Us<strong>in</strong>g the SUR Model <strong>of</strong> Tourism Dem<strong>and</strong> forNeighbour<strong>in</strong>g Regions <strong>in</strong> Sweden <strong>and</strong> Norway6A. Khalik SalmanMid Sweden University, Social science / Economics DepartmentSweden1. IntroductionThis chapter estimates the <strong>in</strong>ternational dem<strong>and</strong> for tourism <strong>in</strong> two neighbour<strong>in</strong>g regions:the objective number 6 (SW:6) <strong>in</strong> Sweden <strong>and</strong> North Norway <strong>in</strong>cluded – Tröndelag (NWT)<strong>in</strong> North Norway, from five different countries: Denmark, the United K<strong>in</strong>gdom,Switzerl<strong>and</strong>, Japan, <strong>and</strong> the United States. For each visit<strong>in</strong>g country, <strong>and</strong> for Sweden <strong>and</strong>Norway, we specify separate equations by <strong>in</strong>clud<strong>in</strong>g the relevant <strong>in</strong>formation. we thenestimate these ten equations us<strong>in</strong>g Zellner’s Iterative Seem<strong>in</strong>gly Unrelated Regressions(ISUR). <strong>The</strong> benefit <strong>of</strong> this model is that the ISUR estimators utilize the <strong>in</strong>formation with<strong>in</strong><strong>and</strong> the relation between the equations present <strong>in</strong> the error correlation <strong>of</strong> the crossregressions (or equations) <strong>and</strong> hence is more efficient than s<strong>in</strong>gle equation estimationmethods such as ord<strong>in</strong>ary least squares. Monthly time series data from 1993:01 to 2006:12are used. <strong>The</strong> results show that the consumer price <strong>in</strong>dex, some lagged dependent variables,<strong>and</strong> several monthly dummies (represent<strong>in</strong>g seasonal effects) have significant impacts onthe number <strong>of</strong> visitors to the SW:6 region <strong>in</strong> Sweden <strong>and</strong> NWT region <strong>in</strong> Norway. We als<strong>of</strong><strong>in</strong>d that, <strong>in</strong> at least some cases, relative prices <strong>and</strong> exchange rates have significant effects on<strong>in</strong>ternational tourism dem<strong>and</strong>.Tourism has important impacts on the economies <strong>of</strong> both develop<strong>in</strong>g <strong>and</strong> <strong>in</strong>dustrializedcountries, result<strong>in</strong>g <strong>in</strong> job creation, additional <strong>in</strong>come for the private <strong>and</strong> public sectors,foreign currency receipts, higher <strong>in</strong>vestment <strong>and</strong> growth. Indeed, tourism has acted as acatalyst to economic restructur<strong>in</strong>g <strong>in</strong> many recipient countries, assist<strong>in</strong>g a shift away fromprimary sector activities, towards greater reliance on services <strong>and</strong> manufactur<strong>in</strong>g. Given thescale <strong>of</strong> tourism’s contribution to the macroeconomic dimension over time, knowledgeconcern<strong>in</strong>g the nature <strong>of</strong> the dem<strong>and</strong> upon which it is based is <strong>of</strong> both theoretical <strong>and</strong>practical relevance. It is well known that tourism dem<strong>and</strong> is responsive to such variables as<strong>in</strong>come, relative prices <strong>and</strong> exchange rates. What is not known is how the responsiveness <strong>of</strong>dem<strong>and</strong> to changes <strong>in</strong> these variables alters dur<strong>in</strong>g a country’s economic transition <strong>and</strong><strong>in</strong>tegration <strong>in</strong>to the wider world <strong>in</strong>itial or subsequent years? Does the sensitivity <strong>of</strong> tourismdem<strong>and</strong> to changes <strong>in</strong> its own prices, or those <strong>of</strong> its competitors, change between differentperiods? Further questions concern the degrees <strong>of</strong> complementarity or substitutabilitybetween tourism dest<strong>in</strong>ations <strong>and</strong> the extent to which these change dur<strong>in</strong>g periods <strong>of</strong>economic transition. Complementarity occurs if holidays <strong>in</strong> different dest<strong>in</strong>ations arepurchased as a package. Alternatively, there may be an <strong>in</strong>tense degree <strong>of</strong> competitionbetween dest<strong>in</strong>ations. Relationships <strong>of</strong> complementarity or substitutability may change over
98Advances <strong>in</strong> Econometrics - <strong>The</strong>ory <strong>and</strong> Applicationstime as lower <strong>in</strong>come dest<strong>in</strong>ations emerge from relative poverty to achieve a higher level <strong>of</strong>development. Little <strong>in</strong>formation is available about this issue. It is not known, for example,whether lower <strong>in</strong>come dest<strong>in</strong>ations tend to become more or less competitive over time,either relative to other develop<strong>in</strong>g countries or relative to more <strong>in</strong>dustrialized nations.Different models have been used to estimate tourism dem<strong>and</strong> <strong>and</strong> some types <strong>of</strong> model aremore appropriate for exam<strong>in</strong><strong>in</strong>g the above questions than others. <strong>The</strong> vast majority <strong>of</strong>studies <strong>of</strong> tourism dem<strong>and</strong> have relied on s<strong>in</strong>gle equation models <strong>of</strong> dem<strong>and</strong>, estimatedwith<strong>in</strong> a static context (for example, Uysal <strong>and</strong> Crompton, 1984; Gunadhi <strong>and</strong> Boey,1986).