Açık Enflasyon Hedeflemesi Döneminde Neo-Fisher Etkisi’nin Geçerliliği: Türkiye Üzerine Ekonometrik Bir Analiz
Sevda YapraklıFiyat istikrarının sağlanmasına yönelik Açık Enflasyon Hedeflemesi (EH) şeklindeki para politikası stratejisi açısından son derece önemli olan iki değişken faiz ve enflasyondur. Çünkü Açık EH’de geleneksel olarak nominal faiz oranları yükseltilerek enflasyon oranları hedeflenen değerlere düşürülmektedir. Ancak 2000’li yılların ortalarından itibaren NeoKlasik ve Neo-Keynesyen görüşlere dayalı tartışmalar ve ülke uygulamaları sonucunda “Nominal faiz oranı ile enflasyon oranı arasında pozitif bir ilişki vardır ve ilişkinin yönü enflasyondan faize doğrudur.” cümlesi ile ifade bulan geleneksel Fisher Etkisi’nin geçerliliği tartışılmaya başlanmıştır. Bu çalışmanın temel amacı nominal faiz ile enflasyon arasında ilişki olup olmadığını, varsa ilişkinin yönünü, ilişkinin yönün faizden enflasyona doğru olduğunu ileri süren, Neo-Fisher Etkisi, bağlamında ekonometrik olarak incelemektir. Bu çalışmanın açık EH dönemi, veri seti ve analiz yöntemi açısından farklılaşarak, literatüre katkı sağlaması beklenmektedir. Bu amaçla Türkiye’de açık EH’nin uygulandığı 2006:01-2021:08 dönemi için devlet iç borçlanma senedi faiz oranı ve tüketici fiyatları indeksi değişkenlerine ait veriler kullanılmıştır. Çalışmada birim kök (Ng-Perron), çoklu yapısal kırılma (Bai-Perron) ve eş-bütünleşme (Johansen) testleri, Dinamik EKK regresyon tahmini ve VEC nedensellik analizi şeklindeki ekonometrik yöntemler kullanılmıştır. Analizler sonucunda değişkenlerin uzun dönemde birlikte hareket ettikleri ve nedenselliğin enflasyondan faize doğru olduğu tespit edilmiştir. Analizlerden elde edilen bulgulara göre Türkiye’de açık EH’nin uygulandığı dönemde Neo-Fisher Etkisi geçerli değildir. Bu bulgu, açık EH döneminde Türkiye’de enflasyon oranı yükseldikçe tasarruf sahiplerinin reel kazançlarını korumak için daha fazla faiz talebinde bulunduklarının göstergesi olarak kabul edilebilir. Analiz sonuçlarından hareketle, Türkiye’de kamu borçlarının reel değerini yükseltmek için enflasyonun aktif olarak kullanılmasının ve diğer temel makroekonomik göstergelerin dikkate alınmasının uygun olacağını söylemek mümkündür.
The Validity of The Neo-Fisher Effect in The Period of Explicit Inflation Targeting: An Econometric Analysis on Turkey
Sevda YapraklıInterest and inflation rates are two critical variables concerning monetary policy strategies in the form of explicit Inflation Targeting (IT) to ensure price stability. Traditionally, nominal interest rates are raised, and inflation rates are reduced to explicit IT’s target values; however, through discussions on Neo-Classical and Neo-Keynesian views, the traditional Fisher Effect has come under scrutiny. This study econometrically examines the relationship and direction of flow between nominal interest and inflation. This study contributes to the literature by providing a unique view of the open IT period, data set, and analysis method, using data regarding interest rates on government domestic debt securities and consumer price index variables from January 2006 to August 2021, when Turkey applied the explicit IT. This study uses econometric methods, such as unit root (Ng–Perron), multiple structural breaks (Bai– Perron), and cointegration (Johansen) tests, Dynamic OLS regression estimation, and VEC causality analysis. The results indicate that the variables move together in the long run, with causality running from inflation to interest. According to the analysis results, the Neo-Fisher Effect is invalid when explicit IT is applied in Turkey. These findings indicate that savers demand more interest to protect their real earnings as the inflation rate increased in Turkey during the explicit IT period. Based on the analysis results, it would be appropriate to use inflation actively and consider other basic macroeconomic indicators to manage the real value of public debt in Turkey.
