Valuing Marine and Coastal ecosystem service benefits: Case study of St Vincent

and the Grenadines' proposed Marine Protected Areas.

Michael Christiea,b, Kyriaki Remoundoua, Ewa Siwickaa and Warwick Wainwrighta

a School of Management and Business, Aberystwyth University, Aberystwyth, Wales, UK,

SY233AL

b Corresponding author - email: mec@aber.ac.uk, tel: 01970 622217

Abstract

This paper reports the results of a choice experiment (CE) that values the ecosystem service

benefits from extending the current network of Marine Protected Areas (MPAs) in St Vincent

and the Grenadines (SVG), Caribbean. We considered two future options: an ‘improved’

scenario in which marine protection is increased, and a ‘decline’ scenario in which current

protection mechanisms are removed. The CE was administered at two sites (the degraded St

Vincent South Coast and the pristine Tobago Cays) and to tourists and local residents. Results

suggest that both groups value health protection, fishing, coastal protection, ecosystem

resilience, and diving / snorkelling. Values are higher for the ‘decline’ scenario compared to

the ‘improved’ scenario. Also, tourists had significantly higher WTP values than locals. Our

analysis also enabled an evaluation of the benefits derived from alternative policy

interventions that may be used to protect and enhance SVG’s marine parks. Stopping

pollution from agriculture run-off and sewage was found to generate the highest ecosystem

service benefits, with restricting over-fishing and bad fishing practices also being important.

We demonstrate economic valuation of marine ecosystem service might be used to design and

target marine conservation policies that maximise welfare benefits.

Keywords: marine and costal ecosystems, ecosystem services, choice experiment, Caribbean

Acknowledgements: We would like to thank the Government of St Vincent and the

Grenadines for supporting this research.

1

1 Introduction: Marine and coastal ecosystems, ecosystem services and

valuation.

Marine and coastal ecosystems are one of the most productive, diverse and valuable

ecosystems on Earth (Souter and Linden, 2000; Spalding et al., 2001; Wilkinson,

2008; Wilkinson and Buddemeier, 1994). For example, coral reefs are thought to host

25% of the World’s fish species (Spalding et al., 2001). However, marine and coastal

ecosystems are currently being threatened worldwide from a wide range of

anthropocentric as well as non-anthropocentric pressures including unsustainable

fishing practices, the development of tourism and urban infrastructure, pollution from

land-based sources, ocean acidification and sea level rise (Allsopp et al., 2009;

Beharry-Borg and Scarpa, 2010; Butchart et al., 2010; Cinner et al., 2012; Secretariat

of the Convention on Biological Diversity, 2010; van Beukering et al., 2007). This

has resulted in a significant loss and degradation of these important habitats (Butchart

et al., 2010; Jameson et al., 1995; Moberg and Folke, 1999) which in turn is likely to

negatively impact the welfare and livelihood of people living in coastal areas (MEA,

2005; van Beukering et al., 2007). Wilkinson (2004) estimates that 30 million people

in coastal and island communities are totally reliant on reef-based ecosystems for their

primary means of food production, income and livelihoods, while Cesar et al. (2003)

estimates that the global reef-based tourism and recreation market is worth US$9.6

billion per annum.

Marine and coastal ecosystems provide a wide array of services, which are of value to

human populations. Waite et al. (2014) classifies marine and coastal ecosystem

services into provisioning services (Food / fisheries, Raw materials, Medicinal

2

resources and Genetic resource), regulating services (Flood / storm / erosion

regulation, Climate regulation), cultural services (Tourism and recreation, History,

cultural and traditions, Science, knowledge and education) and supporting services

(Primary production, Nutrient cycling, Species / ecosystem protection). Increasingly,

however, it is being recognised that a key contributing factor to the loss and

degradation of ecosystems has been a failure of people to fully recognise, and account

for, the range of ‘ecosystem service’ benefits provided by those ecosystems (Costanza

et al., 1997; Daily, 1997; MA, 2005; Sachs et al., 2009; Secretariat of the Convention

on Biological Diversity, 2000; TEEB, 2010). Understanding the values of biodiversity

and ecosystem services and embedding these values in decision-making is essential

for ensuring more equitable, cost effective and sustainable biodiversity conservation

policies (TEEB, 2011a, b). Further, demonstrating the economic and societal benefits

from marine conservation is important to (i) justify expenditures on marine

conservation programmes, and (ii) to provide evidence to help better target resources

to policy actions that maximise societal benefits (or minimise costs). The ecosystem

services approach to policy appraisal provides a mechanism to achieve such ambitions

(TEEB, 2011a, b; Waite et al., 2014).

Over the past decade, there has been a significant research effort to value the

ecosystem services provided by marine and coastal biodiversity (Schuhmann, 2012;

Schumann and Mahon, 2014; van Beukering et al., 2007; Waite et al., 2014), and

explore how these values may be embedded into decision making (TEEB, 2011a, b;

Waite et al., 2014). Most of these studies, however, tend to only address a handful of

ecosystem services with only a limited number of studies having attempted to value

the full range of provisioning, regulating and cultural services.

3

This paper contributes to the literature eliciting values for marine and coastal

ecosystem services by implementing a stated preference choice experiment that aims

to evaluate the benefits derived from protecting and enhancing a range of ecosystem

services delivered through two proposed Marine Protected Areas in St Vincent and

the Grenadines (SVG), Caribbean. The two case study sites are the proposed 2.5 Ha

South Coast marine park that is currently in a degraded condition due to human

pressures and the almost pristine 11.25 Ha Tobago Cays marine park. The evidence

collected in this study demonstrates that locals and tourists value different aspects of

SVG’s marine ecosystems, and we argue that these findings provide useful evidence

as to how MPA conservation policies might best be designed to maximise societal

benefits.

This paper is organised as follows. In the next section we provide some background to

the two case study marine parks in SVG. We then describe the choice experiment that

was used to value a bundle of ecosystem services provided by the two proposed parks.

The results from the CE are then reported. We conclude the paper with some policy

recommendation for the future management of SVG’s MPAs based on the findings of

the choice experiment.

4

2 Case study: Marine Protected Areas in St Vincent and the Grenadines

St. Vincent and the Grenadines (SVG) comprise a series of islands located in the

Eastern Caribbean. The main island of St. Vincent is 345 km2, while the Grenadines

are 44 km2 and comprise a series of smaller islands including Bequia, Mustique,

Canouan, Mayreau, Union Island, Palm Island, Petit St. Vincent, and 28 uninhabited

islets including those of the Tobago Cays. The combined population of SVG is

approximately 109,400 people (World Bank, 2012).

St. Vincent has a mountainous landscape that includes 12,700 ha of tropical forests,

while the Grenadines comprises low dry islands. All of the SVG islands are

surrounded by extensive coral reefs and seagrass beds that are globally significant and

are host to several endemic species (NPRBA, 2009). However, these important

ecosystems are currently being threatened by a number of anthropocentric activities,

including:

Tourism Development - In response to the downturn in the banana industry, the

Government of St Vincent and the Grenadines (GoSVG) made a strategic decision

to fill the void created in the national economy by targeting growth of the tourism

sector. Direct incomes from tourism account for a large proportion of the GDP.

However, uncontrolled tourism development is likely to be a major cause of

ecosystem degradation through: (i) destruction of habitats for the development of

infrastructure; (ii) degradation of habitats from an increased quantity of waste

generated; (iii) damage to habitats as a result of recreational activities; and (iv)

transportation facilitating the introduction of invasive alien species.

