Content uploaded by Jesko Hirschfeld
Author content
All content in this area was uploaded by Jesko Hirschfeld on Jan 31, 2017
Content may be subject to copyright.
Review article
Economic valuation of Baltic marine ecosystem services: blind
spots and limited consistency
Julian Sagebiel1, 2*, Carmen Schwartz1, Mounaim Rhozyel1, Sandra Rajmis1‡, and Jesko Hirschfeld1
1
Institute for Ecological Economy Research, Potsdamer Str. 105, 10785 Berlin, Germany
2
Department of Agricultural Economics, Humboldt-Universita¨t zu Berlin, Unter den Linden 6, 10099 Berlin, Germany
*Corresponding author. tel: +49 30 88 45 94-29; fax: +49 30 8825439; e-mail: julian.sagebiel@ioew.de
‡
Present address: Julius Ku¨hn-Institute, Federal Research Centre for Cultivated Plants JKI, Stahnsdorfer Damm 81, 14532 Kleinmachnow, Germany.
Sagebiel, J., Schwartz, C., Rhozyel, M., Rajmis, S., and Hirschfeld, J. Economic valuation of Baltic marine ecosystem services: blind
spots and limited consistency. – ICES Journal of Marine Science, 73: 991 –1003.
Received 24 May 2015; revised 8 December 2015; accepted 14 December 2015; advance access publication 26 January 2016.
Economic valuation of marine ecosystem services in the Baltic Sea region has gained importance, as policy-makers are recognizing their decline and
focusing on achieving good environmental status there in terms of, for example, reduced eutrophication. Parallel with this development, several
initiatives have been launched, leading to a large number of economic valuation studies. However, current research indicates that neither a
common approach to classifying ecosystem services nor a widely accepted methodological framework for assessing their economic value exist
yet. This paper seeks to shed light on the current stateof the economic valuation of ecosystem services provided by the Baltic Sea through reviewing
all currently available empirical studies on the topic. The results indicate that only afew ecosystem services, including recreation and reduction of
eutrophication, have been extensively monetarily valued, and still lack cross-study methodological consistency, while many other marine ecosys-
tem services have rarely or never been valued with economic methods. Thepaper concludes that existing economic valuation studies provide only
limited practical guidance for policy-makers intending to improve the environmentalstatus of the Baltic Sea. There is a need for more widely shared
agreement on the systematic nature of marine and coastalecosystem services and especially on a coherent methodological framework for assessing
their economic value.
Keywords: Baltic Sea, marine planning, meta-study, nutrient abatement costs, stated preference method.
Introduction
The Baltic Sea is one of the largest semi-closed brackish watersin the
world. It is almost entirely surrounded by land, with limited water
exchange taking place, making it sensitive to environmental
influences, especially from human activities (HELCOM, 2010).
The need for action to maintain and, in some areas, to improve
the environmental status of the Baltic Sea is widely recognized. In
1974, the Baltic Marine Environmental Protection Commission
(HELCOM) was founded by the riparian states surrounding the
sea with the aim of protecting its coastal and maritime ecosystem
services, particularly through reduction of eutrophication. In
2007, HELCOM released the Baltic Sea Action Plan, a coordinated
effort to re-establish good environmental status by 2021.
Analogously, the European Union published the Marine Strategy
Framework Directive in 2008 (European Union, 2008), with the aim
of achieving good environmental status for all European Seas by
2020.Both of thesepolicydocumentsexplicitlymentionthat countries
should use cost–benefit analysis and the ecosystem services approach
to identify theirparticular needfor actionand appropriatemeasuresto
improve Baltic Sea ecosystem services (Matzdorf and Meyer, 2014).
HELCOM (2010) developed a classification of marine ecosystem
services for the Baltic Sea, based on the Millennium Assessment
Report (Millennium Ecosystem Assessment, 2005). However, several
authors have proposed alternative classifications (de Groot et al.,
2002;Hein et al., 2006;Boyd and Banzhaf, 2007;Wallace, 2007;
Fisher and Turner, 2008;Fisher et al., 2009), with some criticizing
the lack of a common method for valuing ecosystem services (Fisher
et al., 2009;Clifton et al., 2014;Hirschfeld and Sagebiel, 2014).
Recent studies have analysed the current state of the economic
valuation literature and pointed out gaps in cost– benefit analyses of
ecosystem services. Remoundou et al. (2009) reviewed studies valuing
marine ecosystem services around the Black and Mediterranean seas,
#International Council for the Exploration of the Sea 2016. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
ICES Journal of
Marine Science
ICES Journal of Marine Science (2016), 73(4), 991– 1003. doi:10.1093/icesjms/fsv264
also assessing their usefulness and the need for further research.
Bo
¨rger et al. (2014) compared the use of valuation studies in
marine spatial planning in the UK and United States, claiming
that the United States is clearly taking a pioneering role, especially
when it comes to using valuation studies in policy-making.
Clifton et al. (2014) reviewed available valuation methods for
marine ecosystem services and identified not only high variability
but also lack of consistency in outcomes. Bertram and Rehdanz
(2013) reviewed the valuation literature on Europe’s seas and dis-
cussed current challenges to cost-benefit analysis in the context of
the Marine Strategy Framework Directive. In the Baltic Sea
context, Bertram et al. (2014) pointed out gaps in valuation of
marine ecosystem services and found there are few studies valuing
their benefits. In a recent book on the economics of ecosystem ser-
vices, Nunes et al. (2014) summarized the call of many researchers
for a unified classification and valuation framework. In sum, all of
these studies have identified limitations in existing concepts of eco-
system service classification and valuation for cost-benefit analysis.
This paper aims to complement previous discussions on valuing
marine ecosystem services by providing facts and figures regarding
the current state of valuation methods and their application to the
Baltic Sea region. All available studies that have valued at least one
ecosystem service provided by the Baltic Sea between 1995 and
2015 are systematically analyzed with regard to methods applied
and ecosystem services valued. In addition to providing a general
synopsis, the two most widely used valuation methods—stated pref-
erence and abatement costs—will be examined in detail. Through
our analysis, we have identified eight blind spots in the economic
valuation of Baltic Sea ecosystem services that are related to coun-
tries, ecosystem services, and ecosystem-service interactions that
have been neglected in the literature. Consequently, we discuss
how these issues can be addressed by policy-makers and researchers
while also pointing out the limitations of previously applied
methods and identifying a need for more methodical work.
Regardless of the ecosystem service valued, we found a need to
develop a coordinated valuation strategy for marine ecosystem
services in the Baltic Sea to provide consistent and meaningful
policy recommendations.
The present study is relevant for policy-makers, practitioners, and
researchers for the following reasons: first, it provides a clearly struc-
tured overview of what dataare available and which ecosystem services
have been valued by which methods. For example, the limited avail-
ability of observed travel behaviour data seems to have forced research-
ers to use stated preference methods instead of revealed preference
methods. Improving our knowledge of available data is indispensable
for setting priorities for future data collection efforts (Bo
¨rger et al.,
2014). Second, revealing blind spots of ecosystem services valuation
may provide guidance for defining future research directions.
Research on the advancement of valuation methods regarding
neglected ecosystem services is recommended to establish a more
consistent valuation framework. Third, the paper provides applied
researchers and policy-makers with an overview of the existing litera-
ture to facilitate conducting of further meta- and cost-benefit analyses.
Material and methods
The data processed in this paper were collected from studies pub-
lished in peer-reviewed international journals and final project
reports from 1995 to the beginning of 2015. We searched Google
Scholar using the following terms: Baltic Sea AND (valuation OR
stated preferences OR revealed preferences OR abatement costs
OR cost-benefit analysis.) This method provided more relevant
results than Scopus or searching databases on marine ecosystem
service valuation, such as http:// www.marineecosystemservices.
org/). Only studies fulfilling the requirement of using primary or
secondary data to value at least one ecosystem service related to
the Baltic Sea with an economic valuation method were taken into
account. Considered methods include market price and cost-based
methods as well as stated and revealed preference approaches.
Studies relying primarily on meta-analysis and literature review
were not incorporated, but studies based on already published
results to conduct new analysis (e.g. benefit transfer, cost-benefit)
were included. For comparison, all monetary values were converted
to Euros using the average exchange rate from the year of publica-
tion, based on the online database OANDA (OANDA, 2015).
Then, values were converted to price levels for 2014 and normalized
for purchasing power using the harmonized indices of consumer
prices and comparative price levels provided by EUROSTAT
(Eurostat, 2015a,b).
For ecosystem services classification, the approach from
HELCOM (2010) was used (see the Ecosystem services classification
and valuation section). Methods were usually named as mentioned
in each study, but some methods were subsumed under a broader
category, as they lacked sufficient observations. The scope included
all nine states neighbouring the Baltic Sea (from now on riparian
states): Denmark, Germany, Poland, Lithuania, Latvia, Estonia,
Russia, Finland, and Sweden.
