Economics of Tourism
ECONOMICS OF TOURISM
1. Introduction
Tourism has been a major growth industry globally for over five decades. Factors under-
pinning this growth include the growth of incomes and wealth, improvements in transport,
changing lifestyles and consumer values, increased leisure time, international openness and
globalization, immigration, special events, education, information and communication
technologies, destination marketing and promotion, improved general and tourism
infrastructure and so on (Matias et al 2007). Since there are economic consequences to all
of these factors it is not surprising that research in the area of tourism economics has
increased substantially during the same period. At the same time, the study of tourism
economics has attracted relatively few research economists compared to other topics, such
as energy and transport economists, within the mainstream discipline.
Although indirectly related to tourism economics, we may argue that the serious study of
the field began in the mid-1960s with the seminal book produced by Clawson and Knetsch
(1966) on the Economics of Outdoor Recreation. Rather prophetically, the book dealt in detail
with environmental issues, which are now considered of crucial importance in tourism
economics. Four years later, Gray (1970) published a very enlightening book on the
interrelation between international travel and trade. From then onwards, tourism gradually
gained momentum among economists; interestingly, however, it was not until 1995 that
Tourism Economics,
i.e. the first academic journal dedicated to the study of tourism economics, emerged. As a
complementary development it is also worth noting the establishment of the International
Association for Tourism Economics in 2007.
Four major observations can be made about the state of research in tourism economics.
⧫ First, there are ongoing areas of research very much within the single disciplinary
mainstream economic methodological framework. Obvious topics include demand
modelling, forecasting, economic impact and industry analysis (Stabler et al., 2010).
⧫ Second, several areas of research in economics have emerged that were either non-
existent two decades ago or were in their infancy. These include Game Theory, Chaos
Theory and climate change economics. These have been applied to tourism.
⧫ Third, there are several research areas relevant to the wider context of tourism studies,
that tourism economists have virtually ignored, or have relatively neglected. These
relate to themes and issues and methodologies of analysis that have been recognized in
other fields of the subject. These include ecological economics, poverty alleviation, and
sustainable development.
⧫ Fourth, tourism economics has become increasingly quantitative over time, paralleling
developments in the economics literature.
Critics have argued that the emphasis on ‘positivist/post positivist’ epistemologies render
the economics of tourism less relevant than it might otherwise be in addressing real world
,2 Economics of Tourism
issues and problems. As Jennings (2007) has argued, quantitative based research has become
the ‘orthodoxy’ for tourism economists and has prevented them from addressing tourism
problems in a more holistic, interdisciplinary way appropriate to the complexity of tourism
phenomena. Jennings’s view is that new and different methodologies and methods must be
employed by tourism economists for theory development, to better serve the industry, and
for policy formulation. Jennings’ review reflects the debate in the wider tourism literature
concerning the continued unwarranted adherence to positivist, quantitative oriented
orthodoxy in the face of tourism’s complexity, rapidly changing characteristics and
instability, quite different from its nature in the 1960s.). For some years now geographers
(and new economic geographers) have taken up a political economy stance. Williams (2004),
advocates a political economy perspective wherein theoretical developments in the
approach have relevance to issues in tourism as illustrated by issues such as
commodification in the sector’s markets, its labor structures and processes and its
regulation.
A discussion of the issues that have been addressed in tourism economics for the past 50 or
so years reveals that the range of issues addressed is perhaps much wider than the criticisms
might imply. We highlight several topics for discussion below.