Over-exploitation and Unsustainable Uses of Biodiversity - Over-fishing, over-

hunting, over-grazing and over-harvesting are major causes of biodiversity loss in

5

the island ecosystems. Overfishing, for instance, can cause significant declines in

fish populations of coral reefs and can have long-lasting negative effects on all

aspects of reef ecology. Furthermore, overharvesting significantly threatens queen

conchs, spiny lobsters and hawksbill among other species.

Pollution and Waste Disposal - Pollution from liquid (e.g. agrochemicals) and

solid waste / sewage, is causing degradation of rivers, subsurface and coastal

water quality, compromising island habitats and having adverse effects on

recreational activities.

Deforestation and Land Degradation – The loss of forests to agriculture (legal and

illegal) in watershed areas is leading to soil erosion and land degradation,

reducing the capacity of the land to provide nutrient cycling and to support

biodiversity. Siltation of rivers and coastal areas due to run-off also threaten

sensitive riverine and coastal ecosystems.

Human activities are clearly impacting SVG’s marine ecosystems, which in turn

undermine the capacity of these habitats to deliver ecosystem services that are

fundamental for people’s well-being and livelihoods. To tackle some of these issues,

as well as to meet its commitments under the Convention on Biological Diversity

(CBD), the Government of SVG (GoSVG) has established a suite of protected areas.

These policies are set out in the GoSVG’s ‘SVG National Parks and Protected Areas

System Plan 2010 – 2014’ report (NPRBA, 2009). Currently, the GoSVG has

designated 35 protected area sites in SVG; of which there is one Marine Park, one

Marine Reserve and six Marine Conservation Areas (Table 1). Following

recommendations from Jackson (2004), the GoSVG now wish to consolidate and

6

upgrade SVG’s marine protected areas. The proposed new system would include five

Marine Parks, three Marine Reserves and three Marine Conservation Areas (Table 1).

Table 1: Current and proposed marine protected area designations in SVG.

Incr

easi

ng le

vels

of p

rote

ctio

n

Designation Current Protection Proposed Protection

MarineParks Tobago Cays

Tobago CaysSt Vincent South Coast

Chateaubelair Islet Marine ParkPetit Byahaut Marine ParkAnchor Reef Marine Park

MarineReserves Tobago Cays

CanouanIsle de Quatre

Mustique

MarineConservation

Areas

BequiaPetit St Vincent

Union Island / Palm IslandCanouan

Isle de QuatreMustique

BequiaPetit St Vincent

Union Island / Palm Island

To provide evidence to support these developments, the GoSVG commissioned this

study to assess the costs and benefits provided by with the proposed expansion of its

marine protected areas network. Specifically, this research will explore local people’s

and tourist’s preferences for a range of ecosystem services provided by the marine

and coastal ecosystems in SVG and the values they attached to the protection and

enhancement of their provision.

3 Research method

Quantifying the costs and benefits associated with changes in the provision of

ecosystem services requires researcher to first understand the complex ecological

linkages between biodiversity (the ecosystem) and ecosystem service provision, and

then perform a valuation study to examine how much people value the changes to

ecosystem service provision (Haines-Young and Potschin, 2008). Accordingly, in this

paper, we first conducted a series of stakeholder workshops to explore how MPA

policy in SVG may affect the capacity of marine ecosystems to deliver ecosystem

7

services. Once the linkages were established, preferences and values were elicited

through a choice experiment. Below, we detail on our research approach.

3.1 Stakeholder workshops: Linking policy interventions to changes in the

provision of ecosystem services.

A series of stakeholder workshops were conducted at the two case study sites in 2011

to collate information that would be used to inform the design of the choice

experiment survey instrument. Stakeholders participating in the workshops included

local policy makers, ecologists, ecosystem managers, fishermen, tourist businesses,

community based organisations, local residents and tourists. The workshops were

organised around a series of discussion topics that aimed to identify: the key

ecosystem service benefits derived from coastal ecosystems in SVG; the key threats to

these coastal ecosystems; the impact of these threats on the provision of ecosystem

services; and possible policy interventions to protect and enhance the delivery of

ecosystem services.

A key outcome of these workshops was a ‘Linkage table’, which quantifies the

strength of the linkage between alternative policy interventions and the provision of

ecosystem services (see Table 2). Policy interventions identified by the stakeholders

as being important for the protection and enhancement of marine ecosystem services

included: the elimination of marine pollution due to sewage discharge and runoff from

yachts, houses and hotels; the prevention of overfishing and bad fish practices; the

reduction of land-based pollution (mainly from agricultural activities) that contributes

to eutrophication and sedimentation; the prohibition of sand mining and coral

extraction; and the introduction of ‘No take zones’ that would ban fishing and

8

anchoring. Following the identification of policy interventions, the stakeholders were

then asked to consider the potential impact of these interventions on the delivery of

six ecosystem services. The services considered included: provisioning services

(fishing), regulation services (coastal protection; and water quality which was linked

to the risk of contracting a water-related disease), habitat services (species richness /

resilience) and cultural / recreational services (beach recreation; and diving /

snorkelling).

The resultant Linkage table is reported in Table 2. It is clear that there is a high degree

of variation in terms of the impacts of the alternative policy interventions on the

delivery of marine ecosystem services. For example, stopping overfishing was

considered to have a high impact on species diversity and diving / snorkelling; but

was not thought to impact water quality / human health. Similarly, stopping sand

mining and coral damage would have a high impact on coastal protection and beach

recreation, but have no impact on fishing, water quality / human health, or diving /

snorkelling. In our analysis, we use this Linkage table to help develop the policy

scenarios for the choice experiment, and to allocate the ecosystem service values

derived from the choice experiment to the alternative policy interventions. This, in

turn, will allow us to measure the economic value associated with each of the five

policy interventions listed in Table 2 in terms of how they impact the delivery of

marine ecosystem services.

9

Table 2: Links between policy interventions and the provision of ecosystem servicesPolicy interventions

Stop Sewage (yachts, houses

and hotels)

Stop Overfishing / bad fishing

practices

Stop Land Based Pollution (mainly agricultural, such

as pesticides, eutrophication and

sedimentation)

Stop sandmining / extraction of

coral

Introduce ‘No take zones’

that would ban fishing and anchoring

Eco

syst

em se

rvic

e im

pact

s

Provisioning:Fishing: increases in fish numbers, size

and desirable species

++ ++ ++ 0 +

Regulating:Coastal protection

Water quality / human health

+

++

+

0

+

++

+++

0

++

0

Habitat:Species diversity,

ecosystem resilience and genetic pool

+ +++ ++ + +++

Cultural / recreational:

Beach recreation

Diving / snorkelling recreation

+

++

+

+++

+

++

+++

0

0

++

Notes: ‘+++’ indicates a high level of linkages; ‘++’ a moderate level; ‘+’ a low level; and ‘0’ no linkage

During the workshops, we also asked the stakeholders to identify the current and

future levels of the ecosystem service delivery at the two case study MPA sites (St

Vincent South Coast and the Tobago Cays) under different policy scenarios. To

achieve this, stakeholders were first asked to describe what they felt may be the worst

and best case scenarios in terms of the provision of each of the six ecosystem services

across SVG as a whole. These descriptions are shown in column 2 and 5 in Table 3.

