Econometric modelling of inbound tourist expenditure in South Africa

Transcription

1 Economeric modelling of inbound ouris expendiure in Souh Africa Paper prepared for CBTS 2011, Brunico, Ialy by Andrea Saayman and Melville Saayman Norh-Wes Universiy, Pochefsroom Campus

2 Agenda Inroducion Lieraure Review Mehod of Research Model and daa Economeric mehod Resuls Conclusion

3 Inroducion Souh African governmen realise poenial of ourism for developmen & growh Growh in arrivals since 1994 improve 20 posiions on he mos visied desinaion lis Arrivals versus spending Susainabiliy? Aim: o deermine sensiiviy of ouris spending o changes in income, prices

4 Lieraure Review Tourism demand: Measure of visiors use of goods and services (Frechling, 2001) Need o make use of service or acquire commodiy and purchase akes place (Song e al., 2010) Difference beween arrivals and spending Daa recording Is applicaion Evoluion of daa series over ime Mos demand models focus on arrivals (Song & Li, 2008) Economeric modelling allows for economic inerpreaion, policy recommendaions and evaluaion Mos popular independen variables income, relaive prices, quaniaive facors, ranspor cos

5 Lieraure Review (2) Tourism demand for Souh Africa Saayman & Saayman (2008) and Seeanah e al. (2010) Arrivals is income elasic, bu price inelasic Low cross-price elasiciies and disance/ ranspor cos plays significan role Song e al. (2010) expendiure as dependen variable lead o more accurae elasiciies for ourism o Hong Kong Prices more prominen in explaining expendiure Income more prominen in explaining arrivals Phakdisoh & Kim (2007) for Laos Lee e al. (1996) for Souh Korea

6 Mehod: Model and Daa Follow a micro-economic approach: ( s Y, P, P, P Z) Dij = F i i j j, Demand is funcion of income, price, subsiue prices, ases and preferences In erms of relaive prices: D s Y P P = F i j j,, Z Pi Pi Pi ij, Demand measured as inernaional ouris spending in SA per day (SA Tourism) Jan 2003 o Dec 2010 Models esimaed for: UK, USA, Germany, France, he Neherlands, Ialy, Ausralia, Brazil, India, China, Boswana

7 Mehod: Model and Daa Income of origin counry real GDP index (IFS) Wi & Wi (1995) wo prices or coss ouriss encouner: Transpor cos price of oil is used (IFS) Cos of living in he desinaion real exchange rae (IFS) P = ( CPI CPI ) * e i j Subsiue price closes African compeiors (IFS) s ( P + P P ) 3 P = + Bo, Ken, Tan,

8 Economeric Mehod Linearised demand funcion: i s = β 0 + β1 ln Yi + β 2 ln Pi + β 3 ln P + β 4 ln D TC + u Inspeced uni roo properies using ADF and PP Some variables I(0), ohers I(1) Follow an ADLM specificaion y p = α0 + αi y i + β j x j + u i= 1 n j= 0 Useful wih shor ime series Expand wih an error-correcion erm o accoun for long-run y n = α 0 + α1 y 1 + β j x j + λz 1 + u j= 0 i

9 Economeric Mehod Since all variables no inegraed of he same order, we follow Bound es approach for coinegraion (Pesaran e al., 2001) ADLM is esimaed wih an EC componen Wald es deermine wheher coefficiens of EC significanly differen from 0 F-criical values provided by Pesaran e al. (2001) BOT BRA AUS USA CHI IND UK FRA GER NL ITA Tes saisic

10 Resuls: Long run elasiciies BOT BRA AUS USA CHI IND UK FRA GER NL ITA Price Income Fuel Subsiue Price Touris spending is relaively price inelasic, excep for ouriss from France and Ialy Demand is income elasic, excep for ouriss from Brazil, India, UK and he Neherlands Fuel elasiciy also generally less han -1 Relaive high subsiue price elasiciies Desinaions are complemens excep for ouriss from Brazil, USA and Ialy where hey are subsiues

11 Resuls: Shor-run elasiciies Shor run elasiciies mainly confirm long-run elasiciies Price changes: Influence spending for up o 3 quarers Show elasiciies around uniy Chinese ouriss have higher price elasiciies Transpor cos is relaive inelasic Income changes: Mosly greaer han uniy ourism is a luxury good Subsiue price elasiciies: Mainly negaive and inelasic complemenary desinaions Ialians view Boswana, Kenya, Tanzania as subsiues posiive elasiciy

12 Resuls: Model diagnosics Counry BOT BRA AUS USA CHI IND UK FRA GER NET ITA Adj R Akaike Schwarz Jarque- Berra LM ARCH [0.71] [0.85] [0.37] [0.63] [0.09] [0.68] [0.43] [0.56] [0.85] [0.61] [0.26] [0.72] [0.64] [0.51] [0.19] [0.49] [0.81] [0.89] [0.71] [0.07] [0.35] [0.80] [0.74] [0.61] [0.98] [0.45] [0.38] [0.78] [0.30] [0.61] Opimal lags beween 1 and 3 Insignifican variables excluded for parsimony Price variables someimes excluded due o mulicollineariy problems [0.62] [0.58] [0.35]

13 Conclusion Aim was o deermine sensiiviy of inbound ouriss expendiure in Souh Africa o changes in price, income and subsiue prices Main findings: Some elasiciies differen han ha of arrivals SA sill relaive price inelasic desinaion Tourism spending in SA is a luxury good Oher similar African desinaions are raher complimens han subsiues Implicaions: Inernaional economic condiion will influence ourism o SA significanly Price compeiiveness remains paramoun some desinaions are already showing subsiuion Souhern Africa mus sar hinking abou markeing and posiioning he region no jus a counry

14 Thank you