Economic & static Project.
Annki lee
ForecastingUSTourisminCaribbeanEconomies.pptxForecasting US Tourism in Caribbean Economies
Introduction
Small islands face interesting obstacles towards growth and stability
Small islands lack: natural resources, economies of scale, close neighbors by land
There is however, much potential for growth in tourism-related ventures
Many small islands rely heavily on the tourism industry for growth and revenue
What Islands We Will Study
There have been previous studies on the Eastern Caribbean Currency Union (ECCU):
Antigua & Barbuda
Dominica
Grenada
St. Kitts and Nevis
St. Lucia
St. Vincent and the Grenadines
These islands rely heavily on tourism from the US: we will focus on US tourism as a source for tourists.
Also, the ECCU is a convenient group of islands to study since some data is readily provided for all islands in the ECCU,
Hypothesis
I expect that tourism in the Caribbean with the US as a source country is heavily affected by income, price, and supply factors
We expect US tourists to these countries to take into account their income.
We expect US tourists to these countries to take into account the costs involved
We expect US tourists to these countries to take into account the ease of access to the island and the things that the island has to offer.
Dependant Variable
TDit – The number of tourist arrivals in island i at time t .
More specifically, the number of stay-over arrivals.
Tourists that spend the night on the island.
We expect these tourists to actually spend money on the island as opposed to cruise ship arrivals.
In the end, stay-overs are more important for government policy since we expect them to bring in the majority of revenue.
Independent Variables
INCt - the average income in the top 40% of income earners in the US at a given time t
We use the top 40% variable since only people with money can afford this kind of travel.
Expected sign is positive.
REERit - the real effective exchange rate between the US and given island i at a given time t
Expected sign is positive.
Independent Variables
OILt - the average crude price of petroleum at a given time t
Expected sign is negative.
AIRit - the number of flights serving a given island i at time t
Expected sign is positive.
FDIit - the level of foreign direct investment for a given island i at time t
Expected sign is positive.
Model
lnTDit = ß0i + ß1 ln INCt + ß2 ln REERit + ß3 ln OILt + ß4 ln AIRit + ß5 FDIit + αidi+ vit
With t= 1,…,20 (1990-2009) and i=1,…6 which represents the island.
di is a dummy variable with i=2,…6 and the country Antigua & Barbuda as the benchmark.
vit is an error term
Results
VARIABLE COEFF. STD. ERROR T-RATIO PVALUE
LNINC 0.53689 0.3488 1.539 0.127
LNREER 0.46599 0.3882 1.200 0.233
LNOIL 0.19797 0.04952 3.998 0.000
LNFDI -0.023843 0.01486 -1.605 0.111
LNFLY 0.077191 0.01560 4.948 0.000
R-SQUARE = 0.9164 R-SQUARE ADJ = 0.9087
Potential Problems with Heteroskedasticity
Our Chi-square test statistic: 3.880
P-value = .04887
We would reject our null hypothesis for any α ≥ .04887
If α ≥ .04887, then we would conclude that there is heteroskedasticity in our data
Corrected Results (HETCOV)
VARIABLE COEFF. STD. ERROR T-RATIO PVALUE
LNINC 0.53689 0.4025 1.334 0.185
LNREER 0.46599 0.3752 1.242 0.217
LNOIL 0.19797 0.04629 4.277 0.000
LNFDI -0.023843 0.01023 -2.331 0.022
LNFLY 0.077191 0.01146 6.736 0.000
R-SQUARE = 0.9164 R-SQUARE ADJ = 0.9087
Other Tests
No multicollinearity
Every pairwise correlation has an absolute value greater than .8
Highest R-Square value of an individual variable on all other independent variables is .6934.
No first order autocorrelation
Model’s Durbin Watson Stat: 2.19545
Since p-value>.10 ; .828221>.10
Conclusions
There is a positive relationship between income and tourism demand.
