in the treated
village (Oi,t,k) and the rate in the control village (Ci,t,k), i.e. Ai,t,k, where Ai,t,k = wk (Oi,t,k―Ci,t,k), is positive.
The value of wk represents the directional effect of the change in indicator k that defines improvement in
well-being (Table 1), i.e. wk=1 if positive change (or an increase) in indicator k represents improvement in
well-being (e.g. proportion of household with electricity) and wk=–1 if negative change (or a reduction) in
indicator k represents improvement in well-being (e.g. malnutrition prevalence, conflict frequencies). The
estimated impact of oil palm based on indicator k over time period t, i.e. Āt,k, was then obtained by fitting
an ordinary linear regression model with Ai,t,k as a response and a binary variable representing the treated
and the control villages and all variables described in Table 2 as predictors. We then normalized the
estimated treatment effects for each indicator of well-being. To obtain the overall efficacy of oil palm in
improving each aspect of well-being m for plantation age h, i.e. Ăm,h, we averaged Āt,k across all indicators
10
k belonging to the same group of well-being aspect m (Table 1) and across time period t belonging to the
same group of plantation age h (Table B4).
While the value of Ăm,h is an informative measure of the overall impact of oil palm in improving
each aspect of well-being for a given plantation age, it is also of interest to understand how efficacy varies
spatially. We conducted a systematic review on past, local-level evaluations of oil palm impacts (Table B1),
and based on this review we stipulated that baseline community livelihoods in the earliest stage of oil
palm development, particularly between subsistence-based livelihoods and market-oriented livelihoods,
play an important role in altering the direction and relative magnitude of oil palm impact on well-being.
An overall positive association between oil palm plantations and socioeconomic well-being was found in
villages where the majority of communities rely on market oriented livelihoods (92% of the studies we
reviewed indicate a positive association). In contrast, an overall negative association was found in villages
where the majority of communities rely on subsistence livelihoods (60% of the studies we reviewed
indicate a negative association). For social and environmental well-being, none of the studies we
evaluated reported positive impacts of oil palm.
To assess the robustness of the matching method against the possible presence of an unobserved
confounder, we applied a sensitivity analysis based on the principle of randomization inference
(Rosenbaum 2005) implemented in R-package rbounds (Keele 2014). This approach relies on the
sensitivity parameter Γ that measures the degree of departure from random assignment of the treatment,
in this case, oil palm plantation villages. A threshold value of Γ, namely ΓC, was calculated at the point at
which hidden bias would eliminate the effect of oil palm plantations. A study is considered sensitive to
hidden bias, i.e. the effect of oil palm plantation development likely can be explained by an unobserved
covariate, if ΓC<1, and a study is considered robust if the value of ΓC is large. The sensitivity analysis
indicated that our estimate on the impact of oil palm plantation on all the well-being indicators for each
time period based on matching was robust to the possible presence of an unobserved confounder, as
indicated by relatively large values of the sensitivity parameter threshold ΓC (ΓC≥1.72) (Table B5).
3. Results
3.1. Market economy characteristics of the baseline livelihoods
Living conditions (electricity, adequate sanitation, non-wood cooking fuel), access to infrastructure
(schools and healthcare facilities), and various credit schemes were better in villages where market-
oriented livelihoods dominated than in villages where most communities depended on subsistence
livelihoods (Fig. A6a-c). The biophysical features of these villages, i.e. proximity to major cities and roads,
may have contributed to creating such conditions (Fig. A7). In villages where subsistence-based livelihoods
dominated, basic infrastructure was generally lacking due to isolation, and households with poor living
conditions were common (Fig. A6a). Despite lacking material wealth, however, malnutrition rates among
infants were lower in subsistence-based livelihood villages (Fig. A6a), which could reflect access to a
greater variety of food sources (fruits, vegetables, and medicinal plants) due to high levels of food self-
sufficiency through farming and forest product collection (Budiharta et al. 2016; Ickowitz et al. 2016).
Moreover, the occurrence of cooperative schemes and micro enterprises (<20 employees, mainly in
forestry related products, such as rattan) were higher overall (Fig. A6c). Nearly 60% of the new oil palm
plantations developed between 2000 and 2015 in Kalimantan were situated in villages where the majority
of communities had relied on subsistence-based livelihoods (Fig. A3b).
11
3.2. Overall effect of oil palm on village well-being
We found an overall increase in basic, physical and financial indicators of well-being between 2000 and
2014, both in villages with oil palm plantation developments and those without such developments across
Kalimantan between 2000 and 2014 (Fig. 2A-C). Conversely, there was an overall decline in social and
environmental measures of well-being (Fig. 2D-E). On average the improvements in basic, physical and
financial well-being occurred more slowly in villages with oil palm plantations compared to those without,
for villages of similar biophysical features and baseline community socioeconomic conditions (Fig. 2F-H).
Moreover, the decline in social and environmental well-being had occurred significantly faster overall in
villages with oil palm plantations compared to villages without plantations (Fig. 2I-J).
Fig. 2. (A-E) Mean rate of change in human-well-being across Kalimantan villages for varying time lags, which
correspond to plantation ages since establishment. (F-J) Difference in the mean rate of change of well-being
over the same time lags in villages with and without oil palm. Large negative values indicate higher reduction of
well-being relative to the counterfactual. Aspects of well-being assessed include: (A,F) basic well-being (living
condition), (B,G) physical well-being (infrastructure), (C,H) financial well-being (income support), (D,I) social
well-being (security and social equity), and (E,J) environmental well-being (natural hazard prevention). Error
bars represent 95% confidence intervals.
3.3. Benefits vary with baseline conditions
Despite the overall negative association between of oil palm and all aspects of human-well-being
(Fig. 2F-J), these associations varied across different baseline community livelihoods at the initial stage of
plantation development (Fig. 3). While the association between oil palm development and basic, physical,
and financial well-being appeared to be negative overall in villages where communities had relied
12
formerly on subsistence-based livelihoods, the association was positive overall in villages where
communities had relied on market-oriented livelihoods (Fig. 3A-C). However, the negative association
between the industry and social and environmental well-being appeared to be more severe in villages that
formerly relied on market-oriented livelihoods compared to villages with livelihoods that were formerly
subsistence-based (Fig. 3D-E).
In villages that formerly relied on market-oriented livelihoods water and air pollution were
reported to be more severe compared to villages that were mostly subsistence-based (Fig. 4E). This could
reflect substantially higher agriculture intensification and the occurrence of more oil palm mills in the
former (Fig. A7f). An increase in frequency of floods and landslides was reported to be greater in oil palm
villages where communities formerly depended on subsistence-based livelihoods (Fig. 4E). The rate of
forest conversion(>60% forest cover) within five years prior to oil palm plantation developments was
comparatively similar to villages that formerly relied on market-oriented livelihoods (Fig. A8). However,
most of the near-intact forest being cleared for plantations in former subsistence-based livelihoods
villages was located at relatively high altitude and had received considerable amount of rain during the
wet season over the last decades (Fig. A7a-b).
Fig. 3. Difference in the mean rate of change of aspects of well-being over varying oil palm plantation ages in
years, in villages with and without oil palm, where most communities formerly depended on: subsistence-
based livelihoods or market-oriented livelihoods. Large negative values indicate a greater decline in well-being
relative to the counterfactual. Aspects of well-being include: (A) basic well-being (living condition), (B) physical
well-being (infrastructure), (C) financial well-being (income support), (D) social well-being (security and social
equity), and (E) environmental well-being (natural hazard prevention). Error bars represent 95% confidence
intervals.
13
Fig. 4. Difference in the mean rate of change of aspects of well-being over varying oil palm plantation ages (in
years) between villages with and without oil palm, where most communities formerly depended on
subsistence-based livelihoods or market-oriented livelihoods. Large negative value indicates higher reduction
of aspect of well-being relative to the counterfactual. Aspects of well-being include: (A) basic well-being (living
condition), (B) physical well-being (infrastructure), (C) financial well-being (income support), (D) social well-
being (security and social equity), and (E) environmental well-being (natural hazard prevention). Detail
explanation for each variable is provided in Table 1.
14
3.4. Benefits vary through time
3.4.1. Socioeconomic effect
In villages that had an existing market economy we observed that basic well-being indicators, such
as adequate sanitation and energy for cooking, improved 6-8 years after a plantation had been
established (Fig. 4A). Physical well-being indicators such as access to healthcare facilities and secondary
schools primarily improved in the early stages (i.e. 2-3 years) of oil palm plantation development (Fig. 4B).
Financial well-being improved mainly through increased access to credit schemes after 6-8 years (Fig. 4C).
On the other hand, for oil palm established in villages where most communities had relied on
subsistence-based livelihoods, we observed the opposite trends on socioeconomic aspects of well-being.
Basic well-being indicators such as access to electricity, adequate sanitation and energy for cooking were
markedly reduced since the early stages (i.e. 2-3 years) until 11-14 years of oil palm plantation
development compared to the control villages (Fig. 4A). Physical well-being indicators, in particular access
to secondary schools, were also reduced (Fig. 4B). Reduced basic and physical well-being relative to the
controls likely occurred due to a combination of the remoteness of villages (Fig. A7c) and the influx of
workers from outside the region. An increase in wage labourers in these villages was substantially higher
compared to the controls in the first 2-3 years of plantation development (Fig. 4D), and the ethnic
diversity of communities also increased after plantations were established (Fig. A9b). Labour influx in
some villages may be related to the existing low population density of these villages (Fig. A7d).
In villages where most communities had relied on subsistence-based livelihoods, financial well-
being indicators (e.g. access to cooperative schemes, number of small enterprises) reduced relative to the
controls after 6-8 years of plantation development (Fig. 4C). This could reflect the impact of increased
specialisation towards cash cropping (Cramb et al. 2009; Kremen et al. 2012) and increased scarcity of
forest products, such as rattan (Meijaard et al. 2014).
