Highlights
Global
Environmental
Change
xxx
(2013)
xxx–xxx
Global
Environmental
Change
xxx
(2013)
xxx–xxx
The
importance
of
local
forest
benefits:
Economic
valuation
of
Non-Timber
Forest
Products
in
the
Eastern
Arc
Mountains
in
Tanzania
M.
Schaafsma
*
,
S.
Morse-Jones,
P.
Posen,
R.D.
Swetnam,
A.
Balmford,
I.J.
Bateman,
N.D.
Burgess,
S.A.O.
Chamshama,
B.
Fisher,
T.
Freeman,
V.
Geofrey,
R.E.
Green,
A.S.
Hepelwa,
A.
Herna
´
ndez-Sirvent,
S.
Hess,
G.C.
Kajembe,
G.
Kayharara,
M.
Kilonzo,
K.
Kulindwa,
J.F.
Lund,
S.S.
Madoffe,
L.
Mbwambo,
H.
Meilby,
Y.M.
Ngaga,
I.
Theilade,
T.
Treue,
P.
van
Beukering,
V.G.
Vyamana,
R.K.
Turner
Centre
for
Social
and
Economic
Research
on
the
Global
Environment,
University
of
East
Anglia,
Norwich
NR4
7TJ,
UK
We
value
four
Non-Timber
Forest
Products
from
the
Eastern
Arc
Mountains
in
Tanzania.
We
transfer
spatially
explicit
models
of
NTFP
collection
across
a
wide
area.
The
total
annual
benefit
flow
is
approximately
USD
42
million.
Households
in
the
lowest
income
quartiles
in
the
area
depend
most
on
these
products.
Conservation
initiatives
need
to
be
coordinated
with
poverty
and
energy
policies.
G
Model
JGEC
1217r
1
Contents
lists
available
at
ScienceDirect
Global
Environmental
Change
jo
ur
n
al
h
o
mep
ag
e:
www
.elsevier
.co
m
/loc
ate/g
lo
envc
h
a
1
2
The
importance
of
local
forest
benefits:
Economic
valuation
of
Non-
3
Timber
Forest
Products
in
the
Eastern
Arc
Mountains
in
Tanzania
4
M.
Schaafsma
Q1
a,
*
,
S.
Morse-Jones
a
,
P.
Posen
a
,
R.D.
Swetnam
b
,
A.
Balmford
b
,
I.J.
Bateman
a
,
5
N.D.
Burgess
b,c,d
,
S.A.O.
Chamshama
e
,
B.
Fisher
a,f
,
T.
Freeman
g
,
V.
Geofrey
h
,
R.E.
Green
b,i
,
6
A.S.
Hepelwa
j
,
A.
Herna
´
ndez-Sirvent
k
,
S.
Hess
l
,
G.C.
Kajembe
e
,
G.
Kayharara
h
,
M.
Kilonzo
e
,
7
K.
Kulindwa
i,m
,
J.F.
Lund
n
,
S.S.
Madoffe
e
,
L.
Mbwambo
o
,
H.
Meilby
n
,
Y.M.
Ngaga
e
,
8
I.
Theilade
n
,
T.
Treue
n
,
P.
van
Beukering
p
,
V.G.
Vyamana
e
,
R.K.
Turner
a
9
a
Centre
for
Social
and
Economic
Research
on
the
Global
Environment,
University
of
East
Anglia,
Norwich
NR4
7TJ,
UK
10
b
Conservation
Science
Group,
Department
of
Zoology,
University
of
Cambridge,
Downing
Street,
Cambridge
CB2
3EJ,
UK
11
c
Conservation
Science
Program,
World
Wildlife
Fund,
P.O.
Box
97180,
Washington,
DC,
USA
12
d
Centre
for
Macroecology,
Evolution
and
Climate,
Department
of
Biology,
University
of
Copenhagen,
Universitetsparken
15,
DK-2100
Copenhagen,
Denmark
13
e
Sokoine
University
of
Agriculture
(SUA),
P.O.
Box
3000,
Morogoro,
Tanzania
14
f
Woodrow
Wilson
School
of
Public
and
International
Affairs,
Princeton
University,
Princeton,
NJ
08544-1013,
USA
15
g
School
of
Applied
Sciences,
Cranfield
University,
Cranfield
MK43
0AL,
UK
16
h
Centre
for
Environmental
Economics
and
Development
Research
(CEDR),
Dar-es-Salaam,
Tanzania
17
i
Conservation
Science
Department,
RSPB,
The
Lodge,
Sandy,
Bedfordshire
SG19
2DL,
UK
18
j
Department
of
Economics,
University
of
Dar
es
Salaam,
P.O.
