Content uploaded by Ashwini Chhatre
Author content
All content in this area was uploaded by Ashwini Chhatre
Content may be subject to copyright.
DOI: 10.1126/science.1199343
, 1606 (2011);331 Science , et al.Lauren Persha
Livelihoods, and Biodiversity Conservation
Social and Ecological Synergy: Local Rulemaking, Forest
This copy is for your personal, non-commercial use only.
clicking here.colleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others
here.following the guidelines can be obtained byPermission to republish or repurpose articles or portions of articles
): May 15, 2011 www.sciencemag.org (this infomation is current as of
The following resources related to this article are available online at
http://www.sciencemag.org/content/331/6024/1606.full.html
version of this article at: including high-resolution figures, can be found in the onlineUpdated information and services,
http://www.sciencemag.org/content/suppl/2011/03/22/331.6024.1606.DC1.html
can be found at: Supporting Online Material
http://www.sciencemag.org/content/331/6024/1606.full.html#ref-list-1
, 4 of which can be accessed free:cites 18 articlesThis article
http://www.sciencemag.org/cgi/collection/ecology
Ecology subject collections:This article appears in the following
registered trademark of AAAS. is aScience2011 by the American Association for the Advancement of Science; all rights reserved. The title CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience
on May 15, 2011www.sciencemag.orgDownloaded from
Social and Ecological Synergy:
Local Rulemaking, Forest Livelihoods,
and Biodiversity Conservation
Lauren Persha,
1
*Arun Agrawal,
1
Ashwini Chhatre
2
Causal pathways to achieve social and ecological benefits from forests are unclear, because
there are few systematic multicountry empirical analyses that identify important factors and their
complex relationships with social and ecological outcomes. This study examines biodiversity
conservation and forest-based livelihood outcomes using a data set on 84 sites from six countries
in East Africa and South Asia. We find both positive and negative relationships, leading to
joint wins, losses, and trade-offs depending on specific contextual factors; participation in forest
governance institutions by local forest users is strongly associated with jointly positive outcomes
for forests in our study.
Human-dominated forested landscapes in
tropical developing countries provide
many different and important ecosystem
services and also sustain the livelihoods of large
numbers of poor peoples (1). Recognition of the
diverse socioeconomic and ecological contribu-
tions of forests has prompted many governments
to pursue policies for improved livelihoods and
conservation outcomes. Notable are forest policy
decentralization reforms that transfer ownership
and management responsibilities to local forest
user organizations (2,3). Such policies have been
introduced in more than two-thirds of the devel-
oping world (4), including 35 of 51 countries in
sub-Saharan Africa (5), and apply to an estimated
one-third of forests in developing countries glob-
ally (6).
Formalized local participation in forest gov-
ernance via decentralization is often viewed as a
key mechanism to provide incentives to local com-
munities to use forests sustainably through en-
hanced local knowledge, stronger accountability,
and perceived legitimacy of forest rules (2,3). Its
effectiveness is also debated because of fears that
decentralization could lead to resource capture by
local elites and remain ineffectual without exten-
sive devolution of rights to local participants and
functional linkages between local decentralized
institutions and well-crafted macro-level govern-
ance institutions (7).
More broadly, policy pathways toward joint
improvements in sustainable livelihoods and bio-
diversity conservation continue to be unclear. De-
spite the inherent complexity of social-ecological
contexts for forest systems, current policy responses,
particularly in terms of explicit management for
trade-offs or synergies across multiple social and
ecological goals, are seldom based on careful anal-
ysis or evidence of factors that lead to improve-
ments across desired sets of social and ecological
outcomes together (8,9).
In scholarly research, relationships between
human livelihoods and biodiversity conservation
are often conceptualized as diametrically opposite
(10,11), but more studies that undertake explicit
quantitative analysis of this relationship are
needed (12). There is evidence of trade-offs, but
case studies also reveal the potential for synergies
(13). However, previous work tends to focus anal-
ysis on these social and ecological outcomes in
isolation from each other, rather than in tandem
as a single outcome constructed across both so-
cial and ecological dimensions. Here we analyze
patterns of outcome relationships between forest-
based household livelihoods and biodiversity
conservation in 84 study sites, and the potential
explanations associated with the joint production
of these two forest benefits.
