Impact of oil price changes on Turkey’s exports

ABSTRACT

Substantial decline in oil prices observed since the second half of 2014 has brought forward discussions on the impact of oil prices on the external balance. This decline is expected to shift real income from oil exporter countries to oil importer countries. In this context, we study the effect of change in oil prices on Turkey’s exports taking into account the fact that Turkey’s export to oil exporting countries has approximately one-third share in total exports, while export to oil importing countries has a share of two-thirds. At the first step, we analyse the effect of oil prices on economic growth of trade partners of Turkey, where countries are grouped according to their net oil export position. We find that the effect of oil prices on different country groups’ growth is asymmetric. In the second stage, for two groups of countries, we estimate growth-export elasticity through Turkey’s export demand function. Finally, we calculate the net effect of oil prices on exports considering different income effects caused by oil price change on each country groups. Results indicate that the net effect of oil prices on the exports is limited given the current export shares.

KEYWORDS:

• Oil prices
• exports
• Turkey
• dynamic panel

JEL CLASSIFICATION:

• F14
• Q43

Acknowledgements

The authors thank Koray Kalafatcılar and Altan Aldan for their contributions. The views and opinions expressed in this study belong solely to the authors and do not under any circumstance reflect the views or opinions of the Central Bank of the Republic of Turkey or its staff.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Classification is from the IMF-World Economic Outlook. The list of countries is in the Appendix.

² Given the recent (extraordinary and temporary) trends of gold trade, export figures exclude gold.

³ Similar estimation strategy is exercised by Rasmussen and Roitman (2011).

⁴ Given that system GMM estimation is sensitive to number of instruments, the results are presented as a range.

⁵ For similar results, see Berument, Ceylan, and Dogan (2010), Rasmussen and Roitman (2011), IMF (2000) and Alkhathlan (2013).

⁶ Similar results are reported by Çulha and Kalafatcılar (2014) and Bozok, Şen Doğan, and Yüncüler (2015).