Box
35096,
Dar
es
Salaam,
Tanzania
19
k
School
of
Geography
and
the
Environment,
University
of
Oxford,
Oxford
OX1
3QY,
UK
20
l
Hess
Environmental
Economic
Analyst,
Elzenlaan
17,
1214
KJ
Hilversum,
The
Netherlands
21
m
Department
of
International
Environment
and
Development
Studies,
Norwegian
University
of
Life
Sciences
(UMB),
P.O.
Box
5003,
NO-1432
Aas,
Norway
22
n
Forest
&
Landscape,
University
of
Copenhagen,
DK-1958
Copenhagen,
Denmark
23
o
Tanzania
Forest
Research
Institute
(TAFORI),
P.O.
Box
1854,
Morogoro,
Tanzania
24
p
Institute
for
Environmental
Studies
(IVM),
Vrije
Universiteit
Amsterdam,
De
Boelelaan
1087,
1081
HV
Amsterdam,
The
Netherlands
25
26
1.
Introduction
27
More
than
800
million
people
worldwide
live
in
or
near
tropical
28
forests
and
savannas,
and
rely
on
these
ecosystems
and
their
29
services
and
welfare
benefits
for
fuel,
food
and
income
(Chomitz
30et
al.,
2007;
Boyd
and
Banzhaf,
2007;
Fisher
et
al.,
2009).
In
31Tanzania,
rural
households
largely
depend
on
agriculture
or
32natural
resources
as
their
main
source
of
income
(NBS,
2009).
33Tanzania
is
one
of
the
poorest
countries
in
the
world,
ranked
148th
34of
the
169
countries
on
the
Human
Development
Index
(UNDP,
352010).
Eighty-nine
percent
of
the
population
lives
below
the
$
361.25/day
poverty
line
(UNDP,
2010).
Poverty
is
mainly
a
rural
37phenomenon:
83%
of
the
households
below
the
national
food
38poverty
line
live
in
rural
areas
(NBS,
2009).
In
Tanzania,
direct
Global
Environmental
Change
xxx
(2013)
xxx–xxx
A
R
T
I
C
L
E
I
N
F
O
Article
history:
Received
3
September
2012
Received
in
revised
form
12
February
2013
Accepted
4
August
2013
Keywords:
Non-Timber
Forest
Products
Environmental
valuation
Benefit
transfer
Ecosystem
services
Forest
conservation
A
B
S
T
R
A
C
T
Understanding
the
spatial
distribution
of
the
quantity
and
economic
value
of
Non-Timber
Forest
Product
(NTFP)
collection
gives
insight
into
the
benefits
that
local
communities
obtain
from
forests,
and
can
inform
decisions
about
the
selection
of
forested
areas
that
are
eligible
for
conservation
and
enforcement
of
regulations.
In
this
paper
we
estimate
transferable
household
production
functions
of
NTFP
extraction
in
the
Eastern
Arc
Mountains
(EAM)
in
Tanzania,
based
on
information
from
seven
multi-site
datasets
related
to
the
behaviour
of
over
2000
households.
The
study
shows
that
the
total
benefit
flow
of
charcoal,
firewood,
poles
and
thatch
from
the
EAM
to
the
local
population
has
an
estimated
value
of
USD
42
million
per
year,
and
provides
an
important
source
of
additional
income
for
local
communities,
especially
the
poorest,
who
mainly
depend
on
subsistence
agriculture.
The
resulting
map
of
economic
values
shows
that
benefits
vary
highly
across
space
with
population
density,
infrastructure
and
resource
availability.
We
argue
that
if
further
restrictions
on
forest
access
to
promote
conservation
are
considered,
this
will
require
additional
policies
to
prevent
a
consequent
increase
in
poverty,
and
an
enforced
trade-
off
between
conservation
and
energy
supply
to
rural
and
urban
households.
ß
2013
Published
by
Elsevier
Ltd.
*
Corresponding
author.
Tel.:
+44
01603
593224;
fax:
+44
01603
591327.
E-mail
address:
m.schaafsma@uea.ac.uk
(M.
Schaafsma).
G
Model
JGEC
1217
1–11
Please
cite
this
article
in
press
as:
Schaafsma,
M.,
et
al.,
The
importance
of
local
forest
benefits:
Economic
valuation
of
Non-Timber
Forest
Products
in
the
Eastern
Arc
Mountains
in
Tanzania.
Global
Environ.
Change
(2013),
http://dx.doi.org/10.1016/j.gloenvcha.2013.08.018
Contents
lists
available
at
ScienceDirect
Global
Environmental
Change
jo
ur
n
al
h
o
mep
ag
e:
www
.elsevier
.co
m
/loc
ate/g
lo
envc
h
a
0959-3780/$
–
see
front
matter
ß
2013
Published
by
Elsevier
Ltd.
http://dx.doi.org/10.1016/j.gloenvcha.2013.08.018
39
dependence
on
ecosystem
services
is
high;
92%
of
rural
households
40
use
firewood
as
their
main
cooking
fuel,
whereas
over
50%
of
the
41
urban
population
uses
charcoal
(NBS,
2009).