Our analysis draws on a global data set of
social, ecological, and governance data on a wide
range of representative forests in human-dominated
tropical landscapes that has been compiled by the
International Forestry Resources and Institutions
(IFRI) research program (14,15). The 84 cases
are drawn from six countries in East Africa (6 in
Kenya, 7 in Tanzania, and 17 in Uganda) and South
Asia (2 in Bhutan, 27 in India, and 25 in Nepal).
We use the nonparametric Chao-1 estimator of tree
species richness as an indicator of forest biodiver-
sity. We use the percent of households that depend
substantially on the forest for subsistence live-
lihoods as an indicator of livelihood contributions
of the same forest (16).
We classify the outcome relationships be-
tween tree species richness and forest-based sub-
sistence livelihoods into categories on the basis
of above- or below-average levels for each of our
two indicator variables, relative to other forests in
thesameforesttypeinthedataset(16). Our ap-
proach focuses on three joint-outcome categories
where (i) species richness and livelihoods contri-
butions are both above average (sustainable forest
systems); (ii) species richness and livelihoods are
both below average (unsustainable forest sys-
tems); and (iii) either species richness is above
average relative to other forests and livelihoods
are below average, or species richness is below
average but livelihoods are above average (trade-
off forest systems).
We find that all possible combinations of
relationships between forest-based subsistence
livelihoods and tree species richness are present
in the data (Fig. 1A). The existence of multiple
patterns of relationships underscores the relevance
of analyses that seek to identify factors respon-
sible for facilitating or impeding trade-offs and
1
School of Natural Resources and Environment, University of
Michigan, 440 Church Street, Ann Arbor, MI 48109, USA.
2
De-
partment of Geography, University of Illinois at Urbana-
Champaign, 607 South Mathews Avenue, Urbana, IL 61801,
USA.
*To whom correspondence should be addressed. E-mail:
lpersha@umich.edu
Fig. 1. (A) Tree species rich-
ness and subsistence liveli-
hoods zscores, standardized
within each forest type. Black
circles indicate East African
cases (n= 30), and gray cir-
cles indicate South Asian
cases (n= 54). (B)Distri-
bution of sustainable, un-
sustainable, and trade-off
outcomes, by region. H-H,
above-average standardized
values for tree species rich-
ness and subsistence live-
lihoods; L-L, below-average
standardized values for tree
species richness and subsistence livelihoods; SL, subsistence livelihoods; TSR, tree species richness.
25 MARCH 2011 VOL 331 SCIENCE www.sciencemag.org1606
REPORTS
on May 15, 2011www.sciencemag.orgDownloaded from
synergies across social and ecological outcomes.
Most of our 84 cases (60%) are characterized
by trade-off relationships, although jointly posi-
tive outcomes across biodiversity and livelihoods
are also well represented (27% of cases). Jointly
negative outcomes are less common (13% of
cases) (Fig. 1B). The distribution of outcomes
across all possible categories suggests that there
is no universally applicable positive or negative
association between livelihoods and biodiversity
to be found in the studied forests. Outcomes are
similarly distributed across the two regions cov-
ered in our study (likelihood-ratio Χ
2
= 2.77, df =
2, P= 0.251), and region was not a significant
predictor of outcomes.
