Paper presented at the 14th International Conference of the ACEI,
Vienna 6-9 July 2006
Economic Valuation of Protection of Archaeological
Artefacts in Great Aamose, Denmark
Preliminary version
Trine Bille*, Thomas Lundhede** and Berit Hasler***
Corresponding author: Trine Bille, e-mail: tbi@akf.dk
*Copenhagen Business School, IVS, Kilevej 14A, 3rd floor, DK 2000 Frederiksberg,
Denmark
** The Royal Veterinary and Agricultural University, Bülowsvej 17, DK-1870
Frederiksberg C, Denmark.
*** National Environmental Research Institute, Frederiksborgvej 399, PB 358, DK
4000 Roskilde, Denmark.
Abstract
This paper presents the economic valuation of a proposed nature and wetland
restoration project in a river basin area, Great Aamose, in Denmark. The aim of the
restoration project is the preservation of cultural heritage in the form of unique
archeological artefacts from Stone-Age villages and sacrificial sites, which are
presently buried within the topsoil. The artefacts are threatened by agricultural
cultivation and drainage, but can be preserved by wetland restoration. A wetland
restoration project will also enhance biological diversity in the area, as well as
landscape features and recreational opportunities. The valuation of the non-marketed
restoration effects in Great Aamose has been performed by using the choice
experiments (CE) method. The results indicate that the population has the highest
willingness to pay (WTP) for ensuring a permanent protection of archaeological
artefacts, compared to the other attributes used in the study, like biological diversity
and access for recreation.
1. Background and purpose
Economic valuation of non-marketed good has increased in scale and scope over the
last 40 years. The greatest number of studies has been conducted within
environmental economics, where a huge number of studies have been done. Far fewer
studies have been done within cultural economics, and even less has been done to
value cultural heritage and archeological sites. Noonan (2003) has made a metaanalytic review of contingent valuation studies done within cultural economics. His
meta-analysis contains data from 72 studies. 3 studies are dealing with archeological
sites, 26 with historical sites, and 7 with heritage. 10 studies are valuing museums
(e.g. Santagata and Signorello, 2000) and 3 studies were valuing theatres (e.g. Bille
Hansen, 1997). The rest of the studies were valuing the arts, broadcast, libraries and
sports. Most studies within this field have though been dealing with preservation of
historical buildings and monuments (like the Stonehenge, Maddison and Mourato,
2001). Navrud and Ready (2002) contains a collection of valuation studies dealing
with historical buildings, monuments and artifacts. Most of these studies show
substantial use values as well as non-use values attributed to cultural assets. Values
for users (visitors or residents) are invariably higher than for non-users, but a general
finding is that non-visitors benefits are positive (Navrud and Ready, 2002). One study
(Whitehead and Finney, 2003) are looking into the willingness to pay for submerged
historic shipwreck. They ask the question: How much are people willing to pay to
maintain shipwrecks in their pristine state? And they find that households are willing
to pay about $35 in a one-time increase in state taxes.
The study presented comprises valuation of a proposed nature and wetland restoration
project in a river basin area, Great Aamose (Store Åmose), situated within an open
and magnificent landscape in West Zealand.
Figure 1. The Great Aamose (Store Åmose) situated in the western part of Zealand
The aim of the restoration project is the preservation of cultural heritage and
archeological artefacts from Stone-Age villages and sacrificial sites, which are
presently buried within the soil profile. The artefacts are threatened by agricultural
cultivation and drainage, but would be preserved by wetland restoration. A wetland
restoration project would also enhance biological diversity in the area, with a positive
impact mainly on conditions for waterfowl, but also on landscape features and
recreational opportunities.
The purpose of the valuation study presented here is to estimate how much the Danish
population is willing to pay for the restoration of Great Aamose. The National Forest
and Landscape Agency and West Zealand County have proposed various scenarios
for the restoration involving permanent protection, thereby securing the future for
ancient artefacts, improving the biological diversity and improving recreational
possibilities in the area. In order to establish the value to society of such
improvements to Great Aamose, the valuation method Choice Experiment (CE) has
been used as the effects are non-marketed and do not have a price.
2. Introduction to the Great Aamose area
The Great Aamose area was originally one of the largest moor- and marshland areas
in the eastern part of Denmark. However, due to cultivation, drainage and regulation
of the Aamose stream, the area has become dry, and a large part of the area is now
under agricultural production (largely not included in a crop rotation). The moor and
marsh area has been cultivated over the past 250 years, and the latest and most
dominant changes were implemented in the 1950s and 1960s, when the stream was
converted to a channel and the area was heavily drained.