As many publications have reported results for several riparian
states, ecosystem services, and/or applied methods, the data were
organized into a multilevel format, with each observationconsisting
of one unique combination of these three categories. For example, a
publication assessing the value of recreation using the travel cost
method and contingent valuation in Sweden was recorded as two
observations, because two methods were used. Table 1illustrates
the data alignment used, including the just-explained example
(number 35). The full table is available in the Supplementary
Appendix and can be readily extended for meta-analysis.
Ecosystem services classification and valuation
Since it is still the most common and frequently used classifi cation
system, the approach from HELCOM (2010) and the Millennium
Ecosystem Assessment Report (Millennium Ecosystem Assessment,
2005) was followed for categorizing ecosystem services into provi-
sioning, regulating, cultural, and supporting services (Table 2).
[There are some drawbacks to the HELCOM ecosystem services clas-
sificationsystem that have implicationsfor valuation (de Groot et al.,
2002;Hein et al.,2006;Boyd and Banzhaf, 2007;Wallace, 2007;Fisher
and Turner,2008;Fisher et al.,2009). For example, ecosystem services
are interdependent or overlapping (Kandziora et al.,2013).] As sup-
porting ecosystem services generally set the basis for generation of the
other services, manyauthors have regarded them as a special case,not
integrated into their primary valuation, to avoid double counting
(Hein et al.,2006).
Authors have used a wide variety of methods to value ecosystem
services (Haab and MacConnell, 2002;Freeman, 2003;Hanley and
Barbier, 2009;TEEB, 2010), including market price and cost-based
methods as well as revealed and stated preference methods. Market
price methods are mainly concerned with estimating the welfare
effects of shifts in demand and supply curves.
Cost-based methods include assessment of abatement, damage,
and replacement or substitution costs, out of which the abatement
cost method has been used most frequently in the Baltic Sea
context, often applied to estimate costs for reducing nutrient
992 J. Sagebiel et al.
emissions to reduce eutrophication. Abatement cost values are
derived from the costs of measures that seek to prevent emissions
or immissions, such as reducing agricultural fertilization, restor-
ation of wetlands, or investment in wastewater treatment plants.
In general, however, cost-based methods lack a solid foundation
in economic theory, as they only rely on changes in costs—
thereby not accounting for market dynamics—and thus are
unable to provide a full picture of welfare changes (Freeman,
2003). Welfare changes could be estimated with, for example,
general equilibrium models, but often the additional value of such
models usually does not justify the costs induced by significantly
greater data requirements and computational efforts.
Revealed preference methods include the travel cost method and
hedonic pricing, both of which rely on observed data, such as the
actual distances tourists have travelled to marine recreational sites
and associated costs. Such methods make use of the assumption
that ecosystem services create value that can be observed in related
markets. High recreational value in some areas may, for example,
be indicated by increased real estate prices compared with other
areas. However, it is often difficult to trace back the sources of
increased prices, as several other variables may affect real estate
values (e.g. neighbourhood characteristics, access to public trans-
port). Thus, such methods are only applicable when the data
include sufficient uncorrelated control variables and independent
observations.
Meanwhile, stated preference techniques, most prominently dis-
crete choice experiments and contingent valuation, can avoid such
problems. They relyon survey-based data generated via hypothetical
decision situations for certain groups of stakeholders and are espe-
cially useful when the valued ecosystem service is not tradable on
markets and, therefore, prices and quantities cannot be directly
observed. (Stated preferences approache will be explained in more
detail below). In some cases, results from existing revealed and
stated preference studies are transferred to other areas, thereby
adjusting important variables such as income, an approach called
benefit transfer.
While market price methods as well as revealed and stated pref-
erence methods are theory-driven and consistent with economic
welfare theory (Freeman, 2003), cost-based approaches rely on
empirical and pragmatic considerations. It is, hence, difficult
to compare results which are produced by different methods
(Hirschfeld and Sagebiel, 2014). Yet different ecosystem services
require different valuation methods (Barbier, 2007,2011;DEFRA,
2007;Bateman et al., 2011b;Bouma and van Beukering, 2015).
Provisioning services usually produce goods that are physically
tradable, such as fish, making the market price method most appro-
priate to measure their value. In contrast, the value of regulating
services is usually not indicated by market prices, but cost-based
approaches or stated preference methods can be used to provide
appropriate approximations. The values of cultural ecosystem ser-
vices only emerge through their (subjective) effect on the wellbeing
of people, such as through the perceived value of recreation, and
are best measured with revealed and stated preference methods. A
more detailed discussion of advantages and disadvantages of these
valuation methods is found in Brouwer et al. (2013).
When it comes to cost-benefit analysis, different strategies exist
for combining results from different methods (Cost-benefit analysis
is not a valuation study in the sense intended here. But one can still
use results of such analysis to indirectly value ecosystem services).
One could, for example, use cost-based approaches for the cost
side of the analysis and revealed and stated preference methods
for the benefit side. Alternatively, stated and revealed preference
methods can complement the cost side, such as when assessing
the costs of a reduction in fish stock due to increased commercial
fishing activities. In addition to loss of future incomes for fisheries
and increased fish prices, stated preference methods can capture
associated non-use values [non-use values are “monetary values
on natural resources and environmental characteristics that are in-
dependent of any present or future use” (Freeman, 2003, p. 137)],
allowing a more holistic picture of costs and benefits.
Results and discussion
This section presents the results of the valuation studies considered
in this paper, providing a comprehensive picture of ecosystem ser-
vices valued and when, where, and how they were assessed. The
first part analyses all of the studies, whereas the second and third
parts examine in more detail those studies that have applied stated
preference and abatement cost methods. We have decided against
conducting a meta-regression because, first, in our case, there is
Table 2. Classification of ecosystem services according to HELCOM.
Provisioning ecosystem services Supporting ecosystem services
Food
Inedible goods
Energy
Space and waterways
Chemicals
Ornamental resources
Genetic resources
Biogeochemical cycles
Primary production
Food web dynamics
Biodiversity
Habitats
Resilience
Regulating ecosystem services Cultural ecosystem services
Impact on climate and air quality
Sediment retention
Reduction of eutrophication
Removal of nutrients
Regulation of pollutants
Recreation
Aesthetic value
Science and education
Cultural heritage
Inspiration
The legacy of the sea
Source:HELCOM (2010).
Table 1. Example of data alignment from the multilevel analysis format used for this study.
Study no. Title Year Country Ecosystem service Method
33 Economic criteria for using wetlands as
nitrogen sinks under uncertainty
2000 Sweden Reduction of eutrophication Abatement costs
34 Economic valuation for sustainable
development in the Swedish coastal zone
2005 Sweden Recreation Travel cost
34 2005 Sweden Habitats Travel cost
35 Economic valuation of sport-fishing in
Sweden
2004 Sweden Recreation Contingent valuation
35 2004 Sweden Recreation Travel cost
Economic valuation of Baltic marine ecosystem services 993
no common effect size (dependent variable) which could be ana-
lysed with such methods. This argument mirrors our main critique
of a lack of consistency in the economic valuation literature con-
cerning the Baltic Sea—a key reason for writing this paper.
Second, specifically for the Baltic Sea, the data are still too sparse
to obtain statistically valid results. Extending the dataset to the
whole of Europe or the entire world is not within the scope of this
paper, but it has already been done (Nelson and Kennedy, 2009;
Ahtiainen and Vanhatalo, 2012).
Overview
In total, 76 studies were reviewed, leading to 388 observations, with
an average study consisting of five observations (e.g. applying one
method to value one ecosystem service in five countries). The
average amount of studies conducted per year did not increase
during the almost 20-year period studied, though there is some
variation between years (Figure 1).
Table3indicates the number of methods, ecosystem services, and
riparian states each study has incorporated into its economic valu-
ation. Only nine studies valued more than one ecosystem service,
and none valued more than three services at once, indicating to us
that Baltic Sea economic valuation studies have hardly incorporated
ecosystem service interactions and related synergies. We will call this
Blind Spot 1. Similarly, 59 studies reported results based on using
only one method. Applying more than one method to a valuation
exercise tends to increase costs and efforts enormously, as it often
requires different datasets. Yet, such methodological contributions
(e.g. recreational values can be measured using travel costs and cross
checked with contingent valuation surveys) appear to be important
for investigating the reliability and robustness of individual
methods as well as the overall validity of economic valuations
(Blind Spot 2). Additionally, as explained in the Ecosystem services
classification and valuation section, holistic valuation of an ecosys-
tem service often requires several methods for capturing values for
different stakeholders. This could be remedied by studies synthesiz-
ing existing results, but, to our knowledge, no such studies exist
(Blind Spot 3). It is noteworthy that, in economic valuation
studies in general, approaches using several methods and valuing
several ecosystem services are still rather uncommon (Chan et al.,
2012;Gilvear et al., 2013). Thus, the first three blind spots we have
indicated come as no surprise.