2. Developments and Current Issues in Tourism Economics
a. Tourism Demand and Forecasting
Demand modelling, one of the most developed and rigorous areas of the economic analysis
of tourism, is a long-established area of economic research and continues to be so. Research
over the past four decades suggests that the range of factors affecting the demand for tourism
is very large. The more prominent factors that have been included in destination demand
modeling are income, (exchange rate adjusted) relative prices, transport costs, marketing and
promotion activity, migration levels and qualitative factors time available for travel, trade and
ethnic ties between the countries; destination attractiveness (for example, culture, climate,
history, natural resources, tourism infrastructure; special events taking place at the
destination; natural disasters; and social threats such as political instability, health issues or
terrorism) (Crouch 1994a, Lim 2006, Saymaan et al 2008). Of these factors, the bulk of
studies indicate that income, and to a lesser extent price, are the most important (Crouch
1992). Still, the focus on income as an influence on tourism flows has been associated with
a relative neglect of wealth as a determining factor (Alperovich and Machnes 1994). Thus,
while the Global Financial Crisis (GFC) certainly reduced incomes on average for millions of
people, perhaps the greatest effect was on their level of wealth due to the decline in value of
their assets including superannuation payouts. While there has always been some
recognition that wealth is important for some tourism markets e.g. Seniors’ tourism, the issue
needs more research (Sheldon and Dwyer 2010)
Demand analysis has recently taken new directions, with greater attention being
increasingly paid to the characteristic’s framework of demand (Lancaster 1966). This is also
associated with the development of the hedonic pricing method (Rosen 1974, Sinclair et al
1990, Clewer et al 1992, Papa theodorou 2001, 2002) and discrete choice analysis (Louvier
2000). More recent studies evaluating a variety of tourism markets are using panel data
, Economics of Tourism
techniques (Naudé and Saayman 2005, Van Der Merwe et al 2007, Saayman and Saayman
2008). When cross-sectional and time series data are combined, as in panel data analysis,
greater insights are gained from the data. Panel studies offer all the advantages of a larger
number of observations; that is, more informative data, less multicollinearity, more degrees of
freedom and more efficient estimates. In tourism demand studies, panel data techniques
allow the inclusion of the variables that are mostly static for one region (such as distance), but
which differ between regions, which is not possible with time series data only. Panel data is
expected to play an increasingly important role in tourism demand analysis.
Over time, the modeling of tourism demand has become more sophisticated and more
complex and different contexts of study, different data sets, use of different variables and
different modeling techniques preclude generalizations (Crouch 1994 a, 1995; Lim 1999, 2006).
Given the importance of a better understanding of demand for destination management,
marketing and policy purposes tourism demand modeling may be expected to continue to
be refined with more input from the econometrics literature (Song and Witt 2000, Li et al
2005, Song and Li 2008).
Forecasting is especially important in tourism because it aids long term planning and is
fundamental to the conduct of modern business and destination management. It is
particularly challenging because: the tourism product is perishable; tourism behavior is
complex; people are inseparable from the production-consumption process; customer
satisfaction depends on complementary products and services; and tourism demand is
extremely sensitive to natural and human-made disasters (Archer 1980, 1994). In a changing
global tourism environment, it is important, for both government policy development and
business planning, to have reliable short-term and long-term forecasts of tourism activity
(Frechtling 2001).
There are two broad approaches to tourism forecasting: qualitative tourism forecasting and
quantitative tourism forecasting (Sheldon and Var 1985). The same as for the area of demand
modeling the forecasting literature is increasingly incorporating ‘state of the art’ statistical
techniques that are new to tourism research (Song and Turner 2006). Song and Witt (2000)
were the first researchers to systematically introduce a number of modern econometric
methods to tourism demand analysis. More recently, modern econometric methods, such as
the autoregressive distributed lag model (ADLM), the error correction model (ECM), the vector
autoregressive (VAR) model, the almost ideal systems approach (AIDS and the time varying
parameter (TVP) models, have emerged as the main forecasting methods in the current tourism
demand forecasting literature. The technical illustration of these methods is in Song and Witt
(2000) and Li, Song and Witt, 2006; Song and Li 2008). There is no single quantitative technique
that gives best forecasting results in all contexts (Song and Li 2008).
Qualitative tourism forecasts are based on the judgments of persons sharing their
experience, practical knowledge and intuition. These judgments are found through polling,
expert opinion, panel consensus, surveys, Delphi technique and scenario writing and are
often used to moderate or “second guess” quantitative forecasts (Frechtling 2001). Qualitative
forecasting is best applied when facing insufficient historical data; unreliable time series;
rapidly changing macro environments; major disturbances; and when long term forecasts are
desired.