<strong>The</strong>se models are not derived from consumer dem<strong>and</strong> theory <strong>and</strong> fail to quantify thechanges <strong>in</strong> dem<strong>and</strong> behaviour that occur over time. Innovations <strong>in</strong> the methodology weresubsequently <strong>in</strong>troduced <strong>in</strong> the form <strong>of</strong> s<strong>in</strong>gle equation models <strong>of</strong> dem<strong>and</strong> estimated us<strong>in</strong>gan error correction methodology were subsequently <strong>in</strong>troduced <strong>in</strong> s<strong>in</strong>gle equation model <strong>of</strong>dem<strong>and</strong> estimated us<strong>in</strong>g an error correction methodology (Syriopoulos, 1995). Kulendran(1996) used a general to specific, error correction model to estimate the Australia dem<strong>and</strong>for tourism <strong>in</strong> the form <strong>of</strong> visits per capita to outbound dest<strong>in</strong>ations <strong>and</strong> demonstrated thatthe model has good forecast<strong>in</strong>g ability. This modell<strong>in</strong>g approach has the advantage <strong>of</strong>explicit treatment <strong>of</strong> the time dimension <strong>of</strong> tourism dem<strong>and</strong> behaviour <strong>and</strong> allows forimproved econometric estimation <strong>of</strong> the specified equations.A More recently approach to tourism dem<strong>and</strong> estimation <strong>in</strong>volves a system -wide approachby us<strong>in</strong>g the ISUR Model (Salman et al. 2010). This system <strong>of</strong> study is particularly useful fortest<strong>in</strong>g the properties <strong>of</strong> homogeneity <strong>and</strong> symmetry which are basic to consumer dem<strong>and</strong>theory. Hence, it provides a stronger theoretical basic for estimat<strong>in</strong>g the cross- priceelasticities <strong>of</strong> dem<strong>and</strong> than the s<strong>in</strong>gle equation approach.This chapter was used system wise approach by the ISUR Model to exam<strong>in</strong>e tourismdem<strong>and</strong> by the UK, Switzerl<strong>and</strong>, Denmark, Japan <strong>and</strong> <strong>The</strong> USA <strong>in</strong> the neighbor<strong>in</strong>gdest<strong>in</strong>ation number 6 (SW:6) <strong>and</strong> (NWT) <strong>in</strong> Norway. <strong>The</strong> UK, Switzerl<strong>and</strong>, Denmark, Japan<strong>and</strong> <strong>The</strong> USA are major orig<strong>in</strong> countries for tourism <strong>in</strong> the dest<strong>in</strong>ations under consideration.SW:6 <strong>in</strong> Sweden <strong>and</strong> NWT <strong>in</strong> Norway are key dest<strong>in</strong>ations, account<strong>in</strong>g for over one-third<strong>of</strong> all receipts from tourism <strong>in</strong> the European Union <strong>in</strong> 2005. <strong>The</strong> absolute value <strong>of</strong> theirreceipts from tourism is very high, at over $ 200 million <strong>in</strong> 2005. <strong>The</strong> choice <strong>of</strong> the countriesas dest<strong>in</strong>ations for analysis is also appropriate ow<strong>in</strong>g to their position as geographicneighbours. Complementarity or substitutability <strong>in</strong> tourism dem<strong>and</strong>, as <strong>in</strong>dicated by thesigns <strong>of</strong> the relative-price elasticities <strong>of</strong> dem<strong>and</strong>, is <strong>of</strong> particular relevance <strong>in</strong> this context.This chapter pays attention to this issue, which has not previously been exam<strong>in</strong>ed for thecase <strong>of</strong> neighbour<strong>in</strong>g countries us<strong>in</strong>g the ISUR approach. Sweden <strong>and</strong> Norway are<strong>in</strong>terest<strong>in</strong>g cases for consideration ow<strong>in</strong>g to their position as economies <strong>in</strong> transition dur<strong>in</strong>gthe period under consideration. By the early 1990s, the start years <strong>of</strong> the period under study,they had adopted new development policies, high dependence on <strong>in</strong>dustry, mov<strong>in</strong>gtowards <strong>in</strong>creased globalization, economic <strong>in</strong>tegration, <strong>and</strong> foreign competition. Swedenhad jo<strong>in</strong>ed the ranks <strong>of</strong> the more developed European economies <strong>in</strong> the period under study.Hence, an <strong>in</strong>novative feature <strong>of</strong> this chapter is its exam<strong>in</strong>ation <strong>of</strong> the evolution <strong>of</strong> tourismdem<strong>and</strong> dur<strong>in</strong>g these countries’ transition to a new technological system <strong>and</strong> globalizationstatus. It also permits exam<strong>in</strong>ation <strong>of</strong> the extent to which the behaviour <strong>of</strong> dem<strong>and</strong> becomemore or less similar over time with respect to changes <strong>in</strong> prices <strong>and</strong> exchange rates. Thus,this study provides useful <strong>in</strong>formation, at the cross-country level, about change <strong>in</strong> a majoractivity with<strong>in</strong> each <strong>of</strong> the economies.