Recent research on benefit transfer has shown that there are large
differences between riparian states that make it difficult to take
results from one and draw conclusions for the whole area
(Bateman et al., 2011a;Ahtiainen et al., 2015). In our data, only
25 out of 76 studies focused on all riparian states simultaneously,
while 48 studies were carried out in only one state, and 3 studies
were done in two or three states (Blind Spot 4). Studies conducted
in all riparian states were mostly abatement-cost based (see the
Abatement cost approaches to estimating costs for reducing Baltic
Sea eutrophication section), relying on secondary data. In some
cases, contingent valuation studies also investigated all states (see
the Stated preference approaches for estimating the value of
HELCOM Baltic Sea ecosystem services section).
Riparian states
Marine ecosystem services were studied most frequently in Sweden,
with 77 observations, followed by Finland, with 50.
All other riparian states were relatively similarly covered, with
between 34 and 41 observations each (Figure 2), which can be
mainly attributed to the fact that several studies included all
riparian states. Considering only studies thatdid not focus on all ri-
parian states (the dark bars in Figure 2) reveals that seven out of nine
states have been studied less than ten times (Blind Spot 5). Only
Sweden and Finland have 43 and 16 observations, implying that
the overall picture on economic values provided by the studies
may be biased towards values from Scandinavia and it may, conse-
quently, be difficult to draw conclusions concerning the whole
Baltic Sea area. In future research, less frequently covered states
could take a more prominent role to complement the rich data
availability from other states. Alternatively, to obtain values valid
for the whole Baltic Sea area, grant givers could concentrate on
Figure 1. Published Baltic marine ecosystem services valuation studies
per year between 1995 and 2015. This figure is available in black and
white in print and in colour at ICES Journal of Marine Science online.
Table 3. Number of methods, ecosystem services, and riparian states
incorporated into economic valuation studies of the Baltic Sea.
X
Number of Xper study
123... 9
Ecosystem services analysed 67 5 4 –
Methods used 59 13 4 –
Riparian states included 48 1 2 25
Total number of studies n¼76
Figure 2. Frequency of each riparian state being covered by valuation
studies (1995–2015), among all studies examined and for studies not
covering all nine riparian states. This figure is available in black and
white in print and in colour at ICES Journal of Marine Science online.
994 J. Sagebiel et al.
supporting transboundary projects. In this respect, the BalticSTERN
project (http://www.stockholmresilience.org/21/research/research-
programmes/balticstern.html) and the recently relaunched
European Union Bonus projects (http://www.bonusportal.org/)
seem to be promising ways forward.
Ecosystem services
Figure 3shows the frequency of different ecosystem service valua-
tions in absolute and relative terms. Reduction of eutrophication
was by far valued most frequently (327 times), using the abatement
cost method (194 times), stated preference methods (57 times),
benefit transfer (39 times), and cost–benefit analysis (22 times).
This appears to be due, first, to eutrophication being a major issue
in the Baltic Sea (Gren et al., 2000) and, second, because changes
in the status of eutrophication have tangible effects for various
stakeholders.
The great focus on reduced eutrophication is therefore no sur-
prise. Yet, although the Baltic Sea is also quite vulnerable to the
adverse effects of pollutants and heavy metals—related to the eco-
system service “regulation of pollutants” (EMEP, 2013)—this eco-
system service has only been valued twice and only with stated
and revealed preference methods (Blind Spot 6).
Recreation has been valued 29 times, usually with stated prefer-
ence methods (Figure 3). Interestingly, food, a provisioning service,
has been valued 17 times with several methods, including stated
preference methods and cost-based approaches, though here the
market price method seems most suitable. However, as with eu-
trophication, changes in the production of food—primarily fish
in the Baltic Sea context—have implications for other ecosystem
services. For example, there may exist non-use values for certain
species of fish that market price methods cannot capture. Only
stated preference methods can measure such non-use values. It
thus seems obvious that interactions between ecosystem services
are important for economic valuation, with the example of fish
showing that different methods may be required to capture the
total economic value of individual ecosystem services.
We now come to the somewhat surprising fact that several eco-
system services have not been valued at all. Some may be difficult
to measure (“genetic resources”, “inspiration”), not threatened or
not alterable (“space and waterways”). Yet others are quite import-
ant, and it is unclear why they have so often been ignored. Consider
“energy”, a provisioning ecosystem service. Although there is a large
body of literature on offshore wind and geothermal energy, and off-
shore wind energy has been abundantly valued with stated prefer-
ence methods, it has not been discussed within the ecosystem
services context. Rather, such studies have mainly been focused on
the aesthetic disadvantages of the visibility of these technologies
and, if at all, fit better into the classification of cultural ecosystem
services concerning “cultural heritage” and “aesthetic value”
(Ladenburg and Dubgaard, 2007,2009;Haggett, 2011).
Five observations were concerned with “biodiversity” and two
with “habitats”, both supporting ecosystem services. As explained
earlier, however, researchers have suggested not including support-
ing services to avoid double counting, demonstrating inconsisten-
cies in valuation approaches as well as their complexity.
Methods
Abatement cost has been the most frequently applied method, with
194 observations (Figure 4), and was exclusively applied for reduc-
tion of eutrophication. The damage cost method has been used 13
times and replacement cost three times. Authors haveless frequently
used stated preference methods compared with abatement cost
ones: our data include 78 observations for contingent valuation
studies and 12 observations for discrete choice experiments.
Meanwhile, revealed preference methods have rarely been used
(six times for the travel cost method and once for hedonic
pricing), which may be caused by limited data availability (Blind
Spot 7). In the UK, a large panel dataset on the recreational activities
of UK residents is readily available (Natural England, 2015), that has
enabled thorough analyses to be undertaken using the travel cost
method (Bateman et al., 2013;Sen et al., 2014). In contrast,
similar data regarding the Baltic Sea are at present rather limited,
which may be the reason for the infrequent application of revealed
preference methods.
Altogether, stated and revealed preference methods were used 97
times to value various ecosystem services, including biodiversity,
habitats, recreation, and reduction of eutrophication. Discrete
choice experiments have been especially useful here, as they allow
the inferring of preferences for different ecosystem services within
a single survey. The rarely used revealed preference methods were
exclusively focused on recreation.
Figure 3. Economic valuation methods used per Baltic Sea ecosystem
service studied (1995– 2015).
Figure 4. Frequency of economic valuation methods applied for
valuing marine ecosystem services related to the Baltic Sea region
(1995–2015). This figure is available in black and white in print and in
colour at ICES Journal of Marine Science online.
Economic valuation of Baltic marine ecosystem services 995
Benefit transfer, as a low-cost alternative to primary data
collection, was found 40 times. Earlier applications of this
method have often, however, been contested as providing in-
accurate results, making it difficult to rely on such studies, espe-
cially if their values are incorporated into cost-benefit analysis
(Blind Spot 8). Nonetheless, recent methodological developments
regarding this approach seem promising and may provide new
ways to obtain more reliable economic values (Bateman et al.,
2011a;Johnston et al., 2015).
Cost-benefit analysis had 22 observations (from six studies) in
our data and other methods that cannot be categorized with those
explained in the Ecosystem services classification and valuation
section totalled 14 observations, coming from 6 studies. These
included calculation of net present values, other cost-based
approaches, and estimations of demand elasticities.
Synthesis
The above analysis has revealed what we consider to be eight blind
spots in the economic valuation literature regarding the Baltic
Sea, the key ones being riparian states not sufficiently covered,
neglected ecosystem services and ecosystem service interactions,
and limited employment of multiple methods. Funding bodies,
such as the European Union and national governments, as well as
researchers will now hopefully take the opportunity to steer their
activities towards attacking these blind spots, which Table 4sum-
marizes in more detail. Funding agencies could support research
that helps to fill gaps here while also aiming at establishing a Baltic
Sea-specific platform to collect and communicate the results of
past and future Baltic Sea economic valuation studies. It is also pos-
sible for existing platforms to integrate such results. HELCOM, for
example, provides geo-referenced data and maps assessing various
ecosystem services, yet provides no estimates of their value (http
://www.helcom.fi/baltic-sea-trends/data-maps).
Researchers can make use of the already existing valuation data,
such as from Sweden, to conduct further secondary analyses while
also focusing their actions towards regions and ecosystem services
that have been less frequently valued. Further, they can tackle emer-
ging issues such as the incorporation of interactions between ecosys-
tem services and make use of state-of-the-art methodological
advances in areas such as benefit transfer. Finally, all relevant
actors can try to work towards building an economic valuation
framework that can serve as a basis for future research activities.
