Content uploaded by A. Özlem Önder
Author content
All content in this area was uploaded by A. Özlem Önder on Dec 14, 2015
Content may be subject to copyright.
An Empirical Analysis of the Determinants of International Tourism Demand :
The Case of Izmir
A. Özlem Öndera, Aykan Candemirb, Neşe Kumrala
a Department of Economics, Ege University, 35100, Izmir, Turkey
b Department of Business Administration, Ege University, 35100, Izmir, Turkey
Abstract
Tourism has become an important sector in Turkey as a growing source of foreign
exchange reserves and employment over the last two decades. After being one of the most
important tourist destinations for decades, Izmir lost its relative importance after 1990’s. With
its historical, cultural values and its nature Izmir still has a significant tourism potential.
Given the importance of this sector for Izmir, this paper investigates the factors
affecting international tourism demand in Izmir using time series data between 1980-2005.
Double logarithmic model is used in estimation. Real exchange rate, GDP per capita of
OECD countires, GDP per capita of Izmir and transportation public capital stock of Izmir are
the variables used to explain international tourist arrivals of Izmir. The empirical results show
that price and income of the tourist generating country are the main determinants of tourism
demand. Income and price elasticites are above one. Local factors related to the development
level of Izmir and transportation public capital stock have no significant effect. Policy
implications derivable from this study suggest that government should encourage alternative
forms of tourism development besides mass tourism via experience economy
Keywords: Tourism demand, Izmir, experience economy.
JEL classification: R11, C13
An Empirical Analysis of the Determinants of International Tourism Demand :
The Case of Izmir
A. Özlem Öndera, Aykan Candemirb, Neşe Kumrala
a Department of Economics, Ege University, 35100, Izmir, Turkey
b Department of Business Administration, Ege University, 35100, Izmir, Turkey
1. Introduction
Tourism has become an important sector in Turkey as a growing source of foreign exchange
reserves and employment over the last two decades. Contribution of tourism to Turkish
economy has also been stressed by some researchers (see, Gunduz and Hatemi-J, 2005).
When the historical background of tourism in Turkey is examined, it is seen that the
development of tourism sector has merely depended on mass tourism and tourism plans has
been prepared to promote coastal activities, namely sea-sand-sun (3S) type of activities.
Moreover although Turkey has high tourism potential, she had not developed alternative
modes of tourism mainly depending on natural, cultural and traditional assets of the country.
Izmir, as being one of the leading tourism destinations at the beginning of modern
tourism activities in Turkey has also problems to realize its tourism potential after 1990’s.
However tourism is an important sector for the region in terms of employment and growth.
Factors affecting international tourism demand in Turkey has been investigated in Turkey
(see, Icoz, et al. 1998, Akis, 1998, and Akal, 2004 for example). However, there appears to
be no study related to the determinants of international tourism demand as far as Izmir is
concerned.
Given the importance of the sector for Izmir, the aim of this paper is to investigate the
factors affecting international tourism demand in Izmir. The rest of the paper is organised as
follows. Section 2 gives some information about tourism sector in Turkey and Izmir. Section
3 explains the theoretical model. Section 4 gives detailed information about data employed
in the study. Section 5 gives the estimates of the study. The final section underscores the main
findings of the study and gives some policy recommendations.
2. Tourism in Izmir
The most strategic port and the third largest city of Turkey, Izmir is located at the
intersection point of many touristic centers. With a mystical history of 7,000 years, Izmir
proudly hosts a large number of extremely important architectural designs and archeological
sites.
On the other hand, Izmir presents its services to cultural tourism with its countless
historical sites reaching to our days from prehistoric times, works and buildings, museums,
typical Mediterranean kitchen, handicrafts, festivals, festivities, and other local cultural
specialties, as well as offering its respectful participation in the belief tourism with a large
collection of sacred places.
Congress tourism is rapidly developing, thermal waters are a very important tourism
resource for Izmir, as well as in all the Aegean Region. In the health tourism sector, Izmir has
numerous facilities. Touristic facilities in metropolitan Izmir are well developed, with a wide
range of possibilities to deserve the appreciation of every taste and savor.
With its historical identity, beaches and sea, mountains and plateaus, thermal waters,
macchia and red pine forests, Aegean agricultural products and soft climate, Izmir presents
many unique opportunities to realize recreational activities and nature-compliant tourism
types (see, Pirnar 2005, for details).
After being one of the most important tourist destinations for decades, Izmir lost its
importance after 1990’s. With its historical, cultural values and its nature Izmir still has a
significant tourism potential. Revival of tourism industry for the city will contribute the
regional and national economy. However the success of a destination is not easy since many
new destinations and variations of touristic products are introduced within the international
tourism market and current destinations and products try harder to strengthen their
competitiveness. (see, Ozdemir, 2005, for details).