The
collection
of
Non-
42
Timber
Forest
Products
(NTFPs)
for
house
construction
and
43
household
use
is
also
widespread,
driven
by
poverty
and
a
lack
44
of
means
to
invest
in
better
quality
housing
and
non-wood
45
substitute
products
Q2
(World
Bank,
2009).
For
these
communities,
46
ecosystem
final
services
benefits
in
the
form
of
NTFPs
provide
a
47
source
of
complementary
cash
income,
or
a
safety
net
when
48
agricultural
yields
are
low
(Anthon
et
al.,
2008;
Ngaga
et
al.,
2009).
49
In
addition
to
timber
extraction,
the
production
of
building
poles,
50
charcoal
and
firewood
has
led
to
overexploitation
of
forests
and
is
51
one
of
the
main
immediate
drivers
(alongside
agricultural
52
expansion)
of
forest
degradation
and
deforestation
in
Tanzania
53
(Hofstad,
1997;
Chiesa
et
al.,
2009;
Ahrends
et
al.,
2010;
URT,
54 2010).
Rapid
population
growth
puts
an
additional
increasing
55
pressure
on
these
natural
resources
in
the
country.
56
The
Eastern
Arc
Mountains
(EAM)
contain
over
21,500
km
2
57
woodlands,
which
are
very
important
for
carbon
storage
on
a
58
landscape
scale
(Willcock
et
al.,
2012),
and
4000
km
2
of
tropical
59
forests
(Platts
et
al.,
2011),
recognised
as
one
of
the
world’s
60
biodiversity
hotspots
(Myers
et
al.,
2000).
Tropical
forest
61
ecosystems
host
at
least
60%
of
the
terrestrial
biodiversity
(Dirzo
62
and
Raven,
2003;
Myers
et
al.,
2000)
and
contain
around
25%
of
the
63
carbon
in
the
terrestrial
biosphere
(Bonan,
2008).
Their
clearance
64
and
degradation
account
for
about
17%
of
annual
CO
2
emissions
65
worldwide
(IPCC,
2006).
Global
concerns
about
biodiversity
66
conservation
and
climate
change
mitigation
are
leading
to
rising
67
international
demand
to
reduce
degradation
and
deforestation
68
resulting
from
the
harvesting
of
timber
and
NTFPs.
However,
while
69
the
benefits
from
CO
2
sequestration
and
biodiversity
protection
70
accrue
to
the
entire
international
community
(Balmford
and
71
Whitten,
2003;
Strassburg
et
al.,
2010),
the
current
welfare
of
72
people
in
local
communities
in
developing
countries,
many
of
73
whom
already
live
near
the
poverty
line,
is
likely
to
decrease
if
74
NTFP
harvesting
is
restricted
(Wunder,
2001).
Accordingly,
the
75
costs
of
supplying
internationally
beneficial
conservation
services
76
would
be
carried
by
the
poorest
and
most
vulnerable
people.
77
The
trade-offs
between
socio-economic
impacts
and
forest
78
conservation
in
forest-rich
countries
with
high
levels
of
poverty
79
and
forest-dependency
are
increasingly
being
considered
in
80
international
conservation
initiatives,
including
the
UN’s
pro-
81
gramme
on
Reducing
Emissions
from
Deforestation
and
forest
82
Degradation
(REDD+,
see
UNFCCC,
2006;
Strassburg
et
al.,
2009)
83
and
the
Convention
on
Biological
Diversity
(CBD,
2002).
REDD+
is
84
aiming
to
mitigate
climate
change
for
the
benefits
of
the
global
85
population
by
reducing
forest
degradation,
with
a
payment
86
mechanism
yielding
co-benefits
for
poverty
alleviation.
Similarly,
87
the
CBD,
in
aiming
to
reduce
biodiversity
loss,
recognises
the
role
88
of
biodiversity
for
human
wellbeing
and
promotes
sustainable
use
89
and
equitable
benefit-sharing
(CBD,
2010).
The
CBD
objectives
90
have
been
integrated
in
the
Millennium
Development
Goals
and
its
91
strategies
to
reduce
extreme
poverty
(Sachs
et
al.,
2009).
92
To
achieve
equity
and
poverty
alleviation
objectives,
effective
93
forest
conservation
policies
should
not
only
be
informed
by
the
94
potential
for
carbon
sequestration
and
biodiversity
protection,
but
95
also
by
the
distribution
of
costs
and
benefits
of
forest
conservation
96
among
stakeholders
at
different
spatial
scales
(Hein
et
al.,
2006;
97
Turner
et
al.,
2010).