Next, using a scoring system in which the worst case scenario = 1 and the best case

scenario = 5, the stakeholders were asked to quantify the levels of ecosystem service

provision that would be attained at each of the two Marine Parks in their current state,

under an improved state (in which the case study sites would be effectively managed

10

as Marine Parks) and in a declined state (in which there was no management of the

parks). The results from this exercise are reported in Table 3.

As expected, the stakeholders indicated that improved management of the marine

parks would result in a higher level of ecosystem service delivery than the current

situation, while a reduction in the levels of management would reduce the levels of

ecosystem service delivery (Table 3). The stakeholders also identified that the Tobago

Cays had the capacity to deliver higher levels of services than St Vincent South Coast:

this largely reflects the fact that the Tobago Cays ecosystems are in a much better

condition than those on the South Coast (Table 3).

This information on the levels of ecosystem service delivery associated with different

management options was then used to design the choice experiment (see Section

3.2.2).

Table 3: Levels of ecosystem services in MPA for the Decline, Current and Improved scenarios

Ecosystem service Worst case scenario description

(1)

South coast

DeclineCurrentImprove

Tobago Cays

DeclineCurrentImprove

Best case scenario description

(5)

Fishing: increases in fish numbers, size and desirable species

No fish for fishing in and / or around MPA

234

345

A sustainable catch of large, desirable fish in and / or around

MPA

Coastal protection No coastal protection with very high risk of

flooding

234

245

Excellent coastal protection with no threat of flooding

Water quality / human health

High risk of gastro intestinal disorder/ skin rashes / ear infections

124

235

No health problems

Species diversity, ecosystem resilience and genetic pool – Existence value

Low species diversity and low resilience

124

245

High species diversity and very resilient

Beach recreation Beach largely eroded away

245

345

Large stable beaches

Diving / snorkelling recreation Poor quality dive site

1 3Excellent dive site

11

24

45

Notes: The numbers in the table relate to the three possible future scenarios: Current situation, Improved MPA management, Declined MPA management, where 1 = the worst case scenario and 5 = the best case scenario.

3.2 The choice experiment: valuing changes to the delivery of ecosystem

services

3.2. 1. Choice experiments

Choice experiments (CE) are a stated preference valuation technique where individual

preferences and values for bundles of ecosystem services are elicited with the use of

questionnaires (Louviere et al., 2000). In CE, respondents are presented with a series

of choice tasks in which they are asked to choose their preferred policy option from a

list of future options. Each option is described in terms of a bundle of attributes

describing the good presented at various levels. In this application, the CE attributes

are the six marine ecosystem services outlined in Table 2.

To analyse choices we use a random parameter logit (RPL) model. The RPL model

allows the coefficients associated with each attribute to vary randomly across

respondents therefore can model preference heterogeneity in the sample (Revelt and

Train, 1998).

Under a RPL specification, the utility a respondent i derives from an alternative j in

each choice situation t is given by:

ijtjtiijt eXU

, where X is a vector of observed attributes associated with each policy alternative

(i.e. the ecosystem services) and ijte is the random component of the utility that is

12

assumed to be independently and identically distributed (iid) and follow a Type 1

extreme value distribution. The probability that an individual i chooses alternative j in

a choice situation t is:

df

XX

kkti

jtiijt

expexp

Pr

,

with f (b) being the mixing distribution. A normal distribution is assigned to the

random parameters to allow respondents to have either positive or negative marginal

utilities for the levels of the attributes (Kataria 2009; Carlsson et al. 2003; Hanley et al

2005). To facilitate the estimation of the welfare measures, the monetary coefficient is

specified as constant, implying fixed marginal utility of money (Train 2003; Revelt

and Train 1998).

Assuming a linear in parameters utility function, the implicit price of each attribute

associated with changes to SVG’s MPAs is calculated as the ratio of the attribute’s

estimated coefficient to the estimated coefficient of the monetary attribute:

payment

attribute

bb

WTP

For a comprehensive discussion of the design, administration and analysis of CE, see

Louviere et al. (2000) and Hensher et al. (2005) .

3.2.2 Survey design and administration

The choice experiment was administered to a random sample of tourists and locals in

the two proposed marine parks. Data were collected through face-to-face interviews

13

by the authors in 2012 and 2013. The survey administration resulted in 710 people

completing the survey: 374 were interviewed at the South Coast marine park (of

which 204 were locals and 170 were tourists), and 336 at the Tobago Cays marine

park (of which 194 were locals and 142 were tourists).

The selection of attributes and their levels under different policy options was guided

by information collected during the stakeholder workshops (see Section 3.1) and

refined following feedback from local and tourist developmental focus groups. The

final design of the CE included six ecosystem service attributes (which are presented

in Table 4). Each attribute could take three levels: currently condition, an

improvement scenario (in which the levels of attributes were enhanced) and a

deterioration scenario (which referred to the future condition of the attributes in the

absence of any policy response). A cost attribute was also included in the design of

the CE, in which the payment vehicle was specified as a tourist tax (for tourists) or an

increase in the general taxation (for locals). The payment attribute could take six

levels ranging from US$0 to $160 per annum. Table 4 below presents the attributes

and their levels for the South Coast site, while Table 5 for the Tobago Cays site.

14

Table 4: Attributes and their levels in St Vincent South Coast

Ecosystem service Decline Current Improve

Fishing:

Number, size and desirability of

fish species in and around the

MPA

Some small fish only

Moderate fish numbers

incl.some large fish. Few

desirable fish.

Moderate numbers of

large fish including some

desirable fish.

Coastal protection: Level of risk

from coastal flooding

Minimal coastal protection

with a moderate risk of

flooding

Moderate coastal

protection with a some

risk of flooding

Good coastal protection

with minimal threat of

flooding

Human health:

Risk of gastoral intestinal

disorder/ skin rashes / ear

infections from poor sea water

quality.

High risk to human health

Moderate risk to human

health

Virtually no risk to human

health

Ecosystem resilience:

Contribution of genetic diversity

to long term resilience of marine

ecosystem

Low genetic diversity with

poor resilience

Low genetic diversity with

short term resilience

High genetic diversity

with medium term

resilience

Beach recreation:

Size and stability of beaches

Small, stable beach

Large, stable beach

Large, actively growing,

stable beaches

Diving / snorkelling:

Diving experience: condition of

coral and abundance of marine

life

Very poor diving

experience: coral badly

degraded and limited

marine life

Poor diving experience:

coral badly degraded and

moderate marine life

Good diving experience:

coral in good condition

and good marine life

15

Table 5: Attributes and their levels in Tobago Cays

Ecosystem service Decline Current Improve

Fishing:

Number, size and desirability of

fish species in and around the

MPA

Moderate numbers of large

fish including some

desirable fish.

Moderate numbers of large

fish including some

desirable fish.

A sustainable catch of

large, desirable fish

Coastal protection: Level of risk

from coastal flooding

Minimal coastal protection

with a moderate risk of

flooding

Good coastal protection

with minimal threat of

flooding

Good coastal protection

with minimal threat of

flooding

Human health:

Risk of gastoral intestinal

disorder/ skin rashes / ear

infections from poor sea water

quality.