The continuous decrease of share for tourism in and around Izmir can be seen from
Figure 1. In the last 25 years Izmir has lost its share with 50% (i.e. from 8% to 4%). The
tourism industry has been growing in Turkey but Izmir were incapable of adapting to this
growth. This has several reasons; the first reason was that the southern Turkey had long and
unoccupied seashore and became the “Turkish Riviera”. After 1995, Antalya province has
become the star of Turkish Tourism with a 30% of share for arrivals (see, Figure 2), the
second reason was that the Aegean Region was occupied by the summer residences which did
not leave enough space for the development of tourism. The third reason was that the facilities
became old-fashioned.
0,00
2,00
4,00
6,00
8,00
10,00
12,00
14,00
1982.
1983.
1984.
1985.
1986.
1987.
1988.
1989.
1990.
1991.
1992.
1993.
1994.
1995.
1996.
1997.
1998.
1999.
2000.
2001.
2002.
2003.
2004.
2005.
2006.
2007.
(% of Arrivals)
Source: Ministry of Culture and Tourism, www.kultur.gov.tr
Figure 1: Share of Izmir in Touristic Activities (% of Arrivals in Turkey)
As can be seen from Figure 2, Izmir is still a destination for traditional tourist sending
countries, however southern Turkey (i.e. Antalya Region) has become one of the top
destinations for Russian tourists. Izmir cannot attract those new tourist sending countries.
Istanbul is increasing its share with the new efforts being performed in the last couple of
years.
0,00
5,00
10,00
15,00
20,00
25,00
30,00
35,00
40,00
2000. 2001. 2002. 2003. 2004. 2005. 2006.
İzmir Antalya İstanbul
Source: Ministry of Culture and Tourism, www.kultur.gov.tr
Figure 2: Share of Izmir, Antalya and Istanbul in Touristic Activities (% of Arrivals in
Turkey)
Figure 3 presents the average length of stays in 2006. As can be seen from the figure
average length of stay in Turkey is 4 days, where Antalya province has 5 days of average stay.
ength of stay and occupancy rates in Izmir.
he occupancy rate reveals important information about the present situation and the potential
Ankara and İstanbul have lower average length of stay due to business trips, congress tourism
etc. Izmir has an average length of stay slightly above 3 days due to its position as a hub with
its port and airport.
Source: Ministry of Culture and Tourism, www.kultur.gov.tr
Figure 3: Average Length of Stay (Days-2006)
0
1
2
3
4
5
Turkey
Antalya
İstanbulAnkara
İzmir
The following table presents the average l
T
for tourism. The occupancy rate for Izmir, excluding the camping sites, cannot exceed 50%
which shows that there is no need for more investment for new facilities whereas the need for
effective marketing efforts can obviously seen. The highest rates are for 4 star hotels and
holiday villages (see Table 1).
Table 1: Average Length of Stay (Days) and Occupancy Rates in Izmir by Type and
lass of Establishment (Izmir 2006)
ype and Class of Establishment Average Length of Stay Occupancy Rate %
C
T
HOTEL Foreigner Citizen Total Foreigner Citizen Total
5 Star 3,0 1,9 3,4 20,78 17,97 38,75
4 Star 3,9 1,5 2,6 31,26 15,58 46,84
3 Star 2,2 1,5 1,6 6,35 26,56 32,91
2 Star 2,8 1,7 1,9 6,93 27,13 34,06
1 Star 1,8 1,1 1,2 3,23 30,62 33,85
Special Licence 2,0 1,5 1,9 20,41 3,96 24,37
Thermal - - - - - -
Boutique - - - - - -
Apart - 6,2 6,2 - 2,57 2,57
TOTAL 3,1 1,7 2,1 17,46 20,69 38,14
MOTEL 14,4 4,6 4,5 3,61 14,66 28,27
BOARDING HOUSE 7,3 2,7 2,9 1,48 12,43 13,91
HOLIDAY VILLAGE 5,9 3,5 4,2 17,59 27,16 44,75
INN - - - - - -
CAMPING 10, 13,8 12,5 24,39 51,76 76,15 4
GOLF EST. - - - - - -
TRAINING EST. 2,8 1,3 1,4 2,62 17,46 20,08
TOURISM COMPLEX - - - - - -
MOUNTAIN HOUSE - - - - - -
GRAND TOTAL 3,3 1,8 2,3 17,29 21,38 38,66
Source: Ministry of Culture and Tou m, www.kultur.gov.trris
iterature for different countries
ee, Crouch, 1994, Lim, 1997, and Li et al. 2005, for a survey). Income and relative prices
are variables. Tourism prices are not in general available. Exchange rates have
3. Model
Tourism demand has been extensively investigated in the l
(s
mostly used
used as a proxy variable for price. Other exogenous variables are also considered like cost of
transportation, attributes of the destination like safety, infrastructure etc.