This
paper
aims
to
provide
insight
into
the
98
distribution
of
local
benefits
within
the
EAM,
by
modelling
and
99
mapping
NTFP
extraction
across
a
wide
spatial
scale.
A
better
100
understanding
of
the
spatial
variation
in
the
(opportunity)
costs
101
and
benefits
of
conserving
ecosystem
services,
conditioned
by
102
factors
such
as
resource
availability
and
population
density
103
(Naidoo
and
Ricketts,
2006;
Pagiola
and
Bosquet,
2009;
Turner
104
et
al.,
2010),
can
help
to
define
priority
areas
where
limited
105budgets
for
forest
and
biodiversity
conservation
would
have
106highest
overall
benefits
(Naidoo
et
al.,
2008).
This
is
especially
107relevant
for
the
montane
and
sub-montane
forests
of
the
EAM
in
108Tanzania,
where
the
benefits
of
protection
of
rare
and
endangered
109species
could
render
extractive
uses
of
these
forests
with
local
and
110national
benefits
problematic
(Burgess
et
al.,
2007,
2010).
111However,
effective
mechanisms
for
realising
stakeholder
benefits
112and
their
possible
redistribution
on
fairness
grounds
have
to
be
in
113place
to
avoid
adverse
poverty
and
equity
effects
of
forest
114conservation
initiatives.
The
equity
effects
of
conservation
115management
will
depend
on
who
is
considered
to
be
a
stakeholder
116and
how
much
they
gain
or
lose
under
a
conservation
policy.
117This
paper
presents
a
unique,
spatially
wide-scale
analysis
of
118NTFP
collection
across
the
EAM
of
Tanzania,
demonstrating
the
119importance
of
natural
resource
extraction
for
income
and
120sustenance
at
the
local
level.
Based
on
a
large
dataset
from
a
121number
of
household
surveys,
we
estimate
spatially
explicit,
122micro-economic
models
of
household
NTFP
collection,
and
transfer
123these
models
to
predict
the
economic
value
of
the
annual
flow
of
124NTFP
extracted
by
2.3
million
households
across
the
study
area
of
12550,000
km
2
.
In
the
next
section,
we
discuss
our
modelling
126approach
and
its
main
strengths.
The
case
study
is
described
in
127Section
3
and
the
results
of
our
analysis
are
presented
in
Section
4.
128In
Section
5,
we
put
our
results
into
a
wider
policy
context
and
129discuss
the
implications
of
our
findings
for
forest
conservation
130policy
and
the
links
with
other
policy
objectives
such
as
poverty
131reduction.
1322.
Methodological
approach
133Increasing
policy
interest
since
the
1980s
in
sustainable
134development,
social
forestry,
indigenous
people’s
rights,
and
the
135commercialisation
of
forest
products,
has
stimulated
a
rapid
136growth
of
the
number
of
studies
on
socio-economic
aspects
of
137NTFP
collection
and
forestry
dependence
(Neumann
and
Hirsch,
1382000).
The
use
of
these
studies
in
assessments
of
natural
resources
139to
inform
decision-making
at
national
level
has
been
limited
for
a
140number
of
reasons.
Most
of
these
studies
are
qualitative
in
nature
141or
describe
forest
dependency
in
terms
of
average
quantities
142extracted
by
households.
They
are
usually
also
rather
localised,
143focusing
on
a
particular
forest
or
community
(Croitoru,
2007)
and
144the
results
do
not
capture
heterogeneity
across
forests,
communi-
145ties
and
other
spatial
contexts.
This
inhibits
generalisation
of
their
146results
and
the
transfer
of
the
models
to
other
locations,
or
over
147more
extensive
spatial
scales
(Godoy
et
al.,
1993).
This
lack
of
148generalisable
information
induces
a
risk
that
NTFP
values
are
149omitted
from
strategic
decision-making
processes
altogether
if
150site-specific
information
is
unavailable,
with
potentially
serious
151effects
on
local
welfare
in
forest-dependent
areas.
There
is
a
152growing
need
at
national
and
international
policy
levels
for
153projections
at
large
spatial
scales
of
the
economic
values
local
154communities
derive
from
forests,
including
the
collection
of
NTFPs
155(Daily
et
al.,
2009).
Moreover,
in
light
of
the
urgency
of
policies
that
156foster
sustainable
development
in
forest
rich
countries
with
high
157poverty
rates,
such
information
has
to
be
provided
in
due
time
and
158in
a
cost-efficient
manner.
159Our
quantitative
bottom-up
modelling
approach
uses
survey
160information
on
actual
household
behaviour
from
multiple
loca-
161tions
over
a
wide
spatial
scale
and
different
spatial
contexts
to
162develop
a
spatially
explicit
and
transferable
household
production
163function.