Moderate risk to human

health

Minimal risk to human

health

No health problems

Ecosystem resilience:

Contribution of genetic diversity

to long term resilience of marine

ecosystem

Low genetic diversity with

short term resilience

High genetic diversity

with medium term

resilience

High genetic diversity

with long term resilience

Beach recreation:

Size and stability of beaches

Moderate, stable beach

Large, actively growing,

stable beaches

Large, actively growing,

stable beaches

Diving / snorkelling:

Diving experience: condition of

coral and abundance of marine

life

Moderate diving

Excellent diving

Excellent diving

16

experience: coral in

reasonable condition and

moderate marine life

experience: coral in

excellent condition and

abundant marine life

experience: coral in

excellent condition and

abundant marine life

The survey questionnaire comprised three parts. The first part introduced the survey

and asked respondents to provide information on the SVG islands that they had

visited and what activities they had undertaken on these islands. Respondents were

then presented with information on the current and proposed system of marine

protection in SVG, including information on the levels of protection provided by

future scenarios of SVG’s marine protected areas. Subsequently, the six ecosystem

service benefits derived from marine protection were introduced and respondents

were asked to indicate how important these benefits are to them. Finally, respondents

were asked to consider how their activities may affect the potential of the marine

parks to deliver the six ecosystem services. This aimed to make respondents aware of

the potential trade-offs that they personally may have to make in order to attain the

ecosystem service benefits. For example, they might have to change the way they fish

to ensure the long-term resilience of coral reefs.

The second part introduced the choice experiment tasks. Respondents were first

informed that the SVG government was planning to revise its marine protection

policies and that depending on how it targets resources, different combinations of

ecosystem service benefits may be attained. Respondents were also informed that in

order to attain these benefits, funds would have to be raised through increases in

taxation / tourist tax. Respondents were then presented with an example choice task

17

(which was described in detail) before being presented with a series of five choice

tasks.

provides an example of a typical choice task used in this study. In each choice task,

participants were presented with three choice options each described in terms of the

six ecosystem service attributes. A consistent ‘Baseline’ scenario was used across all

choice tasks. In the baseline scenario, all the ecosystem service attributes were

presented at the ‘continued decline’ scenario level, while the taxation was set at no

additional cost to participants (i.e. US$ 0). In the choice tasks, the levels of the

ecosystem services and taxation attributes used for the Option A and Option B

scenarios were determined using a main effects orthogonal, fractional factorial

experimental design. This design reduced the number of possible choice combinations

from 4344 choice tasks to 25 choice tasks. To reduce the cognitive burden, each

participant was presented with only five choice tasks.

Option A Option B Baseline(Continued decline)

Fishing:Number, size and desirability of fish species in and around the MPA

Some small fish only

Moderate fish numbers including

some large fish. Few desirable fish.

Some small fish only

Coastal protection: Level of risk from coastal flooding

Good coastal protection with minimal threat of flooding

Minimal coastal protection with a

moderate risk of flooding

Minimal coastal protection with a

moderate risk of floodingHuman health:Risk of gastoral intestinal disorder/ skin rashes / ear infections from poor sea water quality.

Moderate risk to human health

High risk to human health

High risk to human health

Ecosystem resilience:Contribution of genetic diversity to long term resilience of marine ecosystem

High genetic diversity with

medium term resilience

Low genetic diversity with poor

resilience

Low genetic diversity with poor

resilience

Beach recreation:Size and stability of beaches

Large, actively growing, stable

beaches

Large, actively growing, stable

beaches

Small, stable beach

18

Diving / snorkelling:Diving experience: condition of coral and abundance of marine life

Very poor diving experience:

coral badly degraded and limited marine life

Good diving experience: coral in good condition and good marine

life

Very poor diving experience:

coral badly degraded and limited marine life

Cost (EC$ per year) EC$ 20 EC$160 0

I prefer: Option A Option B Baseline q q q

How confident do you feel about this choice? (Not confident) 0 – 1 – 2 – 3 – 4 – 5 (Very confident)

Figure 1: Example of a typical choice task

A number of steps was undertaken to ensure that participants were fully informed and

made thoughtful choices. Specifically, respondents were reminded of the scope of the

marine protection policy, informed about potential substitute sites and provided with

a ‘cheap talk’ statement asking them to fully consider their budget constraints.

Once participants had completed all five choice tasks, they were asked to state their

motivations for their choices, which was used to identify genuine choices from protest

bids; the latter subsequently being removed from the analysis.

In the third part of the survey respondents were asked to provide information on their

socio-economic background. This information was collected to (i) test the

representativeness of the survey sample, and (ii) feed into a theoretical validity test.

4 Results

4.1 Characteristics of our survey respondents

Table 6 summarizes the socio-economic background of our respondents at the South

Coast and Tobago Cays study sites. In both study sites, tourists have higher household

income than locals. Also, in the South Coast locals interviewed are also older than the

tourists. Table 6 also reports the frequency of the activities undertaken by our survey

19

respondents. Across all sites, beach recreation was the most popular activity with

around half of respondents stating that this was their main activity. For locals this is

followed by snorkelling and diving, while 44% of Tobago Cays locals participated in

sailing / yachting. Snorkelling and diving along with wildlife watching were also

popular activities for tourists in Tobago Cays. Testing reveals significant differences

in the frequency of activities. The last column for each sample in Table 6 reports the

relevant p-values.

Table 6: Characteristics of the survey samplesSocio-economics South Coast Tobago Cays

Locals Tourists Signif Diff(P-value)

Locals Tourists Signif Diff(P-value)

Age_Below 30 0.54(0.50)

0.61(0.49)

0.18 0.37(0.48)

0.49(0.50)

0.03

Age_30-60 0.40(0.49)

0.35(0.48)

0.30 0.56(0.50)

0.48(0.50)

0.08

Age_Above 60 0.06(0.24)

0.03(0.17

0.18 0.07(0.25)

0.03(0.18)

0.16

Gender (0=female,1=male)

0.44(0.48)

0.40(0.49)

0.43 0.45(0.50)

0.41(0.49)

0.49

Mean Annual Household income

$16,650 $45,000 0.00 $16,650 $35,000 0.00

Income > $30.000 0.26(0.48)

0.65(0.50)

0.00 0.36(0.48)

0.53(0.50)

0.00

Hiking 5.9 % 11.2% 0.00 5.2 % 5.9 % 0.23

Waterfalls 6.8 % 11.8% 0.00 3.6 % 5.3 % 0.004

Wildlife watching 3.5 % 6.6 % 0.00 13.4 % 13.2 % 0.78

Fishing 3.5 % 6.6 % 0.00 3.1 % 10.3 % 0.00

Beach Activities 52.5 % 42.3 % 0.00 42.3 % 43.3 % 0.69

Snorkelling 17.6% 4.1 % 0.00 32 % 12.5% 0.00

Diving 9.3% 6.1 % 0.00 11.3 % 11.1 % 0.80

Yachting/sailing 10.2% 3.5% 0.00 43.8 % 8.6 % 0.02No of observations

204 170 194 142

20

4.2 Analysis of the choice experiment utility coefficients

The RPL models for the locals and tourists are reported in Table 7 for both the South

Coast and Tobago Cays marine parks. In the table, we report the estimated utility

coefficients (in the top half of the table) and the coefficient’s standard deviation (in

the bottom half of the table). The models were estimated in Nlogit using simulated

maximum likelihood with 100 Halton draws. We used effects coding in the model to

allow us to discriminate among Willingness to Pay to avoid a deterioration of the

current condition of ecosystem services (‘decline’) and Willingness to Pay for

improvements over the status quo (‘improve’). The use of effects coding has been

advocated in the stated preference literature (Bech and Gyrd-Hansen, 2005).