We can not make any definitive statements on the price determinants and how they affect tourism demand.
REER has expected sign, but not significant
OIL has unexpected sign
The number of flights serving a given destination is a significant supply factor.
FDI however has an unexpected sign but if also not significant
Further Remarks
Perhaps using the “top 40%” variable threw off some of our estimates.
This would explain our unexpected elasticities (coefficients)
Why did some variables have unexpected signs in our study compared to different studies?
Different time span
Different variables used
Ideas for Further Study
An update to this study would be appropriate in the future since we can expand the years we looked at.
Further studies on supply factors would be ideal such as type and cost of accommodation
Inclusion of other islands around the world
Average GDP per capita VS top 40%
Which is more appropriate?
Sources
"Air Carriers : T-100 International Segment (All Carriers)." Bureau of Transportation Statistics. Research and Innovative Technology Administration, July 2010. Web. 10 Feb. 2011. <http://www.transtats.bts.gov>.
Craigwell, Roland, and DeLisle Worrell. The Competitiveness of Selected Caribbean Tourism Markets. Central Bank of Barbados. Mar. 2008. Web. 10 Feb. 2011. <http://www.centralbank.org.bb/WEBCBB.nsf/0/CD5AF9F3006204400425766400760026?OpenDocument>.
Gonzales, P. and Moral (1995), “An Analysis of the International Tourism Demand in Spain,” International Journal of Forecasting, Vol. 11, pp. 233-51. 9 Mar. 2011.
"Inward and Outward Foreign Direct Investment Flows, Annual." UNCTADStat. 19 Oct. 2010. Web. 10 Feb. 2011. <http://unctadstat.unctad.org>.
Ledesma-Rodriquez, F.J and M. Navarro-Ibanez (2001), “Panel Data and Tourism: A Case Study of Tenerife,” Tourism Economics, Vol. 7(1), pp. 75-88. 9 Mar. 2011.
Lundberg, Donald E., M. Krishnamoorthy, and Mink H. Stavenga. Tourism Economics. New York: Wiley, 1995. Print.
Mak, James. Tourism and the Economy: Understanding the Economics of Tourism. Honolulu: University of Hawai'i, 2004. Print.
"Mean Household Income Received by Each Fifth (Table H-3)." Current Population Survey. U.S Census Bureau, 2010. Web. 10 Feb. 2011. <http://www.census.gov>.
Önder, A. Özlem, Aykan Candemir, and Neşe Kumral. "An Empirical Analysis of the Determinants of International Tourism Demand: The Case of Izmir." European Planning Studies 17.10 (2009): 1525-1533. Academic Search Elite. EBSCO. Web. 9 Mar. 2011.
"Petroleum: Average Crude Price." International Financial Statistics. International Monetary Fund, 2011. Web. 10 Feb. 2011. <http://www.imfstatistics.org/imf/>.
"Real Effective Exchange Rate” International Financial Statistics. International Monetary Fund, 2011. Web. 18 Feb. 2011. <http://www.imfstatistics.org/imf/>.
Song, Haiyan, and Gang Li. "Tourism Demand Modelling and Forecasting: A Review of Recent Research." Surrey Research Insight. University of Surrey, 2007. Web. 7 Mar. 2011. <http://epubs.surrey.ac.uk/1085/1/fulltext.pdf>.
"Stayover Arrivals from US by Country of Destination." OneCaribbean. Caribbean Tourism Organization. Web. 10 Feb. 2011. <http://www.onecaribbean.org/statistics/tourismstats/>.
Tsounta, Evridiki. What Attracts Tourists to Paradise? International Monetray Fund, Dec. 2008. Web. 10 Feb. 2011. <http://www.imf.org/external/pubs/ft/wp/2008/wp08277.pdf>.
Witt, S. and C. Witt (1995), “Forecasting Tourism Demand: A Review of Empirical Research,” International Journal of Forecasting, Vol. 11, pp. 447-75.