3.4.2. Environmental impacts
Water and air pollution in oil palm landscapes is sometime associated with the production of palm
oil mill effluent (POME) in the oil extraction process after harvesting (Singh et al. 2010). This could explain
why water and air pollution was often reported in the 6-8 years after plantation establishment (Fig. 4E),
around the period when fruit yields have become substantial (Carter et al. 2007). Excessive applications of
pesticides and fertiliser could also contribute to water pollution (Meijerink et al. 2008; Comte et al. 2012).
It has been estimated that for every tonne of crude palm oil produced, about 2.5 tonnes of liquid waste is
generated (Ahmad et al. 2003), although companies vary on how this waste is treated.
3.5. Interacting and possible spillover effects
The influx of workers in remote villages that lack infrastructure facilities (Fig. A6a-b) could be
associated with the reported reduction in electricity provision, clean sanitation, energy for cooking, and
secondary schools, which were all substantially greater in these villages compared to the counterfactual
(Fig. 4A-B). This does not necessarily mean that such facilities are reduced in villages with oil palm
development, but it is rather caused by reduced proportion on the provision of such facilities relative to
village population as the population increased. An increase in waged labourers from outside the region
not only occurs in remote villages where most communities had relied on subsistence-based livelihoods,
but also in semirural villages where communities were formerly dependent on market-oriented
livelihoods. In the latter, the increase in waged labourers was substantially higher relative to the controls
up to 6-8 years of plantation being established (Fig. 4D). This could reflect higher pressure to intensify
agricultural production and labour requirements, both for plantations and oil palm mills that largely occur
in semirural villages (Fig. A7f).
15
Conflicts have occurred more frequently in villages where communities formerly relied on market-
oriented livelihoods, particularly after 6-8 years of plantation age (Fig. 4D). In these villages it is possible
that these conflicts are triggered by communities’ discontent over insufficient company benefit sharing to
local communities (Sheil et al. 2009; Abram et al. 2017) after initiation of harvesting at around four years
(Carter et al. 2007), as well as a signal of rising social inequalities and disparities (Euler et al. 2017). On the
other hand, in villages formerly relying on subsistence-based livelihoods, conflicts mostly occurred at the
start-up phase of plantations (i.e. 2-3 years). In former subsistence-based villages, indigenous
communities dominate (Fig. A9a). In these villages, the conflicts could be a result of the loss of indigenous
land and perceived lack of land compensation at the beginning of plantation development (2-3 years), due
to a weak land tenure system (Colchester et al. 2006).
4. Discussion
There is a widespread belief that oil palm development results in increased socioeconomic well-
being. Yet, in Kalimantan, we found that the socioeconomic effects of the industry were not always equal
across villages, and depended on the biophysical characteristics and baseline socioeconomic conditions of
communities prior to oil palm development. In villages where communities had mostly relied on market-
oriented livelihoods, oil palm development was associated with increased basic, physical and financial
well-being. However, in remote forested villages where communities had often relied on subsistence-
based livelihoods, a substantial reduction in basic, physical and financial well-being was observed, along
with reduced social and environmental well-being. This could reflect harmful implications of rushed
development in remote areas where the government’s regulatory powers and monitoring capacity are
limited (Zoomers et al. 2017). Two-thirds of the new oil palm plantations developed between 2000 and
2015 in Kalimantan were situated in villages where the majority of communities had relied on subsistence-
based livelihoods. Further expansion of oil palm in remote forested landscapes, such as in Kalimantan and
Papua, should therefore be considered more carefully. While poverty can be caused by isolation, abrupt
contact with a market economy can lead to exacerbation of poverty and displacement of communities
who are not well integrated in the market system. Alternative policies that could facilitate local
communities in remote areas to thrive and prosper given the right skills, strengths, and social context
should be sought (Ruiz-Pérez et al. 2004; Shackleton et al. 2011).
For oil palm industries that have been developed, regardless of baseline village livelihood sector,
our findings show that unsustainable livelihoods, increased socioeconomic disparity, and environmental
issues remain major challenges. Local communities are burdened with the socioeconomic and
environmental costs of poorly planned and implemented development, while a small number of rural and
urban elites may take the largest share of economic benefits (Sheil et al. 2009). Taxation of oil palm with
direct production bonuses accruing at regency and district levels could ensure that the economic benefits
of industrial oil palm directly compensate local development costs (McCarthy et al. 2012; Falconer et al.
2015). However, to achieve this would require several vital adjustments in terms of (a) fiscal policy on
collection and redistribution of revenues, (b) monitoring and regulation of tax compliance, and (c) data
sharing and transparencies among different key government ministries involved in the oil palm sector (i.e.
Ministry of Environment and Forestry; Ministry of Agriculture; Ministry of Villages, Disadvantaged Regions
and Transmigration; Ministry of Finance; and Ministry of Agrarian Affairs and Spatial Planning). These
issues are inextricably and causally intertwined, and they ought to be addressed in concert.
The current fiscal policy in Indonesia highlights relatively low levels of tax collection from the oil
palm industry and redistribution of revenues to local governments, with only an estimated 14 % of palm
oil tax revenues being redistributed to local governments (Falconer et al. 2015). Corruption and lack of tax
compliance are recognized as key reasons for low levels of tax collection, and are considered to be the
biggest issues tainting the reputations of many oil palm companies in Indonesia (Falconer et al. 2015; CNN
Indonesia 2017; Mongabay 2018a). In the provinces of West Kalimantan and Riau alone, only 10% and
16
30% oil palm plantation companies have complied with their tax obligations, respectively (Tempo 2016;
Times Indonesia 2018). Indonesia’s Corruption Eradication Commission, known as KPK, has recently been
called upon to intervene in the country’s palm oil sector with the hope of rectifying this problem (Tempo
2017c; Mongabay 2018a). However, the work of the KPK also faces tremendous challenges given
uncertainty in land ownership and permits, and lack of data sharing and transparencies among different
government ministries, as reflected by the slow progress in the implementation of One Map Policy or
Kebijakan Satu Peta (KSP) that attempts to bring together data from different government departments in
a single spatial platform (Kompas 2018). The KSP data portal was eventually launched by the Indonesian
President in December 2018, and this is a substantial achievement. However, the accessibility of the data
is currently restricted only to the President and Vice President, government ministries, and local
government heads, i.e. Governors and Bupati (Head of Regency) (Presidential Instruction No. 20/2018).
Excluding non-governmental organizations and the general public from accessing these data
fundamentally limits the utility and the main goal of the KSP in providing transparency and inclusiveness
(Mongabay 2018b). It is also recognizable that analysing multifaceted geospatial data would require
adequate skills in data processing, modelling, and analysis, and importantly collaborative efforts across
different disciplines and levels of society to provide accurate interpretation in different contexts, in which
the public could greatly assist (Mongabay 2018b). Mechanisms for including public participation in
analysing the data are required for the effectiveness of the KSP, not only to resolve conflicts related to
overlapping land permits, but also to provide foundations for good governance and planning for healthy
and sustainable developments that ultimately benefit rural people in the long run.
While our study provides an understanding of the impact of oil palm development in general
across different types of plantations, it is worth mentioning that there are currently a few mechanisms
that attempt to improve the performance of oil palm plantations to deliver socioeconomic benefits to
villages. Certification schemes, such as the Roundtable on Sustainable Palm Oil (RSPO), could help
considerably, although by 2015, only 12% of the oil palm plantations in Kalimantan that had been certified
(Fig. 1A). Although many of the earlier plantations in Kalimantan and Sumatra contained little forest when
they were RSPO certified, Carlson et al. (2018) demonstrated that most of these plantations reduced
deforestation rates. On the other hand, the impact of this scheme in reducing fire occurrences (Cattau et
al. 2016; Noojipady et al. 2017; Carlson et al. 2018) and delivering socioeconomic benefits to village
communities are questionable (Morgans et al. 2018). Further scrutiny of the RSPO-certified plantation
estate using our analytical framework could help address this knowledge gap.
While the premise regarding variation in the impact of large-scale agriculture, land reform, and
development in general in different locations has been reported in numerous studies (e.g. Mertz et al.
2009; Shete & Rutten 2015), evidence for the heterogeneity of the impact on well-being was lacking for
the oil palm. Lack of understanding and evidence about the conspicuous difference between the impact of
oil palm on market-oriented livelihoods and the impact on subsistence-based livelihoods is (and is used as)
a basis of constant debate in the Indonesian politics on development objectives, particularly between the
Ministry of Environment and Forestry (who effectively has jurisdiction in the majority of forest area and is
responsible for the welfare of nearby communities who largely depend on subsistence-based agriculture,
including farming, fishing and gathering of non-timber forest products), the Ministry of Agriculture (who is
responsible for commercialized agriculture sector and welfare of the peasants), and the Ministry of
Villages, Disadvantaged Regions and Transmigration (who is responsible for advancing developments in
the outer islands of Java and transmigration programs) (Mongabay 2016; Tempo 2017a, b). For the local
government at Kabupaten or Regency level, direction towards supporting versus resisting oil palm is likely
to be informed by the vision of most communities: either to pursue socioeconomic growth or to maintain
socioecological capital. Regencies where the majority of communities depend highly on forest
environmental income (which tend to be of indigenous background with stronger ties to ancestral land)
may place higher importance on maintaining socioecological values (Potter 2009; Sirait 2009).
Acknowledging that the impacts of oil palm developments can be fundamentally different across different
baseline village livelihood systems should inform fruitful multi-level and cross-sectoral government
discussions on the long-term risks and benefits of oil palm plantation on the well-being of local
communities.
17
References
Abram, N.K., Meijaard, E., Wilson, K.A., Davis, J.T., Wells, J.A. et al. (2017) Oil palm–community conflict
mapping in Indonesia: A case for better community liaison in planning for development initiatives.