Stated preference approaches for estimating the value
of HELCOM Baltic Sea ecosystem services
Stated preference applications comprise contingent valuation
surveys and discrete choice experiments. In contingent valuation,
respondents state their willingness to pay for an improved scenario,
Table 4. Blind spots located within the Baltic Sea ecosystem services economic valuation literature (1995– 2015) and possible responses from
policy-makers, funding agencies, and researchers.
nBlind spots Possible responses from funding agencies Possible responses from researchers
1 Interactions between ecosystem services are
often neglected, as 67 of 76 studies valued
only one ecosystem at a time
Incorporate modelling of ecosystem services
interactions in new projects
Use valuation methods (e.g. discrete choice
experiments) that can account for
interactions
2 Different methods are required to value one
ecosystem service (see reduction of
eutrophication and food), though 59 of 76
studies only used one method
Enlarge the role of valuation in such projects,
and encourage researchers with expertise in
different kinds of methods to participate,
taking into account that additional costs can
arise
Consider using various methods to value a
particular ecosystem service. Increase
cooperation with researchers from other
fields
3 There are no synthesizing studies that bring
together results from previous studies,
although such studies could be used towards
obtaining a holistic valuation of ecosystem
services
Support projects that synthesize existing
economic valuation data; establish a platform
that collects and makes available all economic
valuation studies concerning the Baltic Sea
Develop a framework for arriving at a holistic
economic valuation scenario
4 Conclusions for the Baltic Sea as a whole cannot
be easily drawn from results from one
country; yet only 33% of the reviewed studies
incorporated all nine riparian states
Support projects that conduct research in all
riparian states; strengthen European
Union-wide collaboration
Frame new research in ways such that it can
be compared with existing results
5 All states beside Sweden and Finland have been
valued less than ten times individually, which
may create a bias of values towards
Scandinavian preferences
Focus primary data collection support to areas
with limited existing valuation data
Use existing data in Sweden for further
analysis (e.g. cost–benefit). Apply newer
benefit transfer methods to transfer values
from Sweden and Finland to less explored
countries, and especially test their accuracy
6 Several ecosystem services have not been valued
at all
Identify further ecosystem services that are
relevant for EU directives concerning the
Baltic Sea, and support research activities for
them
Develop/apply methods that can be used to
value such ecosystem services
7 Revealed preference methods have been used
only seven times, possibly attributable to low
data availability.
Similarly to the UK, centrally organized primary
data collection should be supported
Use existing data if possible, and complement
stated preference studies with revealed
preference methods
8 Benefit transfer has been used, but recent
methodological advances have revealed
several inaccuracies in the estimated values,
questioning its validity
Support further methodological research on
benefit transfer
Do not rely too much on older benefit
transfer studies; make use of recent
advances to obtain more accurate results
996 J. Sagebiel et al.
compared with baseline or status quo ones. Researchers can use a
variety of elicitation formats, which one can broadly categorize
into open-ended and closed-ended questions. With open-ended
questions, respondents can freely state a willingness to pay value,
while in closed-ended formats, they can either choose from a set
of given values (payment card) or can agree or disagree to pay a
fixed amount (dichotomous choice). In contrast, in discrete
choice experiments, respondents choose among alternatives that
differ in terms of their characteristics or attributes, one of which is
a monetary one, such that respondents face a trade-off between
the other attributes and the costs. The method has the advantage
that each attribute can be assigned an individual monetary value.
Thus, researchers can elicit more detailed information on people’s
preferences and infer values for different ecosystem services
simultaneously. [The introductory literature on discrete choice
experiments and contingent valuation is abundant. Readers are re-
ferred to Alberini and Kahn (2006),Freeman (2003),Haab and
MacConnell (2002), and Louviere et al. (2006).]
Among the stated preference applications used for the Baltic Sea,
11 studies estimated willingness to pay for improvement of water
quality (Table 5). Five of these studies explicitly referred to reduc-
tion of eutrophication, while the others used other ecosystem
services—biodiversity, habitats, food, aesthetic value, recreation—
to describe water quality. Three studies used discrete choice experi-
ments and the remaining ones contingent valuation.
Gren et al. (1997) conducted the first contingent valuation
survey with multiple countries. They included data from an earlier
Swedish contingent valuation survey and used the estimated
values to generate willingness to pay values for all other
non-ex-socialist riparian states. They also included values from
Poland and Lithuania, taken from an earlier draft of Markowska
and Z
˙ylicz (1999) and Z
˙ylicz et al. (1995).
Other studies that have directly focused on reduction of
eutrophication are Atkins and Burdon (2006), who used the
Randers Fjord in Denmark as a case study, and Ahtiainen et al.
(2014), who have conducted the largest Baltic Sea contingent valu-
ation study undertaken to date, covering all nine riparian states and
with over 10 000 respondents. Other contingent valuation publica-
tions have not exclusively focused on reduction of eutrophication.
O
¨stberg et al. (2012), for example, valued improved water quality
defined as reduced noise and litter at recreational sites within the
context of Special Conservation Zones and the EU Water
Framework Directive. Meanwhile, Toivonen et al. (2004) estimated
willingness to pay for recreational fishing, and Ressurreic¸a
˜oet al.
(2012) focused on biodiversity by investigating willingness to pay
for species loss.
Within the discrete choice experiment studies, different ecosys-
tem services were simultaneously covered. Eggert and Olsson
(2009)—the first discrete choice experiment study for the Baltic
Sea context—used coastal cod stock, bathing water quality, and bio-
diversity as attributes, thereby capturing economic values for the
ecosystem services food, recreation, and biodiversity in Sweden.
Konsenius (2010) conducted a discrete choice experiment in
Finland regarding various attributes representing the ecosystem ser-
vices recreation, aesthetic value, and food, whereas Liu and Wirtz
(2010) used Germany as a case study to assign values for the improve-
ment of oil spill management, using a discrete choice experiment to
cover the ecosystem services recreation, regulation of pollutants, and
food.
A closer look at estimated willingness to pay values reveals high
variance among them as their overall framing, the goods to be
Table 5. Studies using stated preference methods to value HELCOM Baltic Sea ecosystem services (1995–2014).
Reference Riparian state Benefit Ecosystem service Method Period Payment vehicle WTP
a
WTP
a
pP
and year
Z
˙ylicz et al. (1995) Poland Reduction of eutrophication Reduction of eutrophication CVM Not specified Yearly tax pP 309 309
Gren et al. (1997) Sweden Reduction of eutrophication Reduction of eutrophication CVM 20 years Yearly tax pP 385 385
Markowska and Z
˙ylicz (1999) Poland, Lithuania Reduction of eutrophication Reduction of eutrophication CVM 10 years Yearly tax pP 61–150 61–150
Toivonen et al. (2004) Denmark, Sweden, Finland Recreational fishing Recreation CVM 1 year Yearly fee pP 57–88 57–88
Atkins and Burdon, (2006) Denmark Reduction of eutrophication Reduction of eutrophication CVM 10 years Monthly payment pP 10 120
Eggert and Olsson (2009) Sweden Improvement of water quality
b
Biodiversity, recreation, food DCE 1 year Yearly fee pP 50 –107 50– 107
Kosenius (2010) Finland Improvement of water quality
c
Recreation, aesthetic value, food DCE 20 years Yearly fee pHH 325 –423 155– 201
Liu and Wirtz (2010) Germany Improvement of oil-spill
management
a
Recreation, regulation of pollutants,
food
DCE Not specified Yearly fee pHH 166 83
O
¨stberg et al. (2012) Sweden Improved water quality/reduction
in noise and litter
Habitats, reduction in
eutrophication
CVM 20 years Monthly fee pHH 3– 10 19–54
Ressurreic¸a
˜o et al. (2012) Poland Prevention of species loss (3 levels) Biodiversity CVM Not specified Onetime payment
pP
47–181 47 – 181
Ahtiainen et al. (2014) All riparian states Reduction of eutrophication Reduction of eutrophication CVM Infinite Yearly tax pP 6 –75 6– 75
WTP, willingness to pay; CVM, contingent valuation method; DCE, discrete choice experiment; pP, per person; pHH, per household; CPL, comparative price levels.
Attributes include coastal waters protected from oil pollution, beaches avoided from oil pollution, Eider ducks avoided from oil pollution, collect ratio of spilled oil to be collected by combat vessels, yearly payment for
using combat facilities.
a
All WTP’s are given in EUR
2014
based on CPL
EU-28
.
b
Attributes include coastal cod stock levels, bathing water quality, and biodiversity levels.
c
Attributes include water clarity, abundance of coarse fish, status of bladder wrack population, mass occurrence of blue– green algae blooms.
Economic valuation of Baltic marine ecosystem services 997
valued, applied methods, and time frames differ. We tried different
meta-regressions with the available data, but the limited number of
observations and high correlations among explanatory variables did
not allow drawing of statisticallyvalid conclusions. Stilla descriptive
analysis reveals some insights. It is notable that, even between studies
that appeared similar in their framing, the resulting values differed
significantly. For example, Ahtiainen et al. (2014) estimated willing-
ness to pay for a 50% reduction of eutrophication at between 6 and
75 EUR per person and year while, for the same reduction and
periodicity, Gren et al. (1997) came to a value of 385 EUR, and
Markowska and Z
˙ylicz (1999) estimated willingness to pay as
being between 61 and 150 EUR. These differences may come from
a historical change in preferences (18-year difference between
Gren et al. and Ahtiainen et al.) or may be due to methodological
standards that have changed over the years but which are not
related to changing preferences. For example, recent studies have
frequently used online panels, while earlier studies used in-person
or telephone interviews (Lindhjem and Navrud, 2011).