The area of Great Aamose is now undeveloped and uninhabited, but in the hunting
and agrarian Stone-Age period, the area was an attractive place to live. From a
cultural historic point of view, the area has a certain, unique status, and the area is
internationally well-known for its well protected Stone-Age artefacts. These artefacts,
buried in the wet soil profile, comprise the remains of villages; skeletons from
humans, domestic and wild animals; fishing and hunting tools; sacrificial objects and
places; as well as ornaments. The soil and the groundwater, which has been at a
constant, high level in the area, preserve the artefacts. A large number of different
artefacts has already been recovered from the area and exhibited in museums. The
main body of the archaeology, however, still remains in the soil and the cultivation
and drainage of the area threaten its endurance. Securing the remaining artefacts for
future generations requires restoration of the wetland in the area and this can be
achieved by dismantling the drainage systems. Especially in the eastern part of the
area in question, settlements of international interest have been identified very close
to the cultivated layer of the soil. In the same area, artefacts are also in evidence in the
deeper layers of the soil profile. In other parts of Great Aamose, there is still 0.5 to 1m
of soil above the artefacts.
The idea behind the conservation measure is to secure access to the artefacts for future
generations and, thus, good material for future archaeological research. In the future,
researchers will most certainly ask other questions about the past than researchers at
present, and they will have access to research techniques other than those at our
disposal today.
3. The scenarios
The Danish Forest and Nature Agency and West Zealand County have proposed
various scenarios for the restoration of Great Aamose, and the characteristics of the
scenarios have been used for the valuation (Danish Forest and Nature Agency, 2001).
Three restoration scenarios are proposed. The area of Great Aamose currently under
protection is 230 hectares, and the scenarios for further designation and restoration
include areas in the range of approximately 600 to 1,500 hectares. The scenarios will
all, but to differing degrees, result in permanent protection of the unique, unearthed
ancient artefacts and cultural values in Great Aamose, securing them for the
generations to come. This will be achieved by wetting the area – raising the water
level will reduce or stop the present destruction of ancient artefacts. In other words,
the artefacts will not be visible to visitors in the area, but visitors and non-visitors will
know that the artefacts are preserved, and information about this will be presented in
the area. The restoration will also result in improved conditions for animal- and plantlife in the area and improved conditions will be secured for several endangered
species. Finally, restoration may also result in better recreational possibilities in the
area, partly by providing the area with a higher discovery value, when nature and
ancient artefacts are protected, and partly by extending the current path and road
system in the area. The various restoration scenarios differ with regard to e.g. the
extent of area restored, the extent of the improvement in conditions for biodiversity
and unique ancient artefacts, and the degree of public access and recreational
possibilities in the local area.
4. The Choice Experiment method (CE)
The CE method was, like other choice modelling techniques, developed for market
analysis (See eg. Louviere 2000), but the method have been increasingly used and
further developed for the valuation of non-marketed goods (Adamowicz, 1995, Boxall
et all 1996, Hanley et al. 1998a, Hanley et al. 1998b, Hanley et al. 2001).
The CE method presents respondents with a number of choices, presented as choice
sets, and the choice situation is created to resemble the market situations that
respondents are used to in everyday life. In order to allow the valuation to be
approached as a market choice, the CE method describes public goods in terms of
attributes, defining the goods and attribute levels, following Lancaster (1966).
In a CE survey, respondents are requested to choose between pre-defined alternatives,
each of these representing different levels of the attributes. In the present study, the
attributes are preservation of artefacts and biodiversity, recreational possibilities
through access to the area (tracks and parking facilities), size of the area and the price
the respondent is willing to pay. In the present study, the price is expressed as an
amount reflecting an increase in annual tax payment, per individual.
The respondents are requested to select their preferred alternative and, in contrast to
the much used Contingent Valuation method, the term “indirect method” is used as
consumer preferences are estimated on the basis of preferred situations and not on the
basis of actual expressed willingness to pay. Respondents are, hereby, provided with
an explicit basis for assessing costs in relation to effects and, therefore, the method is
recommended for valuation of complex problems, where the good is characterised by
several “attributes”. The method is also suitable if the nature of the environmental
good is relatively removed from characteristics possessed by traditional consumer
goods, because the choice situation places the valuation in a situation more reflective
of real market conditions than is the case with other forms of valuation exercise.
Consequently, the power of the CE method is that it is split into attributes and choice
sets, and can avoid response difficulties, reduce problems of multicollinearity and
measure the marginal value of changes.
The method can be described formally by the following utility function. An individual
i’s utility from a good j (Uij ) can be described as a function of a deterministic part (V)
and a stochastic element (ε ) as follows:
(1) Uij = Vij(xj, zi ) + εij
Person i’s utility from alternative j depends on the deterministic part that contains a
vector of the alternative specific characteristics xj, inclusive the price of the alternative
j, as well as a vector for the individual specific characteristics and a stochastic part, εij
(cf. Adamowicz et al. 1994; Bateman et al. 2002).
The probability of a choice between alternative options for changes in preservation is
described as a function of the attributes and the probability for each choice between
the alternatives is analysed by random utility models.
The probability of an alternative being chosen can be expressed in terms of the
logistic distribution and, depending on the nature of the data, different logit models
can be applied (Train, 2003). A multinomial logit model is applied in the present
analysis, and willingness to pay for the non-monetary attributes is estimated as the
marginal rate of substitution between each of the attributes and the monetary attribute:
(2) Willingness to pay = -a/p
where a is the parameter of attribute a, and p is the price-parameter.
