10.08.2010 THE EFFECTS OF CUSTOMS UNION OVER TURKEY`S TRADE Furkan AYCAN Department of Economics, Istanbul Technical University, State University of New York [email protected]We use micro data for Turkey to understand the effects of the Customs Union Agreement on the trade volume and the efficiency between Turkey and the European Union. It is found that the Customs Union Agreement has a significant effect on increasing the trade positively. The previous studies have also shown that Turkey`s foreign trade has been positively affected by the Customs Union Agreement. Research topic: Customs Union and its effect on Turkish trade Research question: Has European Customs Union Agreement affected Turkish import and export? Research hypothesis: European Customs Union Agreement has been beneficial for Turkish trade so far, since after that the import and export of Turkey have substantially improved.
Research topic: Customs Union and its effect on Turkish trade
Research question: Has European Customs Union Agreement affected Turkish import and export?
Research hypothesis: European Customs Union Agreement has been beneficial for Turkish trade so far, since after that the import and export of Turkey have substantially improved.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
10.08.2010
THE EFFECTS OF CUSTOMS UNION OVER TURKEY`S TRADE
Furkan AYCAN
Department of Economics, Istanbul Technical University, State University of New York
area. It is possible to say that both Turkey and the European Union are putting effort in order
to expand the Customs Union in order to cover the uninvolved areas such as services and
public procurement. Also an important part of this agreement is the commitment of both the
European Union and Turkey, to extend and strengthen the Customs Union. After the
agreement upon the membership of Turkey to the Customs Union, the first achievement was
the free trade area that was established between the two sides for the products that are covered
by the European Coal and Steel Community7. A decision of the Association Council of the
European Coal and Steel Community includes trade in agricultural products.
The integration of the Customs Union, have boosted the momentum of Turkey’s trade
expansion to the majority of the world and provided a basis for future regional settlements
that will shape the international trade of Turkey. With the contribution of the changes that
occured as a result of the Customs Union, total volume of Turkey’s foreign trade have
substantially increased its share in the GDP. Right before the Customs Union Agreement in
the year 1995, the share of the volume of Turkey’s foreign trade in the GDP was 30,6%; and
in the year 2008, it has increased to 45%.8
Theoretically, assuming the optimum use of the resources, the highest level of welfare
will be gained when the free trade conditions are developed. However Customs Union begins
with the removal of the trade barriers between the member countries and then implementing
and following a common tariff against the third countries. With this function, Customs Union
is a privileged free trade for the member countries.9
As a result of the initial research about the subject, it can be said that there are both
positive and negative effects. It has been observed that Turkey’s trade has been positively
7 European Coal and Steel Community was a six nation international organization serving to unify Western Europe during the Cold War and create the foundation for the modern-day developments of the European Union.8 Undersecretariat of the Prime Ministryfor Foreign Trade9 Hitris (2003)
affected as a result of the Customs Union Agreement. However, it can also be argued that
some sectors have been negatively affected by the Customs Union Agreement. These sectors
are mainly inward-oriented which produce for the domestic market and after the agreement
coming into action, they became unable to face the competition and were forced to shut down.
It is observed that export in these sectors have substantially decreased. In contrast to that,
some sectors got influenced positively by the effects of the Customs Union Agreement. The
production and the export of these sectors have improved. The automotive industry can be
given as a suitable example to this situation. As Sakıncı argued in his study, the automotive
sector have been positively affected by the Customs Union.10 His econometric findings show
that the industry have been significantly expanded and the export volume have substantially
increased. Ozkale have estimated Turkey’s import demand function with the help of random
effects model by using panel data method. It is estimated that Customs Union have trade-
attractive and creative effects for the main import goods. In the sectoral analysis, it is
observed that some sectors have been positively affected while some of them haven’t got
affected at all.11
Adam and Moutos argued in their study that before the Customs Union Agreement
came into effect, Turkey had a major change in its development policy and strategy. The
change was from import substitution to export oriented policies, that resulted in a significant
decrease in protectionist means.12 With the acception of the agreement, the technical barriers,
which were concluded as big obstacles for trade, have been removed. This removal is
expected to have a positive effect over the Turkey’s trade. Turkey agreed to adopt some basic
mechanisms of the European Union such as standardization, quality and accreditation. In
addition to these, there were objectives to harmonize the regulations and the policies that
10 Sakinci (2008)11 Ozkale (2006)12 Adam and Moutos (2008)
relate to the economic areas such as industrial and intellectual property rights, competition
rules, state aid, the Custom code and administrative cooperation. So, it is not right to conclude
the Customs Union Agreement as a change in tariff and some other structures. It is better to
conclude as series of changes that are likely to have an accumulated effect on total trade level.
