The Power of Linear Methods in Determining the Dominant Fiscal Regime: A Research on TurkiyeKaan Masatcı, Asuman Oktayer
The Fiscal Theory of the Price Level suggests that in a non-Ricardian regime, an inflation targeting monetary policy will no longer capable of controlling price level, and that optimum policy should always include a combination of monetary and fiscal policies. The aim of this paper is to investigate the dominant fiscal policy regime in Turkey using the ARDL Bounds Testing Approach and to compare the results with actual observations. The review period covers 1996–2019 period using a quarterly data set. Based on the assumption that the actual behavior of the government’s fiscal policy can be observed more accurately with structural balance, the sub-components of primary balance (structural and cyclical balance) are also included in the analysis. During the review period, there were important breaks in the Turkish economy that could lead to changes in monetary and fiscal policies. To observe possible policy changes, rather than dividing the sample into different periods, dummy variables were used to reflect the difference between periods. The results of the analysis demonstrate that structural balance more accurately reflects actual behavior of the government’s fiscal policy. Another finding is that the linear ARDL approach is insufficient to capture the changes in the fiscal policy, especially in periods when the fiscal space narrows.
Hakim Mali Rejimin Belirlenmesinde Doğrusal Yöntemlerin Gücü: Türkiye İncelemesiKaan Masatcı, Asuman Oktayer
Fiyat Düzeyinin Mali Teorisi (Fiscal Theory of Price Level, FTPL), enflasyon hedefleme stratejisinin uygulandığı ekonomilerde eğer hakim rejim Ricardocu olmayan ise para politikalarının fiyat düzeyini kontrol etmede yetersiz kalacağını, bu nedenle optimum politikanın her koşulda, para ve maliye politikalarının bir birleşiminden oluşması gerektiğini ileri sürmektedir. Bu çalışmanın amacı, Türkiye’deki hâkim maliye politikası rejiminin ARDL Sınır Testi Yaklaşımının kullanılarak incelenmesi ve elde edilen sonuçların gerçek gözlemlerle kıyaslanmasıdır. İnceleme dönemi 1996-2019 aralığını kapsamaktadır ve çeyrek yıllık veri seti kullanılmıştır. Hükümetin maliye politikasına ilişkin gerçek davranışının yapısal denge ile daha doğru şekilde gözlemlenebileceği varsayımından hareketle, faiz dışı dengenin alt bileşenleri de (yapısal denge ve devresel denge) analize dâhil edilmiştir. İnceleme dönemi boyunca Türkiye ekonomisinde para ve maliye politikalarında değişikliğe yol açabilecek önemli kırılmalar yaşanmıştır. Olası politika değişikliklerini gözlemleyebilmek için örneklemi farklı dönemlere bölmek yerine, dönemler arasındaki farklılığı yansıtacak kukla değişkenler kullanılmıştır. Analiz sonuçları, yapısal dengenin, hükümetin uyguladığı maliye politikasına yönelik gerçek davranışı daha doğru şekilde yansıttığını göstermektedir. Elde edilen bir diğer sonuç, doğrusal ARDL yaklaşımının özellikle mali alanın daraldığı dönemlerde maliye politikasındaki değişimleri yakalamakta yetersiz kaldığıdır.
It seems to be widely accepted in the economics literature that monetary and fiscal policies have different objective functions. In this sense, monetary authority is assigned the task of maintaining price stability, and fiscal authority is assigned the task of stabilizing the budget and public debt stock. However, the Fiscal Theory of the Price Level (FTPL) approach, which was proposed in the 1990s, argues that monetary policy alone cannot control the price level in an economy where the fiscal policy is dominant. Leeper (1991), Sims (1994), Woodford (1994, 1995), and Cochrane (1999, 2001), the pioneers of this perspective, claim that price level is determined by a mechanism called the intertemporal budget constraint under certain circumstances:
where, Bt is the government’s nominal liabilities and Pt is price level. FTPL asserts that in a non-Ricardian regime, the government determines future primary surpluses externally and independently (indifferent to debt stock). In such a case, at a given level of debt at the beginning of period t, equilibrium can only be achieved through price level adjustments. This mechanism implies that price level indirectly depends on whether the dominant regime is Ricardian or non-Ricardian. A notable point to emphasize is that even in circumstances in which the government finances the fiscal deficit with future surpluses, fiscal policy can still be “active” as Cochrane (2020) points. According to Leeper (1991), a non-Ricardian regime remains if the government does not exhibit any fiscal reaction to fluctuations in debt stock. Hence, FTPL explains fluctuations in price level according to the dominant characteristics of the economy’s fiscal regime.
The aim of this paper is to determine the Ricardian and non-Ricardian characteristics of fiscal policy in Turkey. Referencing the FTPL approach, it is assumed that once the dominant regime is defined, inflation in Turkey will be better understood; however, fiscal policy is not only a matter of economics, but is also—perhaps predominantly—a political issue, and is therefore subject to change. In an econometric analysis, these political shifts can be captured by dividing the sample into two or more subsamples and separately conducting the analysis for those parts. Rather than using this approach, in this study, dummy variables are used to capture changes in the fiscal policy regime. As for the reasoning behind this approach, previous literature considers the 2001 Financial Crisis in Turkey to be a turning point in economic policies, generally assessing pre- and post-crisis episodes individually. However, in addition to the 2001 crisis, there are two significant events that could have direct effects on fiscal policies. First, the medium-term (three-year) fiscal plan (MTFP) was implemented in 2006. This plan was proposed as an implicit deceleration of Ricardian fiscal policies and should be included in the analysis. Second, the 2008 global financial crisis had a critical impact on economic policies; therefore, since dividing the sample into four parts would be inefficient, dummy variables were used. Moreover, as Cochrane (1999) asserts, government’s fiscal policy behavior can be observed better in structural balance rather than primary balance. The main contribution of this study is the inclusion of structural balance in the analysis along with primary balance to reveal the actual fiscal policy regime.
Nevertheless, some limitations on the power of empirical analysis may remain for analyzing the dominant fiscal regime. The main objective of this study is to assess the strength of a linear model in determining fiscal dominance. A linear method cannot perfectly fit into the analysis as it treats any negative relationship between the debt stock and the primary balance (or structural balance) as a Ricardian regime; however, in real terms, if a rise in the primary balance (or structural balance) causes a decline in the debt stock, one might claim that it is a Ricardian regime. Conversely, if a negative link remains between the two variables, this rule does not apply (decrease in primary balance, increase in debt stock).
We use the ARDL Bounds Testing Approach to test the relationship between the real debt stock and primary balance (structural balance), as this approach is suitable for the time-series properties of our dataset. The results demonstrate that the inclusion of the dummy variables and structural balance are not successful in capturing the regime shifts throughout the investigation period. Both the MTFP in 2006 and the global financial crisis in 2008 seem to be driving toward Ricardian fiscal policies; however, our second analysis, which includes structural balance, reveals the opposite effect. Fiscal policies are found to be structurally non-Ricardian, even after the MTFP. Finally, when we compare the empirical findings with real-life outcomes, particularly since 2017, while the primary balance (and/or structural balance) decreases, public debt stock increases, indicating a negative link between the variables and suggesting a non-Ricardian regime. We conclude that the findings of linear methods may not be reliable during periods when the fiscal room is narrower.