The choice of forecasting method depends on several considerations including: the level
ECONOMICS OF TOURISM
1. Introduction
Tourism has been a major growth industry globally for over five decades. Factors under-
pinning this growth include the growth of incomes and wealth, improvements in transport,
changing lifestyles and consumer values, increased leisure time, international openness and
globalization, immigration, special events, education, information and communication
technologies, destination marketing and promotion, improved general and tourism
infrastructure and so on (Matias et al 2007). Since there are economic consequences to all
of these factors it is not surprising that research in the area of tourism economics has
increased substantially during the same period. At the same time, the study of tourism
economics has attracted relatively few research economists compared to other topics, such
as energy and transport economists, within the mainstream discipline.
Although indirectly related to tourism economics, we may argue that the serious study of
the field began in the mid-1960s with the seminal book produced by Clawson and Knetsch
(1966) on the Economics of Outdoor Recreation. Rather prophetically, the book dealt in detail
with environmental issues, which are now considered of crucial importance in tourism
economics. Four years later, Gray (1970) published a very enlightening book on the
interrelation between international travel and trade. From then onwards, tourism gradually
gained momentum among economists; interestingly, however, it was not until 1995 that
Tourism Economics,
i.e. the first academic journal dedicated to the study of tourism economics, emerged. As a
complementary development it is also worth noting the establishment of the International
Association for Tourism Economics in 2007.
Four major observations can be made about the state of research in tourism economics.
⧫ First, there are ongoing areas of research very much within the single disciplinary
mainstream economic methodological framework. Obvious topics include demand
modelling, forecasting, economic impact and industry analysis (Stabler et al., 2010).
⧫ Second, several areas of research in economics have emerged that were either non-
existent two decades ago or were in their infancy. These include Game Theory, Chaos
Theory and climate change economics. These have been applied to tourism.
⧫ Third, there are several research areas relevant to the wider context of tourism studies,
that tourism economists have virtually ignored, or have relatively neglected. These
relate to themes and issues and methodologies of analysis that have been recognized in
other fields of the subject. These include ecological economics, poverty alleviation, and
sustainable development.
⧫ Fourth, tourism economics has become increasingly quantitative over time, paralleling
developments in the economics literature.
Critics have argued that the emphasis on ‘positivist/post positivist’ epistemologies render
the economics of tourism less relevant than it might otherwise be in addressing real world
,2 Economics of Tourism
issues and problems. As Jennings (2007) has argued, quantitative based research has become
the ‘orthodoxy’ for tourism economists and has prevented them from addressing tourism
problems in a more holistic, interdisciplinary way appropriate to the complexity of tourism
phenomena. Jennings’s view is that new and different methodologies and methods must be
employed by tourism economists for theory development, to better serve the industry, and
for policy formulation. Jennings’ review reflects the debate in the wider tourism literature
concerning the continued unwarranted adherence to positivist, quantitative oriented
orthodoxy in the face of tourism’s complexity, rapidly changing characteristics and
instability, quite different from its nature in the 1960s.). For some years now geographers
(and new economic geographers) have taken up a political economy stance. Williams (2004),
advocates a political economy perspective wherein theoretical developments in the
approach have relevance to issues in tourism as illustrated by issues such as
commodification in the sector’s markets, its labor structures and processes and its
regulation.
A discussion of the issues that have been addressed in tourism economics for the past 50 or
so years reveals that the range of issues addressed is perhaps much wider than the criticisms
might imply. We highlight several topics for discussion below.
