Transcript Income and Price Elasticities of Croatian Trade – A Panel

The Fifteenth Dubrovnik
Economic Conference
Organized by the Croatian National Bank
Income and Price Elasticities of
Croatian Trade – A Panel Data
Approach
by Vida Bobic
Discussant: K. Zigic CERGE-EI Prague,
Czech Republic
Summary of the Paper
• Objective:
– analysis of factors determining Croatian merchandise trade
• Motivation: large trade deficits of Croatia
• Methodology:
– Panel data for 2000-2007, sectoral level (30 NCEA sectors)
– Models with lagged explained variable estimated by Arellano
& Bond (1991) methodology
– Domestic and foreign commodities treated as imperfect
substitutes
– Independent estimates of export and import function
• Imports assumed to depend on domestic GDP, import price, tariffs,
exchange rate
• Exports function of world GDP, export price, and exchange rate
– Croatia treated as a small country
Comments (1): Motivation
• Page 1:
– “… current account deficit. This deficit is, in turn, for
the most part a consequence of a large deficit in
merchandise trade”
• Small conceptual objection
– Deep current account deficits are mostly results of
deeper macroeconomic disequilibria
– While Croatia indeed has deep trade deficit, the
current account deficit is lower thanks to services
account
Current Accounts of Selected New EU Members
% of GDP
Source: World Development Indicators
Comments (2): Econometric Methodology
• Page 9, equation 2 – demonstration of
Arellano & Bond framework
– Shouldn’t the equation include xit instead of xit ?
– Could Blundell & Bond (1998) further improve the
methodology?
Comments (3a): Variables
• Tariffs omitted from export equation because:
– “The very large number of countries to which goods
are being exported and the resulting equally large
number of different tariffs being applied to those
exports makes the construction of a single tariff
indicator very difficult”
– But: EU accounts for about 60% of Croatian exports
and it has common tariff schedule
– How about using the EU common tariff as proxy?
Comments (3b): Variables
• Tariff measure used in the import equation is
derived as tariff revenues / total value of
imports
• Problem: interaction between tariff rate and
volume of imports, higher protection does not
always mean higher revenue from the tariff
• Possible solution:
– Simple non-weighted average tariffs are often used
in trade literature (they are imperfect too, but may
lead to less biased results in the presence of highly
protected sectors)
Conclusion
• Author met the main objective (estimation of
income and price elasticities)
• Estimates are made with the use of more
reliable methodology than many previous
applied papers (simple fixed effects or OLS are
not uncommon)
• Results seem to be in line with both previous
estimates and expectations. However, look
more into the issue of quality upgrading
References
• Arellano, M., & Bond, S. (1991). Some tests of
specification for panel data: Monte Carlo
evidence and application to employment
equations. Review of Economic Studies, 58, 227–
297.
• Blundell, R. W., & Bond, S. R. (1998). Initial
conditions and moment restrictions in dynamic
panel data model. Journal of Econometrics, 87,
115–143.
• World Bank: World Development Indicators