Table 7: Results of the RPL models for ecosystem services derived from SVG’s MPAAttribute South Coast Tobago Cays

Locals Tourists Locals TouristsFish (decline) 0.05

(0.125)-0.60***(0.160)

-0.22*(0.127)

-0.41***(0.132)

Fish (improve) -0.01(0.109)

0.26*(0.134)

0.02(0.117)

0.03(0.130)

Coastal protection (decline)

-0.44***(0.098)

-0.45***(0.13)

-0.54***(0.116)

-0.54***(0.121)

Coastal protection (improve)

0.38***(0.129)

0.30*(0.16)

0.35***(0.130)

0.22(0.135)

Health (decline) -1.21***(0.142)

-1.35***(0.22)

-0.72***(0.124)

-0.36***(0.138)

Health (improve) 1.15***(0.129)

1.26***(0.18)

0.73***(0.127)

0.39***(0.114)

Ecosystem (decline)

-0.44***(0.131)

-0.63***(0.160)

-0.84***(0.153)

-0.61***(0.146)

Ecosystem (improve)

0.52***(0.113)

0.78***(0.084)

0.60***(0.127)

0.58***(0.134)

Beach (decline) -0.25(0.234)

-0.07(0.28)

-0.27(0.246)

0.17(0.254)

Beach (improve) -0.02(0.227)

0.15(0.28)

0.05(0.232)

0.37(0.251)

21

Diving (decline) -0.067(0.123)

-0.20(0.158)

-0.252**(0.118)

-0.25*(0.145)

Diving (improve) 0.26*(0.135)

0.28*(0.152)

0.28**(0.132)

0.12(0.133)

Tax -0.04***(0.006)

-0.03***(0.003)

-0.07***(0.008)

-0.02***(0.003)

Alternative Specific Constant

-3.41***(0.373)

-3.21***(0.47)

-3.85***(0.408)

-1.69***(0.324)

Log likelihood -691.3325 -637.29 -726.00 -633.28

No of observations 1020 850 970 710

Parameters standard deviation

Fish (decline) 0.62***(0.181)

0.78***(0.256)

0.34(0.228)

0.08(0.223)

Fish (improve) 0.53**(0.21)

0.69***(0.238)

0.66***(0.207)

0.72***(0.179)

Coastal protection (decline)

0.08(0.218)

0.46*(0.246)

0.42*(0.243)

0.37*(0.223)

Coastal protection (improve)

0.15(0 .189)

0.45(0.321)

0.25(0.310)

0.47**(0.222)

Health (decline) 0.04(0.310)

0.44(0.485)

0.06(0.261)

0.51**(0.253)

Health (improve) 0.62***(0.186)

0.58**(0.258)

0.13(0.438)

0.17(0.385)

Ecosystem (decline)

0.05(0.221)

0.12(0.349)

0.50(0.354)

0.39(0.274)

Ecosystem (improve)

0.25(0.196)

0.72***(0.277)

0.80***(0.191)

0.47*(0.264)

Beach (decline) 0.93***(0.234)

0.44(0.485)

1.23***(0.209)

1.20***(0.190)

Beach (improve) 0.93***(0.234)

1.12***(0.240)

1.23***(0.209)

1.20***(0.190)

Diving (decline) 0.07(0.174)

0.87***(0.276)

0.05(0.275)

0.58**(0.265)

Diving (improve) 0.65***(0.243)

0.34(0.342)

0.42(0.304)

0.29(0.238)

Notes: Figures in cell are model coefficients, with standard errors in parenthesis.***Indicates significance at 1%, **Indicates significance at 5%,*Indicates significance at 10%.

Across all the choice models reported in Table 7 the majority of the estimated

coefficients are highly significant indicating that the selected ecosystem service

attributes are indeed important determinants of individual choice. The key exception

22

is beach recreation, which was not significant in any of the models. The model

coefficients also have the expected signs with negative coefficients for the

deterioration scenarios, and positive coefficients for improvement scenarios.

Respondents are thus more likely to select alternatives associated with improvements

in Ecosystem resilience, Human health, Coastal protection, Fishing and diving

opportunities, while less likely to choose alternatives were Ecosystem resilience,

Human health, Coastal protection and Fishing are in decline. Conforming to economic

theory, the coefficient for the Tax was significant and negative, suggesting that

respondents are less likely to select alternatives associated with a higher tax. The

negative and significant coefficient on the alternative specific constant indicates that

respondents desire to move away from the status quo and in principle favour the MPA

policy.

Results also point to significant differences between the preferences of locals and

tourists. In the South Coast, tourists were more concerned about changes to Fishing

than locals, while in the Tobago Cays only the locals were willing to pay to improve

coastal protection and diving. Differences between the two study sites are also

apparent. For example, locals at the Tobago Cays appeared concerned about declines

in Fishing and Diving, whereas locals at South Coast were not willing to pay to hedge

against a deterioration scenario. Equally, tourists at the South Coast held positive

values for improvements in Fishing, Coastal protection and Diving, while

improvements in these services did not significantly contribute to the utility of their

counterparts in Tobago Cays.

23

Finally, across all respondent groups, the standard deviations of the coefficients in the

random utility models are statistical significant indicating that marginal utilities do

vary in the population and are thus correctly treated as random.

4.3 Willingness to Pay for SVG’s Marine Protected Areas

Table 8 reports the marginal WTP values for each of the marine ecosystem service for

locals and tourists in South Coast and Tobago Cays marine parks respectively. WTP

was estimated using the Wald method. Standard errors and 95% confidence intervals

are estimated using the bootstrap method proposed by Krinsky and Robb (1986).

Table 8: Willingness to pay estimates (US$ per household per annum) for SVG’s marine parks.

EcosystemService

South Coast Tobago Cays

Locals Tourists Signif Dif

Poe test(p-value)

Locals Tourists Signif DifPoe test

(p-value)

Fish (decline) $1.28[-4.15, 8.03]

-$20.56[-31.23 -10.39] 0.001 -$3.22

[-6.58, 0.48]-$18.24

[-29.47, -6.98] 0.007

Fish (improve)

-$0.24[-5.57, 4.95]

$8.80[0.18, 18.63] 0.046 $0.23

[-3.21, 3.54]$1.49

[-10.52, 12.72] 0.426

Coastal protection (decline)

-$10.63[-15.73, -6.41]

-$15.31[-23.54, -7.32] 0.167 -$7.81

[-11.02, -4.78]-$24.06

[-35.60, -14.28] 0.002

Coastal protection (improve)

$9.01[3.18, 14.78]

$10.26[-0.20, 20.38] 0.422 $5.05

[1.42, 8.36]$9.78

[-1.70, 21.25] 0.223

Health (decline)

-$29.00[-40.24, -

21.92]

-$46.14[-50.39, 35.06] 0.018 -$10.45

[-14.44, -7.13]-$16.05

[-28.70, -4.76] 0.14

Health (improve)

$27.50[21.18, 37.64]

$42.95[34.45, 53.71] 0.013 $10.56

[7.21, 14.16]$17.61

[7.95, 28.56] 0.083

Ecosystem (decline)

-$10.58[-16.42, -4.73]

-$21.57[-30.87, -12.12] 0.027 -$12.17

[-15.97, -8.43]-$27.21

[-40.87, -15.05] 0.002

Ecosystem (improve)