Applied Geography 78, 33-44.
Ahmad, A.L., Ismail, S. & Bhatia, S. (2003) Water recycling from palm oil mill effluent (POME) using
membrane technology. Desalination 157, 87-95.
Alkire, S., Chatterjee, M., Conconi, A., Seth, S. & Vaz, A. (2014) Global Multidimensional Poverty Index
2014. OPHI Briefing 21, Oxford: University of Oxford.
Allen, E.R. (2016) Analysis of Trends and Challenges in the Indonesian Labor Market. ADB Papers on
Indonesia No. 16. Asian Development Bank, Metro Manila, Philippines.
Alwarritzi, W., Nanseki, T. & Chomei, Y. (2015) Impact of oil palm expansion on farmers’ crop income and
poverty reduction in Indonesia: An application of propensity score matching. The Journal of
Agricultural Science 8, 119.
Austin, P.C. (2011) Optimal caliper widths for propensity-score matching when estimating differences in
means and differences in proportions in observational studies. Pharmaceutical Statistics 10, 150-
161.
Basiron, Y. (2007) Palm oil production through sustainable plantations. European Journal of Lipid Science
and Technology 109, 289–95
Belcher, B., Achdiawan, R. & Dewi, S. (2015) Forest-based livelihoods strategies conditioned by market
remoteness and forest proximity in Jharkhand, India. World Development 66, 269-279.
Blaber-Wegg, T., Hodbod, J. & Tomei, J. (2015) Incorporating equity into sustainability assessments of
biofuels. Current Opinion in Environmental Sustainability 14, 180-186.
BPS (Bureau of Statistic Indonesia) (2010) Sensus Penduduk 2010. Jakarta, Indonesia.
BPS (Bureau of Statistic Indonesia) (2015) Village Potential Statistics (PODES) 2000, 2003, 2005, 2008, and
2014. Jakarta, Indonesia.
Budiharta, S., Meijaard, E., Wells, J.A., Abram, N.K. & Wilson, K.A. (2016) Enhancing feasibility:
Incorporating a socio-ecological systems framework into restoration planning. Environmental
Science & Policy 64, 83-92.
Byerlee, D., Falcon, W.P. & Naylor, R.L. (2017) The Tropical Oil Crop Revolution: Food, Feed, Fuel, and
Forests. Oxford University Press, Oxford, UK.
Carlson, K.M., Curran, L.M., Asner, G.P., Pittman, A.M., Trigg, S.N. et al. (2013) Carbon emissions from
forest conversion by Kalimantan oil palm plantations. Nature Climate Change 3, 283-287.
Carlson, K.M., Heilmayr, R., Gibbs, H.K., Noojipady, P., Burns, D.N. et al. (2018) Effect of oil palm
sustainability certification on deforestation and fire in Indonesia. Proceedings of the National
Academy of Sciences 115, 121-126.
Carter, C., Finley, W., Fry, J., Jackson, D. & Willis, L. (2007) Palm oil markets and future supply. European
Journal of Lipid Science and Technology 109, 4307–314.
Castiblanco, C., Etter, A. & Ramirez, A. (2015) Impacts of oil palm expansion in Colombia: What do
socioeconomic indicators show? Land Use Policy 44, 31-43.
Cattau, M.E., Marlier, M.E. & DeFries, R. (2016) Effectiveness of Roundtable on Sustainable Palm Oil
(RSPO) for reducing fires on oil palm concessions in Indonesia from 2012 to 2015. Environmental
Research Letters 11, 105007.
18
CNN Indonesia (2017) KPK Temukan 63 Ribu Wajib Pajak Industri Sawit Kemplang Pajak. 4 May. Available
at: https://www.cnnindonesia.com/nasional/20170503174824-12-212023/kpk-temukan-63-ribu-
wajib-pajak-industri-sawit-kemplang-pajak.
Colchester, M., Jiwan, N., Andiko, Sirait, M., Firdaus, A.Y. et al. (2006) Promised Land. Palm Oil and Land
Acquisition in Indonesia: Implications for Local Communities and Indigenous Peoples. Forest Peoples
Programme / Perkumpulan Sawit Watch, Jakarta, Indonesia.
Comte, I., Colin, F., Whalen, J.K., Grünberger, O. & Caliman, J.P. (2012) Agricultural practices in oil palm
plantations and their impact on hydrological changes, nutrient fluxes and water quality in Indonesia:
A review. Advances in Agronomy 116, 71.
Cooke, M.F. (2012) In the name of poverty alleviation: Experiments with oil palm smallholders and
customary land in Sabah, Malaysia. Asia Pacific Viewpoint 53, 240-253.
Costanza, R., Fisher, B., Ali, S., Beer, C., Bond, L. et al. (2007) Quality of life: An approach integrating
opportunities, human needs, and subjective well-being. Ecological Economics 61, 267-276.
Cramb, R.A., Colfer, C.J.P., Dressler, W., Laungaramsri, P., Le, Q.T., Mulyoutami, E. et al. (2009) Swidden
transformations and rural livelihoods in Southeast Asia. Human Ecology 37, 323-346.
Daniel, T.C., Muhar, A., Arnberger, A., Aznar, O., Boyd, J.W. et al. (2012) Contributions of cultural services
to the ecosystem services agenda. Proceedings of the National Academy of Sciences 109, 8812-8819.
Dehejia, R.H. & Wahba, S. (2002) Propensity score-matching methods for nonexperimental causal studies.
Review of Economics and Statistics 84, 151-161.
Dennis, R.A. & Colfer, C.P. (2006) Impacts of land use and fire on the loss and degradation of lowland
forest in 1983–2000 in East Kutai District, East Kalimantan, Indonesia. Singapore Journal of Tropical
Geography 27, 30-48.
Directorate General of Estate Crops Indonesia (2017) Tree Crop Estate Statistics of Indonesia 2015-2017:
Palm Oil. Ministry of Agriculture, Jakarta.
Euler, M., Krishna, V., Schwarze, S., Siregar, H. & Qaim, M. (2017) Oil palm adoption, household well-
being, and nutrition among smallholder farmers in Indonesia. World Development 93, 219-235.
Euler, M., Schwarze, S., Siregar, H. & Qaim, M. (2016) Oil palm expansion among smallholder farmers in
Sumatra, Indonesia. Journal of Agricultural Economics 67, 658-676.
Falconer, A., Mafira, T. & Sutiyono, G. (2015) Improving Land Productivity through Fiscal Policy: Early
Insights on Taxation in the Palm Oil Supply Chain. Climate Policy Initiative. Available at:
https://climatepolicyinitiative.org/wp-content/uploads/2015/12/Full-Report-English-Early-Insights-
on-Taxation-in-the-Palm-Oil-Supply-Chain.pdf.
Ferraro, P.J. (2009) Counterfactual thinking and impact evaluation in environmental policy. New Directions
for Evaluation 2009, 75-84.
Feintrenie, L., Chong, W.K. & Levang, P. (2010) Why do farmers prefer oil palm? Lessons learnt from
Bungo district, Indonesia. Small-Scale Forestry 9, 379-396.
Fick, S.E. & Hijmans, R.J. (2017) WorldClim 2: new 1-km spatial resolution climate surfaces for global land
areas. International Journal of Climatology 37, 4302-4315.
Food and Agriculture Organization of the United Nations Statistics Division (FAOSTAT) (2016) Crops;
Indonesia; Oil, Palm and Oil, Palm Fruit; Average Production of Commodity; 1990-2014. Available at:
http://faostat3.fao.org/browse/Q/QC/E.
Friedman, J.H. (2001) Greedy function approximation: a gradient boosting machine. The Annals of
Statistics 29, 1189-1232.
Gatto, M., Wollni, M., Asnawi, R. & Qaim, M. (2017) Oil palm boom, contract farming, and rural economic
development: Village-level evidence from Indonesia. World Development 95, 127-140.
19
Gaveau, D.L.A., Salim, M. & Arjasakusuma, S. (2016a) Deforestation and Industrial Plantations
Development in Borneo. Center for International Forestry Research (CIFOR) Dataverse, V1.
Gaveau, D.L., Sheil, D., Husnayaen, M.A.S., Arjasakusuma, S., Ancrenaz, M. et al. (2016b). Rapid
conversions and avoided deforestation: examining four decades of industrial plantation expansion in
Borneo. Scientific Reports 6, 32017.
Geospatial Information Agency (BIG) (2016) Peta Propinsi Kalimantan Barat, Kalimantan Selatan,
Kalimantan Tengah, Kalimantan Timur, dan Kalimantan Utara. Jakarta, Indonesia.
Hansen, M.C., Potapov, P.V., Moore, R., Hancher, M., Turubanova, S. et al. (2013) High-resolution global
maps of 21st-century forest cover change. Science 342, 850-853.
Hodbod, J. & Tomei, J. (2013) Demystifying the social impacts of biofuels at local levels: Where is the
evidence? Geography Compass 7, 478-488.
Huffman, G.J., Bolvin, D.T., Nelkin, E.J., Wolff, D.B., Adler, R.F. et al. (2007) The TRMM multisatellite
precipitation analysis (TMPA): Quasi-global, multiyear, combined-sensor precipitation estimates at
fine scales. Journal of Hydrometeorology 8, 38-55.
Ickowitz, A., Rowland, D., Powell, B., Salim, M.A., & Sunderland, T. (2016) Forests, trees, and
micronutrient-rich food consumption in Indonesia. PLoS ONE 11, e0154139.
Jarvis, A., Reuter, H.I., Nelson, A. & Guevara, E. (2008) Hole-filled SRTM for the Globe Version 4. URL:
http://srtm.csi.cgiar.org.