5. The design of the study and the questionnaires
5.1. The attributes
As mentioned above, the size of area, the extent of improvement in biodiversity, level
of protection of the unique archaeological artefacts in the wet soil, and degree of
public access and recreational possibilities in the area are all used as attributes in the
valuation.
Quantitative attribute levels have some obvious advantages as they can be connected
to dose-response functions associated with the effects of the wetland restoration and
other management options in the area. Such dose-response functions could, for
instance, be used to estimate e.g. the number of artefacts or number of waterfowl
protected per hectare wetland restored. However, it has not been possible to obtain
such dose-response functions regarding the available data on the effects. Qualitative
formulations have therefore been used, describing the attribute levels in more or less
the same wordings as in the scenarios formulated by the end-users of the result, as far
as these were understandable for the respondents. The qualitative approach is used in
many former studies (cf Bergström & Dorfman 1994, Boiesen et al. 2005, Hanley et
al. 2005, Hasler et al. 2005, Stenger & Willinger 1998).
For describing preservation of the artefacts, the following attribute levels have been
chosen:
 Continued devastation: Important artefacts, including internationally unique
sacrificial and settlement sites will be destroyed within a number of years.
 Reduced devastation: The tempo of the devastation is reduced significantly for
some of the most important artefacts, but the devastation is not brought to an end.
 Protection now and in the future: The artefacts will be protected within the soil,
now and in the future, so that they can be excavated and exhibited in future.
The respondents were presented with the following information: “Great Aamose is
well known internationally for its numerous and well-preserved artefacts from the
Stone-Age, including both the hunter-gatherer and the agrarian periods. The artefacts
comprise settlement remains, angling tools, other tools, skeletons from humans and
animals, sacrificial sites and ornaments. Some of the artefacts have already been
excavated and exhibited in museums, but the largest part is still buried in the soil
profiles. A considerable portion of these hidden artefacts is now close to devastation
due to drainage activities as part of agricultural production. If the artefacts are to be
protected for the future, the water-table has to be raised.”
The description used for the effects concerning flora and fauna is as follows:
 Low diversity: A large number of animals, but distributed largely within a
limited number of common species. Vulnerable and rare species are threatened
by distinction due to dry conditions and cultivation.
 Some diversity: Many animals distributed within a greater number of common
species, including small birds. Vulnerable and rare plants are threatened by
distinction only in few places.
 High diversity: Many common and rare animal species, especially bird species.
Rare species are protected against dry conditions and cultivation.
The respondents were further informed that: “Animal life in Great Aamose mostly
comprises common species such as deer, pheasant, hare and wading birds as well as
duck species. In parts of Great Aamose, there are plants which are rare on Zealand,
for example heather plants. The rare plant species are threatened by conditions which
are too dry.”
The concept of “biodiversity” had the benefit that it does not describe only one
species or some chosen species, but describes nature quality in more general terms.
Some earlier studies indicate that the willingness to pay for one single species is much
larger than for nature quality in general (cf Boisen et al. 2005), but results from other
studies are more mixed (cf Veisten et al. 2004).
Access to the area is described as:
 Restricted access via a few tracks and paths
 Extended access to a larger part of the area via a wider network of paths and
tracks
The respondents were informed that the public today has limited access to the area by
foot or bicycle, that there is a narrow track into the area, and that there is only a total
of 2 km’s path in that part of the area, which is currently under nature protection.
The areal extent can change incrementally through 6 levels, from 230 to 1,750
hectares, and the respondents were informed that: “The extent of the present area
under nature protection is 230 hectares. One hectare is 100 x 100 metres, i.e. the
approximate size of a football ground. The total area under nature protection could
extend to up to 1,750 hectares. For comparison, the Skjern River Basin in the western
part of Jutland is in the region of 2,200 hectares.” The Skjern River Basin is
mentioned as this area is well known to people all over Denmark.
The choice of payment vehicle forms a substantive part of the survey design, and
should have a plausible connection with the good it is being used to value (Garrod &
Willis, 1999). Hence, the payment vehicle must be perceived as realistic, fair and
equitable for all respondents to prevent non-responses and protest responses. Entrance
payment could be used, but would delimit the analysis to use values. According to the
results of our focus group tests, an annual tax was found to be a reliable payment
vehicle. In this connection, it should be noted that annual payments have been chosen
as opposed to a one-off payment. In terms of the levels of the cost attribute, 6 levels
were applied, stated in DKK. The respondents were informed that the costs of
implementing the policy alternatives are assumed to be covered by the Danish
consumers and that all consumers will contribute equally to implementation of the
alternatives. Moreover, it is emphasised that the stated amount (willingness to pay)
represents a sum over and above their present tax payment. This text is intended to
induce respondents to provide as valid and reliable responses as possible, and to
discourage exhibition of strategic behaviour. Highlighting that all consumers must
contribute, and that they must do so equally, is intended to discourage respondents
from free-riding. Likewise, stipulating that the stated amounts are additional to the
current tax payment is considered relevant, as it may serve to remind respondents that
the amount they are asked to pay in the present survey is in fact only one among many
expenses that they have to consider in their day-to-day lives.