$12.37[7.71, 18.16]

$11.73[3.57, 21.34] 0.446 $8.67

[5.48, 12.25]$25.98

[15.27, 39.18] 0.002

Beach (decline)

-$6.00[-18.99, 4.86]

-$2.58[-22.47, 16.20] 0.375 -$3.90

[-11.14, 3.18]$7.61

[-15.45, 30.20] 0.168

Beach (improve)

-$0.55[-12.39, 10.01]

$5.28[-13.88, 23.31] 0.298 $0.76

[-6.20, 7.23]$16.68

[-5.16, 38.81] 0.091

Diving (decline)

-$1.59[-8.18, 4.05]

-$6.89[-18.57, 3.75] 0.022 -$3.66

[-7.49, -0.18]-$11.16

[-25.20, 1.71] 0.134

24

Diving (improve)

$6.32[0.22, 13.14]

$9.46[-0.23, 20.22] 0.307 $4.09

[0.53, 8.00]$5.53

[-5.88, 18.20] 0.413

Decline in all services -$56.52 -$113.05 -$41.21 -$89.11

Improvement to all services $54.41 $88.48 $29.36 $77.07

AGGREGATE VALUATION OF SVG’s MPA POLICYLower bound Upper bound

($’000) ($’000)

Decline in all services -$661 -$1,823 -$4,638 -$7,152 -$77 -$1,330 -$2,08 -$3,529

Improvement to all services

$636 $1,755 $3,630 $5,597 $55 $947 $1,801 $3,052

Note: Values shown in bold relate to those ecosystem services that were significant in the choice models are therefore are robust value estimates. The coefficients of those services not in bold were insignificant in the choice models, and therefore these WTP estimates are not robust estimates. 95% confidence intervals are reported in parentheses.

WTP estimates suggest that both locals and tourists were willing to pay more to hedge

against future deterioration in ecosystem services provision than for improving the

delivery of these services. This finding is not uncommon in valuation studies and

reflects the fact the people tend to value what they already have more than increasing

current provision (Christie and Rayment, 2012). Mirroring the result from the utility

coefficients’ estimation WTP values suggest significant differences between tourists

and locals. To examine whether the values for the two groups are similar we perform

the complete combinatorial test proposed by Poe et al. (2005)1. The null of equal

WTP can be rejected for the majority of the attributes in both sites, suggesting that

tourists generally had higher values than locals. Much of this can be explained by that

fact that tourists, on average, have higher annual incomes (Table 6). Finally our

results point to significant differences in values between sites. WTP estimates were

generally higher in the South Coast compared to the Tobago Cays. This finding was

initially surprising since the reefs and sea grass beds at the Tobago Cays are in much

better condition than those in the South Coast. However, further discussions with 1 This test calculates every possible difference between the two empirical distributions generated by the Krinsky-Robb procedure and computes the proportion of negative values in the distribution of differences to approximate a one-sided p-value for the null of equal WTP values between the two groups.

25

respondents and stakeholders suggest that respondents did not consider the Tobago

Cays to be under significant threat, whereas in the South Coast there was clear

evidence of damage. Thus, respondents were more aware of the impacts of human

activities in the South Coast, and therefore had higher values for policies to protect

them.

In terms of the values for specific ecosystem services, preventing a deterioration in

Human health was the attribute that was most highly valued in the South Coast among

both locals (-$29.00 per household per year) and tourists (-$46.14 per household per

year) followed by improvements to Human health (Locals = $27.50 per household per

year; Tourists = $42.95 per household per year)2. In Tobago Cays, both locals and

tourists were willing to pay? most to hedge against future deteriorations in ecosystem

resilience (Locals = -$12.17 per household per year; Tourists = $25.98 per household

per year) followed by improvements to Human health for locals ($10.56 per

household per year) and improvements in Ecosystem resilience by tourists ($25.98 per

household per year).

The second panel of Table 8 provides aggregate value estimates for the changes to the

provision of ecosystem services at the two marine parks under the improved and

decline scenarios. The aggregate values were estimated by multiplying the mean

household values (Table 8) with a lower and upper bound estimate of the number of

local households and tourists impacted by the proposed MPA designations. In terms

of locals, the lower bound estimate for the South Coast was the number of households

2 To examine whether low risk to human health is a dominant attribute we calculate the share of

choices that are associated with low health risk. These are 64% of the choices in South Coast.

However, only 28.6% of the sample is always selecting the low health risk option which does

not provide evidence of health risk being a dominant attribute.

26

in the three districts on the South Coast (Kingstown, Calliaqua and South Grenadines

= 11,687 households), while the lower bound estimate for the Tobago Cays was the

number of households in Mustique, Canouan and Union islands (1,872 households).

For both MPA sites, the upper bound of locals impacted was the number of all

households in SVG (32,262 households). For tourists, the lower bound estimate of the

number of tourists visiting the South Coast was approximated by the number of

visitors staying in tourist accommodation in the South Coast (41,028), while the upper

bound was the total number of tourists visiting St Vincent (63,263). For the Tobago

Cays, the lower bound estimate of tourists was estimated using the number of Marine

Park permits sold to individuals (27,355), while the upper bound was based on

number of visitors arriving in the islands of Mustique, Canouan and Union Island

(39,602).

The aggregate local value of ecosystem services that would be delivered through

improvements is estimated to range between US$0.63m and US$1.75m at the South

Coast and between US$0.05m and US$0.94m at the Tobago Cays. The locals’ values

to protect against a deterioration in ecosystem services ranges between -US$0.66m

and -US$1.82m at the South Coast and between -US$0.07m and -US$1.33m at the

Tobago Cays (Table 8). The equivalent tourists’ values for improvements is estimated

between US$3.63m and US$5.59m per year at the South Coast, and between

US$1.80m to US$3.05m at the Tobago Cays, while the benefits from preventing a

deterioration in services is found to range between –US$4.63m and –US$7.15m at the

South Coast and between –US$2.08m and –US$3.53m at the Tobago Cays (Table 8).

Thus, overall, tourists attain higher aggregate benefits from the MPA policies than

locals. These higher values are based on (i) higher individual WTP values (which are

27

to a large extent a reflection of the higher incomes of tourists compared to locals), and

(ii) the fact that there are more tourists using the MPAs than locals.