Jelsma, I., Schoneveld, G.C., Zoomers, A. & Van Westen, A.C.M. (2017) Unpacking Indonesia’s independent
oil palm smallholders: An actor-disaggregated approach to identifying environmental and social
performance challenges. Land Use Policy 69, 281-297.
Junaidi (2012) Perkembangan Desa-Desa Eks Transmigrasi dan Interaksi dengan Wilayah Sekitarnya Serta
Kebijakan ke Depan (Kajian di Provinsi Jambi). PhD Thesis. Bogor Agricultural University (IPB), Bogor.
Available at: https://repository.ipb.ac.id/jspui/bitstream/123456789/58350/1/2012jun.pdf.
Keele, L.J. (2014) Perform Rosenbaum Bounds Sensitivity Tests for Matched and Unmatched Data. R
Package Version 2.1.
Kinnaird, M.F., Sanderson, E.W., O'brien, T.G., Wibisono, H.T. & Woolmer, G. (2003) Deforestation trends
in a tropical landscape and implications for endangered large mammals. Conservation Biology 17,
245-257.
Klasen, S., Meyer, K.M., Dislich, C., Euler, M., Faust, H. et al. (2016) Economic and ecological trade-offs of
agricultural specialization at different spatial scales. Ecological Economics 122, 111-120.
Kompas (2018) Presiden Jokowi: Banyak yang Ketakutan dengan Kebijakan Satu Peta. 11 December.
Available at: https://nasional.kompas.com/read/2018/12/11/13471861/presiden-jokowi-banyak-
yang-ketakutan-dengan-kebijakan-satu-peta.
Kremen, C., Iles, A. & Bacon, C. (2012) Diversified farming systems: an agroecological, systems-based
alternative to modern industrial agriculture. Ecology and Society 17, 44.
Krishna, V., Euler, M., Siregar, H. & Qaim, M. (2017) Differential livelihood impacts of oil palm expansion in
Indonesia. Agricultural Economics 48, 639-653.
Lee, B.K., Lessler, J. & Stuart, E.A. (2010) Improving propensity score weighting using machine learning.
Statistics in Medicine 29, 337-346.
Lee, J.S.H., Ghazoul, J., Obidzinski, K. & Koh, L.P. (2014) Oil palm smallholder yields and incomes
constrained by harvesting practices and type of smallholder management in Indonesia. Agronomy
for Sustainable Development 34, 501-513.
Li, T.M. (2016) Situating transmigration in Indonesia’s oil palm labour regime. In: The Oil Palm Complex:
Smallholders, Agribusiness and the State in Indonesia and Malaysia (eds. R. Cramb & J. McCarthy),
pp. 354-377. Singapore: NUS Press.
20
Linkie, M., Smith, R.J. & Leader-Williams, N. (2004) Mapping and predicting deforestation patterns in the
lowlands of Sumatra. Biodiversity and Conservation 13, 1809–1818.
Margono, B.A., Potapov, P.V., Turubanova, S., Stolle, F. & Hansen, M.C. (2014) Primary forest cover loss in
Indonesia over 2000-2012. Nature Climate Change 4, 730-735.
McCarthy, J.F., Gillespie, P. & Zen, Z. (2012) Swimming upstream: local Indonesian production networks in
“globalized” palm oil production. World Development 40, 555-569.
Meijaard, E., Achdiawan, R., Wan, M. & Taber, A. (2014) Rattan. The Decline of a Once-Important Non-
Timber Forest Product in Indonesia. CIFOR Occasional Paper No. 101. Center for International
Forestry Research (CIFOR), Bogor, Indonesia.
Meijaard, E., Garcia-Ulloa, J., Sheil, D., Carlson, K., Wich, S.A. et al. (2018) Oil Palm and Biodiversity – A
Situation Analysis. IUCN Oil Palm Task Force, Gland, Switzerland.
Meijerink, G.W., Langeveld, J.W.A. & Hellegers, P.J.G.J. (2008) Biofuels and Water: An Exploration.
Wageningen UR.
Mertz, O., Padoch, C., Fox, J., Cramb, R.A., Leisz, S.J. et al. (2009) Swidden change in Southeast Asia:
understanding causes and consequences. Human Ecology 37, 259-264.
Merz, B., Aerts, J.C.J.H., Arnbjerg-Nielsen, K., Baldi, M., Becker, A. et al. (2014) Floods and climate:
emerging perspectives for flood risk assessment and management. Natural Hazards and Earth
System Sciences 14, 1921-1942.
Ministry of Environment and Forestry Indonesia (MEF) (2010) Peta Kawasan Hutan. Jakarta, Indonesia.
Ministry of Environment and Forestry Indonesia (MEF) (2016) Peta Penutupan Lahan (PPL) 1990-2015.
Jakarta, Indonesia.
Ministry of Environment and Forestry Indonesia (MEF) (2017) Peta Kesatuan Hidrologi Gambut. Jakarta.
Moeliono, M. & Limberg, G. (2012) The Decentralization of Forest Governance: Politics, Economics and the
Fight for Control of Forests in Indonesian Borneo. Routledge.
Mongabay (2016) Menanti Moratorium Sawit, Berikut Alasan Mengapa Kebijakan Ini Urgen. Mongabay
Indonesia. 22 August. Available at: http://www.mongabay.co.id/2016/08/22/menanti-moratorium-
sawit-berikut-alasan-mengapa-kebijakan-ini-urgen.
Mongabay (2018a) Indonesia’s Anti-Graft Agency ‘Eager to Intervene’ in Palm Oil Sector. 25 October.
Available at: https://news.mongabay.com/2018/10/indonesias-anti-graft-agency-eager-to-
intervene-in-palm-oil-sector.
Mongabay (2018b) Ketika Kebijakan Satu Peta Rilis, Bagaimana Akses Publik? 22 December. Available at:
https://www.mongabay.co.id/2018/12/22/ketika-kebijakan-satu-peta-rilis-bagaimana-akses-
publik/.
Morgans, C.L., Meijaard, E., Santika, T., Law, E., Budiharta, S. et al. (2018) Evaluating the effectiveness of
palm oil certification in delivering multiple sustainability objectives. Environmental Research Letters
13, 064032.
Noojipady, P., Morton, D.C., Schroeder, W., Carlson, K.M., Huang, C. et al. (2017) Managing fire risk during
drought: The influence of certification and El Niño on fire-driven forest conversion for oil palm in
Southeast Asia. Earth System Dynamics 8, 749.
Obidzinski, K., Andriani, R., Komarudin, H. & Andrianto, A. (2012) Environmental and social impacts of oil
palm plantations and their implications for biofuel production in Indonesia. Ecology and Society 17,
25.
Obidzinski, K., Dermawan, A. & Hadianto, A. (2014) Oil palm plantation investments in Indonesia’s forest
frontiers: limited economic multipliers and uncertain benefits for local communities. Environment,
Development and Sustainability 16, 1177-1196.
21
Oldeman, L.R. & Frere, M. (1982) Technical Report on a Study of the Agroclimatology of the Humid Tropics
of Southeast Asia. Food & Agriculture Organizations.
Palupi, S., Sukapti, Y.S., Maemunah, S., Prasetyohadi, P., Tømte, A. et al. (2017) Privatisasi Transmigrasi
dan Kemitraan Plasma Menopang Industri Sawit: Resiko Hak Asasi Manusia dalam Kebijakan
Transmigrasi dan Kemitraan Plasma di Sektor Industri Perkebunan Sawit. The Institute for Ecosoc
Rights & Norwegian Center for Human Rights. Jakarta. Available at:
https://www.jus.uio.no/smr/english/about/id/news/docs/buku_privatisasi_transmigrasi_2017.pdf.
Phalan, B. (2009) The social and environmental impacts of biofuels in Asia: an overview. Applied Energy
86, S21-S29.
Potter, L. (2009) Oil palm and resistance in West Kalimantan, Indonesia. In: Agrarian Angst and Rural
Resistance in Contemporary Southeast Asia (eds. Caouette, D. & Turner S.), pp. 125-154. Routledge.
Potter, L. (2012) New transmigration ‘paradigm’ in Indonesia: Examples from Kalimantan. Asia Pacific
Viewpoint 53, 272-287.
Resosudarmo, I.A.P. (2004) Closer to people and trees: will decentralisation work for the people and the
forests of Indonesia? The European Journal of Development Research 16, 110-132.
Ridgeway, G., Southworth, M.H. & R Development Unit (2015) Generalized Boosted Regression Models. R
package version 2.1.1.
Rist, L., Feintrenie, L. & Levang, P. (2010) The livelihood impacts of oil palm: smallholders in Indonesia.
Biodiversity Conservation 19, 1009-1024.
Ruiz-Pérez, M., Belcher, B., Achdiawan, R., Alexiades, M., Aubertin, C. et al. (2004) Markets drive the
specialization strategies of forest peoples. Ecology and Society 9, 4.
Rosenbaum, P.R. (2005) Heterogeneity and causality: Unit heterogeneity and design sensitivity in
observational studies. The American Statistician 59, 147-152.
Shackleton, S., Delang, C.O. & Angelsen, A. (2011) From subsistence to safety nets and cash income:
exploring the diverse values of non-timber forest products for livelihoods and poverty alleviation. In:
Non-Timber Forest Products in the Global Context (eds. Shackleton, S., Shackleton, C. & Shanley, P.),
pp. 55-81. Berlin: Springer.
Santika, T., Ancrenaz, M., Wilson, K.A., Spehar, S., Abram, N. et al. (2017a) First integrative trend analysis
for a great ape species in Borneo. Scientific Reports 7, 4839.
Santika, T., Meijaard, E., Budiharta, S., Law, E.A., Kusworo, A. et al. (2017b) Community forest
management in Indonesia: Avoided deforestation in the context of anthropogenic and climate
complexities. Global Environmental Change 46, 60-71.
