Proceedings in pdf format. - Sociotechnical Systems Engineering ...

200%0%19802045holidays. The share of domestic in total holidays onlystarts to fall if the lower income class are rich enough toafford a holiday abroad; with the estimates of Equation(4), this happens if average income exceeds $360,000, ahigh number.For the total (domestic and foreign) number oftourists, the world total is 12.0% higher if we include theinterpolated tourist numbers, that is, 4.0 billion versus3.6 billion tourists. The observed world total includesthose countries for which we have observed bothdomestic tourists and international arrivals. For domestictourists only, the observations add up to 3.1 billiontourists, and 3.5 billion tourists with interpolation, a12.1% increase.Note that Equations (3) and (4) can be used to deriveinternational departures, just like Equation (2). Thecorrelation coefficient between these two alternatives is99.8%. We prefer (2) for its simplicity.WTO (2002) contains data on the number of nightsforeign tourists stay in selected countries. Dividing bythe number of foreign tourists, this leads to the averagelength of stay, S. This can be modelled as(Equation 5)−6 −1 −4S = 2.13− 2.58 H −1.91⋅ 10 A + 2.06⋅ 10 T + 1.72⋅10Ci i i i i0.61 0.62 0.79 0.40 0.78where H is a dummy for measurement in hotels only (asopposed to all establishments). All parameters aresignificantly different from zero. The income per capitain the destination country does not affect the length ofstay. Equation (5) says that tourists stay longer in hottercountries, in smaller countries and in countries withlonger coasts; tourists spend less time in the destinationcountry if they are accommodated in a hotel.WRI (2002) has data on the total expenditures ofinternational tourists. Dividing by the number of arrivalsand their length of stay, this yields expenditure pertourist per day, E, which can be modelled as(Equation 6)2E = − 611+ 0.029Y + 295 P R = 0.31; N = 47i i i200 0.007 71where P is the ratio of the purchasing power parityexchange rate to the market exchange rate. Expendituresincrease linearly with the average per capita income inthe holiday country. This is as expected. Surprisingly,there is no significant relationship between the averageincome of the tourists and their expenditures. There isalso no significant relationship between expenditures andincome distributions, a measured by the Gini coefficient,in either the destination or the origin country. Per capitaincome is measured in market exchange dollars. Thesecond explanatory variable in (6) is the ratio ofpurchasing power and market exchange rates. This ratiois high (up to 5) for the least developed countries andaround 1 for developed economies. Holidays are moreexpensive in poorer countries, probably becauseinternational tourists tend to be restricted to luxuryresorts.100%80%60%40%20%0%100%90%80%70%60%50%40%30%20%10%0%19801985199019952000200520102015202020252030203520402045205020552060206520702075208020852090209521001980198519901995200020052010201520202025203020352040204520502055206020652070207520802085209020952100100%80%60%40%20%0%100%80%60%40%20%0%19801985199019952000200520102015202020252030203520402045205020552060206520702075208020852090209521001980198519901995200020052010201520202025203020352040204520502055206020652070207520802085209020952100SISSSANAFCHISEASASSAMCAMMDEFSUCEEANZJPKWEUCANUSAFigure 1. The regional distribution of domestic tourists (top, left), international departures (top, right), internationalarrivals (bottom, left) and tourism receipts (bottom, right) for the A1B scenarios without climate change. The regionsare, from top to bottom: Small Island States; Sub-Saharan Africa; North Africa; China, North Korea and Mongolia;South East Asia; South Asia; South America; Central America; Middle East; Former Soviet Union; Central and EasternEurope; Australia and New Zealand; Japan and South Korea; Western Europe; Canada, and the USA.Annual Proceedings of Vidzeme University College “ICTE in Regional Development”, 2006106