A Climate for Change

(^66) Tourism Human Development Report - Croatia 2008
Box 4-3: Models to estimate the changes in visitation by tourists and tourist welfare
Estimating the changes in visitation to Croatia by
foreign and domestic tourists can be undertaken
using participation or travel cost models, using ag-
gregate data, as is the case with the Hamburg Tour-
ism Model or the participation model being used in
the second step of the PESETA study. Modelling of
the impacts of climate change on tourism is currently
institutionally fragmented in the EU. In addition, the
coverage of Croatia in these models is not very good
and this needs to be improved for use by Croatian
policy makers. However, research institutes in other
parts of Europe already have a comparative advan-
tage in this field, both in modelling and handling
the large geographic databases. Therefore, future
research in this area would benefit Croatia by the col-
laboration between Croatian and EU researchers to
improve existing or develop new aggregate partici-
pation (or travel cost) models and related databases.
While leaving the aggregate modelling of tourist be-
haviour to other European centres, it may also make
sense for Croatian researchers to develop recreation
demand models for individual sites or areas, using
survey data from individual respondents. There are
three reasons for this. First, economic activity in the
key beach tourism sector accounts for a large share
of Croatia’s GDP. These kinds of models, based on in-
dividual data, tend to shed much more light on the
climatic and environmental determinants of recre-
ation demand than do aggregate models. Second,
not only are these disaggregated models better at
predicting tourist behaviour, they can also be tai-
lored to the environmental nuances of specific sites,
such as Plitvice and other karst-based systems, that
are among Croatia’s must valuable environmental
assets. This enables the disaggregated recreation
demand models to better manage both the direct
and indirect effects of climate change on recre-
ational demand behaviour.
Recreation demand modelling at individual sites
or multiple sites using survey data on individuals,
either from a target population or on-site surveys,
is a fairly well developed field and quite complex.
A comprehensive study of an individual site (Na-
tional Park Plitvice, for example) could cost signifi-
cant resources and take several years to complete.
This would involve designing, administering and
processing survey data, estimating the parameters
of the recreation demand models, and then using
them to simulate the impacts of climate change and
the adaptation policies needed. However, if this sort
of study were done through the university system
grants programme, the costs would be significantly
less, since the only major costs would be the surveys
themselves and consultations/ trainings from econo-
mists abroad. The benefits of conducting such stud-
ies would be that they would provide more accurate
information than provided by aggregate models,
both about the impacts of climate change and pos-
sible responses to avoid these impacts.
Modelling the economic impacts of climate change
Aggregate participation models and disaggregated
recreational demand models can be used to project
the effects of climate change on visitation. However,
additional data is needed to calculate the effects on
the local economy and then to simulate the impact
from changes in local revenue on the larger econ-
omy. First, Croatia probably needs to step up its ef-
forts to systematically survey tourist expenditures at
many locations, on an on-going basis, using a homo-
geneous approach to sampling and surveying. This
data can be used in conjunction with participation
data to estimate changes in tourist expenditures due
to the impacts of climate change on visitation. Pro-
vided there is sufficient climate variability in the data
set, this data can also be analysed to determine if the
expenditures of tourists are in any way influenced
by climate change (if visitation is consistent). This
climate effect on expenditures can be incorporated
into the above calculations. A Croatian Input-Output
table or, better still, Computable General Equilibrium
(CGE) model could then be used to translate these
changes in local spending into their effects on na-
tional income and employment.