Can Local Ecosystem Services Valuation Studies be Up-scaled for use in Global Assessments?
Dr Mike Christie
1
$
$
$
$
$
$
$
$
Overview• Global assessments of biodiversity value
• Research question
• Case study 1: Value transfer within the UK BAP
• Case study 2: Valuation in developing countries.
• Conclusions
2
3
Global assessments of biodiversity value: • Costanza (1997) Nature:
Value of all of the World’s ecosystem
services = $33 trillion / yr
• TEEB (2010) Annual global economic impact of
biodiversity loss = $2 - 4.5 trillion / yr
4
Global assessments of biodiversity value … • tend to be based on ‘value transfer’ where local
empirical ecosystem service valuation studies are aggregated or up-scaled to the global assessment.
• Value transfer is the process by which values are transferred (and adjusted) from a study site to a new policy site.• Unit value transfer• Function value transfer
5
Value transfer in global assessments
Study site
Policy sites
Can values be effectively
transferred across regions?
How suitable are local studies to feed into global assessments?
• Local valuation studies tend to: – Utilise a wide variety of valuation approaches; – Address different ranges of ecosystem services; – Have specific policy context settings; and – Be of varying quality.
• Arguably, this variation makes it difficult to extract synthesis conclusions to feed into global assessments.
6
Can Local Ecosystem Services Valuation Studies be Up-scaled for use in Global Assessments?
7
Key research questions
• Does value transfer work (in a controlled setting)?• How accurate are value transfers?• Can values be transferred across regions?
• If values can’t be transferred across regions, what are the implications for global assessments (and in particular assessments in developing countries)?
Case study 1: Testing consistency of value transfer across UK BAP
• Choice experiment used to value ecosystem services associated with the UK BAP
• Administered across 12 UK regions.
8
Choice experiment
example: UK BAP
9I prefer:I prefer: No BAPI prefer: No BAP Action Plan A Action Plan B
Can VT consistently perform well if differences between the study and policy sites are controlled?
– Can the value from a single study site predict the value of another policy site?
– Can the value from a number of study sites predict the value of another site?
– Value transfer test
10Transfer error = (|Predicted WTP – Observed WTP|) . Observed WTP
11
Policy sites
Scot
land
Wal
es
N. I
rela
nd
Nor
th E
ast
Nor
th
Wes
t
Sout
h Ea
st
Sout
h W
est
East
of
Engl
and
East
M
idla
nds
Gre
ater
Lo
ndon
York
and
H
umbe
r
Wes
t M
idla
nds
All U
K
Ave
rage
Stud
y si
tes
Scotland 0
(0)72
(74)68
(52)158
(215)376
(570)63
(45)131
(107)962
(1787)78
(31)87
(73)88
(50)78
(49)135
(151)191
(267)
Wales133227)
0(0)
65(39)
503(1205)
264(493)
66(30)
110(162)
370(565)
61(24)
73(77)
70(51)
46(28)
101(96)
155(249)
N. Ireland978
(2262)473
(1002)0
(0)543
(826)330
(357)63
(41)261
(230)1054
(1791)56
(30)396
(815)153
(140)57
(69)460
(750)402
(692)
North East113
(184)78
(58)83
(70)0
(0)327
(723)60
(34)117
(130)420
(679)65
(30)118
(103)65
(48)50
(30)282
(266)148
(196)
North West231
(330)100
(111)55
(34)417
(990)0
(0)63
(33)92
(112)368
(736)66
(33)93
(78)65
(35)50
(33)105(88)
142(217)
South East538
(981)281
(411)127
(178)399
(555)731
(1309)0
(0)295
(519)941
(1425)37
(28)358
(405)81
(37)61
(83)397
(378)353
(525)
South West267
(328)99
(103)122
(129)570
(1325)523
(1158)66
(40)0
(0)481
(710)54
(37)180
(195)68
(67)52
(41)174
(129)221
(355)
East of England578
(1178)252
(516)93
(86)468
(937)394
(847)61
(32)123
(111)0
(0)68
(28)266
(415)83
(49)54
(33)251
(381)224
(384)
East Midlands930
(1653)441
(726)155
(198)843
(1559)860
(1479)88
(115)410
(546)806
(960)0
(0)541
(582)220
(249)81
(73)588
(514)496
(721)
Greater London98
(121)39
(26)52
(22)282
(554)151
(186)52
(37)99
(83)533
(938)67
(30)0
(0)63
(22)59
(21)56
(67)129
(175)
York and Humber266
(406)137
(167)84
(62)248
(446)372
(701)44
(10)150
(275)743
(1186)58
(21)279
(335)0
(0)45
(23)181
(144)217
(314)
West Midlands475
(877)215
(377)86
(168)429
(874)601
(1274)38
(32)163
(196)610
(918)37
(22)279
(335)87
(89)0
(0)278
(256)274
(451)
All UK218
(344)56
(20)64
(18)191
(381)153
(263)67
(26)92
(73)428
(637)79
(20)63
(54)62
(21)66
(16)0
(0)128
(156)
% times that ‘pooled’ data reduced transfer errors 73% 91% 82% 91% 91% 9% 0% 73% 0% 100% 100% 18% 67%
Value transfer test: transfer errors
Value transfer: results
12
• Large variations in size of the transfer errors when single study site used: 37% up to 1054%
• Pooling data across several study sites generally (but not always) reduces transfer errors: 56% to 428%.