Based on the literature, the international tourism demand model can be written as
follows
tttttt TRANSPGDPPCGDPOPCEXCHTOUR
ε
β
β
β
β
α
+
+
+
++= 4321 (1)
where represents tourist arrivals, is the real Exchange rate, s the
ita GD pita of the
l stock,
The per capita GDP of the OECD countries are used as a proxy for income. It is expected
tha
4. Data
order to measure the tourism demand we have used tourist arrivals. Another measure could
in logarithmic form.
t
TOUR t
EXCH t
DPOPC , iG
per cap P of the OECD countries, t represents GDP per ca Izmir,
t
TRANSP is the transportation public capita
ε
GDPPC
t is error term and t is time.
t tourist arrivals increase as the income increases. The real exchange rate is defined as the
price of the tourist generating countries currency in terms of host countries currency. The
coefficient of the real exchange rate is expected to be negative. GDP per capita of the region
is an indicator of the level of economic development, which could promote tourist arrivals.
The coefficient of the GDP per capita for Izmir is expected to be positive. Transportation
infrastructure is expected to have a positive impact on tourism. If the tourism demand model
is estimated in the log-log form, the coefficients could be interpreted as elasticities.
In
be tourism receipts which were not available for Izmir. Data on tourist arrivals are obtained
from Turkish Statistical Institute (TURKSTAT). Data on real exchange rate are collected
from electronic delivery system of Central Bank of the Republic of Turkey. Data on GDP per
capita of OECD countries based on purchasing power parity (PPP) are obtained from OECD
datasets. Per capita income and transportation public capital stock data for Izmir are obtained
from Önder et al. 2007. Except for data on GDP per capita for Izmir and transportation capital
stock annual data from 1980 to 2005 were available. Data for GDP per capita for Izmir and
transportation capital stock were only available from 1980 to 2001. All the variables are used
5. Estimation
Table 2 presents the results of estimation of Equation 1. The equation is estimated by method
of e, Witt, 2000 for different methods). The dependent variable is tourist
ficients of real exchange rate and GDP per
apita of the OECD countries are as expected and significant. Price elasticity, -1.578 indicates
a price
ults of the Model I 1980-2001
Dependent Varaible: TOUR
t-ratio
least squares (se
arrivals and real exchange rate, GDP per capita of the OECD countries, GDP per capita of
Izmir and transportation capital stock of Izmir.
As can be seen from the table the coef
c
elastic demand for international tourism. The results also show that income elasticity is
above 2 indicating that tourism is a luxury good which is in line with previous literature on
tourism demand (see Eilat, and Einav, 2004, for example) All the diagnostic tests are
satisfactory.
Table 2: Res
Variable Coefficient
Constant -8.952 -0.60
Et
XCH -3.59***
GDPOPCt2.117 3.25***
Tests
y 9)
-1.578
GDPPCt1.3946 0.77
TRANSPt-0.336 -0.72
Adj R20.872
Diagnostig
Normalit 0.45 (0.7
Autocorrelation χ2(1) 61)
cedasticity F(8,22) .39)
0.26 (0.
Heteros 1.165(0
Notes: Normality is the Jarque-Bera Test has Chi-squared distributi n under null normality of
uals. Autocorrelation is a Lagrange multipl r test for first order serieal
hite h city test has F statistic under the null of
. p parenthesis.
o
resid ie
correlation. W eteroscedasti
homoscedastic error values are in
The coefficient of the GDP per capita of Izmir and transportation capital stock of Izmir
nd to be insignificant indicating that the local fact
are fou ors of Izmir does not have a
significant effect on international tourist arrivals.
We reestimate the model by excluding these two insignificant variables. Table 3
presents estimation results of the new model. Adjusted R2 is higher in this model. The price
and income elasticities are again above one and significant. This implies that income and
price policies has a strong effect on international tourism demand for Izmir. However policies
related to local factor are found to be ineffective.
Table 3: Results of the Model II 1980-2005
Dependent Varaible: TOUR
Variable Coefficient t-ratio
Constant -0,342 -0.19
EXCH -1.267 -4.20***
DPO 14.36***
Adj R20.891
g Tests
t
G PCt1.956
Diagnosti
Normality 60)
relation χ2(1) 67)
asticity F(5,20) 46)
1.03 (0.
Autocor 0.17 (0.
Heterosced 0.95(0.