4.4 Benefits from different policy scenarios

WTP values from this study can be used to estimate the economic benefits that may

be attained from the various policy interventions aiming to achieve the protection and

enhancement of ecosystem services in MPAs. For this analysis we combine the

information on the links between policy interventions and the provision of ecosystem

services (Table 2) with the values of ecosystems services at the case study sites (Table

8). To combine the information, we multiply the aggregate WTP values by a

weighting factor of 1 if there are high levels of linkages (i.e. ‘+++’ in Table 2), by

0.66 if moderate levels of linkages (i.e. ‘++’) and by 0.33 if low level of linkages

(‘+’). To avoid double counting of benefits we also make an adjustment to the

weighting factors so that the contribution of the different policy interventions equals 1

across the individual services. To illustrate, Fishing was identified in Table 2 to have

moderate links with the ‘stopping of sewage’, ‘stopping overfishing’, and ‘stopping

land based pollution’ policy interventions, and low level links with introducing ‘No

take’ zones. Since moderate links were weighted by 0.66 and low level links by 0.33,

this gave a total weighting score of 2.32 across all policy interventions affecting the

provision of fish. To avoid problems of double counting, all weightings for Fish

services were divided by 2.32 so that the weighting coefficient used for Fish services

equalled 1 across all interventions. Thus, the weighting used in the analysis for the

moderately linked interventions was 0.286, while that for the low level linked

interventions was 0.143. The output from our analysis of the ecosystem service

benefits attained from the policy interventions is reported in Table 9. Based on this

28

analysis, it is predicted that stopping land based pollution in the South Coast would

generate the greatest ecosystem services benefits (US$1.43m / yr), followed by

controlling sewage (US$1.37m / yr), stopping sand mining / extraction of coral

US$0.51m / yr) and introducing ‘No take’ zones (US$0.47m / yr). The highest values

for the Tobago Cays were found for controlling land based pollution (US$0.27m / yr)

and for stopping sewage (US$0.23m / yr) and introducing ‘No take’ zones sewage

(US$0.17m / yr). These figures should be considered as lower bound estimates since

(i) they are based on only those ecosystem services that were valued by respondents

and (ii) the values are aggregated using the lower bound estimate the affected

population. However, they do provide useful information on the potential benefits

associated with alternative policy interventions that may be used to enhance marine

protection at different locations in SVG, and these values may then be directly

compared to the costs of controlling these damaging activities.

Table 9: Economic benefits (US$’000 / yr) of policy interventions to enhance MPAs.Stop Sewage

(yachts, houses and

hotels)

Stop Overfishin

g / bad fishing

practices

Stop Land Based Pollution (mainly agricultural, such

as pesticides, eutrophication

and sedimentation

Stop sandmining / extraction

of coral

Introduce ‘No take

zones’ that would ban fishing and anchoring

St V

ince

nt S

outh

Coa

st

Fishing 103 103 103 0 52Coastal protection 66 66 66 199 131Water quality / human health 1,042 0 1042 0 0

Species diversity, ecosystem resilience and genetic pool

62 189 124 62 189

Beach recreation 0 0 0 0 0Diving / snorkelling 102 155 102 0 102TOTAL VALUE (US$’000/ yr) 1,375 512 1,437 261 473

Toba

go C

ays

Fishing 0 0 0 0 0Coastal protection 22 22 22 65 43Water quality / human health 165 0 165 0 0

Species diversity, ecosystem resilience and genetic pool

40 122 81 40 122

Beach recreation 0 0 0 0 0Diving / snorkelling 11 16 11 0 11

29

TOTAL VALUE (US$’000/ yr) 237 160 277 106 176

5 Discussion and policy implications

This study aimed to provide an assessment of the economic value of ecosystem

services provided by Marine Protected Areas in SVG. To meet this aim, we

implemented a choice experiment to estimate the values that both locals and tourists

have for a range ecosystem services delivered at two marine parks in SVG (the

proposed St Vincent South Coast marine park and the Tobago Cays Marine Park).

Our results demonstrate that both locals and tourists have high values to protect

against a deterioration in current levels of service provision (i.e. the ‘decline’

scenario) but also significantly value improvements in the level of service provision

(the ‘improved’ scenario) (Table 8).

Our analysis further points to differences in the values between the two sites. At the

South Coast, the ecosystem services that were most highly valued included Human

health, Ecosystem resilience and Fishing, while at the Tobago Cays the highest valued

services were Ecosystem resilience and Coastal protection (Table 8). Although we did

not collect data to identify the reason why preferences may differ at the two marine

parks, further discussions with respondents and feedback from local and tourist

developmental focus groups pointed to potential contributing factors which include:

(i) that the condition of the ecosystems were very different between the two MPAs

(the South Coast is degraded and the Tobago Cays is pristine), and (ii) the level of

external pressures on the MPAs is different (there are greater human-induced

pressures at the South Coast). Based on these observations, we conclude that one

30

cannot directly infer that there is a single bundle of ecosystem services that is

important and valued across all MPAs and therefore policies need to be tailored to

individual sites. This finding also has implication for future applications of the value

transfer methods, which allows values from a study site to be transferred to a policy

site under consideration. Our results suggest that values are not directly transferable

and attention should be given to differences in the current condition of, and pressures

on, those sites.

We also find that tourists have significantly higher values for marine ecosystem

services than locals (Table 8). Furthermore, our analysis also points to significant

differences in preferences between tourists and locals in both sites. For example, at

the South Coast, fishing was significant for tourists, but not for locals. The fact that

different groups of people have different preferences and values may have policy

implications in terms of who benefits from alternative policy scenarios that deliver

different bundles of services. Furthermore, our research findings may have

implications in terms of how funds for protecting MPAs might most equitably be

raised. Our research suggests that there is a case for raising funds from tourists,

through e.g. a tourist tax, to fund marine conservation programmes in SVG.

6. Conclusions

This study has generated a wealth of evidence on the economic values delivered by

the protection and enhancement of MPAs in SVG. There are clearly some

uncertainties and caveats with this data, many of which we have highlighted above. In

presenting the results we have attempted to account for these uncertainties by

presenting the economic values as either lower bound conservative estimates or where

possible as ranges. We argue that the lower bound estimates are robust and therefore

31

can be used in policy analysis with confidence, while the upper bound values might

best be considered as representing the potential value of services delivered.

In this paper, we only assess the benefits derived from alternative policy interventions

that may be used to deliver the ambitions of SVG’s new MPA policy. The GoSVG

now needs to compare these benefit estimates with the costs of expanding marine

protection, to determine whether future expansions of the marine parks can be

justified within a cost benefit framework. It is our hope that our value evidence will

contribute towards the continued protection of SVG’s marine resources, whilst also

providing opportunities to reduce poverty and enhance well-being.

Finally, although the research reported here directly links to the evaluation of the

proposed expansion of SVG’s MPAs, the way in which the research was set up should

allow the findings to be readily transferred to MPA appraisals in other parts of the

Caribbean and possibly further afield. Specifically, in the choice experiment, we

presented the ecosystem service attributes using a continuous condition score scale (1

= worst condition and 5 = good condition). Thus, our analysis of the choice

experiment allows us to identify the value of a unit change in the condition of the

ecosystem services, which in turn can be readily transferred to other case study sites.

7. References

Allsopp, M., Page, R., Johnston, P., Santillo, D., 2009. State of the World’s Oceans.

Springer, Dordrecht.

Alpizar, F., Carlsson, F., Martinsson, P., 2003. Using Choice ExPeriments for Non-

Market Valuation. Econ Issues 8: 83–110.

32

Bech, M., and Gyrd-Hansen, D., 2005. Effects coding in discrete choice experiments.

Health Economics 14, 1079–1083.

Beharry-Borg, N., Scarpa, R., 2010. Valuing quality changes in Caribbean coastal

waters for heterogeneous beach visitors. Ecol Econ 69, 1124-1139.

Butchart, S.H.M., Walpole, M., Collen, B., van Strien, A., Scharlemann, J.P.W.,

Almond, R.E.A., Baillie, J.E.M., Bomhard, B., Brown, C., Bruno, J., Carpenter, K.E.,

Carr, G.M., Chanson, J., Chenery, A.M., Csirke, J., Davidson, N.C., Dentener, F.,

Foster, M., Galli, A., Galloway, J.N., Genovesi, P., Gregory, R.D., Hockings, M.,

Kapos, V., Lamarque, J.-F., Leverington, F., Loh, J., McGeoch, M.A., McRae, L.,

Minasyan, A., Morcillo, M.H., Oldfield, T.E.E., Pauly, D., Quader, S., Revenga, C.,

Sauer, J.R., Skolnik, B., Spear, D., Stanwell-Smith, D., Stuart, S.N., Symes, A.,

Tierney, M., Tyrrell, T.D., Vie, J.-C., Watson, R., 2010. Global Biodiversity:

Indicators of Recent Declines. Science 328, 1164-1168.