Willingness to pay is further influenced by the assumptions
researchers make. Sometimes they want to obtain conservative esti-
mates of willingness to pay to avoid critique of inflated values or
just to provide lower bounds (willingness to pay is larger than ...).
An interesting example is the study by Gren et al. (1997), where the
authors used Polish willingness to pay values from Z
˙ylicz et al.
(1995). Unlike the reported values in Z
˙ylicz et al., they assumed non-
responses as zero willingness to pay values. As the response rate was
50%, willingness to pay was consequently halved in their study. This
example demonstrates the challenge of understanding economic
valuation studies. In decision-making processes, it is difficult for
policy-makers to elicit meaningful values from such studies, as
there are no clear guidelinesand several assumptionshave to be made.
A sensitive characteristic of stated preference studies is payment
vehicle, that is, the way the respondent is supposed to pay for a pro-
posed improvement (Morrison et al., 2000;Ivehammar, 2009).
Proposed payment vehicle definitions varied strongly among
Baltic Sea valuation studies. For instance, three studies referred to
willingness to pay of households, whereas the remaining eight
studies referred to per-person values. Four studies referred to a
“tax”, while the others referred to a “fee” or “payment”. Further,
the studies differed in terms of their periodicities, meaning the
time frames within which proposed payments are to be made.
Two studies considered a one-time payment, five studies focused
on a period between 10 and 20 years, one study had an infinite peri-
odicity, and three studies did not report their periodicity.
Recent work on stated preference methods has revealed further
methodological challenges that influence welfare estimates, includ-
ing protest responses, i.e. cases where respondents do have a willing-
ness to pay but state it to be zero, because they have constraints that
prevent them from revealing their true willingness to pay
(Meyerhoff and Liebe, 2008), experimental design complexity
(Meyerhoff et al., 2015;Weller et al., 2014;Oehlmann et al., 2014),
and elicitation format (open vs. closed ended) in contingent valu-
ation (Kealy and Turner, 1993). We cannot drawfurther conclusions
on whether willingness to pay in the studies presented here are influ-
enced by such methodological challenges but can at least note, based
on findings from the literature, that they do exist in general.
We do conclude that it is cumbersome for policy-makers to use
the currently available results for practical planning activities and
for researchers to incorporate them into cost-benefit analyses, as it
would seem hardly possible to trace back the reasons for the
various outcomes they are confronted with.
Abatement cost approaches to estimating costs
for reducing Baltic Sea eutrophication
In the literature, the abatement cost method was exclusively used for
estimating the costs of measures to reduce eutrophication, with two
research questions being dominant. The first question focused on
comparison of costs for individual measures to reduce eutrophica-
tion, whereas the second dealt with the identification of cost-
effective or coordinated solutions (i.e. where all countries cooperate
on an overall cost-effective reduction option). While the first ques-
tion is more general, thesecond can be consideredto be seeking a sci-
entific response to the Baltic Sea Action Plan, as most studies applied
different models to identify strategies that may help in achieving its
goals at minimum cost. These research questions are of course inter-
related, as applying most cost-efficient measures is a prerequisite for
achieving cost-effective reduction of eutrophication.
Concerning the first research question, most studies distin-
guished between agricultural land use measures, such as improved
fertilization techniques, cultivation of catch crops and restoration
of wetlands, and direct abatement measures, which had a primarily
technical character, such as increasing capacities of wastewater man-
agement in purification plants. Costs of land use changes were
assessed by estimating opportunity costs for alternative land uses
(Gren, 2008a). The costs of reducing the use of fertilizers were
usually calculated as the change in farmers’ surplus that would
result from a given reduction in fertilizer application (Kiirikki,
2003). Direct measures were generally assessed based on economet-
ric methods, estimating the additional costs that arise when a
measure is implemented (Gren et al., 1997, 2000). Costs for direct
abatement measures ranged between 2 and 86 EUR kg
21
of abated
nitrogen, between ,1 and 243 EUR kg
21
of abated phosphorus,
and between 2 and 25 EUR kg
21
for the abatement of both nutrients
(Table 6).
The most cost-efficient measures for nitrogen reduction were
improved fertilization techniques, restored wetlands, and improved
treatment of wastewater, whereas for phosphorus reduction, they
were the use of phosphate-free detergents and more effective waste-
water treatment. Although most studies provided values per kg and
used similar measures, direct comparability was not always given, as
they differed in terms of examined nutrients (either nitrogen or ni-
trogen and phosphorus), reduction targets, discount rates, as well as
quantity and specification of measures. This comparability problem
becomes obvious when considering the wildly differing highs for
abatement costs of only one nutrient compared with the relatively
low upper limit for abating both.
The policy background for the second research question regard-
ing the identification of cost-effective abatement solutions is com-
mitment to a preliminary nutrient-reduction scheme agreed upon
by all riparian states in 2007, based on the concept of setting
maximum allowable nutrient inputs in water and air necessary to
reach good environmental status for the Baltic Sea (see HELCOM
(2013) for more details). Parallel with the HELCOM targets,
several studies compared regionally differentiated coordinated solu-
tions with unilateral solutions (Table 7).
A coordinated solution means that all riparian states would co-
operate on an overall cost-effective reduction option. Meanwhile,
a unilateral solution means agreement to a policy with uniform
proportional load reductions (Gren, 2008b). Such policies have
been common for nitrogen reduction regarding the Baltic Sea,
where riparian states have agreed to reduce their loads by a certain
proportion in relation to their initial loads (Hjorth, 1992). A
998 J. Sagebiel et al.
central finding of these studies is that it appears to be worthwhile to
focus nutrient-abatement efforts on economic sectors, riparian
states, and regions within those states, which have the greatest
potential for nutrient reduction (Ahlvik et al., 2014). The most cost-
effective measures have been reported to be restoring wetlands and
increasing the capacities of wastewater management in purification
plants and cultivation of catch crops. The nitrogen abatement costs
for a unilateral solution at a 50% reduction target wereonly reported
in Ollikainen and Honkatukia (2001),at242 EUR kg
21
nitrogen
being abated. For the same reduction target, the costs of the coordi-
nated solutions ranged between 1 and 162 EUR kg
21
. Abatement
costs at a 50% reduction target for phosphorus ranged between
70 and 650 EUR kg
21
for a unilateral solution and 0 to 352
EUR kg
21
for a coordinated solution.
Similar to our observations regarding stated preference approaches,
the abatement cost studies also differed in terms of the details of the
methodsused.Differentreductiontargetsandvariouswaysofestimat-
ing the success of implemented measures make comparisons difficult.
Another challenge was the observation of two nutrients (phosphorus
and nitrogen) in some studies and only one in others. Additionally,
costsperkgincreasewithrisingambitionsoftheselectedreduction
target. Due to these increasing margi nal costs of abatement, reduction
costs per kg are lower for a 20% reduction target than for a 50% one.
Thus, comparing such values would require accounting for the often
not clearly understood non-linearities of abatement costs.
Conclusions
In this paper, a total of 76 empirical studies conducted between 1995
and 2015 focusing on marine ecosystem service valuation in the
Baltic Sea area have been considered. Two findings from our analysis
appear to be especially relevant for policy-makers and researchers
who plan to conduct or investigate valuation studies in the future.
First, there is an imbalance among ecosystem services valued
and countries covered. While the regulating ecosystem service
reduction of eutrophication and the cultural ecosystem service re-
creation have frequently been the subject of valuation studies,
thus far most Baltic Sea marine ecosystem services, as categorized
by the HELCOM commission, have only rarely been valued or
have not at all been the subject of economic assessment. Since infor-
mation in monetary terms is usually easy to understand and an
essential part of political decision-making, increased efforts
should be made to undertake studies on the economic valuation
of such understudied marine ecosystem services. Further, the
number of conducted studies among the different riparian states
varied strongly. While data availability in Scandinavian countries
has been rather plentiful, authors have hardly covered Eastern
European countries. Consequently, estimated values based on
these data have had only limited capability when formulating state-
ments regarding the Baltic Sea riparian states in general. Another
blind spot is related to the treatment of interactions between
Table 6. Studies calculating abatement costs for various measures to reduce nutrient immissions into the Baltic Sea (2000– 2012).