We also examined how a set of hypothesized
social and ecological mediating factors affects the
observed outcomes, using ordered logistic regres-
sion analysis to identify important predictors of
the likelihood of obtaining jointly negative, trade-
off, or positive outcomes. Our model included
three independent variables that are a focus of
much theoretical and empirical work related to
biodiversity conservation and sustainable liveli-
hoods outcomes: forest size, formal participation
(as conferred through policy) of local forest users
in forest rulemaking (hereafter, “rulemaking par-
ticipation”), and dependence on the forest for
extractive commercial livelihoods (in our data,
primarily charcoaling, small-scale timber harvest-
ing, fuelwood, and collection of nontimber forest
products for cash income). Forest patch size is
a key factor relating to potential species richness,
as well as to prospects for sustainable forest man-
agement due to monitoring and enforcement chal-
lenges related to scale (17); rulemaking participation
is highlighted as important for obtaining local
knowledge necessary to improve forest resources,
promoting legitimacy over forest rules, and en-
gendering management accountability (18); a high
level of commercial extractive forest use may be
viewed as reducing the likelihood of sustainable
forest outcomes across subsistence livelihoods and
species richness, due to higher access inequities
for poor households and negative impacts on bio-
diversity objectives (19–22).
Our results indicate that forest systems are
more likely to have sustainable outcomes (above-
average tree species richness and subsistence
livelihoods) when local forest users participate
in forest rulemaking (z= 2.21, P= 0.027),
whereas unsustainable forest system outcomes
are more likely when users do not participate in
rulemaking (z=−2.62, P=0.009;Tables1and2).
The size of the forest and the extent to which the
forest provided commercial livelihoods to house-
holds are also important factors associated with
Table 1. Marginal effects of ordered logit regression (rulemaking participation held
at its median value, all other variables at their means). Dependent variable: tree
species richness and subsistence livelihoods joint outcome, with three categories:
low-low, trade-off, or high-high. N=84.WaldtestX
2
(3) = 9.33, P= 0.0252,
pseudo R
2
= 0.0903. Independent variables: “forest size,”logarithm of forest
size (ha); “rulemaking participation,”local forest user participation in forest
rulemaking (0 = No, 1 = Yes); “commercial livelihoods”(percent of local forest
users who depend substantially on the forest for cash income).
Independent variable Marg. effect Std. error zP>z[95% CI ] X
Outcome category: low-low (unsustainable forest systems)
Forest size −0.036 0.019 −1.820 0.069 −0.074 0.003 5.821
Rulemaking participation −0.119 0.045 −2.620 0.009 −0.208 −0.030 0.000
Commercial livelihoods −0.215 0.122 −1.750 0.080 −0.456 0.026 0.207
Outcome category: trade-offs
Forest size 0.001 0.014 0.070 0.943 −0.027 0.029 5.821
Rulemaking participation −0.158 0.101 −1.570 0.117 −0.355 0.039 0.000
Commercial livelihoods 0.006 0.087 0.070 0.944 −0.164 0.177 0.207
Outcome category: high-high (sustainable forest systems)
Forest size 0.035 0.018 1.980 0.048 0.003 0.068 5.821
Rulemaking participation 0.277 0.125 2.210 0.027 0.031 0.522 0.000
Commercial livelihoods 0.209 0.092 2.270 0.023 0.028 0.390 0.207
Fig. 2. Predicted probabilities of (A) unsustainable, (B) trade-off, and (C)
sustainable forest system outcomes when local forest users participate in forest
rulemaking (solid black line) and when they do not participate in forest
rulemaking (dashed black line), as forest size increases. The gray shaded area
between probability curves in each panel shows the decrease in the predicted
probabilities of unsustainable forest systems and trade-offs, and the increase
in the predicted probability of sustainable forest systems, when local forest
users participate in forest rulemaking.
www.sciencemag.org SCIENCE VOL 331 25 MARCH 2011 1607
REPORTS
on May 15, 2011www.sciencemag.orgDownloaded from
obtaining either sustainable or unsustainable for-
est system outcomes, with a higher likelihood of
sustainable outcomes as forest size and commer-
cial livelihoods dependence increase [Table 1;
note that the amount of benefits from commercial
livelihoods, though important at the individual
level, constitutes a small proportion of liveli-
hoods for many households and is relatively low
overall for our data (16)].