Furthermore, a so-called “cheap talk” section is included. According to Cummings
and Taylor (1999:650), who introduced the cheap talk concept, cheap talk may be
defined as an attempt to eliminate hypothetical bias by including an explicit
discussion of the problem. Cummings and Taylor (1999) found that cheap talk could
effectively mitigate hypothetical bias in the cases considered. Subsequently, the effect
of different cheap talk designs, implemented in various contexts, has been
investigated in a number of studies, and the experiences hereof have been mixed.
While Carlsson et al. (2004) and Murphy et al. (2003) found the inclusion of a cheap
talk section to have a positive effect, Samnaliev et al. (2003) found it had no effect.
The cheap talk used in this study is presented below:
“Results from similar studies have shown that people have a tendency to overestimate how much they are actually willing to pay for implementation of the various
policy measures. Before you mark your selection, therefore, we would ask you to be
entirely sure that you are willing and able to pay the stated sum associated with an
alternative.”
Despite the inconclusive effect of cheap talk, it has been decided to incorporate the
above-mentioned cheap talk passage in the design of the present study. Compared
with other studies it should, however, be emphasised that the script used in the present
study is significantly shorter than usually applied. As an example, the cheap talk in
Cummings’ and Taylor’s (1999) article amounts to more than half a page. In relation
to the present study, the inclusion of such an elaborate cheap talk section is
considered inappropriate, as it is expected that the resulting negative effect arising
from increasing the length of the questionnaire would by far outweigh the potential
positive effect arising from the cheap talk, itself.
5.2. The choice sets
Consulting the literature, it is often recommended, sometimes even required, that a
status-quo alternative, or an “opt-out” option, is included in the design, as failure to
do so may imply that respondents are forced to choose alternatives, which they do not
desire (cf. Bateman et al, 2002). If this is the case, the observed choices should not be
interpreted as expressions of respondents’ true preferences, implying that they cannot
be used as the basis for deriving valid estimates of welfare changes. A status-quo
alternative, describing the baseline situation that will prevail if current initiatives are
maintained and if no further action is taken, is therefore included in the present study.
This alternative is characterised by “continued devastation”, “low diversity” and
“restricted access” to the 230 hectare area.
Apart from the status quo alternative, which features in all choice sets, each choice set
contains 2 alternatives. According to Bateman et al. (2002), it is important to ensure
that respondents are not asked to perform too complex tasks, as this may induce
respondents to provide unreliable answers or resort to using simplifying decision
strategies instead of the compensatory decision strategies, which are assumed in CE.
In the present context, it is considered appropriate to operate with a choice set size of
3 alternatives per set. The combination of the alternatives into choice sets is
performed using a fractional factorial design, which comprises a part of the potential
choice sets. The design is orthogonal, which means that the attribute levels vary
independently and without correlation between the levels and the attributes. The
design is also balanced, i.e. each attribute levels and the combinations of them are
represented with the same frequency. The fractional factorial design contains 18
choice sets, divided into 3 blocks of 6 choice sets. Each respondent was then
presented with one block. An example of a choice set is presented in Figure 2.
Figure 2. A Choice Set.
Please choose one of the following three options:
Present situation
Alternative 1
Alternative 2
The area of the
protected area in
hectares
230
230
1,250
Biodiversity
Low
Low
Some
Protection of the
ancient artefacts
Continued
devastation
Reduced devastation
Protection now and in
the future
Public access
Restricted
Extended
Restricted
Extra payment in
annual tax
0 DKK
225 DKK (30 EUR)
850 DKK (113 EUR)
I prefer
(please only tick one of the boxes)



6. The data collection and tests
The final questionnaire contained both the choice questions and questions about the
respondents attitudes and habits.
Three focus groups were used to test the questionnaire before submission, especially
the choice options, the wording of the questions and the background information,
and these tests provided valuable information on the understanding of the attributes,
the choices and the information presented. One of the experiences gained was that
the cognitive understanding of biodiversity and nature quality is much more difficult
than the understanding of artefacts and their preservation, even though these
artefacts are hidden and most likely will never be seen by the respondents.
The survey was sent out in the period 26th of May to the 3rd of June 2005. The
professional survey institute GALLUP was used for the sampling of data, using
GALLUP’s internet panel, which contains about 35,000 individuals. Overall, the
survey was sent out to 3,200 individuals, and 1,636 respondents had answered after
one week, the cut-off imposed for response to the survey. This resulted in the
requested number of responses and a response rate of around 51 percent.