A
full
explanation
of
this
approach
is
described
in
164Schaafsma
et
al.
(2012),
and
a
detailed
description
is
provided
in
165the
Supplementary
Material
–
Methods
and
Results.
Essentially,
166our
approach
involves
four
steps:
(1)
estimating
the
household
167‘‘production
function’’
of
NTFP
collection;
(2)
transferring
168this
function
across
the
total
study
area,
using
secondary
data
M.
Schaafsma
et
al.
/
Global
Environmental
Change
xxx
(2013)
xxx–xxx
2
G
Model
JGEC
1217
1–11
Please
cite
this
article
in
press
as:
Schaafsma,
M.,
et
al.,
The
importance
of
local
forest
benefits:
Economic
valuation
of
Non-Timber
Forest
Products
in
the
Eastern
Arc
Mountains
in
Tanzania.
Global
Environ.
Change
(2013),
http://dx.doi.org/10.1016/j.gloenvcha.2013.08.018
169
for
non-surveyed
areas;
(3)
aggregating
household
level
extraction
170
over
all
households
in
the
study
area,
and
(4)
turning
NTFP
171
quantities
into
economic
values.
172
This
approach
has
three
main
advantages.
The
first
is
that
the
173
estimated
annual
flows
of
ecosystem
values
reflect
the
realised
174
monetary
benefits
accruing
to
the
local
communities,
rather
than
a
175
projected
potential
flow
from
the
underlying
stocks.
Potential
176
harvesting
rates
do
not
reflect
the
actual
NTFP
benefits
that
can
be
177
derived,
because
they
will
be
constrained
by
physical
access
178
problems
such
as
steep
slopes,
and
because
markets
may
not
be
179
sufficiently
large
(Sheil
and
Wunder,
2002)
or
prices
not
180
sufficiently
high
to
cover
extraction
costs
in
remote
areas.
So
181
the
potential
stock
will
not
be
fully
harvestable,
and
it
is
still
open
182
to
question
what
the
sustainable
resource
take
rate
might
be.
The
183
second
related
advantage,
compared
to
top-down
approaches,
is
184
that
the
modelled
household
production
functions
(step
1)
are
185
based
on
micro-level
data
about
individual
decision-making
and
186
the
factors
that
affect
whether
and
how
much
to
collect.
In
our
187
bottom-up
approach,
the
models
empirically
capture
values
as
188
perceived
by
local
communities.
Top-down
approaches,
on
the
189
other
hand,
typically
start
with
forest
availability
and
production
190
to
express
values
per
hectare
(Batagoda
et
al.,
2000).
However,
they
191
fail
to
capture
the
effect
of
typical
household
characteristics
that
192
influence
the
decision
to
collect
NTFPs,
such
as
the
time
and
costs
193
involved
in
collection,
available
labour
(after
fulfilling
other
194
income
generating
activities)
and
capital,
market
access
and
195
demand,
transportation
options,
and
the
potential
gains
to
the
196
household
budget
of
selling
NTFPs
(de
Beer
and
McDermott,
1989).
197
The
third
strength
is
that
our
approach
uses
data
from
different
198
areas
with
different
socio-economic,
spatial
and
biological
199
conditions
and
can
therefore
assess
whether
these
factors
200
influence
the
cost
of
collection,
demand
and
availability
of
various
201
NTFPs.
NTFP
harvesting
efforts
and
forest
degradation
typically
202
vary
spatially
(Robinson
et
al.,
2002,
2008).
Forest
quality,
for
203
instance,
is
often
lower
near
villages
or
population
centres
(e.g.,
204
Ndangalasi
et
al.,
2007;
Ahrends
et
al.,
2010),
due
to
variation
in
205
NTFP
harvesting
behaviour
as
predicted
by
economic
theory:
the
206
distance
from
the
household
to
the
NTFP
harvesting
location
is
207
positively
correlated
with
the
opportunity
costs
of
labour
and
time
208
spent
to
collect
NTFPs
(e.g.,
Amacher
et
al.,
1996;
Ko
¨
hlin
and
Parks,
209
2001;
Pattanayak
and
Sills,
2001).
The
spatial
distribution
of
210
harvesting
efforts
is
also
affected
by
forest
accessibility,
forest
211
protection
status
and
enforcement
(Robinson
and
Lokina,
2009,
212
2011).
213
The
variability
of
NTFP
products
in
terms
of
the
frequency
of
214
collection
and
use,
the
areas
where
they
are
available,
their
215
marketability
and
legal
context,
imply
that
household
production
216
functions
will
differ
across
NTFPs.
Therefore,
we
develop
separate
217
models
for
each
NTFP,
showing
the
relationship
between
the
218
quantity
of
a
NTFP
extracted
by
an
individual
household
(our
219
dependent
variable)
and
land
cover
suitability
and
household
220
characteristics
(our
explanatory
factors).