To better understand how interactions be-
tween forest size and rulemaking participation
together affect the likelihood of obtaining jointly
positive outcomes, we examined how the pre-
dicted probability of each of the three outcomes
varies across the range of forest sizes in our data
set, in the presence and absence of rulemaking
participation by local forest users (Fig. 2). We
find that rulemaking participation is associated
with a lower probability of less desirable out-
comes (unsustainable forest systems and those
characterized by trade-offs) and a higher proba-
bility of sustainable forest system outcomes, across
smaller and larger forests. But our results suggest
that participation in rulemaking may be especial-
ly important in promoting positive outcomes in
small forest fragments, where greater challenges
to achieving jointly positive outcomes across bio-
diversity and livelihoods already exist (Fig. 2).
We conclude that working toward formal partic-
ipation of local forest users in rulemaking pro-
cesses for use and management of forests from
which they draw their livelihoods (irrespective of
whether such activities are sanctioned under pre-
vailing rules) is an important way to increase the
probability of obtaining more positive outcomes
across social and ecological dimensions.
Further work is needed to understand the
causal mechanisms that underlie such outcomes.
One proposed mechanism is that rulemaking par-
ticipation provides an opportunity for local forest
users to contribute more specific and locally rele-
vant information on forest resources and dynam-
ics of use for a given forest, which in turn leads to
the construction of rules that are viewed as legit-
imate and better suited for local forest conditions.
Rulemaking participation may also help shift
incentive structures for forest users to undertake
decisions aimed toward a more balanced prioriti-
zation between activities that maintain good for-
est conditions and benefit flows over longer time
horizons (hence biodiversity conservation in-
directly) and shorter-term livelihoods benefits.
In constructing our model, we also tested for
the significance of additional variables, particular-
ly market distance and population density, which
have been found in prior work to be associated
with biodiversity and livelihoods when treated
as independent outcomes. However, we do not
find a statistically significant association between
these other factors and our joint outcome cate-
gories (16). Species richness does not encompass
all facets of biodiversity, and species compo-
sition is also particularly important for assessing
biodiversity conservation (9). Our use of tree
species richness as an indicator of biodiversity is
supported by a strong positive correlation of this
variable with an abundance-based similarity in-
dex of tree species composition calculated for
each of our cases in comparison with a minimally
disturbed reference forest within the same forest
type (16).
There are some important regional differ-
ences in the broader set of biophysical, socio-
economic, and institutional factors associated
with the East African versus South Asian cases
(table S2). Forests are larger on average in East
Africa, and a greater proportion of households
rely on the forests for commercial income. We
also find differences in the strength of association
of some of these explanatory and broader con-
textual factors between the two regions, even as
overall patterns of outcomes in the relationship
between tree species diversity and subsistence
livelihoods are similar. We suggest that this may
point to the likelihood of multiple pathways for
achieving these outcomes, differentiated, for in-
stance, across varied regional contexts and key
factors that also likely operate at broader scales.
Recognition that forested ecosystems simul-
taneously generate multiple services is embodied
in forestry policy decentralization reforms that
have doubled the area of forest land under com-
munity ownership or management in the past 15
years (6). These efforts rest on an assumption that
synergies across multiple forest outcomes can be
achieved, yet existing scholarship provides only
limited guidance on avenues by which policies
might better promote these synergies. Our data
and analysis from a large number of cases from
different countries, forest types, and social contexts
suggest that trade-offs favoring forest conser-
vation objectives or immediate human livelihoods,
as well as jointly positive or negative results, are
each possible across the many different contexts
that comprise the human-dominated forested land-
scapes of East Africa and South Asia. Although
achieving desirable outcomes across potentially
competing social and ecological objectives is a
complex process, our analysis suggests that jointly
positive outcomes are far more likely when forest
users participate in rulemaking aspects of forest
governance. Our findings are particularly relevant
for small forest patches in human-dominated land-
scapes (especially forests under 200 ha), which
often present a particular challenge for achieving
jointly positive results.
References and Notes
1. W. D. Sunderlin et al., World Dev. 33, 1383 (2005).
2. A. M. Larson, F. Soto, Annu. Rev. Environ. Resour.
33, 213 (2008).
