Presented at the Ist International Symposium on Innovative Approaches in Scientific Studies (ISAS 2018), Kemer-Antalya, Turkey, Apr 11, 2018

SETSCI Conference Proceedings, 2018, 2, Page (s): 175-176 , https://doi.org/

Published Date: 23 June 2018    | 1210     9

To investigate how foreign trade deficits and budget deficits of Turkey navigated pre and post customs union (CU) with European Union. In the study, for the budget deficits (BD), General Directorate of Budget and Fiscal Control (BUMKO) and also for the foreign trade deficits (FTD), Turkey Statistical Institute (TUIK) records were used. Records from 1975-1995 before UC, and 1996-2016 for UC were used. Series has been converted into Turkish Liras (TL) unit according to the foreign exchange sales rates of the Central Bank. Budget (BD) and foreign trade deficit (FTD) series which took place between 1975 and 2016 according to Turkey. Statistical Institute (TUIK) and the General Directorate of Budget and Fiscal Control (BUMKO), were used in the study. Foreign trade series consist of US Dollar (USD), Budget Deficit Turkish Lira (TL) and New Turkish Lira (YTL). Firstly, by using the USD exchange rate, the series are converted into TL and YTL units. Subsequently, by using the 1975 Istanbul Chamber of Commerce (ITO) wholesale price index (TEFE), the series are reduced to 1975 fixed prices. Finally, with the reason for the transition to YTL unit in 2006, six zeroes were taken from the serial members until 2006 to equalize the units. EViews 9 package program is used for statistical tests. In order to operate with time series, the series have to be stationary. The stationary test was performed using Augmented Dickey_Fuller (ADF) method. For both BT and DT series, stationary was determined by 1 st differential, so a cointegrated regression model was established and error correction mechanism applied. The significance of the model was tested by F method. The aim of reaching the result is considered by interpreting the regression equations belonging to models. (Pre-CU) For the first differences, the ΔBD1 and ΔFTD1 series are stationary (ADFΔBD1 p = 0.0000, ADFΔFTD1 p = 0.0000) and ΔBD1 and ΔFTD1 are co-integrated from the 1st level I (1). The error correction coefficients (EC1) were obtained. The EC1 coefficients are stationary at level values (ADFEC1 p = 0.0031). ΔBD1 dependent variable, ΔFTD1 and 1 delayed ECt-1 series are independent variables. The regression equation is: ΔBD1 = 3.008866 + 0.379167 ΔFTD1 - 0.841345 EC1(t-1). The equation is significant (F p = 0.000193). Adjusted R2=0,591344. Durbin-Watson d=1.791467. (Post-CU) For the first differences, the ΔBD2 and ΔFTD2 series are stationary (ADFΔBD2 p = 0.0007, ADFΔFTD2 p = 0.0000) and ΔBD2 and ΔFTD2 are co-integrated from the 1st level I(1). The error correction coefficients (EC2) were obtained. The EC2 coefficients are stationary at level values (ADFEC2 p = 0.0031). ΔBD2 dependent variable, ΔFTD2 and 1 time delayed ECt-1 series are independent variables. The regression equation is: ΔBD2= 1.694511 - 0.321704.EC2(t-1) - 0.512637.ΔFDT2. The equation is significant (F p=0.000383). Adjusted R2=0,557041. Durbin-Watson d=1.221074. The 1-unit increment in pre-GB foreign trade deficit reflects an increment of 0.379 units in the budget deficit. After GB, 1 unit increment in foreign trade deficit reflected as 0.513 unit decrement in budget deficit. In other words, after CU, we can reach a result like "budget revenues have increased dependency on import taxes".  

Keywords - Foreign Trade Deficits, Budget Deficits, ADF Test, Cointegration, Error Correction, Customs Union