2. Developments and Current Issues in Tourism Economics
a. Tourism Demand and Forecasting
Demand modelling, one of the most developed and rigorous areas of the economic analysis
of tourism, is a long-established area of economic research and continues to be so. Research
over the past four decades suggests that the range of factors affecting the demand for tourism
is very large. The more prominent factors that have been included in destination demand
modeling are income, (exchange rate adjusted) relative prices, transport costs, marketing and
promotion activity, migration levels and qualitative factors time available for travel, trade and
ethnic ties between the countries; destination attractiveness (for example, culture, climate,
history, natural resources, tourism infrastructure; special events taking place at the
destination; natural disasters; and social threats such as political instability, health issues or
terrorism) (Crouch 1994a, Lim 2006, Saymaan et al 2008). Of these factors, the bulk of
studies indicate that income, and to a lesser extent price, are the most important (Crouch
1992). Still, the focus on income as an influence on tourism flows has been associated with
a relative neglect of wealth as a determining factor (Alperovich and Machnes 1994). Thus,
while the Global Financial Crisis (GFC) certainly reduced incomes on average for millions of
people, perhaps the greatest effect was on their level of wealth due to the decline in value of
their assets including superannuation payouts. While there has always been some
recognition that wealth is important for some tourism markets e.g. Seniors’ tourism, the issue
needs more research (Sheldon and Dwyer 2010)
Demand analysis has recently taken new directions, with greater attention being
increasingly paid to the characteristic’s framework of demand (Lancaster 1966). This is also
associated with the development of the hedonic pricing method (Rosen 1974, Sinclair et al
1990, Clewer et al 1992, Papa theodorou 2001, 2002) and discrete choice analysis (Louvier
2000). More recent studies evaluating a variety of tourism markets are using panel data
, Economics of Tourism
techniques (Naudé and Saayman 2005, Van Der Merwe et al 2007, Saayman and Saayman
2008). When cross-sectional and time series data are combined, as in panel data analysis,
greater insights are gained from the data. Panel studies offer all the advantages of a larger
number of observations; that is, more informative data, less multicollinearity, more degrees of
freedom and more efficient estimates. In tourism demand studies, panel data techniques
allow the inclusion of the variables that are mostly static for one region (such as distance), but
which differ between regions, which is not possible with time series data only. Panel data is
expected to play an increasingly important role in tourism demand analysis.
Over time, the modeling of tourism demand has become more sophisticated and more
complex and different contexts of study, different data sets, use of different variables and
different modeling techniques preclude generalizations (Crouch 1994 a, 1995; Lim 1999, 2006).
Given the importance of a better understanding of demand for destination management,
marketing and policy purposes tourism demand modeling may be expected to continue to
be refined with more input from the econometrics literature (Song and Witt 2000, Li et al
2005, Song and Li 2008).
Forecasting is especially important in tourism because it aids long term planning and is
fundamental to the conduct of modern business and destination management. It is
particularly challenging because: the tourism product is perishable; tourism behavior is
complex; people are inseparable from the production-consumption process; customer
satisfaction depends on complementary products and services; and tourism demand is
extremely sensitive to natural and human-made disasters (Archer 1980, 1994). In a changing
global tourism environment, it is important, for both government policy development and
business planning, to have reliable short-term and long-term forecasts of tourism activity
(Frechtling 2001).
There are two broad approaches to tourism forecasting: qualitative tourism forecasting and
quantitative tourism forecasting (Sheldon and Var 1985). The same as for the area of demand
modeling the forecasting literature is increasingly incorporating ‘state of the art’ statistical
techniques that are new to tourism research (Song and Turner 2006). Song and Witt (2000)
were the first researchers to systematically introduce a number of modern econometric
methods to tourism demand analysis. More recently, modern econometric methods, such as
the autoregressive distributed lag model (ADLM), the error correction model (ECM), the vector
autoregressive (VAR) model, the almost ideal systems approach (AIDS and the time varying
parameter (TVP) models, have emerged as the main forecasting methods in the current tourism
demand forecasting literature. The technical illustration of these methods is in Song and Witt
(2000) and Li, Song and Witt, 2006; Song and Li 2008). There is no single quantitative technique
that gives best forecasting results in all contexts (Song and Li 2008).
Qualitative tourism forecasts are based on the judgments of persons sharing their
experience, practical knowledge and intuition. These judgments are found through polling,
expert opinion, panel consensus, surveys, Delphi technique and scenario writing and are
often used to moderate or “second guess” quantitative forecasts (Frechtling 2001). Qualitative
forecasting is best applied when facing insufficient historical data; unreliable time series;
rapidly changing macro environments; major disturbances; and when long term forecasts are
desired.
The choice of forecasting method depends on several considerations including: the level