Notes: Normality is the Jarque-Bera Test has Chi-squared distributi n under null normality of
utocorrelation is a Lagrange multiplier test for first order serieal correlation. White
oscedasticity test has F statistic under the null of homoscedastic error. p values are in
6.
In this stimated international tourism demand for Izmir, which is an important
tourism destination in Turkey. We used double logarithmic model for the time series data
between 1980-1995.
o
residuals. A
heter
parenthesis.
Conclusion
study we e
The results of the study show that price and income of the tourist generating country
are the main determinants of tourism demand. Income and price elasticites are above one.
Local factors related to the development level of Izmir and transportation public capital stock
ave no significant effect. This results implies that tourism demand in Izmir is mainly
depend
the recent years. Homeros
wrote his world famous Iliad here and the oldest temple for goddess Athena was built in
mir.
stinations in the 1990’s.
, artisan food alongside culinary, environmental,
h
s on the factor of tourism generating countries. Local factors of Izmir are not effective
in demand generation, which is typical for mass tourism generating regions.
Izmir has a distinct place from the aspect of being an important center for tourism and
having a developed infrastructure as well as having rich natural, cultural and historical
resources and having a close location to tourist sending countries.
As being one of the leading tourism destinations at the beginning of modern tourism
activities in Turkey, Izmir has a very important potential for tourism with its location on the
Aegean shore, 7000 years of history and its organic agriculture in
Iz Besides these, Izmir was a home for Hitites, legendary tribe of Amazones, Phrigyians
and many others.
Despite all these natural, historical and cultural richness the city cannot benefit this
potential. Thus the demand for tourism is quite low and the city stayed at the back of many
shining tourism de
Policy implications derivable from this study suggest that government should
encourage alternative forms of tourism development besides mass tourism via experience
economy (including arts craft and design
archaeological and the variants of cultural tourism) for Izmir. Organic agriculture and food
industr
sitors and the revenues can be increased. Izmir can also
be a leading city for health tourism with its thermal resources. The city can increase its
ctiviti
y depending of organic agriculture has a remarkable place in the tourism policy in
relation with experience economy. From this point of view Izmir has a great potential and the
government started activities to form clusters for organic agriculture and considerable amount
of support will be supplied. In such a cluster, culinary tourism can be developed, organic
farms and a gourmet institute can be founded, world famous chefs and cooks can be educated
and food festivals can be organized.
Besides organic agriculture the richness of the 7000 years of history can be utilized to
present the visitors meaningful experiences (performances about daily living in ancient times
etc.). By this means the number of vi
a es with conferences, festivals, concerts, exhibitions, fairs, sports, cultural events.
References
kal, M. (2004), “Forecasting Turkey’s revenues by ARMAX model” TourismManagement,
25, 5, 565-580.
kis, S. (2004), “A Compact econometric model of tourism demand for Turkey” Tourism
, 19, 1, 99-102.
earch, 32, 4, 12-23.
conomics 36, 1315-1326.
9-504.
fore 82-89.
9.
niversity I.I.B.F. MANISA, 12, 1,
Önder dağ M. (2007) The effects of public capital stock on
ust-2
Ozdem , P. (2005), ‘Tourism Product Diversification as a Strategy to Stimulate
A
A
Management
Crouch, L. (1994), “The study of international tourism demand:a survey of practice” Journal
of Travel Res
Eilat and Einav, (2004) “Determinants of International Tourism:a three dimensial panel data
Analsis”, Applied E
Gunduz L. and Hatemi-J. A., (2005), Is the tourism-led growth hypothesis valid for Turkey,
Applied Economics Letters, 12, 49
Icoz O. T. Var, and Kozak M. (1998), “Tourism demand in Turkey”, Annals of Tourism
Research, 25, 1, 236-240.
Li, G. Song H. , and Witt, S. F. (2005) “Recent developments in econometric modeling and
casting, Journal of Travel Research, 44, 1,
Lim C. (1997), “Review of international tourism demand models, Annals of Tourism
Research, 24, 4, 835-84
Pirnar, I. (2005), ‘The Importance of Tourism For Izmir and Suggestions For
Development’, Yonetim ve Ekonomi, Celal Bayar U
47-60.
A. Ö., Deliktaş E., and Kara
regional convergence in Turkey47th Congress of ERSA, Paris, 29 Aug
September.
ir Meric
Tourism Demand in Izmir and the Use of Delphi Technique’, Unpublished Ph
Thesis, Doku
D
z Eylul University.
www.I
Witt, S. F. (2000),Tourism demand modeling, and forecasting :modern econometric
approaches, Elsevier.
zmirturizm.gov.tr
www.kultur.gov.tr