In
this
NTFP-specific
221
approach,
it
is
possible
to
capture
such
differences
between
the
222
NTFPs,
unlike
an
aggregate
model
in
which
estimates
of
total
NTFP
223
income
is
used
as
the
dependent
variable.
This
may
also
in
turn
224
allow
for
more
targeted
restriction
on
NTFPs
where
this
is
deemed
225
necessary
for
sustainable
forest
management.
226
Our
approach
thus
combines
the
strengths
of
micro-level
227
analysis
of
household
behaviour
with
those
of
large
spatial
scale
228
projections
of
forest
values.
The
household
production
functions
229
provide
a
spatially
explicit
evaluation
of
actual
household
NTFP
230
collection
and
production.
They
can
therefore
be
‘transferred’
231
across
the
study
area,
for
which
the
data
is
representative,
to
show
232
how
NTFP
collection
varies
with
socio-economic,
biophysical
233
and
ecological
factors.
NTFP
collection
and
its
benefits
can
234
therefore
be
estimated
for
the
entire
study
area
in
a
relatively
235rapid
and
cost-effective
manner,
avoiding
the
prohibitive
costs
of
236interviewing
all
households
in
the
area.
237A
limitation
of
such
a
spatially
extensive
estimation
of
ecosystem
238use
is
inevitably
its
accuracy
at
local
levels.
The
underlying
239assumption
of
function
transfer
is
that
the
relationship
between
240the
explanatory
and
dependent
variables
is
constant
between
241households
in
and
out
of
the
sample
(Rosenberger
and
Stanley,
2422006).
Function
transfer
is
expected
to
lead
to
more
accurate
results
243than
value
transfer
(Navrud
and
Ready,
2007),
where
the
mean
value
244is
taken
to
estimate
the
value
of
a
non-surveyed
site,
because
it
245allows
for
the
effects
of
contextual
factors
(but
see
Rosenberger
and
246Phipps,
2007;
Matthews
et
al.,
2009).
The
validity
of
our
approach
247hence
depends
on
the
quality
of
the
NTFP
collection
data,
the
248representativeness
of
the
sample,
and
the
specification
of
the
NTFP
249model
(Boyle
et
al.,
2009).
To
improve
accuracy
at
finer
spatial
scales,
250additional
local
analyses
are
recommended
for
local
policy
251development,
such
as
conservation
schemes
that
include
some
252form
of
compensation
to
individuals
or
households.
2533.
Case
study
254The
EAM
consist
of
13
mountain
blocks
extending
from
255southern
Kenya
to
eastern
Tanzania
with
a
total
area
of
over
25650,000
km
2
(Fig.
1).
The
dominant
natural
land
cover
is
miombo
Fig.
1.
Case
study
area.
Note:
The
NTFP
villages
reflect
the
villages
in
our
datasets
where
household
data
on
NTFP
collection
has
been
collected.
The
EAM
block
delineation,
based
on
Platts
et
al.
(2011),
reflects
the
area
for
which
NTFP
values
are
estimated.
The
river
basin
boundaries
reflect
the
larger
study
area
of
the
Valuing
the
Arc
project.
Source:
based
on
Schaafsma
et
al.
(2012).
M.
Schaafsma
et
al.
/
Global
Environmental
Change
xxx
(2013)
xxx–xxx
3
G
Model
JGEC
1217
1–11
Please
cite
this
article
in
press
as:
Schaafsma,
M.,
et
al.,
The
importance
of
local
forest
benefits:
Economic
valuation
of
Non-Timber
Forest
Products
in
the
Eastern
Arc
Mountains
in
Tanzania.
Global
Environ.
Change
(2013),
http://dx.doi.org/10.1016/j.gloenvcha.2013.08.018
257
woodland,
covering
approximately
42%
of
the
total
area,
of
which
258
10%
is
‘‘disturbed
miombo’’,
in
the
form
of
woodland
with
scattered
259
crops.
There
are
various
types
of
forests
depending
on
the
altitude:
260
lowland
forests
at
basin
levels,
sub-montane
and
montane
forests,
261
and
upper
montane
forests
at
highest
elevations
(Burgess
et
al.,
262
2007).
Apart
from
NTFPs,
important
EAM
ecosystem
services
263
include
the
provision
of
timber,
the
regulation
of
river
flows
for
264
drinking
water,
irrigation
and
hydropower,
and
carbon
storage
265
(Fisher
et
al.,
2011a).
Approximately
21%
of
the
EAM
blocks
are
266
protected
(Swetnam
et
al.,
2011),
including
75%
of
the
remaining
267
forests
and
24%
of
undisturbed
miombo
woodlands
(Platts
et
al.,
268
2011).
