erc/metu
INTERNATIONAL CONFERENCE IN ECONOMICS  IV
September 13-16, 2000, Ankara

Domestic and Imported Core Inflation in Turkey

Kamuran Malatyalı (SPO)
Cengiz Cihan (SPO)

Abstract

As the practice of inflation targeting has become widely resorted monetary policy framework in the world, the concept of core inflation is becoming a dominant reference point both in the literature and application. The crux of the concept relies on capturing the effect of monetary policy actions in the inflationary development. To extract the trend inflation, and thus determining the monetary part of inflation process, is the subject matter of core inflation studies.

Inflation process is a monetary phenomenon. Therefore, the development of studies on core inflation involved in determining the monetary part of inflation by extracting the seasonal, administered or temporary shock factors. Hence, Bryan and Cecchetti (1993) adopts a view of determining the monetary part of inflation by applying VAR analysis. For Turkey such a study was performed by Sakarya, Yurtoglu and Duvan (1999).

Although inflation is a monetary phenomenon and it seems logical to derive this part of inflation as the core inflation it might still well be argued that this is not sufficient to capture the real core inflation part. It is a proven fact that there is a vertical long run Phillips curve which represents that all type of shocks are output neutral in the long run, including the monetary ones. Hence, in this study we aim to approach the problem of core inflation in this perspective.

Using the guidelines proposed in Quah and Vahey(1995) we adopt the rationale that core inflation is to be identified as the component of inflation that has no long-run effect on real output. The authors clarify core and non-core parts of inflation by using a structural VAR methodology. In this respect, core shocks are the ones which have no long-run effects on real output while the non-core shocks such as taxes, energy shocks, price controls and productivity changes affect real output. Based our study on this view we further aim to capture the domestic and imported parts of core inflation as proposed in Bjornland(2000). As a small, open, non-oil producing economy Turkey imports a large amount of her inflation. Thus, it is crucial for the monetary authorities to detect the price signals clearly. Hence, in determining the imported part of core inflation we identify two types of prices, in addition to the domestic prices; firstly the oil prices which creates shocks on a non-oil producing country and secondly, the weighted average of price levels of Turkey’s major trading partners. Using this approach, this study is important being the first study to isolate imported core component from the domestic one in Turkey.

The data used in this study are industrial production index, consumer price index (1987=100), weighted average of price levels of Turkey’s main trading partners and average crude oil prices. The sample size of the study ranges within January,1987-January,2000.

Economic Research Center
Middle East Technical University
06531 Ankara Turkey
Phone: + 90 312 210 3044, 210 2003
Fax: +90 312 210 1244
e-mail: metuerc@metu.edu.tr