The relationship between nominal interest and inflation and the direction and size of the relationship are important for countries’ macroeconomic policies concerning price stability. Explicit Inflation Targeting (IT) is one of the primary monetary policy strategies applied to ensure price stability. Explicit IT intervenes upon the nominal interest rate to eliminate the deviations in the targeted inflation rates; traditionally, the nominal interest rates are raised, and the inflation rates are reduced to the targeted values. Accordingly, there is a negative relationship between the nominal interest rate and inflation; however, the economic developments that emerged, especially after the 2008 Global Crisis and the studies of Neo-Keynesian economists, have revealed a positive relationship between interest and inflation in the short term. Through discussions based on Neo-Classical and Neo-Keynesian views, the validity of the traditional Fisher Effect (which claims that there is a positive relationship between the nominal interest rate and the inflation rate and that the direction of the relationship is from inflation to interest) has come under scrutiny.
Most theoretical-applied studies on explicit IT and policy implementations in practicing countries have shown a positive relationship between interest and inflation rates; however, the direction of the relationship is from nominal interest to inflation. The positive effect of interest on inflation began as the Neo-Fisher Effect in the literature; thus, the direction of the relationship between the two variables has become important for academics and policymakers.
This study econometrically examines the direction of the relationship between nominal interest and inflation, which is important in terms of price stability. For this purpose, econometric analyses are made using monthly data from January 2006 to August 2021, when Turkey applied explicit IT. In line with the findings, this study examines whether the Neo-Fisher Effect is valid. The second part of this study is handled in three parts; explanations are made about the literature concerning applied studies examining the interest-inflation relationship. The third chapter first discusses the data set and method used in the research, and then the results of the analyses are presented and interpreted. The last part provides general evaluations and policy recommendations.
To determine whether the nominal interest-inflation relationship is valid based on the Neo-Fisher Effect, this study uses monthly data on government domestic debt securities interest rates and CPI/inflation variables from January 2006 to August 2021, when the deficit EH was applied in Turkey. The year 2003 (2003=100) is the basis for the CPI, representing inflation. The interest rate includes fixed-rate cash borrowings realized through the auction method and shows the monthly average cost of domestic cash borrowing (compounded % per year). The moving average method is used to adjust the seasonal effects of this study’s main variables.
In addition, shadow variables (K), related to the structural break periods determined by the Bai–Perron test, are created and added to the estimation equations. The inflation and nominal interest rate variables are given a value of 1 for June 2009–November 2017 and June 2015–November 2018, and 0 for other months. Furthermore, this study investigates the existence of the relationship between nominal interest and inflation using econometric analysis methods. This study’s main analysis methods are unit root (Ng–Perron), multiple structural break (Bai–Perron) and cointegration (Johansen) tests, Dynamic EKK regression estimation, and VEC causality analysis.
Since the test statistics for MZ, MZt , MSB, and MPT tests are smaller than Ng and Perron’s critical values, the basic hypothesis of “There is a unit root” is rejected; the first difference [I(1)] values of interest rate and inflation–CPI variables become stationary. The Bai–Perron structural break tests show two breaks in this study’s variables, and the Johansen cointegration tests indicate that the variables act together in the long run.
Dynamic OLS regression findings with a structural break dummy variable show that a 1% increase in the inflation rate, in the long run, leads to a 0.67% increase in the nominal interest rate; however, the positive effect of the increase in the interest rate on inflation is statistically insignificant. Similarly, the VEC causality analysis results show that there is no unidirectional causality running from the nominal interest rate to inflation, and therefore the Neo-Fisher Effect is not valid in Turkey. Dynamic OLS and VEC causality analysis results indicate that savers demand more interest to protect their actual earnings as inflation rates rise in Turkey during the explicit IT period. Based on the analysis results, it is possible to say that it would be appropriate to use inflation actively and consider other basic macroeconomic indicators to manage the real value of public debt in Turkey.