A positive feature of using a professional institute is the ready-selected, representative
panel. The experience with the internet solution, moreover, turned out to be positive
in other ways. First of all, our experience is that the internet survey was less
cognitively demanding than a mail questionnaire, as the questions and the choice sets
are presented individually. Furthermore, the possibilities for presenting information,
including photographs, are good. The survey mode also turned out to be as
representative as other survey modes – elderly people, over 70, are not represented in
the sample, but experience reveals that this age group is also highly underrepresented
in postal surveys (Hasler et al 2005; Olsen & Lundhede 2005). Also, the survey could
be performed in one week’s time using the internet, which is fast compared with
postal surveys. This proved advantageous where the survey had to be finished within
a short time-span. Moreover, the budget was restricted and the internet solution was
the cheapest alternative compared with various postal survey options. Last, but not
least, the internet survey made it possible to perform a pre-test on the first batch of
data collected. After testing the data from this first batch of 100 respondents, the
survey was continued the same day.
7. The results
7.1. The Willingness to Pay
The willingness to pay for protection of the artefacts, access and nature restoration is
analysed using multinomial logit models, cf. Section 5. The attributes size of the area,
access and price are quantitative variables, while the qualitative attributes biodiversity
and protection of artefacts are dummy variables. The parameters for these variables
should be interpreted as the utility of the change from the present situation to the
protection level presented. From Table 1, it can also be seen that there is a so-called
alternative specific constant, which is a dummy-variable, describing preferences for
the status quo situation which are not described by the presently chosen attributes. In
general, it is recommended that this constant should be included (Bennet &
Adamowicz, 2001).
The estimated parameters in the main effect model are apparent from Table 1.
Table 1. Parameter estimates for the main effect model
Parametervalue
Alternative specific constant
Price (annual tax)
Some biodiversity
High biodiversity
Reduced devastation of artefacts
Protection of the artefacts now and in the future
Extended access
Area (DKK per 100 ha)
N
Log L
2
Adjusted pseudo R2
-0.73852
-0.00095
0.05189
0.51534
0.79738
1.13179
-0.12477
0.01498
9,060
-7,718
4,470.85
0.224
Standard
deviation
0.05735
0.00003
0.03543
0.03533
0.03705
0.03760
0.02638
0.00312
P-value
0.00
0.00
0.14
0.00
0.00
0.00
0.00
0.00
WILLINGNESS
TO PAY
DKK per person
per year
55 (7,3 EUR)
543 (72 EUR)
840 (112 EUR)
1,192 (159 EUR)
-131 (-17,5 EUR)
16 (2,1 EUR)
From Table 1, it can be seen that the parameter “some biodiversity” is insignificant,
whereas all the other parameters differ significantly from zero. The results show that
the population has a significantly positive willingness to pay when it comes to
reducing the destruction of ancient artefacts which at the moment are lying in the soil
profile of Great Aamose and threatened by destruction by the current use of the moorand marshland. This willingness to pay is estimated at approximately DKK 800 (107
EUR) per person in additional annual taxation. The willingness to pay for ensuring a
permanent protection of such ancient artefacts is estimated at approximately DKK
1,200 (160 EUR) per person per year.
Furthermore, the results show that the population has a significantly positive
willingness to pay for high biodiversity in Great Aamose, amounting to
approximately DKK 500 (67 EUR) per year in additional tax per person. This
willingness to pay covers changing the currently “low biodiversity of Great Aamose”
to a higher level described as “high biodiversity”. Low diversity (the present
situation) is described as “many animals distributed on a few, common species, while
vulnerable and rare plants are threatened by reclamation and cultivation”. A high
diversity is described as “many common and rare species, in particular species of
birds, and rare plants are protected against reclamation and cultivation«.
The analysis also indicates a positive willingness to pay for additional restored area.
The willingness to pay for extension of the area restored is estimated at approximately
DKK 16 (2 EUR) per 100 additional hectares. The willingness to pay is, also in this
case, calculated as a annual willingness to pay, per person. The result indicates that
the willingness to pay increases with the size of area.
The estimate of the willingness to pay for access to Great Aamose via an extended
path and road system in the area is, however, significantly negative. This means that,
on average, the respondents experience a non-benefit (e.g. a negative benefit) by
establishing further access to the area over and above the existing path and road
system. This negative benefit may stem from the fact that the respondents find that
new paths and roads in the area correlate negatively with the conditions for plant- and
animal-life as well as the protection of ancient artefacts. Besides, the majority of the
respondents did not have any prior knowledge of Great Aamose and do not intend to
visit Great Aamose in future, even if the area is restored.
As anticipated, the picture is quite different when looking at the section of the
population that lives relatively close to Great Aamose. This section of the population
has a positive willingness to pay for increased access to Great Aamose of
approximately DKK 430 (57 EUR) per year per person.