Pole
cutting,
charcoal
production
and
timber
harvesting
are
269
prohibited
in
Protected
Areas
and
licensed
under
other
manage-
270
ment
schemes.
Nevertheless,
illegal
extraction
of
NTFPs
and
timber
271
continues
in
Protected
Areas,
caused
by
multiple
and
interrelated
272
factors,
including
weak
enforcement
of
conservation
policies
and
273
poverty.
274
The
total
population
of
the
EAM
blocks
is
estimated
at
275
2.3
million
(based
on
Platts
et
al.,
2011),
with
a
mean
household
276
size
of
4.6.
Most
people
living
in
rural
Tanzania
depend
to
some
277
degree
on
the
collection
of
NTFPs,
a
situation
that
can
also
be
found
278
in
many
other
African
countries
(e.g.,
Shackleton
and
Shackleton,
279
2000,
2006;
Ambrose-Oji,
2003;
Mamo
et
al.,
2007;
Kamanga
et
al.,
280
2009;
Palmer
and
MacGregor,
2009).
In
the
EAM,
people
collect
281
firewood,
charcoal,
poles,
thatch,
fruits,
vegetables,
honey,
bush
282
meat,
and
medicines,
and
use
a
wide
range
of
species
(e.g.,
Luoga
283
et
al.,
2000;
Turpie,
2000;
Monela
et
al.,
2005;
Anthon
et
al.,
2008;
284
URT,
2008;
Robinson
and
Lokina,
2011).
In
this
study,
we
focus
on
285
the
first
four
of
these
NTFPs
and
we
therefore
provide
a
short
286
description
of
their
importance
for
urban
and
rural
livelihoods
and
287
the
trends
in
collection.
288
Firewood
is
collected
by
most
households
themselves,
but
only
289
2%
of
households
sell
it
onwards
(NBS,
2003).
As
demand
for
290
firewood
has
increased
due
to
population
growth,
the
availability
291
of
dead
wood
is
now
limited
in
some
areas.
In
such
cases,
people
292
have
increasingly
started
to
collect
live
wood,
which
can
threaten
293
the
sustainability
of
forest
use.
Substitution
to
alternative
energy
294
sources
or
more
fuel
efficient
stoves
is
still
very
limited
(Arnold
295
and
Ko
¨
hlin,
2003).
296
Whereas
the
rural
community
relies
mainly
on
firewood
for
297
cooking,
the
urban
population
commonly
uses
charcoal
(75%
of
298
households
in
Dar
es
Salaam
and
54%
in
other
urban
areas,
NBS,
299
2009).
Charcoal
production
takes
place
in
rural
areas.
In
the
lower
300
woodland
and
forest
areas
of
the
EAM,
charcoal
production
is
301
practised
for
commercial
purposes,
mainly
by
men
(Luoga
et
al.,
302
2000;
Anthon
et
al.,
2008).
Local
communities
are
seasonally
or
303
occasionally
involved
in
charcoal
production,
primarily
outside
304
planting
and
harvesting
seasons.
According
to
official
statistics
305
(NBS,
2003),
40%
of
charcoal-producing
households
sell
their
306
produce,
but
this
proportion
is
likely
to
be
higher
in
reality.
307
Charcoal
makers
sell
their
products
to
middlemen
who
transport
it
308
to
the
major
urban
centres
(Malimbwi
and
Zahabu,
2008).
Full-
309
time
charcoal
producers
often
move
around
the
country
to
new
310
production
sites.
311
Another
important
NTFP
used
by
many
rural
families
is
poles
312
(Burgess
and
Clarke,
2000;
Persha
and
Blomley,
2009),
used
for
the
313
construction
of
houses.
The
commercialisation
of
pole
cutting
is
314
small
with
only
6%
of
collecting
households
selling
their
poles,
315
mainly
to
neighbours
(NBS,
2003).
Due
to
diminishing
pole
316
availability
near
to
villages
in
some
areas,
villagers
are
increasingly
317
less
likely
to
sell
poles
(Robinson
and
Kajembe,
2009).
Some
318
households
now
prefer
to
build
brick
walls,
which
they
sometimes
319
finance
by
small
loans
(Freeman,
2010).
Bricks
are
currently
more
320
expensive
than
poles
and
only
available
to
richer
families.
Since
321
bricks
are
usually
dried
using
firewood,
increasing
brick
use
may
322
reduce
the
availability
of
dead
wood
for
firewood
consumption.
323Thatch
is
widely
used
for
roofing,
because
it
is
considered
to
be
324cheap
and
also
a
traditional
building
material
(Monela
et
al.,
3252005).
In
miombo
areas,
grass
species
that
provide
useful
326thatching
material
are
abundant
(Campbell
et
al.,
2008).