According to the study of Harrison, the most beneficial effect of Customs Union for
Turkey is the accessibility of the third country markets. Turkey’s annual national income is
predicted to increase by %1 – 1,5 after the market accessibility as a result of the Customs
Union, especially together with the application of the common trade strategy with the
European Union.13 Besides that, a study by Mercenier and Yeldan states the adverse effects of
the Customs Union. This study proposes that the Customs Union would lead to revenue losses
for the government sector and in GDP, unless the trade reforms are increased and the barriers
should be removed.14 According to their argument, only the private sector would benefit from
the Customs Union Agreement.
In a research made by Halıcıoğlu , the static effects of the Customs Union over the
Turkish economy have been examined with partial equilibrium analysis. In order to do this,
the effects that lead to expansion or deflection of trade have been estimated, and also how
these criterias would be affected if Turkey would be a member of another trade block. As a
result of these queries, it is mentioned that the Customs Union has been concluded as the
second best advancement after the application of free trade. In addition to this, it is calculated
that the public welfare will increase by %1.28 of the national income.15
Neyaptı, Taskin, Ungor also attempted to investigate the effects of the Customs Union
over Turkey’s foreign trade, with the help of some demand functions. Positive effects have
13 Harrison (1996)14 Mercenier and Yeldan (1997)15 Halicioglu (1997)
been spotted between the change in Turkey’s national income and its export, as well as the
change in those countries, who trade with Turkey, and their import. This study argues that the
Customs Union have increased the international trade with the member countries of the
European Union, and also caused a shift over Turkey’s international trade, from the other
countries to European countries.16
After the establishment and the implementation of the Customs Union Agreement, it is
possible to observe a regular increase in Turkey’s export to the European and the other
country markets.17 Especially the stability in Turkey’s export to the members of the European
Union is very much influenced by the agreement, which eventually led to being Turkey’s
most important trade partner in both export and import, with a total share of approximately
%50 in year 2008.18 According to the data retrieved from the Undersecretariat of the Prime
Ministry for Foreign Trade, it is possible to see the improvement in the economic relations
between Turkey and the European Union have improved after 1996.
Table 1. Turkey’s Import, Export with EU and Foreign Trade Balance between years 1986-
Phillips Perron, 1st difference test on ln(GDPturkey)Null Hypothesis: D(LGDP1) has a unit rootExogenous: ConstantBandwidth: 0 (Newey-West using Bartlett kernel)
*MacKinnon (1996) one-sided p-values.Warning: Probabilities and critical values calculated for 20 observations and may not be accurate for a sample size of 16
Residual variance (no correction) 0.033570HAC corrected variance (Bartlett kernel) 0.033570
Phillips-Perron Test EquationDependent Variable: D(LGDP1,2)Method: Least SquaresDate: 08/03/10 Time: 11:29Sample(adjusted): 1993 2008Included observations: 16 after adjusting endpoints
C 0.111971 0.054511 2.054101 0.0591R-squared 0.589957 Mean dependent var 0.004485Adjusted R-squared 0.560669 S.D. dependent var 0.295513S.E. of regression 0.195872 Akaike info criterion -0.306240Sum squared resid 0.537123 Schwarz criterion -0.209666Log likelihood 4.449916 F-statistic 20.14278Durbin-Watson stat 1.963932 Prob(F-statistic) 0.000511
Phillips Perron test, 1st difference on ln(GDPitaly)Null Hypothesis: D(LGDP2) has a unit rootExogenous: ConstantBandwidth: 0 (Newey-West using Bartlett kernel)
*MacKinnon (1996) one-sided p-values.Warning: Probabilities and critical values calculated for 20 observations and may not be accurate for a sample size of 16
Residual variance (no correction) 0.009032HAC corrected variance (Bartlett kernel) 0.009032
Phillips-Perron Test EquationDependent Variable: D(LGDP2,2)Method: Least SquaresDate: 08/03/10 Time: 11:31Sample(adjusted): 1993 2008
Included observations: 16 after adjusting endpointsVariable Coefficient Std. Error t-Statistic Prob.