The relationships involved in the estimated willingness to pay for ancient artefacts,
larger area, protection of biological diversity and access all seem plausible. The
average willingness to pay is largest when it comes to preserving the nationally and
internationally, unique, ancient artefacts in Great Aamose, whereas the estimated
willingness to pay for an increased biodiversity in the area is approximately half this
level. This smaller willingness to pay is plausible as the focus is not simply the
protection or creation of a unique biodiversity or the maintenance of species which
are not found elsewhere, but is equally the improvement of biodiversity in the specific
local area in question. Insofar as ancient artefacts and biodiversity are concerned, it is
largely existence value and bequest value which prevail, i.e. benefits to society in
general and not simply to the users of the area. Therefore, it makes sense that the
average willingness to pay for ancient artefacts and biodiversity is, by and large,
similar in the population as a whole to that of those people living in the area local to
Great Aamose.
Access to the area is a different matter. Here, it is use value which is important, and
this more so for the population living near Great Aamose. Therefore, it makes sense
that this section of the population has a positive willingness to pay for increased
access to the area, whereas the population of the rest of the county has a lower
willingness to pay, and perhaps even a negative willingness to pay.
7.2 The aggregate benefits
If the aggregate value of the three scenarios, proposed by the Danish Forest and
Nature Agency and West Zealand County, is calculated by multiplying the average
willingness to pay for the various effects in the three scenarios by the size of the
Danish population, the resulting values are very large. These results are shown in
Table 2, where the aggregate benefits for the entire population are calculated on the
basis of the estimated willingness to pay, as mentioned above – first by calculating the
estimated willingness to pay for each scenario per person and then for the entire
population.
Table 2. Aggregate benefits of scenarios for restoration of Great Aamose
Aggregate annual
willingness to pay per
person.
Aggregate benefits
per year.
Aggregate benefits
per year.
Aggregate benefits
per year.
1) Primary
assumptions, all
individuals over 18.
2) Sensitivity analysis
for all households.
3) Individuals in the
local area.
Scenario
1
approx. DKK 800 (107
EUR)
approx. DKK 3.4
billion (0.5 billion
EUR)
approx. DKK 2 billion
(0.27 billion EUR)
approx. DKK 31
million (4.1 million
EUR)
2
approx. DKK 1,500
(200 EUR)
approx. DKK 6.1
billion (0.8 billion
EUR)
approx. DKK 3.7
billion (0.5 billion
EUR)
approx. DKK 32
million (4.3 million
EUR)
3
approx. DKK 1,900
(253 EUR)
approx. DKK 7.8
billion (1 billion
EUR)
approx. DKK 4.6
billion (0.6 billion
EUR)
approx. DKK 42
million (5.6 million
EUR)
1) Number of individuals over 18 in the entire country according to Statistics Denmark: 4,184,858 as per
1st January 2005.
2) Number of households in the entire country according to Statistics Denmark: 2,498,621 as per 1st
January 2005.
3) Calculated for inhabitants in the counties Tornved, Jernløse, Tølløse, Stenlille and Dianalund (approx.
28,000 as per 1st January 2005).
As indicated in the table, the overall benefits of the three scenarios have been
calculated on the basis of three different underlying assumptions – i.e. for all adult
individuals in Denmark; for all households in Denmark, as sensitivity analysis; for
local inhabitants near Great Aamose. The aggregate benefit results from these three
calculation assumptions range from DKK 31 million – DKK 3.4 billion (4.1 million –
0.5 billion EUR) per year for Scenario 1; DKK 32 million - DKK 6.1 billion (4.3
million – 0.8 billion EUR) per year for Scenario 2; DKK 42 million – DKK 7.8
billion (5.6 million – 1 billion EUR) per year for Scenario 3. The question is to what
extent these amounts can be considered realistic for the population’s actual
willingness to pay for the restoration of Great Aamose. With an infinite time-frame
and a discount rate of either 3 percent or 6 percent, the calculated net present value is
very large, as the calculated aggregate benefits for each scenario are calculated on the
basis of the population’s annual willingness to pay.
The following causes may explain why the aggregate benefits are considered to be
overestimated:

Moral satisfaction resulting in a willingness to pay that is too high: Other types of
analyses have previously indicated various types of problems with this kind of
analysis, e.g. the “warm glow” phenomenon (see e.g. Kahneman and Knetsch,
1992). This implies that the respondents would like to pay for achieving the good
feeling of doing something worthwhile for the environment or cultural heritage.
The willingness to pay is, therefore, not so much linked to the actual
project/change, but is rather a positive expression of the respondents’ attitudes
towards more general problems. It is possible to imagine that this could be the
case, to a certain degree, in this analysis. The respondents are willing to pay for
preservation of ancient artefacts and for increased biodiversity. Via the
questionnaire, they are presented with the opportunity of expressing their
willingness to pay for these effects in the Great Aamose project. However, at the
same time they may have “emptied their accounts” with regard to the willingness
to pay for improvements of biodiversity and preservation of ancient artefacts, in
general.

A lack of understanding of the scope of the project: The questionnaire only covers
the population’s willingness to pay for restoration of Great Aamose and the
respondents were not asked whether they would give the same amount per
restoration project if there were 10 different restoration projects (e.g. the current
Danish national park projects) in various places of the country. This type of bias is
called embedding and has also been the subject of discussion in the literature. In
any case, willingness to pay is presumed to decline for each new restoration
project that is introduced.

