THE ROLE OF INTERNATIONAL TOURISM IN IMPROVING LIVING STANDARDS IN UZBEKISTAN: A MULTI-EQUATION ECONOMETRIC ANALYSIS WITH ARDL BOUNDS TESTING

Abstract

This study investigates the multidimensional impact of international tourism on household living
standards in Uzbekistan over the period 2000–2023, employing a comprehensive three-equation econometric
framework alongside an Autoregressive Distributed Lag (ARDL) bounds testing approach. Three dependent
variables operationalize distinct dimensions of living standards: (i) GDP per capita (GDPPC) as an income
welfare proxy; (ii) the national poverty headcount rate (POV) as a distributional welfare indicator; and (iii) the
Human Development Index scaled score (HDI) as a multidimensional welfare composite. International tourism
is measured by the logarithm of annual tourism receipts in USD million (ln(TOUR)), sourced from the World
Bank World Development Indicators and the United Nations World Tourism Organization. Augmented Dickey-
Fuller (ADF) unit root tests confirm that all primary variables are integrated of order one — I(1) — in levels but
stationary in first differences, satisfying preconditions for ARDL cointegration analysis. Three OLS regression
models — estimated with heteroskedasticity-robust HC3 standard errors and validated through Breusch-
Pagan, Shapiro-Wilk, Jarque-Bera, Durbin-Watson, and VIF diagnostic tests — yield theoretically consistent
and statistically significant results. Model 1 demonstrates that a 1% increase in tourism receipts is associated
with a 0.409% increase in GDP per capita (β = 0.409, p < 0.001, Adj. R² = 0.874), confirming the tourism-led
income growth hypothesis. Model 2 establishes that a 1% increase in tourism receipts is associated with a
0.340% reduction in the poverty headcount rate (β = −0.340, p < 0.001, Adj. R² = 0.961), providing statistically
significant evidence for the pro-poor tourism hypothesis. Model 3 reveals that a 1% increase in tourism receipts
is associated with a 1.821-unit increase in the HDI score (β = 1.821, p < 0.001, Adj. R² = 0.962), confirming a
positive human development impact. The ARDL bounds test confirms long-run cointegration (Bounds F = 9.69,
exceeding the Pesaran et al. (2001) upper critical bound of 4.01 at 5%), with a long-run tourism elasticity of GDP
per capita of 0.370 and an error correction term of ECT = −0.311, implying convergence to long-run equilibrium
at approximately 31% per year. Granger causality tests suggest unidirectional causality from tourism to income,
poverty, and HDI, though formal significance is constrained by the sample size. Pearson correlations strongly
support all modeled relationships (r = 0.923 for ln(TOUR)–lnGDPPC; r = −0.949 for ln(TOUR)–lnPOV; r = 0.967
for ln(TOUR)–HDI; all p < 0.001). These findings carry direct policy implications for the tourism sector targets
set out in the Decree of the President of the Republic of Uzbekistan No. PF-21 dated February 16, 2026, “On
additional measures to consistently continue reforms and bring them to a new stage within the framework of the
country’s development priorities until 2030”, particularly the targets of increasing the annual number of foreign
tourists to 20 million and raising tourism services exports to USD 6 billion [1]