Hakan Kara

Inflation Targeting and Exchange Rate Pass-Through: The Turkish Experience

Using a vector autoregression model, we show that the pass-through from imported inflation to domestic inflation has weakened substantially and slowed after the adoption of inflation targeting in Turkey. We argue that this finding is due mainly to several features—such as enhanced credibility of the central bank, changing behavior of the exchange rate, and a shift in expectation formation—possibly acquired by the implementation of a successful inflation-targeting regime. These observations suggest that adopting an inflation-targeting regime in itself may help to reduce exchange rate pass-through.

Inflation expectations in Turkey: learning to be rational

Turkey implemented an ambitious restructuring of the economy in the past several years, including the adoption of inflation targeting along with a floating exchange rate regime. Inflation came down from almost triple digits to single digits between 2001 and 2005. This particular episode of the Turkish economy sets a genuine case study for investigating the possible changes in the behaviour of inflation expectations upon a regime shift. Accordingly, this study analyses inflation expectations in Turkey, focusing especially on the post-2001 transition phase. We first conduct classical tests of unbiasedness and efficiency using aggregate survey data between August 2001 and October 2007 to get a statistical benchmark for rationality; we find that classical tests reject full rationality hypothesis for all series except next months Consumer Price Index (CPI) inflation expectations. Then, we carry out Time-Varying Parameter (TVP) estimates based on a Kalman filter to see how the coefficients in the classical test equations evolve over time. This framework allows us to see whether there is convergence to rationality in terms of unbiasedness and efficiency. We find that forecast performance has improved through time, as the coefficients on the test equations shows movement towards values implied by unbiasedness and efficiency hypotheses, supporting the learning hypothesis.

Monetary transmission with multiple policy rates: evidence from Turkey

ABSTRACT Quantitative easing policies have led to persistent divergence between officially announced policy rates and short-term money market rates in many economies, making it challenging to assess the stance of monetary policy in the aftermath of the global financial crisis. Lack of data variation in short-term interest rates across time dimension has made it difficult to identify the monetary transmission mechanisms. In order to shed some light on this topic, we make advantage of a specific period from Turkey during which the central bank deliberately allowed the policy rates to diverge frequently from the interbank rates due to capital flow management purposes. Using bank-level flow data from this episode, we investigate the relationship between various short-term interest rate measures and bank loan/deposit rates through panel estimation methods. Our findings suggest that interbank rates are more relevant than central bank’s officially announced rates for the transmission of monetary policy when the two diverge from each other persistently. Interbank rates particularly play a key role in the pricing of loans and deposits. These findings provide helpful guidance for evaluating the monetary stance under unconventional policies.