The willingness to pay of individuals versus households: The questionnaire covers
the willingness to pay of the individual and, consequently, the number of adult
individuals in the entire country constitutes the correct aggregation basis.
However, it may be argued that the respondents had their overall household
budget in mind when answering the questions. The result is that the aggregate
willingness to pay for households is somewhat lower than that based on
individuals.

Annual payments versus one-off payments: In the analysis the respondents were
asked to state their annual willingness to pay. Their willingness to pay may not
have differed if they had been asked to state a one-off amount (an amount once
and for all), but this would have resulted in uncertainty concerning which implicit
discount factor the respondents use and, consequently, it was decided to use an
annual amount in the analysis. It can be argued that the respondents could be
expected to have stated their willingness to pay as a one-off amount and,
therefore, that the current values should be considered, as part of a sensitivity
analysis, as lower estimates.
It would be interesting to test the importance of the above-mentioned factors and their
effect on the results, but this was not possible within the framework of the present
study.
Overall, the conclusion is that the population is willing to pay a considerable amount
for restoration of an area such as Great Aamose with a view to preserving ancient
artefacts, improving biodiversity and access – in this order.
The willingness to pay for changes in the Great Aamose area, estimated in the present
study, has been used by the Danish Environmental Protection Agency in a Cost
Benefit Analysis of the effects of restoration of the area. Three scenarios were
considered, and the purpose of calculating the costs and benefits, and hereby the netbenefits, was to obtain a socially efficient basis for decisions on the degree to which
the moor- and marshland in Great Aamose should be protected and restored to
wetland. The direct costs and the costs of lost agricultural rents, as well as marketed
benefits from e.g. hunting and angling, were calculated by the Danish Environmental
Protection Agency for the purposes of the analyses. On the basis of the results of the
cost benefit analysis, as well as other analyses, the Minister of Environment decided
in Autumn 2005 to implement a restoration project in Great Aamose. The particular
restoration scenario has not yet been selected, but the project itself was initiated in the
beginning of 2006.
8. Conclusions
The results indicate that the population has a significantly positive willingness to pay
for ensuring a permanent protection of archaeological artefacts; approximately DKK
1,200 (160 EUR) per person per year in additional tax per person. Furthermore, the
population has a significantly positive willingness to pay for a major increase in the
biodiversity of Great Aamose, amounting to approximately DKK 500 (67 EUR) per
year in additional tax per person. The analysis also indicates a positive willingness to
pay for additional restored areas at about DKK 16 (2 EUR) per year per 100 hectares.
The willingness to pay for recreation and access to Great Aamose via an extended
path and road system in the area is, however, significantly negative. This may stem
from the fact that the respondents find that the extension of paths and roads in the area
correlates negatively with the conditions for plant- and animal-life, as well as the
protection of archaeological artefacts. Moreover, the majority of the respondents did
not have any prior knowledge of Great Aamose and did not intend visiting the area in
the future, even if the area was to be restored. As anticipated, however, the population
living relatively close to Great Aamose has a positive willingness to pay for increased
access, approximately DKK 430 (57 EUR) per person per year.
References
Adamowicz, W, J. Louviere and M. Williams (1994): Combining revealed and stated
preference methods for valuing environmental amenities. Journal of Environmental
Economics and Management. 26(3), pp. 271-92
Adamowicz, W. (1995): Alternative Valuation Techniques: A Comparison and
Movement to a Synthesis. In Willis, K.G. & J.T. Corkindale (eds.): Environmental
Valuation, Cab International.
Bateman, I.J., Carson, R.T., Day, B., Hanemann, M., Hanley, N., Hett, T., Jones-Lee,
M., Loomes, G., Mourato, S., Özdemiroglu, E., Pearce, D.W., Sugden, R. & Swanson,
J. (2002): Economic Valuation with Stated Preference Techniques. A Manual,
Edward Elgar Publishing, Cheltenham.
Bennet, J. & Adamowicz, V. (2001): Some Fundamentals of Environmental Choice
Modelling. In: Bennet, J. & R. K. Blamey (eds.): The Choice Modelling Approach to
Environmental Valuation. Edward Elgar Publishing Limited, UK. pp. 37-69.
Bergstrom, J.C. & Dorfman, J.F. (1994): Commodity Information and Willingness-toPay for Groundwater Quality Protection, Review of Agricultural Economics 16(3),
413-425.
Bille Hansen, Trine (1997): The Willingness-to-Pay for the Royal Theatre in
Copenhagen as a Public Good, Journal of Cultural Economics 21(1), 1-28
Boiesen, J., Jacobsen, J.B., Thorsen, B.J., Strange, N. & Dubgaard, A. (2005):
Værdisætning af de danske lyngheder. Working Paper, KVL, Frederiksberg.