Reference Riparian state
Reduction
target Measures
Results in million
EUR
2014
(CPL) per year
Discount
rate
Costs in EUR
2014
(CPL) kg
21
Nitrogen
Bystro¨m (2000) Sweden 50% Wetlands 64 – 13
Agriculture 89 – 17
Kiirikki (2003)
a
Finland 2100 t year
21
Sewage – 3% 6
Agriculture – 3% 28
Larsson et al. (2005)
b
Sweden 21% Agriculture – – 5
Helin et al. (2006) Finland 50% Agriculture 35 –48 – 7–9
Schou et al. (2006)
c
All 160 000 t Wetlands/agriculture/
sewage
60/895/12 3% 2
HELCOM and NEFCO (2007) All 50% Sewage – – 47–102
Agriculture – – 13
Mewes (2012) Germany 25% Agriculture 34 – 10
d
Phosphorus
Kiirikki (2003)
a
Finland 62 t year
21
Sewage 3% Below 1
HELCOM and NEFCO (2007) All 50% Detergents 207 – 64
Sewage – – 132– 280
Mewes (2012) Germany 25% Agriculture 35 – 345
d
Nitrogen and phosphorus simultaneously
Elofsson (2003) Sweden, Russia,
Poland
50% Agriculture 1489 – 4
d
HELCOM and NEFCO (2007) All 50% Sewage 484– 1155 – –
Agriculture 1402 – –
Combination and
additional measures
3525 – 25
Ahlvik et al. (2014)
e
All 50% Agriculture/wetlands/
sewage/detergents
1476 3.5% 2
d
a
All results are for nitrogen equivalents.
b
Cost calculations are divided into farm short-term, farm long-term, and social costs.
c
Deprecation period of 20 years.
d
Based on own calculations from data provided by Mewes, 2012 (average value from different regions) and Ahlvik et al., 2014 (total costs/total kg N +P).
e
Computation of the least-cost solution to reach good environmental status of the Baltic Sea (according to BSAP) within 40 years.
Economic valuation of Baltic marine ecosystem services 999
ecosystem services. While such interactions have been emphasized
in the literature on ecosystem service assessment and in the
natural sciences (Bennett et al., 2009), nearly all of the Baltic Sea
valuation studies that we analysed neglected their combined
effects—positive and negative—thus hindering a more comprehen-
sive understanding of the values of ecosystem services. In addition,
no synthesizing study aimed at bringing together estimated values
from different ecosystem services was found, making a holistic
view on ecosystem service values hardly possible.
Second, the analysis conducted here of stated preference and
abatement cost studies has shown that there is great variation in
estimated values, making it difficult for policy-makers to predict
benefits and costs, such as concerning measures for reduction of
eutrophication. The blind spots we have pinpointed appear to
have strong implications for cost– benefit analyses, as not incorpor-
ating important values may lead to biased results when adding up
the costs and benefits of a measure and, consequently, may lead to
imprecise or even misguided policy recommendations. Using
meta-analyses which incorporate methodological, regional, and
other variables to explain estimated values may be a good means
for obtaining more reliable results (Nelson and Kennedy, 2009).
However, meta-analyses require a large number of observations,
and the number of stated preference studies related to ecosystem
services in the catchment area of the Baltic Sea is not yet sufficiently
large. Additionally, the “effect size”, the dependent variable in a
meta-regression, needs to be the same across studies, which does
not hold true for the presented cases. As a consequence, authors
primarily provided literature reviews instead of meta-analyses
(Elofsson, 2010a;Bertram and Rehdanz, 2013).
Toestablisha sound basis forcomparable primarystudies andsub-
sequentmeta-analyses and cost– benefit analyses regardingthe Baltic
Sea region, a more unified valuation framework is required. In the
early 1990s, the National Oceanic and Atmospheric Administration
Panel developed general guidelines for conducting contingent valu-
ation studies (Arrow et al.,1993). A similar approach adapted to the
Baltic Sea is justified because, compared with other seas, the Baltic
Sea is more vulnerable to eutrophication and its consequences due
to it being semi-brackish and having limited water exchange, com-
bined with large-scale industrial and agricultural activities in the
riparian states (Gren et al., 2000). In other EU policy contexts, for
example the Water Framework Directive, initial attempts have
already been made in this direction (Brouwer et al.,2009), but
making greater efforts towards developing a more systematic and reli-
able economic valuation methodological toolbox for marine ecosys-
tem services could provide more transparent information for
policy-makersseeking to find efficient solutionsto improve the envir-
onmental status of the Baltic Sea’s marine environment.
Finally, it is important to mention that the Baltic Sea is well
covered with economic valuation studies compared with other seas
and several issues we have mentioned here are of general nature and
valid forother geographicalareas. Thus, ouroverall recommendation
is not necessarily to increase the number of valuation studies, but
rather to aim for a more coordinated future research agenda, which
fills the existing gaps, while making use of the large body of values
that already exist.
We propose that neglected riparian states such as the former
socialist countries should become the focus of new research or,alter-
natively, new projects should be set up in a transboundary manner
to cover all riparian states similarly. Research focus should also be
placed on neglected ecosystem services such as regulation of pollu-
tants, to complement already existing values such as those for the
ecosystem service reduction of eutrophication. Methodological
challenges should also be tackled from all sides. Standards for differ-
ent valuation methods could be established,such as guidelinesfor the
use of payment vehicles in stated preference studies. Developing a hol-
istic picture covering all riparian states and ecosystem services should
be a statedaim of European Unionresearch activities, as only thenwill
sound policyrecommendations become possible, based on more ac-
curate identification of ecosystem services that require additional
resources for their protection. With a more comprehensive picture
of where help is most crucially needed, measures for improving the
state of the Baltic Sea can then be targeted more specifically and
more accurately.
Supplementary data
Supplementary material is available at the ICESJMS online version
of the manuscript.
Table 7. Studies on coordinated vs. uniform abatement costs for the reduction in nitrogen and phosphorous immissions into the Baltic Sea
(1997–2010).
Reference
Reduction
target
Number of
measures
Result in million EUR
2014
(CPL) per year Cost in EUR
2014
(CPL) kg
21
Unilateral Coordinated Unilateral Coordinated
Nitrogen
Gren (2001) 50% – 4318
a
2689
a
– 1–217
Ollikainen and Honkatukia (2001) 50% – 140 084 21 072 324
a
48
b
Gren (2008a) 0–50% 14 – ,3300
c
30
c
Elofsson (2010b)
d
24% 14 4858 4082 1– 75 0 –71
Phosphorus –
Ollikainen and Honkatukia (2001) 50% – 10 229 1094 301
b
32
b
Gren (2008a) 0–70% 10 – ,2 200
c
– 211
c
Elofsson (2010b)
d
50% 8 4858 4082 75 –697 0–377
Nitrogen and phosphorus simultaneously
Gren et al. (1997) 50% 21 659– 6500 110– 1810 – 5
b
a
Cost– benefit analysis; figures are net-costs.
b
Based on our own calculations from data provided by Ollikainen and Honkatukia (2001): total costs/total kg N and P.
c
Figures are based on a 50% reduction target.
d
Elofsson reports zero marginal costs for those basins where only one nutrient is binding, i.e. the other is abated “for free”.
1000 J. Sagebiel et al.
Acknowledgements
We gratefully acknowledge support from the German Ministry for
Education and Research within the SECOS (Grant Number:
03F0666) and KLIMZUG-RADOST (Grant Number: 01LR0807H)
projects. Special thanks go to Eva-Maria Brodte for comments
on an earlier draft as well as to Carolin Hoffmann, Daniel Bo
¨ss,
Heinrich Bo
¨ing, and Stefanie Doll for support in preparation of the
manuscript. We are also grateful to Christopher Hank for several
comments on content and language. Finally, we thank the editor
Claire Armstrong and two anonymous reviewers for their valuable
comments.
References
Ahlvik, L., Ekholm, P., Hyytia
¨inen, K., and Pitka
¨nen, H. 2014. An
economic–ecological model to evaluate impacts of nutrient abate-
ment in the Baltic Sea. Environmental Modelling and Software, 55:
164–175.
Ahtiainen, H., Artell, J., Czajkowski, M., Hasler, B., Hasselstro
¨m, L.,
Huhtala, A., Meyerhoff, J., et al. 2014. Benefits of meeting nutrient
reduction targets for the Baltic Sea—a contingent valuation study
in the nine coastal states. Journal of Environmental Economics
and Policy, 3: 278– 305.
Ahtiainen, H., Artell, J., Czajkowski, M., and Meyerhoff, J. 2015.
Performance of different approaches in international benefit trans-
fer: insights from a nine country experiment. Paper prepared
for the 21st Annual Conference of the European Association of
Environmental and Resource Economists, Helsinki, 24– 27 June
2015. http://www.webmeets.com/eaere/2015/prog/viewpaper.asp?
pid=762 (last accessed 7 September 2015).
Ahtiainen, H., and Vanhatalo, J. 2012. The value of reducing eutrophi-
cation in European marine areas—a Bayesian meta-analysis.
Ecological Economics, 83: 1–10.
Alberini, A., and Kahn, J. R. 2006. Handbook on Contingent Valuation.
Elgar Original Reference. Elgar, Cheltenham.