Santika, T., Meijaard, E. & Wilson, K.A. (2015) Designing multifunctional landscapes for forest
conservation. Environmental Research Letters 10, 114012.
Scoones, I. (1998) Sustainable Rural Livelihoods: A Framework for Analysis. IDS Working Paper 72. Sussex:
Brighton Institute of Development Studies, University of Sussex.
Sekhon, J.S. (2015) Multivariate and Propensity Score Matching with Balance Optimization. R package
version 4.9-2.
Sen, A. (1985) Commodities and Capabilities. Amsterdam, North-Holland.
Sheil, D., Casson, A., Meijaard, E., Van Noordwjik, M., Gaskell, J. et al. (2009) The Impacts and
Opportunities of Oil Palm in Southeast Asia: What Do We Know and What Do We Need to Know?
CIFOR Occasional Paper No. 51. Center for International Forestry Research (CIFOR), Bogor,
Indonesia.
Shete, M. & Rutten, M. (2015) Impacts of large-scale farming on local communities’ food security and
income levels–Empirical evidence from Oromia Region, Ethiopia. Land Use Policy 47, 282-292.
22
Singh, R.P., Ibrahim, M.H., Esa, N. & Iliyana, M.S. (2010) Composting of waste from palm oil mill: a
sustainable waste management practice. Reviews in Environmental Science and Bio/Technology 9,
331-344.
Sirait, M. (2009) Indigenous Peoples and Oil Palm Plantation Expansion in West Kalimantan, Indonesia. The
Hague: Cordaid Memisa.
Tempo (2016) Baru Sepertiga Perusahaan Sawit Bayar Pajak. 24 August. Available at:
https://bisnis.tempo.co/read/798659/baru-sepertiga-perusahaan-sawit-bayar-pajak/full&view=ok.
Tempo (2017a) Industry Minister Airlangga Hartarto: Our Objective is to Save the Industry. May 23.
Available at: https://magz.tempo.co/konten/2017/05/23/LU/33190/Industry-Minister-Airlangga-
Hartarto-Our-Objective-is-to-Save-the-Industry/40/17.
Tempo (2017b) Minister of the Environment and Forestry Siti Nurbaya: Protecting Peatlands is an Absolute
Must. May 23. Available at: https://magz.tempo.co/konten/2017/05/23/LU/33189/Minister-of-the-
Environment-and-Forestry-Siti-Nurbaya-Protecting-Peatlands-is-an-Absolute-Must/40/17.
Tempo (2017c) KPK: Pengelolaan Komoditas Kelapa Sawit Rawan Korupsi. 24 April. Available at:
https://nasional.tempo.co/read/869099/kpk-pengelolaan-komoditas-kelapa-sawit-rawan-
korupsi/full&view=ok.
Times Indonesia (2018) Ekspedisi Gubernur: Bang Midji Siapkan Moratorium Sawit di Kalbar, Ini
Alasannnya. 15 October. Available at:
https://www.timesindonesia.co.id/read/186159/20181015/153742/ekspedisi-gubernur-bang-midji-
siapkan-moratorium-sawit-di-kalbar-ini-alasannnya
Tosca, M.G., Randerson, J.T., Zender, C.S., Nelson, D.L., Diner, D.J. et al. (2011) Dynamics of fire plumes
and smoke clouds associated with peat and deforestation fires in Indonesia. Journal of Geophysical
Research: Atmospheres 116, D08207.
United Nations (1996) Report of The World Summit for Social Development. United Nations: New York,
NY, USA. Available at: https://www.un.org/development/desa/dspd/world-summit-for-social-
development-1995.html.
United Nations General Assembly (2014). Open Working Group Proposal for Sustainable Development
Goals (A/68/970). United Nations: New York, NY, USA.
Villela, A.A., D'Alembert, B.J., Rosa, L.P. & Freitas, M.V. (2014) Status and prospects of oil palm in the
Brazilian Amazon. Biomass and Bioenergy 67, 270-278.
Wakker, E., Watch, S. & de Rozario, J. (2004) Greasy Palms: The Social and Ecological Impacts of Large-
Scale Oil Palm Plantation Development in Southeast Asia. AIDEnvironment, Amsterdam. Available
at: https://www.cabdirect.org/cabdirect/abstract/20056701996.
Wells, J.A., Wilson, K.A., Abram, N.K., Nunn, M., Gaveau, D.L. et al. (2016) Rising floodwaters: mapping
impacts and perceptions of flooding in Indonesian Borneo. Environmental Research Letters 11,
064016.
Zen, Z., Barlow, C., Gondowarsito, R. & McCarthy, J.F. (2016) Interventions to promote smallholder oil
palm and socio-economic improvement in Indonesia. In: The Oil Palm Complex: Smallholders,
Agribusiness and the State in Indonesia and Malaysia (eds. R. Cramb & J. McCarthy), pp. 78-108.
Singapore: NUS Press.
Zoomers, A., van Noorloos, F., Otsuki, K., Steel, G. & van Westen, G. (2017) The rush for land in an
urbanizing world: From land grabbing toward developing safe, resilient, and sustainable cities and
landscapes. World Development 92, 242-252.
23
Appendix A. Supplementary Figures
Fig. A1. The change in the extent of oil palm plantations in (a) Kalimantan, and (b) Sumatra, every 5 years
between 2000 and 2015, by type of plantation holder: Roundtable on Sustainable Palm Oil (RSPO) certified
plantations, RSPO non-certified plantations, Non-RSPO non-certified plantations within oil palm concessions,
and Non-RSPO non-certified outside oil palm concessions (independent smallholders).
Fig. A2. Change in the distribution of oil palm plantations in Kalimantan every five years between 2000 and
2015.
24
Fig. A3. (a) Land cover types, including primary or secondary forest, barren or agricultural land, and (b) primary
livelihood of communities, including subsistence-based livelihoods, plantation (polyculture), and other sectors
(horticulture, aquaculture, agricultural services, and non-agricultural sectors), where new oil palm plantations
have been developed in Kalimantan in 1996-2000, 2000-2005, 2005-2010, and 2010-2015.
Fig. A4. (a) Aspects (dimensions) and indicators of well-being used in the Multidimensional Poverty Index (MPI,
Alkire et al. 2014), the Sustainable Livelihood Approach (SLA, Scoones 1998), and the Nested Spheres of Poverty
(NESP, Gonner et al. 2007). (b) Association between aspects and indicators used in our study with those
encapsulated in MPI, SLA and NESP, with the associated aspect represented by the capital letter in the second
column. For MPI, the aspects include: Education (Ed), Health (H), and Living standards (L). For SLA, the aspects
include: Human (Hu), Social (S), Natural (N), Physical (P) and Financial (F). For NESP, the Core includes: Health
(H), Wealth (W), and Knowledge (K); and the Context includes: Natural (N), Economic (E), Social (S), Political (Po),
and Infrastructure (I).
25
Fig. A5. The distributions of continuous variables characterizing oil palm plantation and control villages in Kalimantan, before and after matching, aggregated across
time periods used in the analyses. Variables include: mean elevation (ELEV), mean slope (SLOP), mean distance to large cities or arterial roads (CITY), mean human
population density (POPB), mean distance to transmigration areas (TRNS), mean long-term mean monthly rainfall during the dry and wet season (SDRY and SWET), and
mean monthly mean rainfall during the dry and wet season between 2000 and 2014 (TDRY and TWET). The degree of overlap between the distributions of variables for
oil palm villages and controls increased after matching, indicating increased similarity in the characteristics of oil palm villages and matched controls. Vertical lines
indicate the mean value for oil-palm villages and controls, with the gap between the two lines decreased after matching.
26
Fig. A6. Baseline community welfare prior to oil-palm plantation developments: (a) Basic wellbeing (living
condition), (b) Physical wellbeing (infrastructure), (c) Financial wellbeing (income support), (d) Social well-being
(security and social equity), and (e) Environmental wellbeing (prevention of natural hazards), in villages with
different main livelihood sectors: subsistence-based livelihoods (SL; swidden farming of dryland rice (inter-
cropped with banana, maize, and cassava, and typically supplemented by market exchange for forest and/or
agroforest products) and inland fishing) and market-oriented livelihoods (ML; including polyculture plantations
(e.g. coffee, rubber, oil palm, coconut), horticulture, aquaculture, agricultural services, and non-agricultural
sectors). Detail explanation of socioeconomic indicator is provided in Table 1.
27
Fig. A7. Biophysical characteristics of villages with different main livelihood sectors: subsistence-based
livelihoods (SL; swidden farming of dryland rice (inter-cropped with banana, maize, and cassava, and typically
supplemented by market exchange for forest and/or agroforest products) and inland fishing) and market-
oriented livelihoods (ML; including polyculture plantations (e.g. coffee, rubber, oil palm, coconut), horticulture,
aquaculture, agricultural services, and non-agricultural sectors). Biophysical characteristics include: (a)
elevation: ELEV, (b) slope: SLOP, (c) distance to major cities and roads: CITY, (d) human population density:
POPB, (e) distance to transmigration areas: TRNS, (f) distance to palm-oil mills: OPML, (g-h) long term monthly
rainfall during the dry and wet seasons: SDRY and SWET, and (i-j) monthly rainfall between 2000 and 2014
during the dry and wet seasons: TDRY and TWET.
28
Fig. A8. The rates of annual forest loss over a five year period prior to oil palm developed between 2000 and
2014, by initial forest cover in villages where most communities had relied on subsistence-based livelihoods or
market-oriented livelihoods.