*MacKinnon (1996) one-sided p-values.Warning: Probabilities and critical values calculated for 20 observations and may not be accurate for a sample size of 17
Residual variance (no correction) 2.06E-07HAC corrected variance (Bartlett kernel) 4.68E-07
Phillips-Perron Test EquationDependent Variable: D(LPOP1)Method: Least SquaresDate: 08/03/10 Time: 11:31Sample(adjusted): 1992 2008Included observations: 17 after adjusting endpoints
C 0.461661 0.027797 16.60824 0.0000R-squared 0.945049 Mean dependent var 0.015204Adjusted R-squared 0.941385 S.D. dependent var 0.001996S.E. of regression 0.000483 Akaike info criterion -12.32190Sum squared resid 3.50E-06 Schwarz criterion -12.22388Log likelihood 106.7362 F-statistic 257.9693Durbin-Watson stat 0.262716 Prob(F-statistic) 0.000000
Phillip Perron test, 2nd difference on Dln(POPitaly)Null Hypothesis: D(DLPOP2,2) has a unit rootExogenous: ConstantBandwidth: 13 (Newey-West using Bartlett kernel)
*MacKinnon (1996) one-sided p-values.Warning: Probabilities and critical values calculated for 20 observations and may not be accurate for a sample size of 14
Residual variance (no correction) 3.69E-06HAC corrected variance (Bartlett kernel) 4.36E-07
Phillips-Perron Test EquationDependent Variable: D(DLPOP2,3)Method: Least SquaresDate: 08/03/10 Time: 11:34Sample(adjusted): 1995 2008Included observations: 14 after adjusting endpoints
C 3.94E-05 0.000556 0.070938 0.9446R-squared 0.436068 Mean dependent var -6.93E-05Adjusted R-squared 0.389073 S.D. dependent var 0.002654S.E. of regression 0.002075 Akaike info criterion -9.386538Sum squared resid 5.16E-05 Schwarz criterion -9.295244Log likelihood 67.70577 F-statistic 9.279154Durbin-Watson stat 1.786429 Prob(F-statistic) 0.010157
Phillip Perron test, 1st difference on ln(TRADE)Null Hypothesis: D(LTRADE) has a unit rootExogenous: ConstantBandwidth: 1 (Newey-West using Bartlett kernel)
*MacKinnon (1996) one-sided p-values.Warning: Probabilities and critical values calculated for 20 observations and may not be accurate for a sample size of 16
Residual variance (no correction) 0.022120HAC corrected variance (Bartlett kernel) 0.022279
Phillips-Perron Test EquationDependent Variable: D(LTRADE,2)Method: Least SquaresDate: 08/03/10 Time: 11:35Sample(adjusted): 1993 2008Included observations: 16 after adjusting endpoints
R-squared 0.415020 Mean dependent var 0.111754Adjusted R-squared 0.280025 S.D. dependent var 0.150972S.E. of regression 0.128102 Akaike info criterion -1.069663Sum squared resid 0.213330 Schwarz criterion -0.873613Log likelihood 13.09213 Durbin-Watson stat 1.552459