Arrow, K., Solow, R., Portney, P. R., Leamer, E. E., Radner, R., and
Schuman, H. 1993. Report of the NOAA Panel on Contingent
Valuation.
Atkins, J. P., and Burdon, D. 2006. An initial economic evaluation of
water quality improvements in the Randers Fjord, Denmark.
Marine Pollution Bulletin, 53: 195– 204.
Barbier, E. B. 2007. Valuing ecosystem services as productive inputs.
Economic Policy, 22: 178– 229.
Barbier, E. B. 2011. Progress and challenges in valuing coastal and
marine ecosystem services. Review of Environmental Economics
and Policy, 6: 1– 19.
Bateman, I. J., Brouwer, R., Ferrini, S., Schaafsma, M., Barton, D. N.,
Dubgaard, A., Hasler, B., et al. 2011a. Making benefit transfers
work: deriving and testing principles for value transfers for similar
and dissimilar sites using a case study of the non-market benefits
of water quality improvements across Europe. Environmental and
Resource Economics, 50: 365– 387.
Bateman, I. J., Harwood, A. R., Mace, G. M., Watson, R. T., Abson, D. J.,
Andrews, B., Binner, A., et al. 2013. Bringing ecosystem services into
economic decision-making: land use in the United Kingdom.
Science, 341: 45–50.
Bateman, I. J., Mace, G. M., Fezzi, C., Atkinson, G., and Turner, K.
2011b. Economic analysis for ecosystem service assessments.
Environmental and Resource Economics, 48: 177– 218.
Bennett, E. M., Peterson, G. D., and Gordon, L. J. 2009. Understanding
relationships among multiple ecosystem services. Ecology Letters,
12: 1394–1404.
Bertram, C., Dworak, T., Go
¨rlitz, S., Interwies, E., and Rehdanz, K. 2014.
Cost– benefit analysis in the context of the EU Marine Strategy
Framework Directive: the case of Germany. Marine Policy, 43:
307–312.
Bertram, C., and Rehdanz, K. 2013. On the environmental effectiveness
of the EU Marine Strategy Framework Directive. Marine Policy, 38:
25–40.
Bo
¨rger, T., Beaumont, N. J., Pendleton, L., Boyle, K. J., Cooper, P.,
Fletcher, S., Haab, T., et al. 2014. Incorporating ecosystem services
in marine planning: the role of valuation. Marine Policy, 46:
161–170.
Bouma, J. A., and van Beukering, P. J. H. 2015. Ecosystem Services -
From Concept to Practice. Cambridge University Press,
Cambridge, UK. 287 pp.
Boyd, J., and Banzhaf, S. 2007. What are ecosystem services? The
need for standardized environmental accounting units. Ecological
Economics, 63: 616–626.
Brouwer, R., Barton, D., Bateman, I., Brander, L., Georgiou, S.,
Martı
´n-Ortega, J., Navrud, S., et al. 2009. Economic Valuation
of Environmental and Resource Costs and Benefits in the Water
Framework Directive: Technical Guidelines for Practitioners.
Research Report, AquaMoney.
Brouwer, R., Brander, L., Kuik, O., Papyrakis, E., and Bateman, I. 2013.
A synthesis of approaches to assess and value ecosystem services in
the EU in the context of TEEB. VU University Amsterdam. http://
lukebrander.com/wp-content/uploads/2012/05/Brouwer-et-al-2013-
A-synthesis-of-approaches-to-assess-and-value-ecosystem-services-
in-the-EU-in-the-context-of-TEEB.pdf (last accessed 15 October
2015).
Bystro
¨m, O. 2000. The replacement value of wetlands in Sweden.
Environmental and Resource Economics, 16: 347– 362.
Chan, K. M. A., Satterfield, T., and Goldstein, J. 2012. Rethinking
ecosystem services to better address and navigate cultural values.
Ecological Economics, 74: 8–18.
Clifton, J., Cullen-Unsworth, L. C., and Unsworth,R. K. F. 2014. Valuing
and evaluating marine ecosystem services: putting the right price on
marine environments? Environment and Society: Advances in
Research, 5: 66– 85.
DEFRA. 2007. An Introductory Guide to Valuing Ecosystem Services.
Department for Environment, Food and Rural Affairs, London.
de Groot, R. S., Wilson, M. A., and Boumans, R. M. J. 2002. A typology
for the classification, description and valuation of ecosystem func-
tions, goods and services. Ecological Economics, 41: 393–408.
Eggert, H., and Olsson, B. 2009. Valuing multi-attribute marine water
quality. Marine Policy, 33: 201– 206.
Elofsson, K. 2003. Cost-effective reductions of stochastic agricultural
loads to the Baltic Sea. Ecological Economics, 47: 13 –31.
Elofsson, K. 2010a. The costs of meeting the environmental objectives
for the Baltic Sea: a review of the literature. AMBIO, 39: 49– 58.
Elofsson, K. 2010b. Cost-effectiveness of the Baltic Sea Action Plan.
Marine Policy, 34: 1043– 1050.
EMEP. 2013. EMEP Centres Joint Report for HELCOM—Atmospheric
Supply of Nitrogen, Lead, Cadmium, Mercury and Dioxines Furanes
to the Baltic Sea in 2011. Technical Report, 2/2013. HELCOM, Oslo.
http://www.helcom.fi/Lists/Publications/Atmospheric%20Supply
%20of%20Nitrogen,%20Lead,%20Cadmium,%20Mercury%20and
%20Dioxines%20Furanes%20to%20the%20Baltic%20Sea%20in%
202011.pdf (last accessed 18 September 2015).
European Union. 2008. Directive 2008/56/EC of the European
Parliament and of the Council of 17 June 2008 establishing a frame-
work for community action in the field of marine environmental
policy (Marine Strategy Framework Directive). Official Journal of
the European Union, 51: 19–40.
Eurostat. 2015a. Comparative price levels of final consumption by
private households including indirect taxes (EU28¼100). http://
ec.europa.eu/eurostat/tgm/table.do?tab=table&init=1&language=
en&pcode=tec00120&plugin=1 (last accessed 29 August 2015).
Eurostat. 2015b. Harmonized Indices of Consumer Prices (HICPs). http://
ec.europa.eu/eurostat/tgm/table.do;jsessionid=5v8FxxTRfC8dtw5lx
DRk2NwI_r34Gbwp-hJ4d5xViWgu3c7gvvFD!-63100377?tab=table&
Economic valuation of Baltic marine ecosystem services 1001
plugin=1&language=en&pcode=teicp000 (last accessed 29 August
2015).
Fisher, B., and Turner, R. K. 2008. Ecosystemservices: classification for
valuation. Biological Conservation, 141: 1167– 1169.
Fisher, B., Turner, R. K., and Morling, P. 2009. Defining and classifying
ecosystem services for decision making. Ecological Economics, 68:
643–653.
Freeman, A. M. 2003. The Measurement of Environmental and
Resource Values: Theory and Methods. Resources for the Future,
Inc., Washington, DC.
Gilvear, D. J., Spray, C. J., and Casas-Mulet, R. 2013. River rehabilitation
for the delivery of multiple ecosystem services at the river network
scale. Journal of Environmental Management, 126: 30 –43.
Gren, I-M. 2001. International versus national actions against nitrogen
pollution of the Baltic Sea. Environmental and Resource Economics,
20: 41–59.
Gren, I-M. 2008a. Costs and benefits from nutrient reductions to the
Baltic Sea. Costs and benefits from nutrient reductions to the
Baltic Sea Ing-Marie Gren Valuable comments are obtained from
Katarina Elofsson. Swedish Environmental Protection Agency.
Gren, I-M. 2008b. Adaptation and mitigation strategies for controlling
stochastic water pollution: an application to the Baltic Sea.
Ecological Economics, 66: 337–347.
Gren, I-M., So
¨derqvist, T.,and Wulff, F. 1997. Nutrient reductions to the
Baltic Sea: ecology, costs and benefits. Journal of Environmental
Management, 51: 123–143.
Gren, I-M., Turner, R. K., and Wulff, F. 2000. Managing a Sea: the
Ecological Economics of the Baltic. Earthscan, London.
Haab, T. C., and MacConnell, K. E. 2002. Valuing Environmental and
Natural Resources: the Econometrics of Non-market Valuation.
Elgar, Cheltenham.
Haggett, C. 2011. Understanding public responses to offshore wind
power. Energy Policy, 39: 503–510.
Hanley, N., and Barbier, E. B. 2009. Pricing Nature: Cost–Benefit
Analysis and Environmental Policy. Edward Elgar Publishing,
Cheltenham, UK. 361 pp.
Hein, L., van Koppen, K., de Groot, R. S., and van Ierland, E. C. 2006.
Spatial scales, stakeholders and the valuation of ecosystem services.
Ecological Economics, 57: 209–228.
HELCOM. 2010. Ecosystem Health of the Baltic Sea 2003– 2007:
HELCOM Initial Holistic Assessment.