Fig. A9. (a) Ethnic and religious diversity of oil-palm villages where communities formerly relied on subsistence-
based livelihoods (SL) and market-oriented livelihoods (ML). We used indicators of percentage of villages with
more than one ethnic group (where higher value indicates higher ethnic diversity in communities) and
deviation (i.e. standard deviation) in the number of different religious buildings (where lower value indicates
more religiously diverse communities). (b) Difference in mean rate of change in ethnic diversity for villages with
and without oil-palm development, where most communities formerly depend on subsistence-based
livelihoods and market-oriented livelihoods.
29
Appendix B. Supplementary Tables
Table B1. Past evaluations on the social impact of industrial-scale oil-palm plantations on rural well-being in developing countries (in chronological order of references for
each aspect of well-being), showing a generally positive association between oil palm and socioeconomic aspect of well-being and negative association between oil palm
and social and environmental aspects of well-being.
Aspect of
well-being
Indicator of
well-being Study area Time frame
Case-
control
study
҂
Subsistence
-based
livelihoods
§
Oil palm
impact ‡ Reference
4 communities in West Kalimantan province 2008 N Y — Sirait (2009)
Bungo district, Jambi province, Sumatra 2007-2009
N N + Feintrenie et al. (2010)
Bungo district, Jambi province, Sumatra 2007-2009
N N + Rist et al. (2010)
78 villages in 8 provinces in Sumatra, Kalimantan & Sulawesi 2010 N N + Budidarsono et al. (2012)
Dabat village, Sanggau district, West Kalimantan province 2010 N N + Julia & White (2012)
Manokwari district, West Papua province 2011 N Y + Obidzinski et al. (2012)
Kubu Raya district, West Kalimantan province 2011 N Y — Obidzinski et al. (2012)
Boven Digoel district, Papua province 2011 N Y — Obidzinski et al. (2012)
3 oil palm estates in Riau province, Sumatra 2012 N N — Sinaga (2013)
2 communities in Polochic Valley, Guatemala 2009-2011
N Y + Mingorría et al. (2014)
1 oil palm estate in Bayelsa, Nigeria 2013 N N + Ohimain et al. (2014)
271 households in Riau province, Sumatra 1990-2015
Y N + Alwarritzi et al. (2015)
100 municipalities in Colombia 1993-2009
Y YN + Castiblanco et al. (2015)
245 households in Jambi province, Sumatra 2010 N N + Cahyadi & Waibel (2016)
Kutai Kertanegara district, East Kalimantan province 2014 N N + Dharmawan et al. (2016)
682 households in Jambi province, Sumatra 2012 Y N + Euler et al. (2017)
98 villages in Jambi province, Sumatra 1992-2012
Y N + Gatto et al. (2017)
Socioeconomic
Income,
consumption,
employment,
and
developments
682 households in Jambi province, Sumatra 2012 Y N + Krishna et al. (2017)
Summary N = 12
Y = 5
YN = 1
N+ = 11 (92%)
Y+ = 2 (40%)
YN+ = 1 (100%)
‡ Oil palm impact on indicator wellbeing: Positive (+), Negative (—)
҂ Whether or not case-control study was conducted: Yes (Y), No (N)
§ Whether or not the majority of communities within study area have relied on subsistence-based livelihoods prior to oil palm development: Yes (Y), No (N), YN (Mixed)
30
Table B1. (Continued)
Aspect of
wellbeing
Indicator of
wellbeing Study area Time frame
Case-
control
study
҂
Subsistence-
based
livelihoods §
Oil palm
impact ‡ Reference
4 communities in West Kalimantan province 2008 N Y — Sirait (2009)
Bungo district, Jambi province, Sumatra 2007-2009
N N — Rist et al. (2010)
Community along Lemanak River and
resettlement site along Tinjar River, Sarawak, Malaysia
1990-2010
N Y — Cramb & Sujang (2011)
41 households in Bumbun village, Carey Island, Malaysia 2007 N Y — Lai (2011)
Pru district, Brong Ahafo, Ghana 2009 N Y — Schoneveld et al. (2011)
Dabat village, Sanggau district, West Kalimantan province 2010 N N — Julia & White (2012)
Manokwari district, West Papua province 2011 N Y — Obidzinski et al. (2012)
Kubu Raya district, West Kalimantan province 2011 N Y — Obidzinski et al. (2012)
Boven Digoel district, Papua province 2011 N Y — Obidzinski et al. (2012)
Jambi province, Sumatra 2012 N N — Manik et al. (2013)
3 oil palm estates in Riau province, Sumatra 2012 N N — Sinaga (2013)
2 communities in Polochic Valley, Guatemala 2009-2011
N Y — Mingorría et al. (2014)
100 municipalities in Colombia 1993-2009
Y YN — Castiblanco et al. (2015)
4 communities in Berau district, East Kalimantan province 2005-2010
N N — Elmhirst et al. (2015)
1 community in East Kutai district, East Kalimantan province 2005-2010
N Y — Elmhirst et al. (2015)
Meliau, Sanggau district, West Kalimantan province 2010-2012
N N — Li (2015)
Kampung Lebor, Serian, Sarawak, Malaysia 2012 N Y — Yong & Pang (2015)
Coastal villages in West Kalimantan province 2013-2015
N N — de Vos (2016)
Social Security,
social equity,
gender
equity
682 households in Jambi province, Sumatra 2012 Y N — Euler et al. (2017)
Summary N = 8
Y = 10
YN = 1
N+ = 0 (0%)
Y+ = 0 (0%)
YN+ = 0 (0%)
‡ Oil palm impact on indicator wellbeing: Positive (+), Negative (—)
҂ Whether or not case-control study was conducted: Yes (Y), No (N)
§ Whether or not the majority of communities within study area have relied on subsistence-based livelihoods prior to oil palm development: Yes (Y), No (N), YN (Mixed)
31
Table B1. (Continued)
Aspect of
wellbeing
Indicator of
wellbeing Study area Time frame
Case-
control
study
҂
Subsistence-
based
livelihoods §
Oil palm
impact ‡ Reference
Beluran district, Sabah, Malaysia 2007-2008
N Y — Dayang Norwana et al.
(2011)
Manokwari district, West Papua province 2011 N Y — Obidzinski et al. (2012)
Kubu Raya district, West Kalimantan province 2011 N Y — Obidzinski et al. (2012)
Boven Digoel district, Papua province 2011 N Y — Obidzinski et al. (2012)
Wilberforce Island, Nigeria 2013-2014
N N — Ohimain & Izah (2014)
Siak watershed, Riau province, Sumatra 2009-2010
N N — Comte et al. (2015)
Bungku village, Jambi province, Sumatra 2013-2014
N N — Merten et al. (2016)
Environmental
Water, air or
soil quality
Kutai Kertanegara district, East Kalimantan province 2014 N N — Dharmawan et al. (2016)
Summary N = 4
Y = 4
YN = 0
N+ = 0 (0%)
Y+ = 0 (0%)
YN+ = NA
‡ Oil palm impact on indicator wellbeing: Positive (+), Negative (—)
҂ Whether or not case-control study was conducted: Yes (Y), No (N)
§ Whether or not the majority of communities within study area have relied on subsistence-based livelihoods prior to oil palm development: Yes (Y), No (N), YN (Mixed)
32
References Table B1
Alwarritzi, W., Nanseki, T. & Chomei, Y. (2015) Impact of oil palm expansion on farmers’ crop income and
poverty reduction in Indonesia: An application of propensity score matching. The Journal of Agricultural
Science 8, 119.
Budidarsono, S., Dewi, S., Sofiyuddin, M. & Rahmanulloh, A. (2012) Socioeconomic impact assessment of palm
oil production. Technical Brief No. 27. World Agroforestry Centre (ICRAF).
Cahyadi, E.R. & Waibel, H. (2016) Contract farming and vulnerability to poverty among oil palm smallholders in
Indonesia. Journal of Development Studies 52, 681-695.
Castiblanco, C., Etter, A. & Ramirez, A. (2015) Impacts of oil palm expansion in Colombia: What do
socioeconomic indicators show? Land Use Policy 44, 31-43.
Comte, I., Colin, F., Grünberger, O., Whalen, J.K., Widodo, R.H. et al. (2015) Watershed-scale assessment of oil
palm cultivation impact on water quality and nutrient fluxes: a case study in Sumatra (Indonesia).
Environmental Science and Pollution Research 22, 7676-7695.
Cramb, R. & Sujang, P.S. (2011) ‘Shifting ground’: Renegotiating land rights and rural livelihoods in Sarawak,
Malaysia. Asia Pacific Viewpoint 52, 136-147.
Dayang Norwana, A.A.B., Kanjappan, R., Chin, M., Schoneveld, G.C., Potter, L. et al. (2011) The local impacts of
oil palm expansion in Malaysia; An assessment based on a case study in Sabah State. Center for
International Forestry Research (CIFOR) Working Paper No. 78, 1-17.
de Vos, R.E. (2016) Multi-functional lands facing oil palm monocultures: A case study of a land conflict in West
Kalimantan, Indonesia. Austrian Journal of South-East Asian Studies 9, 11-32.
Dharmawan, A.H., Mardiyaningsih, D.I. & Yulian, B.E. (2016) Expansion of oil palm plantation and changes in
social, economic and rural ecology: A case study in Kutai Kartanegara. Project Working Paper Series OPAL
No. 01/2016.
Elmhirst, R., Siscawati, M. & Sijapati Basnett, B. (2015) Navigating investment and dispossession: gendered
impacts of the oil palm ‘land rush’ in East Kalimantan, Indonesia. In: Land Grabbing, Conflict and Agrarian-
Environmental Transformations: Perspectives from East and Southeast Asia. Chiang Mai, Thailand.
Euler, M., Krishna, V., Schwarze, S., Siregar, H. & Qaim, M. (2017) Oil palm adoption, household welfare, and
nutrition among smallholder farmers in Indonesia. World Development 93, 219-235.