Thatch
327collection
is
expected
to
have
a
less
detrimental
effect
on
forests
328than
fuel
wood
or
pole
collection,
and
is
an
important
ecosystem
329service
to
local
communities.
Thatch
is
not
traded
on
a
regular
330basis.
331To
test
and
demonstrate
our
approach,
we
acquired
four
332existing
datasets
on
NTFP
collection
in
the
EAM
and
set
up
333collaborations
with
three
other
projects
to
supplement
these
data
334and
extend
our
spatial
coverage
(see
Supplementary
material
–
335Data).
From
these
datasets,
household
information
from
villages
336within
40
km
of
the
EAM
boundaries
was
selected.
This
selection
337resulted
in
a
pooled
dataset
with
over
2000
observations
from
60
338villages.
The
availability
of
multiple
multi-site
datasets
of
339household
level
observations
on
NTFP
collection
in
Tanzania
340provided
the
opportunity
to
innovate
and
develop
spatially
explicit
341household
production
functions.
3424.
Economic
valuation
of
actual
NTFP
flows:
results
3434.1.
Forest
and
woodland
income
and
dependency:
sample
statistics
344The
sample
statistics
show
that
NTFPs
are
of
great
importance
345to
villagers
in
the
EAM
area
(see
Supplementary
material
–
Data).
346More
than
60%
of
houses
are
constructed
with
poles
and
half
of
the
347sample
has
thatched
roofs
(see
Supplementary
material
–
Table
348A.2).
For
13%
of
households
the
main
source
of
household
income
is
349forest
related,
including
timber
and
NTFP
collection.
NTFP
income
350(cash
and
non-cash)
accounts
on
average
for
20%
of
total
household
351income,
which
is
comparable
to
the
results
of
a
meta-analysis
of
352over
50
NTFP
studies
worldwide
by
Vedeld
et
al.
(2007),
which
353estimated
that
forest
environmental
income
represented
22%
of
354the
total
income
of
communities
living
near
forest
in
developing
355countries.
The
annual
median
household
income
of
the
sample
356corresponds
to
$
1.89
per
household
per
day
PPP-corrected,
357equivalent
to
a
daily
income
per
person
far
below
the
poverty
line.
358We
used
the
UNDATA
(2010)
PPP
conversion
factor
of
the
local
359currency
to
international
dollars
of
2007:
TSH
521,600
=
$
1.
The
360number
of
people
living
below
the
basic
needs
poverty
line
in
our
361sample
is
higher
than
census
data
indicate
(38%
in
rural
areas,
see
362NBS,
2009);
nevertheless
it
is
clear
that
the
households
in
the
363sample
are
very
poor.
364Income
is
unequally
distributed:
the
GINI-coefficient
of
our
365overall
sample
is
61%
(a
Gini
coefficient
of
0
percent
implies
366perfect
equality,
whereas
100
percent
implies
maximal
inequal-
367ity).
Excluding
NTFP
income
from
the
calculation
increases
368inequality
and
the
GINI-coefficient
to
65%.
Thus,
according
to
369our
data
access
to
NTFPs
reduces
inequality.
Splitting
the
sample
370into
income
quartiles
(Table
1)
shows
that
NTFP
income
(cash
371and
non-cash)
of
the
poorer
groups
is
lower
in
absolute
terms
372but
higher
relative
to
the
total
household
income,
compared
to
373richer
househ old s.
This
result
confirms
findings
by
earlier
socio-
374economic
studies
(e.g.,
Cavendish,
2000;
Mamo
et
al.,
2007;
375Kamanga
et
al.,
2009).
Of
course,
the
terms
rich
and
poor
should
376be
interpreted
with
caution,
as
the
mean
annual
household
377income
of
the
richest
group
is
only
TSH
2
million
(PPP
$
4123).
378In
our
sample,
richer
households
are
less
involved
in
the
379collection
of
firewood
and
thatch,
but
they
are
more
likely
to
380produce
cha rcoal.
In
terms
of
quantity,
they
collect
more
381firewood
and
poles,
compared
to
poorer
households.
Differences
382in
quantities
for
charcoal
and
thatch
are
not
significant
at
the
5%
383level.
These
figures
confirm
that
NTFPs
reduce
relative
inequali-
384ty,
and
are
an
especially
important
source
of
income
for
the
385poorest
in
these
communities.
M.
Schaafsma
et
al.
/
Global
Environmental
Change
xxx
(2013)
xxx–xxx
4
G
Model
JGEC
1217
1–11
Please
cite
this
article
in
press
as:
Schaafsma,
M.,
et
al.,
The
importance
of
local
forest
benefits:
Economic
valuation
of
Non-Timber
Forest
Products
in
the
Eastern
Arc
Mountains
in
Tanzania.
Global
Environ.
Change
(2013),
http://dx.doi.org/10.1016/j.gloenvcha.2013.08.018