• Implications: Even in a controlled setting (within same country and using same valuation instrument), the accuracy of VT is questionable!
If values can’t be transferred across regions, what are the implications for global assessments (and particularly
assessments in developing countries)?
– To what extent are there local valuation studies in developing countries to feed into global assessments?
– Do current approaches to valuation work in developing countries?
– What is current best practice for valuation in developing countries?
13
An Evaluation of Economic and Non-economic Techniques for Assessing the Importance of Biodiversity and associated
Ecosystem Services to People in Developing Countries.
14
Dr Mike Christie
Aims• To review the extent to which valuation has
been applied in developing countries.
• To review the methodological, practical and policy challenges associated with valuation in developing countries.
• To develop good practice guidance for valuation in developing countries.
15
To what extent are there local valuation studies in
developing countries to feed into global assessments?
• No. of studies that valued biodiversity in developing countries:– EVRI:
• 195 economic studies – Web of Knowledge:
• 183 economic studies • 101 non-economic studies.
• Developing country studies represent 1/10th of all published biodiversity valuation studies.
16
0
200
400
600
800
1000
1200
1400
1960-1964
1965-1969
1970-1974
1975-1979
1980-1984
1985-1989
1990-1994
1995-1999
2000-2004
2005-2008
NU
MB
ER O
FPA
PER
S
YEARS
HIGH INCOME
UPPER MIDDLE INCOME
LOWER MIDDLE INCOME
LOWER INCOME
TRANSITIONAL INCOME
Cumulative total of biodiversity valuation studies
sourced from EVRI for all National Income States from
1960- 2008.
Methodological challenges• Low levels of literacy, education and language
– Creates barriers to valuing complex environmental goods– Makes it difficult to utilise traditional survey techniques such as
questionnaires and interviews. – More deliberative and participatory approaches to data collection may
overcome these issues.
• Informal or subsistence economies– People may have little or no experience of dealing with money. – People find it difficult to place a monetary value on complex and
unfamiliar environmental goods / services.
• Valuation methods have been developed in developed countries– Are the current best-practice guidelines appropriate for applications in
developing countries?
17
Practical challenges• Extreme environmental conditions
– Affect the researcher’s ability to access areas or effectively undertake research.
• Lack of local research capacity to design, administer and analyse research projects– Involvement of local people is considered essential within the research
process to ensure that local nuances / values are accounted for.
• Sometimes easier to administer valuation studies in developing countries– Response rates are typically higher– Respondents are receptive to listening and considering questions posed– Interviewers are relatively inexpensive (allowing larger sample sizes).
18
Policy challenges• Lack of local research capacity
– Lack of awareness of valuation methods and of the importance of biodiversity to people.
• Lack of empirical valuation studies in developing countries– Little evidence to illustrate the importance of biodiversity to people.– Makes future value transfer difficult.
• Existing research is often extractive– Often little engagement with local communities, researchers or
decision makers. – Research therefore has very little impact on the welfare of local
people.
19
Conclusions: Best practice guidance• The way people in developing countries think about the natural
environment is different to those in developed countries. – They have much closer ties to their natural environment. – Their knowledge is often implicit and / or experiential knowledge– Low levels of literacy and education mean that most people will have little or
no scientific understanding of their natural environment. – The above means that people from developing countries may have difficulty
in expressing their value for natural resources.
• Given the above, standard approaches to valuation are unlikely to effectively reveal the preferences of people in developing countries.
• Valuation may be more effective if: – Local researchers are used throughout the research process– Deliberative, participative and action research approaches are incorporated
into the valuation methods.
20
Can Local Ecosystem Services Valuation Studies be Up-scaled for use in Global Assessments?
• Depends on level of accuracy needed!
• Generally, value transfer performs poorly (even in controlled settings) and therefore it may be questionable as to whether it can effectively be utilised in global assessments.
• Currently there are only a limited local studies in developing countries, and novel valuation approaches need to be developed to expand this knowledge base.
21
References• Christie, M., Fazey, I., Cooper, R., Hyde, T., Deri, A., Hughes, L., Bush, G., Brander, L.,
Nahman, A., de Lange, W. and Reyers, B. (2008). An Evaluation of Economic and Non-economic Techniques for Assessing the Importance of Biodiversity to People in Developing Countries. Defra: London.
• Christie, M., Fazey, I., and Hyde T (2010). Economic Valuation of the Benefits of the UK Biodiversity Action Plan. Defra: London
• Kenter J, Hyde T, Christie M and Fazey I (2011 In press). The importance of deliberation in valuing ecosystem services in developing countries – evidence from the Solomon Islands. Global Environmental Change Human and Policy Dimensions.
• Fazey, I, Kesby, M, Evely, A, Latham, I, Wagatora, D, Hagasua, J-E, Reed, M, S, Christie, M (2010). A three-tiered approach to participatory vulnerability assessment in the Solomon Islands. Global Environmental Change. 20, 713-728., doi:10.1016/j.gloenvcha.2010.04.011
22