3. A. Agrawal, A. Chhatre, R. Hardin, Science 320,
1460 (2008).
4. A. Agrawal, E. Ostrom, Polit. Soc. 29, 485 (2001).
5. E. Barrow, K.-R. Jones, I. Nhantumbo, R. Oyono,
M. Savadogo, “Cutomary practices and forest
tenure reforms in Africa: Status, issues and lessons”
(Rights and Resources Initiative, Washington,
DC, 2009).
6. W. D. Sunderlin, J. Hatcher, M. Liddle, “From Exclusion to
Ownership? Challenges and opportunities in advancing
forest tenure reform”(Rights and Resources Initiative,
Washington, DC, 2008).
7. K. P. Andersson, C. C. Gibson, F. Lehoucq, World Dev.
34, 576 (2006).
8. E. Ost rom, Proc. Natl. Acad. Sci. U.S.A. 104,
15181 (2007).
9. T. A. Gardner et al., Ecol. Lett. 12, 561 (2009).
10. W. M. Adams et al., Science 306, 1146 (2004).
11. S. Wunder, World Dev. 29, 1817 (2001).
12. R. L. Chazdon et al., Biotropica 41, 142 (2009).
13. L. Naughton-Treves, M. B. Holland, K. Brandon, Annu.
Rev. Environ. Resour. 30, 219 (2005).
14. E. Wollenberg, L. Merino, A. Agrawal, E. Ostrom,
Int. Forest. Rev. 9, 670 (2007).
15. www.snre.umich.edu/~ifri
16. Materials and methods are available as supporting
material on Science Online.
17. J. L. Hill, P. J. Curran, J. Biogeogr. 30, 1391 (2003).
18. J. C. Ribot, A. Agrawal, A. M. Larson, World Dev. 34,
1864 (2006).
19. B. M. Belcher , Int. Forestry Rev. 7, 82 (2005).
20. B. Belcher, M. Ruiz-Perez, R. Achdiawan, World Dev. 33,
1435 (2005).
21. T. A. Gardner, J. Barlow, N. S. Sodhi, C. A. Peres,
Biol. Conserv. 143, 2293 (2010).
22. F. E. Putz, G. M. Blate, K. H. Redford, R. Fimbel,
J. Robinson, Conserv. Biol. 15, 7 (2001).
23. We thank E. Ostrom, T. Hayes, two anonymous reviewers,
and members of the Development, Sus tainable
Livelihoods and Conservation (DESUL ICO) group fo r
comments on an earlier draft of this paper, and
R. Kornak and J. England for assistance with data
cleaning. We gratefully acknowledg e the many
researchers affiliated with IFRI as Coll aborating
Research Cent ers for their data contributions and
ongoing involvement with the IFRI network. This work
was funded by the Ford Foundation, the MacArthur
Foundation, National Science Foundation grants
BCS-0703073 and CNH-0709545, and a Uni versity
of Michigan OVPR/Rackham grant for an Annual
Institute on Joint Outcomes related to Sustainability.
Supporting Online Material
www.sciencemag.org/cgi/content/full/331/6024/1606/DC1
Materials and Methods
Figs. S1 to S3
Tables S1 and S2
References
10.1126/science.1199343
Table 2. Predicted probabilities of obtaining unsustainable, trade-off, or sustainable forest system
outcomes as local forest users gain participation in forest rulemaking.
Outcome category
Predicted probability of
outcome (%): Forest
rulemaking participation
by local forest users
Change in probability as
local forest users gain
rulemaking participation
(95% CI)
No Yes
Unsustainable outcome (low-low) 0.17 0.05 −0.12 (−0.21, −0.03)
Trade-off outcomes 0.68 0.52 −0.16 (−0.32, 0.00)
Sustainable outcome (high-high) 0.16 0.44 0.28 (0.07, 0.48)
25 MARCH 2011 VOL 331 SCIENCE www.sciencemag.org1608
REPORTS
on May 15, 2011www.sciencemag.orgDownloaded from