Carlsson F., Frykblom, P., Liljenstolpe,C., 2003. Valuing wetland attributes: an

application of choice experiments. Ecol Econ 47, 95-103.

Cesar, H., Burke, L., Pet-Soede, L., 2003. The economics of worldwide coral reef

degradation. Cesar Environmental Economics Consulting, Arnhem, The Netherlands.

Christie, M., Rayment, M., 2012. An economic assessment of the ecosystem service

benefits derived from the SSSI biodiversity conservation policy in England and

Wales. Ecosystem Services 1, 70-84.

Cinner, J.E., McClanahan, T.R., Graham, N.A.J., Daw, T.M., Maina, J., Stead, S.M.,

Wamukota, A., Brown, K., Bodin, O., 2012. Vulnerability of coastal communities to

key impacts of climate change on coral reef fisheries. Global Environ Chang 22, 12-

20.

Costanza, R., d'Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg,

K., Naeem, S., O'Neill, R.V., Paruelo, J., Raskin, R.G., Sutton, P., van den Belt, M.,

1997. The value of the world's ecosystem services and natural capital. Nature 387,

253-260.

33

Daily, g.C., 1997. Nature's Services. Societal Dependence on Natural Ecosystems.

Island Press, Washington, D.C., p. 392.

Haines-Young, R., Potschin, M., 2008. England's Terrestrial Ecosystem Services and

the Rationale for an Ecosystem Approach: Full Technical Report. DEFRA, London.

Hanley, N., Wiktor A., Wright, R.E., 2005. Price vector effects in choice experiments:

an empirical test. Resourc Energ Econ 27 (3), 227-234.

Hanemann, W.M., 1994. Valuing the Environment through Contingent Valuation.

The Journal of Economic Perspektives, 19-44.

Hensher, D.A., Rose, J.M., Greene, W.H., 2005. Applied Choice Analysis. A Primer.

Cambridge University Press, Cambridge.

Jackson, I., 2004. Master Plan: System of Protected Areas and Heritage Sites, St.

Vincent and the Grenadines. Ivor Jackson and Associates.

Jameson, S.C., McManus, J.W., Spalding, M.D., 1995. State of the Reefs: Regional

and Global Perspectives. ICRI., US Department of State., Washington D.C.

Kataria M., 2009. Willingness to pay for environmental improvements in hydropower

regulated rivers. Energ Econ 31:69–76.

Krinsky, I., Robb, A.L., 1986. ON APPROXIMATING THE STATISTICAL

PROPERTIES OF ELASTICITIES. Rev Econ Stat 68, 715-719.

Louviere, J.J., Hensher, D.A., Swait, J.D., 2000. Stated Choice Methods. Analysis and

Application. Cambridge University Press, Cambridge.

MA, 2005. Millennium Ecosystem Assessment. Ecosystems and Human Well-being.

Island Press, Washington D.C.

34

MEA, 2005. Millennium Ecosystem Assessment. Ecosystems and Human Well-being.

Island Press, Washington D.C.

Moberg, F., Folke, C., 1999. Ecological goods and services of coral reef ecosystems.

Ecol Econ 29, 215-233.

NPRBA, 2009. SVG National Parks and Protected Areas System Plan 2010 - 2014.

National Parks Rivers and Beaches Authority, SVG.

Poe, G.L., Giraud, K.L., Loomis, J.B., 2005. Computational methods for measuring

the difference of empirical distributions. American Journal of Agricultural Economics

87, 353-365.

Revelt, D., Train, K., 1998. Mixed logit with repeated choices: Households' choices of

appliance efficiency level. Rev Econ Stats 80, 647-657.

Sachs, J.D., Baillie, J.E.M., Sutherland, W.J., Armsworth, P.R., Ash, N., Beddington,

J., Blackburn, T.M., Collen, B., Gardiner, B., Gaston, K.J., Godfray, H.C.J., Green,

R.E., Harvey, P.H., House, B., Knapp, S., Kumpel, N.F., Macdonald, D.W., Mace,

G.M., Mallet, J., Matthews, A., May, R.M., Petchey, O., Purvis, A., Roe, D., Safi, K.,

Turner, K., Walpole, M., Watson, R., Jones, K.E., 2009. Biodiversity Conservation

and the Millennium Development Goals. Science 325, 1502-1503.

Schuhmann, P., 2012. The Valuation of marine ecosystem goods and services in the

Wider Caribbean Region., CERMES Technical Report no 61. University of the West

Indies - Cave Hill, Barbados.

Schumann, P.W., Mahon, R., 2014. The Valuation of marine Ecosystem Goods and

Services in the Caribbean: A literature review and framework for future valuation

efforts. Ecosystem Services, Forthcoming.

Secretariat of the Convention on Biological Diversity, 2000. Sustaining Life on Earth.

How the Convention on Biological Diversity promotes nature and human well-being.

Secretariat of the Convention on Biological Diversity, Montreal.

35

Secretariat of the Convention on Biological Diversity, 2010. Global Biodiversity

Outlook 3., Montreal.

Souter, D.W., Linden, O., 2000. The health and future of coral reef systems. Ocean &

Coastal Management 43, 657-688.

Spalding, M.D., Ravilious, C., Green, E.P., 2001. United Nations Environment

Programme, World Conservation Monitoring Centre. World Atlas of Coral Reefs.

Univeristy of California Press, Berkeley.

TEEB, 2010. The Economics of Ecosystems and Biodiversity: Ecological and

Economic Foundations. Earthscan, London.

TEEB, 2011a. The Economics of Ecosystems and Biodiversity in Local and Regional

Policy and Management. Earthscan, London.

TEEB, 2011b. The Economics of Ecosystems and Biodiversity in National and

International Policy. EarthScan, London.

van Beukering, P., Brander, L.M., Tompkins, E., McKenzie, E., 2007. Valuing the

Environment in Small Islands: An environmental economics toolkit. Joint Nature

Conservation Committee, Peterborough.

Train, K.E., 2003. Discrete Choice Methods with Simulation. Cambridge University

Press, Cambridge, UK.

Waite, R., Burke, L., Gray, E., van Beukering, P., Brander, L., McKenzie, E.,

Pendleton, L., Schuhmann, P., Tompkins, E., 2014. Coastal Capital: Ecosystem

Valuation for Decsion Making in the Caribbean. World Resources Institute,

Wahsington DC.

36

Wilkinson, C.R., 2004. Status of the coral reefs of the world: Volumes 1 and 2.

Australian Institute for Marine Sciences, Townsville, Australia.

Wilkinson, C.R., 2008. Status of the Coral Reefs of the World. Global Coral Reef

Monitoring Network and Reef and Rainforest Research Centre, Townsville, Australia.

Wilkinson, C.R., Buddemeier, R.W., 1994. Global climate change and coral reefs:

implications for people and reefs. IUCN, Gland, Switzerland.

World Bank, 2012.

37