Feintrenie, L., Chong, W.K. & Levang, P. (2010) Why do farmers prefer oil palm? Lessons learnt from Bungo
district, Indonesia. Small Scale Forestry 9, 379-396.
Gatto, M., Wollni, M., Asnawi, R. & Qaim, M. (2017) Oil palm boom, contract farming, and rural economic
development: Village-level evidence from Indonesia. World Development 95, 127-140.
Julia & White, B. (2012) Gendered experiences of dispossession: oil palm expansion in a Dayak Hibun
community in West Kalimantan. Journal of Peasant Studies 39, 995-1016.
Krishna, V., Euler, M., Siregar, H. & Qaim, M. (2017) Differential livelihood impacts of oil palm expansion in
Indonesia. Agricultural Economics 48, 639-653.
Lai, W.T. (2011) Gender and livelihoods: A case study of the Mah Meri and the oil palm plantations of Carey
Island. Asian Journal of Women's Studies 17, 66-95.
Li, T.M. (2015) Social Impacts of Oil Palm in Indonesia: A Gendered Perspective from West Kalimantan. Center
for International Forestry Research (CIFOR) Vol. 124.
Manik, Y., Leahy, J. & Halog, A. (2013) Social life cycle assessment of palm oil biodiesel: a case study in Jambi
Province of Indonesia. The International Journal of Life Cycle Assessment 18, 1386-1392.
Merten, J., Röll, A., Guillaume, T., Meijide, A., Tarigan, S. et al. (2016) Water scarcity and oil palm expansion:
social views and environmental processes. Ecology and Society 21, 5.
Mingorría, S., Gamboa, G., Martín-López, B. & Corbera, E. (2014) The oil palm boom: socio-economic
implications for Q’eqchi’ households in the Polochic valley, Guatemala. Environment, Development and
Sustainability 16, 841-871.
33
Obidzinski, K., Andriani, R., Komarudin, H. & Andrianto, A. (2012) Environmental and social impacts of oil palm
plantations and their implications for biofuel production in Indonesia. Ecology and Society 17, 25.
Ohimain, E.I., Emeti, C.I. & Izah, S.C. (2014) Employment and socioeconomic effects of semi-mechanized palm
oil mill in Bayelsa state, Nigeria. Asian Journal of Agricultural Extension, Economics and Sociology 3, 206-
216.
Ohimain, E.I. & Izah, S.C. (2014) Possible contributions of palm oil mill effluents to greenhouse gas emissions in
Nigeria. British Journal of Applied Science & Technology 4, 4705-4720.
Rist, L., Feintrenie, L. & Levang, P. (2010) The livelihood impacts of oil palm: smallholders in
Indonesia. Biodiversity Conservation 19, 1009-1024.
Schoneveld, G., German, L. & Nutakor, E. (2011) Land-based investments for rural development? A grounded
analysis of the local impacts of biofuel feedstock plantations in Ghana. Ecology and Society 16, 10.
Sinaga, H. (2013) Employment and income of workers on Indonesian oil palm plantations: food crisis at the
micro level. Journal of the Science of Food and Agriculture 1, 64-78.
Sirait, M. (2009) Indigenous peoples and oil palm plantation expansion in West Kalimantan, Indonesia. The
Hague: Cordaid Memisa.
Susanti, A. & Burgers, P.P.M. (2011) Oil palm expansion in Riau province, Indonesia: Serving people, planet,
profit? The European Report on Development.
Yong, C. & Pang, W.A. (2015) Gender impact of large-scale deforestation and oil palm plantations among
indigenous groups in Sarawak, Malaysia. In: Gender and Land Tenure in the Context of Disaster in Asia (pp.
15-31). Springer International Publishing.
34
Table B2. Survey questionnaire in PODES that used to generate the data represented in Table 1.
Aspect of
well-being Variable (abbreviation) Associated survey question in PODES
(original translation in Bahasa Indonesia in italic font)
Proportions of households with poor housing
conditions (POOR)
Number of households occuring in slums or poor housing neighbourhoods.
Jumlah rumah tangga yang tinggal di pemukiman kumuh.
Proportions of households with electricity
(ELCT)
Number of households with electricity.
Jumlah rumah tangga pengguna listrik.
Cooking fuel for majority of households (COOK)
Cooking fuel for the majority of households. (1=electricity or liquefied petroleum gas (LPG), 2=kerosene,
3=wood or others).
Bahan bakar yang digunakan sebagian besar rumah tangga untuk memasak. (1=listrik atau LPG,
2=minyak tanah, 3=kayu bakar atau lainnya).
Toilet facilities for majority of households
(TOLT)
Toilet facility for the majority of households. (1=own toilet, 2=joint toilet, 3=public toilet, 4=no toilet).
Tempat buang air besar sebagian besar rumah tangga. (1=jamban sendiri, 2=jamban bersama, 3=jamban
umum, 4=bukan jamban).
Basic
(Living conditions)
Child malnutrition incidence in the last year
(MLNT)
Number of child malnutrition cases in the last year.
Jumlah balita penderita gizi buruk setahun terakhir.
Distance to nearest healthcare facility (HEAL) Distance to the nearest health facilities (hospital, doctor’s practice, clinic, public health center, village
health center).
Jarak ke sarana kesehatan terdekat (rumah sakit, praktek dokter, poliklinik, puskesmas, puskesdes).
Distance to nearest primary school (PSCH) Distance to the nearest primary schools.
Jarak ke Sekolah Dasar (SD) terdekat.
Physical
(Infrastructure)
Distance to nearest secondary school (SSCH) Distance to the nearest secondary (junior high) schools.
Jarak ke Sekolah Menengah Pertama (SMP) terdekat.
Number of active village cooperative schemes
or other schemes (COOP)
Number of cooperative schemes that are actively operated (including KUD, Kopinkra, Kospin, etc).
Jumlah koperasi yang masih aktif beroperasi (termasuk KUD, Kopinkra, Kospin, dll).
Number of credit for farmers or communities
(CRDT)
Number of credit facilities received by communities in the last year (including KUR, KKP, KUK, KPR, etc).
Banyaknya fasilitas kredit yang diterima penduduk selama setahun terakhir (termasuk KUR, KKP, KUK,
KPR, dll).
Financial
(Income support)
Number of small industries (<20 employees)
(SIND)
Number of small or micro-scale industry (less than 20 employees).
Jumlah industri kecil dan mikro (kurang dari 20 pekerja).
35
Table B2. (Continued)
Aspect of
well-being Variable (abbreviation) Associated survey question in PODES
(original translation in Bahasa Indonesia in italic font)
Frequency of conflicts among communities in
the last year (CNFL)
Frequency of conflicts among communities in the last year.
Jumlah kejadian perkelahian massal selama setahun terakhir.
Proportion of families with agricultural wage
labourers (AGLB)
Number of families with at least one family member working as an agricultural wage labourer.
Jumlah keluarga yang ada anggota keluarganya menjadi buruh tani.
Social
(Security and social
equity)
Suicidal rates in the last year (SUIC) Number of suicidal victims in the last year.
Banyaknya korban bunuh diri yang terjadi selama setahun terakhir.
Water pollution over the last 3 years (WPOL) The occurrence of water pollution in the last year (1=none, 2=yes but the issue has not been formally
reported by communities, 3=yes and the issue has been formally reported by communities).
Pencemaran air selama setahun terakhir. (1=tidak ada, 2=ada tetapi tidak ada pengaduan dari
masyarakat, 3=ada dan ada pengaduan dari masyarakat).
Air pollution over the last 3 years (APOL) The occurrence of air pollution in the last year (1=none, 2=yes but the issue has not been formally
reported by communities, 3=yes and the issue has been formally reported by communities).
Pencemaran udara selama setahun terakhir. (1=tidak ada, 2=ada tetapi tidak ada pengaduan dari
masyarakat, 3=ada dan ada pengaduan dari masyarakat).
Environmental
(Natural hazard
prevention)
Frequency of floods and landslides over the last
3 years (FLOD)
Frequency of floods, flash floods, or landslides in the last three years.
Banyaknya kejadian banjir, banjir banding, dan tanah longsor selama tiga tahun terahkir.
36
Table B3. The use of PODES data in studies on socioeconomic and developments across Indonesia (in alphabetical order of references).