Boxall, P.C., Englin, J. & Adamowicz, W.L. (2003): Journal of Environmental
Economics and Management, 45, 213-230
Carlsson, F., Frykblom, P. & Lagerkvist, C.-J. (2004): Using Cheap-Talk as a Test of
Validity in Choice Experiments, Working Papers in Economics no. 128, Department
of Economics, Gothenburg University.
Cummings, R.G. & Taylor, L.O. (1999): Unbiased Value Estimates for
Environmental Goods: A cheap Talk Design for the Contingent Valuation Method,
American Economic Review, 89(3):649-665.
Danish Forest and Nature Agency (2001): Åmosen – Vestsjællands Grønne Hjerte.
Handlingsplan for naturgenopretning og beskyttelse af kulturmiljøet i den østre del af
Store Åmose. Miljøministeriet, Skov- og Naturstyrelsen, København.
Garrod, G. & K.G. Willis (1999): Economic Valuation of the Environment. Methods
and Case Studies. Edward Elgar.
Hanley, N., Macmillan, D., Wright, R., Bullock, C., Simpson, I., Parrison, D. &
Crabtree, B. (1998a): Contingent Valuation versus Choice Experiments: Estimating
the benefits of Environmentally Sensitive Areas in Scotland. Journal of Agricultural
Economics 1(15).
Hanley, N., Wright, R.E. & Adamowicz, V., (1998b): Using Choice Experiments to
Value the Environment: Design Issues, Current Experience and Future Prospects.
Environmental and Resource Economics 11(3-4), pp. 413-28.
Hanley, N.D., Mourato, S. & Wright, R.E. (2001): Choice Modelling Approaches: A
Superior Alternative for Environmental Valuation? Journal of Economic Surveys
15(3), pp. 435-62.
Hanley N. & Wright, R.E. (2005): Estimating the economic value of improvements in
river ecology using choice experiments. Paper presented at the 14th annual meeting of
European Environmental Economists, Bremen.
Hasler B., Lundhede T., Martinsen L., Neye S., Schou J.S. (2005): Valuation of
protection and purification of groundwater by choice experiments and contingent
valuation methods. Technical Report from NERI nr. 543.
Kahneman, D. & Knetsch, J.L. (1992): Valuing Public Goods: The Purchase of Moral
Satisfaction, Journal of Environmental Economics and Management, vol. 22, no. 1,
January 1992, pp. 57-70.
Lancaster, K.J (1966): A New Approach to Consumer Theory. The Journal of
Political Economy. Vol 74, 2, pp. 132-157.
Louviere, J.J., Hensher, D.A. & Swait, J.D. (2000): Stated Choice Methods. Analysis
and Application. First edition, Cambridge University Press, Cambridge.
Maddison, D. & Mourato, S. (2001): Valuing Different Road Options for Stonehenge,
Conservation and Management of Archaeological Sites 4(4), 203-212
Murphy, J.J., Stevens, T. & Weatherhead, D. (2003): An Empirical Study of
Hypothetical Bias in Voluntary Contribution Contingent Valuation: Does Cheap Talk
Matter? Working Paper No. 2003-2, Department of Resource Economics, University
of Massachusetts Amherst.
Navrud, S. & Ready, R.C. (eds.) (2002): Valuing Cultural Heritage. Applying
Environmental Valuation Techniques to Historic Buildings, Monuments and
Artefacts, Edward Elgar, Northampton, MA
Noonan, D.S. (2003): Contingent Valuation and Cultural Resources: A Meta-Analytic
Review of the Literature, Journal of Cultural Economics, 27(3-4), 159-176
Olsen , S. B. & Lundhede, T. (2005): Rekreative værdier ved konvertering til
naturnær skovdrift. En værdisætningsundersøgelse udført vha. metoden Discrete
Choice Experiments. Specialerapport, akf-forlaget, Copenhagen
Santagata, W. & Signorello, G. (2000): Contingent Valuation of a Cultural Public
Good and Policy Design: The Case of “Napoli Musei Apert”, Journal of Cultural
Economics, 24(3) 181-203
Samnaliev, M., Stevens, T. & More, T. (2003): A Comparison of Cheap Talk and
Alternative Certainty Calibration Techniques in Contingent Valuation, Working Paper
No. 2003-11, Department of Resource Economics, University of Massachusetts
Amherst.
Stenger, A. & Willinger, M. (1998): Preservation value for groundwater quality in a
large aquifer: a contingent valuation study of the Alsatian aquifer. Journal of
Environmental Management, 53,99. 177-193.
Train, K.E. (2003) Discrete Choice methods with Simulation. cambridge. 334 pp.
Veisten, K., Hoen, H.F., Navrud, S. & Strand, J. (2004): Scope insensitivity in
contingent valuation of complex environmental amenities. Journal of Environmental
Management, 73. pp. 317-331
Whitehead, J.C. & Finney, S.S. (2003): Willingness to Pay for Submerged Maritime
Cultural Resources, Journal of Cultura Economics, 27(3-4), 231-240