HELCOM. 2013. Summary report on the development of revised
Maximum Allowable Inputs (MAI) and updated Country
Allocated Reduction Targets (CART) of the Baltic Sea Action Plan.
Baltic Marine Environment Protection Commission, Copenhagen.
http://www.balticnest.org/download/18.7060460140fffe0c07a2e/
1381789016719/Summary+report+on+the+development+of+revised+
maximum+allowable+inputs+%28MAI%29+and+updated+country+
allocated+reduction+targets+%28CART%29+of+the+HELCOM+
BSAP_3October2013.pdf.
HELCOM, and NEFCO. 2007. Draft final report on economic analysis
of the BSAP with focus on eutrophication.
Helin, J., Laukkanen, M., and Koikkalainen, K. 2006. Abatement costs
for agricultural nitrogen and phosphorus loads: a case study of
crop farming in south-western Finland. Agricultural and Food
Science, 15: 351–374.
Hirschfeld, J., and Sagebiel, J. 2014. Holistische Bewertung von
O
¨kosystemleistungen-A
¨pfel, Birnen und Biodiversita
¨t. O
¨kologisches
Wirtschaften, 29: 25–26.
Hjorth, R. 1992. Building international institutions for environmental
protection: the case of Baltic Sea environmental cooperation. http://
www.diva-portal.org/smash/record.jsf?pid=diva2%3A255954&dswid=
467 (last accessed 15 December 2014).
Ivehammar, P. 2009. The payment vehicle used in CV studies of envir-
onmental goods does matter. Journal of Agricultural and Resource
Economics, 34: 450–463.
Johnston, R. J., Rolfe, J., Rosenberger,R. S., and Brouwer, R. (Eds). 2015.
Benefit Transfer of Environmental and Resource Values. The
Economics of Non-Market Goods and Resources. Springer
Netherlands, Dordrecht.
Kandziora, M., Burkhard, B., and Mu
¨ller, F.2013. Interactions of ecosys-
tem properties, ecosystem integrity and ecosystem service indica-
tors—a theoretical matrix exercise. Ecological Indicators, 28: 54 –78.
Kealy, M. J., and Turner, R. W. 1993. A test of the equality of
closed-ended and open-ended contingent valuations. American
Journal of Agricultural Economics, 75: 321–331.
Kiirikki, M. 2003. Cost effective water protection in the Gulf of Finland:
focus on St. Petersburg. Finnish Environment Institute, Helsinki.
Kosenius, A-K. 2010. Heterogeneous preferences for water quality attri-
butes: the case of eutrophication in the Gulf of Finland, the Baltic
Sea. Ecological Economics, 69: 528–538.
Ladenburg, J., and Dubgaard, A. 2007. Willingness to pay for reduced
visual disamenities from offshore wind farms in Denmark. Energy
Policy, 35: 4059– 4071.
Ladenburg, J., and Dubgaard, A. 2009. Preferences of coastal zone user
groups regarding the siting of offshore wind farms. Ocean and
Coastal Management, 52: 233–242.
Larsson, M. H., Kyllmar, K., Jonasson, L., and Johnsson, H. 2005.
Estimating reduction of nitrogen leaching from arable land and
the related costs. AMBIO, 34: 538–543.
Lindhjem, H., and Navrud, S. 2011. Using Internet in stated preference
surveys: a review and comparison of survey modes. International
Review of Environmental and Resource Economics, 5: 309– 351.
Liu, X., and Wirtz, K. W. 2010. Managing coastal area resources by stated
choice experiments. Estuarine, Coastal and Shelf Science, 86:
512–517.
Louviere, J. J., Hensher, D. A., Swait, J. D., and Adamowicz, W. 2006.
Stated Choice Methods: Analysis and Applications. Cambridge
University Press, Cambridge.
Markowska, A., and Z
˙ylicz, T. 1999. Costing an international public
good: the case of the Baltic Sea. Ecological Economics, 30: 301 –316.
Matzdorf, B., and Meyer, C. 2014. The relevance of the ecosystem ser-
vices framework for developed countries’ environmental policies:
a comparative case study of the US and EU. Land Use Policy, 38:
509–521.
Mewes, M. 2012. Diffuse nutrient reduction in the German Baltic Sea
catchment: cost-effectiveness analysis of water protection measures.
Ecological Indicators, 22: 16–26.
Meyerhoff, J., and Liebe, U. 2008. Do protest responses to a contingent
valuation question and a choice experiment differ? Environmental
and Resource Economics, 39: 433– 446.
Meyerhoff, J., Oehlmann, M., and Weller, P. 2015. The influence of
design dimensions on stated choices in an environmental context.
Environmental and Resource Economics, 61: 385– 407.
Millennium Ecosystem Assessment. 2005. Ecosystems and Human
Well-being: Synthesis. Island Press, Washington, DC.
Morrison, M. D., Blamey, R. K., and Bennett, J. W. 2000. Minimising
payment vehicle bias in contingent valuation studies.
Environmental and Resource Economics, 16: 407– 422.
Natural England. 2015. Monitor Engagement with the Natural
Environment: Technical report from the 2013– 2014 survey—
JP010. http://publications.naturalengland.org.uk/publication/
4750201384337408?category=47018 (last accessed 1 September
2015).
Nelson, J. P., and Kennedy, P. E. 2009. The use (and abuse) of
meta-analysis in environmental and natural resource economics:
an assessment. Environmental and Resource Economics, 42:
345–377.
Nunes, P. A. L. D., Kumar, P., and Dedeurwaerdere, T. 2014.
Introduction. In Handbook on the Economics of Ecosystem
Services and Biodiversity, pp. 1– 14. Ed. by P. A. L. D. Nunes, P.
Kumar, and T. Dedeurwaerdere. Edward Elgar Publishing,
Northampton, MA.
1002 J. Sagebiel et al.
OANDA. 2015. Historical Exchange Rates. http://www.oanda.com/
currency/historical-rates/(last accessed 4 November 2014).
Oehlmann, M., Weller, P., and Meyerhoff, J. 2014. Complexity-induced
status quo effects in discrete choice experiments for environmental
valuation. Annual General Meeting of the “Verein fu
¨r
Socialpolitik” (Association for Social Politics) 2014. Session:
Environmental Economics II, No. G01-V1.
Ollikainen, M., and Honkatukia, J. 2001. Towards efficient pollution
control in the Baltic Sea: an anatomy of current failure with sugges-
tions for change. AMBIO, 30: 245–253.
O
¨stberg, K., Hasselstro
¨m, L., and Ha
˚kansson, C. 2012. Non-market
valuation of the coastal environment—uniting political aims, eco-
logical and economic knowledge. Journal of Environmental
Management, 110: 166–178.
Remoundou, K., Koundouri, P., Kontogianni, A., Nunes, P. A. L. D., and
Skourtos, M. 2009. Valuation of natural marine ecosystems: an eco-
nomic perspective. Environmental Science and Policy, 12: 1040– 1051.
Ressurreic¸a
˜o, A., Zarzycki, T., Kaiser, M., Edwards-Jones, G., Ponce
Dentinho, T., Santos, R., and Gibbons, J. 2012. Towards an ecosys-
tem approach for understanding public values concerning marine
biodiversity loss. Marine Ecology Progress Series, 467: 15– 28.
Schou, J. S., Neye, S. T., Lundhede, T., Martinsen, L., and Hasler, B. 2006.
Modelling cost-efficient reductions of nutrient loads to the Baltic Sea.
NERI technical report, 592. http://www.mare.su.se/dokument/
evaluation/Economy/Schou%20et%20al%20(2006).pdf(last accessed
24 February 2014).
Sen, A., Harwood, A. R., Bateman, I. J., Munday, P., Crowe, A., Brander,
L., Raychaudhuri, J., et al. 2014. Economic assessment of the
recreational value of ecosystems: methodological development and
national and local application. Environmental and Resource
Economics, 57: 233–249.
TEEB. 2010. The Economics of Ecosystems and Biodiversity: Ecological
and Economic Foundations. Ed. by Pushpam Kumar. Earthscan,
London.
Toivonen, A-L., Roth, E., Navrud, S., Gudbergsson, G., Appelblad, H.,
Bengtsson, B., and Tuunainen, P. 2004. The economic value of
recreational fisheries in Nordic countries. Fisheries Management
and Ecology, 11: 1– 14.
Wallace, K. J. 2007. Classification of ecosystem services: problems and
solutions. Biological Conservation, 139: 235–246.
Weller, P., Oehlmann, M., Mariel, P., and Meyerhoff, J. 2014. Stated and
inferred attribute non-attendance in a design of designs approach.
Journal of Choice Modelling, 11: 43–56.
Z
˙ylicz, T., Georgiou, S., and Bateman, I. 1995. Contingent valuation of
eutrophication damage in the Baltic Sea region.
Handling editor: Claire Armstrong
Economic valuation of Baltic marine ecosystem services 1003
