Effects of the customs union with the European Union on the market structure and pricing behaviour of the Turkish manufacturing industry

Abstract

Turkey established a Customs Union with the European Union in 1996. This study aims to analyse the effect of that Customs Union on the market structure and the pricing behaviour of the Turkish manufacturing industry for the period 1994 to 2000. To that end, the price cost mark-up equation of 12 manufacturing industry sectors is estimated using the import and export ratios with European Union countries together with control variables. A second equation is also estimated for the concentration ratio index, taking the trade ratios with European Union countries as explanatory variables. The estimation method is panel data covering eight years and 12 cross-section units. Estimation outcomes indicate that the export and import ratios of trade with European Union countries have a negative relation with the price cost mark-up in the manufacturing sector. It is concluded that increased imports with union countries have created a positive wealth and efficiency effect upon the Turkish manufacturing industry, due to falling price cost mark-ups. Similarly, the concentration ratio equation estimation outcomes indicate that increased imports with union countries have induced a decline in the concentration ratio for the manufacturing industry during the period in question.

Notes

¹ See Dixit and Norman (1980), Brander (1981), Venables (1985) and Baldwin and Venables (1995).

² A detailed account of the trade liberalization policy can be found in Senses (1984 and 1988), Baysan and Blitzer (1990) and Togan (1994). For an evaluation of trade liberalization policy from a comparative perspective see Odekon (1988).

³ Examples of studies analysing the effect of an increase in imports from all sources on manufacturing industry profitability are Katırcıoğlu (1990), Foroutan (1991), Levinsohn (1993), Yalçın (2000) and Yeldan et al . (2000).

⁴ In order to account for the joint determination concentration ratio and price cost mark-up, a two-equation system is estimated by 2SLS. As the coefficient of the price cost mark-up in the concentration ratio equation proves statistically insignificant, the recursive system estimation is preferred.

⁵ Prior to the fixed effect model estimation, the F-test is performed to test for the behavioural differences among sectors. The F-test confirms the presence of sectoral differences. Accordingly, the fixed effect model estimation was performed by assuming time invariant differences among cross section units in terms of the intercepts.

⁶ Before the fixed effect model estimation, and to test for behavioural differences among sectors, the F-test was performed, supporting the presence of sectoral differences. Accordingly, the fixed effect model estimation was performed by assuming time invariant differences among cross section units in terms of the intercepts.

⁷ Since the import variable turns out significant in the concentration ratio equation estimation, one should consider a multicollinearity problem between the concentration ratio variable and the import variable for the price cost mark-up equation estimation. Though, multicollinearity test outcomes indicate no level of harmful collinearity between the two variables.

⁸ In the price cost mark-up equation, and in terms of the values of t-statistics, the lagged export variables are not significant. However, the lagged price cost margin variables are statistically significant in the export equation. For detailed information on Granger causality tests, see Madalla (1992, pp. 393–4).