Study Use of PODES data References
Scope Area Year Aspect
Barron et al. (2016) Distributions of conflicts Indonesia 2003 Population demographics, economic,
institutional
Barron et al. (2009) Characteristics of social conflicts Indonesia 2003 Population demographics, economic,
institutional
Budidarsono et al. (2012) Socioeconomic impact of oil palm Indonesia 2008 Social, economic, infrastructure
Cameron & Shah (2013) Social impact of cash transfer programs Indonesia 2006 Population demographics, social,
economic
Cameron & Shah (2015) Risk-taking behaviour given exposure to
natural disaster
Indonesia 2008 Social, economic, environment
Cooley et al. (2016) Impact of maternal undernutrition
on child health
Indonesia 2011 Social, infrastructure
De Juan et al. (2015) Religious conflicts Indonesia 2003 Social
Grimm et al. (2015) Impact of electricity on fertility Indonesia 1996, 2000, 2003,
2006, 2008
Infrastructure, livelihood
Hasan et al. (2013) Early education and development Indonesia 2003, 2006, 2008,
2011
Population demographics, social,
infrastructure
Hashiguchi & Higasikata (2017) Human capital externalities Indonesia 1996, 2006 Social, economic, infrastructure
Jagger & Rana (2017) Impact of REDD+ on social welfare Kalimantan 2008, 2011 Infrastructure, land use
Korkeala & Obidzinski (2012) Oil-palm expansion and household welfare Kalimantan 2003 Land use
Lewis (2016) Impact of local political fragmentation on fiscal
and service
Indonesia 2003, 2006, 2008,
2011, 2014
Social, economic, infrastructure
Maharani & Tampubolon (2014) Impact of fiscal decentralisation on child
immunisation
Indonesia 2011 Population demographics, economic,
infrastructure
Makovec et al. (2016) Impact of restrictive emigration policies on
social welfare
Indonesia 2006 Population demographics, social,
economic
Miguel et al. (2005) Impact of social capital on industrialization Indonesia 1986 Population demographics, social,
economic, institutional
Miteva et al. (2015) Impact of forest management certification on
human wellbeing
Kalimantan 2000, 2006, 2008 Population demographics, infrastructure,
land use
Liu & Yamauchi (2014) Impact of population growth on income Indonesia 2000, 2006 Infrastructure, environment
Olsson & Valsecchi (2015) Impact of fiscal decentralisation on economic
and social welfare
Indonesia 2003, 2006, 2008 Infrastructure
37
Table B3. (Continued)
Study Use of PODES data References
Scope Area Year Aspect
Parmanto et al. (2008) Multidimensional visualization of village health
statistics
Indonesia 2003 Social, environment
Rokx (2010) Provision of health services Indonesia 1996, 2006 Population demographics, social,
infrastructure
Sparrow (2008) Effectiveness of health card programs Indonesia 1996 Social, economic, infrastructure
Sparrow & Pradhan (2014) Efficiency of district’s public health and education
programs
Indonesia 2003, 2006, 2008 Population demographics, infrastructure
Sujarwoto & Tampubolon (2016) Access to internet facilities Indonesia 2011 Population demographics, economic,
infrastructure
Sugiarto et al. (2017) Impact of climate change on water resources and
agriculture
East Java 2014 Economic, infrastructure, land use,
environment
Tan Soo (2017) Valuation of air quality Indonesia 2006 Economic
Vothknech & Sumarto (2011) Impact of violent conflicts on economic growth Indonesia 2003, 2006, 2008 Social, economic
Widayati et al. (2014) Livelihoods and access to clean water South Sulawesi
2011 Population demographics, social,
economic
Yamauchi et al. (2009) Dynamics of village economies Indonesia 1996, 2006 Infrastructure
38
References Table B3
Barron, P., Jaffrey, S. & Varshney, A. (2016) When large conflicts subside: The ebbs and flows of violence in
post-Suharto Indonesia. Journal of East Asian Studies 16, 191-217.
Barron, P., Kaiser, K. & Pradhan, M. (2009) Understanding variations in local conflict: Evidence and implications
from Indonesia. World Development 37, 698-713.
Budidarsono, S., Dewi, S., Sofiyuddin, M. & Rahmanulloh, A. (2012) Socioeconomic Impact Assessment of Palm
Oil Production. Technical Brief No. 27: palm oil series. Bogor, Indonesia. World Agroforestry Centre
(ICRAF), SEA Regional Office.
Cameron, L. & Shah, M. (2013) Can mistargeting destroy social capital and stimulate crime? Evidence from a
cash transfer program in Indonesia. Economic Development and Cultural Change 62, 381-415.
Cameron, L. & Shah, M. (2015) Risk-taking behavior in the wake of natural disasters. The Journal of Human
Resources 50, 484-515.
Cooley, P.C., Poulos, C., Rineer, J.I., Rogers, S.M., Scruggs, M.D. et al. (2016) Forecasting the Impact of Maternal
Undernutrition on Child Health Outcomes in Indonesia. RTI Press Publication No. RR-0028-1612.
Research Triangle Park, NC: RTI Press.
De Juan, A., Pierskalla, J.H. & Vüllers, J. (2015) The pacifying effects of local religious institutions: An analysis of
communal violence in Indonesia. Political Research Quarterly 68, 211-224.
Grimm, M., Sparrow, R. & Tasciotti, L. (2015) Does electrification spur the fertility transition? Evidence from
Indonesia. Demography 52, 1773-1796.
Hasan, A., Hyson, M. & Chang, M.C. (2013) Early Childhood Education and Development in Poor Villages of
Indonesia: Strong Foundations, Later Success. World Bank Publications (2013).
Hashiguchi, Y. & Higashikata, T. (2017) Human Capital Externalities in Indonesian Cities. IDE Discussion Paper
No. 672. Institute of Developing Economies, Japan External Trade Organization (JETRO).
Jagger, P. & Rana, P. (2017) Using publicly available social and spatial data to evaluate progress on REDD+ social
safeguards in Indonesia. Environmental Science and Policy 76, 59-69.
Korkeala, O. & Obidzinski, K. (2012) A Household Welfare Perspective on The Expansion of Palm Oil Production
in Indonesia. Economics Department Working Paper Series No. 4212. University of Sussex.
Lewis, B.D. (2016) Local Political Fragmentation: Fiscal and Service Delivery Effects in Indonesia. Working Papers
in Trade and Development No. 2016/16. ANU College of Asia and the Pacific, The Australian National
University.
Liu, Y. & Yamauchi, F. (2014) Population density, migration, and the returns to human capital and land: Insights
from Indonesia. Food Policy 48, 182-193.
Maharani, A. & Tampubolon, G. (2014) Has decentralisation affected child immunisation status in Indonesia?
Global Health Action 7, 24913.
Makovec, M., Purnamasari, R., Sandi, M. & Savitri, A. (2016) Intended vs. Unintended Consequences of
Migration Restriction Policies: Evidence from a Natural Experiment in Indonesia (No. 2016-13). ISER
Working Paper Series.
Miguel, E., Gertler, P. & Levine, D.I. (2005) Does social capital promote industrialization? Evidence from a rapid
industrializer. The Review of Economics and Statistics 87, 754-762.
Miteva, D.A., Loucks, C.J. & Pattanayak, S.K. (2015) Social and environmental impacts of forest management
certification in Indonesia. PLoS One 10, e0129675.
Olsson, O. & Valsecchi, M. (2015) Resource Windfalls and Local Government Behavior: Evidence from a Policy
Reform in Indonesia. Working Papers in Economic No. 635. University of Gothenburg.
Parmanto, B., Paramita, M.V., Sugiantara, W., Pramana, G., Scotch, M. et al. (2008) Spatial and
multidimensional visualization of Indonesia's village health statistics. International Journal of Health
Geographics 7, 30.
39
Rokx, C. (2010) New Insights into the Provision of Health Services in Indonesia: A Health Workforce Study. World
Bank Publications.
Sparrow, R. (2008) Targeting the poor in times of crisis: the Indonesian health card. Health Policy and Planning
23, 188-199.
Sparrow, R. & Pradhan, M. (2014) Activity Measures for Health and Education Outcomes in Indonesia. TNP2K
Working Paper 15-2014. Jakarta, Indonesia: Tim Nasional Percepatan Penanggulangan Kemiskinan
(TNP2K).
Sugiarto, Y., Atmaja, T. & Wibowo, A. (2017) Developing vulnerability analysis method for climate change
adaptation on agropolitan region in Malang district. IOP Conference Series: Earth and Environmental
Science 58, 012044. IOP Publishing.
Sujarwoto, S. & Tampubolon, G. (2016) Spatial inequality and the Internet divide in Indonesia 2010–2012.
Telecommunication Policy 40, 602-616.
Tan Soo, J.S. (2017) Valuing air quality in Indonesia using households’ locational choices. Environmental and
Resource Economics,1-22.
Vothknecht, M. & Sumarto, S. (2011) Beyond the Overall Economic Downturn: Evidence on Sector-Specific
Effects of Violent Conflict from Indonesia. DIW Discussion Papers No. 1105.
Widayati, A., Khasanah, N., Prasetyo, P.N. & Dewi, S. (2014) Upland Landscape Management for Drinking Water
Provision-Biang Loe Catchment, Bantaeng, South Sulawesi. AgFor Livelihood Conservation Strategy 01.
Bogor, Indonesia. World Agroforestry Centre (ICRAF) Southeast Asia Regional Program.
Yamauchi, F., Muto, M., Chowdhury, S., Dewina, R. & Sumaryanto, S. (2009) Spatial Networks, Labor Supply,
and Income Dynamics. IFPRI Discussion Papers No. 897.
40
Table B4. PODES censuses used to assess the change in village well-being (indicated in check marks) over
varying plantation ages: 2-3, 6-8, and 11-14 years, and the number villages included as treatment units and
control units for each time period of analysis.
PODES census year
Number of unit Plantation
age
Analysis
time frame 2000 2003 2006 2008 2011 2014
Treated Control
2000-2003 √ √
100
5673
2003-2006 √ √
51
5579
2006-2008 √ √
116
5241
2008-2011 √ √
343
5170
2-3 years
2011-2014 √ √
70
4829
2000-2006 √ √
100
5673
2000-2008 √ √
99
5223
2003-2011 √ √
50
5168
2006-2014 √ √
116
4827
6-8 years
2008-2014 √ √
70
4829
2000-2011 √ √
99
5153
2003-2014 √ √
50
4826
11-14 years
2000-2014 √ √
99
4812
Table B5. The robustness of the matching analysis against hidden bias due to an unobserved confounder, as
indicated by the minimum value of the sensitivity parameter ΓC across wellbeing indicators, for the overall
analysis and each of the livelihood subgrouping analysis (subsistence-based livelihoods and market-oriented
livelihoods), for each time period. ΓC=1 means that unobserved confounder needs to be at least as strong as the
included effects to have an influence. Thus, larger min(ΓC) indicates a more robust analysis against hidden bias.
Analysis type according to
plantation age
Analysis
time frame min(ΓC)
2000-2003 2.24
2003-2006 1.88
2006-2008 1.74
2008-2011 2.16
2-3 years
2011-2014 1.98
2000-2006 1.82
2000-2008 2.01
2003-2011 1.92
2006-2014 1.72
6-8 years
2008-2014 2.06
2000-2011 2.11
2003-2014 1.94
11-14 years
2000-2014 2.04
